ML20213A406: Difference between revisions

From kanterella
Jump to navigation Jump to search
(StriderTol Bot change)
(StriderTol Bot change)
 
Line 13: Line 13:
| document type = NUREG REPORTS, TECHNICAL SPECIFICATIONS & TEST REPORTS, TEST/INSPECTION/OPERATING PROCEDURES
| document type = NUREG REPORTS, TECHNICAL SPECIFICATIONS & TEST REPORTS, TEST/INSPECTION/OPERATING PROCEDURES
| page count = 298
| page count = 298
| project =
| stage = Other
}}
}}



Latest revision as of 11:48, 7 October 2021

International Code Assessment and Applications Program. Annual Report
ML20213A406
Person / Time
Issue date: 03/31/1987
From: Hanson R, Jenks R, Ting P
NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:
References
NUREG-1270, NUREG-1270-V01, NUREG-1270-V1, NUDOCS 8704270543
Download: ML20213A406 (298)


Text

_. - - --

NUREG-1270 Vol.1 l International Code Assessment and Applications Program .

Annual Report l

U.S. Nuclear Regulatory Commission Office of Nuclear Regulatory Research E s I

7 T27318 1270 R 2

PDR

.  % n y

, l l

1 NOTICE Availability of Reference Materials Cited in NRC Publications Most documents cited in NRC publications will be available from one of the following sources:

1. The NRC Public Document Room,1717 H Street, N.W.

Washington, DC 20555

2. The Superintendent of Documents, U.S. Government Printing Office, Post Office Box 37082, Washington, DC 20013-7082
3. The National Technical Information Service, Springfield, VA 22161 Although the listing that follows represents the majority of documents cited in NRC publications, it is not intended to be exhaustive.

Referenced documents available for inspection and copying for a fee from the NRC Public Docu-ment Room include NRC correspondence and internal NRC memoranda; NRC Office of Inspection and Enforcement bulletins, circulars, information notices, inspection and investigation notices; Licensee Event Reports; vendor reports and correspondence; Commission papers; and applicant and licensee documents and correspondence.

The following documents in the NUREG series are available for purchase from the GPO Sales Program: formal NRC staff and contractor reports, N,RC-sponsored conference proceedings, and NRC booklets and brochures. Also available are Regulatory Gulds, NRC regulations in the Code of Federal Regulations, and Nuclear Regulatory Commission Issuances.

Documents available from the National Technical Information Service include NUREG series reports and technical reports prepared by other federal agencies and reports prepared by the Atomic Energy Commission, forerunner agency to the Nuclear Regulatory Commission.

Documents available from public and special technical libraries include all open literature items, such as books, journal and periodical articles, and transactions. Federal Register notices, federal and state legislation, and congressional reports can usually be obtained from these libraries.

Documents such as theses, dissertations, foreign reports and translations,and non-NRC conference proceedings are available for purchase from the organization sponsoring the pub;ication cited.

Single copies of NRC draft reports are available free, to the extent of supply, upon written request to the Division of Technical Information and Document Control, U.S. Nuclear Regulatory Com-mission, Washington, DC 20555.

Copies of industry codes and standards used in a substantive manner in the NRC regulatory process are maintained at the NRC Library, 7920 Norfolk Avenue, Bethesda, Maryland, and are available there for reference use by the public. Codes and standards are usually copyrighted and may be purchased from the originating organization or, if they are American National Standards, from the American National Standards institute,1430 Broadway, New York, NY 10018.

NUREG-1270 Vol.1 International Code Assessment and Applications Program Annual Report Manuscript Completed: March 1987 Date Published: March 1987 Prepared by P. Ting Office of Nuclear Regulatory Research U. S. Nuclear Regulatory Commission Washington, D.C. 20555 R. Hanson Idaho National Engineering Laboratory P.O. Box 1625 Idaho Falls, ID 83415 R. Jenks Los Alamos National Laboratory Los Alamos, New Mexico 87545 Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission Washington, DC 20555

p. . ,.

h.....

O

r PREFACE This is the first annual report of the International Code Assessment and Applications Program (ICAP). The ICAP was organized by the Office of Nuclear Regulatory Research, United States Nuclear Regulatory Commission (USNRC) in 1985. The ICAP is an international cooperative reactor safety research program planned to continue over a period of approximately five years. To date, eleven European and Asian countries / organizations have joined the program through bilateral agreements with the USNRC. Seven proposed agreements are currently under negotiation.

The primary mission of the ICAP is to provide independent assessment of the three major advanced computer codes (RELAP5, TRAC-PWR, and TRAC-BWR) developed by the USNRC. However, program activities can be expected to enhance the assessment process throughout member countries. The codes were developed to calculate the reactor plant response to transients and loss-of-coolant accidents. Accurate prediction of normal and abnormal plant response using the codes enh=nces procedures and regulations used for the safe operation of the plant and also provides technical basis for assessing the safety margin of future reactor plant designs. The ICAP is providing required assessment data that will contribute to quantification of the code uncertainty for each code.

The first annual report is devoted to coverage of program activities and accomplishments during the period between April 1985 and March 1987.

l l

l iii

l EXECUTIVE

SUMMARY

j o Background The Office of Nuclear Regulatory Research, United States Nuclear Regulatory Commission (USNRC) has developed three advanced thermal-hydraulic computer codes for analyzing plant response to postulated transients and loss-of-coolant accidents in light water reactors. These codes are: the RELAPS (Reactor Leak and Pressure Excursion) Code, the TRAC-PWR (Transient Reactor Analysis) Code, and the TRAC-BWR Code. After more than a decade of research effort, the major part of the code development has been accomplished. However, the qualitative and quantitative assessment of each code has not yet been completed. Code f assessment provides the necessary technical basis for determining the accuracy and validity of each code over its range of applicability.

To determine the code accuracy and applicability for reactor. safety analysis, the codes must be rigorously assessed against a matrix of experiments that covers a broad range of experiment types and facility scales. The determination of code accuracy and applicability is made based on a body of assessment data consisting of comparisons between code predictions and measured data. An independent code assessment effort is underway and will provide a major portion of the needed information.

The independent assessment effort consists of domestic and international programs. The international code assessment process is being carried out through the International Code Assessment and Applications Program (ICAP). The ICAP is an international cooperative reactor safety research program organized by the Office of Nuclear Regulatory Research, USNRC in 1985 to provide an independent assessment of the three major thermal-hydraulic codes. (A fourth advanced thermal-hydraulic ccde, COBRA / TRAC, is also receiving limited assessment under the ICAP). The program is planned to continue over a period of approximately five years.

v

o Objectives of the Program The objectives of the ICAP are to obtain an international consensus on the acceptability of the codes and to support the quantification of code accuracy. Also, through this effort, code deficiencies are discovered and reported by the code users. 'The code improvements necessary to resolve the deficiencies will be prioritized and made according to regulatory needs, o Goals and Cost Effectiveness The ultimate goal of the International Code Assessment and Applications Program is to use the quantification of code uncertainty to determine the accuracy of a full scaled nuclear power plant analysis with a '

best-estimate thermal hydraulic code for specific reactor transient types.

The international assessment will supplement the domestic assessment, thereby broadening the data base available for code assessment. The international code assessment effort is particularly important to the domestic nuclear safety research effort because the ICAP is providing assessment results for transient analyses which are not currently planned in domestic assessment programs.

The code uncertainty quantification requirement of a significant number of assessment studies for each type of reactor transient implies substantial resource expenditure. A typical assessment study will cost about $50,000, not including the cost of generating the test data. With consideration o' facilities cost, the international code assessment effort would translate to well over $75 million of reacto: safety research over the duration of the program.

l 1

The ICAP members benefit from the use of the USNRC developed codes as  ;

well as the technical assistance provided through the ICAP. Therefore, the international cooperative effort enables each participant to achieve i

vi

optimum use of limited resources to obtain state-of-the-art thermal-hydraulic codes for use in the safety related efforts of their respective countries.

o Countries Having Agreements with USNRC The USNRC is obtaining international code assessment results and test data through bilateral agreements with foreign countries possessing nuclear plants and/or experimental test facilities. Currently, the USNRC has bilateral agreements with eleven European and Asian countries / organizations. These are Belgium, Finland, France, Germany, Italy, Korea, Spain, Sweden, Switzerland, United Kingdom, and the American Institute of Taiwan. Agreements are under negotiation with Austria, Brazil, European Joint Research Center at ISPRA, Japan, Mexico, Netherlands, and Yugoslavia.

o Accomplishments as of March 1987 Accomplishments realized in the ICAP through March 1987 are many and varied. The program has developed a management document which describes the manner in which the ICAP will proceed to reach the stated program goals. The program has held an initial planning meeting, two Program Management Group Meetings, two technically oriented Specialist Meetings, and an ICAP Task Group Meeting. User guidelines have been published for each of the thermal-hydraulic codes. Similarly, a quarterly newsletter is published for each of the thermal-hydraulic codes. A significant effort has been expended addressing code uncertainty quantification with reports prepared for a time-averaged uncertainty quantification methodology and a time-dependent uncertainty quantification methodology. The effort of the ICAP on code uncertainty quantification has been in conjunction with another task to determine code scalability and applicability. Nine formal studies by ICAP members have been published by the USNRC in NUREG/IA reports.

l vii

A TRAC-PF1/M001 workshop was held at Los Alamos National Laboratory (LANL) in support of ICAP and domestic code user needs. In addition, the ICAP helped support workshops for the RELAP5/M002 code at the Eidgenossisches Institut fur Reaktorforschung (EIR) in Switzerland and for the TRAC-BD1/M001 code at the Idaho National Engineering Laboratory (INEL). Training was provided for two representatives from the American Institute of Taiwan in which two small break code assessment studies were performed at the INEL using the RELAP5/M002 code. The thermal-hydraulic codes have been subjected to the scrutiny of the international participants of ICAP, and numerous errors and recommendations have been directed into the code development effort since the inception of the ICAP.

o Regulatory Significance The underlying objective of conducting research on the thermal-hydraulic behavior of light water reactors is to prevent severe accidents. Thermal-hydraulic research provides the technical basis for ensuring power reactors are designed and operated to minimize the risk of occurrence of a severe accident.

The products of the research are validated computer codes, experimental data, and thermal-hydraulic analyses. These products are used in a variety of ways to give technical defensibility to regulatory policy and decisions. They provide the regulatory staff with an independent analytical capability for auditing licensee analyses as well as for performing studies on generic issues.

The products of the ICAP are experimental data and thermal-hydraulic code assessments with which to validate the codes for their end use in the regulatory process. The use of best estimate codes is now being introduced into the regulatory process. From the USNRC perspective, this move imposes a new requirement, that of quantifying the accuracy of the codes. The accuracy quantification, in turn, introduces new requirements with respect viii

to the form and number of code assessments'needed. The nedessary uniformity and number of' assessments can be' accomplished most cost effectively by sharing costs and products throughout the. international

- community. That role is being fulfilled by~ the ICAP. The structure of-ICAP provides additional benefits through'the increased technical quality

- and generality of the code improvements which are developed from.the ICAP code assessment data base.

4 4

i

[ ix

CONTENTS PREFACE ..................................................... iii EXECUTIVE

SUMMARY

..................................................... v ACKNOWLEDGEMENTS ..................................................... xv

1. INTRODUCTION ..................................................... 1
2. DESCRIPTION OF THE PROGRAM ....................................... 4 2.1 Program Organization ....................................... 4 2.2 Program Coordination ....................................... 5 2.3 USNRC Contract Support ..................................... 6 2.3.1 Contract Support Objectives ........................ 6 2.3.2 Contractor Services and Products ................... 6 2.4 Contents of ICAP Agreements ................................ 8 2.4.1 Program Cooperation ................................ 8 2.4.2 Forms of Cooperation ............................... 8 2.4.3 Technical Scope of Agreement ....................... 9 2.4.4 Administration of Agreement ........................ -9 2.4.5 Patents ............................................ 10 2.4.6 Exchange of Scientific Information and Use of Program Results .................................... 10 2.4.7 Final Provisions ................................... 10 2.5 Guidelines for Performing and Documenting Code Assessment Studies ......................................... 11 2.6 Tangible Products .......................................... 12
3. MAJOR FINDINGS DERIVED FROM THE ICAP ............................. 15 3.1 Code Modifications ....................................... 16 3.1.1 TRAC-PWR ........................................... 16 3.1.2 RELAP5 ............................................. 24 3.1.3 TRAC-BWR ........................................... 28 3.2 Documentation Modifications ................................ 32
4. STATU S O F TH E P ROG RAM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 4.1 Code Assessment Matrices ................................... 36 1

4.2 ICAP Assessment Calculations and Summaries ................. 36 l l

r xi

-4.3~ Code Uncertainty Analysis .................................. 39 4.3.1 Method Overview .................................... - 39 4.3.2 - Data Selection and Review .......................... 40 4.3.3- Calculation Review-................................. 40 4.3.4 Key Parameter Selection ............................ 41 4.3.5 Production of Difference Data'...................... 41 4.3.6 Uncertainty Statement .............................. 42 4.3.7 Scaling Considerations ............................. 42 4.4 Milestones-of Contract Support ............................. 43 4 . 5_ 'ICAP Reports ............................................... 43 4.6 ICAP Meetings .............................................. "49 4.7 Newsletters and Workshops .................................. ~ 49 4.7.1 Newsletters ........................................ 50 4.7.2 Workshops .......................................... 50-4.8- Laboratory Vi sits by ICAP Participants . . . . . . . . . . . . . . . . . . . . . 51

5. REGULATORY SIGNIFICANCE AND CONCLUSIONS .......................... 52

- 6. REFERENCES ....................................................... 53 J

APPENDIX A--ICAP DESIGNATED CONTACTS .................................. A-1

- APPENDIX B--TYPICAL INTERNATIONAL AGREEMENT ........................... B-1 APPENDIX C--REPORTED USER PROBLEMS .................................... C-1 APPENDIX D--CODE ASSESSMENT MATRICES .................................. D 1 APPENDIX E--ASSESSMENT

SUMMARY

METHODOLOGY ............................ E-1 APPENDIX F--ICAP REPORT SUMMARIES ..................................... F-1 APPENDIX G--ICAP MEETING MINUTES ...................................... G-1 4 APPENDIX H--BIBLIOGRAPHY OF PAPERS PRESENTED AT ICAP MEETINGS ......... H-1 FIGURES j

1. Organization of the ICAP ......................................... 4 xii y , - -- . - - , -~ +,9 ~ - -

- --- -, -- sr, ,-

4 TABLES

1. Summa ry of TRAC-PWR u se r probl ems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-
2. S umma ry o f RE LAP 5 u s e r . p robl em s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17-
3. Summa ry lo f TRAC-BWR u ser p robl ems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4. Summary of'ICAP Calculations ..................................... 37.
5. Milestones for documentation produced by the laboratories ........ 44
6. Milestones.for the services provided by the laboratories ......... 45 7 .-- ^ Reports received by the USNRC through the ICAP ................... 46 I

s I

1 L

l t

J i

i I

4 1

xiii

' ACKNOWLEDGEMENTS The authors wish to thank those who have contributed to the completion' of_the ICAP Annual Report. At the Idaho National Engineering- Laboratory, .

those. persons are G. Case, J. Jacobson, E. Johnson, N.'Larson, G. Singer, and G. Wilson. Specific thanks goes to G. Case who wrote major sections of

'the document. -At Los Alamos National Laboratory,'those persons are.

.T. Knight, J. Spore, D. Liles, B. Boyack, L. Garcia, V. Metzger, . and J. Holmes. At the U. S. Nuclear Regulatory Commission, those persons are L. Shotkin, F. Odar, D. Bessette, P. Larkins, and S. Zungoli.

xv

. 1. INTRODUCTION The International Code Assessment and Applications Program (ICAP) is an international cooperative reactor safety research program organized.by the Office of Nuclear Regulatory Research, United States Nuclear Regulatory Commission (USNRC) in 1985 to provide an independent assessment of USNRC sponsored thermal-hydraulic codes. The codes were developed for analyzing plant response to postulated transients and loss-of-coolant accidents in light water reactors. The program'is planned to continue over a period of approximately five years. The assessment results from the ICAP, along with the domestic code assessment studies, are used to qualitative.ly evaluate the code and also to quantify code uncertainty. The ultimate goal of the code assessment program is to use the quantification of code uncertainty for scaled applications of the code to determine the accuracy of the code for nuclear power plant application.

Countries possessing nuclear plants and/or experimental test facilities have been invited to participate in the ICAP by signing a l

bilateral agreement with the USNRC. Currently, the Office of Nuclear Regulatory Research has bilateral agreements with eleven countries / organizations. These are Belgium, Finland, France, Germany, Italy, Korea, Spain, Sweden, Switzerland, United Kingdom, and the American Institute of Taiwan. Agreements are under negotiation with Austria, Brazil, European Joint Research Center at ISPRA, Japan, Mexico, Netherlands, and Yugoslavia. Appendix A presents a complete list of ICAP designated contacts. The ICAP members have the same material interest in cooperating in the field of thermal-hydraulic research with the objective of improving and ensuring the safety of reactors on an international basis. The need is recognized to equally share both the resources resulting from this research and the effort required to develop those resources. The terms in these bilateral agreements are simple. The USNRC provides the thermal-hydraulic codes and related technical support through Los Alamos National Laboratory (for the TRAC-PWR code) and the Idaho National Engineering Laboratory (for the RELAP5 and TRAC-BWR codes) for the participants. In return, the ICAP participants supply code assessment 1

l

results.to the USNRC according-to a test-matrix which is mutually agreed upon. Technical information exchange among the ICAP members is achieved by published reports, individual code newsletters, and holding regular international meetings.

The ICAP results are needed for qualitatively and quantitatively assessing the code, and ultimately for determining the code uncertainty relative to full-scale application of the code. Code uncertainties are determined from the differences between code calculations and corresponding measured data. The international code assessment efforts supplement the USNRC sponsored domestic code assessment activities.

The code uncertainty quantification methodology requires several assessment studies for each type of reactor transient which represents a substantial resources expenditure. Assuming an-input deck has been developed, the cost for a typical assessment study is'about fifty thousand dollars ($50K). It is estimated that, with the aid of the ICAP, the USNRC will realize a cost avoidance of ten million dollars in the assessment of the three best-estimate codes (TRAC-PWR, TRAC-BWR, and RELAPS). This does not include the facility cost. With consideration of the facilities costs, the international assessment effort translates into well over $75 million of reactor safety research over the duration of the program. Considering the cost of required data in the code assessment program, the benefits to the USNRC are extensive since most new data are obtained from multi-million-dollar foreign facilities which the USNRC does not operate or maintain.

In addition to benefits realized by the USNRC, the international participants also benefit from the ICAP through use of the codes and technical assistance provided by the U.S. national laboratories.

The ICAP is about one year old and it has, to date, produced a number of technical papers, code related documents, and assessment reports. The ICAP reports, written by international participants, carry the special .

identification of NUREG/IA when published by the USNRC. Nine NUREG/IA reports have been published and distributed. Other major program

, 2

milestones completed include an initial planning meeting,'two Program Management Group Meetings, two Specialist Group Meetings, and a. Task Group Meeting. In addition, the program has supported three workshops, one for-each code.

This annual report is devoted to covering major activities and accomplishments of the ICAP from April 1985 through March 1987.

This report is divided into six sections, with an Executive Summary and nine appendices. Section 2 is devoted to describing the program, program organization, contract support, contracts of the bilateral agreements, program coordination, and tangible products of the program.

Section 3 presents major findings from the ICAP code assessment activities. Section 4 is devoted to a status reports of'the program.

Experimental tests and calculations completed by the participants, code uncertainty quantification methodology under development for the program, completed major milestones of the contract support (INEL and LANL), and completed technical papers and reports, as well as other ICAP accomplishments are discussed. Section 5 discusses the regulatory significance of the ICAp. The program's impact on regulatory decision making is also discussed. A list of references appears in Section 6.

t 4

l l

3

2. DESCRIPTION OF THE PROGRAM.

This section highlights the management aspects of the program; including program organization and coordination, USNRC contract support,-

contents of bilateral agreements, guidelines for performance and documentation of code assessment, as well as, tangible products of the program.

The Office of Nuclear Regulatory Research developed a document i describing the guidelines and procedures by which the International Code Assessment and Applications Program is managed.I The document is being revised. A NUREG report of the document will be issued in April 1987. .It will be updated periodically, as needed, as the program develops.

2.1 Program Organization The organization of the ICAP is summarized in Figure 1.

a BILATERAL AGREEMENTS BETWEEN TFE ICAP MEMBERS AND THE USNRC THE ICAP PROGRAM MANAGEMENT GROUP THE CODE ASSESSMENT ICAP TASK GROUPS SPECIALISTS GROUP Figure 1. Organization of the ICAP.

wrooses

- 4 l

The governing bilateral agreements of the program set'forth the rights l

.and obligations of the-ICAP members and the USNRC. The program planning, .

execution, and review of program status are coordinated through the Program l.

j Management Group. Each signatory has a. designated representative in the

{~ Program Management Group. The group is required to meet ~at"least once a ,

[ -year and may meet more often at the discretion of its members.'

i 4 The ICAP code users may meet twice a year at Specialist Group Meetings I' to present assessment results and exchange information on user experience.

All code users within the ICAP may attend the specialist meetings, 7

i .. However, attendance is not required.

4 Task Groups are formed, as needed, by the Program Management Group to

! resolve specific technical issues such as code changes, plant: applications,-

model development, supporting experimental work, etc. All members may attend task group meetings, but attendance is not required. A task group l[ ' may meet as.often as the Program Management. Group deems necessary at locations in member countries.

2.2 Program Coordination L

l As developer of the codes and common signatory to the bilateral

) agreements, the USNRC assumes responsibility for program coordination as

' well as code support and interpretation. -The USNRC designates a Program Manager whose primary responsibility is the execution of the ICAP bilateral agreements. Duties include organizing and conducting Program Management

! Group, Task Group, and Specialist Group Meetings, and coordinating the publication of ICAP reports, newsletters, topical reports, and the annual report, i

Code support and technical interpretation are provided by two U.S.

I national laboratories, the Idaho National Engineering Laboratory-(INEL),

and Los Alamos National Laboratory (LANL) under contract with the USNRC.

i

}

I i 5 I _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ - - _ _ - _ - _ _ _ - _ _ _ - _ _ _ _ - _ - _ _ _ _ _ - _ _ - _ _

The contractors assist'the USNRC in carrying out the program at the direction of the USNRC, attend, and participate in all appropriate ICAP meetings.

2.3 USNRC Contract Support

'2.3.1 Contract Support Objectives

'The-USNRC has contracted the Idaho National Engineering Laboratory

'(INEL) and Los Alamos National ~ Laboratory (LANL) to provide specific services in support of the ICAP. Because the major thermal-hydraulic codes are developed and maintained at these two national laboratories, the INEL and LANL are best able to provide the technical assistance necessary for ICAP activities. The general objectives of the contract support cover five areas:

~

(1) Maintain computer codes (frozen versions of RELAP5, TRAC-BWR, and TRAC-PWR) and user manuals for distribution to the ICAP members, as required by agreements; (2) Provide user assistance to ICAP members, as required; (3) Evaluate assessment work performed by the ICAP members and improve the code as the result of this evaluation; (4) Perform code uncertainty studies using the international and domestic code assessment results; (5) Provide assistance to the USNRC, as required.

2.3.2 Contractor Services and Products The following specific services and products are specified in the work scope of the two contractors (INEL and LANL):

6

_(1) Conduct training workshops and individual training to ICAP code users, as required; (2) Consultation and troubleshooting for specific problems, as required; (3) Support ICAP meetings as' requested by the NRC; (4) Provide documentation on code developmert, assessment,-and applications to international participants as required by international agreements and as requested by the NRC; (5) Provide code input decks and other information required for execution of the bilateral agreements; (6) Publish quarterly newsletters containing ICAP assessment results, code updates, and general code related information; (7) Review and evaluate assessment reports prepared by foreign participants. Prepare summaries of the evaluations; (8) Perform integrated code uncertainty studies to quantify the code accuracy; (9) Develop code user guidelines for each code; (10) Provide error corrections and user convenience enhancements in the frozen versions of the three thermal-hydraulic codes (RELAP5, TRAC-BWR, and TRAC-PWR);

(11) Develop model improvements for the next version of each code; (12) Provide additional services, as requested by the USNRC.

l 7 I J

. Major milestones and accomplishments completed by the two contractors are described in Section 4.4 of this report.

2.4 Contents of ICAP Agreements

. Bilateral agreements between the United States Nucle'ar Regulatory Commission (USNRC) and countries or international research organizations are based on a mutual interest for cooperation in.the field of thermal-hydraulic research with the objective of improving reactor safety on an international basis. Many agreements are extensions of earlier arrangements between the USNRC and the membe' countries for the " Technical Exchange and Cooperative effort in the field of Nuclear Safety Research and Development." These agreements form the foundation for the ICAP. The ICAP's purpose is to integrate the research into a comprehensive program.

The contents of ICAP agreements are described in the following sections, and a typical agreement is shown in Appendix B.

j 2.4.1 Program Cooperation I The cooperative research effort between the USNRC and an ICAP member is implemented in accordance with the provisions of the agreement and subject to applicable laws and regulations in force in the respective countries. Each cooperative program has a defined duration with provision for renewal. A common objective shared by all ICAP members is to assess.

the USNRC supplied thermal-hydraulic computer codes.

2.4.2 Forms of Cooperation Cooperation between the parties may take the following forms:

(1) Exchange of information in the form of technical reports, experimental data, correspondence, newsletters, visits, joint meetings, etc.;

8

-(2) Temporary assignment of personnel of one party, or-its j contractors, to laboratories or facilities owned by ~the other party or in which it sponsors research; (3) Execution of joint research programs and projects;.

(4) Use by one party of facilities owned by the other party or .in which research is being sponsored by the other party.

2.4.3~ Technical Scope of Agreement The U3NRC provides, over the duration of the agreement, the requested code (s) and code manual (s), user guidelines,. technical assistance as required by the agreement, code maintenance and support, experimental facility information and test data, as well as research reports, all as required by the agreement, in the area of thermal-hydraulic research.

The ICAP member shall provide over the duration of the agreement the l results from code assessment studies, recommended modeling modifications and improvements for the codes, as well as technical reports containing the results of any analysis performed using USNRC codes to resolve reactor safety issues.

Some bilateral agreements may contain additional specified terms and conditions mutually agreed upon.

2.4.4 Administration of Agreement This article of the agreement contains a number of provisions including description of representatives to coordinate and determine the detailed implementation of the agreement, restrictions concerning dissemination of proprietary or other confidential or privileged information, assignment of technical personnel for program access authorization to technical information, etc.

9 4

2.4.5 Patents c

j This article of the agreement deals with all aspects of inventions or discoveries conceived or first actually reduced to practice in the

implementation of a bilateral agreement. The details of terms and 4 conditions in this article have been reviewed thoroughly by both parties for legal implications before signing the agreement.

2.4.6 Exchange of Scientific Information and Use of Program Results A number of important issues concerning exchange of scientific information and use of results are dealt with in this article. The following issues are covered:

(1) Approval for publishing and distributing results;

, (2)' Restrictions on disseminating and using proprietary, confidential, or privileged information for commercial purposes; (3) Restrictions on disseminating USNRC supplied codes for commercial purposes; i

(4) Liability for damages of any type that may result from the use of the USNRC codes.

2.4.7 Final Provisions This article contains other administrative terms and conditions.

Duration of agreement, protection of proprietary, confidential and privileged information, termination of the agreement, costs arising from implementation of the agreement, as well as modification or extension of the agreement, are included in the article.

i 10

2.5 Guidelines for Performing and Documenting Code Assessment Studies Guidelines for performing and documenting code assessment studies by the ICAP members are provided in the " Guidelines and Procedures for the International Thermal-Hydraulic Code Assessment and Applications Program" document (Reference 1) developed by the Office of Nuclear Regulatory Research in April 1986; hereafter, referred to as the " guidelines and procedures" document.

Basically, a code assessment consists of a code calculation simulating a test for which qualified experimental data are available. Comparison of the calculated and measured response of the reactor : stem provides insights into the capabilities of the code, code accu.acy, and code applicability.

Guidelines for performance of code assessment recommended to the ICAP members are outlined below:

l (1) Perform a base case calculation using best-estimate techniques and recommended code options; (2) Perform sensitivity studies to explore the effect of input modeling, boundary conditions, nodalization, and code options; (3) Study the code calculation to verify whether the code predicts the phenomena observed in the experiment (a list of important phenomena for light water reactor transients and loss-of-coolant accidents are provided in the guidelines and procedures document);

(4) Evaluate the code documentation, including the user manual, user guidelines, documentation of code models, corrections and code improvements to determine their usability and correctness.

11

2.6 Tangible Products Tangible products of the program include:

(1) NUREG/IA reports:

ICAP assessment results are published as NUREG/IA reports after reviews by the USNRC and its supporting national laboratories; (2) Code user manuals for the frozen versions of the RELAPS, TRAC-BWR, and TRAC-PWR codes. The user manuals are updated to reflect user experience and code improvements made based on the

assessment results; (3) Guidelines and Procedures for the International Thermal-Hydraulic Code Assessment and Applications Program. This document will be updated to reflect recommendations made by the ICAP members and code users and experience from conducting the program; (4) Technical papers:

Technical papers may include documentation of code assessments, code uncertainty quan.ification efforts, code applicability work, and regulatory significance analyses; (5) Topical reports:

These reports include ICAP Task Group reports, code development and application reports generated by the supporting laboratories, and reports documenting the investigation and resolution of other related technical issues; 12 y -v-n-

(6) Code Uncertainty Analyses:

!~

The supporting laboratories will perform code uncertainty analyses based on ICAP assessment data. Analysis results will be

, ~ used to determine overall code accuracies; (7) Code Improvements:

Code improvements will be made periodically and a new frozen code version released. The period between new code versions will be determined by the USNRC with due consideration of the recommendations of other ICAP members; (8) ICAP Newsletters:

The supporting laboratories publish quarterly RELAP5, TRAC-PWR, and TRAC-BWR newsletters to provide ICAP members with information on program activities, assessment results, code improvements, and significant issues concerning the program; (9) Presentations at ICAP Meetings:

The ICAP participants, supporting laboratories and the USNRC present numerous technical papers at ICAP meetings. These papers serve as a major avenue of information exchange among code users internationally and domestically; (10) Formal training workshops for the ICAP participants:

4 The supporting laboratories conduct training workshops for the ICAP participants as required; 13

(11) ICAP Annual Report:

The annual report summarizes program activities and accomplishments during the previous year.

l 14

3. MAJOR FINDINGS DERIVED FROM THE ICAP This section presents major findings relevant to code use and code '

errors that have resulted from ICAP assessment activities.

The USNRC code development effort at the supporting laboratories has received feedback on err ors in the codes and their documentation from ICAP members. ICAP members identify errors and often provide suppo'rting investigation which enhances the code development activity. The active involvement of ICAP members in identifying and assisting in resolv~ing code and documentation errors and shortcomings enables the USNRC to maintain the highest quality codes at minimum expense.

The USNRC desires to focus future code development efforts on aspects of the codes for which development efforts will result in the most significant improvements in overall code performance. A thorough assessment of one version of.a code can best identify what code development activities will most improve code performance. Hence, the USNRC has released a frozen version of each code for use in the international community. Only updates to provide user conveniences and to correct coding that was not written as intended are provided for the frozen code's.

Instances in which a code correlation or model precludes reasonable calculation of available data are designated as performance deficiencies, j No code updates to fix performance deficiencies will be made to frozen versions of the codes. Resolutions of performance deficiencies will be provided when revising a code after a thorough assessment of the frozen version has been performed. New frozen versions of RELAPS and TRAC-PWR will be released mid-1987 and will include resolutions to performance deficiencies identified during the current assessment period. A release date of 1990 is planned for subsequent frozen versions of all three codes.

ICAP users currently provide a significant portion of the total feedback on code quality that code development personnel receive from 15

f domestic and international sources. Since the beginning of the program, nearly half of the TRAC-PF1/ MODI user problems and approximately one third of the RELAP5/M002 problems were reported by ICAP users.

The following subsections describe code improvement and documentation update activities performed at the supporting laboratories for the respective codes.

3.1 Code Modifications Modifications to the codes are prompted by ICAP user experience. A user problem is a documented difficulty in applying a code. User problems include difficulties in starting a calculation, code failure during a calculation, and other difficulties encountered while using the codes. In addition, calculated results that either are nonphysical or do not predict experimental behavior are identified for future code improvement activities. All code users are encouraged to submit problems encountered during assessment activities to the appropriate code development organization. Code development personnel evaluate each submitted problem and provide a response to the problem.

Table 1 summarizes the major TRAC-PWR code user problems resulting from ICAP activities. Similarly, Tables 2 and 3 summarize the major user problems for the RELAPS and TRAC-BWR codes, respectively. Appendix C highlights specific user concerns for the TRAC-PWR, RELAP5, and TRAC-BWR codes. Of these, the major findings and their associated benefits are discussed in the following subsections. Some of the major findings and associated benefits may also have regulatory significance.

3.1.1 TRAC-PWR Three major issues identified that relate to code input were:

16

- .,n- ,

. .~h t

TABLE 1. .

SUMMARY

OF TRAC-PWR USER PROBLEMS '

' Area of Concern Number

~ Input Problems 12 Modeling Deficiencies '

27 Code Errors ,

=

20 Output Problems ~ . 13 Code Implementation Difficulties 5 Code Documentation Deficiencies 12 Other Concerns j, 1 11 Total ( 100-TABLE 2.

SUMMARY

OF RELAP5 USER PROBLEMS Area of Concern Number Input Problems 17 ,

Modeling Deficiencies 6 Code Errors 82 Output Problems 4 Code Implementation Difficulties 11

' Code Documentation Deficiencies 6 Other Concerns 14 Total 140 TABLE 3.

SUMMARY

OF TRAC-BWR USER PROBLEMS

  • Area of Concern Number Input Problems 2 Modeling Deficiencies 1

, Code Errors 4

! Output Problems 6 Code Implementation Difficulties 2 Code Documentat ion Deficiencies 0 Other Concerns , _1 Total 16

a. Only problems pertaining to TRAC-BF1 (released August 1986) are summarized.

17

(1) INPUT ERROR MESSAGES:

Several users encountered difficulties that relate to input-deck processing and interpretation.of error messages associated with incorrect ii.put models. Their feedback produced improved code error checking, which subsequently made the code easier to use.

BENEFIT:

Moderate gain in ease of code usage and in reduction of undetected input errors that can lead to incorrect models.

(2) EXTRACT

Users were concerned that a current version of EXTRACT was not being maintained. An independent version of EXTRACT was developed to address this concern.

BENEFIT:

Reduced costs associated with multiple restarts and parametric calculations. Accelerated analysis.

(3) DATA ORGANIZATION:

Some users were_ concerned about the organization of the data in the input deck. They found that either the data were not organized to their liking or the organization lacked flexibility. Some input descriptions in.the code manual were

i ircproved, and improved user guidelines were provided.

BENEFIT:

i Improved ease of modeling.

18 y

Users identified the following nine major issues that relate to model deficiencies:

(1) CODE SPEED:

Many users want improved code performance. The three-dimensional two-step numerics in TRAC-PF1/ MOD 2, will provide improved performance. Modeling techniques to enhance code running speed were investigated. Work was initiated on code vectorization of

.the code to improve performance.

BENEFIT:

Reduced costs. Faster analyses.

(2) HEAT STRUCTURES:

Some users wanted a more general heat-transfer modeling capability so that they could specify heat transfer on a cell-by-cell basis for all components. This capability will be available in MOD 2, BENEFIT:

Broader code application. More accurate heat-transfer modeling capability. Users not limited by specialized models (STGEN, VESSEL,etc.).

(3) STEAM SEPARATOR:

Users wanted to be able to identify liquid carry-over and steam carry-under in the separator. As a result of this need, a variable efficiency separator model was developed and distributed

.to users. A model description and input specifications will be included in the April 1987 TRAC-PWR newsletter.

i 19

BENEFIT:

Improved modeling of steam separation and improved prediction of steam generator liquid inventory.

(4) FUEL DAMAGE:

Users requested more extensive fuel-damage models-in the code.

In response, work continued with TRAC /MELPROG that is specifically designed to perform post-core-meltdown studies in conjunction with a TRAC-PWR systems analysis.

BENEFIT:

Expanded modeling capability for predicting post-core-meltdown system behavior.

(5) VESSEL MULTIPLE SOURCES:

Several code users wanted the capability to connect more than one external one-dimensional ~(10) component-to the same VESSEL cell.

A multiple-source model that allows this capability has been developed.

BENEFIT:

Expanded modeling capability, broader application, and easier modeling assure a more accurate VESSEL model and reduce chances of modeling errors.

(6) GENERALIZED ID COMPONENT:

Some users wanted a generic component so that they would not be l forced to use multi-cell components (e.g. , PUMP for the momentum -

source, VALVE for the variable junction area, and single 20

connection for the VESSEL cells). Plans that would make a l

generalized ID component available in M003 have been developed.

BENEFIT:

Reduced modeling complexity. More cumurate and consistsnt user application. Faster running code.

(7) CONSTITUTIVE EQUATIONS:

Many users have suggested improvements that they believe are needed to model heat-transfer coefficients under varying fluid conditions more accurately. In response, plans that will improve the overall constitutive package in future code versions have been developed. The new versions are planned for late 1987 and 199D.

BENEFIT:

Improved prediction of important physical phenomena. More accurate prediction over broader range of fluid conditions.

Improved prediction of peak clad temperature (PCT).

(8) NODALIZATION DEPENDENCY:

There was a growing demand for improved guidelines.regarding noding and nodalization dependencies during 1986. A program to study noding dependency and to improve user guidelines was initiated.

BENEFIT:

Improved accuracy of simulation. In some cases, the use of more nodes has not improved code predictions. Complete nodalization guidelines will reduce computation costs and increase the accuracy of calculations.

21 I

(9) ACCUMULATOR:

Nitrogen injection rates have caused some users diffi_culty.

Accumulator guidelines will be made available as future changes-to the " TRAC User's Guide."

BENEFIT:

Improved modeling of emergency core-coolant injection,'that allows more accurate prediction of system inventory and i time-to quench during refill for a large break loss-of-coolant accident (LBLOCA).

The major code error issues identified by users are:

(1) HORIZONTALLY STRATIFIED FLOW:

Users reported errors in the stratified flow model. An intense.

investigation produced changes to correct the problem.

BF.NEFIT:

More accurate prediction of horizontal reflux cooling flow behavior during low-flow situations that may occur during natural circulation.

(2) WALL HEAT TRANSFER:

An error in the wall heat-transfer package that caused the code to use an incorrect conductivity was corrected.

BENEFIT:

More accurate prediction of stored energy in the structural material. This improvement is not important for rapid transients or large vessels where wall effects are not dominant.

22

.(3) PRESSURE LOSSES AT TEES:

Some users reported that pressure losses at TEES appear to be incorrect. The problem is being investigated for future resolution as resources permit.

l l

l BENEFIT:

Improved prediction of system pressure losses and, hence, flow behavior in ID components.

The major issue raised relative to code output considerations was:

. CODE DIAGNOSTICS:

Some users reported that they had difficulty relating error diagnostics to code problems. In response, improved diagnostic messages and more diagnostic options (IDIAG, etc.) were developed. Also, improved output documentation with annotated listings of TRCOUT and TRCMSG was included in the " TRAC User's Guide."

BENEFIT:

Input model debugging easier. Less chance of user error in modeling. Reduced analysis time and reduced calculation cost- .

t Relative to code implementation, the following major issue was identified:

FORTRAN 77:

-Users wanted code FORTRAN updated to FORTRAN 77. In response, TRAC-PF1/M002 uses ANSI FORTRAN 77 throughout.

23

A BENEFIT:

Expanded programming capability, especially in regard to user interfaces that require character variable and string manipulation. ,

I 3.1.2 RELAp5 Users identified the following two major issues relative to code input concerns:

(1) MODELING FLEXIBILITY:

Several users have requested a code update that will permit the connection of hydrodynamic components in ways that currently are not permitted. Requests under investigation include permitting multiple connections to a time dependent volume, specification of a check valve at a pump junction, and specification of a crossflow junction at a valve and a time dependent junction.

BENEFIT:

Easier,and more realistic modeling of thermal-hydraulic systems.

(2) SI UNITS IN CSUBV TABLE:

A request was received to allow the CSUBV table for motor and servo valves to be in SI units. Currently, only British units are allowed.

BENEFIT:

A change will make the code easier to use and decrease the chance of analyst errors in the international community.

24

Four major issues were identified that relate to code model l

deficiencies. Resolution of the model deficiencies will be addressed when the next frozen version of the code is produced. These major issues are:

i (1) REFLOOD BEHAVIOR:

Code users have reported void fraction oscillations and inadequate prediction of rod temperature' profiles in calculating reflood behavior.

BENEFIT:

Improved modeling of reflood behavior and prediction of fuel rod temperatures. Better prediction of LOCA severity.

(2) CHF CORRELATION:

A late prediction of dryout, coupled with an underprediction of core inventory, suggests an inadequate CHF correlation is used in the code.

BENEFIT:

More accurate prediction of core thermal behavior during dryout.

Better prediction of rod temperatures and fuel integrity.

(3) STEAM GENERATOR DRYOUT:

The code failed to predict dryout in the steam generator for a volume that contained mostly superheated steam.

BENEFIT:

1 Elimination of nonphysical predictions in the steam generator. I Improved calculation of the fluid state.

25

4 3

(4)- PUMP CAVITATION MODEL:

a The code. pump model is ' ineffective in simulating pump performance.

degradation for-pump _ cavitation if fluid at the inlet is only slightly subcooled. There is potential for' cavitation on the impeller where the flow is accelerated.

BENEFIT:

More accurate simulation of pump behavior and recirculation: flow

in certain off-normal conditions.

The major code error issues identified by' users are:

(1) TWO-STEP NUMERICS:

1 I

i The two-step (nearly-implicit) solution option is intended'to reduce calculation time and costs. However,. users have reported some instances in which use of the two-step option resulted in

) longer run times or different (from the one-step option) '

calculation of hydraulic behavior. ,

f BENEFIT:

i Faster-running code with no compromise in simulation quality.

(2) WATER PROPERTY FAILURES:

On occasion, the code has failed with a water = property error when very low pressure is calculatedf or when the critical pressure is approached.

i BENEFIT:

Proper simulation of all fluid states that may arise during postulated transients.

J 26-w -p- -p w q. - - -g- - - , a- e e

l (3) HEAT TRANSFER:

l A condition was reported in which heat from a heat structure was supplied equally to the liquid and vapor phases, even when only one phase was present. Another user ~ reported inability to_ reach a steady state because the code was oscillating between two heat transfer correlations. Some type of smoothing between the correlations is needed. Alsc, an error in the calculation of heat transfer in the presence of a noncondensable gas was corrected.

BENEFIT:

Better calculation of thermal behavior 'in the system. Improved accuracy of simulations.

(4) ACCUMULATOR MODEL:

User reported problems motivated changes to the code's accumulator model. Code modifications include changes to permit a restart calculation and proper calculation of accumulator liquid level. Investigation is continuing to discover why the t

code fails when the accumulator empties into a volume that contains only a noncondensable gas.

BENEFIT:

Improved calculation of accumulator behavior. Improved r

simulation of emergency core coolant injection systems.

(5) CRAY VERSION OF CODE:

Several editorial errors specific to the CRAY version of the code were identified by users. Updates submitted by users or provided by code development personnel fixed the errors.

~~

27

BENEFIT:

Machine dependent differences in calculations can be evaluated.

Broad use of the code assists a thorough assessment program.

3.1.3 TRAC-BWR l

The current frozen version of TRAC-BWR, TRAC-BF1, was released in j June 1986. TRAC-BF1 contains extensive modifications and . improvements i relative to the previous frozen version, TRAC-BD1/M001. Because TRAC-BF1 has been available for only a few months,-little user feedback pertaining to the code has been received to date. Thus, a discussion of major user concerns pertaining to TRAC-BF1 is inappropriate at this time. The following discussion presents major issues identified during the assessment of TRAC-BD1/ MODI, and describes how the issues were resolved in TRAC-BF1.

The following four major issues were identified relative to model deficiencies:

(1) NEUTRON KINETICS:

The point kinetics model of TRAC-BD1/M001 was found to be inadequate for ATWS calculations. A one-dimensional neutron kinetics model is a more realistic way of representing axial power variations during all BWR transients. A one-dimensional neutron kinetics model has been implemented in TRAC-BF1.

BENEFIT:

More accurate predictions of BWR transients, especially ATWS.

Improved code capability in modeling BWR systems.

l 28

(2) CONDENSATION MODEL FOR STRATIFIED VERTICAL FLOW:

TRAC-BD1/M001 grossly overpredicted condensation rates in volumes with stagnant or moving stratified liquid levels. The new condensation model for stratified vertical flow in TRAC-1F1-results in a realistic prediction of interfacial heat exchange and condensation rates.

BENEFIT:

4 4

Improved calculation of the fluid's hydrodynamic state in relatively slow transients.

(3) INTERFACIAL DRAG CORRELATION:

The TRAC-BD1/M001 interfacial drag correlation resulted in unsatisfactory predictions of liquid level during boiloff and reflood transients. Work performed at EIR, Switzerland showed that a new correlation developed by Bestion in France resulted in excellent agreement with measured data. The Bestion correlation for interfacial drag has been implemented in TRAC-BF1 for rod bundles.

i 1

BENEFIT:

I Improved simulation of boiloff and reflood behavior in BWR rod bundles.

(4) INTERFACIAL HEAT TRANSFER:

i-A new annular flow condensation correlation for interfacial heat transfer, which provides better agreement with measured data than the TRAC-BD1/ MODI correlation, was implemented in TRAC-BF1.

29

^

BENEFIT:

Improved calculation of the thermal-hydraulic fluid behavior'in annular flow condensation.

The following five items addressed in the TRAC-BF1 development effort-resulted in a significantly faster running code:

(1) COURANT LIMIT VIOLATION:

Courant limited TRAC-BD1/M001 calculations resulted in long computer run times and high calculation costs. A two-step

-Courant limit violating solution technique, installed in TRAC-BF1, allows code users to run slow transients with large time steps.

BENEFIT:

Savings in computation time and cost.

'(2) SEPARATOR / DRYER MODEL:

With the implementation of the Courant violating numerics in TRAC-BF1, the time step limit imposed by the TRAC-BD1/ MODI separator / dryer model became unacceptable. The model was modified for TRAC-BF1 to allow the use of larger time steps.

i BENEFIT:

Savings in computation time and cost.

(3) TURBINE MODEL:

The turbine model in TRAC-B01/M001 was designed to provide a basic capability to model BWR' main steam turbines. The model is satisfactory for calculations with time step size equal to or 30

less than a tenth of a second.' In order to guarantee ~

computational stability with the TRAC-BFI two-step numerics, an implicit turbine model was developed and installed in TRAC-BF1.

l BENEFIT:

Savings in computation time and cost.

(4) CONTROL SYSTEM:

A modification of the TRAC-BD1/ MODI control system logic was developed for TRAC-BF1. To avoid inaccuracies due to large time steps allowed in the hydraulics solution, the new control logic selects its own time step and solves interconnected control blocks implicitly.

BENEFIT:

Faster running code with no degradation of co~ntrol system effectiveness. Lower analysis costs. ,

(5) PRE-LOAD PROCESSOR:

The TRAC-BF1 preload processor allows a reduction in the amount of storage required to run the code. The preload capability determines, on the basis of an individual input deck, which-subroutines will actually be needed for a particular run and only loads those routines needed before executing.

1 BENEFIT:

Shorter turnaround time leads to faster analyses. Reduction in computer costs for memory during long runs.

31

Two major issues relating to user conveniences were resolved during the TRAC-BF1 development effort. They are:

(1) CONTROL SYSTEM LOGIC:

The control system in TRAC-B01/ MOD 1 required the code user to carefully number all control blocks according to the flow of signals. The control system has been modified so that in TRAC-BF1 the user may number the blocks in an arbitrary fashion, and the code will establish the correct order of treating them according to their function and the paths of the signals.

BENEFIT:

Improves ease of modeling for the user. Facilitates modification of existing control systems. Saves analyst time, and hence, reduces analysis costs.

(2) PORTABILITY:

TRAC-BF1 has been updated so essentially all coding is ANSI standard FORTRAN 77 and IBM compatible.

BENEFIT:

The code is easily adaptable to virtually any computing environment.

3.2 Documentation Modifications Documentation updates are provided to correct errors in current code manuals, to clarify or elaborate upon existing documentation, and to keep documentation up to date with improvements made to the codes. ICAP members are notified of code error corrections to the codes through the various ICAP newsletters for the respective codes. The newsletters contain descriptions of all modifications made to the frozen code versions.

32

i i

l l Changes or corrections to code manuals and user guidelines are attached to l .the newsletters when necessary. These attached pages can directly replace pages in the current' documents, thereby retaining the clarity of the code documents while ensuring they are up to date. Updates to the TRAC-PWR Code Manual and User's Guide were distributed-in October 1986 and January 1987.

The RELAP/M002 Code Manual was released in August 1985 (Volume 1) and December 1985 (Volume 2). Updates to three pages of the RELAP5/M002 Code Manual were produced and distributed to ICAP members in April 1986.

TRAC-BD1/M001 user guidelines were published in November 1985. The code manual for TRAC-BF1 was published in August 1986.

Documentation concerns for the TRAC-PWR, RELAP5, and TRAC-BWR codes l- are presented in Appendix C. The major documentation issues relative to the TRAC-PWR code are as follows:

(1) INCONSISTENCIES IN CODE MANUAL:

Many users have found inconsistencies in the old manual. A newly revised code manual in a binder format has been issued. The new format has helped to keep the code manual current with the latest version of the code.

BENEFIT:

Improved input-modeling accuracy. Accelerated use of advanced and improved modeling capabilities. Reduced troubleshooting costs.

(2) TRIPS AND CONTROLS:

Several users expressed concern that trips and control modeling was difficult to understand and implement. An effort to revise and expand the discussion of TRAC-PWR control-system modeling was  ;

initiated. More modeling flexibility as well as more

" stand-alone" control systems were provided (e.g., constrained 33

steady-state self-initialization and PI and PID controllers).

Guidelines and examples for control system modeling were also included'in the new User's Guide.  !

'l i

BENEFIT: l Improved and more accurate modeling of plan't or test-facility control systems. More accurate simulation of operator or equipment-actions. Improved understanding of automatic plant control systems.

Documentation issues relative to the RELAP5 code are as follows:

(1) ERROR CORRECTIONS:

Various errors in the code manual have been reported by code users. An extensive revision of the manual will be completed by

.the end of March 1987. The revistrn will expand upon and clarify much of the code manual, as well as correct the reported errors.

BENEFIT:

Clearer'information on code input, system models, and solution techniques facilitates code use and results in improved analyses.

(2) SELF-INITIALIZATION:

A self-initialization option was completed for RELAPS/ MOD 2 in 4

September 1986. The option automatically drives a model to steady-state, thus saving time and effort of the code user.

Documentation describing-the self-initialization option and its use will be included in the March 1987 revision of the code manual.

34 1

- . . . .. - .- -- .~ . -

-t BENEFIT:

l Reduction in analyst and calculational computer time to. perform model initialization. Savings in time'and costs of performing l analyses'.

t l-No major issues relative to th'e TRAC-BWR' code have- been identified '

since the TRAC-BF1 code.and manual were released in August 1986.

l' I

s t

4' i

.i 4

I i

i I

.I 1

i i.

1

]

35

+

. . - , _ ,-. _ , _ . . .. . _ _ _ ._ , . _ . . , . . _ _ - . _ . . . - ~ , . . . . , . ,- . . _ - . , , ,

4. STATUS OF THE PROGRid Thissectionlistsanddescrfbesspecificact.s ities and accomplishments of the ICAP between April 1985 and Parch 1987. Items relevant to code assessment matrices, assessments performed by ICAP members, specific program milestones, and the status of the USNRC's code uncertainty quantification effort are presented.

4.1 Code Assessment Matrices The ICAP produces a system.atic code assessment effort which will provide a data base to enhance the code development effort. The data base will also be used for quantifying code uncertainty. At the foundation of this program is a defined matrix of assessment studies to be performed.

This matrix must consist of a sufficient range of experiments to produce a broad data base for determining code applicability to full-scale plants.

Three code assessment matrices currently exist in the ICAP. The matrices summarize the code assessment activities planned by each ICAP participant for each code. ICAP participants agree to perform a set of assessment studies for one or more of the subject USNRC thermal-hydraulic computer codes as fulfillment of each specific bilateral agreement with the USNRC. Tables 01, D2, and D3 in Appendix D present these matrices for the RELAP5, TRAC-PWR, and TRAC-BWR codes.

4.2 ICAP Assessment Calculations and Summaries Reports of code assessment studies performed by ICAP members are a major benefit the USNRC receives from its involvement in the ICAP. This ,

subsection briefly summarizes the assessment reports forwarded to the USNRC by ICAP participants, to date. A formal review method ~for the assessment reports is presented in Appendix E. In addition, some identified code shortcomings of importance to code development and/or regulatory concerns are presented. Table 4 lists the ICAP assessment studies that have been submitted, including information that characterizes the respective 36

I TABLE 4.

SUMMARY

OF ICAP CALCULATIONS

~

Code / Country Transient Facility Scale Reference

' TRAC-PD2 blowdown Super Cannon 2 F.R. Germany TRAC-PF1/ MODI large break' LOFT 1/60 3 United Kingdom RELAP5/M002 small break LOFT 1/60 4 Switzerland RELAP5/M002 intermediate FIX-II' 1/777 5 Sweden break-RELAP5/ MOD 2 SGTR DOEL-2 1/1 6 Belgium RELAPS/MGD2 . dryout separate small 7 Sweden effects RELAP5/M002 boil-off/ NEPTUN 1/1 8 Switzerland reflood RELAP5/ MOD 2 critical flow MARVIKEN large 9 Sweden RELAP5/M002 level swell MARVIKEN large 10 Sweden RELAP5/M002 start-up DOEL-4 1/1 11 Belgium TRAC-BD1/ MODI boil-off/ NEPTUN 1/1 12 Switzerland reflood' l

37 l

1

-- , , . - .- - - - - --n,-. ., . . , - -

l studies. To.date, the listed reports have received cursory review only; however, full reviews are in progress and are expected to be completed by September 1987.

Two assessment reports (References 2 and 3) pertaining to TRAC-PWR have been submitted to the USNRC by ICAP members over the past year.

Reference 2 assessed the code against Super Cannon and Heiss Dampf Reactor blowdown data. Reference 3 assessed the code against a simulated large break loss-of-coolant accident (LOCA). Both reports assessed nonfrozen versions of the codes. Their conclusions may or may not apply to the official frozen version. Appendix F contains a brief summary of these documents.

Eight assessment reports pertaining to the frozen version of RELAP5/ MOD 2 have been submitted to the USNRC. One of the reports, Reference 11, was received in February 1987, and thus has not received any review, to date. Of the seven reports given cursory review, four assessed the code against separate effects data, and three assessed the code against data from integral facilities. The separate effects studies addressed rod bundle boil-off and reflood behavior, post dryout heat transfer, level swell in a large vessel, and critical flow. The integral assessment reports involved small and intermediate break loss-of-coolant accidents, and steam generator tube rupture transients. Small, intermediate, and full scale facilities provided the data for the integral assessments. A summary of each assessment report, including significant findings of the report, is contained in Appendix F.

Several RELAPS/ MOD 2 shortcomings were identified in the submitted reports. Three studies (References 4, 5, and 7) found that RELAP5/M002 calculated dryout later than was experimentally observed. The Critical Heat Flux (CHF) correlation in the code was identified as the cause of the late calculation of dryout. Three independent assessments (References 6, 8, and 10) noted problems in the interphase drag package of the code.

Excessive level swell in rod bundles and the riser region of the steam genecator was attributed to excessive interphase friction. Reference 10 also indicated that the interphase friction becomes too low at high vapor 38

void fractions. Also it was noted that due to different geometries, (rod bundles,-vessels, pipes)-the friction correlation should be component l dependent. Reference 6 and Reference 8 suggestedLthat inappropriate application of the RELAP5/M002 slug flow regime was responsible for the failure of the code to reproduce the observed phenomena. The addition of flow regimes to RELAP5/M002 was proposed to be the proper. solution to the code deficiency.

One assessment report (Reference 12) pertaining-to TRAC-BD1/ MODI was received from an ICAP member in 1986. The report assessed the code against full scale bundle boil-off and reflood experiments. The report iden'tified deficiencies in the bubbly / slug interfacial friction correlation in the code. It was also noted that TRAC-BD1/ MOD 1 numerics produce calculations that are very sensitive to the form of the correlations used in the code.

Especially in calculating reflood behavior, the numerics used in the code were found to overshadow the physical phenomena being calculated. A brief summary of this assessment report is included in Appendix F.

4.3 Code Uncertainty Analysis 4

The USNRC intends to use international and domestic assessment studies to form a data base which will be used to quantify code uncertainty. The quantification of code uncertainty initially will be used to assist the direction of code development and assessment activity resources.

Ultimately, quantification of code uncertainty will contribute to the determination of code applicability and associated uncertainty for predicting plant transient behavior. A code uncertainty quantification methodology is being developed. The key elements of the methodology are presented and discussed in the following subsections.

4.3.1 Method Overview The proposed method of uncertainty quantification calls for producing a separate statement of code uncertainty for each transient of. interest and for each sub-interval of each transient. Typically, a specific phenomenon dominates the response of a facility during a specific interval of a 39

transient. Because the-codes may calculate different phenomena with varying accuracy, separate statements of code uncertainty for the sub-intervals of each transient can best indicate which parts of the codes require further development and which parts are adequate. Also, because interaction often occurs between code correlations and/or models, it is necessary to provide an uncertainty stateme, for'several key parameters to 4

ensure the effect of ccmpensating errors is not significant.

The proposed method of code uncertainty quantification prescribes quantification through a statistically based comparison of code calculated values and measured data. ICAP assessment activities will provide much of the necessary data base.

4.3.2 Data Selection and Review The quantification of code uncertainty requires a comparison of code calculations and measured data. The data chosen for the comparison must satisfy two requirements. First, the uncertainty of the measured values used for comparison must be well known. Second, the data must come from a facility that exhibits phenomena expected in a full sized plant. The ultimate goal of uncertainty quantification is to determine code applicability to nuclear reactors. Thus, the uncertainty statement must be relevant to phenomena expected in those reactors.

In addition to the above requirements, the data should come from facilities for which detailed descriptions and drawings are available.

Facility descriptions are essential to the production of an accurate input model of the facility for use in code calculations.

4.3.3 Calculation Review Code calculations for assessment studies must represent a best estimate application of the code. The code should be applied in'a manner consistent with existing recommended user guidelines and the code manuals.

The facility model should accurately represent the facility configuration and be nodalized so as to capture the system response of the experiment.

40

The nodalization must also be compatible with the' instrumentation in the facility to allow direct comparison of the measured and calculated responses.

4.3.4 Key Parameter Selection The selection of key' parameters to be used for the quantification of code uncertainty requires first, that the selected parameters capture the important phenomena, and second, that sufficient parameters are selected in an effort to identify compensating errors that may exist. A select group of key parameters that characterize the transient res'p'onse of the system and the related severity of the transient is sufficient. The statistical analysis should address the key parameters associated with the important phenomena and processes of the transient.

A consideration relative to the quantification of code uncertainty is calculating the correct response for the wrong reason. The presence of compensating errors may be determined by performing an evaluation of several key parameters for each assessment study. If good agreement is

achieved for several key parameters throughout the system, it is likely that the phenomena are being correctly calculated. If the code is not calculating the observed phenomena, the uncertainty statement should simply indicate the code deficiency and provide feedback to the code development effort. Careful selection of key parameters will ensure that the code capability to duplicate the correct phenomena can be evaluated and that an
appropriate statement of code uncertainty can be made.

l 4.3.5 Production of Difference Data After the suitability of the experimental data is assured, the nodalization and code options used are determined to be appropriate, and the key parameters are identified, the uncertainty calculation can be performed. The transient is first subdivided into time intervals during j which dominant phenomena that drive the transient response do not change.

The time intervals are unique for each transient type and may be unique l l

41 4

relative to the parameter observed. If the code fails to duplicate the observed phenomena'in.an early time interval, calculations of code uncertainty-in later time. intervals may be unnecessary since the calculated response is likely to be time dependent.

' Definition of the time intervals is followed by preparing difference data for.the specified time intervals during which the cods calculates.the  ;

I observed response. As the first step in preparing the difference data _;

within each time interval, the code-calculation data are interpolated to l provide comparable calculated and measured data points. The differer.ce data are obtained for a specific' parameter by subtracting the measured value from the calculated value at each point in time.

4.3.6 Uncertainty Statement .

i The uncertainty statement provides a usable single valued statement

]'

about the ability of the code to calculate the response of a particular key parameter during an icentified time interval for a specific type of a

transient. The uncertainty statement is a statistical representation of 3

the difference-data. The uncertainty statement produced is the time 1 averaged bias of the code minus experimental data and associated variance bands.

The differer.cr data population should consist of difference data from a number of experiments and from several facilities. The main restriction on elements of the population is that they must originate from experiments that are similar in nature and exhibit the same dominant thermal-hydraulic 1 phenomena.

4.3.7 Scaling Considerations The ultimate objective of code uncertainty quantification is to arrive at an uncertainty statement derived from <Lbscaled facilities that is valid 4

for future full-scale code applicaticos. The methodology relative to using the available subscale studies and isolated full scale data to determine code uncertati.ty for full scale applications is in the development stage.

42-

/

Clearly,:the accuracy of code predictions should not be' dependent on facility scale. If scale dependencies are identified during the

! uncertainty quantification analysis, the dependencies should be l quantified. If unacceptably large, they will be eliminated through future code development efforts.

4.4 Milestones of Contract Support The Idaho National Engineering Laboratory (INEL) and Los Alamos National Laboratory (LANL), at the direction of the USNRC, provide administrative and technical support for the ICAP. Laboratory milestones important to the support of ICAP activities are presented in this section.

Laboratory milestones important to ICAP activit.ies fall into two categories. First are those milestones which exist for the purpose of ICAP documentation support. Table 5 presents these milestones. Second are I those milestones which describe other services provided by the laboratories. Table 6 lists these milestones.

In addition to activities represented by specific milestones, the supporting laboratories provided technical support for the ICAP participants. Numerous user problems were reported and resolved, and guidance was provided for the proper use of the codes. (Significant user concerns are listed in Appendix C for the respective codes.)

4.5 ICAP Reports ICAP participants have provided written assessment reports to the USNRC. Assessment reports contain code assessment results and information pertaining to modeling and user experience. In addition, the participants have provided other reports that contribute to the development and improvement of the codes. The USNRC has published some of the submitted reports as NUREG/IA documents. Tabla 7 contains a list of a'll submitted reports. Also listed are the country contributing the report and the type of information the report contains. The code assessment reports submitted to the USNRC by ICAP members were dise nsed in Section 4.2.

43

TABLE 5. MILESTONES FOR DOCUMENTATION PRODUCED BY THE LABORATORIES Product Delivered Delivery Date Laboratory RELAP5 Newsletter Vol 1 No 1 7/85 INEL TRAC-BWR Newsletter Vol 1 No 1 7/85 INEL RELAP5/M002 Code Manual Vol 1 8/85 INEL TRAC NEWS Vol 1 No 1 (TRAC-PWR) 10/85 LANL RELAP5 Newsletter Vol 1 No 2 10/85 INEL TRAC-BWR Newsletter Vol 1 No 2 10/85 INEL RELAP5/M002 Code Manual Vol 2 12/85 INEL 4

TRAC NEWS Vol 2 No 1 1/86 LANL RELAP5 Newsletter Vol 2 No 1 1/86 INEL TRAC BWR Newsletter Vol 2 No 1 1/86 INEL TRAC-PWR User Guideline 2/86 LANL TRAC-PWR Code Manual 3/86 LANL TRAC NEWS Vol 2 No 2 4/86 LANL RELAP5 Newsletter Vol 2 No 2 4/86 INEL TRAC NEWS Vol 2 No 3 7/86 LANL RELAPS Newsletter Vol 2 No 3 7/86 INEL TRAC-BWR Newsletter Vol 2 No 3 7/86 INEL Updates to TRAC-PWR Documentation 10/86 LANL TRAC NEWS Vol 2 No 4 11/86 LANL RELAP5 Newsletter Vol 2 No 4 11/86 INEL TRAC-BWR Newsletter Vol 2 No 4 11/86 INEL-TRAC NEWS Vol 3 No 1 2/87 LANL RELAP5 Newsletter Vol 3 No 1 2/87 INEL 44

TABLE-6. MILESTONES'FOR THE SERVICES PROVIDED BY THE LABORATORIES Services Date Laboratory Support for ICAP Initiation Meeting. 4/85 INEL/LANL Support for-First Management Group Meeting 10/85- INEL/LANL Support for First Specialist Group Meeting - 6/86 INEL/LANL

~

Support for First Task Group Meeting 6/86 INEL/LANL I Support for Second Management Group Meeting 11/86 INEL/LANL 4

i Support for Second Specialist Group Meeting 1/87 INEL/LANL 1

Support for Water Reactor Safety Meeting 10/85 INEL/LANL

Support for Water Reactor Safety Meeting 10/86 INEL/LANL i

TRAC-PWR Workshop 3/86 LANL

$ RELAP5 Training (onsite) 10/86 to 1/87 INEL TRAC-BF1 Release 6/86 INEL-4 4

1 i

I.

4-4 i

45 l

l

_________.____.____._______i____..___..________._____.___.____._______ ___.__m

~

3 '/ '

c e i , :?

TABLE 7. 'REMRTS RF.CEIVED BY THE USNRC THROUGH -THE ICA9

  • 9 r ,

,' (- t Contributing- Type _of Report Title ,JCountry Information Reference-

\

Assessment of TRAC-PD2 F.R.' Germany- TRAC-PD2 2 Using Super Cannon and Assessment-HDR Experimental Data I ,

l'

,, Analysis of-LOFT Experimental United Kingdom TRAC-PE1 (3  ;

LP-LB-1 Using TRAC-PF1/M001-As's essment i i

j.

RELAP5/ MOD 2 Assessment: . Switzerland / RELAP5/ MOD 2 s

4 OECD-LOFT Small Break ' Assessment _

Experiment LP-SB-3

~

1 Assessment of RELAP5/M002, Sweden RELAP5/M002 5/

, Cycle 36, Against FIX-II Assessment

- Split Break Experiment No. 3027 .

'j n '

Assessment Study of RELAPS Belgium RELAP5/ MOD 2 6 f$[, ',y Cycle 36.01 Based on the Assessment 4 'S DOEL-2 Steam Generator m,.

Tube Rupture Incident of ,

June 1979-Assessment of RELAP5/M002 Sweden RELAP5/f,0D2 7 Against Twenty-Five Dryout As se ssmeni.,,

Experiments' Conducted at the Royal Institute of Technology Implementation and Switzerland RELAP5/M002 8 Assessment of a New Bubbly / Assessment Slug Flow Interfacial Friction Correlation in -

~ f , [) ]'

  • I RELAP5/M002/36.02 ,,

, ./ c ,,

Assessment of RELAPS/ MOD 2 Sweden RELAPE/ MOD 2 "r 9 Against Critical- Flow Data h y Assessment;'.

from MARVIKEN Tests J1T 11 and CFT 21 Assessment of RELAP5/M002 Sweden- RELAP5/ MOD 2 10 Assessment Against MARVIKEN Jet Impingement Test 11 Level 4 Swell 46

TABLE 7. (continued)

Contributing. Type of Report Title Country Information Reference Assessment Study of Belgium RELAP5/M002 11-RELAP5/M002 Cycle'36.04 Assessment Based on Spray Start-up-Test for DOEL-4

/

Assessment of TRAC-BD1/ MOD 1 Switzerland TRAC-BD1 '12 with Boil-off and Reflooding Assessment Experiments: Model Improvements and Numerical Problems c NEPTUN: Information about Switzerland Facility 13 the Reflood Experiments Description 5012 - 5056 Preliminary ' Investigations Switzerland Facility 14

,on the Repeatability of Description Thermal-Hydraulic Data Obtained at the NEPTUN Bundle Reflooding Experiments Influence of the Wetting F.R. Germany Code 15' State of a Heated Surface Development on Heat Transfer and Pressure Loss in an Evaporator Tube Heat Transfer Processes F.R. Germany Code 16' during Intermediate and Development Large Break Loss-of-Coolant Accidents (LOCAs)

Thermal Mixing Tests in a Finland Code 17 Semiannular Downcomer With Development Intera'cting Flows From Cold Legs Assessments of CHF Sweden Code 18 Correlations Based on Development Full-Scale Rod Bundle Experiments Optimum Evaluation Japan Code 19 Improvement Report 6: Development TRAC-PF1 Code Constitutive Equations 47

l l

.i

)

i The USNRC has received, from ICAP members, a number _of' documents, other than assessment reports, that contain useful information for the code development and code assessment community. Reference 13 and Reference 14 contain information on the NEPTUN reflood experimental facility.

Reference 13 presents detailed information, needed for code modeling, on the construction (geometry and materials) of the facility. It also presents a full discussion of the NEPTUN instrumentation. Reference 14 discusses the repeatablity of NEPTUN experiments. To establish how accurate any computer simulation of the data can possibly be, it is necessary to determine experiment repeatability.

References 15, 16, and 17 discuss phenomena' pertinent to thermal-hydraulic analysis. Reference 15 describes various mechanisms and -;

regimes of heat transfer and. pressure loss in vertical and horizontal evaporator tubes. Reference 16 analyzes heat transfer behavior observed in a rod bundle during simulated loss of coolar.t accidents (LOCAs).

Reference 17 presents experimental results of thermal mixing of high pressure injection (HPI) water in a two-fifths scale model of a PWR vessel.

Reference 18 compares various CPC correlations with full scale rod bundle data; the report recommends which correlations should be used in reactor system codes on the basis of the data comparison. The report shows that the Biasi correlation, currently used in RELAPS is not adequate for properly predicting core thermal behavior. Reference 19, submitted by a prospective ICAP member, documents the coding of constitutive equations in selected models employed in TRAC-PF1/M001.

Reports other than assessment reports can also aid the USNRC in its code development and code assessment activities. For example, facility descriptions are necessary to create and verify the accuracy of code input decks. The quality of any assessment is highly dependent upon the code nodalization and code options used. Hence, detailed descriptions of facilities are needed to perform and evaluate assessment studies.

Similarly, complete details of the experimental data, including how they l were measured, must be available for quality assessment work to be 48 l

performed. The analyst must be aware of the accuracy of the data and the-nature of any anomalies in the data to properly compare the calculated values to the measured data.

Descriptions and analyses.of physical phenomena observed in experiments are valuable information for the code development effort. Only with a clear understanding cf the actual phenomena, can the code developer determine if models employed in the code are_ appropriate for properly calculating observed behavior. Reports studying the applicability of various correlations may provide the developer with information necessary to determine which correlations are appropriate, 4.6 ICAP Meetings Regular meetings of ICAP participants provide a forum for exchanging 5

technical and administrative information pertinent to the ICAP. To date, six meetings have been held. A general meeting, called to initiate the l ICAP, was held April 1985. Since then, two Program Management Group i Meetings, two Specialist Meetings, and a-Task Group Meeting have been held. The Program Management Group Meetings were held in conjunction with the USNRC annual Water Reactor Safety Information Meetings of November 1985 and November 1986. The First Specialist Meeting was held in Erlangen, Germany, in June 1986. The Second Technical Meeting was held in Winfrith, United Kingdom, in January 1987. The Task Group Meeting was held in Erlangen, Germany, in June 1986.

Complete. minutes and proceedings from each of the ICAP meetings are contained in Appendix G. A bibliograghy of the papers presented at these meetings is contained in Appendix H.

4.7 Newsletters and Workshops INEL and LANL, in support of the ICAP, publish quarterly RELAP5, TRAC-PWR, and TRAC-BWR newsletters and conduct workshops and onsite l training programs for ICAP participants as required. The following subsections discuss specifics of newsletter and workshop support of the ICAP.

49

4.7.1 Newsletters Quarterly TRAC-PWR, RELAPS, and TRAC-BWR newsletters, published by the supporting laboratories, keep ICAP users up to date on information that impacts code use. The following items are included in each newsletter: 1) descriptions of code changes such as corrections of code errors and new user conveniences;-2) code documentation changes including corrections to, additions to, and clarification of existing documentation; 3) descriptions of user reported code performance problems and the status of each' problem's resolution; 4) summaries of assessment studies indicating the code's ability to model observed phenomena; and 5) code user guidelines. The newsletters also contain l_istings of current code-related literature and information on upcoming ICAP meetings.

ICAP members are encouraged'to contribute articles and information for inclusion in any section of the newsletters. The newsletters provide a mechanism for sharing code experience and ideas among all ICAP members and the USNRC. Thus, the newsletters promote high quality assessments that provide the USNRC with a complete picture of code capabilities and shortcomings, as well as information needed for code uncertainty quantification.

The TRAC-PWR newsletter was first issued in October 1985. The RELAP5 and TRAC-BWR newsletters were first issued in July 1985. Currently, the i

limited use of the TRAC-BWR code does not warrant regular publication of the TRAC-BWR newsletter. The TRAC-BWR newsletter is published on a quarterly basis, provided that sufficient material to warrant publication is available. The format of the newsletters has evolved since 1985 to better respond to code user needs. Copies of all newsletters are included as Appendix I.

4.7.2 Workshops A workshop to train ICAP participants in using the TRAC-PWR code was held at LANL in March 1986 to inform TRAC-PFI/M001 users about specific 50

heat transfer models in the core. Although the workshop was specific in nature, a broad range ofLrudimentary skills were also covered to give the user sufficient appreciation for the depth of the code.

~In September 1985, a workshop.to train ICAP participants in using RELAP5/M002 was held at EIR, Switzerland. INEL personnel spent the five day workshop instructing 30 ICAP users in all aspects of the code.

Also in September 1985, a TRAC-BWR workshop on all aspects of the code was held at the INEL. ICAP users were among the workshop participants.

4.8 Laboratory Visits by ICAP Participants Four ICAP participants visited the supporting laboratories for training and administrative activities-consistent with the program scope.

Two participants visited the INEL, and two visited LANL.

R. Y. Yuann and K. S. Liang, of the American Institute of Taiwan, visited the INEL to receive training in using RELAP5/ MOD 2. The visit lasted from October 1986 through January 1987. The training assignment culminated in the production of an assessment report evaluating the code's ability to predict Semiscale small break LOCA data. The assessment performed by the trainees was documented and will be issued as an International Agreement NUREG/IA.

G. Yadigaroglu of Switzerland, visited LA.NL during September 1985. He requested a stripped version of TRAC-PWR to assess the heat transfer package and investigate other solution methods. The visit was concluded with an agreement in principle to provide a stripped version of the code to l the Swiss Technical Institute.

During March 1986, M. Andreant, also of Switzerland, attended a LANL workshop on heat transfer models in TRAC-PF1/ MOD 1. The scope of the'

)

workshop is described in Section 4.7.2.

51

5. REGULATORY SIGNIFICANCE AND CONCLUSIONS The underlying objective of conducting research on the thermal-hydraulic behavior of licht water reactors is the prevention of' severe accidents. Thermal-hydraulic research provides the technical basis. l I

for assuring that power reactors are designed and operated so as to minimize the risk of occurrence of a severe accident.

The products of the research are validated computer codes, experimental data, and thermal-hydraulic analyses. These products are used in a variety of ways to give technical defensibility to regulatory policy and decisions. They provide the regulatory staff with an independent analytical capability for auditing licensee analyses, as well _as for performing studies on generic issues.

The products of the ICAP are experimental data and thermal-hydraulic code assessments which are used to validate the codes for their end use in the regulatory process. Work accomplished to date has provided international feedback of code errors and deficiencies for the TRAC-PWR, TRAC-BWR, and RELAPS frozen code versions. These efforts have enhanced the code documentation, code user guidelines, and improved the code capability to simulate plant transients. These efforts will continue in the future-years of the ICAP.

The use of best. estimate codes is now being introduced into the regulatory process (e.g, the current USNRC effort to modify the ECCS rule). From the USNRC perspective, this move imposes a new requirement, that of quantifying the accuracy of the codes. The accuracy quantification, in turn, introduces new requirements with respect to the form and number of code assessments needed. The necessary uniformity and number _of assessments can be accomplished most cost effectively by the sharing of costs and products throughout the international community. That role is being fulfilled by the ICAP. The structure of ICAP provides additional benefits through the increased technical quality and generality of the code improvements which are developed from the ICAP code assessment data base.

52

1 l

1

6. REFERENCES l 1. Bessette, D.E., Guidelines and Procedures for the International Thermal-l Hydraulic Code Assessment and Applications Program, NUREG-1271, to be published.
2. Neumann, U., Assessment of TRAC-PD2 Using Super Cannon and HDR Experimental Data, NUREG/IA-0001, 1986.
3. Coddington, P., Analysis of LOFT Experiment LP-LB-1 Using TRAC-PF1/ MOD 1, AEEW-R2039, WINFRITH UKAEE,1986.*
4. Guntay, S., RELAP5/M002 Assessment: OECD-LOFT Small Break Experiment LP-SB-3, EIR, 1986.*
5. Eriksson, J., Assessment of RELAP5/M002, Cycle 36, Against FIX-II Split Break Experiment No. 3027, NUREG/IA-0005, STUDSVIK/NR-85/99, 1986.
6. Stubbe, E.J., Assessment Study of RELAP5 Cycle 36.01-Based on the DOEL-2 Steam Generator Tube Rupture Incident of June 1979, NUREG/IA-0008, TE, NU/EST/ml, 1986.
7. Sjoberg, A. and Caraher, D., Assessment of RELAP5/M002, Against Twenty-Five Dryout Experiments Conducted at the Royal Institute of -

Technology, NUREG/IA-0009, STUDSVIK/NP-86/66, 1986.

8. Analytis, G. Th., Implementation and Assessment of a New Bubbly / Slug Flow Interfacial Friction Correlation in RELAP5/M002/36.02, TM-32-86-10, EIR, 1986.*
9. Rosdahl, 0. and Caraher, D., Assessment of RELAP5/ MOD 2 Against Critical Flow Data from MARVIKEN Tests JIT and CFT 21, NUREG/IA-0007, STUDSVIK/NP-86/99, 1986.
10. Rosdahl, 0. and Caraher, D., Assessment of RELAP5/M002 Against MARVIKEN Jet Impingement Test 11 Level Swell, NUREG/IA-0006, STUDSVIK/NP-86/97, 1986.
11. Moeyaert, P. and Stubbe, E., Assessment of RELAP5/M002 Cycle 36.04 Based on Spray Start-up Test for DOEL-4, Tractebel, Brussels, Belgium, October 1986.*
12. Analytis, G. Th., Assessment of TRAC-BD1/ MOD 1 with Boil-Off and Reflooding Experiments: Model Improvements and Numerical Problems, EIR, 1986.*
13. Frei, E. and Stierli, F. , NEPTUN: Information about the Reflood Experiments 5012-5056, TM-24-83-17, EIR, 1984.*
14. Guntay, S. , Preliminary Investigations on the Repeatability of Thermal-Hydraulic Data Obtained at the NEPTUN Bundle Reflooding Experiment, TM-32-83-19, EIR, 1983.*
  • See ICAP Contact List, pages G-31 thru-G-38, for contact and organization.

53

15. .Kohler, W. and Hein, D., Influence of the Wetting . State of _ a Heated Surface on Heat Transfer and Pressure Loss insan Evaporator Tube, NUREG/IA-0003, 1986.

- 16. Vojtek,~I., Heat Transfer Processes During Intermediate an'd Large Break Loss-Of-Coolant Accidents (LOCA), NUREG/IA-0002,1986.

[ - 17. Tuomisto, H. and Mustonen, P., Thermal Mixing Tests in a Semiannular Downcomer With Interacting Flows From Cold Legs, NUREG/IA-0004, October 1986.

- 18. Sardh, L. and Becker, K. M. , Assessments:of CHF Correlations Based on ,

H Full-Scale Rod Bundle . Experiments, KTH-NEL-36,1986.*

19. - Akimoto, T., et. al . , Optimum Evaluation Code Improvement REPORT-6-TRAC-PF1 Code Constitutive Equations, NRC 1710, 1985.*

4 4

4 4

  • See ICAP Contact List, pages G-31 thru G-38, for contact and organization.

54

APPENDIX A ICAP DESIGNATED CONTACTS The International Code Assessment and Applications Program (ICAP) is an. international cooperative reactor safety research program organized by tre USNRC in 1985 to provide an independent assessment of the USNRC's

. thermal-hydraulic codes developed for analysis of plant response to the postulated-transients and loss-of-coolant accidents in light water reactors. Countries possessing nuclear and/or experimental test facilities I

are invited to participate-in the ICAP by signing a bilateral agreement with the USNRC. Currently, bilateral agreements have been signed with Belgium, Finland, France, Germany, Italy, Korea, Spain, Sweden, Switzerland, Taiwan, and the United Kingdom. Proposed agreements are under negotiation with Austria, the European Joint Research Center at ISPRA,

~

Japan, Mexico, the Netherlands, and Yugoslavia.

This Appendix contains a list of the names and addresses of each participant's designated contact (s) in the ICAP.

1 i

l l

A-1 l

~ _ . - - - . . - . _ . - .

Page No. 1 03/13/87 ,

ICAP CONTACTS COUNTkY NAME STATUS ADDRESS PHONE TELIX/ CODE ICAP IRIS TELECDPY (3M) INTEREST pep 5ER R5G VERIFICATION DR. IIIE STUBBE 011-32-2-234-5208 64860 TRACTB REIAP5 YES NO 4 BE!f.IIM 1 NUCLEAR DEPARIM NT TRACTEBEL RUE DE IA SCIENCE, 31 B-1040 BRUSSELS, BEICItM FINIAND MR. H. HotNSTROM 1 NUC12AR D GINE DING LABORATORY 011-358-0-64-8931 122972 VTIMASF REIAPS YES YES-VALTION TEKNIELINEN TUTKIMUSKESKUS P. O. BOX 169 SF-00181 HELSINKI FINIAND 71tJUICE DR. MIQIEL REDCREUX & INSTITUT DE PROTECTION 011-4654-7158 RELAPS YES NO ET DE SURETE NUCLEAIRE TRAC-PWR CEA COBRA-TF f

N CD/FAR, B.P. No. 6 92260 PONID AY-AUX-ROSES FRANCE I PRANCE tet. B. ADROGUDt 3 CEA (CADARAOIE) DERS/SEAREL RELAP5 NO NO 3 08 ST PAUL LEZ 11 TAO-PWR DURANCE CEDEX COBRA-TF PRANCE FRANCE let. R. L. P00 LARD 3 INSTITUTE DE PROTECTION REIAP5 NO NO ET DE SURETE NUCLEAIRE T1 TAC-PWR CEA (DBitA-TF CIN/FAR, B.P. NO. 6 92260 PONTENAY-AUX-ROSES PRANCE P1tANCE DR. B. SPIleLut 3 Q3 MISSION OF ATOMIC DERGY 011-33-76-88-4687 ItEIAP5 YES 10 0 CENG/SE11f TRAC-PWit 85X - 38041 GRElIDBLE CEDEX CDBitA-TF PRAIICE GEIWWlY PROF. E. F. HIQWN 1 NYNFT PUR REAIGUltSIGIERIEIT MtA 011-49-89-3200-405 5215110 MSte ItEIAP5 YES IID PMWN2r m TItAC-IIEt D-8046 GARCHING TItAC-PWR

Pare No. 2 03/13/87 ICAP CONIACTS COUNTRY MAME STATUS ADDRESS PHONE TELEK/ CODE ICAP IRIS TRrsmPY (3M) INTEREST PE!)SER RBG VDtIFICATION FEDERAL REPUBLIC OF GERMANY GDtMANY MR. F1 TANK WINKLER 1 KRAFIWDtK UNION AKTIENCESEILSQiAFT (KW) 011-49-9131-18-2156 629290 KW REIAP5 YES NO HAP 9tERBAQERSTRASSE 12+14 11 TAC-BWR

[ POSTFACH 3220 TRAC-PWR D-8520 DtIANGEN FEDDtAL REPUBLIC OF GERMANY GDtMANY MR. GUNTER JACDBS 3 m KARLSRUIE Q1BH 011-49-724-82-2435 782%84 ZKFKD REIAP5 YES YES INSTITUT FUR NEUIRONENPHYSIK UND TRAC-BWR REAKIORTECHNIK POSTFAQ{ 3%0 D-7500 KARLSRUHE 1 FEDDtAL REPUBLIC OF GDtMANY GERMANY MR. I. V0JTDC 3 GESELLSQlATI PUR REAKIORSIQIDtHEIT Mt4 REIAP5 NO NO 3 FURSGIUNGSGELANDE 11 TAC-BWR D-8046 GARQIING TitAC-PF1 FEDDtAL REPUBLIC OF GUtMANY GDtMANY DR. KLAUS WOIJDtT 3 GESELLSCHAFT PUR REAKTORSICHDtHEIT Mb4 011-49-89-3200-4405 5215110 GRSPO REIAPS YES 10 0 FORSQfUNGSCELANDE TItAC-BWR D-8046 GAltQtING TRAC-PF1 FEDUtAL REPUBLIC OF GDtMANY GDtMANY DR. RINNINSLAND 5 PROJDCI NUCLEAR SIQIDtHEIT 011-49-7244-82-6484 7826484 ZKFKD REIAP5 YES ND KBtNFORSQEUNGSZENDtUM KARLSRtBIE QSH 11 TAC-BWR POSTFAQt 3640 D-7500 KAltLSRtalE 1 FIDBtAL REPUBLIC OF GEltMANY ITALY PGt. G. SAPONARO 1 DEEA-DISP 011-39-6-8526-2817 612167 EIIEURI REIAP5 YES NO SET!DitE SVILUPPI E RICBtOE TRAC-BWit VIA VITALIAII0 BitANCATI, 48 TItAC-PWR 001 % ROMA ITALY i

_ _ . . ._ , . - _ - - . . - -. - ~ . .. _ .- --

Page No. 3 03/13/87 ICAP Q)lfIACTS QXJNIRY NAME STATUS ADDRESS PHONE TELEX / CODE ICAP, . IRIS TFT FmPY (3M) INTEREST IWSER REG VERIFICATION ITALY Lit. V. MANCUSD 3 ENEA-TUtM 011-39-6-3048-4804 6132% ENEACAI REIAPS YES 10 0 CRE CASACCIA S. P. AM GUILIARESE 301 C. P. NO. 2400 00100 ItaMA ITALY ITALY DR. PA01D MARSILI 3 DfEA-DISP 011-39-6-282-5028 612167 DIEURI RELAP5 YES 10 0 VIA VITALIANO BRANCATI, 48 TRAC-BWR 00144 ROMA TRAC-PWR >

ITALY ITALY DR. E. NEGRDfrI 3 ENEA-TERM 011-39-6-3048-4265 613296 ENEACAI RELAP5 YES NO CRE CASACEIA S. P. ANGUI11ARESE 301 ROMA - 00060 ITALY y ITALY DR. GIORGIO PALAZZI 3 DfEA-TERM 011-39-6-69 & 3039 613296 ENEACAI REIAPS YES NO g CRE CASACCIA S. P. ANGUILIARESE 301 C. P. NO. 2400 00100 R0pm ITALY JAPAN DR. KAZUD SATO I DEPT. OF REACTOR SAFETY RESEAkt3I 011-81-29-282-5028 03632340 JIOKAIJ REIAP5 NO NO JAPAN ATOMIC DIEltGY RESEARG INSTITUIE TRAC-BWR TOKAIR-tENtA, NAKA-QRI, IBARAKI-KDE 31911 TRAC-PWR JAPAN COBRA-TF JAPAN DR. H. AKIMDIO 3 DEPT. OF REACTOR SAFETY RESEAltG 011-81-29-282-5275 03632340 JTOKAIJ REIAP5 NO 11 0 JAPAN ATOMIC ENERGY ItESEAltOf INSTITUIE TRAC-BWit TOKAI-PENTA, IIAKA-GUlt, TRAC-PWR IBAltAKI-KEN 31911 COBRA-TF JAPAll JAPAN DR. YOSHIO IGJRAO 3 DEPT. OF REACIOR SAFETY RESEARG 011-81-29-282-5272 03632340 JTOKAIJ REIAPS 11 0 11 0 JAPAll ATOMIC E NWABfM IIISTITUTE TitAC-BWR 70EAI-ptJRA, MAKA 'ANI, IBAltAKI-MEll 31911 TRAC-PWR

,~ ~ __. _ _ .___ _ _ . . . _ _ _ _ _

Page No. 4 03/13/87 ICAP CONIACTS CX!NTRY NAME STAIUS ADDRESS PHONE TELEX / CODE ICAP IRIS TELECDPY (3M) INTEREST lepeER It3G VERIFICATION JAPAN COBRA-TF KOREA DR. SANG HOON IJI 1 KOREA ADVANC D ENERGY RESEARQI INSTITUTE 011-82-042-820-2601 KAERI K45553 REIAP5 YES NO NUCLEAR SAFETY CDfrDt TRAC-PWR P. O. BOX 7 COBRA-TF DAEUK - DANJI, Q100NG - MAN, 300-31 KOREA KOREA  ?* B. D. CHUNG 3 KOREA ADVANC D ENEWGY RESEARQI INSTITtTTE REIAP5 ' 11 0 NO NUCLEAR SAFETY CDfrDt TRAC-PWR P. O. BOX 7 COBRA-TF DAEDUK - DANJI. OIDONG - MAM 300-31 KOREA KOREA DR. HHO JUNG KIM 3 KOREA ADVANC E ENERGY RESEARCH INSTITUTE 011-82-042-820-2613 KADtl K45553 REIAP5 YES NO NUCLEAR SAFETY CENTER TRAC-PWR P. O. BOX 7 COBRA-TF

')AEDUK - DANJI, OIDONG - NAM, 300-31 y KOREA e

u, f01.TINATIONAL DR. HDtBERT STADTKE 1 (%)p9 FISSION OF THE EUROPEAN CDB90NITIES 011-39-332-789986 43380042 EURAT(M REIAP5 YES NO JOINI RESEAROI CENTER ISPRA ESTABLISIDENT l I-21020 ISPRA (VARESE)

ITALY MULTINATIONAL DR. W. RIEBOLD 3 00p9tISSION OF THE DNt0 PEAN QP90NITIES 011-39-3-3278-9898 43380042 EURAT(M REIAP5 YES NO i JOINI RESEARQI CENTRE ISPRA ESTABLIS1953ff C.P. NO. 1 I-21020 ISPRA (VAR N E)

ITALY i

SPAIN DR. JOSE DECAR14S 1 CONSEJO DE SEGURIDAD NUCIIAR 011-34-1-456-1812 45869 CSIDE REIAP5 YES 11 0 SOR ANGEIA DE IA CitUZ 3 EXT. 384 TRAC-BR HADRID 16 TitAC-PWR SPAIN SPAIN DR. JOSE M. IZQUIDtDO 1 CollSEJO DE SEGURIDAD IIUCI2AR 011-34-1-456-1812 49051 CSIBE ItEIAP5 YES YES i

-_ . _ . _ . . -. _ _ _ . - _ _ . _ , _ . _ _ _ . ~_. __ , ,_.

Page No. 5 i 03/13/87 ICAP CONIACTS COUNIRY NAfE STATUS ADDRESS PHONE TE!2X/ CODE ICAP IRIS TELE 00PY (3M) INTEREST lEISER REG VERIPICATION SOR ANGEIA DE LA CRUZ 3 EXI. 358 TRAC-PWR MADRID 16 SPAIN SPAIN DR. JOSE PUGA 3 ENUSA 011-34-1-254-1408 43042 INtAME REIAPS YES 'YES 4 C/ SANTIAGO RUSINOL, 12 TRAC-PWR E-28040 MADRID SPAIN 1

DR. 00DBJORN SANDERVAG 1 STUDSVIE IISItGITII0fIE AB 011-46-15-52-1868 64013 STUDSS REIAP5 YES YES SWEDEN S-61182 NYEOPING TItAC-PWR SWEDD SWITZERIJWO DR. O. IERCIE3t 1 SAFEIY DIVISION 011-41-56-992688 827417 EIRG REIAP5 YES NO EIDGDOSSIGIES IIISTITUI PUR TRAC-BWR P M E (EIR) 3D G-530314NtENLIIIGElf b SWITZERIAND SWITZIltIAND DR. S. N. AESAN 3 SAFETY DIVISI0lt 011-41-56-992710 82417 ElitG REIAP5 YES NO EI-IOES IIISTITUI PUR 41-56-902327 TItAC-PWR ItEAETURPoltSORNIG (EIR)

G-5303 IdURENLIIIGEls SWITZERIAIS SWITZIltIJUB DR. G. TH. ANALYTIS 3 EIR ltEIAP5 11 0 11 0 5303 samnus TIIGEN TitAC-stet suttyser m SWITZERIAle DIt. SALIH GUNIAY 3 SAFETY DIVISION 011-41-56-992-684 53714 EIROI REIAP5 YES 11 0 EIDGDOSSIMS IIISTITUT PtR TRAC-NR

, REAIGURP m

  • N E (IIR)

G -5303 N YlIGElf 4 SWIT7 m m

Page No. 6 03/13/87 ICAP CONTACTS CElffRY NAME STATUS ADDRESS PHONE TELEX / CODE ICAP IRIS TII.ECOPY (3M) INTEREST PO WER RBG VIRIPICATION TAIWAN DR. YI-G'N QEN 1 HEAD, REACIOR TEQWOIDGY DIVISION 011-886-2-381-4014 34154 CAEC ItEIAP5 YES NO INSTITUTE OF NUCLEAR BfBIGY RESEARQt TRAC-BWR P. O. BOX NO. 3-3 IL'NG-TAN TAIWAN TAIWAN MR. J. E. HSIUE 3 ATOMIC POWER DEMIDPPENT 011-886-2-3 % -7124 NONE YES NO TAIWAN POWDt COMPANY 242 ROOSEVELT ROAD, 3RD SECTION TAIPEI TAIWAN TAIWAN DR. LIH-YIH LIA0 3 INSTITUTE OF MUCIJ.AR ENERGY RESEARQI 011-886-2-381-4014 34154 CAEC REIAP5 YES NO P. O. BOX NO. 3-3 IlieG-TAN p TAIWAN N

THE NETHERIANDS DR. J. SPEIIMAN 1 DfDtGIEDNDUt20Dt CDrIltUPt NEDBt!AND 011-599-31-2246-4949 57211 REACP IIL REIAP5 YES NO WESTDtDUINWEG 3 POST BOK 1 NL-1755 2G PETTEN THE NETHDtIARIDS THE NETHERIANDS DR. P. M. SIUDP 3 ENERGIE0IentZODC CBfIRtM IIEDet!AIS REIAPS NO NO WESIIItDUINWBG 3 POST BOK 1 NL-1755 2G PETTEN THE NED8Enf AM THE NEIMDtIANDS DR. L. WINTUtS 3 NE11ERIANDS DIEEGY ItESEAltQt POUlmATION, 011-599-31-2-246-4462 57211 REACPIE. REIAPS YES YES ICN P. O. BOK 1 1755 ZG PETTEN THE N AMht

.1

I ..e ~.

03/13/87 ICAP (XMfTACIS STATUS ADDItESS PHONE TILEK/ @DE ICAP IRIS COUNTRY NAME TELECDPY (3M) INTE3 TEST IEDSER RBG VERIFICATION 011-44 305-63111 41231 AT0fGdHG REIAPS YES NO UNITED KINGDOM Pet. JESSE P1IL 1 UNITED KINGDOM ATOMIC DE3tGY AUDIORITY It00M NO.166, BUIISING A32 EXT. 3451 TRAC-PWR AIUMIC INE31GY ESTABLISIGENT WINFRITH DORGESTE3t DDRSET UI2 80H UNITED KINGD(St 3 CINUtAL ELECTRICITY GEIE3tATING BOAltD 011-44-452-65-2198 43501 CEGBCDG REIAP5 YES NO UNITED KINGDOM DR. KEITH ARDitON GUEItATIOlt DEVIIDPPENT AND CMtST. DIVISION TRAC-PWR BARNETI WAY BARNWOOD, GIDUCESTE3t GIA 7R5 UNITED KINGDOM AE3tE HAltWELL REIAP5 NO NO UNITED KLM IGt. P. BIACK 3 DIDOUT TRAC-PWR OXON OX11 ORA UNITED KINGDOM 3 UNITED KINGDOM AIUMIC ENEltGY AUTHORITY 011-44-305-63111 41231 ATOMWHG REIAP5 YES NO y UNITED KINCDOM Pet. IAN BRITTAIN h ATOMIC ENEltCY ESTABLISI9ENT, WINFRITH DORCHESTot DORSEr DT2 8Du EXT. 2039 TRAC-PWR UNITED KINGDOM 3 UNITED KIIIGDOM ATOMIC ENEltGY AUTHORITY 011-44-305-63111 41231 ATOPRdHG REIAPS YES YES UNITED KINCDOM DR. WALIACE M. BRYCE ATOMIC ENEItGY ESTABLISIDE3ff WINFRITH EXT. 3249 TRAC-PWR DOItGESTER DORSET Iff2 8DH UNITED KINGDOM UNITED KINGDOM DR. P. CODDINGTON 3 UNITED KIIIGDOM ATOMIC IIENGY AUDEEtITY REIAP5 11 0 10 0 ATCptIC IIIENGY ESTABLISIDElff, WI!WRITH TRAC-PWR DoltOESTElt DaltSET DT2 8DH UltITED KI%BOM 3 CBtL 0372-374488 ET22 7SE ItEIAP5 NO NO UNITED KINGDOM IEt. M. CONEY KELVIN AVE 31UE -TRAC-PWR LEATIEIIIEAD UNITED KINGDOM 3 CINTItAL EIACUtICITY GBE3tATING anaan cII-44-1-634-7630 REIAPS YES YES UIIITED KIIIGDOM IGt. GItAHAM PitAIICE GMPUTIIIG CElintE TRAC-PWIt

rage no. e 03/13/87 ICAF CONTACTS COLWIRY MAME STATUS ADDRESS PHONE TELEX / CODE ICAP IRIS TELE (X)PY (3M) INTEREST PO WER REG VERIFICATION 85 PARK STREET IDMDON, S319DY UNITED KINGD(M G"ITED KINCDCM Pet. PETER C. HALL 3 C DIR*L ELECIRICITY GENERATING BOARD 011-44-45-265-2529 43501 - REIAPS YES YES CDERATION DEVEIDPPE3fT AND (IINST. DIVISION TRAC-PWR BARNETT WAY BARNWOOD, GIDUCESTDt CIA 7R5 UNITED KINGDOM (U TED KINCDOM pet. NICK NEWMAN 3 CENTRAL ELECTRICITY GENERATING BOMtD 011-44-372-37-4488 REIAPS YES YES KELVIN AVDUE, LEATHE3tHEAD TRAC-PWR SURREY, KT22 7SE UNITED KINGDOM USA P5t. PHILIP TING 1 UNITED STATES NUCLEAR REGUIATORY 301-443-7920 908142 NRC-BIG-WSH REIAPS YES YES CDPMISSION NO VERIFICATION NO. TRAC-BWR e

NIQIOLSON IANE BUI1 DING, M/S NL-005 TRAC-PWit 5650 NIQa0LSon gANE COBRA-TF ROCKVIILE, le 20814 USA USA DR. IAWRDCE D. BUXTON 2 SANDIA NATIONAL 1ABORATORIES 844-6537 FIS REIAP5 YES YES DIVISION 6444 505-844-6537 TRAC-BWit P. O. BOK 5800 TitAC-PWIt AIJUQUERQUE, IWW PEKI(X) 87183 USA USA PGt. ROBERT C. HANSON 2 EG6G IDAHO, INC., TSB 583-9701 PTS 9109775915 USDOEID REIAP5 YES NO P. O. BOK 1625 208-526-9701 583-9591 PTS TRAC-BIR IDAHO FAILS, ID 83415 583-9500 PTM TRAC-FIR USA CONA-TF USA PGt. RICK JDES 2 IDS AIAMOS NATIONAL Tanneam 363 2021 PTS 843-3670 PTS REIAPS YES YES

_ . _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . ._ . _ _ . . . . _ . _ _ _ =

Page No. 9 C3/13/87 ICAP COIrIACIS TE!2X/ CODE ICAP IRIS STATUS ADDRESS FHONE c)tnr11tY NAPE penER ItBG TE12tDPY (3M) INTEREST VERIFICATION 505-M 7-2021 843-5113 FTS TRAC-BISt MS K-555 TRAC-FWR IDS AIAMOS, IEW IEXIOD 87545 USA 843-3670 FTS ItE1APS YES 11 0 2 EDS ALAMOS IIATIONAL r aanma1tIRY 843-3113 FTS USA DR. TIIAD ENIQtT J 505-M7-3113 843-5113 FTS TRAC-BWR P. O. 30K IM3 IDS AIAICS, IEW IEXICO 87%5 TRAC-FWit USA 666-2475 FTS 666-3000 FIS REIAP5 YES 10 0 USA DR. U. S. ROHATGI 2 BR00ERAVDE NATIONAL IABORAIDRY

' ASSOCIATED UllIVERSITIES. IIIC. 516-282-2123 666-2%7 FTS TitAC-BWR TRAC-FWR UPIUlt, IDIIG ISIJUID, IEEW YORK 11973 USA 9109775915 USDOEID ItEIAPS YES YES Nt. Celti E. WILS0It 2 EGEC IDAIIO. IIIC., TSB 583-9511 FIS USA TRAC-gult P. O. BOK 1625 208-526-9511 583-9591 FTS 583-9500 FTS IDAHO FALLS, ID 83415 TitAC-FWR I

COBRA-TF 2> USA e

H O

843-2023 FIS ItEIAP5 YES NO l USA DR. BRENT BOYACK 3 005 AIJIWS NATIONAL rAanmaTORIES 505-M7-2023 TitAC-BWR F. O. BOK 1M3 IDS AIAMOS, IEW IEXIOD 87%5 TRAC-FWit USA 9109775915 USDOEID REIAPS YES YES MS. JAII L. J N 3 3GEG IDAIIO, IIIC., TSB 583-9762 FTS USA 208-526-0061 583-9591 FTS TitAC-BIE F.O. 3GE 1625 IDAND FAILS, ID 83415 583-9500 FTS TRAC-FWIt COBItA-TF I USA 9109775915 USDOEID REIAF5 YES YES USA PR. GARY U. JOISISEE 3 3GEG IDAIW, IIIC., TSS 583-98% FTS 208-526-98 % 583-9591 FTS TitAC-BWR F. O. 30K 1625 1D4110 falls, ID 83415 583-9500 FTS TRAC-Flat

  1. i 1

8 M-1068 FTS REIAP5 YES YES USA DIt. LIlbtBITRA IIADIA WE!YK 3 SAISIA IIATICAAL FAnnmaTORIES 505-844-00 M 8M-8917 FTS TRAC-Ftm DIVISICII M64

- -- .- - .- . . _ _ - _ - -- _-- - _- -. - - . - . . - - . ~ - . - ~ - - - - - - - - - - - - --

rage leo. 10 03/13/87 ICAP CONTACTS COLWTRY NAME STATUS ADDRESS PHONE TE12X/ CODE ICAP IRIS TEIE00PY (3M) INIntEST IWWER REG VDt1FICATION P. O. BOX 5800 ALBUQUDOUE, NEW IEXIO) 87185 USA USA DR. S. MIQiAEL MODRO 3 EG6G IDAHO, INC., TSA 583-9402 FTS 9109775915 USDOEID ltEIAP5 YES YES P. O. BOX 1625 208-526-9402 583-9591 FTS IDAHO FALIS, ID 83415 583-9500 FIS USA USA DR. PUAT ODAR 3 UNITED STATES NUCLEAR REGUIAIORY 443-7836 FTS 443-7836 FTS REIAPS YES YES COPMISSION 301-44'. 7902 NO VERIFICATION NO. TRAC-Bem NIQt0LSON LANE BUIIDING, M/S NL-007 TitAC-PWR 5650 NIQt0LSON IANE ROCEVillE, PO 20814 USA p USA DR. JEAN FIDtRE SURSOCE 3 D.ECTRIC POWER RESEARQI INSTITUIE ItELAP5 NO 10 0 x 3412 HILLVIEW AVENUE TRAC-BWR P. O. BOX 10412 TRAC-PWR PAID ALTO, CALIFORNIA 94304 USA USA M2. PAUL BODEtDt" 4 UNITE STATES 14UC12AR REGUIAICItY 634-3267 FTS ItEIAPS YES NO 030tISS10N 202-634-3267 TRAC-BWR ADVISORY C0petITTEE 018 ItEACIOR SAFBCUAltDS

+ H STitET BUTIDING, M/S H-1016 WASHINGIDIt, D.C. 20555 USA USA DR. IVAN CATTON 4 LMIVERSITY OF CALIF 0ItNIA 213-825-5320 ItEIAP5 NO NO 5732 BOELTER HALL TItAC-BWR l IDS ANGE12S, CA 91364 USA USA DR. RDPSIEY DUFFEY 4 D'2CTRIC POWBt FNmM IIISTITUIE 415-855-2097 415-855-2954 FTS REIAPS No HD

?A12 HILLVIEW AVBIUE 415-855-2717 FTS TRAC-BWit P. O. BOK 10412 TitAC-FWit

Page No. 11 03/13/87 ICAP CONIACIS G UlfIRY NAME STATUS ADDRESS PHONE TELEX / (XOE ICAP IRIS TFf M PY (3M) INTUtEST MDIBDt REG VERIFICATION PAID ALIO, CALIfDRNIA 94304 USA USA DR. M. WAYNE LODGES 4 L31TED STATES MUCLEAR RDAJIATORY 492-7403 FIS 492-1617 FIS RELAP5 YES NO (XPetISSION 301-492-7483 492-7371 FIS TRAC-BWR PHILLIPS BUIIDING M/S P-1022 7920 NORPUID AVDIUE BEIHESDA, MD 20814 USA USA DR. YI-HSIUNG HSII 4 UNITD STATES NUCLEAR REGUIATORY 492-7952 FIS 492-7617 FIS REIAP5 YES NO CoretISSION 301-492-7952 492-7371 FTS FHILLIPS BUIIDING, M/S 544 7920 NORFULK AVENUE BETHESDA, MD 20814 USA USA DR. ROBERT C. JONES 4 UNITED STATES NUCIIAR REXX?LATORY 492-8004 FTS 492-7617 FIS RE!APS YES YES 301-492-8004 492-7371 FIS 7 CoretISSION PHILLIPS BUIIDING M/S P-1132 TRAC-BWR q 7920 NORFO!K AVDIUE BETHESDA, MD 20814 USA 4 UNITE STATES NUCLEAR REGULATURY 492-8997 FIS 492-7617 FTS REIAPS YES NO USA MR. NORMAN IAUBEN (XPetISSION 301-492-8997 492-7371 FTS TRAC-BWR FHILLIPS BUIIDING, M/S P-11324 7920 NORFULK AVENUE BETHESDA, MD 20814 USA BAB00CE AND WIIDOX m. 804-385-3705 804-385-3663 REIAPS YES YES USA DR. ROBERT J. SOKHA133t 4 P. O. BOX 10935 671-1060 FTS LYNOIBURG, VIRC!NIA 24506-0935 USA 4 UNITED STATES NUCLEAR REGULATORY 492-7303 FIS 492-7617 FTS REIAP5 YES NO USA DR. BRIAN SHBON

_ _ - - -- - . .- . _. . _ . . .. _ m. , _

! .g. e i o3/13/a:

I ICAP GINTACTS i

COUNTRY IEAPE STATUS ADDRESS PHONE TE2JX/ CODE ICAP IRIS TII2CDPY (3M) INTEREST penER R3G VERIPICATION GretISSION 301-492-7303 492-7371 PTS TRAC-BIR PHIllIPS BUTISING, M/S P-1132 TRAC-PWR 7920 IIORP0tX AVEIEUE BETIESDA, le 20814 USA USA DR. S. SLM 4 UNITED STATES NUQ2AR REGRATUItY 301-492-8158 ,492-7617 FIS RELAP5 YES IID GWetISSION 492-7371 FTS PHIILIPS BUILDING, M/S P-314 7920 NORPOL2 AVEHLT.

BEINESDA, le 20814 O

I l

USA / GERMANY PGt. SAM A. NAFP 3 KRAFIVERK LMION AKTIENGESE11SOIAPT (EWU) 011-49-9131-18-3905 629290 EW REIAP5 YES ND l C/0 57142 BISG. 18 011-49-9131-7018 ImAC-Inst

! hap 0ERBAOER STR.12+14 TRAC-PWR D-8520 ERIAIGEN FIDERAL REPUBLIC OF GERMANY b

Las l

Status

Description:

l 1. Primary contact for a specific research agramment between the IEC and an ICAP emeber country, i 2. Primary contact for the NRC or one of its contractors.

3. Ingaged in work supportisig the ICAP.
4. Indivihaels interested in the ICAP, generally at the behest of IRC-RES.
5. Primary contact for a specific organization within a country.

APPENDIX B TYPICAL INTERNATIONAL AGREEMENT The forms and extent of cooperation between International Code Assessment and Applications Program (ICAP) members and the USNRC are detailed in bilateral agreements. Bilateral agreements exist between the i USNRC and each ICAP member. The agreements specify the responsibility of l

l each party within the ICAP. A sample agreement is presented in this appendix.

f 1

1 i

l l

)

5

.I 4

i i

l J

t i

i .

I i

l l

B-1 ,

q l

l l

1 j

AGREEMENT l

ON THERMAL-HYDRAULIC RESEARCH j BETWEEN l THE UNITED STATES NUCLEAR REGULATORY COMMISSION AND THE Considering that the United States Nuclear Regulatory Comission, hereinafter referred to as USNRC, and the , hereinafter referred to as  :

1. Have a mutual interest in cooperation in the field of thermal-hydraulic research with the objective of improving and thus ensuring the safety of reactors on an international basis;
2. Recognize a need to equitably share both the resources resulting from this research and the effort required to develop those resources; And considering that:
3. The USNRC and are, respectively, an authorized participant under and a successor to an arrangement between the United States Nuclear Regulatory Comission and the of for the Exchange of Technical Information and Cooperation in the Regulatory and Safety Research Matters dated .

They ? ave therefore AGREED as follows:

B-2

2 ARTICLE I - PROGRAM COOPERATION The USNRC and the in accordance with the provisions of this Agreement and subject to applicable laws and regulations in force in their respective Countries, will join together for cooperative research in thermal-hydraulic programs sponsored by the USNRC as well as those sponsored by the .

The cooperative progran for a five-year period will include thermal-hydraulic code development and assessment studies, technical information exchange, and training and user assistance for personnel.

The specific objective of the program is to assess the USNRC-supplied thermal-hydraulic computer codes. Information generated by the will serve as a means of assessing these codes.

ARTICLE II - FORMS OF COOPERATION Cooperation between the parties may take the following forms:

A. Exchange of information in the form of technical reports, experimental data, correspondence, newsletters, visits, joint meetings, and such other means as the parties agree.

B. Temporary assignment of personnel of one party or of its contractors to laboratories or facilities owned by the other party or in which it spon-sors research; each assignment shall be considered on a case-by-case basis and shall be the subject of a separate attachment-of-staff agreement between the parties.

C. Execution of joint programs and projects, including those involving a division of activities between the parties; each joint program and project shall be considered on a case-by-case basis and shall be the subject of a separate agreement between parties.

B-3

3 D. Use by one party of facilities that are owred by the other party or in which research is being sponsored by the other party; such use of facilities shall be the subject of separate agreements between the parties and may be subject to commercial terms and conditions.

E. Any other form agreed between the parties.

If either party wishes to visit, assign personnel, or use the facilities cwned or operated by entities other than the parties to this Agreement, the parties recognize that approval of such entities will in general be required regarding terms upon which such visit, assignment, or use shall be made.

ARTICLE III - SCOPE OF AGREEMENT A. USNRC Scope of Responsibility i The USNRC shall provide over the duration of this Agreement the following I

specified goods and services related to nuclear reactor thermal-hydraulic l research:

l

1. Thermal-Hydraulic Code - RELAP5/M002 and TRAC PF1/M001 will be provided in source form on machine readable media. Complete available documentation consisting of a code manual, documentation I

of models and significant algorithmic aspects of the implementation,

!. and user guidelines will be included.

2. Code Know-How - Transfer of code know-how will be provided through workshop meetings that will be held in the countries participating in the International Code Assessment and Applications Program (ICAP).
representatives will be invited to attend and participate in these meetings. ICAP meetings may be held periodically outside of the USA.

B-4

4

3. Code Maintenance and Support - Code maintenance and support will be provided through U.S. laboratories which will gather information on l code usage and on experience from all users of the codes in the U.S.

and abroad. This information will be analyzed, evaluated, synthe-sized, and published as newsletters. These newsletters will be l provided to the . The USNRC will supply updated versions of the code resulting from the International Code Assessment and Applications Program. Bilateral meeting may be held as necessary to discuss details of model improvements made by the USNRC.

4 The USNRC will supply experimental facility information (nodalization diagrams when available) and test data to the for code calculations as mutually agreed upon.

5. International Code Assessment and Applications Program - The USNRC has developed an international code assessment and applications progran through various bilateral agreements. Assessment activities are being coordinated with the similar programs in U.S. laboratories and industry. Annual international meetings are held for presenta-tion and discussion of results obtained from these programs. The will be invited to participate in and contribute to these meetings. Guidelines and Procedures for ICAP program coordination J

are issued by the USNRC to ICAP participants.

6. Reports - The USNRC will provide the with research reports in the area of thermal-hydraulic research (NUREG and contractor reports) upon official request by the .

B. Scope of Responsibility 1 The shall provide over the duration of this Agreement the following specified goods and services related to nuclear reactor thermal-hydraulic research:

i l

B-5

~ '-

y 37: g - ~

y .

.5-k i y

1. Code Assessment -'The' will supply-the.results from code

' assessment.st'udies (3 assessment studies per code per year - 30 for duration of this Agreement) using TRAC-PWR and RELAP5 during the term-of this A reement.

5 The assessment matrix will discussed and mutually

!' agreed upon. Where studies' include proprietary or other confidential

! or privileged information, the-USNRC will treat such information according to the provisions of Article VI. ,

) 2. Code Assessment Reports .The will publish code assessment

}.

reports in English. The USNRC will have the nonexclusive right of

publishing assessment reports containing nonproprietary information

~

]- as NUREG/IA relnrts with proper reference to the _ _ _ . - The content of i assessment reports will be discussed between the USNRC and the to .

j make them compatible with USNRC reports on code assessment and those i to be written by other parties taking part in the international code assessment and applications program. The USNRC provides suggestions in its ICAP Guidelines and Procedures document for a standardized l format to be used to facilitate comparison between the various assessment calculations to be performed by other countries and groups. This format will permit computerized information retrieval.

4 The code assessment reports will contain:

4 t

e 4

a. Comparisons of code calculations and test data;
b. Explanations of reasons for differences between predictions and

! test data; i

c. Sensitivity studies (when needed, as mutually agreed upon, with different nodalizations and/or code updates as suggested and implemented by the );

d.. Comparisons showing the accuracy of predictions of selected key parameters (as mutually agreed upon for a specific transient);

i 4

> 3-6 I

p.

gs ,4- i (

,,A Et h{

l 6

e. Suggestions for user cuidelines-for accurate and fast 'l calculations;
f. Suggestions for further code improvements; i
g. Input listings; l
h. Conclusions on capabilities of the code; and 7; e
1. Data and information necessary to perform uncertainty analysis to quantify the code accuracy.

n' If a report contains proprietary or other confidential or privileged #

information, it will be restricted as described in Article VI of this Agreement.

3. Code Model Modifications - The will provide the USNRC with nodeling modifications and improvements made to the codes. Bilateral meeting may be held as necessary to discuss details of model improve-ments made by the .

j 4. Safety Results - The will provide the USNRC with reports containing the results of any analyses performed using USNRC codes to resolve reactor safety issues.

t ARTICLE IV - ADMINISTRATION OF THE AGREEMENT A. The USNRC and the will e ch designate one' representative to coordinate and determine the detailed implementation of this Agreement. This designated representative will be referred to as an Administrator of this Agreement. The Administrators may, at their discretion, delegate this i

responsibility to the appropriate individuals with respect to a given issue.

B-7 e

v -

g r-----e , ,y , y ,~, - -~~-

7 B. The Agreement states restrictions concerning dissemination of proprietary or other confidential or privileged information. Other information that may be restricted includes matters related to organization, budget, personnel, or management. l C. The USNRC and the will endeavor to select technical personnel for assignment to the program who can contribute positively to the program.

USNRC and technical personnel assigned to the program for extended periods will be considered visiting scientists (nonsalaried) within the i program and will be expected to participate in the conduct of the analysis and experiments of the program a', directed.

D. Each party to this Agreement will have access to all reports written by its partner's technical personnel assigned to the respective programs that derive from participation in those programs.

E. Article III outlines the technical program to be performed under this .

Agreement.

F. Travel costs, living expenses, and salaries will be borne by the parties who incurred them unless specified otherwise.

ARTICLE V - PATENTS A. With respect to any invention or discovery conceived or first actually reduced to practice in the implementation of this Agreement:

1. If conceived or first actually reduced to practice by personnel of a party (the Assigning Party) or_its contractors while assigned to the other party (the Recipient Party) or its contractors:
a. The Recipient Party shall acquire all right, title, and interest in and to such invention or discovery and any patent application or patent that may result in its own country and in third countries; and B-8

8

b. The Assigning Party shall acquire all right, title, and interest in and to such invention, discovery, patent application, or patent in its own country.
2. If conceived by or first actually reduced to practice by a party or its contractors as a direct result of employing information that has been communicated to it under this Agreement by the other party or its contractors but not otherwise agreed to under a cooperative effort covered by subparagraph 3 below:
a. The party so conceiving or first actually reducing to practice such invention or discovery shall acquire all right, title, and interest in and to such invention or discovery and any patent application cr patent that may result in its own country and in third countries; and
b. The other party shall acquire all right, title, and interest in and to such invention, discovery, patent application, or patent in its own country.
3. For other specific forms of cooperation, including exchange of samples, materials, instruments, and components for special joint j research projects, the parties shall provide for appropriate distri-bution of rights to inventions. In general, however, each party should formally determine the rights to such inventions in its own country, and the rights to such inventions in other countries should be agreed to by the parties on an equitable basis.
4. Notwithstanding the allocation of rights covered under subparagraphs 1 and 2 above, in any case where one party first l actually reduces to practice after the execution of this Agreement an invention, either conceived by the other party prior to the execution of this Agreement or conceived by the other party outside of the cooperative activities implementing this Agreement, the parties shall B-9

9 provide for an appropriate distribution of rights, taking into account existing comitments with third parties; provided, however, that each party shall determine the rights to such invention in its own country.

B. The party owning a patent covering any invention referred to in paragraph A above shall license the patents to nationals or licensees of the other party, upon request of the other party, on nondiscriminatory terms and conditions under similar circumstances. At the time of such a request, the other party will be informed of all licenses already granted under such patent.

C. Each party shall take all necessary steps to provide the cooperation from its inventors required to carry out the provisions of this article. Each party shall assume the responsibility to pay awards or compensation required to be paid to its employees according to the laws of its country.

ARTICLE VI - EXCHANGE OF SCIENTIFIC INFORMATION AND USE OF THE RESULTS OF PROGRAM A. Both parties agree that, pending the grant by the transmitting party of approval to publish, information developed or transmitted under this Agreement will be freely available to governmental authorities and organ-izations cooperating with the parties. Such information, except as noted below in paragraphs B and C may, as required by the administrative proce-dure in its own country, also be made available to the public by either party through customary channels and in accordance with the normal proce-dures of the parties.

B. It is recogni7ed by both parties that, in the process of exchanging information or in the process of other cooperation, the parties may provide to each other proprietary or other confidential or privileged information. Proprietary infomation is defined as information:

l 1. Of a type customarily held in confidence by commercial firms; l B-10

I 10

2. Not generally known or publicly available from other sources; i
3. Not having been made available previously by the transmitting party or others without an agreement concerning its confidentiality;
4. Not already in the possession of the receiving party or its contrac-tors; and
5. Held in confidence by.the owner.

1 The term "other confidential or privileged information" means information other than " proprietary information" that is protected from public disclo-

sure under the laws and regulations of the country providing.the.informa-tion and that has been transmitted and received in confidence.

I Documents containing such information that are made available hereunder and that bear the following (or substantially similar) restrictive-designation:

].

"Except as set forth in the Agreement dated between the i USNRC and the this document containing proprietary or other i

j confidential or privileged infonnation shall not be disseminated outside the recipient's organization without prior approval .of. (name of transmitting party)."  :

4 r

shall be respected by the receiving party and shall not be used for.

commercial purposes, made public, or disseminated in any manner unspeci-fied by or contrary to the terns of-this Agreement without the consent _of the transmitting party.

C. Proprietary or other confidential or privileged.information shall be clearly marked by the transmitting party and shall be used only in.the furtherance of nuclear safety programs'or to resolve licensing' issues by; i B-11

___. __ , .. . _ ~ _ .. _ . . - -

II the receiving party. Its dissemination will, unless otherwise mutually agreed, be limited as follows:

1. To persons within or employed by the receiving party and to other concerned government agencies of the receiving party;
2. To prime or subcontractors of the receiving party for use only within the country of the receiving party and within the framework of.its contract (s) with that party while engaged in work relating to the subject matter of the information so disseminated;
3. To organizations licensed.by the receiving party to construct or 4

operate nuclear reactors, provided that such information is used only

! within the terms of the license and in work relating to the subject matter of the information so disseminated; and 1

4. To contractors of licensed organizations in subparagraph 3 receiving such information for use only in work for that licensee within the scope of the license and relating to the subject matter of the information so disseminated; PROVIDED that the information disseminated to any person under paragraph C l shall be on an as-needed, case-by-case basis, only to persons who are citizens of the U.S. and and shall be pursuant to an-agreement of confidentiality prohibiting the recipient (s) from any further dissemina-tion or use.

D. The application or use of any information exchanged or transferred between the parties under this Agreement shall be the responsibility of the party receiving the information, and the transmitting party does not warrant the suitability of the information for any particular use or application.

i E. It is understood that all USNRC computer codes disseminated under this )

Agreement are considered privileged information unless otherwise noted,  !

are protected as such by the USNRC, and should be treated likewise by.the B-12 r ,

I

12

. They are, in particular, subject to all of the provisions of this Article with the exception that they need not be marked with the restric-tive designation. The codes are subject to this protection in both object and source form and as recorded in any media. The codes shall be dissemi-I nated only according to this Article and shall not be disseminated outside of without prior written approval from the USNRC.

F. It is understood that the USNRC makes no warranties whatsoever for the ability or suitability of any USNRC code or other analytical technique to perform in any particular manner for any particular purpose, or to accom-plish any particular task. It is further understood that the USNRC-accepts no liability for damages of any' type that may result from the use of the USNRC codes or other analytical techniqus provided under this Agreement.

All reports published within the scope of this Agreement shall be in G.

English.

H. The USNRC computer codes and other related analytical techniques that may be disseminated under this Agreement are for the purpose of studying reactor safety and shall not be used for commercial purposes; that is, for financial or other benefit not concerned with the study of reactor safety.

USNRC codes or other analytical techniques shall not be advertised l directly or by implication to obtain contracts related to the construction, servicing, or refueling of nuclear facilities, nor shall

, advertising imply that the USNRC has endorsed any analysis or techniques.

The further agrees not to provide training in the use USNRC code to 1

other organizations or to citizens.

B-13

13 ARTICLE VII - FINAL PROVISIONS A. This Agreement shall enter into force upon signature of the parties and shall remain in force for a period of five years or until all obligations under the Agreement are fulfilled, whichever comes last. All information protected by provisions of this Agreement as proprietary, confidential, privileged, or otherwise subject to restriction on disclosure shall rcmain so protected indefinitely unless mutually agreed to in writing.

B. Either party may withdraw from the present Agreement after providing the other party written notice at least 480 days prior to its intended date of withdrawal. The party not withdrawing shall reserve the right to deter-mine if the withdrawal will result in the other party receiving a dispro-portionate share of the expected benefit from this Agreement. If so, both parties will endeavor to reach an equitable settlement of the matter through negotiation.

C. All costs arising from inplementation of this Agreement shall be borne by the party that incurs them except when specifically_ agreed to otherwise by both parties.

D. The parties to this Agreement reserve the right to modify or extend the specific activities described in Article III within the intended scope of the Agreement upon written concurrence of its Administrators.

i, E. If the portion of the research program of either party that is pertinent to this Agreement is substantially reduced or eliminated, the technical scope described in Article III may be adjusted to substitute research of equivalent progransnatic interest upon mutual agreement of the parties.

F. The USNRC and the recognize the benefits of international coopera-tion and will endeavor to obtain a mutually agreeable continuation of this Agreement before its expiration.

B-14 l

l

I 14 l FOR THE UNITED STATES NUCLEAR REGULATORY COMMISSION _  ;

BY:

TITLE: Executive Director For Operations DATE:

4 j

PLACE:

i FOR THE j

]-

1 BY:

i TITLE:

4 r i DATE: .  ;

i 1

I- PLACE: - .

\

5 ,

i i

t I

.i i 1

, -B-15 4

APPENDIX C

. REPORTED USER PROBLEMS Code development groups for the USNRC's thermal-hydraulic reactor

~

codes provide resolutions to problems code users encounter in the codes and/or their documentation. Code development personnel at the Idaho National Engineering Laboratory (INEL) and Los Alamos National Laboratory.

(LANL) keep track of reported user problems and the status of the resolution of'each problem.

This appendix contains brief descriptions of reported. user problems and the status of each resolution for the TRAC-PF1/ MODI, RELAPS/ MOD 2, and

TRAC-BF1 codes. The contents of the appendix were supplied by development personnel for each code. Problems-reported since October 1985 are presented.

}!

}

l I

i t

i 1

l 1

l i

T l

C-1 l s.

I

C.i MAJOR REPORTED USER PROBLEMS C.

1.1 INTRODUCTION

In this section we present 100 of the major problems or concerns expressed by ICAP-involved users during FY86. These problems were catalogued into our

" problem-tracking" process as they were received.

We have divided this section according to the type of difficulty encountered by the user. The major areas are input problems, modeling deficiencies, code errors, output prcDiems, code implementation difficulties, code documentation deficiencies, and other concerns. Table C.1 gives the total number of user-reported problems in each area.

C (ABLE C.1 Area of Concern Numoer Presented Input Problems 12 Mooeling Deficiencies 27 Coce Errors 20 Output Problems 13 Code Implementation Difficulties 5 Code Documentation Deficiencies 12 Other Concerns 11 TOTAL 100 Each specific prcblem identifies tne source, paves a "tform" number used to identify specific problems in the problem-tracking process, the problem description, and the status o' tre problem. The statur not only indicates the resciution of the proolem, but also gives the reaoer a chronology of events leading to that resolution, if availaole.

C.1.2 INDUT PROBLEMS SOURCE: UKAEE TFORM: Not assigned.

DESCRIDTICN: Negative fr1Ction factors at changes of area, STATUS: Some users had asked for an option to allow input of negative friction factors. Los Alamos does not uanction nonphysical models. The USNRC reiterated this position by requesting that Los Alamos " instruct all TRAC-PWR users at Los Alamos engaged in code assessment, appitcations, or experimental projects. that arbitrary or unphysical changes in code models Cannot be defended unoer tecnnical review and are thus unaccep'able" (letter. L. Shotkin, USNRC, to L. H. Sullivan. Los Alamos December 12, 1985). It was decided that any modifications necessary to improve modeling at area changes will be included in'our long-range model improvement plans.

SOURCE: KWU and SNL TFORM: Not assigned.

DESCRIPTION: "Need an option for reactivity feedback cependent on the fluid density. Reactivity data are not avhilable as a function of the incependent variables used by TRAC."

STATUS: Fluid density is a function of temperature and void fraction, which are the independent variables in the TRAC model. Equivalent densits enanges resulting from temperature changes or void enanges do not have the same reactivity worth. he decioed no change to this program feature was merited.

SOURCE: SNL TFORM: Not assigned.

DESCRIPTION: " Closely related variables often widely separated in iriput . "

STATUS: In order to maintain upwarcs compatability, we decided not to change this program feature.

SOURCE: SNL TFORM: Not assigned.

DESCRIPTION: "Cannot input mass flows, must use velocities."

STATUS: Mass flow input is allowed for FILLS. Multiphasic velocities are input variables. We do not plan to change this program requirement.

SOURCE: SNL TFORM: Not assigned.

DESCRIPTION: " Data ordering between components not always consistent (e.g., TEES vs PIPES and 3-D vs 1-D)."

STATUS: In order to maintain upwards compatability, we have no plans for C-2

changing this program feature.

SOURCE: SNL TFORM: Not assigned.

DESCRIPTION: "Some input data not checked (e.g., rod radii, fuel censity and GRAVs greater than 1.0)."

STATUS: An update was provided in EC12.2 (RADWRN) which causes a fatal input error when slab or rod radii are not monotonically increasing in VESSEL component. Idents FTDCRRJ and FTDVSRJ in EC13.3 issue warning messages to user if the fraction of theoretical fuel density is greater than 1.0, equal to 0.0 or negative. Other input-error checking improvements will be in future coding.

%B60103c SOURCE: INEL DESCRIPTION: "How do you determine cell number for signal variable pointing to cell in secondary side TEE of STGEN component?"

STATUS: The problem was resolved by providing user with some basic information. To determine cell number, use the TEE component number and let the cell number be the total number of cells on the TEE's primary side (NCELL1) plus the TEE's secondary-side cell numcar.

We receiveo a call from user with request for assistance. We looked in TRAC User's Guice for information under the signal variables section. Code manual indicated that the old convention of counting " phantom" cells was no longer necessary. User had tried this method and had not gotten correct results. We told user we would lock into this, check User's Guide furtner anc get back to him. Talked with knowledgeable engineer, who said there are two methods to determine seconcary-side TEE cell:

1. Icentify internal separate steam-generator component individually and reference that number in signal varisole oefinition card. Then count, starting with main tube until all cells are added. Then add 1 for phantom cell anc continue counting until desired side tube cell is reacned. That is the number to reference in signal variable definition.
2. Identify STGEN component number and start counting on primary side until all cells are added up. Add 1 cell (phantom) to count for each component before desired secondary component. Count cells in intermediate components as normal (including phantom cell between main and side tube in tees) until desired seconcary cell is reached. Reference that number in signal va r- t abl e definition. Orcer of components is as listed in secondary comocnent input of STCEN and count in direction of increasing cell numbers.

we found similar useful information tr. User's Guioe on page V-14 of TRAC INPUT DECK ASSEMBLY AND EXECUTION section. We ran standalone test cases to verify these instructions as the new manual description was not conststent with those in the user guice or old manual, and phantom cell might not have been relevant.

Test runs validated information given to user "or both methods. This meant trat information in new code manual was incorrect as it did not refer to counting extra " pseudo" or " phantom" cells as the old manual did.

We decided that a revised explanation regarding counting in the steam-generator component should be written for the code manual.

SOURCE: SNL TFORM: Not assigned.

DESCRIPTION: " Changing the wall roughness requires code modifications."

STATUS: Closed. We will allow pipe roughness factor as input in M002.

SOURCE: Los Alamos TFORM: 860221a DESCRIPTION: "Wants error message if user tries to input less than one heat-transfer node."

STATUS: Closed. This user convenience was provided with ident NDCKRJ in EC13.0.

SCURCE: Los Alamos TFORM: 860708c DESCRIPTION: " Code does not run past first time step with a certain combination of NOAIR and IEOS input values. There were 10 outer iterations and then it died with message 'cannot reduce time step further.' "

STATUS: Problem was resolved. This was the first successful use of the new trouble shooting software, TSH00TER.

The user came to us with the problem. and we examined the TRAC input. There were lots of TF1D failures. He was using a pipe to model an accumulator and had default NOAIR option. User had set NOAIR=0 and IEOS=1.

We ran TSH00TER with a keyword search on " air" There were 5 " matches." The C-3

data base it searched was approaching 200 tforms in length. The user examined the TSH00TER output and later told us that one of the trouble forms (851211a) he found related to a very similar problem that a user was having back in December, 1985. Changes were made (NOAIR=1 and IEOS=i) and problem progressed normally. We discussed this with other staff members, and it seemed that there is a posstble inconsistency between the NOAIR option (=O) and IE05 option on Main-Data Card 1 set to 1. (This informa? on was in tform found with TSH00TER).

It was suggested that we look into this and possibly have code check this and issue a warning message to users.

A check for this incompatibility and warning message will be provided with future ccding.

SOURCE: BNL 1 TFORM: 860107a l DESCRIPTION: 'Would like to have EXTRACT for frozen TRAC 12.1 plus ECs through EC12.3. Says the maximum rate of change of power value in his VESSEL input limits the rate at which nis power can drop after scram for a LOCA."

STATUS: Closed. EXTRACT compatible with EC12.7 is available, and a TRAC-independent version will be available soon.

The user had a VESSEL component that needed to be changed on e restart, and he did not want to make all the changes by hand. We informed him that we had an EXTRACT for 12.3 but thought it would take some work at his site to implement it on his CDC 7600. We said we would send it on an IBM diskette if it wasn't too large a file and if there were no other complications.

Also told him that other offsite users (see tform 860102a) had discovered that if they did not preset core memory to indefinite values. TPOWR and IPOWR options did not work. Recommended be ensure that compilation option for coce was correct in this regard. Told him maximum rate of change parameter. RRPWMA (card 11 field 4), was intended to be a 11miting value.

He commented that it would be nice if TRAC would allow a single card change on restart instead of requiring that entire component be replaced for a single ena r.ge . (This is wny he wanted EXTRACT. py the way.)

We informed him that EXTRACT was a created witn a set of updates that would require having someone with UPDATE exoerience as well as FORTRAN compilation and loeding knowledge in order to successfully implement.

We attempted to locate a verston 12.3 EXTRACT. but the only one available was intended only for use on our Crays and would not work on 7600s without further effort. We discussed the toea that in the future an effort should be undertaken to install EXTRACT directly into TRAC using IF DEFs. so that a new version would not have to be generated every time a change was made to TRAC.

!t was suggested that this would take approximately 3 to 4 weeks of dedicated effort to accomplish. In this case. With the lintted resources available to the user, it was suggested that he simply employ a text editor to make the necessary changes to the ISEL component using values from the desired restart TRCOUT edit. This shoulc te about 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br />, whereas efforts to implement EXTRACT on his system cou ske weeks.

User was given Los Alamos recommendations and concurred. He catd his primary difficulty was contending with the batch system and slos turnaround f or big batch jobs Itke TRAC.

SOURCE: Los Alamos TFORM: 860424a DESCRIPTION: " Set reverse k at both boundaries of steam generator TEE (single cell) to 1e30 but still got reverse liquid velocity."

STATUS: The users had set reverse k at both boundaries of steam-generator TEE (single cell) to 'e30. Their intention was to limit reverse liquid flow uhtle allowing vapor flow. Using this technique and a BREAK connected to the TEE.

they hoped to allow nitrogen injection later in the transient when they would replace vapor field with gas field.

However, they found that during a time when main tube of TEE was liquid full (alpha =O) and break was all vapor (alpha =1.0) and there was a pressure drop of about 3 MPa from the main tube to the side tube of the TEE. there was no positive liquid velocity through the side tuoe. In fact, the output showed itquid velocity of -1.2e-10. a positive vapor velocity of about 102 and an FF of 1e30.

They questioned the output because they did not expect it with the given pressure differential. They used TRAC-PF1/ MOD 1 with EC12.6.

We talked with users regarding REvK inconststencies. We also examined the code and found that conversion of KFACs to FRICs occurs in ELGR and there is some C-4

error checking on' values'.gt. ie20.-We examined code manual'regarding TEE and found that steam-separatoa model was' invoked by setting FRICs greater than 1e20. There was concern that REV FRICs might not be treated same as FRICs for-separator logic.

It was noted that warning to users existed regarding use of'large FRIC values' elsewhere in their. input model and could. inadvertently invoke separator model wnere it is not desired (especially,-in a pipe).

l It was agreed after much discussion that user would attempt to recreate problem

-and supply input deck for further investigation.

end i

J i

f i

I i

i 4

1 C-5

_ _ ._ _ _ - , -_ _ _ _ . . _ . , _ _ _ . . _ . . - . _ _ ~ , . - - . _ , .

C.1.3 MODELING DEFICIENCIES SOURCE: INEL TFORM: Not assigned.

DESCRIPTION: Heat structures in plenum.

STATUS: May be developed as a user convenience fol;owing implementation of the heat structure module.

SOURCE: 2D/30 and workshop TFORM: Not assigned.

DESCRIPTION: Decreasing maximum time step causes variation in instantaneous results without convergence.

STATUS: The problem arises principally because of the quasi-steady-state analysis technique employed by all codes of this type in providing the necessary constitutive relations. We are currently working on resolving this problem. Our goal is to substantially reduce the sensitivity of the cooe results to the time step.

SOURCE: ICAP TFORM: Not assigned.

DESCRIPTION: "Modeling of coolant seal leakage into primary Lystem."

STATUS: Except for a pump small-break loss-of-coolant accident, this does not represent a modeling oeficiency for full-scale plants. It is envisioned that tnis capacility will be available with a generalized 1-D component moce). It is envisioned that the component will incorporate all the features of the current specialized 1-D models and significantly enhance future vectorization capabilities and run times.

SOURCE: QRS and SNL.

TFORM: Not assigneo.

DESCRIPTION: " User ioentifies liquid carry-over ano steam carry-under in separator. Perfect separator for steam generator is too limited."

STATUS: Variable efficiency separator model is being developed. Work is underway and should be completed Dy the end of FY86. User feature is available with the SEPD ident in EC13.3.

SOURCE: GRS and UKAEE TFORM: Not assigned.

DESCRIPTION: "Coce is too slow. Would like to see 3-D-2-step and carallel processing in TRAC."

STATUS: Code SpeeQuo ennancements are under review will be available in M002 Code.

SOURCE: UKAEE TFORM: Not assigned.

DESCRIPTION: "Coce needs improved interphase area for condensation in stratified flow."

STATUS: This relates to overall constitutive package improvements that will be addressed in the near future.

SOURCE: UKAEE TFORM: Nnt assigned.

DESCRIPTION: " Interface sharpener too sharp. Interface frtCtion and sharpener are not consistent witn each other."

STATUS: Work is currently underway.

SOURCE: KWU, INEL, UKAEE and SNL TFORM: Not assigned.

DESCRIPTION: " Code should allow heat transfer between cells having no fluid connection. Generalized heat slabs are for steam generators only, but are needed for all components. Cannot specify heat transfer on cell-by-cell basis for 1-D components. Can have internal slabs for 3-D cnly and cannot connect between cells or other components. No heat slabs are allowed for the plenum Component."

STATUS: Users were informed that these capabilities will be available with the generalized heat structure model that has been implemented into M002.

SOURCE: KWU TFORM: 851204b OESCRIPTION: "Need better fuel rod models including relocation, variable gao thickness, etc."

STATUS: Gap thickness is a user input, but is not variable with respect to rod length. This feature will be considered for implementation in TRAC at our next prioritization meeting with US NRC. We have no plans for changing the program structure to allow fuel rod relocation at this time. TRAC /MELPROG, a separate fuel-damage package, simulates rod relocation during meltdown.

C-6

We received a hand-written document (" Attachment F*) from Sam Naff, NRC to F.

Ocar, dated 7/22/85. This contained feedback on TRAC problems from meetings with KWU and GR$ and included:

1. Vojtek (GRS) showed time-step dependence. Overall global results good, but timing of events varied for different time steps. F. Motley agreed to look into this.
2. Glaser (GRS) showed problem with Karlstein model which gave non-realistic temp. rise in dead ended drain line. F. Motley identified a potential problem with drain valve leakage, but said he would investigate the

, problem at LANL with input deck he took back.

! 3. Plank (KWU) asked that the following be included in the code: a.

Heat transfer between cells having no fluid connection. b. Better fuel rod mocels including relocation, variable gap thickness, etc..c. Reactivity-feedback dependent on fluid density. d. Flow regime information in graphics file output. Also did not like having annular flow option in code when it gave i bad results and was recommended not to be used (nff=2 or -2). Also concernec

! with pressure losses at TEES being off.

4. Hobbhahn (GRS) had questions about slight assymetries in vessel he
perceived to be time-step dependent.

! 5. Following output improvements were reouested: a. Heat transfer and i mode in graphics file b. Flow regime indication in graphics c, . Friction factor in printed output d. An indication of changes in choke flow or its location e. User-specified variable output in graphics file After receiving information (in December 1985) we made effort to respond to several of these concerns with the following Chronology:

(12/85) Item 1 was addressed and fixed.

(7/16/86) Better fuel-rod models will be in mod 3 (see Vuly 86 TRAC NEWS) .wali pipe roughness also promised in mod 2.

(7/16/86) Regarding need for user-specified variable output in graphics file, user currently can provide additional variables to graphics files by defining appropriate signal variables and control blocks. No further work planned.

(9/5/86) Update NFFIXi re-instates nff=2 option with warning to user.

(9/12/86) Better fuel-rod models and wall pipe roughness will be addressed in mod 2 code.

(9/24/86) At code proolem review meeting it was agreed to attempt to continue' looking at Plank's (item 3) concern regarding pressure loss at TEES. but un a very low priority basis. Have lineited information avaliable from user.

SCURCE: KWU TFORM: 851204b ,

DESCRIPTION: " Annular flow wall friction optior. in the 1-D components gives excessive pressure drops.

STATUS: For most applications the annular flow option is not recommended.

Annular flow option (NFF=2) will be discouraged with future warning messages in the coding. Because part of the problem is the assumed pipe roughness, we will a11ow pipe roughness factor as input in M002.

SOURCE: SNL TFORM: Not assigned.

. DESCRIPTION: "Only 90-degree vessel connections are allowed."

STATUS: We informed user that we have no plans for changing this part of the code at this time. However, development activities aimed at producing a multiple source connection capability for the VESSEL component were started.

In audition, we discussed the possibility of allowing full-space orientation for the VESSEL (eg -- turning the VESSEL on its side).

SOURCE: SNL TFORM: Not' assigned.

DESCRIPTION: 'The material that represents tne gap gases is limited to special cases such as no hcIlow rods."

STATUS: We decided not to remove the constraints on material properties for MATRD. Users may define a new material in the Materials Properties Data section to account for different gap gases, if they wish.

SOURCE: SNL TFORM: Not assigned.

DESCRIPTION: " Liquid and gas velocities forced to be equal in pump."

- STATUS: If this assumption is not made, then one has to partition the pump j momentum source. We decided change was not merited.

! SOURCE::& SNL TFORM: Not assigned.

DESCRIPTION: "The forced use of multicell components (e.g., PUMP for

! the momentum source. VALVE for the variable junction area and single i connection for the VESSEL cells).*

STATUS: Ne informed user that a generte component with variable junction area and variable momentum source modeling capability was under review for a future I

C-7 1

y , .-. ,- . .---,-- , ,-- ,, . - - , - -

- - - , , , -~ . . m- g,-,. - -,-- , -

version.'This capability would not be implemented in TRAC-PF1, however.

SOURCE: GRS and Los Alamos TFORM: Not assigned.

DESCRIPTION: " Multiple-source vessel cell."

STATUS: Users were given this cacability with EC12.7.

SOURCE::8 UKAEE TFORK: Act assigned.

DESCRIPTIt.N: " TRAC uses h1 = .02 rho 1 ( Cp1 )V1 for interfacial heat transfer. By using turbulent (Bankoff, Thomas) model you get better agreement."

STATUS. Users were informed tt 4t the suggested model (Bankoff. Thomas) requires an entrance length thot is not available in the code. An earlier assessment showed current model was adequate, however.

SOURCE: GRS TFORM: Not assigned.

DESCRIPTION: "Better techniques for the flow regime specifications using transport equations are needed."

STATUS: We informed user that we.would also like to pursue this activity, but we did not nave sufficient resources and other activities took priority.

SOURCE: Los Alamos TFORM: 851003a DESCRIPTION: " Problem with hot rod temperatures in CORE and VESSEL. Rod temperatures for floating nodet in hot roo where zirconium / water reaction is considered may be too low."

STATUS: User was able to provide reasonable prediction of rod temperatures based on further interpretation of the output. Guidelines may be provided in future updates to TRAC User's Guide. A description of the metal-water reaction modeling was developed and distributed as a change to the User's Guide.

We talked with user. There are..two types of nodes in the rod heat transfer calculat,1ons: fixed nodes and floating nodes. The fixed nodes do not renode in response to fine-mesh calculation recuirements, whereas floating noces are

~

renoded in response to the fine-mesh calculation reautrements.

The code only calculates the temperature based on 2r/ water reaction in the fixed noces. Radial conduction is calculated for both tyoes of nodes. To -

ob+rin the temperature at intermediate floating cells curing a 2r/ water reaction. the crde interpolates between two closest fixed nodes. Thus. in intermediate floating cells where the temperature is at or above the Zr/ water reaction temperature (where there is a sudden rapid increase in heat generation r&te). the temperature inay be underpredicted oy strictly an interpolation method.

It was suggested that using a " binary switch" type of logic might invoke 2r/ water reactten calculation in the floating cells, as well. if the temperature was determined to be above the minimum value, but it was noted that TRAC was not cesigned to be a degraced core analysis code, and it was quite possible that renoding with finer fixed meshes mignt provide more accurate 2r/ water reaction information, if cesired.

SOURCE: USNRC TFORM: 851121b DESCRIPTION: "The licuid de-entrainment in the upper plenum and entrainment tnrough the hot leg are not accurately modeled."

STATUS: Work completed by P. Shire at Los Alamos using isolated upper plenum model with correct boundary conditions indicated that liquid is properly de-entrained in the upper plenum provided fine noda112ation is employed (see paper, LA-UR-86-2281 July 1986).

SOURCE: USNRC Jnd UK45E TFORM: 851121b DESCRIPTION: " Vapor void fraction in the core is not accurately predicted. ,

Interface sharpener too sharp. Interface friction and sharpener are not consistent with each other."

STATUS: While this information was not new to us, a resurgence of interest by external users has helped to focus more attention on the need for improvements to TRAC constitutive packages. Several improvements will be going into MOO 2 version of the code.

SOURCE: UKAEE TFORM: 860102b DESCRIPTION: Reports problems with accumulator modeling: Accumulator never empties. "The TRAC-PF1/ MOO 1 code allows the accumulator to be modelled using a C-8

PIPE component ith the accumulater model flag IACC set to 1 or 2." "With IACC

=2, it ts noticed that the accumulator tank effectively never empties, because the exit 11ould flow rate is given by:

m-dot-sub-1 = (rho-1) (V-1) A (1-alpha),

where the value used for the void fraction. alpha. is that of the fluid cell just upstream of the exit junction. Therefore, as alpha --> 1 then m-dot-sub-1

--> 0.*

STATUS: We received a letter addressing problem from user dated December 12, 1985. He was concerned as to whether result was an intended consequence of the

. accumulator model, or should the exit value of alpha remain at zero until the tank is empty?

We sent a letter of acknowledgement to user, thanked him for his problem statement and told him we would be considering this at our next prioritization l' meeting. Discussions with developers indicated that emptying of accumulator oepends on option selected by user. The behavior observed by user was just a

, well-known example of numerical diffusion due to conor-calling. Our normal recommendations to users had suggested using a small cell at the bottom of ACCUM to mitigate this. At the pricritization meeting, the development of additional guicelines was assigned to be included with future change-pages to the TRAC user's Guide.

SOURCE: Los Alamos TFORM: 860603a DESCRIPTION: " Time-step limit in VESSEL cell appears too low considering existing flow lengths and fluid velocities."

STATUS: The section in the TRAC User's Guide on Courant time-step control has been expanded. Potential importance for time-step control of user-input VOLs and FAs is now documented. Also, commenting in TRAC subroutine NEWOLT was improved.

SOURCE::6 ICAP TFORM: 860707a DESCRIPTION: "Wants default INVAN option changed to 1 with associated change to documentation."

STATUS: C1ossd. Corrections needed to allow proper operation of the INVAN flag have been installed in EC12.9 with update ident FXINVAN. Changing INVAN default anc associated assessment activity will be consioered for MOD 2 version of the code.

SCURCE: KWU TFORM: 860707d DESCRIPTION: "Wants improved accuracy of thermodynamic properties in TRAC."

STATUS: Tnis concern was brougnt to our attention at Eriangen meeting in dune 1986. User said that they specifically wanted improvement of the accuracy for the pressure range from 1.Oe5 Pa to 450.Oe5 Pa. for liquid temperatures from 273.15K to 713.94K and vapor temperatures from 273.15K to 3OOO.OK.

Although some users have expressed concern that the saturation line may be off a minor amount in places, this does not represent a major problem in modeling system performance. We currently believe that the correlations in subroutine THERMO are accurate enough for the transients of interest at the present time.

The greater uncertainties exist in wall and interfacial heat transfer and the user's modeling of the physical system, and reduction of these errors presently represents a much higher priority.

SOURCE: KWU TFORM: 860707e DESCRIPTION: "Wants propagation and tracing of sharp steam / water fronts within the 1-D cells."

STATUS: This concern was brought to our attention at a meeting in Erlangen in June 1986. The front of concern was identifteu as a "two-phase mixture front in 1D components." The M002 code will track the vertical interface with the separator model turned on.

SOURCE: KWU TFORM: 860707f DESCRIPTION: "We would like to see the capability of the constitutive equations for a two-phase steam / water mixture extended above the critical pressure, P > 22.12 MPa."

STATUS: User was informed (via TRAC NEWS) that this capability already existed for the current code with error corrections through EC13.0. The super-critica fluid is currently modeled as a two-phase steam / water mixture. The thermodynamic correlations have been extended into the super-critical region.

This was implemented in all versions after 11.8, and we have not observed any difficulties associated with properties above the critical pressure.

C-9

. .. - . . _ m . . _ _ .

i' Additional changes will be made as necessary in response to specific TRAC problems as.they arise.

SOURCE: KWU 1 TFORM: 8607079 )

, DESCRIPTION: "Would'like to see modeling of nucleation delay in all components.'

STATUS: This concern was brought to our attention at a meeting in Erlangen in June 1986. User was informed that we do not see this as a serious modeling_

4

+ deficiency in TRAC. .Modeling of nucleation delay may be of some significance

' for components with large liquid volumes and small structural surface area. +

However, we currently have no intention of incorporating-this model into the '

code. There are no accepted correlations that would account for the broad range of structural configurations. surface characteristics, and 11 Quid impurities necessary to construct an'auequate general model.

end j

i I

t l

t i

r

?+

4 4

4 i

l i

s C-10

.. . .= - . . .

C.1.4 CODE ERRORS SOURCE: ICAP TFORM: Not assigned. . .

DESCRIPTION: TCHF not calculated under nucleate boiling conditions.

TCHF should be less than TMIN.

STATUS: A correction to these problems has been implemented as an update to version 12.-i (HTFIX).

SOURCE: SNL TFORM: Not assigned.

DESCRIPTION: "Some SVs not defined if correct code options not chosen (e.g., core average properties)."

STATUS: To obtain tne core average liquid temperature requires that the kinetics option be turned on. We will highlight those SVs dependent on code options in future user guidelines.

i SOURCE: ICAP TFORM: Not assigned.

DESCRIPTION: "There are some discontinuities in these calculations that have slightly inclined planes. The discontinuities are evident when compared to i similar calculations performed for horizontal planes."

STATUS: This proolem relates to the switching in the logic to determine horizontally stratified flow. Ident STEST in EC12.6 addresses this prob 1em.

SOURCE: SNL d

TFORM: Not assigned.

DESCRIPTION: " Signal variables do not work properly for plenum components."

STATUS: Update ioent SVPTS. whicn has gone into EC12.6 defines the pointer variables needed to access signal variao1es cefined in a PLENUM component.

SCURCE: KWU TFORM: 860327a DESCRIPTION: "The vapor mass flow rate for the graphics output file.

1 should be proportional to the veper fraction, alpha ."

j STATUS: Closed. Fix was made available in EC12.7 (ident VMFRFX).

l' SOURCE: SNL a

TFORM: 860701d i

DESCRIPTION: "Some uninitialized variables in code still exist. All problem variables not properly initialized before first time step."

STATUS: Closed.

(6/20/86)

' Apparently Sandia had tried the L2-5 or 6 LOFT analysis with the preset of memory in TRAC set to positive indefinite. The Los Alamos proceoure is to preset memory to negative indefinite. Sandia generated an update that defined some of the undefined integers, which paralleled an ongoing- low-priority task that was already underway. We reviewed the Sandia updates to see if they found some that we had missed. Our Configuration Control Person (CCP) reported that Sandia updates did not mesh with our program library to produce a running code tnus requiring further investigation by staff. EC12.8 contains updates FXNCMNX~

and PCCDER which preset NCMN. NCMX and KK.

(8/86) Besides FXUNDEF in EC12.8 EC12.9 update UPDCLN fixes additional undefined variables as does EC13.0 update CCFLX2. Additional preinitialization activity is ongoing.

SOURCE: SNL. INEL and EIR TFORM: 860706b DESCRIPTION: " Common block VLTAB has different lengths in different routines and PTAB arrays have different lengths in different routines."

STATUS: Closed.

(7/4/86) Users said that their 1caders were not swallowing the PTAB common block. They were getting message "long common found in subroutine OUTER" or something like that. One user also said that he had a problem with MEMADJ which is called from SETLCM. He said that they were running on a Cray-1s but MEMADJ is not available. We told users that our loader was less particular, but 4 that other offsite users had also seen similar message, and we would create a fix to solve the problem.

T (7/86) Problem.with VLTAB and PTAB common blocks resolved in EC12.8 with update CMNSIZE.

SOURCE: GRS and UKAEE TFORM: Not assigned.

, DESCRIPTION: "Possible nonphysical switches from stratified to plug flow."

STATUS: Closed.

Changes were made to the test on volume velocity (VLVC) in subroutine HTIF to C-11

make the stratified flow test compatible with FEMOM (EC12.5 ident ALVLS). Any addtttonal changes required will be made in response to specific TRAC problems as they artse and are provided to us (with data comparisons).

SOURCE: UKAEE I

TFORM: Not assigned.

DESCRIPTION: "Coce has velocity discontinuity in condensation heat l transfer." l STATUS: Closed.

Coding was reviewed. Coding w6s not in error. Additional changes will be made as necessary in response to specific TRAC problems as they arise and are provided to us with cata comparisons.

SOURCE: UKAEE TFORM: 85110Sa DESCRIPTION: " Error in wall heat transfer package. Error occurs in the value of conductivity used in evaluating the term ERAC in subroutine DFHT. TRAC uses liquid Conductivity when vapor conductivity should actually be employed."

STATUS: Closed.

(11/18/85) - Talked witn UK and informed user that we had 'eceived his letter and appreciated the feedback. Discussed prior work and incicated that long term work on correlations was being sorted out at the present time. This specific error and several others were found over 1 and 1/2 years ago by Ivan Vojtek, during his stay at Los Alamoc as the Gurman resident engineer. Ivan made a number cf recommendations for error corrections and improvements to the heat transfer package. During our investigation of a snort-term fix to the post-CHF in order to better predtet quenchtng for LOFT, we incluoed Ivan's error ccrrections as part of the package. This work was summarized tn a memo to Dennis Liles on November 29, 1984 The conclusion was that while the errors where definitely errors, corrections to them lead to poorer predictions of the reflood problems considered because of the coupled nature of the drag, interfacial and wall heat transfer packages. Thus these errors in wall heat transfer were cetng offset by other errors in the drag and interfacial pacnages to produce the generally good results the code currently obtains. Therefore we chose not to make these corrections even though they were technically correct until the constitutive packages could be investigated as a whole. (Note that the same can be said to the INVAN flag question which the U.K. has also raised.) While this answer will may not completely satisfy all users., it is a valid answer, and one hhich often arises for a " mature" code.

(9/24/86) Fixed with ident UPFILM in EC12.9.

SOURCE: UKAEE TFORM: 860407b DESCRIPTION: " Code error exists for inflow from a BREAK component with IVDV = 0."

STATUS: Closed. Fixed in EC12.9 with ident FXVDV1.

SOURCE: UKAEE TFORM: 860407c DESCRIPTION: "Because the nitrogen temperature is not allowed to fall below 273.15 K in TRAC, the nitrogen pressure is overestimated during the latter part of accumulator discharge during a typical large-break LOCA."

STATUS: Closed.

(4/1/86) Received letter addressing problem from user (dated 1C March 1986).

(4/7/86) Distributed letter for review.

(7/11/86) Renewed effort to review problem submitted.

(9/19/86) Informed user (October issue of TRAC NEWS) that accumulator moceling guideltnes would be provided in future changes coco documentation.

SOURCE: Los Alamos TFORM: 860827c DESCRIPTION: Logic error exists in line STGN3X.61 for the STGEN graphics.

STATUS: Closed. Fixed with ident FXHTLOSS in EC13.0.

SOURCE: Los Alamos TFORM; 860827d DESCRIPTION: The user's calculation runs approximately six times slower with the VESSEL mass-error checker turned on.

STATUS: Closed.

(8/26/86) Received VAX MAIL message with problem. It was recommended that we suggest users employ direct inversion option if at all possible as problem appears to La related to Gauss-Setdel with Coarse Mesh Rebalance methods.

However, problem does not disappear just by selecting DI.

l l By allowing a larger value for acceptable /essel mass error, user can improve l run times. It has been recommended that some work be done to vessel mass error l

time-step-limit logic in the code.

l (9/16/86) Fixed with ident VMERFXD in EC13.0.

C-12

SOURCE: Los Alamos TFORM: 860905b DESCRIPTION: " Bad scalar argument" run-time error when code tries to take sovare root of a negative number in subroutine SOUNO.

STATUS: Closed.

User had found a problem with previous choked-flow fixes. To calculate new sonic velocity for steam we had used an expression that involved tne difference oetween the stagnation enthalpy and the throat enthalpy to the 1/2 power.

Apparently. user had a steam line tha+ had a stagnation enthalpy less than the throat enthalpy. Therefore our coding tried to take SQRT of a negative number.

(9/5/86) Update developed that fixed the problem. For user's problem. the steam line had a small quantity of water with some superheated steam. The coding in the SOUND subroutine re-evaluates the GAMMA (CP/CV) If the void fraction is less than one. Therefore. the expresstons that used GAMMA to expand the superheated steam down to the throat pressure used a GAMMA of 1.135 whereas the GAMMA should have been 1.3. With the GAMMA of 1.135 the throat enthalpy was calculated to be larger than the stagnation entnalpy. Ideal gas assumptions cannot be applied to steam Close to the saturation line.

Two changes were made to initial coding for cho41ng of superheated steam.

1. If the void fraction is less than one, then the old method for estimating the throat conditions and the choked flow rates is used.
2. If the stagnation conditions are such that the new method was to be used, then the choked flow velocity is calculated based on the difference in-enthalples. This is the same as before and based on ideal gas approximation, and the maximum of the two velocities is used.

(9/19/86) Fixed with UPDSNO2 in EC13.0.

SCURCE: GRS TFORM: 860905c DESCRIPTION: "In VALVE. reverse FRIC does not work for a certain valve-opening table.*

STATUS: Closed.

(8/24/86) Received note of problem that showed that when the valve component-action taole gives constant flow area vs. time. rev-friction works.

However, when the valve area is ramped open over O.Ss. the rev-friction does not work.

The way the code is set up. the Cooe uses the input cell DXs and H0s to convert input K factors to FRICs during the initialization step and never recalculates the FRICs later: in fact. the code overwrites the K factorb with the calculated FRICs. ano the inout numbers are lort anc not passed tnrough the dump on restart. For a VALVE that is initialized closed. tne HD at the valve interface is set to sometning like 1.Oe-10 in either RVALVE or IVALVE regardless of wnat is input in the HD array. The calculation of the FRIC at the VALVL interface takes place later and uses an HD of 1.Oe-10. resulting in a very small FRIC (essentially zero). If FRICs are input, there it no problem: when the VALVE 1s closec. a small HD 1s used resulting in a large K in the momentum equation (because the VALVE is closed. the velocity is zero and no effect): when the VALVE is open (the condition under whien the K factor is normally determined).

the nominal HD is used to yield the Correct K in the momentum ecuation.

(8/26/86) FRIC problem in the 1-0 valve was fixed with VLVFRIC1 in EC13.1.

SOURCE: Los Alamos TFORM: 860916a DESCRIPTION: " Choked-flow rate is too low out one of the breaks in USPWR model."

STATUS: Closed. Problem may be related to ident FXUNDEF provided for EC12.8 to preinitialize the choked-flow flag. ICFLG. The fix set ICFLG to zero in RCOMP. However. ICFLG should be set to zero only once per component and not each time RCOMP is called.

(9/16/86) User had a problem with the choked-flow model with version 12.8.

This was before choked-flow fixes went into 12.9. Apparently he noticed that the choked-flow rate was very low out one of the breaks. The problem was with the FAUNDEF update supplied by SNL for version 12.8. FXUNDEF set ICFLG to zero in the RCOMP routine. ICFLG is a variable in the fixed length table that is untQue for a given component. ICFLG determines which set of choked flow multipliers to use in the choked flow model. If the user had a TEE in his/her model. and if ICFLOW was set to 2. then there was a potential for ICFLG to be set to zero. If ICFLG was zero, then the code would pick up an indeterminate value for the choked flow multiplier. Each time RCOMP was called with SNL's update. ICFLG was set to zero, then if the user had input any ICFLG flags, then ICFLG was set to that input. For a TEE component. RCOMP is called twice, once for the primary and once for the secondary. If the user input all zeros for the secondary. then the primary side input was ignored and ICFLG was set to zero by the update. ICLFG should be set to zero only once per component, and not each ime RCOMP is called.

torracted with ident UPICFLG in EC13.0.

SOURCE: SNL i

l C-13 i

i

)

._ =m _. . ._ . .

4 L

4 TFORM: 850911e .

DESCRIPTION: " Friction is miscalculated when liquid is removed from.

2 ' countercurrent flow calculation. A 1-D; flooding problem was run in which itquid field was effectively removed leaving air and vapor fields. TRAC still calculated liquid velocity (very small) which snould not have affected vapor veloci t ies, but i t t did."

STATUS: Closed. .

(9/12/85) Discussed -problem and history with staff. Problem pertained to a 1D flooding calculation run in which liquid field was effectively removed leaving air and vapor fields. TRAC still calculated liquid velocity (very small) which should not have affected vapor velocities. but they did. An update was provided to user (private to Sandia) called FXCC which appeared to fix the peculem.. However when liquid field was put back in there were still discontinuities.in the vapor .velectty whenever the liquid velocity changed sign. To give us a better indication of the significance of the bug, we asked.

for a draft of user's assessment report.

(10/8/85) Received draft of assessment report. .

I' '(9/24/86)- Reviewed at cooe problens review meeting. . We received draft of .

a user's report. We investigated the problem on a low-priority bacts.

4 (10/6/86) Staff indicated that user's problem does not show up unless tne pipe i is connected to a pienum component. We fixed the discontinuity that user

. discovered with ident FIXWFV in EC13.1.

+

4 SOURCE: KWU 1 TFORM: 851204b DESCRIPTION: " Pressure losses at TEES appear to oe incorrect."

STATUS: Open. Concern wt11 be addressed, as resources permit, on a low-priority basis.

-SOURCE: GRS i

TFORM: BGO910a DESCRIPTION: "Early pressure drop (about 1.6 MPa) in M001 EC12.6 steady state calculation did not occur for identical inout data set with MOD 1 EC12.5."

STATUS: Open.

(9/10/86) Received package from GR$ with grapns of pressure'vs time for steacy state calculat ions f er both ec12.8 and ec12.5 4 Computer. time for 10s problem time was as follows:

VSN 12.5: 1256 CPTIME VSN 12.8: 11624 CPTIME (9/24/86) We have asked user to send an IBM floppy with their input deck demonstrating this problem. We are investigating, h end i

I l

t 1,

.1 j C-14 i

. - -. - .~ -. .

i i

C.1.5 OUTPUT PROBLEMS 4

SOURCE: INEL

.. TFORM: Not assigned.

!' DESCRIPTION: Difficulty relating error diagnostics to code problems.

STATUS: We decided we would try to provide training workshops to foreign and '

domestic users as time and budget permit and on an individual basis to help them develop skills in TRAC input debugging techniques. In addition, we have'

] an ongoing effort to improve TRAC error diagnostics.

i- ~ SOURCE: GRS. KWU TFORM: Not assigned.

DESCRIPTION: "We need the heat transfer and mode in the graphics file."

STATUS: We informed users that the liquid and vapor heat transfer coefficients as well as the flow regime are available in the graphics output provioed they select wall heat transfer (NODES .ge. 1). These are cell-wise variables designated as HL. HV and IDR, respectively.

-SOURCE: GRS, KWU TFORM: Not assigned.

DESCRIPTION: " Friction factor needs to be identified in printed output.'

STATUS: We informed users that the friction factor was identified by the column heading "FF" in the printed cutput_ We also told users that an 1-annotated output listing showing thth parameter is provided tn the TRAC User's Gutoe.

SOURCE: GRS, KWU TFORM Not assigned.

, DESCRIPTION: " Output should give an indication of changes in chone flow and 5

its location.'

STATUS: User wss informed that when the flag. FF, is equal to.-1.111e11 in the .,

2 component output, the flow is choked. We also indicated that the TRAC User's f

Guide would show not only an anr.otated TRCOUT file, but similarly, annotated TRACIN and TRCMSG files, as well.

4 SOURCE: SNL

TFORM: Not assigned.

, DESCRIPTION: " Component mass not available as signal variable and graphics 4

output."

STATUG: No changes were made to tne code, because the component mass is readtly available by using control blocks with input of cell liquid and vaoor j density and void fraction (from signal variable) and cell volume (from inout geometry). We informed user that control blocks can be used in component action

! tables just itke signal variables and are available througn grepnics output

, (just one of numerous applications of control blocks).

! SOURCE: SNL TFORM: Not assigned, i~

DESCRIPTION: ' Mass flows are not in major edits."

STATUS: No action was taken, because mass flows are output for PUMP and VALVE i'

interfaces and are available through user-defined variables (ISVN=68. 69, 70) in a major edit. Furthermore, information user had requested was readily available, and most graphics post-processors (our TRAP. for example) should have the Capability to perform manipulation of basic variables to produce e additional information in a form desirable for user's specific needs. In 3-addition. signal variables can be used to edit mass flows in the printed output.

I SOURCE: SNL TFORM: Not assigned.

DESCRIPTION: "The SG error message should specify which wall areas are inconsistent.'

, STATUS: We informed user that to avoid this problem in the future, we would j reduce input requirements where appropriate. For example, in this case, one i'

should only have had to enter one wall area (either outer or inner), the corresponding radius, and the wall thickness. Fix will be provided with future coding changes.

I SOURCE; Los Alamos

'! TFORM: 850916a

. DESCRIPTION: " Overt error in STGEN input did not result in any TRAC error

, message. The code exited because of a type 306 operand range error. A traceback pinpointed the error in STGEN1. By double-checking the input, it was 4

found that NSJUN (STGEN input card 6 field 3) was too small. The correct input number was supplied, and the problem ran. "

} STATUS: Closed.

! (9/16/86) A traceback pinpointed the error in STGEN1. We double-checked input j and found that NSJUN (STGEN input card 6 field 3) was tno small. User input I

l C-15 4

4

-en , n,--~n . , - , , -- . -- - , , - . . , ~,v.-. . , , - -----.--n -- .

. ~. -- . . __ _ . _ ~ _ _ _ _ - __. .__ _ . _ .

C 4

correct number and problem ran.

(9/17/86) We called user and confirmed that he had corrected problem. Offered to send him discussion of STGEN arrays in forthcoming USER GUIDE. He said_he would not need that now, but would prefer to see future versions of TRAC have error checking improved in the STGEN component. Staff had previously looked 4 into this, and it appeared that placement of tnis message in the code was not-trivial. Examination of problem indicated that error message could be installed provided correct check parameter could be identified. It was suggested we pursue this with STGEN developer. Frank Addessio.

Fix will be provided in future coding.

SOURCE: GRS and KWU TFORM: 851204b

. DESCRIPTION: -"Need user-specified variable output in graphics file."

i STATUS: Closed. We indicated to users that they could currently provide additional variables to graphics flies by defining appropriate rtgnal variables.

I SOURCE: UKAEE TFORM: 860702a DESCRIPTION: "Wants mass flow ra.tes'to be output.in major edits."

STATUS: Closed.

' (6/4/86) Received letter (23 May 1986) from user stating that *We would like to a retterate SANDIA's request for mass flow rates to be output in major edits -

see January TRAC NEWS (Vol. 2 No. 1) page 9."

(7/16/86) Informed users that feature would be provioed in future coding, t

(ju186 TRAC NEWS). .

SOURCE: SNL .

. T FORM :- Not assigreed.

f DESCRIPTION: " Form losses are not written on dump file ."

! STATUS: Closed. We informed user that this user-convenience feature would be provided in future coding changes to EXTRACT. The objective would be to have EXTRACT produce a file that the user coes not have to subsequently re-edit.  ;

SOURCE:. Los Alamos t TFORM: 860317a DESCRIPTION: "Possible problem with TRAC graDhics when mass flow is oestred )

for first cell connected to VESSEL."

I STATUS: Closed.

1 (3/9/86) Received note with oescription of problem from user. He was trying to-obtain mast balance using TRAP to pick up integrateo values for upper plenum injection, hot leg flow, hot leg inventory and core exit flow. He found tnat he got large error in mass balance when hot leg flow in Cell 1 was integrated i with TRAP. Wnen cell 2 was used mass balance was very good. Also got strange result that integral of flow at cell 1 exit of not legs was much smaller than  ;

cell 2 exit of hot legs.

(3/17/86) Communicated problera statement to staff for review.

j (9/19/86) It was determined that problem may be attributable to TRAP postprocessor and will be fixed in future coding.

SOURCE: Los Alamos

! TFORM: 860705a i OESCRIPTION: *There is a difference between peak outer surface j temperature indicated in TRCOUT ano that given by signal variable 2SJ" STATUS: Closed.

(6/23/86) Received memo from user. He states:

"During TRAC analysis for SCTF Run 605 comparison I noted that there is a difference between value printed in TRCOUT known as

  • peak outer surface temperature of average rods
  • and signal value which 18 used as trip actuator.

l In this case trip was set at 1013K. At first agreement was excellent, but i

gradually signal value fell behind and finally tripped about 15s late."

A taDie was provided Comparing the TRC00T and signal variable values, r (7/1/86)--Talked with user and tsked him to provide input deck, output files, a noding diagrams, following procedure given in TRAC User's Guice Appendix A. If possible. Trouble-shooting would be much easter and his feedbacb would be easier to pursue to problem resolution. User indicated he used signal vartable j

25. but we did not have input deck used for errant calculation to confirm this.

(9/19/26) Debug materials (input. output, etc.) were not available to recreate .

this problem. Closed.

end f

i-t t

l f

1 C-16 i

i

,--. , . . e-- , ., , -- ,, , - . . . . , , . - , , , , - , . , . , , , , ---

C.1.6 CODE IMPLEMENTATION DIFFICULTIES SOURCE: JAERI TFORM: 860706c OESCRIPTION: "Would like to have first two time steps in future test problem output sent with magnetic tape transmittals of the code."

STATUS: Closed. User also asked to have LTSS FTN manual from Los Alamos for interpretation of coding. But, most importantly, he seid that in future test problem output JAERI would like output for first two time steps added.

(6/11/86) Received reouest from user at ICAP meeting in Erlangen. FRG.

(7/6/86) Legged JAERI's request. Assigned task to technician pool.

(7/16/86) First implementation of the new test problems initiated with EC12.8.

SOURCE: KWU TFORM: P60707h DESCRIPTION: "Would like to have all DATA statements collected into a single location in the TRAC code."

STATUS: Closed.

The request was one of several made by user during a presentation at the Friday, June 13th meeting in Erlangen. He said that they would like to see collection of all DATA statements (used to enter initial values *or variables in common blocks) into a single block data subroutine.

(7/15/86) Most DATA statements are already collected into a single location in a routine called ELKDAT.. We informed user that we would move other DAT' statements to that routine, if appropriate in future updates. Reported position in July 1986 TRAC NEWS.

SOURCE: KWU TFORM: 8607071 DESCRIPTION: "Would like to see TRAC coding upgraced to ANSI standard FORTRAN 77."

STATUS: Closed.

(7/6/86) This request was one of several made by user during a presentation at the Friday, June 13th meeting in Erlangen.

(7/15/86) We told user that we believed that TRAC was very close to being compatible with FORTRAN 77. Additional updates will be provided as specific inconsistencies with FORTRAN 77 are brought to our attentior.. Information reported in July 86 TRAC NEWS.

SOURCE: UKAEE TFORM: 860723a DESCRIPTION: "Wants Los Alamos to avoid using *before.

  • sequence. *selyank, aseipurge,
  • copy and = define in updates Los Alamos sends to them."

STATUS: Closed.

(7/16/86) Received copy of letter dated July 9, 1986 from users. They use an inhouse program, called PROMUS, to maintain their programs, instead of UPDATE or HISTORIAN. They said it would greatly aid them if we could avoid using the following HISTORIAN commar.ds: *BEFORE,

  • SEQUENCE, *SELYANK, *SELPURGE,
  • COPY and
  • DEFINE. We have not used any of these in our updates since we last used a
  • BEFCRE in update VMGUD12 of version 12.1.

(10/1/86) The " TRAC Programming Guidelines" was updated to include a statement that the six Historian commands of concern to UKAEE should be avoided. We informed users that in future updates we would attempt not to use these specific HISTORIAN directives.

SOURCE: Los Alamos l TFORM: 860918a DESCRIPTION: " Code should warn user when loaded with core memory preset to zero."

STATUS: Closed.

(9/18/86) Entered problem into trouble-form tracking system. Created update for deck TRAC, the same as created for M002 in September of 1985. Added additional comment cards to inform code developers purpose of coding and that loader will find ACKLOD and ICKLOD " undefined" during construction of code, and that is acceptable.

NOTE: We have found that if core is preset to zero, users will not get "zero-time edit."

(9/19/86) in MOD 2 Problem resolved with ident WRNLODT in EC13.0 and with ident WRNLOD2 version 3.1. It warns user that results nay differ from results at Los Alamos, if core memory is preset to zero. Change submitted to code manual to discourage users from presetting memory to zero, end C-17

C.1.7 OTHER CONCERNS SOURCE: ICAP TFORM: Not assigned.

DESCRIPTION: When would the TRAC User's Guide be available?  !

STATUS: Closed. {

Users were informed that the final document underwent internal review and was sent to printer early December 1985. It was distributed in early February 1986.

SOURCE: ICAP TFORM: Not assigned.

DESCRIPTION: How do you use the telecommunications capability afforded by the VAX UPDATES utility?

STATUS: Closed.

Information pertaining to VAX UPDATES was given in the October 1985 issue of TRAC NEWS. We encouraged users to examine that specific issue and call us if they had any further Questions.

SOURCE: ICAP TFORM: Not assigned.

DESCRIPTION: How are error-correction update sets sent to users and how outckly can they be sent?

STATUS: Closed.

We informed users of the transmittal mechanisms availaole. The present methocs were: (1) hardcopy listings (fast. error prone), (2) mag tape (slow). (3) IBM (or Apple) diskettes (mooerately fast) and ( 4 ) telephone mooem transfer from our VAX UPDATES uttitty (almost immediate).

l SOURCE: UKAEE

[ TFORM- Not assigned.

DESCRIPTION: How long until M002 is released, and what wie) it contain?

STATUS: Closed.

We responded that MOD 2 would incorporate all model improvements deemed necessary as the result of code assessment activities performed with MODI under the International Code Assessment and Applications Program (ICAP). In addition, generalized heat structure modeling capability and an improved constitutive oackage should be available. The approximate release date given was dune 1987 SOURCE: ICAP TFORM: Not assigned.

DESCRIPTION: How close shculd test problems compare for successful implementation?

STATUS: Closed.

We explained to users that. " ideally " test problem output at different sites should compare exactly with calculattons performed at Los Alamos. However, if differences were noted in the last cecimal place, this agreement might be su*ficient and could be due to different hardware employed for handling round-off. Other differences in machine hardware and software might lead to slight (last decimal-place) differences as well. We also stressed that it was most important that some sort of a software-compare utility be employed to focus on differencec so that the true extent of any variance could be 4 determined. l

'A 1 SOURCE: INEL TFORM: Not assigned.

DESCRIPTION: How do you interpret TRCMSG and TRCO'JT files?

STATUS: Closed.

We informed users that the best available guides to interpretation of TRCMSG and TRCCUT flies are centsined in Appendixes I and G, respectively, of the TRAC Usea's Guide. There you can find complete annotated listings that explain the various output segments in a clear, concise,5and handy format. We encouraged users to carefully examine those appendixes to clarify their understanding of the output.

SOURCE: UKAEE TFORM: 360702c ~

DESCRIPTION: What details regarding the proposed TRAC "generalised 1-D component model* can you give us at this time?

STATUS: Closed.

(6/4/86) Received letter (23 May 1980) from user stating that "we note with interest in TRAC NEWS your comments relating to a new ageneralised 1-D component model", and would very much appreciate some details of this new feature." We informed user that we could not give any details regarding the proposed modeling feature as it was too early in the development stage.

However, it was envisioned that the component would incorporate all the features of the specialtred 1-D models and significantly enhance future C-18

4 I N- p

~k i - -

1.

1- 2

/ .

I vectorization Capabilities and run times.

SOURCE: 3NL TFORM: 860706a DESCRIPTION: What is range of time steps you have seen at Los Alamos for 1D calculations for slow transients?

STATUS: Closed.

(6/30/86) User inquired if we had cone calculations with a 1D representation of the VESSEL to save costs. Also, he wanted to know what was largest time step he could expect for 1D modeling durirg a slow transient like a LOSP.

(7/16/86) Informed user that the time step depends on what is happening in the system in terms of ecutoment or operator actions. For example, a valve that is controlled on e pressure differential that is rapidly changing Can inhibit the time step. In these cases, we have seen time steps as low as 0.03s. However.

when equipment actions are not rapidiv changing and the physical phenomena vary slowly, time steps as high as 5 to fCCs have been employed. Normally, when the time step appears to be unbounded, the analyst will place a limit on time-step (

size te give a reasonaole number of calculated data points over the duration of the transient (at least 500).

SOURCE: KWU TFORM: 860707b DESCRIPTION: Is it possible to model.the 3-D effects at the hot-leg ECC injecticn point witn TRAC-PF1/ MOD 17 If so, what mocel would you suggest?

STATUS: C)osed.

(7/6/86) The user's request was one of several mace curing a presentation at a Friday, June 13th meeting in Erl qgen. User had expresseo the same concern the previous afternoon and suggested that an alternative might be scught for hot-leg modeling oy perhaps using two 3D VESSELS or extending 3D modeling into the hot leg.

(7/16/86) Because of the lack of turbulence modeling, we did not recommend using the 3-D VESSEL to investigate multidtmensional flows in pipes. Besides to get any kind of detail, one would have to use very small cells, which would severely Itmit the time-step size. Also. TRAC did not allow the VESSEL to be turned on its side.

SOURCE: GRS and UKAEE TFORM: 851205b DESCRIPTION: What is being done to make code run faster?

STATUS: We respondad that scme coding was being rewritten to allow future vectortzation whenethe 30 SETS numerics enhancements become available (pessibly in MOD 2 ne MOD 3). For several slow transients. like steam generator tube ruptures and steady-state calculations, speedups by a factor of 3 to 6 had oeen observed with time steps increastng by an crder of magnitude and more.

SOURCE: UKAEE TFORM: 860717a CESCRIPTION: What ir the purpose of update ident TCHFFX in EC12.47 Tnis update changes a subroutine that had previously been " purged."

STATUS: We acknosledge user's concern and informed him that since we accumulate updates over a period of time'oefore we make a new version of the code. it is possible within one update or over several updates to replace a subroutine with a new subroutine, but also to have updates that modify the old subroutine. Of course the important question was whether or not the line replaced by the TCHFFX update was also fixed in the new version of HTCOR. It '

was. We had establisned guidel.ines on updating TRAC that Old not allow for purging of decks. From then on. replacatent of an old subroutine with,a new subroutine would be accompitshed by deleting the old subroutine and inse-ting'-

the new subroutine. Attempting to modify a deleted line would henceforth result in a warning message or error from HISTCRIAN. .'

end ' \\

I l

C-19

C.1.8 CODE DOCUMENTATION DEFICIENCIES SOURCE: INEL TFORM: Not assigned.

DESCRIPTION: Inconststencies in-manual STATUS: We informed user (stated in the preface to October 1985 of TRAC NEWS) that we plan to issue errata to the manuel as Attaenments to future issues of the newsletter. We encouraged users to help us identify any errors they find so-that we could continue to improve documentation.' We acknowledged that recent errors had oeen corrected for the final version of the TRAC-PF1/ MOD 1 4 manual that was released in early 1986. j SOURCE: UKAEE '

TFORM: Not-assigned.

DESCRIPTION: " Code documentation is incomplete, and user guidelines are needed."

STATUS: We notified user that final code manual in binder format would be

-distributed in early 1986. A new document, the TRAC User's Guide, was at the printer and would be sent to users (in binder format) in danuary 1986. Both documents helped fill the need addressed by the users concern. In addition, user was told that future-issues of TRAC NEWS should serve to provide additional code documentation requirements. We also let the user know that if other documents become necessary, we would provide them as funding ortorities permitted.

SOURCE: INEL TFORM: 8607C1a SESCRIPTION: "We are experiencing some difficulties with trip-control-trips m'd control block 101."

'r&TUS: Closed.

O ~ ci/86) User was trying to do something with the NPA at Kirtland. He was ve r vg to use a TRAC input ceck supplied by Los Alamos. The code sted because ci wree error with the trip-controlled trip input. .ITNT was one, and it had to be 6. least two. User was going to try a positive two so that it would be an "or" (ie, a summer) situation rather than an."and" (ie, a multiplier). We said we would look at the input deck to figure out the problem. We ran the deck and got the input error encountered by the user. We then examined the output to further understand the problem.

(6/12/86) We looked at the input deck and there appeared to be e problem with the trip input. First of all there were two trips inout witn the same TRIPID (ie. IDTP = 213). Second there were-two trip-contro11ed trips input. However, both trip-controlled trips have INTN eoual to one, and the minimum value that was acceptable to TRAC was two. The IDTN = 219 trip was commented "FOR HPI."

It need to have two trios to sum. Eut it only had the IDTP=212 trip, which was a low pressure trip for HPI. The IDTN . 2559 trip was commented "FOR TSV AND MFW", it also needed to have two trips to sum. It only had the IDTP = 111 which was a reactor power trip that also tripped on low pressure. IDTP = 111 tripped on a low pressure of 130.E5 Pa, while IDTP = 212 tripped on a low pressure of 11.135E6 Pa. We examined our original deck created and run at Los Alamos.

(6/13/86) We reviewed the input deck and talked to previous users. Tne TMI input ceck that was sent to INEL was a holding tank deck that was used to keep several different sets of trips and controls in. The easiest change that would result in a running deck was to change INTN to two for both of the-trip-controlled trips and to add the IDTP=111 trip to botn cf the trip-controlled trips. If INTN was positive, then the trips would be summed.

(6/16/86) The proposed fixes wcrked. We ran the steady state out 14 time steps. We Called user, and he indicated ,that he had the problem running on Kirtland machines, and that he would try the Los Alamos machines. We gave him a short "how to" on how to run interactively on the Los Alamos Crays.

(6/19/86) User called about trying to transfer files to the VAX from an IBM PC.

He said he was having problems with the land lines. We told him we would check into it. He balled back and we told him about KERMIT. User sata he would try a dedicated 4600 baud line and see if that was any less noisy.

(6/20/86) User called and said that he had got a version of KERMIT that was supposed to help them transfer files to the Kirtland machine and that they had used that to transfer the file that he wanted to transfer.

(6/26/86) Responded to user's request for help with using control block #101 (function of one independent variable." He was getting message " array dimensioned for O pairs" and TRAC didn't like that. Talked with other Los Alamos user. He offered his experience on " things to check," when using tnis control block type. We called user and talked with him about control block

  1. 101. Page 142 of code manual indicates that table pairs are required as input. There are two other places in input where numbers must be consistent:
1. Page 237 Card Number 1 (control block cards), fourth field (variable ICB2). He had eleven table pairs, so, as last sentence of variable description indicates this entry must be "the number of entry pairs in the control block's function table" for ICBN (control-block operation number) =101. For his C-20

control block, therefore. ICB2=11, 2. Page 227 Main-Data Card 7 third field (variable NTCF). Reading the variable description compels the user to input "the total number of table entries." This is two times the number of of table pairs. Therefore, if John only had one type 101 CB with 11 pairs. NTCF would be 2 x 11 = 22. However, he had another CB with table pairs. so be must add 2 times those pairs to 22 to get the total number of table entries which is NTCF. He had NTCF = 0. That is why he got error message.

(7/15/86) We informed user that enanges would be made to code manual to make it easier to locate this information.

SOURCE: INEL TFORM: 860701b DESCRIPTION: "Had difficulty modeling ANS+20*4 decay heat, determining what new azimuthal flow areas to input when number of sectors in VESSEL model was reduced, and interpreting HSX values in the input."

STATUS: User was instructed to obtain the 23 9.~nn decay heat model from the TRAC user's Guide and multiply 1.2 times the energy fraction foa each group and input the result into TRAC. Instructions were also given net to change the decay constants. Additional user guidelines will be provioed which accross these concerns in future changes to the TRAC User's Guide.

(6/12/86) User was attempting to put togetner the TRAC input deck for the 2D/3D USPWR calculatten that INEL was doing for Los Alamos. He was trying to figure out how to use a signal variable to multiply 1.2 times the ANS decay heat that was being calculated by the TRAC default model. We told him to use the 23 group decay heat model in the TRAC users Guide and multiply 1.2 times the energy fraction for each g oup and input the result into TRAC. We also told nim to leavo the decay constants alone.

(6/16/86) User called about the USPWR deck, specifically about the FA-T in the core barrel region. The FA-T in the core barrel alternated from 0.1 to O.9.

Since he was reducirg the numbea of theta sectors from 8 to 4 he was interested in which FA-T be should use for his 4 theta model. Our suggestion was 0.1 since that would give him about the right flow loss in the theta direction for the core barrel region. User also indicated that he had f ound a place in the VESSEL on level 1 where tne HXS input had repeat parameters of r 14

.019 r 18 .136. He said tnis did not make sense for an eight-theta VESSEL.

and he thouDht it probably should be r 24 .019 r 8 .136...

(6/26/86) User called and asked if upper case was ok in TRAC input. For example 1.5E-20 instead of 1.5e-20. We told him on our 7600 system, and apparently theirs, no. Our Cray compiler accepts upper case statements (tested with small program).

(7/15/86) We informed user that additional user guidelines will be provided whicn address these concerns in future changes to tne TRAC User's Guide.

(1/14/87) Additional guidelines were proviced as attachments to the January 1987 newsletter.

SOURCE: INEL TFORM- 860704b DESCRIPTION: "How are GRAVE oetermined for parallel connections to TEES?"

STATUS Closed.

(6/24/86) User said he wanted to verify his understanding of pages iv-5 and tv-6 of TRAC User's Guide. His model incorporated a TEE with a parallel branch tube connection. We told him that page iv-5 of guide explained tnat for a parallel TEE, the GRAV was computed as follows:

GRAV = change in elevat.on/ length of flow path change in elevation is the difference in cell center elevations.

flow path length = DX (branch cell)/2 + DX (junction Ce11)/2.

He was a little confused by Figure iv-2 depicting parallel TEE configuration.

It showed the branch cell edge lining up with the junction cell (jce11+1) edge and a dotted line going from jcell center to branch cell edge. We told him the branch cell connection was centered at the junction cell center.

(7/16/86) The TRAC User's Guide indicates how the GRAVs are determined for TEES. As a help to future readers, we decided we would provide an example in Section IV-2a which would clearly illustrate GRAV calculation for TEES.

SOURCE: Los Alamos TF0WM: 860704c DESCRIPTION: "IVSSBF option does not work."

STATUS: Closed.

(6/24/86) Received memo from user. He stated:

"In TRAC analysis for the SCTF Upper Plenum Study it was suggested that I model the upper plenum as a 2-D peessure VESSEL with fills at the bottom representing the inlet condition (core exit) and breaks at the top representing the outlet condition (hot log entrance). This sounds like great stuff for testing a core model, but it doesn't work. I asked some of the old timers about this and the concensus was that it has not been a code feature for at least five years and possibly never was in a released version of the code."

(9/3/86) The option had been added to the code several years ago and tested.

The option went into version 9.7. The option provides for the first and last axial levels, the cells therein, to act as ftxed FILLS / BREAKS. As such, they I provide a fixed pressure boundary (BREAK) and fixed velocity boundary (FILL - I l

l C-21

only the axial velocity makes any sense). The input fluid conditions (TL. TV.

ALP. P) provice the necessary state conditions for the fluid flowing into the rest of the VESSEL component. The axial velocittes on the level acting as a FILL are fixed in time (the code apparently allows for P and VL and VV to be functions of radial and azimuthal location). As such, we did not think that the option would work in the user's specific application because he would like to specify the presbure and flows as functions of time as in the normal FILL and BREAK input specifications. Several staff indicated that IVSSBF once worked and was used by others to adapt TRAC for use in analyzing flow in oil wells.

Further review and discussion brought about the discovery that the VESSEL component reovires at least one source connection even if IVSSBF is non-zero.

We tried a test case with a non-zero IVS$8F and one VESSEL souace connection and the problem ran. After much discussion and testing it was decided to retain

  • IVSSBF option and update the documentation to instruct user how it should be employed.

SOURCE: KWU ,

TFORM: 860707c  !

DESCRInTION: "They would like to have a singlo, flow-regime-dependent 1 two-phase friction muittplier."

STATUS: Closed.

(7/6/86) This request was one of several made by user during a presentation at the Friday. June 13th meeting'in Erlangen.

At the meeting we had indicated that code has a cruce " flow regime map" that evaluates friction differently depending on void fraction.

(7/15/86) Upon further review it was determined that this capanility already exists. The way the coce partitions wall friction implies a two-phase multiplier. Additional description of this will be contained in documentation of TRAC models and correlations.

SOURCE: INEL TFORM: 860317e DESCRIPTION: "How do you get FILL table evaluated during steady state?'

STATUS: Closed.

(3/11/86) Recetved call from user with Questions. He was doing a secondary-stoe main-feedwater small break analysis of a Semiscale experiment using as base model an input deck received from SNL. He ran steady-state calculation with no trip control on FILL table and using time as independent signal variaD10. He could not understand why table was not evaluated during calculation. We referred him to page 110 of draft manual for dtscussion of steady-state solution metaodology. There are seven major differences between transient and steady-state calculational schemes. Difference number 7 states:

" Control-procedure component actions are evaluated ouring the steady-state calculation Only if they are controlled by the trios and their trip is ON.*

Dtfference number 3 states:

"T"1ps with positive identification numbers are not evaluated during steady-state calculations. Thus, even though conditions may exist that would cause the set status of a trip to change, the set status of the trip remains constant at its input value during the steady-state calculation." The upshot of these statements was that in order for his fill table to be evaluated during the steady state, he had to set it up so that it would be trip controlled with a negative trip id number. We Conveyed this information to him. He also asked whether or not there was a mecnanism employing trips and controls to save the value of " time" when a given trip was set to "on" thereby allowing use of that value of time in other control blocks or trips. He had already devised a rather complex strategy for accomplishing this and was just wondering wnether or not we had a special signal variable, trip or control Llock for doing this.

We told him we were not aware of such a specific feature, but several strategies were possible for accomplishing this, which was the power of our flexible control capability in TRAC.

(9/19/86) We informed user that a new FILL option (10) will be available in MOD 2 that snould provide control whether or not the calculation is steady state.

SOURCE: UKAEE TFORM: 860702b DESCRIPTION: "Many UK users are having difficulty understanding and using trip and control procedure input "

STATUS: Closed.

(6/4/86) Received letter (23 May 1986) from user Coddington stating that *Many new code users (and more experienced ones for that matter) still find extreme difficulty in understanding the TRIP and CONTROL procedure. That is both the input required when setting up the TRIPS etc and that within the component input, e.g. ' Rate factor tables,' etc."

(7/5/86) Minutes of ICAP meeting showed that JAERI had problems with trip and control use, as well, and indicated user guidelines would be most helpful.

(9/19/86) Closed. Changes to the TRAC User's Guide will be made to make trip and control modeling easier as specific problem areas are identified.

C-22

SOURCE: UKAEE TFORM: 860703c DESCRIPTION: " Requests further information regarding stratified flow modeling in TRAC-PFi/ MOD 1."

STATUS: Closed.

(6/24/86) Loggeo letter from user dated 20 May 1986 with attachments describing TRAC-PFi modelling of Stratified Flow and listing 4 potnts which require further clarification from the code developers "since they are not discussed in the user manual." These were as follows:

1 A critical velocity criterton is used to determine stratified flow regime bouncary. but addition of extra 11 Quid level gradient head term to licuid velocity equation is unaffected by whether or not this criterion is satisfied.

2. .The additional head term is only added when the inclination of the pipe is less than 0.57 oegrees. However. stratified flow interchase friction is used for angles up to 3C degrees. Please clarify.
3. The basic test for stratification is due to Taitel and Dukler. According to this reference (see letter), an extra factor of ROOT-PI would be expected in the exp*ession for critical gas velocity in TRAC. It is not clear wnether the omission is intentional or an error.
4. Two other tests for strattfication are made; one on 11oute velocity (for void fractions .le. 0.51 ano one on gas velocity. Please explain.

(7/86) Information will be provided in models and methods document.

SOURCE: KWU TFORM: BEO70?f DESCRIPTION: "Would like to have cetailed description of TRAP and EACON system routines."

STATUS: Closed.

(7/6/86) inis reauest was one of several made by user during a presentation at the Friday. June 13tn meeting in Erlangen. Changes to future TRAP and EXCCN cocumentation will be made to provice this information.

SOJRCE: GRS TFORM: 860708a DESCRIPTION: "Would like some gJidelines regarding vent valve modeling."

STATUS: Closeo. Additional user guice11nes will be previoed which address these concerns in future enanges to the TRAC User's Guide.

(7/8/96) Request from user logged. A study was presented by user at the ICAP meeting June 1986 in Erlangen FRG. For that study, involving reficod onenomena, they had to impose a O.025s maximum time step. because any larger caused the code to bomb on their Amcani computer. Also. there were seve*e nign-freauency vent-volve osctilations. He expressed concean that they could not model inertia of vent valve, replacement of water when vent valve was cpening, or condensation effects when steam hit downcomer liquid. He indicated that tney had not turned water packer option on. At the meeting we said that we would review our B & W criculations to see if we saw something similar. User said that he would prepare some information regarding vent valve difficulties and send them to us. JAERI had remarked that problem might be due to possibly too small a bubble diameter downstream of vent valve in oowncomer--a condensation problem.

(7/9/86) Concern was expressed that accurate specific guidelines would be difficult to provide in the near term. In order to develop accurate guidelines for vent valve medeling, experimental data on behavior of vent valves during severe condensation transients must be obtained. If data could be obtained, we wou1D need to analyze them and the experiment in depth.

(9/19/86) We informed user that additional user guidelines will be provided that address these concerns in future changes to the TRAC User's Guide.

end C-23

a > - ~ -- .

s e

C.2 PROACTIVE FEEDBACK FROM ICAP PARTICIPANTS C.

2.1 INTRODUCTION

In this section we present 22 of the major "proactive feedback" activities supplied by ICAP-involved users during FY86. These activities represent '

activities in which ICAP participants became actively' involved-in the problem-resolution process. As with the " reactive" feedback of problem and concerns given in tne previous section, proactive f eedoack was catalogued into -

our " problem-tracking'_. process as it was received.

We have divided this.section according to.the type of code activity the user was involved in. The major areas are user-supplied code corrections, user-supplied code improvements.' user-supplied code user-conveniences and user-supplied code guioelines. Table C.2 gives the total number of user-reported problems In each area.

TAELE C.2' Area of Activity -Number Presented Code Corrections 14 Code Improvements 1 User-Conveniences 2 Code Guidelines 5 TOTAL 22 Each specific activity toentifies the source, a "tform* numoer used to identify

, a specific activity in the problem-tracking process, the problem description.

1 and the status of the problem. The status not only indicates the resolution of the problem, but also gives the reacer a chronology of events and user contributions leading to that resolution.

C.2.2 USER-SUPPLIED CODE CORRECTIONS SOURCE: GRS and Los Alamos TFORM: 851212b OESCRIPTION: 'Divtsion by zero occurs when void fraction almost 1.00 (within 1.35 e-13) and RIDREG=6 (natural or forced convection to single-phase vapor) for code with EC12.5 implemented. Algorithm apparently accepts RIDREG=6. but then loops back to coding for RIOREG=2. At that point. logic causes TCHF to be set = to TSAT and subsequent cenominator of (TCHF-TSAT) causes division by O problem."

STATUS: Received message and explanation from user who was informed that we would distribute information to staff for review. After staff review. we found

, that the exact location in code where TCHF is set = to TSAT is at HTCOR 477 The diviston by 0 occurs at HTCOR 494 Got necessary input and output files from user to recreate problem. Fix suggested by staff was tested and found to remedy the user's comolaint. Fixed with error correction FXHTCOR in EC12.5 and error correcticn HTCOR02 in EC12.6.

SOURCE: EXXON and SNL t TF02M: Not assigned. '

DESCRIPTION: *Some uninitialized variables in code still exist. All problem

' variables not properly initialtzed before first time step (e.g. . GAMMA)."

4 STATUS: The NS00 integer variable in subroutine DMPIT. which was causing a 1 problem to outside users. Will be preinttialized in a future error-correction set. Currently, users should attempt to use a compilation option which presets i memo"y to negative indefinite if at all possible.

! SOURCE: UKAEE i TFORM: 860102a DESCRIPTION: " Variable ISTDY is not defined prior to entering routine IVSSL so

, that with a machine that initia11ses variables to zero RPOWR is< set to the

initial reactor power RPOWRI. This negates the use of the name115t variables IPOWR and-TPOWR to turn the power on in a steady state calculation - since it is always ' turned on' at the start of the run.'

STATUS: Closed.

4 (1/2/86) Received letter addressing problem from user (dated 12 December 1985).

They suggested that ISTDY should be set equal to STDYST in routine INPUT'or-

,some other suitable point.

C-24

1 (1/3/86) Sent letter of acknowledgement'to user and. thanked him for his problem statement and also for his suggested solution. (Not often did we get problems from users with suggestions on how to fix them so explicitly attached, as well!)

(7/86) Error corrected with update NOISTDY1 in EC12.8.

SOURCE: KWU TFORM: 860327a DESCRIPTION: "The torque T-sub-f modelling in oUMP not accurate" STATUS: Closed.-

. (3/18/86) Received onckage with cover-letter and IBM floppy diskette from KWU containing proposed code enanges.

(3/27/86) Drafted memo-to staff re KWU input and incorporated into CODE PROBLEMS section of TRAC NEWS (April.1986 issue).

(9/19/86) User was informed (October issue of TRAC NEWS) that changes would be provided in future coding. Initial review indicated that the user's $uggestion that frictional toroue used in the calculation of the pump speed should oe a constant with the sign dependent on the direction of the rotation may not have

been appropriate. If the pump were initialized at-zero speed. the constant-l frictional toroue would tnen cause the pumo to accelerate away from zero.

SOURCE: UKAEE TFORM: 860702e i DESCRIPTION: " Format statement in PREINP will not complie."

STATUG: . Closed.

(6/4/861 Received letter (23 May 1966) from user.providing us with a list of 4
  • errors" that were corrected at Winfrith to get code to run on their Cray. The error descriction was as follows:

" Correction to a FORMAT statement in subroutine PREINP. The FORMAT statement in 0408 will not comptle on our CRAY computer at Harwell.

53H should be replaced by 54H in the hollerith Count."

(7/2/86) We also had noticed problem when we used our CFT114 compiler. We had corrected tnis preolem with EC12.7 update ident TKERRt.

SOURCE: UKAEE TFORM: 860702f t DESCRIPTION: ' Missing

  • on the ENDIF in ident IFDEFC."

! STATUS: Closed.

(6/4/86) Received letter (23 May 1986) from user providing us with a list of

" errors" that were corrected at Winfrith to get code to run on their Cray. The error aescription was as follows:

, " Missing

  • ENDIF in IDENT IFDEFC line 0040.*

(7/2/86) We corrected this problem with update toent Fr.END in EC12.S. The correction was made in subroutine R1 MACH.

SOURCE: UKAEE TFORM: 860702g DESCRIPTION: " Problems with undefined variable KK when no Control block

, data is entered, and Cray core is preset to zero."

! STATUS: Closed.

(6/4/86) Received letter (23 May 1986) from user'providing us with a itst of i

  • errors" that were corrected at Winfrith to get code to run on their Cray. The error description was as follows:

'KK not defined in line 0349 of RCNTL. when no control block data is entered. When CRAY core is set positive indefinite this causes a failurs with message ' insufficient space for control block table

, storage.' If core is set to zero there is no effect."

4 They indicated that the corrections "is to move lines RCNTLO354 to RCNTLO355 to before RCNTLO349."

(7/15/86) Problem resolved with update PCODER in EC12.8 which used a slightly different approach to define KK, We did not have the problem on our machine.

3 because we preset core to negative indefinite, l'

SOURCE: UKAEE TFORM: 860702n i DESCRIPTION: "Some machine-dependent differences require that certain 4

variable names to be changed in subroutine REVSSL.*

f STATUS: Closed.

) (6/4/86) Received letter (23 May 1986) from user providing us with a list of-t

  • errors
  • that were corrected at Winfrith to get code to run on their Cray. The error description was as follows:

"In subroutine REVSSL the reference to A in line REVSSLO509 should be changed to I A. Also the use of WIARR in line REVSSLO759 should be

changed to WARRAY. This is to make data type and format compatible."

1 (9/18/86) Fixed with ident DTYPERJ in EC13.1.

SOURCE: UKAEE

, TFORM: 8607021 DESCRIPTION: " Missing RETURN statement in subroutine BITS

  • STATUS: Closed.

C-25

, - . _ . . _ _. . _ _m . _- _ _ - _ .u_._ _ _~. 2

.(6/4/86) Received letter (23 May 1986) from user providing us with a list of

" errors" that were corrected at Winfrith to get code to run on their Cray. The error cescription was as.folloss: j

"... missing RETURN statements following entry points OF1123(B, N) and i ON1123(B. N) in routine BITS (B. N)."

(9/24/86) Fixed.with ident BITSRJ in EC13.0.

SOURCE: UKAEE TFORM: 860702j OESCRIPTION: "IF test arQument may'be incorrect for definition of LCORE in

' subroutine TF3DE when there-is no downcomer."

STATUS: Closed.

(6/4/86) Received letter (23 May 1986) from user providing us with a list of

" errors" that were corrected at Winfrith to get code to run on their Cray. -The.

erro".

".. wedescription was as believe there follows:

is an ' error

  • in'TF3DE in the definition of LCORE.

-1e~

(1) .

IF(IZ.LE.ICRL.OR.12.GT.ICRU.OR.IR.GT.IDCR)LCORE=. FALSE.

should read (2)

IF(IZ.LE.ICRL.OR.12.GT.ICRU.0R.IR.GT.ICRR)LCORE=. FALSE.

in those instances where there is no downcomer."

They stated also that they "can see tne virtue of keeping (1) for a' general

  • vessel component, however if the vessel component is used to model a separats cffects experiment with no downcomer IDCR is zero and therefore LCORE is set to be . FALSE."

j (8/1/86) Update LCTFx in EC12.9 had corrected this previously observed 4 problem.

i SOURCE: UKAEE TFORM: 860703b DESCRIPTION: Code needs imDroved horizontal stratified flow.

STATUS: User wanted us to review work he and etners in tne UK had done that indicates e mtseing term in TRAC momentum equation for the liquid phase.

We logged a letter dated May 20 1986, with attachments oescr.'ing the 4 TRAC-PF1/M001 modelling of Stratified Flow.. Item i of the attachment to that letter contained the general form of the velocity equation and compared it to the TRAC equation. The neglected term was_ indicated and a simple case (the

" cam" problem" was presented which showed'the imoact (TRAC neglects half the-hydrostatic head). He stated "the error in TRAC is simply corrected. All that is needed is to substitute calculated level Otfference between cells. delts-h.

for difference in height between centers of pressure which is what TRAC currently uses."

He further elaborated on this with words of caution related to applicability.

l In August 1986, we called user with request for Dam Break problem input.

l There had been considerable effort expended prior to this reauest to attempt to duplicate the problem. we reviewed earlier file that a developer had put together on the misstng term in the TRAC momentum equations for horizontal flow. The explanation by UK on their proposed fix did not appear clear to us at the time. However, by reviewing the equations and the TRAC Coding we identifted the potentially necessary changes.

We tried to duplicate the reported test problem (a " dam break") that was used to first identify this problem and did not get the same reported results. TRAC

without any changes resulted in the deDression wave hitting the closed end of the pipe in less than 4 seconds. However. P. Black reported a value of 7.922 1 seconds in a letter to D.C. Leslie of Queen Mary College Oct 1. 1985.

We needed to know if user had modified TRAC to handle stratified horizontal flow in a square channel. TRAC can only currently handle stratified horizontal flow in a circular pipe. since the formula that relates water level within the pipe to the local void fraction and pipe diameter assumes a circular pipe' . We could see that it would also make method of characteristics solution much easier if the pipe was assumed to be square.

{

In a further-attempt to duplicate UK's results we tried to modify TRAC so that the dam problem was for a square channel. We still did not get the same results.

To proceed further, we needed their input deck. We called UK and asked them to send us their input deck or more description on their input model. We wanted to f resolve this problem, but before implementation of the proposed fix we had to be able to verify the improvement.

In September, 1986 we received the Dam Break input from user and attachments

! C-26

l

.l related to this study. 3 At our code problems review meeting a Oeveloper said that'the input deck sent from UK did not give results obtained at UK with their stripped down version, another developer indicated tnat we nad discovered the missing sort (pt) term (part of the reported problem) independently of the British, but the correction had not been made. We agreed to address this. problem Concurrent with the stratified flow Concern.

we met in October 1986 to discuss misstng sort (pi) term and attempted to resolve problem.-That meeting resulted sn scrutiny of previous momentum equation development' focusing on treatment of ' jump conditions" at liquid / vapor interface in the momentum equations and review of Ishit's treatment. Bottom-line was that UK was correct and missing d-alpha /dx term should be added to coding. Also, missing sqrt(pt) term should be added.

However, the momentum equations 00 treat the shear and vapor thrust terms-in the interfacial momentum transfer-terms (* jump conditions"). Also, it was believed that the uncertainty 1.n the interfacial shear. terms for stratified flow was as large as the error caused by the neglected term. Also, the perceived inadequacies in the coding were dominant'only in an extremely limited subclass of real thermal-hydraulic systems (notably the cam-break problem) and were not typically encountered in full nuclear power plant systems for which 4 the code was designed.

l Upon further examination in October and early November of'1986. we realized we did have the correct d-alpna/dz term. but our implementation in terms of liquid

, level height differences was not the same as that proposed by the UK.

3 Problem was resolved with UPTMCM3 in ident 13.3. Term was found to be in

, error, but not missing.

, SOURCE: UKAEE

, TFORM: 860721a 4

OESCRI.OTION: "The TEE joining cell flag MSCT used in FEMOM ts not set correctly for TEES which are steam generator subccmponents.'

STATUS: Closed.

. (7/16/86) Received letter (9 July 1986) from user providing us with a itsting of the-corrected coding.

, (7/21/86) Letter described problem as fo11ows: . .

i j

"The TEE .toining cell flag MSCT used in F CMCM is not set correctly for TEES which are steam generator subcomponents. This means that the values of Ox and GRAVM asscciated witn the first junction of a side branch of a steam generator TEE will deoend on the Component ordering in a deck, and are Unitkely to take 1

the intended values. MSCT should be set in subroutine STGEN1 as shown in the attached listing.'

The listing of tne corrected routine is with letter in UK letters notebook.. "

(9/24/86) Coding changes suggested by user were forwarded to staff further for review.

(9/29/86) Closed. Coding changes reviewed and additional FORTRAN necessary to

( 1mplement the proposed changes were installed with ident MSCTCR in EC13.1.

SOURCE: SNL TFORM: 860827b DESCRIPTION: User found some undefined variables and typos.

STATUS: Closed.

(8/20/86) Received call from user who identified several undefined variables that he thought were causing code to blow-up. They were:

' 1) IFVTU

2) LOOPCininsubroutine subroutineCONSTE ACCUMi- --'After He was correct checking in that here. it was it was undefined.

determined that his update system was not working, j' 3) Two typos in update UFHGAM -- We had already found them, and an update was in.

4) A typo in line CCFL1.33 -- We had already found it, and an update was in.
5) CPow and ENINP in subroutine WPLEN -- He was right in that they are undefineo, but they were in coding that was never executed.
6) NCPU in subroutine GRAF --

After checking here. it was determined that.he had omitted update ident SYS00.

Therefore, four of six problems were immediately resolved by developer.

(9/19/86) UPDCLN in EC12.9 and CCFLX2 in EC13.0 provide corrections for problems 1 and 5.

SOURCE: Los Alamos TFORM: 860905a OESCRIPTION: The user gets fatal error using forward and reverse friction

, factors (KFACs) as instructed in the Code Manual for the N#RC1=2 option. The problem arises when user has TEE side-tube connected to PRIZER with different directional numbering for cells.

C-27

STATUS: 'Open.

l

'(8/25/86) Received problem from user in writing as described below:

"A problem arises with the use of NFRC1 option. As described in the recently released TRAC Code Manual. 'This variable can be used to force input of forward and reverse loss coefficients to all one-dimensional components . . . The f irst array provides loss coefficients that are used with positive velocities and the second array provides loss coefficients used with negative velocities.'

"The problem arises when the forward and reverse loss coefficients are i different and the adjoining components do not have the same directional numbering. Consider, for example, the case tnat I encountered where I have a hot-leg TEE with -the secondary representing the surge line and a pressurizer attached to the end of the TEE. The pressurizer, as is our convention, has cell 1 at tne top and celt 9 at the bottom adjoining.the-secondary of the hot-leg TEE. The secondary TEE convention is fixed: the two-cell secondary has cell one at the junction with tne primary leg of the TEE and cell 2 at the interface with tne primary.

y

" Going from the TEE secondary to the pressurizer with positive velocities, the K-loss is represented by a sudden expansion. Going from the pressurizer to the TEE secondary, the K-loss is represented by a contraction.

"If the Code Manual instructions are followed, we would have the following:

K-loss

! TEE secondary Pressurizer Forward about 1.0 about 0.5 i Reverse about 0.5 about 1.0 i "However, TRAC will give a fatal error if the Manual instructions are followed. It requires the matching of TEE forward-pressurizer forward and TEE l reverse-pressurtzer reverse. Witn this approach you ao not even know in advance whether the K-factor is gotng to be 1.0 or 0.5. Apparently, the code i should sense that the component numbering is not in the same direction or sense

! ano match the TEE forward with the secondary reverse and the TEE reverse with tne pressurizer forward."

'(9/24/86) At code problem review meeting it was agreed that this preolem would be fixec. .

(2/22/87) Problem resolved with ident VLVFIX1 in EC13.3.

end il i

d I

i h

I i

C-28 i

I l

l l

l C.2.3 USER-SUPPLIED CCJE IMPROVEMENTS SOURCE: UKAEE TFORM: 850913a

, DESCRIPTION: "Some critical flow model improvements are needed ...

l STATUS: User called and informed us that a colleague of'his l' had sent a set of changes they had made'to the code about two months ago.

l The user had met us in Italy and discussed these changes at a conference.

He wanted to know if we had received the code changes and what we thought of them.

The status of user.'s work was reviewed with development staff; We called user to give him response on the status of his request. Told him that some initial work had been done to assess the conclusions ne had drawn. Some.

preliminary work had been performed that indicated UK assertions might.not be correct, but more calculations were required to really know one way or the other. We informed user that we would follow up.on tne additional calculations and perform any fix that might be warranted.

4 He indicated that their calculations definitely showed wrong' mass flows wnen' flow is slightly superheated. The error occurs because of a poor choice of system entropy. The code incorrectly calculates flow state at cell edge. They believed this was a serious error in TRAC that should be Corrected.

User was pleased to know that there was work being done to address UK's 1

concern.

User's initial request was satisfied. However, additional concerns had developed relating to critical flow. A preliminary assessment-was performed of~

the choking-related requests sent by UK. We examined letter and attachments from user offering suggestions for code improvements and corrections.

We met with developer of the choking model in TRAC (now with'another division at the laboratory). We obtained suggestions from developer on how to pursue a

calculation to determine validity of user's suggestion. Preliminary estimate was 5 person days to perform hand calculations and furtner assessment of other suggestions.

We called user end asked how important the resolution of the choking condition i problem was at this time. We told him that it would take at least a week.to resolve. that we were staff limited. and that NRC approval would be required before further work.

He said that this was not affecting the progress of their work, but thought that the code needed to be improved in the areas they had alluded to. He alsc had more Comments on the choking model and after revtewing them would get back to me (botn phone and letter). He thought that perhaps a packege of work i 4 presented to NRC in this area would be better received once they had completed '

l their final review.

During the ICAP Specialist meeting at Erlangen in June 1986 e paper was presented by UK that presented their final review. The paper was entitled

" TRAC-PF1/M001 Basic and separate Effects Data." by C. Rtenards. His j presentation detailed some aspects of critical flow modeling with TRAC.

Some of the items of concern indicated by Richards are as follows.

Large nodalization effects evident. We need guidelines Guidelines should also give reasons for their use. Guideltnes especially needed for

, . critical flow modeling for SELCCAs.

i Two-phase critical flow model differs from HEM.

4 There is a two-phase undocumented feature in the code which indicates tha. some work has been done in this regard.

T SOUND subroutine has possible error and needs more accuracy.

He also noted that in the critical flow model in TRAC an iteration is performed to determine the characteristic of the matrix, but then never changes original conditions calculated by HEM.

Choking depends on alpha and L/D and the range of L/D determines the method for calculating alpha. He said that by increasing length of pipe a small amount (to L/D just over 1.5). flow was doubled in an example problem. This he found intuitively nonphysical.

He also alluded to some superheat problems. It was clear that we definitely needed more information and direct interaction with-the UK to resolve these l

C-29

., - -., , . . . . ~ - . . . . ._ . . -. .. .- . . _ _. . . .

^

i i

)

difficulties.

' ~

Further communication between UK and Los Alamos and dedicated efforts at both sites allowed much headway to be made toward rectifying this problem.

In September 1986.' update ident UPDSND, added to error-correction set EC12.9

i improvea superheated steam choking. The TRAC code manual was changed to reflect

.that improvement. UPDSND2 in EC13.0 and UKCHOK1 in EC13.3 provide additional.

4 corrections to.the' choking model.

l - The details of the work performed to resolve this problem were presented in the j INFORMATION section of October 1986 TRAC NEWS. We planned no further changes

,. to the choking model, end 1

+

h a

t I

'I i

5 i

i i 2

t 1 l i

?

i 1

1 l

4 4

)

t

)

i l

h' s

C-30

i. .w , ,e ---...m.,-. , - - - ,-e.,v v v -+ - - - - - - - - - + ,7vr---,  % . v,-w-.- --m-- , - + - , , ,rr-- -, , o--m-,4.-y- eyo ,w,-- -

e

I I

1 C.2.4 USER-SUPPLIED CODE USER-CONVENIENCES SOURCE::& KWU TFORM: 860327a DESCRIPTION: "Better job tdentification and Quality assurance neeced.

STATUS: Closed. Enhancements to job identification and qtaaltty assurance are ongoing and will be included in future coding as appropriate.

SOURCE: UKAEE TFORM: 860704a DESCRIPTION: "Would like to supply EXTRACT for TRAC version compatible with EC12.5."

STATUS: Closed.

We received a letter dated 28 May 1986 with listing of updates to perform the following:

1. Cope with steam generator components when forward and reverse FRICS are specified
2. Cope with the PLENUM component. In their version the index K is recefined. incorrectly. at the Deginning of subroutine MPLEN. (It is correctly set in the calling routine, REPLEN. they said.) Therefore, they have deleted the line. They also note that EXTRACT will not allow reconstruction of data decks incorporating dimensionless form losses (RFACs) rather that FRICS. They would like to get their hands on a version which allows this feature. if available. We decided to incluoe this feature in future EXTRACT coding' changes.

We did not intend, nor did we have resources, to maintain the version of EXTRACT ortginally sent to the UK. Any changes or improvements that the UK made to EXTRACT we were interested in, but we did not have the resources to incorporate into the coce. This was mostly due to the fact that we did not have a mechanism in place to treat the British developmental efforts in coordination with Los Alamos developmental efforts.

However, we had plans to develop a version-independent version of EXTRACT that would be contained on the same program library as TRAC is and would use the same Common blocks and input decks.

Work was underway to provide EXTRACT compatible with latest TRAC version. We were attempting to include KFAC treatment in EXTRACT coding changes.

end C-31

C.2.5 USER-SUPPLIED CODE GUIDELINES SOURCE: UKAEE TFORM: 860702d DESCRIPTION: "Some NAMELIST variables have not been documented."

STATUS: Closed.

(6/4/86) Received letter (23 May 1986) from user stating that "We note that.the number of NAMELIST variables has increaseo over tne years and there are therefore cuite a number that have not been documented. We have produced an internal note on these which you are quite welcome to if you think it will be of benefit."

(7/2/86) Searched files, and similar concern had be previously expressed by a Los Alamos user. Several parsons responded to this documentation inquiry. We cecided to update the code manual as soon as possible.

(7/6/86) Task assigned to technician pool.

(7/7/86) Received letter dated 30 June 1986 from user with his itst of TRAC-PF1/MCD1 Name11st variables. Gave list to tech pool to use with documentation task.

(7/16/86) We added new variable descriptions to the code manual.

SOURCE: UKAEE TFORM: P60702k DESCRIPTION: "Whe7 code is restarted and restart deck Contains TRIP data, it will use this data to reinitialize the TRIP status variables internal to the Code."

STATUS: The trips, control blocks and signal variables should not be included in the restart input decks unless they are being changed from the original definitinn. User guidelines will be updated to make this clear.

(6/4/86) Received letter (23 May 1986) from user providing us with an appendix describing the problem and offering warntnOs to users. Some excerpts follow:

"The aoove {1ack of status change) reselted from the fact that when the code is restarted, and the restart input deck contains TRIP data it will use this data to reinitialise the TRIP status variables internal to the code." Tney went on to elaborate on how cara input can partially alleviate this problem, but

"...may not be simple if for example the trip is in a holding state, i.e., if a trip set point has been reached but the problem time does not exceed the set point and delay time." The consequences of not getting the state exact were stated and recommendations for circumventing this problem were presented. It seemed that the recommendations that limit changing the trip cards for restarts imposed an excessive burden on the user, however.

(7/20/86) Review of feedback allowed for generation of new ideas for user guidelines. Tne trips, control blocks and signal variables should not be included in the restart input decks unless they are being changed from the original definition. We informed user that guidelines would be updated to make this clear. Recetved written comments on UK trip Qutdelines from staff reviewers and incorporate revised guidelines into October 1986 TRAC NEWS.

SOURCE: UKAEE TFORM: 860703a DESCRIPTION: "Wants review of UK paper discussing Momentum Losses in Pipes."

STATUS: Closed.

(6/4/861 Received letter (23 May 1986) from user providing us with an appendix that included the complete paper by P. Black. The paper focused on equation nonconservattsa and relevance of equation nonconservatism. A simple example demonstrates author's ideas.

(6/23/86) Copies of letter with attachment sent to staff for review.

(9/19/86) A detailed explanation of the momentum flux terms was prepared for the forthcoming models and method document. A simple review of the UK paper was given to user and is as follows. We agree, for abrupt expansions, that TRAC finite difference approximations give the correct pressure Change. And for abrupt contractions, the TRAC finite difference approximations overpredict the pressure drop.

The negative NFF option correctly adds a negative form loss for abrupt contractions to get the correct pressure drop. The negative NFF option adds no correction for abrupt expansions. For smooth area changes. TRAC currently requires a fine nodal 12ation to obtain the correct change in pressure. Future plans are for the development of a smooth area change model in addition to the abrupt area change model. Currently, an improved two-phase multiplier model for abrupt area change is being considered for future coding.

SOURCE: UKAEE TFORM: 860717b DESCRIPTION: "Wants review of Outdeline by P. Coddington and C. Richards entitled ' Momentum Losses at TEES.' the second in a series of papers that examine the fluid momentum or pressure losses predicted by TRAC."

STATUS: Informed user (in October 1986 TRAC NEWS) that momentum losses at TEES would be addressed in the Models and Methods document.

C-32

SOURCE: UKAEE TFORM: 860717c DESCRIPTION: "Wants a review of the guiceltne by P. Coddington and C.

Richards entitled ' Pressure Losses at TRAC-PF1 / MOD 1 Component Boundaries,' the third in a series o* papers that examine the fluid momentum or the pressure losses predicted by TRAC."

STATUS: Informed user (in October 1966 TRAC NEWS) that pressure.1osses at M001 boundaries Will be addressed in the Models and Metnous document, end n

C-33

I RELAPS REPORTED USER PROBLEMS Code performance problems reported to the RELAPS Code Development group by both foreign and domestic code users are described in this section. Each problem reported to the development staff is assigned an identifying number for tracking purposes. Each problem described below is preceded by the date reported, identifier, and code version involved. Advice on modeling may include all aspects of user discretionary input as well as novel or unique approaches to modeling specific situations. (Note: U. P. denotes user problem; U. S. denotes update submittal.)

  1. 1 CATEGORY OPEN - Work in progress.

12/17/85 Cognizant Engineer: RAR U. P. #86012.1 Last Entry: 06/03/86 R5/M2/36.02 An ICAP member reported several problems in the coding specific to the CRAY machine, as well as an error in the horizontal stratification liquid entrainment model at a crossflow junction. The updates the user submitted specific to the CRAY errors, have been checked out, and are referenced as U. S. #86002, and were incorporated with other updates which created RELAPS/M002/ Cycle 36.03. (See U. P. #86012.2 in Category #5). The horizontal stratification error is caused by an incorrect stratification detection check when the junction is crossflow. The ICAP member has suggested a modification that would lower the velocity used in the stratification check and thus allow the liquid entrainment model to be activated. The coding is under examination. This error is related to U. P. #86029 in this category.

01/08/86 Cognizant Engineer: RAR U. P. #86022.2 Last Entry: 01/08/86 R5/M2/36.02 An INEL user reported a problem with the two-step option where the code ran slower and had more time step advancements than the one-step option. The input deck was renodalized to valve off dead end volumes in the feed water system that was causing the repeats. When run at the Courant limit, both options now have the same number of advancements. When the two-step was run at a largar time step the two-step (1.21 CPU time / simulated time) still ran slower than the one-step (0.95 CPU time / simulated time). An update has been generated that will be included with the updates to create RELAP5/M002/ Cycle 36.06 and is referenced as V. S. #87002, that improves the velocity matrix used in the two-step (0.79 CPU / time / simulated time) runs faster. This problem is currently still under investigation to obtain faster speed. For the remainder of this problem, see U. P. #86022.1 in Category #4.

04/14/86 Cognizant Engineer: RAR U. P. #86023 Last Entry: 04/14/86 R5/M2/36.02 C-34

An INEL user reported a problem in trying to simulate a MARK I containment vent system in which the code obtained unphysical results due to problems in the choking model. It was noted if the user changed the area of the volume immediately downstream of a junction that-is choked, different flow rates would result. Due to problems with the choking model for the case of an abrupt area change junction, in the interim, we recommend the use of the smooth area change with an appropriate loss factor until a fix is formulated. The error is still under investigation.

02/06/86 Cognizant Engineer: CSM U. P. #86028 Last Entry: 06/03/86 R5/M2/36.02 An ICAP member reported a problem in which unphysical sound speeds were calculated in the presence of noncondensible gas. The equations in the RELAPS/ MOD 2 users manual and the coding for the sound speed in fluid containing noncondensibles were investigated. The manual reflects coding found in subroutine ISTATE for initial conditions and time dependent volume sound speeds. The unphysical sound speeds were calculated in subroutine STATEP. The solution methods used in ISTATE and STATEP to calculate sound speed are different. The investigation is continuing to find a common solution scheme that will produce physical sound speeds.

02/10/86 Cognizant Engineer: RAR U. P. #86029 Last Entry: 06/03/86 RS/M2/36.02 An INEL user reported a problem for an ICAP member in which a posttest analysis (ROSA-IV Facility Test SB-CL-05) was run and exhibited two code problems. The first problem dealt with having to use a discharge coefficient greater than one at the choked junction. The second problem was concerned with the horizontal stratification entrainment model at a crossflow junction. Both of these problems are still under investigation. The second error is now understood and is related to U. P. #86012.1. in this category.

04/17/86 Cognizant Engineer: RAR U. P. #86050 Last Entry: 06/03/86 RS/M2/36.03 An INEL user reported a problem in trying to execute an ATWS problem. The code continually fails with water property errors with the state near the critical point. This problem has been known for some time. (See U. P. #86003 in Category #7.) An update has been generated to repair this problem. The up6 ate adds coding to the thermodynamic properties subroutines (STATEP, ISTATE, and TSTATE) when the pressure is above the critical pressure. When this occurs, the thermodynamic properties are set the same for both the vapor and liquid variables, since there is only one phase. In addition, the phasic void fractions are both set to 0.5 and the phasic interfacial heat transfer coefficients are both set to 0.0. The code will now run successfully over the critical pressure, but routinely fails slightly below the critical pressure. The update is currently being tested on a simple ten volume push problem just under the critical pressure. Work is continuing on this problem. This error is related to U. P. #86057 in this category.

C-35

05/22/86 Cognizant Engineer: RAR U. P. #86057 Last Entry: 06/03/86 R5/M2/36.02 An ICAP member ran an ATWS - safety valve calculation and the code failed with a water property error. Investigation showed the failure occurred ,when the pressure reached the critical pressure. The input deck has been received, and work is continuing on this problem on the INEL CDC computer. This user problem is related to U. P. #86050 in this category.

06/23/86 Cognizart Engineer: CSM U. P. #86068 Last Entry: 06/23/86 R5/M2/36.02 An ICAp member reported a problem in a calculation of a steam line break (2 x 100%). A sudden decrease in break mass flow was detected. It was noted

^

that this haopened when the flow regime in the break volume became horizontally stratified. The flow in the break junction was choked all of the time. When the line was tilted approximately 15 degrees, no transition in the flow regime was detected, and the break flow curve was smooth. Investigation is continuing

! on this problem.

4 07/15/86 Cognizant Engineer: CSM U. P. #86076 Last Entry: 07/15/86 i R5/M2/36

. An ICAP member reported unphysical results in executing a study to investigate the code's capability to predict CHF-location and post-dryout wall temperatures. The following results were noted:

1. The predicted critical heat flux shows numerical discontinuities in subroutine pREDNB.
2. The disagreement on the CHF problem may indicate an error in the implementation of the correlation.
3. The Biasi correlation, developed for round tubes, is inadequate for fuel bundles.

A magnetic tape was received which contained the user input deck. This input i deck has been loaded on the INEL computing system, and progress is continuing on this user problem, f

08/18/86 Cognizant Engineer: RAR/RJW 1 U. P. #86086 Last Entry: 08/19/86 i RELAP/SCDAP/ TRAP-MELT An INEL user reported a problem in trying to simulate a TMI plant in which the code failed in the heat transfer correlation subroutine CONDEN because the temperature was too low (274K). Modifications in the mass transfer model at high void fraction to the integrated code (RELAP/SCDAP/ TRAP-MELT) have been j developed, and the problem runs successfully. These modifications are being i considered for RELAP5/M002. Investigation is continuing on this user problem.

C-36 l

08/27/86 Cognizant Engineer: DMK U. P. #86089 Last Entry: 08/27/86 RS/M2/36.04 l

An INEL user reported unphysical results in a calculation where predictions of sonic velocity contain large, rapid oscillations while corresponding pressure and density remain smooth. Work is continuing on this reported user problerc.

09/23/86 Cognizant Engineer: DMK U. P. #86096 Last Entry: 09/23/86 R5/M2/36.04 An ICAP member reported a problem with the code where if heat slabs are connected to a real volume (boundary volume) it causes problems. Heat is supplied equally to both liquid and vapor phases even if one of the phases is not there. Investigation is continuing on the reported user problem, and it was found that the code developers agree that a heat flux boundary condition is in error. The fact that this error occurred for a " normal" volume (as opposed to a time dependent volume) implics that the error may be a general error for all heat transfer processes. However, this is unclear without detailed studies of the coding. The user supplied resolution is erroneous in that heat transfer is apportioned to each phase based on the fraction of heat transfer surface area in contact with that phase. Consequently, the parameters of importance in VOIDG (IV0L) for the vapor phase and VOIDF (IVOL) for the liquid phase instead of QUALA (IVOL). Final resolution, however, will be precise in that the correct phasic wall surface area will be used as described in the basic models discussed in the RELAP5/M002, Volume 2 Users Manual, pp 101-104 and Table 1, pp 105-114.

C-37

)

FY-87 REPORTED USER PROBLEMS CATEGORY #1 - WORK IN PROGRESS 10/06/86 Cognizant Engineer: RAR/ TAC U. P. #87001 Last Entry: 10/06/86 RS/M2/36.04 An INEL user reported a problem in running TYPPWR3 with the semi-implicit option.' Unphysical, rapid oscillations in heat transfer related quantities in the gas and fluid in the core starting at 50 seconds of simulation time. This may be producing the rapid void oscillations that probably affect the time step control. It only occurs at large time steps (e.g. Courant limit). The oscillations can be minimized by cutting the time step or by using more old-time weighting in the under-relaxation of the interfacial coefficients.

Investigation is continuing on this reported user problem.

10/13/86 Cognizant Engineer: RAR U. P. #87002.2 Laat Entry: 12/11/86 RS/M2/36.04 An INEL user reported a problem where the code fails on a generic restart (proprietary deck). It obtains numeric problems in subroutine CONDEN. It returns a negative to real or double power error. An update has been prepared that will set the FAIL flag to true and write a message, so that the code will stop gracefully. This update will be incorporated with other updates to create RELAP5/M002/ Cycle 36.05, and is referenced as U. S. #86012. The user then cut the maximum time step, the code ran, however temperature oscillations occurred. Investigation is continuing on this reported user problem.

10/20/86 Cognizant Engineer: RAR U. P. #87004 Last Entry: 10/27/86 R5/M2/36 An ICAP member reported a problem in trying to simulate the Marviken Jet Impingement Test 11 (saturated steam flow) and the Marviken critical Flow Test 21 (subcooled and two phase flow). The calculations indicate that the code overpredicted the experimental flow rates by 10 to 25 percent.

Investigation is continuing on this user problem.

10/24/86 Cognizant Engineer: RJW '

U. P. #87008 Last Entry: 10/24/86 R5/M2/36.02 An ICAP member reported a problem where it appears that the calculation of RKSLOB in subroutine RRKIN is believed to be wrong. Investigation is continuing on this reported user problem.

11/24/86 Cognizant Engineer: RAR U. P. #87009 Last Entry: 11/24/86 R5/M2/36.04 C-38

An ICAP member reported a problem in running a single loop calculation that contains a pump component. The mass flow rates in the pump junctions become quite oscillatory after 9 seconds. This was calculated using the semi-implicit i

hydrodynamic scheme. The user suggested to the RELAPS developers to make the pump head implicit in the flow rate in the semi-implicit hydrodynamic scheme, i

and transmitted updates to us which do this. This was already done for the nearly-implicit hydrodynamic scheme. Work is continuing on this reported user problem.

12/09/86 Cognizant Engineer: DMK U. P. #87015 Last Entry: 12/09/86 RS/M2/36.04 An ICAP member reported a request for a new capability in the RELAP5/ MOD 2 code to allow the CSUBV table for motor and servo valves to be in SI units. At present, only British units are allowed. This option is being considered.

12/16/86 Cognizant Engineer: CSM U. P. #87017 Last Entry: 12/16/86 RS/M2/7 An ICAP member noted oscillations in the void fractions and discontinuous jumps at the quench front region and nucleate boiling on a calculation. The user submitted an update to modify TGSAT to avoid the oscillations. This update is being considered for future cycles of the code.

12/16/86 Cognizant Engineer: DMK U. P. #87018 Last Entry: 12/16/86 RS/M2/36.04 An INEL user reported a request for a new capability, wherein the code is supposed to reset everything to time =0 when a restart from steady-state to transient is done. However, when Workshop Problem 3 is restarted from Workshop Problem 2, the plots made show data is loaded from t=0 to 120 sec (from the S.S. run) and the Problem 3 results continue from 0=120 see to the end of Problem 3 results. The data is plotted over the incorrect time interval.

Investigation is continuing on this requested new capability.

12/16/86 Cognizant Engineer: DMK U. f. #87019 Last Entry: 12/16/86 RS/M2/36.04+.05 upd An INEL user reported an input processing failure where indefinite results when search variable = thermodynamic property for a time dependent volume. In this case the search variable =TEMPF in PIPE component 990, the time dependent volume was component 980. The error occurred because, ISTATE resolves properties in the component order, hence, TDV 980 was referencing pipe 990 which had not been resolved yet. This makes the user unable to initialize a problem with a search variable used as above. A 'workaround' is possible by defining the search variable as a control variable and defining the time control variable initial C-39

value as the correct initial value. Investigation is continuing on this user problem.

12/30/86 Cognizant Engineer: CSM U. P. #87020 Last Entry: 12/31/86 R5/M2/36.04 An ICAP member reported a problem in which an input processing failure occurred in sample problem TYPPWR2. The user found an error in the R-strategy printing where the volume equilibrium initial flags of the pump volumes were printed as 5 even though the input values were 0. The user traced the error to subroutine RPMPVNJ where the variable IHF had been used incorrectly, as well as an additional line should be inserted to avoid the error. The coding of this subroutine is being checked for this error. When the RELAPS code developers executed the sample problem TYPPWR2, no error occurred in input processing.

01/07/87 Cognizant Engineer: RJW U. P. #87021 Last Entry: 01/07/87 R5/M2/36.04 A domestic user is modeling the Oconee plant and wanted to conncct 7 junctions to a time dependent volume (with VOIDG=1). Code allows only 1 junction to be connected to a time dependent volume. The user then used a single volume (with a large volume). The code got into mass error problems and was controlling on this large volume. The user was told to use 7 time dependent volumes (one for each of the 7 junctions) instead. The user suggests that more than one junctions be allowed to be connected to a time dependent volume.

01/12/87 Cognizant Engineer: RAR U. P. #87024 Last Entry: 01/12/87 R5/M2/36.04 An INEL user was performing a steady state balance of plant calculation for the Browns Ferry plant on the NPA. It was noted that the pressure was incieasing in time in the turbines rather than remaining constant, which suggests that power was not being removed. Further examination of the output indicates that the power, torque, and efficiency are not initialized for the I-level and time 0 printouts. Examination of the input deck indicates that the volumes and junctions upstream of the low pressure turbines are not the numerically preceding volumes and junctions. An update has been generated that will be combined with other updates to create RELAPS/ MOD 2/ Cycle 36.06 and is referenced as U. S. #87002, that will initialize the power, torque, and efficiency. The deck was corrected to put the turbines in numerical order, and the calculation then failed at 25 seconds with a water property error. Further examination of the deck showed dead-end turbine volumes and questionable connections. The dead-end volumes were connected to time dependent volumes, and the questionable connections were properly connected. The problem then ran successfully to 50 seconds, but there were flow oscillations, and choking at many junctions.

Investigation is continuing on this reported problem.

01/26/87 Cognizant Engineer: RAR C-40

U. P. #87027 Last Entry: 01/26/87 R5/M2/36.04 +S.I.

An INEL user reported unphysical results occurred on a GERDA Test 1605AA l calculation while running the self-initialization option in both semi-implicit l and two step modes, where a small amount of steam appeared in the primary hot leg. For the two-step mode these voids caused high frequency oscillations in l flow and pressure. For the 1-step mode this did net occur. This prevents the l calculation of a precise steady-state initializatim using the two-step mode.

Investigation is continuing on this reported user '.roblem.

01/26/87 Cognizant Engineer: Md U. P. #87028 Last Entry: 01/26/87 R5/M2/36.04 An INEL user reported a problem where the lead-lag control variable (previously steady in a Pygmalion deck) undergoes a step change in value on the first time step when the initialization flag is set to zero. The code assumes the integral term is zero but it would be better if the code calculated the value.

This user problem is under investigation.

01/29/87 Cognizant Engineer: RAR U. P. #87029 Last Entry: 01/29/87 R5/M2/36.04 + S.I.

An ICAP member at the INEL was running a KWU plant problem using the self-initialization option. Oscillations were observed in the steam generator pressures and mass flows. These oscillations went away when the user switched from the nearly-implicit scheme to the semi-implicit scheme. When the switch between schemes occurred, mass flow changes occurred in the primary system which went away after 5 seconds. Investigation is continuing on this reported user problem.

02/02/87 Cognizant Engineer: RAR U. P. #87030 Last Entry: 02/02/87 RS/M2/36.04 (B&W)

A domestic user (Newsletter member) reported a problem where several inconsistencies related to heat structure initialization were observed while trying to obtain steady state initial conditions for an Oconee model. An approach was followed while trying to obtain steady state initial conditions: 1

1. Control systems are used to adjust feedwater flow and primary system makeup / letdown. The STDY-ST code option is chosen to reduce heat structure capacitance and test for an overall approach to steady state (the TRANSNT option is used in the enclosed deck although the j inconsistencies still remain).
2. Once steady state has been achieved and a corresponding RSTPLT file is written, Pygmalion is used to update the volume properties, junction velocities, control system initial conditions and pump and valve characteristics in the original input deck. Pygmalion does not update ,

heat structure mesh point temperatures, l

3. The updated deck is run while only changing the TRANSNT option to C-41 l

determine if the deck is indeed at a steady state.

The results obtained from the updated deck execution do not yield the expected continuation of the original steady state run. Instead, the updated deck must be executed to a new steady state. The potential cause of this inconsistency was determined by comparing the appropriate major edit from the steady state run with the first major edit from the updated run. All parameters which are updated with Pygmalion are identical. Not all of the heat structure mesh point temperatures, however, are the same. Discrepancies are observed with heat structures which are adjacent to branch volumes. The temperature distribution calculated during heat structure initialization suggests that the incorrect velocity is used when calculating the appropriate Reynolds number and this heat transfer coefficient. A diskette was received from the user containing the input deck, and sample output. The RELAP5 code developers have transferred the input decks to the INEL CDC, and have cxecuted them. Investigation is continuing on this reported user problem.

02/04/87 Cognizant Engineer: RJW U. P. #87031 Last Entry: 02/04/87 l R5/M2/36.04 l

An ICAP member reported a problem where the code will compile, however will not 4 execute with the NO LCM option. Investigation is continuing on this reported user problem.

02/11/87 Cognizant Engineer: DMK U. P. #87032 Last Entry: 02/11/87 RS/M2/36.04 An INEL user obtained an input processing error when executing a modified McGuire deck in which the pump torque friction coefficients were input in British units. The code input was set to be British units and the output was j set to be S.I. units. The numbers printed out for torque friction coefficients were identical to those input meaning the the coefficients were not converted to S.I. units. Investigation in continuing on this reported user problem.

02/24/87 Cognizant Engineer: RAR U. P. #87035 Last Entry: 02/24/87 RS/M2/36.04 An ICAP member at the INEL was running a KWU plant problem using the

self-initialization option. An energy balance of the primary system showed that the total pump power is not added properly to the system energy. Of the Qh term of the pump power only the wall friction dissipation is added to the system energy. There is a discrepancy in checking the steady sate by means of an energy balance. The code should also add the form loss (user, abrupt,)

dissipation to the system energy. Investigation is continuing on this user problem.

02/25/87 Cognizant Engineer: CSM U. P. #87036 Last Entry: 02/25/87 RS/M2/36.04 C-42

i l

An INEL user reported a problem where an input processing failure occurred when' a branch junction was input with the wrong "to" and "from" volumes, so that the message "******** junction connections do not involve component volumes" was printed during input processing. However, input processing was still

! " completed successfully", and the problem executed. .This user problem is under investigation.

, 02/26/87 Cognizant Engineer: RAR U. P. #87037 Last Entry: 02/26/87 R5/M2/36.04 A domestic user is running an "N" reactor calculation and the code fails with a water property error at the minimum time step when significant amounts of ECS water began to fill the lower portion of the core. The user reduced the maximum requested time step, but the calculation still failed. The user also used the equilibrium option and removed some nearby heat structures, but the calculation still failed. It was suggested that the user model the core with equal size tubes rather than unequal size tubes. The user asked if the reflood option could be used for this application (horizontal core). The situation was r

discussed with the original developer of the reflood model, who indicated that-the model was originally developed for a vertical core. The developer thought the code would run,'but that because the reflood heat transfer correlations were from vertical studies and because no post CHF flow regimes exist in horizontal volumes, the results may be suspect. This problem is under investigation.

02/26/87 Cognizant Engineer: RJW/RAR U. P. #87039.2 Last Entry: 02/26/87 R5/M2/36.04 A domestic user was running a steady-state benchmark calculation with a loss-of-offsite power. The user first ran the problem with some direct heating in the source data cards 1CCCG701 through ICCCG799. The user then reran the i calculation with no direct heating. The user thought the 2 calculations should l

give the same total power generated. It was pointed out to the user that the

' code does use the internal source plus the direct heating multipliers in source data cards in adding power to this system. For the remainder of this problem see U. P. #87039.1 in Category #6. The user also reported a problem with

initializing a new trip as true at restart. The user restarted the problem at 100 seconds, put in 100 seconds in the TIME 0F quantity in the new trip card, but the trip was not true at the start of the restart (100 seconds). The user had to set TIMEOF to 99.999 seconds in order to get the trip to be true at 100 seconds. This problem is under investigation. .

02/26/87 Cognizanc Engineer: RAR U. P. #87040 Last Entry: 02/26/87 R5/M2/36.04 An INEL user was running a separate effects calculation for a LPIS break line test. The user observed that for this mostly steam problem, the mass error grew to 25% of the original mass. The calculation was rerun on the SELAP C-43

. . _ . - . - . - -. . . -- _ . . . ~.

4 (RELAP/SCDAP/ TRAP-MELT) code to check'the mass error, since the code's RELAPS portion has been modified to handle mass error problems when there is~ mostly steam in the calculation. The' mass error here grew to:only'3% of the original' mass'. 'The problem in the RELAP5 code is under investigation.

t' 1

1

j found that this user problem had been previously reported and resolved. The "

l codtag is contained in the update set to create Cycle 36.05, and is referenced j as U. S. #86009. 2 I

l 12/29/87 Cognizant Engineer: RJW/CSM  :

1 U. P. #87023 Last Entry: 01/13/87 RS/M2/36.04 An INEL user reported a problem where the code falls with an input processing j error with an overflow when trying to process heat structures in subroutine 4 HTISST. The user found that when a boundary volume is hooked to centerline of cylindrical rod when boundary volume hooked to outside of cylinder, cal rod, then the problem runs. A check was installed in HTISST to catch this type of user error (hooking boundary to centerline of solid cylindrical heat i structure). This update is referenced as U. S. #87001, and will be  ;

I incorporated with other updates to create RELAPS/M002/ Cycle 36.06.

I i 02/19/87 Cognizant Engineer: RAR O. P. #87033 Last Entry: 02/23/87 l RS/M2/36.05(testcycle)

I An INEL user reported a code execution failure, (CPU arithmetic overflow) when restarting a Semiscale steady state problem. The error occurred in subroutine VIMpLT, which calculates the velocities in the 2-step hydredynamic scheme.

After investigation, an error was found in the update to go into Cycle 36.05, t where the bulk steam enthalpy is stored in the variable DSNDPP. This variable

! was zeroed out in subroutine ISTATE in the base code. The user reran the

! restart problem without DSNODP set to zero and the code ran successfully 1 without failures. This update is referenced as U. S. #86012 and will be l

incorporated with other updates to create RELAPS/ MOD 2/ Cycle 36.05.

02/26/87 Cognizant Engineer: RAR/TTB U. P. #87038 Last Entry: 02/26/87

) C-60 i

- _ . - - . . . . _ - - . - , - - - - . - _ - , - - . _ - - - - - - _ _ - - - _ - - - -. w.

RS/M2/36.04 An INEL user was running a system calculation and used GANJA (part of the ISDMS l package) to convert some RELAPS variables on the RSTPLT file to a CWAF file format. The user then used the graphics package MAGNUM (also part of the ISDMS package) to obtain plots of these variables. One of the RELAP5 variable request codes used was HTRNR (heat flux), however, these plots showed the label (and units) to be heat transfer rate (W) instead of heat flux (W/m2) for this variable request. The error was traced to an error in the units code used in GANJA, which was corrected. During the course of the investigation, it was j discovered that the RELAPS variables (HTRNRO, HTRNRN, HTRNSO, HTRNSN) in the  !

i common block HTRSRCM, which are used by the variable request code HTRNR, were i

incorrectly labeled " rate" rather than flux in the comment common block ,

HTSRCMC. This update will appear in Cycle 36.06, and is referenced as U. S. #87002.

=

i i

i i

i 1

1 i

l i

C-61

RELAPS REPORTED USER PROBLEMS

  1. 5 Category 1 CLOSED - Updates in currently released version.  ;

10/02/85 Cognizant Engineer: DMK U. P. #86001 Last Entry: 03/25/86 RS/M2/36.02

)

An INEL user reported that the relief valve model was not working properly. He was comparing the code with EPRI data obtained from a CE safety valve  ;

facility. The valve model did not open as in the test data he was trying to simulate. An error was found in the valve opening force term, an update was

! created, and the user is continuing the assessment problem with this update, as well as working with the RELAPS group to optimize the relief valve model. This update is referenced as U. S. #86001, and was incorporated with other updates l which created RELAPS/M002/ Cycle 36.03.

10/08/85 Cognizant Engineer: RAR U. P. #86002.1 Last Entry: 03/25/86 R5/M2/36.02 1

An INEL user modeling a MARK I containment facility reported a problem in which the code failed on a water property error. An error was found in the noncondensible gas coding for the transition between no noncondensible and some

noncondensible. An update was created which corrected this error. The user used this update and reran the calculation. This update is referenced as

) U. S. #86002, and was incorporated with other updates which created  ;

RELAP5/M002/ Cycle 36.03. The calculation ran successfully, however large unphysical flow oscillations developed at some of the junctions. This problem is being examined, but at a low priority. (See U. P. #86002.2 in Category #2).

11/13/85 Cognizant Engineer: DMK 1

U. P. #86005 Last Entry: 03/25/86 RS/M2/36.02 1

An INEL user reported a problem in which the code failed when using the internal plotting package. It was found that subroutine AXISDV can get a zero for the axis length, which in turn causes a divide by zero in subroutines

PLOT 20. An update was created for the user which resolved his problem. This I

update is referenced as U. S. #86001, and was incorporated with other updates which created RELAPS/M002/ Cycle 36.03.

11/22/85 Cognizant Engineer: DMK U. P. #86006 Last Entry: 03/25/86 RS/M2/36.02 i An INEL user reported a restart /renodalization failure. Errors occurred or, restart in the relief valve model, which were not present with a NEW problem.

After checking, it was found that the test of the input flag for the relief C-62

valve was incorrect in subroutine ICOMPN. An update was created, and the problem was successfully rerun. This update is referenced as H. S. #86001, and was 'ncorporated with other updates which created RELAP5/M002/ Cycle 36.03.

12/11/85 Cognizant Engineer: LYN '

U. P. #86009 Last Entry: 03/25/86 RS/M2/36.02 A user from Sandia National Laboratory reported a problem he encountered when executing a reflood problem on a CRAY machine. Errors were found and corrected in the coding specific to the CRAY machine. This update is referenced as j U. S. #86003, and was incorporated with other updates which created ,

i RELAPS/M002/ Cycle 36.03

! 12/12/85 Cognizant Engineer: LYN

U. P. #86010.1 Last Entry: 01/05/86 j R5/M2/021 An ICAP member reported a problem in running a reflood case. An error was found in the reflood heat transfer index for obtaining material properties in subroutine MDATA2. This update is referenced as U. S. #85028 and was ,

j incorporated with other updates which created RELAPS/M002/ Cycle 36.02. For the l remainder of this user problem see U. P. #86010.2 in Category #3.

l l

j 12/12/85 Cognizant Engineer: RAR O. P. #86011 Last Entry: 03/25/86 l RS/M2/36.02 A user at the INEL reported a problem when trying to simulate a Bellefonte i plant model where there was noncondensible present in a pure vapor, volume. The

! wrong heat transfer rate was calculated for this case. An error was found in I the noncondensible heat transfer subroutine and an update was created. The l user used this update and has encountered no further problems. This update is l referenced as U. S. #86002 and was incorporated with other updates which 1

created RELAPS/M002/ Cycle 36.03.

I I' 12/17/85 Cognizant Engineer: RAR  :

U. P. #86012.2 Last Entry: 03/25/86  ;

RS/M2/36.02 I,

i A user from the United Kingdom reported several problems in the coding specific l to the CRAY machine, as well as a suspected error in the horizontal

stratification mouel. The suspected error is in the process of being
examined. (See U. P. #86012.1 in Category #1.) The updates the user submitted

! specific to CRAY errors, have been checked out, and are referenced as 1 U. S. #86002, and were incorporated with other updates which created

) RELAPS/M002/ Cycle 36.03.

}

! 12/17/05 Cognizant Engineer: DMK 3 U. P. #86013 Last Entry: 03/25/86

) C-63

j

1 l

l l

l R5/M2/36.02 An INEL user reported a problem in simulating a Semiscale M002-A plant model.

The problem occurred in input processing where incorrect values were obtained for the accumulator elevations. The printed elevation change represents the elevation change from the geometric centroid of the accumulator component to the exit of the surge line. It was found that there was an error in the calculation of the geometric centroid. This error affects only the input level printout, it is not considered significant for most applications. The correction update is referenced as U. S. #86001, and was incorporated with other updates which created RELAP5/M002/ Cycle 36.03.

01/02/86 Cognizant Engineer: RAR V. P. #86018 Last Entry: 03/25/86 R5/M2/36.02 An ICAP member reported errors in the CRAY version of RELAP5/M002. These corrections were prepared, tested, and incorporated into the updates which created Cycle 36.03. This update is referenced as U. S. #86002.

01/20/86 Cognizant Engineer: RAR U. P. #86024 Last Entry: 03/25/86 R5/M2/36.02 An INEL user reported a problem in which a calculation failed with a negative to real power. The problem had noncondensible gas present in a time dependent volume with heat structures. An error had already been reported for noncondensible gas in the presence of heat structures, and an update had been created (U. S. #86002). We recommended the use of the update to fix the problem. The user reran the calculation with the update and the code no longer failed. The user then noted that the correct heat flux to the environment could not be obtained. The user adjusted the thermal conductivity table and obtained the correct heat flux to the environment. The calculation was rerun with no further problems.

03/03/86 Cognizant Engineer: RAR/RJW U. P. #86032.1 Last Entry: 05/12/86 R5/M2/36.02 A RELAP5/M002 domestic user reported a problem with a plant calculation in which the core power was increasing when it should have been decreasing. A problem existed with the reactor kinetics feedback package. An error was introduced into this package when RELAPS/M002 went from a one-energy equation model to a two-energy equation model. In the one-energy equation model, the varieble TEMP is defined as the equilibrium temperature of a volume. The feedback model used TEMP for the moderator temperature. In the two-energy model, the equilibrium temperature is not computed. TEMP is now set to the saturation temperature corresponding to the pressure of the volume. The feedback incorrectly refers to TEMP. In a PWR, a largo errer is introduced (usually reactivity increase) when the pressure drops from the normal operating value to the saturation pressure of the 11guld. This pressure drop also causes the saturation temperature to decrease. This error would be expected to be C-64

small in large typical BWR simulations. The fix was to set moderator temperature to the spatial density weighted average of the liquid and vapor ,

temperatures. We generated an update to fix this problem and tur:cessfully tested it on a similar plant calculation. This update was incorporated with other updates which created RELAPS/ MOD 2/ Cycle 36.04, and is referenced as i U. S. #86008. In the course of checking out this problem, a hand calculation l seemed to showed that the bias reactivity was incorrectly calculated when British units are used with temperature coefficients (awi, aft), it was found

! that the user should use temperature in Rankine rather than Fahrenheit under ,

! these circumstances in order to get the correct bias reactivity. Words to this '

effect will appear in the next addendums to the users' manual. (See  !

l U. P. #86032.2 in Category #6.)

1 .

) 03/25/86 Cognizant Engineer: RAR

! U. P. #86043 Last Entry: 05/12/86 R5/M2/36.02 l

An INEL user reported a problem in which the code failed in the subroutine PSTONB. The code tried to use a negative viscosity in a correlation equation that took the negative viscosity to a real power. The negative viscosity was

traced to an incorrect property return from the steam table STH2X3 called from j subroutine PSTDNB. The steam table should have returned vapor properties but i instead returned liquid properties. The code was corrected to use the vapor density rather than the Ilquid density under such a situation so that when the ,

viscosity is then calculated, it will be correct. An update was created, and l l the user reran the calculation with no further problems. This update is t referenced as U. S. #86008 and was incorporated with other updates which '

created RELAP5/M002/ Cycle 36.04.

l 03/25/86 Cognizant Engineer: CSM O. P. #86044.2 Last Entry: 05/12/86 i RS/M2/36.02 An ICAP member reported a problem with standard Workshop Problem 2. Unphysical I results were noted in the mass flow rates and reactor power which resulted in a  !

large mass error. This problem is still under investigation. (See  !

U. P. #86044.1 in Category #2.) In addition, some apparent dead coding was '

found in the subroutine PHAINT by the user. This coding was located, coded,

.! and checked out and was incorporated with other updates which created RELAPS/M002/ Cycle 36.04, and is referenced as U. S. #86006. l 03/26/06 Cognizant Engineer: DMK U. P. #86045 Last Entry: 05/12/86 i RS/M2/36.02 ,

i h i'

An INEL user reported two problems with the steady state option. The first was ,

an error in the label print statement in the steady state printout, and the -

!' second was an abnormally long run time to reach steady state. The first problem was corrected by fixing the " write" statement to output the correct label, and the second paoblem was fixed by branching to the " constant rate" portion of the steady ste checking subroutine if the three time-averaged i

lines were within the sunvergence criterion. These updates were coded, checked  ;

t C-65 i

out and were incorporated with other updates which created RELAP5/M002/ Cycle 36.04. This update is referenced as U. S. #86005.

04/15/86 Cognizant Engineer: RJW U. P. #86049 Last Entry: 05/12/86 RS/M2/36.03

) An INEL user reported a problem in trying to execute the Nuclear Plant Analyzer j (NPA) in the interactive mode. The problem would execute in both batch and a replay mode. An error was found in the CDC version of the NPA when the RELAP5 j group created Cycle 36.03, and added a new comdeck, NPACOM. An update was created, the user reran the calculation with no further problems. This update i

is referenced as U. S. #86007, and was incorporated with other updates which

q created RELAP5/M002/ Cycle 36.04.

i i I l 04/18/86 Cognizant Engineer: DMK U. P. #86051 Last Entry: 05/12/86 RS/M2/36.03 j i

j An INEL user reported a problem in trying to simulate a Zion model in which I occumulator and plotting input failures occurred. However, accumulators were I

not renedalized and plots were not requested. The user could not restart the  !

l problem. The RELAPS developers suggested the user renodalize, but the code j still failed with an indefinito. Errors were corrected in the accumulator trip  !

word packing, logic was corrected to prevent load of plot files when no plots  !

! are requested, and misspelled variables were corrected to prevent indefinites  ;

when the accumulator is renodalized. Coding was created, and the user reran <

i the calculation with these updates and obtained another problem. (See

! U. P. #86053 in this Category.) This updato is referenced as U. 5. #86005, and l i was incorporated with other updates which created RELAPS/M002/ Cycle 36.04, i

04/22/86 Cognizant Engineer
RJW

.I U. P. #86052 Last Entry: 05/12/86 l RS/M2/36.03 An INEL user reported a problem where the code does not recognize extended LOGICAL trips. Those are cards 20610010 through 20620000, which the user correctly inputted. The subroutine RTRIP, which reads in the cards for *

] oxpanded logical trips, was corrected so these cards will now be recognized.  !

! This update is referenced as U. S. #86007, and was incorporated with other l updates which created RELAPS/M002/ Cycle 36.04 I

  • J 04/24/86 Cognizant Engineer: DMK l U. P. #86053 Last Entry: 05/12/86 l j R5/M2/36.03 j i  ;

j An INEL user reported a problem in which the code failed after the calculation l 1 had been restarted and an indefinite occurred on the first time step. The i

! indefinite occurred in the subroutine STACC, however, the error is in the  !

j noncondensible model. The modification extending the noncondensible common  !

block STATEC, did not include extending the length of the block written to the  :

i C-66 I

l l

restart / plot file. Hence, on restart these terms were indefinites. An update was created, checked out, and was incorporated with other updates which created ,

RELAPS/M002/ Cycle 36.04, and is referenced as U. S. #86005. '

05/01/86 Cognizant Engineer: RJW l U. P. #86054 Last Entry: 05/12/86 l R$/M2/36.03 ,

An INEL user reported a problem with the code where the user added more control l

! systems to the input deck (B&W safety reliability) and an FTB #16 error }

! occurred during input processing. An update was created which corrects errors  ;

l in subroutine RMFLOS for large input decks. This update was incorporated with  !

! other updates which created RELAP5/ MOD 2/ Cycle 36.04, and is referenced as [

j U. S. #86007. i i

05/28/86 Cognizant Engineer: CSM U. P. #86061 Last Entry: 05/28/86

, RS/MP36.02 r

An ICtp member reported a problem wherein the code failed when trying to execute the internal plotting package. The error had been already corrected in

i

~

the updates which created Cycle 36.03, and is referenced as U. S. #86001.

These updates had been previously transmitted to thei ICAP members, and the j RELAPS development group suggested that this new cycle be created to alleviate- ,

j this plotting problem.

l 1 l

1 l  !

1  !

l I

l

i.  !

L l

i i i I

i I

1 C.G7

l RELAPS REPORTED USER PROBLEMS ,

l l ,

  1. 6 CATEGORY  !

CLOSE0 - User guidelines provided. 1 10/30/85 Cognizant Engineer: LYN U. P. #86004 Last Entry: 10/30/85 RS/M2/36.02 An INEL user reported a problem in which the code failed due to a water property error in the core region of a steady-state RESAR plant model. The ,

problem had previously executed properly on Cycle 21. After looking into this problem, it was noted that the input deck was produced by PYGMAL10N, and the deck was set up initially with a non-zero vold fraction (1.0E-2) in the core.

The problem could be circumvented by specifying the core initially to be 11guld full. No update was required, the user regenerated his input deck, and reran .

his calculation with no further problems. I 11/26/85 Cognizant Engineer: RAR U. P. #86007 Last Entry: 11/26/85 RS/M2/36.02

+ Exxon updates i

A user from Exxon Nuclear Company submitted a problem in which unphysical results occurred in pressure on the first time step for the GE-large vessel blowdown problem. The problem was resolved by running at a smaller time step size.

Note: Users are cautioned against using a large maximum time step size at problem initiation. The very first time step taken is the maximum time step input by the user. Since the first time step is not Ilmited by the material Courant limit, unphysical results may occur if a rapid transient coincides with the start of the problem.

12/18/85 Cognizant Engineer: RAR/RJW U. P. #06015 Last Entry: 12/18/85 RS/M2/16.02 A RELAPS/M002 domestic user reported a problem in which the test case failed in input processing when trying to add a new component. It was investigated, and was found that a typing error was found in the input deck. The typo was corrected, and the test case was rerun with no further problems.

l 01/07/86 Cognizant Engineer: RAR/RJW/DMK l U. P. #86020.1 Last Entry: 01/07/86 R5/M2/36.02 A domestic user reported a problem in executing a steady state calculation (ANO-2 turbine trip) in which a high primary pressure (compared to RELAP5/M001), was obtained which was due to input problems with direct C 68

. _ _ _ _ _ _ _ _ _ _ _ . _ . _ . _ ~ -

heating. The user had not partitioned the direct heating among the heat slabs. The user corrected the input deck by inputting the correct partitioning of the direct heating and raran the calculation, but could not obtain l

steady-state. Recommendations were made to change the input deck (the user had not previously specified the steady-state option) and the steady state portion of the calculation was completed successfully. When the transient portion of the calculation was begun, the code ran slowly. Suggestions were made to speed up the calculation by increasing the length of the volumes that were Courant l

! limited. For the remainder of this problem, see U. P. #86020.2 in Category #1.

01/07/86 Cognizant Engineer: RAR/RJW l U. P. #86021 Last Entry: 01/07/86 R5/M2/36 (Private IBM version) l i A domestic user reported a problem in executing the TYPPWR2 problem on an IBM i machine, and obtained different results than those contained on the microfiche i which was transmitted with Cycle 36. It was found that the user had used the l TYPPWR2 input d ck used for Cycle 21, and not the recommended Cycle 36 input i deck. The new Input deck was sent to the user, the calculation was rerun and j obtained basically the same results as on the microfiche, t I 01/23/86 Cognizant Engineer: RAR

! U. P. #86025 Last Entry: 01/23/06

! R5/M2/36.02 l

An IM L user reported a problem where a calculation failed due to high vapor j temperatures in a volume without heat structures attached. The error was  :'

l Investigated, and it was found that the user had an error in the input deck, I

wherein noncondensible gas was specified in a volume that should have had just steam / water. The user corrected the input deck and raran the calculation with no further problems. No update was required for this error.  !

l' 02/05/86 Cognizant Engineer: RAR i U. P. #86027 Last Entry: 02/05/86 i

, R5/M2/36.02 i An INEL usar reported a problem in which the code failed with a CPU error on I'

restart. This problem was traced to the input deck in which the user had i specified on Card 100 a NEW problem rather than a RESTART problem. The input I i deck was corrected and the calculation raran with no further problems. No i update was required for this user problem.

I I 02/19/86 Cognizant Engineer RAR  !

) U. P. #86030 Last Entry: 02/19/86  ;

I RS/M2/36 j 1

b A domestic user reported a problem in executing the TYPPWR2 input deck in which input processing errors occurred. The problem was investigated, and it was found that when their version of RELAPS/M002/ Cycle 36 was compiled they did not input the correct " define" cards for their specific computer, and therefore f were inadvertently trying to execute the NpA (which the INEL does not include j C-M i

i

on transmittals of RELAP5/M002). In addition, the code failed with a divide by l zero. The error was fixed in Cycle 36.01, and it was suggested that the problem be rerun with the updates to Cycle 36.01 and Cycle 36.02. The user obtained the updates and executed the calculation with no further problems. No update was required for this problem.

03/03/86 Cognizant Engineer: RAR/RJW U. P. #86032.2 Last Entry: 03/03/86 ,

R5/M2/36.02 i A RELAP5/M002 domestic user reported a problem with a plant calculation in which the core power was increasing when it should have been decreasing. (See  !

U. P. #86032.1 in Category #5.) In the course of checking out this problem, a I hand calculation seemed to show that the bias reactivity was incorrectly  ;

calculated when British units aret used with temperature coefficients (awl, i aft). It was found that the user should use temperature in Rankine rather than Fahrenholt under these circumstances in order to get the correct bias l reactivity. Words to this effect will appear in the next addendums to the L users' manual.

03/04/06 Cognizant Engineer: CSM U. P. #86037 Last Entry: 03/04/06 R5/M2/36.02 ,

An INEL user reported a problem in executing a TMI-2 simplified model '

calculation where the code falls with a water property error at the minimum ,

time step. The calculation was rerun using a smaller requested maximum time '

stop, and the code executed. However, large mass flow oscillations occurred in the core which appeared to be unphysical. After investigation of this problem, '

It was decided that the user rerun the calculation using the combined code ,

SCDAP/RELAPS/ TRAP MELT. No further problems occurred.

03/07/06 Cognizant Engineer: RJW U. P. #86040 Last Entry: 03/07/06 R5/M2/36.02 i An INEL user requested that the parameter number associated with the interactivevariables(Cards 801 999) be changed from 1000000000 to a smaller number to avoid user input errors. The number 1 is used as a component. Other  ;

numbers are also used for volume numbers, junction numbers, etc. The number l 1000000000 la the smallest number that is not used. There is no plan to chinge  ;

this.  !

l 03/11/86 Cognizant Engineer: CSM U. P. #06041 Last Entry: 03/11/06 I R5/M1/025 and RS/M2/36.01 A domestic user reported a significant dif ference between REL APS/M001 and l RELAP5/M002 calculations of form loss for an abrupt area change junction that has a small area ratto. Input decks for both codes were received and the results rep 1tcated on the INEL CYBER. Due to the small junction area, the user c t

C 10 f

l was advised to use a smooth rather than an abrupt area change and to input the a junction form loss factor. (See RELAPS/M002 Code Manual, Volume 2, page 14).

No update or code change was required for this user reported problem.

03/25/86 Cognizant Engineer: CSM  :

U. P. #86044.1 Last Entry: 05/12/86 '

RS/M2/36.02 l

! An ICAP member reported a problem with standard Workshop Problem 2. Unphysical l results were noted in the mass flow rates and reactor power which rasulted in a j large mass error. After investigation it was found that Standard Workshop 1

Problem 2 is run in a transient mode, and was replaced by PROB 25, which is the j same problem run in a steady-state mode. The transient mode Problem 2 was run 4

with a power and flow that produced in intermittent CHF condition that caused l the reported unphysical results. No update was required for this problem. For j the remainder of this problem see U. P. #86044.2 in Category #5. <

i

, 04/09/86 Cognizant Engineer: CSM i j U. P. #86047 Last Entry: 04/09/86  !

1 RS/M2/36.02 i

j An ICAP member reported a problem where discrepancies were found between the i users' CDC-cYBER 180-830 (with no round optinn on) calculated sample problems Workshop Problem 2 and TYPPWR2 and the transmittal tape results for the  !

corresponding problems. The sample problem EDHTRK gave identical results.  ;

Machine hardware and software differences can cause solution discrepancies  !

particularly when and after the solutions have undergone disturbances such as

{ water packing or significant heat transfer thresholds. No update was required j for this reported user problem, i

1 Note: Users can expect slight differences in the machine round option, and results may vary.

t 04/14/86 Cognizant Engineer: RAR

, U. P. #86048 Last Entry: 04/14/86 R5/M2/36.02

( An INEL user was running a separate effects steam line calculation, beginning j with stagnant conditions and then ramping up the steam flow. The calculation i failed with a negative to a real or duuble power. The failure was traced to the condensation heat transfer CONDEN. The user had set a trip wrong which resulted in a valve closing when it should have remained open. This resulted  :

! In unphysical and unrealistic conditions and then the code failed. There is no l

update required for this reported user problem, i i  !

t i 05/06/06 Cognizant En3tneer: CSM l U. P. #86056 Last Entry: 05/06/06  ;

j R5/M2/36.02 4

! An INEL user running a separate effects (sub systems) superheated hydrogen, i steady-state calculation, noted the code reduced the time step due to large i i

C 71 ,

i

mass error. After investigation of the input deck, it w?s recommended the user change the input model by removing the large single volume that was controlling the mass error. The user reran the calculation with no further problems. No update was required for this reported user problem.

05/27/86 Cognizant Engineer: CSM U. P. #86058 Last Entry: 02/23/87

RS/M2/36.02 An ICAP member reported a problem where the code failed with a mode 6 error in subroutine QFHTRC when the reflood option was specified. In checking out this

. problem, it was decided that the RELAPS code development group needed this i

! input dock for further investigation. Contact was made with the ICAP member l and the input deck was requested. After conversations with the user it was found that when Cycle 36.02 was installed, the user had requested a "0EF NOS" statement, rather than a "DEF NOS-BE" statement. The user recompiled the version and reran the calculation with no further problems.

i 05/27/86 Cognizant Engineer: CSM i U. P. #86059 Last Entry: 02/23/87 I RS/M2/36.02 An ICAp member reported a problem in executing a large break LOCA calculation where the code failed with an error mode 1 in subroutine MDATA2 (line 86). The input deck was requested from the user, however after conversations with the user, it was found that when Cycle 36.02 was installed, the user had requested a "0EF NOS" statement, rather than a "DEF NOS-BE" statement. The user recompiled the version and reran the calculation with no further problems.

l 05/27/86 Cognizant Engineer: CSM I U. P. #86060 Last Entry: 02/23/87 R5/M2/36.02 An ICAp member reported a problem in executing a REBEKA (post test calculation on reflood) calculation. A restart was requested, and the code failed with an

, error mode 4. The input deck was requested, however, after conversations with

the user, it was found that when Cycle 36.02 was installed, the user had requested a "DEF NOS" statement, rather than a "DEF NOS-BE" statement. The 4

user recompiled the version and roran the calculation with no further problems.

06/16/86 Cognizant Engineer: RJW U. P. #86064 Last Entry: 06/23/86 R5/M2/36.04 Two INEL users were running an Oconee plant model with the new and improved control system (users' own control system). The code failed with arithmetic overflow in the subroutine CONVAR (control variables). Af ter investigation of this problem, it was found the user had an error in the update flie. The user fixed the updato and reran the calculation with no further problems. No update was required for this reported user problem, as the updates are for this particular calculation crer.ted by the user.

C-72

06/23/86 Cognizant Engineer: CSM i U. P. #86066 Last Entry: 06/23/86 l

R5/M2/36.03 An ICAP member reported a problem installing the updates to create RELAP5/ MOD 2/ Cycle 36.03. The user submitted the errors via phone which caused problems in compiling their version of Cycle 36.03. The updates were traced to the main RELAPS program. When the INEL-CDC computer had problems with the FTNS compiler, it was decided to re-create Cycle 36.02 to see if there were FORTRAN problems. None were indicated by the sample problems. When Cycle 36.03 was created, the interim cycle (36.50) was used as the base cycle, therefore the "PTITLE" cards used the incorrect update reference numbers. Two card changes were sent to all ICAP members and recipients of Cycle 36.03 to replace the Cycle 36.50 title cards.

06/27/86 Cognizant Engineer: RAR U. P. #86071 Last Entry: 08/26/86 R5/M2/36 An ICAP member reported problems in trying to install Cycle 36 on a CRAY machine. Errors were noted in trying to execute the sample problems NULL, EDHTRK, AOKI, and PROB 2. Corrections to Cycles 36.01, 36.02, 36.03, and 36.04 have corrected these problems. The user was informed to install all the updates to create Cycle 36.04 before executing the sample problems.

07/02/86 Cognizant Engineer: CSM U. P. #86073 Last Entry: 10/31/86 RS/M2/36.04 An INEL user reported a problem in executing a B&W OTIS calculation where unphysical results were obtained on a time dependent junction with a trip.

After investigation, it was found that wrong entries wore inputted in the TDJ tables. The user was informed of this, and the user reran the calculation with' no further problems.

07/14/86 Cognizant Engineer: CSM U. P. #86074 Last Entry: 07/14/86 RS/M1/028 A domestic user reported a problem in trying to execute a calculation where unphysical results were obtained in the pump curve data. It was noted that the user did not input enough homologous pump curves into the input deck.

Recommendations were made to the user to add more pump curves. No update was required for this reported user problem.

07/14/86 Cognizant Engineer: RJW U. P. #86075 Last Entry: 10/14/86 R5/M2/36.04 C-73

An INEL user requested the capability to apply the min / max limits to the integral prior to scaling the integral reset. rate (PROP-INT CNTRLVAR). Updates received from the user several cycles ago (in the code now) have been modified. Guidelines were reported to the user.

08/26/86 Cognizant Engineer: CSM U. P. #86088 Last Entry: 08/26/86

, R5/M2/36.02 1

A domestic user reported a problem in'trying to restart a calculation adding I noncondensible gas (nitrogen) and the code failed. After investigation of this problem, it was noted that the code will not accept a noncondensible gas addition (card #110 on restart) on a renodalization/ restart if it was not entered at the start of the calculation. User guidelines were provided, and words to this effect will appear in the next addendum to the user manual.

08/27/86 Cognizant Engineer: DMK U. P. #86090 Last Entry: 10/14/86 R5/M2/36.04 An INEL user was running a calculation simulating a Chernobyl reactor and obtained unphysical results using the check valve option 1 (static pressure controlled check-valve (no hysteresis)). When option 0 (static pressure / flow controlled check valve (has hysteresis effect) was used,'the results appeared okay. Guidelines were reported to the user explaining that the check valve model in the code are "on-off" switch models and because of their on-off nature, then unphysical behavior can be observed. In the case of this reported problem, option 0 was a check valve model that better approximated real valve behavior than the option +1 model. Hence a more physical result was obtained.

09/02/86 Cognizant Engineer: RAR U. P. #86091 Last Entry: 09/02/86 R5/M2/36.04 An INEL user running a B&W OTIS small cold leg break calculation, reported a problem in which the code failed on high pressure (>3000 psi). The RELAP5 developers suggested to the user to cut the maximum time step and rerun the

- ca_l cul ati on. The user reran the calculation, and'the code did not fail, and the results and plots looked okay.

09/17/86 Cognizant Engineer: RAR U'. P. #86094 Last Entry: 09/17/86 RS/M2/36.01 A domestic user reported a problem in trying to attempt-to create Cycle 36.04 from Cycle 36.01 and the following segload directive error occurred in the dayfile: RSTGEN - undefined segment, RSTGENX - undefined segment. It was found that the segments RSTGEN and RSTGENX were part of B&W updates for noncondensible heat transfer and auxiliary feedwater model which were included in their version. The user is resolving the problem with B&W.

C-74

FY-87 REPORTED USER PROBLEMS-CATEGORY #6 - USER GUIDELINES PROVIDED 10/22/86 Cognizant Engineer: RJW U. P.#87007 Last Entry: 10/27/86 RS/M2/36.02 An ICAP member reported a problem in trying to restart a " big case" (SB-2) changing the geometry to two volumes (renodalization) and redefining two trips. The problem contained 43 vessels, 95 secondary volumes, and 900 control variables. The user found it impossible to continue the transient unless changes to the restart dump frequency and restart from a preliminary restart dump was made. The user experienced a code failure stating " insufficient SCM space to handle restart". After investigation into this user problem, it was noted that the user had a CYBER 830, and when installing RELAP5/ MOD 2 should have specified " define LCM" rather than " define no LCM". The user was informed to recompile the version with LCM.

11/24/86 Cognizant Engineer: CSM U. P. #87010 Last Entry: 11/24/86 R5/M2/36.04 An ICAP member reported discrepancies in comparing a RELAp5 result to a KWU fuel case, and the results showed the gap conductance too low. The RELAPS developers created an update to make the subroutine GAPCON concentric. The calculation was rerun with this update and the fuel node temperatures and the centerline temperature matches KWU code results. However, this update will not be incorporated into the frozen version of the code, due to the fact that RELAPS is a best estimate code and its fuel model uses a non-concentric gap model taken from FRAP. This has been shown to be more ac; urate in FRAP assessment. This user problem was designated as a user guideline.

12/05/86 Cognizant Engineer: RJW U. P. #87013 Last Entry: 12/16/86 4

R5/M2/36.04 A domestic user reported a problem where the code fails when using expanded trip logic (supposedly fixed in Cycle 36.04). The user sent the input deck, to the RELAPS code developers. The input deck was loaded onto the CDC computer, and run through input processing, however, the same problems occurred. After q investigation it was found that the user entered card numbers instead of trip

, numbers where trip numbers are required on logical trip input data. No update was required for this reported user problem.

1 12/08/86 Cognizant Engineer: RAR U. P. #87014 Last Entr : 12/08/86 RS/M2/36.04 An ICAP member reported a problem when running a 2-loop KWU plant model to a

steady-state using the self-initialization features. The steam flow to the turbine (time dependent volume) showed large unphysical oscillations, but the C-75

junction area did not. Investigation shows a small amount of liquid appearing in the steam lines seem to be causing problems with the choking model. The user increased the user inputted form losses in the junctions connecting the 2 steam generators, and the oscillations went away. No update was required for this reported user problem.

12/09/86 Cognizant Engineer: RAR U. P. #87016 Last Entry: 12/09/86 R5/M2/36.04 An ICAP member reported a problem in running a KWU plant problem using the self-initialization option, where large void oscillations were observed in the separator volumes. A guideline was recommended to the user to use large (~100) user input loss coefficients for vapor outlet and liquid fallback junctions.

No update was required for this reported user problem. l Wote: Separators in general have many internal surfaces that lead to flow resistances above that of an open region. For this reason, additional energy loss coefficients may be required at the appropriate separator' junctions. We have recommended that these be obtained from handbook values or adjusted to match a known pressure drop across the separator. In some cases, it may be necessary to use large loss coefficients (~100) at the vapor outlet and liquid fallback junctions in order to remove void oscillations in the separator volume.

12/16/86 Cognizant Engineer: DMK U. P. #87018 Last Entry: 02/12/87 R5/M2/36.04 An INEL user reported a problem wherein the code is supposed to reset everything to time =0 when a restart from steady-state to transient is done.

However, when Workshop Problem 3 is restarted from Workshop Problem 2, the plots made show data is loaded from t=0 to 120 sec (from the S.S. run) and the Problem 3 results continue from 0=120 sec to the end of Problem 3 results. The data is plotted over the incorrect time interval. This is not a code problem.

Workshop Problem 3 was restarted as a transient from a steady-state restart file, hence the problem was reset to time =0. However, because trip 501 was set to begin the transient at 150 seconds, then the problem simply ran the first 120 seconds as a duplicate of Workshop Problem 2. The fix was to set trip 501 to 30 seconds, and the calculation ran with no further problems.

02/26/87 Cognizant Engineer: RJW/RAR U. P. #87039.1 Last Entry: 02/26/87 R5/M2/36.04 A domestic user was running a steady-state benchmark calculation with a loss-of-offsite power. The user first ran the problem with some direct heating in the source data cards 1CCCG701 through 1CCCG799. The user then reran the calculation with no direct heating. The user thought the 2 calculations should give the same total power generated. It was pointed out to the user that the code does use the internal source plus the direct heating multipliers in the source data cards 1CCCG701 through ICCCG799 in adding power to this system.

C-76

Further user guidelines will be included in Volume 2 of the manual, and the Newsletter Service to explain this further. For the remainder of the user problem, see U. P. #87039.2 in Category #1.

l l

C-77

i RELAP5 REPORTED USER PROBLEMS

  1. 7 CATEGORY CLOSED - user problem merged with another similar problem.

, 10/18/85 Cognizant Engineer: RAR 4

U. P. #86003 Last Entry: 06/03/86 R5/M2/36.02 A user from Brookhaven National Laboratory reported a problem with a water property error in which the code fails. The failure occurred when the critical g

point was exceeded. The thermody.tamic properties subroutines require modification to permit calculations beyond this point. The problem has been merged with U. P. #86050 in Category #1.

12/17/85 Cognizant Engineer: RAR/RJW U. P. #86014 Last En":ry: 03/03/86 R5/M2/36.02 1 A user at the INEL reported a water property error when executing the TYPPWR2 sample problem on the CRAY machine. This user problem has been merged with

U. P. #86017 (below), and #86036 in Category #2.

I 1

?

12/30/85 Cognizant Engineer: RAR i U. P. #86017 Last Entry: 03/03/86 RS/M2/36.02 An INEL user reported a problem with the TYPPWR2 sample problem in which thw code fails after 800 seconds of transient time. It occurred in the condensation heat transfer correlation. This user problem has been merged with

U. P. #86014 (above) and #86036 in Category #2.

1 C-78

_ .. ~ . . . _ . ._- . _ . - .

      • TRAC-BF1 DEVELOPMENT FORM ***

CORRECT FUEL R0D PROPERTIES FOR USER INPUT MATERIALS PCL/ TYPE - RAl/E PROBLEM STATUS: RESOLUTION LEVEL -3.

Open -- Item is completed pending integration checkout.

NATURE OF THE PROBLEM: TYPE - E.

This update involves an error that caused unsatifactory results.

DEVELOPER (S): R. W. Shumway l

REPCRTED BY: Mike Analytis ADDRESS: EIR, Switzerland DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BF1 (G1J1)

PROBLEM STATEMENT:

1 EIR (Mike Analytis) found that putting a very small conductivity i at a rod surface gave unrealistic results.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

The problem was traced to Subroutine RODHT. R0DHT was using an interface material property.from the wrong location in the material when the material wastinput by the user.

To correct the problem, Subrcptine MFR00 must call the property routines twice, once ,for the CK array and once for the CKIR array. The same is true ,for RHO and CP.

1

~

( FIRST 4 CHARACTERS OF DECK AND COMDECKS CHANGED OR ADDED BY THIS UPDATE:

MFR0 F

e o

?

I l

C-79

      • TRAC-BF1 DEVELOPMENT FORM ***

SIMPLIFY THE R00 CONDUCTION SOLUTION. N0 GAP REQUIRED.

PCL/ TYPE - RA2/C PROBLEM STATUS: RESOLUTION LEVEL =1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE - C.

This update involves a model change.

DEVELOPER (S): R. W. Shumway REPORTED BY: R. W. Shumway ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87-VERSION USED: TRAC-BF1 (GlJ1)

PROBLEM STATEMENT:

Currently rods must have a gap and there are a lot of IF tests needed to keep track of the node in question relative to the cladding.

The new method pre-sets the solution coefficients (involving the conductivity and radii) and does the solution with a D0 loop without IF tests. A node is a node. This will help in future vectorization.

Update RA4 is needed with this to get the RHO *CP product and conductivity based on the average temperature.

VERSION TO BE CORRECTED: This update will be made to the next code version. The update will be not applied to the released version.

FIRST 4 CHARACTERS OF DECK AND COMDECKS CHANGED OR ADDED BY THIS UPDATE:

RODH C-80

      • TRAC-BF1 DEVELOPMENT FORM ***

FIX MAJOR EDIT FOR AVERAGE R00 ENERGY TRANSFER PRINT 0UT. ,

PCL/ TYPE = RA3/0 PROBLEM STATUS: RESOLUTION LEVEL -3.

Open -- Item is completed pending integration checkout.

NATURE OF THE PROBLEM: TYPE = 0.

This update involves a nonfatal code output problem or an output request by a user.

DEVELOPER (S): R. W. Shumway 4 REPORTED BY: R. W. Shumway DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BF1 (G1J1)

PROBLEM STATEMENT:

The rod surface area was not being multipied times the convective heat flux. The printout was supposed to be in watts but.was a mix.

The area was multiplied times the vapor convection but not the liquid.

FIRST 4 CHARACTERS OF DECK AND COM0ECKS CHANGED OR ADDED BY THIS UPDATE:

WROD i

')

l l

I j

i C-81

      • TRAC-BF1 DEVELOPMENT FORM ***

CHANGE MFROD TO BASE CONDUCTIVITY ON THE AVE TEMP.

PCL/ TYPE - RA4/C DEVELOPER (S): R. W. Shumway REPORTED BY: R. W. Shumway ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 DATE OF PROBLEM REPORT: 3/13/87 i DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BF1 (GlJ1)

PROBLEM STATEMENT:

Currently the thermal conductivity needed between two nodes is based on the temperature of the node on the left.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

The thermal conductivity needed between two nodes will now be based on the average temperature of the two nodes.

The RHO *CP product is used by RA2 update. This product will be computed in Subroutine MFR00.

VERSION TO BE CORRECTED: This update will be made to the next code version. The update will be not applied to the released version.

FIRST 4 CHARACTERS OF DECK AND COMDECKS CHANGED OR ADDED BY THIS UPDATE:

MFR0 i

C-82

      • TRAC-BF1 DEVELOPMENT FORM ***

WRITE OUT FINE MESH INCONSISTENCY MESSAGE FOR CHAN PCL/ TYPE - RAS /0 PROBLEM STATUS: RESOLUTION LEVEL -3.

Open -- Item is completed pending integration checkout.

NATURE OF THE PROBLEM: TYPE - 0.

This update involves a nonfatal code output problem or an output request by a user.

DEVELOPER (S): R. W. Shumway l

REPORTED BY: R. W. Shumway ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 l

DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BF1 (G1J1)

PROBLEM STATEMENT:

The output for the input of the CHAN component tells you something is wrong with your fine mesh input, but not specifically what.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

This update gives you the values used that generated an error message for the fine mesh input and suggests how to remedy the problem.

FIRST 4 CHARACTERS OF DECK AND COMDECKS CHANGED OR ADDED BY THIS UPDATE:

FCHN C-83

      • TRAC-BF1 DEVELOPMENT FORM ***

FIX GRAPHICS ENGINEERING UNITS CODE FOR TORQUE MULTIPLIER PCL/ TYPE - RA6/0 PROBLEM STATUS: RESOLUTION LEVEL -3.

Open -- Item is completed pending integration checkout.

NATURE OF THE PROBLEM: TYPE - 0.

This update involves a nonfatal code output problem or an output request by a user.

DEVELOPER (S): R. W. Shumway REPORTED BY: R. W. Shumway ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BFI (GlJ1) 1 PROBLEM STATEMENT:

The graphics output for the engineering units code for the the torque multiplier (TORQM) is wrong.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

The torque multiplier currently has torque units but it is a ratio of torques and should be unitless. Therefore the graphics output has been changed to make it unitless.

THIS UPDATE REQUIRES A CHANGE TO THE MANUAL:

In the BF1 Manual, the units code for all occurences of the plot variable TORQM in Tables 11.2-8 through II.2-10 should be changed from 90 to 77.

FIRST 4 CHARACTERS OF DECK AND COMDECKS CHANGED OR ADDED BY THIS UPDATE:

IGPM l

l C-84 l

      • TRAC-BF1 DEVELOPMENT FORM ***

CHANGE THE BUILT-IN PUMP CONTROLLER TORQUE-M MIN TO 0.

PCL/ TYPE = RA7/C PROBLEM STATUS: RESOLUTION LEVEL -1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE - C.

This update involves a model change.

-DEVELOPER (S): R. W. Shumway

. REPORTED BY: R. W. Shumway

' ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 PROBLEM STATEMENT:

The torque multiplier minimum for the built in flow controiler of the control system is currently .75. This means that the users must have there loop resistance matched fairly well with their

, pump size.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

Having a lower minimum torque, multiplier will allow the user to use an over sized pump.

f f

C-85

      • TRAC-BF1 DEVELOPMENT FORM ***

BFIMANUALCORRECTIONSTOSTEk)YSTATECONTROLLERINPUT PCL/ TYPE = RA8/D PROBLEM STATUS: RESOLUTION LEVEL -5.

' Open -- Item is completed pending integration checkout.

NATURE OF THE PROBLEM: TYPE - D.

This update involves a documentation deficiency.

DEVELOPER (S): R. W. Shumway REPORTED BY: R. W. Shumway ,

ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 PROBLEM STATEMENT:

The TRAC-BF1 manual is incorrect in the description of three variables for steady state controller input.

THIS UPDATE REQUIRES A CHANGE TO THE MANUAL:

Page 265 cf the manual should be corrected by changing "W3-R Initial" to "W3-R Expected steady state".

Page 266 should be corrected by changing, under Flow Controller Input

Description:

"W3-R Initial recirculation ... Torque)." to "W3-R Expected steady state torque / rated torque."

and under Pressure Control System Input

Description:

"W3-R Time zero" to "W3-R Expected steady state".

i I

C-86

~ . .

      • TRAC-BF1 DEVELOPMENT FORM ***

REPLACE INEFFICIENT LOGIC IN VESSEL LEVEL PCL/ TYPE = WAl/P PROBLEM STATUS: RESOLUTION LEVEL -1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE - P.

This update is a programming improvement.

DEVELOPER (S): W. L. Weaver REPORTED BY: W. L. Weaver ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BFI (G1J1) l PROBLEM STATEMENT:

The vessel level tracking logic cycles through the level data if the level model is off. This is inefficient.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

This update changes the vessel level tracking logic so that the code doesn't cycle through the level data if the level model is off.

This will speed up the code a little.

FIRST 4 CHARACTERS OF DECK AND COMDECKS CHANGED OR ADDED BY THIS UPDATE:

VSL3 i

C-87

. ~ . _ , _ .

l

      • TRAC-BF1 DEVELOPMENT FORM *** l PARTITION VESSEL STRATIFICATION FORCE BY PHASIC INERTIA PCL/ TYPE - WA2/N PROBLEM STATUS: RESOLUTION LEVEL =1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE = N.

This update involves adding a new model.

DEVELOPER (S): W. L. Weaver REPORTED BY: W. L. Weaver ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 DATE OF PROBLEM REPGRT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BF1 (G1J1)

PROBLEM STATEMENT:

Water packing problems are aggravated by an uneven radial void profile in the bottom of the bypass.

USER OR DEVELOPER COMENTS ON POSSIBLE SOLUTION TO PROBLEM:

This update changes the partitioning of the stratification force from half on each phase to a partition where the force is divided according to the ratio of the phasic inertias. The liquid is usually more massive and needs more force to get it moving in the right direction. This should help the water packing problem in the BWR/6 run by evening out the radial void profile in the bottom i'

of the bypass. The update also removes some redundant computations of the radial stratification time step.

FIRST 4 CHARACTERS OF DECK AND COM0ECKS CHANGED OR ADDED BY THIS UPDATE:

TF3E 1

C-88

      • TRAC-BF1 DEVELOPMENT FORM ***

WRITE MESSAGE TO EXPLAIN EXTRA MAJOR EDIT OUTPUT PCL/ TYPE - WA3/0 PROBLEM STATUS: RESOLUTION LEVEL -1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE - 0.

This update involves a nonfatal code output problem or an output request by a user.

DEVELOPER (S): W. L. Weaver REPORTED BY: W. L. Weaver ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BF1 (GlJ1)

PROBLEM STATEMENT:

Currently the code writes an edit for reasons not explained to the user.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

This update writes a message to explain why a mysterious edit i may appear in the output.

The code makes a dump and edit every 1800 CPU seconds whether the user has asked for one or not. This is to guard against the user losing all of a run if a hardware failure occurs and the user has chosen infrequent dumps. It also changes the logic so that the user scheduled edits and dumps appears properly and are not rescheduled by the extra edit and dump based on CPU time.

The update also calls the dump routine when a trip is activated in addition to the short edit and graphics calls because the user might want to restart after a trip activation. This extra dump doesn't affect the user scheduled dumps.

FIRST 4 CHARACTERS OF DECK AND COM0ECKS CHANGED OR ADDED BY THIS UPDATE:

TCAL TIMC TRAN C-89

$0c TRAC-BF1 DEVELOPMENT FORM oco l

CORRECT EXTRACT FOR SEPARATOR /ORYER COMPONENT PCL/ TYPE - WA5/0 PROBLEM STATUS: RESOLUTION LEVEL -1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TfPE - 0.

This update involves a nonfatal code output problem or an output request by a user.

DEVELOPER (S): W. L. Weaver REPORTED BY: W. L. Weaver ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BFI (GlJ1)

PROBLEM STATEMENT:

Currently part of the separator / dryer component extract output is incorrect.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

This update corrects the extract routines for the separator / dryer component so that the extracted cards agree with the format as specified in the input manual. Extract should have been changed when the input format was changed but was overlooked at that time.

FIRST 4 CHARACTERS OF DECK AND COMDECKS CHANGED OR ADDED BY THIS UPDATE:

WESE C-90

      • TRAC-BF1 DEVELOPMENT FORM ***

CORRECT DIFFICIENCIES IN SEPARATOR / DRYER COMP 0NENT PCL/ TYPE - WA6/E PROBLEM STATUS: RESOLUTION LEVEL -1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE - E.

l This update involves an error that caused unsatifactory results.

DEVELOPER (S): W. L. Weaver REPORTED BY: Paul Dotson ADDRESS: Los Almos National Laboratory, Los Almos, NM DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BF1 (G1J1)

PROBLEM STATEMENT:

Two errors in the separator / dryer have been pointed out by Paul Dodson of LANL. The separator / dryer component will work correctly if both separator and dryer options or just i the dryer option are activated but will not give the correct

carryover if just the separator option is activated.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

I There are two errors for this situation. The first error is the limit on the void fraction in the joining cell which gives too much carryover and the second is the fact that the velocities on the dryer is not homogeneous while the joining cell void fraction was computed using homogeneous flow. This update corrects these two problems by removing the limit on the void fraction in the joining l cell from Subroutine SPDI and by making the interfacial friction on the dryer face very large( 1.E+10 ) to give homogeneous flow on this face.

4 FIRST 4 CHARACTERS OF DECK AND COMDECKS CHANGED OR ADDED BY THIS UPDATE:

SPD1 TFIE l

4 i

j C-91 l 1

i

      • TRAC-BFI DEVELOPMENT FORM ***

FIX STABILIZER TO CONVECT NON CELL AVERAGE PROPERTIES PCL/ TYPE - WA9/E PROBLEM STATUS: RESOLUTION LEVEL -1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE - E.

This update involves an error that caused unsatifactory results.

DEVELOPER (S): W. L. Weaver REPORTED BY: W. L. Weaver ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 l DATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BF1 (GlJ1)

PROBLEM STATEMENT: l Currently a problem exists in the stabilizer whenever a noncell average property is convected across a cell boundary.

The problem was manifested as a very large component mass error.

This occurs in the separator-dryer component and whenever the level model is activated.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

The fix is to modify the phasic velocity and its derivative by the ratio of the desired void fraction to the cell center void fraction. The ratios are used in the flux terms for every cell edge and are saved at the ends of each component for use in the stabilizer and in the component mass balance. The separator option in the turbine component modifies the phasic velocities directly and doesn't need further modification.

FIRST 4 CHARACTERS OF DECK AND CONDECKS CHANGED OR ADDED BY THIS UPDATE:

QVTE ZBND ZVLQ ZVLR CNST TEEL TEE 2 TEE 3 TFl! CHN2 CHN3 FLUX J1D- PIP 2 PIP 3 PMP2 PMP3 SETB VLV2 VLV3 VSL2 VLS3 STBM BRK1 BRK2 BRK3 Fill FIL2 FIL3 IBRK ICHN IFIL INNE IPIP IPMP ITEE IVSL TEEX VSLI IVLV BLKD C-92

      • TRAC-BF1 DEVELOPMENT FORM ***  !

SIMPLIFY TEE COMPONENT LOGIC PCL/ TYPE - WAA/P DEVELOPER (S): W. L. Weaver REPORTED BY: W. L. Weaver ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID. 83415 1

OATE OF PROBLEM REPORT: 3/13/87 DATE OF LAST FORM CHANGE: 3/13/87 VERSION USED: TRAC-BF1 (GlJ1)

PROBLEM STATEMENT:

The current coding for the tee compenent logic is unduly complicated.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

This update simplifies the logic in the tee component anich makes the tee component into the other tee based com>onents such as a jet purp, a separator / dryer, a turbine, or a f(edwater heater.

4 The control flags for the various tee based componer ts were set j is several different subroutines and were passed to the lowest level

} subroutines several differelt ways. This update racionalizes the way c in which the flags are set, where they are set, and how they are passed i to other subroutines. The flags are set in the high level tee component i

routines TEEL, TEE 2, TEE 3 and are passed thru the labelled Common Block ZS00AT which formerly contained only separator / dryer data. A separate j flag is assigned to each tee based component to highlight the different options based on the tee component. This update will make it easier to follow the logic for the different tee options.

FIRST 4 CHARACTERS OF DECK AND COM0ECKS CHANGED OR ADDED BY THIS UPDATE:

QVTE CHN2 INNE PIP 2 ZSDD TEE 2 TFID TFIE TF1I VLV2 FTEE ITEE RETE STBM TEE 1 WTEE TEE 3 l

C-93

      • TRAC-BF1 DEVELOPMENT FORM ***
- CORRECT OUTPUT FORMAT FOR CONTROL SYSTEM INPUT LISTING l

PCL/ TYPE - sal /0 DEVELOPER (S): G. L. Singer l

F PROBLEM STATUS:. RESOLUTION LEVEL -3.

{

Open -- Item is completed pending integration checkout.

NATURE OF THE PROBLEM: TYPE - 0.

! This update involves a nonfatal code output problem or an output

{

request by a user.

. DATE OF PROBLEM REPORT: 1/19/87 j DATE OF LAST FORM CHANGE: 1/19/87 REPORTED BY: R. W. Shumway ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, ID 83415 VERSION USED: TRAC-BF1 (GlJ1) ,

t PROBLEM STATEMENT:

If more than 99 control system blocks are used, the output l

for the control system listing during input processing uses l

the one hundreds place digit for carriage control, giving one page of output per line.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

This can be fixed by inserting a blank carriage control

character in a format statement in Subroutine FCONT.

I THE FOLLOWING CHANGES ARE RELATED TO THIS UPDATE:

RK2 i This update corrects the control system input when more j than 99 blocks are used by changing a format statement i in Subroutine FCONT.

I i

! FIRST 4 CHARACTERS OF DECK AND COMDECKS CHANGED OR ADDED BY THIS UPDATE:

I

FCON i

i I

i

! C-94 l

      • TRAC-BF1 DEVELOPMENT FORM ***

REWRITE THE INPCVI ROUTINE FOR GENERALITY AND EFFICIENCY PCL/ TYPE = SA2/P DEVELOPER (S): G. L. Singer PROBLEM STATUS: RESOLUTION LEVEL =1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE = P.

This update is a programming improvement.

DATE OF PROBLEM REPORT: 3/3/87 DATE OF LAST FORM CHANGE: 3/3/87 REPORTED BY: G. L. SINGER ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, Id. 83415 VERSION USED: TRAC-BF1 (GlJ1)

PROBLEM STATEMENT:

The portable INPCVI routine does not allow special characters such as &. It is also slow compared to the original version. More seriously, it appears not to be able to handle input card identifiers of greater than 99999999 on the IBM computer.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

The INPCVI routine needs rewritten for efficiency making use of FORTRAN-77 capabilities; in the process the use of special characters should be allowed and it will be made to work for card numbers up to about 250000000.

THIS UPDATE REQUIRES THE FOLLOWING INPUT CHANGES:

Alphanumeric data enclosed in quote will be allowed to contain special characters.

When the portable INP free field input routines replaced the original INP routines, all of the INP routines were largely rewritten except for the INPCVI routine that is miniminally modified from the Japanese version. While the Japanese version is quite portable, it was originally written in Fortran-66 and could not take advantage of Fortran-77 features that could have made it more efficient.

The INPCVI routine will be rewritten using character type l data and internal read-write statements and will process j an entire field at a time instead of one character at a '

time. In the process, special characters will be permitted in alphanumeric fields.

In addition, the Japanese version did not need very large card identifiers for any of the code to which it was previously applied. This new version will extend the maximum size allowed for card numbers.

Also, see the problem described in update SA2 4 C-95

      • TRAC-BF1 DEVELOPMENT FORM ***

CORRECTIONS TO ALLOW THE IBM VERSION TO COMPILE AND EXECUTE PCL/TYN = SA3/V DEVELOPER (S): A. Haghighat and G. L. Singer PROBLEM STATUS: RESOLUTION LEVEL -1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE - V.

This update involves a code implementation problem or system dependency problem.

DATE OF PROBLEM REPORT: 2/3/87 DATE OF LAST FORM CHANGE: 2/3/87

REPORTED BY
A. Haghtghat i

ADDRESS: Pennsylvania State University, Nuclear Engineering Dept.,

231 Sackett Bldg, University Park, PA 16802 VERSION USED: TRAC-BF1 (G1J1)

, PROBLEM STATEMENT:

l EG&G converted TRAC-BF1 for use on the IBM computer as best it could without actually using an IBM. Pennsylvania State

~

University (PSU) has a contract to check out the IBM version.

! The code as given to PSU would not compile or execute for several reasons. This update, when completed, will contain all of the changes need to execute on the IBM computer at PSU.

The following errors prevented compilation of the program:

some parameters were not defined for the IBM, single precision constants and double precision variables exist in many intrinsic function such as MAX, explicit REAL statements override the IMPLICIT DOUBLE PRECISION statements causing allignment problems, data statements may not appear twice-for the same variable on the IBM, some NPA routines not currently used on the IBM are not compilable on the IBM.

The following errors prevented execution on the IBM: some timer routines were not bypassed (or dummies used) for the IBM, the variable ITITLE should be changed TITLEI to make it floating point, and various NPA (Nuclear Plant Analyzer) routines should be skipped on the IBM version.

Also, the user must remember to turn off graphics, traceback and timer routine by setting N0GRAF - 7, NTRACE - -1, and IST = -1.

1 In addition, users of Historian must be aware of differences between it and CDC Update with respect to treatment of

  • WEOR cards, and make sure that the coding for the CDC only other computers only is not compiled. Also, the input file must be declared with a word length of 90 (or Subroutine ALPHTN must be changed so that the format reading 90 characters reads 80.

Also, see the problem described in update SA2.

C-96

PAGE 2 0F PCL SA3 PLAN: Completion of this update is awaiting successful checkout of TRAC-BF1 by PSU.

THIS UPDATE REQUIRES A CHANGE TO THE MANUAL:

Additional instructions, hints, and control card listings for IBM usage of TRAC-BF1 will need to be added to the manual.

l i

i i

i i

i I

I C-97 i

      • TRAC-BF1 DEVELOPMENT FORM ***

'HAVE INPUT CHECKING CHECK JUNCTION NUMBERS FOR MATCHES PCL/ TYPE = SA4/I DEVELOPER (S): G. L. Singer PROBLEM STATUS:' RESOLUTION LEVEL -1.

Open -- Work on this item is in progress.

NATURE OF THE PROBLEM: TYPE = I.

This update involves a user input problem or code input checking problem.

DATE OF PROBLEM REPORT: 2/04/87 l' DATE OF LAST FORM CHANGE: 2/04/87 REPORTED BY: R. A. Shumway ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, Id. 83415 VERSION USED: TRAC-BFI (GlJ1)

PROBLEM STATEMENT:

Input processing will fall with a system error (mode 04b on the CDC) instead of a TRAC diagnostic message when user has '

an input error in junction numbers.

USER OR DEVELOPER COMMENTS ON POSSIBLE SOLUTION TO PROBLEM:

When the correct total number of junctions is entered, so that the code does not fail gracefully on this message, but when a junction ~  :

number on one. component is not matched by any other component, the code r will fail ungracefully. That is, instead of the user getting a TRAC '

generated message explaining what he did wrong, he will get a system ,

error (a mode 04b error on the CDC). The specific case in which this l was observed to happen had a fill as the component that whould have been joined to the junction and it had JUNI =0, that is, a disconnected fill.  !

t I

I i

l C-98

cco TRAC-BF1 DEVELOPMENT FORM 000 BF1 Manual Corrections and Additions through Feb. 1987 PCL/ TYPE - SAS/D DEVELOPER (S): G. L. Singer PROBLEM STATUS: RESOLUTION LEVEL -5.

Open -- Item is completed pending integration checkout.

NATURE OF THE PROBLEM: TYPE - D.

This update involves a documentation deficiency.

DATE OF PROBLEM REPORT: 3/17/87 DATE OF LAST FORM CHANGE: 3/17/87 REPORTED BY: G. L. SINGER ADDRESS: EG&G Idaho, Box 1625, Idaho Falls, Id. 83415 VERSION USED: TRAC-BF1 Manual PROBLEM STATEMENT:

Many additical graphics variables were added to the code after the BF1 manual was prepared. Also, the structure of the records on the TRAC Update Program Library was changed subsequent to the manual. Various other manual corrects need to be made.

THIS UPDATE REQUIRES A CHANGE TO THE MANUAL:

GRAPHICS Section 11.2.9 should be replaced with the following.

      • start of change ***

II.2.9 Plotting Capability A wide variety of plot data are generated and stored by TRAC-BFl. The set of data points for a single item of data (such as the pressure in component number 5 in cell 3) at a series of time points is called a channel. Each channel has a unique identifier generated by TRAC-BFI that is used by auxiliary plot programs to retrieve the data. This channel identifier consists of up to four pieces. For component related data the four pieces are the variable name (alphanumeric), the component number (a two digit integer), the axial level or rod group number ( a two digit integer), and a cell number ( a two digit integer). The alphanumeric and numeric parts of the channel identifier are separated by a minus sign with no blanks allowed. Each of the integers must be a two digit number with a leading zero required if the number is less than 10.

For variables that are not component related, such as the time step size (DELT), the number "001" is used for the rest of the identifier following the variable name and minus sign, except for the control <

system. For the control system the number used is the user control l block number, left filled with zeroes to give 3 digits if less than 100.  ;

I C-99

PAGE 2 0F PCL SA5 The user may obtain a complete list of the channel identifiers created for his particular TRAC-SF1 run either by specifying N0GRAF - 1 on the CHECK 0UT card (by setting the seventh variable to an integer 1) or by running the auxiliary program BSUM on his HUNI output file. The only exception to the use of three digits for these variables is that the time channel is simply "TIMET-1".

Some examples of channel identifiers are "PV-020001", "MODCHANW-123456",

and "T0TPOW-001". In the first case the identifier indicates the pressure in component number 2, cell 1. The second indicates the channel wall heat transfer mode in component 12, rod group 34, cell 56, where the component must be a CHAN. The third indicates the total reactor power.

Lists of the plot variable names generated by the code are found in Tables II.2-8, II.2-9, and II.2-10 in three different formats. The first is an alphabetical list by the plot variable name. All duplicates i due to the variable being used by more than one component are eliminated. The plot variable names are then listed alphabetically showing the components that use the names. Finally, the plot variables ,

are listed according to the order they are encountered in the code -

logic, showing them component by component.

In the tables for the plot variable names the headings have the following meanings. "Name" is the plot variable name. " Description" is the TRAC internal description that is used to identify to the TRAC programmers and manual users the items to be written to the graphics file. The " Units code" is a number used by the ISDMS plotting programs to supply a label for the y-axis. The part of this y-axis label that is the units is shown under " Units value" without the rest of the label that may be found in the MAGNUM manual. The heading " Rod / Level" indicates whether part of the channel identifier must include a nonzero number for the rod, in the case of the CHAN component, or for the level, in the case of the VESSEL component. A nonzero value indicates that, for the CHAN or VESSEL, this number must be included. Thus, a channel identifier of the form NAME-CCLLNN would be used for variables with a rod or level, and an identifier of the form NAME-CC00NN would be used for the other variables, where NN is the component number, LL is the rod or level number, and NN is the cell number. The heading " Type" indicated to programmers where the data is in the TRAC data base (0 -

selected variables, 2 - real variable length table, 3 - integer variable length table, 4 - array data, 5 = rod or level data, 6 - control system data).

Programmers adding new plot variables should be aware of the reason for using exactly six or three digits (or one for TIMET-1) in the numeric field. They must also realize that the all channel names must be unique C-100

PAGE 3 0F PCL SAS for all characters prier to the n'th p sition whnre n is the position of the left most minus sign. (For example, ABC- and AB- are not unique to the'left of position 3 and should not both be used as identifiers.) The reason for these requirements is thav. the tollating sequence on the CDC and- ASCII collating sequence are different for non-alphabetic characters. Therefore, to maintain machine indeper. dent collating sequences (and identical coding regardless of machine), the above rules must be observed. i

      • end of change ***

Tables II.2-8 through II.2-10 should be updated to reflect the following-additional variables.

      • start of change ***

, NAME DESCRIPTION UNITS UNITS ROD /

CODE VALUE LEVEL ALPA ALPHA AB0VE LEVEL 95 (NONE) 0 ALPB ALPHA BELOW LEVEL 95 (NONE) 0 j

ALPC CORE LEVEL AVERAGE V0ID FRACTION 95 (NONE) 3 BORRC CORE LEVEL AVERAGE BORON CONCENTRATION 350 (NONE) 0 l CTRLF CONTROL FRACTION 77 (NONE) 0-EGV GLOBAL EIGENVALUE 77- (NONE) 0 i ERMC COMPONENT PERCENT MASS ERROR 56 PERCENT

, ERMSCT SYSTEM PERCENT MASS ERROR 56 PERCENT FRICIT THETA INTERFACIAL FRICTION COEFFICIENT 208 (NONE)

. FRICIZ AXIAL INTERFACIAL FRICTION COEFFICIENT 208 (NONE)

FRICIR RADIAL INTERFACIAL FRICTION COEFFICIENT 208 (NONE)

HAIL H.T. COEF.* AREA, LIQUID TO INTERFACE 155 W/K 0 HAIV H.T. COEF.* AREA, VAPOR TO INTERFACE 155 W/K- 0 HLWALLI INSIDE WALL H. T. COEF. TO LIQUID 155- W/M2-K 0 4

HVWALLI INSIDE WALL H. T. COEF. TO VAPOR 155 W/M2-K 0

HLWALLO OUTER WALL H. T. COEF. TO LIQUID 155 W/M2-K 0
HVWALLO OUTER WALL H. T. COEF. TO VAPOR 155 W/M2-K 0

, LVLEV~ LEVEL VELOCITY 235 M/S 0 MODR00 R00 HEAT TRANSFER MODE 51- (NONE) 1.

MODWALLI INNER WALL MODE 51- (NONE).0 i PDHL DECAY POWER HYDRAULIC LEVEL 351- (NONE)'O

! PDHN DECAY POWER NEUTRONIC LEVEL 351 (NONE) 0

, QCONV CONV HEAT FLUX FOR GROUP 388 1 4

QRAD RAD HEAT FLUX FOR GROUP 388 1 QTOT TOTAL ~ HEAT FLUX FOR GROUP 388 1 i RADHL

  • RAD HTC LIQUID LIQUID FOR GROUP 155 W/M2-K 1 RADHV RAD HTC LIQUID VAPOR FOR GROUP 155 W/M2-K 1 4

RHL CONTAINMENT COMPARTMENT LIQUID MASS 229 KG- 0 R0DHL R00 HTC LIQUID FOR GROUP 155 W/M2-K 1 RODHV ROD HTC VAPOR FOR GROUP 155 W/M2-K 1-RODTIN ROD PLOT TEMPERATURE FOR GROUP 84 K 1 RODTL R00 SURFACE TEMPERATURE LIQUID, FOR GRP 84 K 1 RODTV R00 SURFACE TEMPERATURE VAPOR, FOR GROUP 84 K 1

RPOWL PROMPT POWER HYDRAULIC LEVEL 152 (NONE) 0
RPOWN PROMPT POWER NEUTRONIC LEVEL 152 (NONE) O C-101 4

- _ ... _ ,= m_ . -. _ , , - - - _ , . , , _ . . , , _ ,. _ ,_ _ _.. .. , ..- -, __ - - , -

,,m . .;..-._

PAGE 4 0F PCL SA5-TFC AVE FUEL TEMP 84 K 0 TMC AVE MODERATOR TEMP 84 K 0 TSURF R00 SURFACE TEMP 84 K 1

.TWALLI INNER WALL T 74 W/M2-K 0 1 ZPOWN NEUT. AX. POWER FRACTION 77 (NONE) 0 ZPOWR AXIAL POWER FRACTION 77 (NONE) 0 t

      • end of change ***

In these tables,- delete variables' ALV and CHTI. Also in the table, change the word " internal" to " interval" everywhere it occurs.

On page 271, change'P-201 to PV-20001 and TOTPOW-1 to TOTPOW-001.

, In Table II.2-12 change " ALPHA-30507 MFLOW-502" TO 2

" ALPHA-030507 MFLOW-050002" and "HL-20304" to "HL-020304".

Change to section III.4.6 of the TRAC-BF1 manual by adding the following l information at the bottom on page 320.

      • start of change ***

i i' Preload assumes that the memory needed for the input segment exceeds that needed during initialization. However, this may not be the case for runs that required large original input decks and short restart i decks, since short restart decks require very little memory for the i unprocessed input data. Therefore, the user should always use the

larger value of LAST calculated for the original and restart runs.

! Failure to do this may result in TRAC failing with a message that there 4 is insufficient small core memory.

! .It should also be noted that there is no check for sufficient memory for the temporary space needed for reading the restart tape. This has never i been observed to cause a problem, but it is conceivable that a run could be devised that would cause memory to be exceeded in subroutine RDREST.

      • end of change ***

i Change to section III.7.3 of the TRAC-BF1 manual:

      • start of change ***

4 111.7.3 STRUCTURE AND USE OF THE CDC UPDATE LIBRARY 1

The record stra-ture of the CDC update program library is now organized l according to code portability considerations. The source cards for the TRAC-BF1 program are maintained on a CDC update program library in 13 parts that are separated on the program library by *WEOR cards. When a full update is made (UPDATE,F.), these 13 parts become thirteen separate ih 4

C-102 i _. _ _ .__ _ ._ _ _. _ _ - _ . _ . _ , _ . . _ . . _ . - _ . , . _ _ . _ , . _ . _ . _ _ _ _ _ . . _ _ _ _ _

PAGE 5 0F PCL SA5 files on the compilo file, as follows. N:;ts that cany of the files are empty since these versions are not'yet available. We would appreciate your sharing any necessary machine dependent coding with us so that we can include it on the next transmittal tape. Please note the word "necessary" since we are attempting to eliminate as many machine dependencies as possible.

1. Main program and block data
2. The bulk of the TRAC subroutines that can generally be expected to be machine independent.

l 3. TRAC documentation routines (definitions of l

variable names, subroutine purposes, etc.).

4. Reserved for maximum portability coding.

(eventually this will allow the use of binary 1/o instead of buffered 1/o, etc.

5. Coding specific to CRAY CTSS operating system.
6. Coding specific to CRAY C0S operating system.
7. Coding specific to CDC NOS operating system.
8. Coding specific to the IBM.
9. Coding specific to the MASSCOMP.
10. Coding specific to the VAX.
11. Coding specific to the INEL CDC NOS system.

(system text)

12. Coding specific to the INEL CDC NOS system.

(compass routines)

13. Miscellaneous routines of possible interest.
      • end of change ***

V0ID FRACTION TRANSITION ZONE WIDENING.

The void fraction change necessary to cause a transition from bubbly flow to annular flow was set at 0.1 in previous code versions.

This transition window has been widened to 0.25 in the current code.

The change was done by altering equation I.1.3-3 in NUREG/CR-4391 page 21 to read:

ACA = (1 + 4/ gamma)/Co - 4/ gamma - 0.15 The new end of the transition region is:

ACA5 = ACA + .25 C-103

~ -. - . _ _ _ _ _ -

PAGE 6 0FLPCL SA5-

!. This transition window is used in both the interfacial shear and interfacial heat transfer subroutines. A new feature of this code.

- version is that the window is also used in the wall heat transfer logic.

SUBC00 LED BOILING USE OF THE TRANSITION WINDOW N

The void transition window is also calculated in nucleate and

transition boiling. Between a void fraction of ACA and ACA5 the subccoled boiling flux is ramped to zero. This change helps the use of the subcooled boiling model be closer to its developed data base plus it prevents direct liquid flashing when the liquid fraction approaches zero.

INTERFACIAL HEAT TRANSFER FOR FILMS When un-entrained water is calculated to be in a cell it is assumed to be in a film on the wall. A Stanton number of 0.02 is used to obtain the ' interfacial heat transfer coefficient for both the liquid and the vapo.. This is described on page 28 of the NUREG/CR-4391. Equation I.1.3-43 is the equation for the film velocity used in the Stanton number. Three problems with this approach are: 1. the Stanton number of 0.02 was suggested for direct contact condensation on a water jet, 2.

The velocity equation is for laminar films, 3. The velocity equation yields values of thousands of meters /second for thick films. The solution to these problems was to adopt an equation from " Interfacial Heat Transfer in Countercurrent Flows of Steam and Water", EGG-RTH-2495, January 1987 by Mohamed M. Megahed. The equation for the heat tranfer coefficient from the liquid to the interface is:

Hil - 0.0771*Cp*M*C/D C = (gc*(rol/visl)**2)**1/3 0 - ((rol/r.v)**.5*Cp*(Tsat-TI)/Hfg)**2/3 o

  • Rey **2/3
  • Cir Rey - 4*M/(Cir
  • visl)

. Cir - 3.14* Hydraulic diameter M = flow rate of un-entrained rol - liquid density rov - vapor density visl- liquid viscosity

Cp - liquid specific heat at constant pressure j Hfg = heat of vaporization
gc - gravitational constant

} Tsat= saturation temperature T1 = liquid temperature

! If the liquid is not subcooled:

J Hil = 1.e8 i l The vapor side coefficient is the maximum from the Dittus-Boelter j equation and Nusselt number equals 4.

1 1

i j

i C-104

PAGE 7 0F PCL SA5 MISCELLANE0US CHANGES The following code changes affect the contents of the manual in minor ways and are probably not of interest to most users, but are included here for completeness.

      • start of change ***
  • / #*WJC N LOOSEN WATERPACKING CRITERIA
  • /
  • / ##ENGPR W L WEAVER
  • /
  • / THIS UPDATE LOOSENS THE WATERPACKING CRITERIA FOR PRESSURE
  • / DECREASES ( STRETCH ) S0 THAT THE PACKER DOESN'T COME ON
  • / WHEN NOT NEEDED. A PRESSURE DECREASE CORRESPONDING TO 5.
  • / DEGREES OF SUPERHEATING BASED ON THE LIQUID TEMPERATURE IS
  • / ALLOWED WHERE 5. DEGREES WAS CHOSEN TO BE CONSISTANT WITH
  • / THE TIME STEP CONTROL ON LIQUID TEMPERATURE.
  • /
  • / ##END
  • /
  • / #*RLH C(CNT) DO NOT AVERAGE WALL H.T. ON FIRST TIME STEP
  • / ##ENGPR R. W. SHUMWAY
  • / THE INITIAL WALL H.T.C'S ARE ZERO AND THEY SHOULD NOT BE
  • / AVERAGED ON THE FIRST TIME STEP UNLESS IT IS A RESTART AND
  • / THE OLD VALUES ARE AVAILABLE. IHTAVE KEEPS TRACK OF THIS.
  • / ##END RLH
      • end of change ***

C-1G5

. APPENDIX D CODE ASSESSMENT MATRICES y l This appendix contains the code assessment matrices that currently

!' exist in the' International Code Assessment and Applications Program-  ;

(ICAP). The matrices summarize the code assessment activities planned by

, each ICAP participant for.the USNRC major thermal-hydraulic codes.

_ Tables D1, 02, and D3 present planned assessment matrices for the RELAPS, 3 TRAC-PWR, and TRAC-BWR ccdes, respectively.

l-

}

I s

j l

l 4

I l'

1 a

1 r

k D-1 f

E -

E x E 3 =

2 5 W W Q- g-

-: = E W e a .

g w = =

5 g N I E I E

~=

E w W-E . . . W . . = W

-- . g- g a a . = c W W w == w w -

= s

  • 4 $5 mE $4 E "" E E 20 EE m g E5 8 5

E= 'M 5 W w" -m S- 5 E -

5 . 5~u 5 5 5 w

-- g = a t. . - - - -g - - -- u w - --

g~ = mw Sg gg gy W W W gg y a a g g b b E Eh EE es -

=E e- -

= W s a w =- = W W e E -w g g'e-MI.e--

WC wo E WB 6 ==

ES ma

.85 na 85 nu e

= = 5 E:-= E- - - - o m

o b ~

N Ud

=

- m m- mm PI wt l G W

Sw E 8 E E.

8 8 8

%% g d d d d= d j w- = a a M m =

5 E . 3 ~

~ " "

E - n EE E E E E E E E E s& &

m e m m W k a

ka E a

a E

a E

a kaa E a E ay oa E k a a am t amg E a

m a a e

h 2e E a a a m

a a

S WW W W W WWWWW WWWW ..

  • W- WW W n.

t: E E m: M S-3 - - - - - - 2 2 - -

C" -

E E 5 5 -X 5 5 5 " "

5 E B"

c" =

d a d

=

d a

d e

a d d t d

= = a - -

d g g W W g g d d

- 3 m d

a o '. EW

  • E W W W WWW E E WW W E.W I v=u -- 6 6 w

m m = - -

1 ..

Mw O-W.

J : - -

m*

g

=

a e n  ?

-? 3m 3 m

~ a a - -. - -

& Y Y w- sa 1 M

8 2

=

d

=

C

=

d =d n Em *

= m C E E E

  • m w

= .

E a

Ea 8a % a

= = , - = =w = = w = = =w w w w w w W w w C 5 6 5 5 d 5 5 5 a d M E " E" E E E E E E E "

1. E E E E E E E E .t 5 W- E
6. m w

h b

6. h b

h h

w.

nn

=

to d

n/b

= W s/>

T- ==

- w w

w w

w w

w w

w w

w

. I M $ $ gz . $.

4 W . . . . . . . . . m

=

x

=

E b h. h b. b h b b h a b . . =. =. . . =.

U 2 E E E E 2 E 2 2 R R- E & 2 P .E n

5 5 '$ $ $ $ g

.- 2 2

m 5

0 2 . 2. .2 m

2 .I _!

W =

n

= 5 . . 5 R I

%= 2 2 a a a a a a d a 'd a a d d d d d- d. W 23 5 WW d a d i a a a a d 2 a C C C C C C D-3

i: }; ) ' ' !tl H

1 1N S 1 3 0 N N 3 1 0 H 3 v8 N

  • ll 33 A
  • S! I' 0 1 13 f* G 1 V N

3 3 9 t O N M W3O A' v

A' V

C h

3 3 W' 30 W' A3 8 3 3 b d 3 6

3 3 e R aW 3

8 O G G G 0 G G t G G G 0 O G 3 1

8 6

d 6

H 3 1

1 1 9 3 0 d6 3

0Q S C N D i hu Ml Dl Dl 3

N I Ml Ml hl Dl MI N i h D pu I

3 N

l l 9 9 N 3 3 A 9 8 l W W W k m 0 3 3 2 N

1 V 1

3 1

1' 1 0 1 3 n

g s 3

a l

8 3

u s

3 H H3 3

U 3

8 3

H N H 3 3 1 u

1 d

3 d

3 3

3 1 1 3 3 V 3 1 0 0 3 3 1 1 1 1 1 1 1 1 1 1 3 1 J 1 3 3 E 1 3 0 3 0S 1 1 l 1 1 3 3 03 0 N 2 3 3 3 W11 d 0

3 1 3 1 N1 0

3 0

3 3 M 3

0 3

0 M 3 0

3 0

3 0 8 0 M 0 0 0 3 0

H 0 d 0 5 0 3

l' 0 m*

3 33 3 3 1S 1S 1 1 A

~ 1 5 8 1 1 J 0 N 3 3 3 3 3 4

3 d

3 4

3 0

3 8 M 3 8

3 d 6 M 6 S 1 J J 1 V 1 ] 13 4 6 8 9 V N1 6

3 3 1 H 5 T kl i d S

3 3 0 NO S N 0 S 0

Z GI 03 0 1 Z1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 0

1 W

3 i d W 1 S J S 0 3

A N

D l v V V u 3 3 3 3 3 0 0 0 d d 1 I A M 8 M l1 S 1 1

_ 8 3 1 l Z 1 1 0 1 1 9 - - - -

5 2 - - - N 8 u 8 3 1 4 8 Z 3 S S 5 1 N 5 1 1 S S S S S S S S S S S S S S S S S S S S S S a W W a J V W A1 J V

i V W d1 d v W W 8 W m W a A W R 1

M l 3

1 1 3

l 3

1 3 3 1

3 1

3 1

3 1

3 3 3 1 1 3

1 3 3 1 1 3 3 1

3 1 l 3 3 1

31 3 0 3 d 4 8 8 8 5 8 8 8 8 8 5 8 d 5 1 d 3 5 a 3 4 1

(

P. . 5 5 $ $ 5 a 1 3

1 3

1 3

1 3

1 3

n. 3 3

3 J

3 3

3 3

3 J

u J 3

3 3

3 3

3 3

3 3

_. ug pI 1 N

3 W l v h 1 9

1 V

1 V

1 V

1 7 V T 1 V

1 V

l v

e 1

V 3

1 V

3 1

3 1

V 3

1 V

3 1

t l

v n

l V

3 I

V d

1 V

3 WWWW 1 W 8 8 8 v 8 3 5 B u 9 8 V 9 9 9 3 9 3 oi S 9 9 9 9 3 9 3 9 3

9 9 9 3 3 3 W 3 3 3 3 av W3 u A 1

3 1

N 3

1 k

3 1

N 3

1 N

3 1

N 3

1 N

3 1

N 1

N 3

1 N

3 1

N 1

N 1

N 1

N J

3 4

3 4

3 J

1 J

3 1

N 1

N 1

N 1

N

_. )u l 1 I l I I I I I I I 1 I I I S S S 5 S I I I 1

- *i IS 03

- 1 3

_ 7 N N N N N 8 a l 3 3 3 3 3 V.. l 1

9 9

9 9

9 a 1 1 1 1 1 1 S S $ S S S 5 S S S 3

1 5

I S

1 5

1 1

5 1

J 1

3 3 3

V 3

0 3

0 3

0 3

d 3

J 3

J 3

3 3

J 3

4 3

d 3

J 3

d J S

3 3 3 3 3 9

0 0 4 0

. J 1 1 1 5 S 5 S S 5 S S S 9 9 9 9 1 1 1 1 h

C @ O O C C O C O D 0 D a

C D O C 0 0 O O G C h

_ H 8 d h d V

a n

d V

W u

b d 1

8 W V i

e s t D 3 M 3 W S 5 t W

I S

W 1

S W

1 H1 M

3 W b G N

o M

d N

0 Q O N

WM C

r 0 D J

N 3

J 0 D A

d O H h 3

x 1

0 H

1 0

H 0

H d

V S

4 V

S a

s A #S a S S d

V S

J S

V W S

W S a Rs S V

S V

S S M

S V

S M S M S

_ 1 * *

]

V H H

H* 9' S S' 9' 9' S' 9 9' $

S S S

S S

$ S O

9 9' 1 * *

  • C u' B* d* 5" b*

W W d* u* B

_. 3 3 a W W" W* d u W* W W* $* W W* W* J W W W d d W*

S G

_ Nm e

e 5

A 1

C W

Q N

V M

A A A A A 1 A A A 1 A A A A A A A A A i

N a

i ti 1

N i

n t W 1 V W l v

1 V 4 1 l V V 1

1 h v l

h W i V r l v V 1

v

_ uo nm 3 3

l j

J t

s 1

1 1

1 I

I I

I i

l 1

1 1

1 1

1 i

l 1

1 1

i l

1 1

1 1 II t!

i l

I I

i l

l i.

l,  ! iijl! - ft ; f i  !,

{l! if [

-!I

,4 i y_'-.=c-en E te 8"

.=

a W a = =

daW s- a m a

=E -

g - a =

1 g? = g"

  • g a 2 E "

E i

gE WE w w- =w= w . =w .w w w = .

e --

o o N W _

g =

=

's-a E a w r ww w

= . o e- =5 -w a a s -e --

- r= ! .m

! = =rr=r = g! a

= =

= ,

= e E = = = r r r 55 E E E E E E E E 55gm i s s i 5E E E

=g- sie S=E E= = m a r e m W W = = = .-

~ - = e = w w =

ba $$ $5 2 2 2 2 E 2 2 2 -

E E 5 $ $ $ $ 5 E E -

E W

= g WE

=C $

f E

Y

.I W = 0 W t a

a E .n. a a a a k % a k k & E k. . .% k a a a a k

g . . . . .

. k & % 3 4 B W W WWWWWWWWWWWWW a a a a a a a W E W Wa T W W a a a a n

V.

  • 5 C 3:

C S M 5" ----

X X X 5

- - e e"

- 5 =

c- = d d e e e e e e e" - e o" ~

3 & = = = = = = gg u~ Ir W W Eg. -g E g

-~= == = = . Emm a m

o. .

W

a. =

OS -

h* $

G Y

m a a a E = E m m I $

J e g g g a y g g n a a a a a agIaaai1a

= = * = a a a* gggB

, a a a a a a 5

4 i

=

a a =

I 555iiE y a a a a a e a a a a a EIg g . .

v a

g

..sg n E g g M N O N O O O ja3 -

5 SC 6 d d * *

  • 1
= = = = d =d =d =d =d= d= d= d= d g E g. . g

= .

W

= E E g We e

. W eW D-5

L 4,-- *- - + . 4. -a 4 -4_ A - ,e i. = y. L a,.e-

+ - m s l.

?

i E 1

-m R

= . . . I *E .W. .W .. . . ,. . . . . . .

WWWWWW

=5 . . . .

-W W W W W =,~g -W W g = - m

g-W a

WWWW

-=WW m - W a v 2 . m s- a s

= *

  • M W W W - = W W W - W W W a- a- r n = - n-.m um t
  • W :- W a =- W -=W - -

=

WE E n W W x W W W W W W W W W W W W W Ea E E E E -

SQ WWWW E W W W W W W W W W W W W W "a W W W E

$W "

2 2 2 kE 2 2 R -2 R R R R 2 2 2 2 2 k E ,2

  • 2 2 W

r = .

3 EE w-

st!

i i

f 1 6  ? ~ ?

=

w n.

W T'

a

= =.a a

i a

g - * - r - = - w - e - - - - - - - - - - - - - - - - - . -

1

< m * - W w W w W w = w w w = r w w e w = w A

I T

1 e:

s* - -

C" 3 5 9c .~ 3 -

d d

- .a O- kg- "E  ;

1' o - -

v, g G M

, C3 W 1 -

LAJ

  • J C w w
c. 2

.EX3 -

W d W H -

w 5 E a 4

i e

i E 5 O O E E E sO E '

s ss 5s ms asg a - m 5g sa g s ss a g -

e e i

+

s v 5s y .

a" -s

= e = = = - =m ee em s = =s g = W3

- s

= = W W 3 des i 5 m g g g 9a a a a a a a a a a a a a a 4 4 m e 3 h 5 m 3 5 & h m w w y w i E - . . E E E = !S 5 i i i i i i i i i i i i i i i i i i i i i i i i f eft 4

s*

! .~ = 5 5 5 = h M E a 5 = # 5 4

se
a e c : s : e a a e e a a a w E E F W" 5 g" 5 W W W W W W W W - W W W W W W W

.,~2 g g - - - - .- - - - - - .- ,+ - E.. E,, . E.

. ra e, - - . W W W W W W W W W W W W W W W S = n a a D-6

- . - , , - . , . . . -- ,e -------c-- ~-ma ., ---n, ,,-rw--,---,, -.,---.--,>-~-evn--- -.-----,w-~

me - w -

E W W E E

m a a - =- .

= a = . . o W

m - E E E X X

2 Ea *

- . . . . . n W W W W W .* W = ' 5 :. C *

  • E- W
  • g .

E 3 "E WWWWWW -

E g- E E E F E E M - E ~

g E 5 E g E

E 5 g

5 5 5 5 5 -

e = ed - h9 c

5 g

5 e W g g -

c g W = E-w W =e o E t o

u = o E g m

s -

o o

= r n

  • w *
  • 3 g

" " E" go " g 3 g S S E Ea E E k L EE E 2 2 2

  • 4 E E ** R O "

E

=

i. EE CC G

- B

=f WR E

e t N = 3 3 4

g - -

y . - - - - - - - ..

M Y $ Y $ Y Y Y n

T.

E N 2 5 W M

$:E c e = = = e =

w- b 2w w b b w

& w2 w p" _

- C o"

E

~

W WWWWW W W WWW

! I e 3e3 g3 y3 3 u-

-- ==

It m a = a a a y

s

[ [ [ !

a m a m A-o" m- g d e W M3 - . .

8 bI Ed WW E

E 7 7 ? - - 3 3 a E W

  • w E E = =w =w ba b

s E E W

  • g E g = & E- 3 E E EEEEEEEh W

= l l li....._

8W s g g e aas aa

- m m , ,

=

. . . = = = e e m EWWWWWW E E & E B 'M M W W W W y

M

= ,

y aa y y M

, , =

gE B

E gE E 2 L

a a a a a E B B l

o E E a a EaE E aE aE Wa W a a .

E $ $ $ $ $ $ $ $ $ 5 $ $ $ $ $ $ $ $ i $ $$ $

.- i a

f H I E E e e a e e s- -

" }

.E a a a e a a a E E E E EE E E E D E E 5 5 s EE E E b E E E E E E 5 E 5 5 5 5 E 5 5 5 5 5 5 i D-7

2 1

O. -

& 3 -

E v E W

-c -

E g

  • E
g. ~

1 a

me g d E w - - -

1 -

E

= 5a M 1

W .W W g ~

_ d WWW W d W E W W W W W W W E E - I d . O g E= = " d = ~$ " O -G- = E E E E E E g

= = = = = = = e

- s .e = 5 ad -

W SE

~ W 2e 4a W Q C-

  • O- -b- O

- ~

~

~

=

QC

.. a

~ ~

  • O- ."==S == == * -C g t m -
  • g S W W W .a ~8 g W W d A W- s a E
  • W W E
  • M W y.

4

$g 2 2

  • S $$ $5 $$ "

S S $ $ E' $ $ E S " R D R S E z

W G

! WE CC

=

  • n - . . n n 3

$I b b b b b b 5 1

p n n e p p n g E 2 e

%nd o%n n

% E

=

k E kn %n p E1 Ed 3 n& kno e

=  % &' a a d

a e a

u 5 W' = = W' d

= W" d d d

. . W" h - W' = W' d = d

  • d. . c. d i

d n

3!s -

E e,ue o

!?

m c" X X X 5 5 5 5 5 5 I X 5 2 = 5 - w 9 ~- 5 d d d d d d d d d d d d d d d d d d W W 1

c" M 5 3 5 = = = = = = = = . . . . . . . g a

.b=.

!E -

N 5

- 5- fI fI e I E fff8 I f'$.5 I $

4 M c" -

W .

J =

e m C 5 E m

  • w H
  • ~

d - - - - - - W -

E e*

  • E E W

B a

B s

B a

E E B Bs B

= = a a E .E m c E Eu E E.

u .S W 1

I l

= = = = = = = = = = = = = = = 2 W 4

5 5- 5 5 a

5 a 5_ g W o Wo Wm "a g g g Wo Wm W Wm W Wm W :. : - - m E

  • E M -* a"E "#

M E E E E E E E E E E E E E E E E 3 ~

-  ? ? ? ? ? ? ?  ; ? ;? ?; ?;? ?; ?;? ; ;  ; C ~ e $2 W d. a d d a & C ~ ~ ~

5 8 E E E E E E E E E E E E E E E E E E E E E F.

5 E

- K a a -

E' a3 - d d d d d d e = 5 W W W =

g KEE E2 5EO

= K NE g_ g g g 5 5 i se g = = = = e = E E g

  1. .. g- ]

E .. ,

29 w 3 5 5 5 5 3 3 3 2 3_ _2 #_ 3 0 M_ E- O E 9 D-8

a 1 g .A 5 p.4. m _. _.s . e_ wen r> , . .- , ,.,,.._ ._ . . - --+_

l I

m' E-g = =

=

B ".

D

- ~

E- 8 8 EE ru

$ $ ~ *

== g =e a wR m w w-

-G E

EE 4

a oc n En W

, = - .A =

1, i

i i

1 w - . . - - -

e W w w w w w w E a

i 4

^

t.

1 W3 E E C U U M N E s2 e M = e e = = =

C

{

, y- -

& b g c e

- - - -b- &

C= e w

e e< w w w <w< <w < e O-u IE 3 3 E I. 3 3 3 m. 3.

= = = <

4 v* EO WWWWW l's WW

! N EfB e i O-w-

a. r Ei
  • E ~W s- ==g 2

- - g -d ~

g = e E E W - -

1 1

g i

= = = = = =

4 b h h h k b i

~

3 d d d d 2 2 n E R~ E E mE . E. E E.

,. f . 6 . . .

g e * * *

  • a e

- E E E E* E E E E j .=

42a E O .

' .$s 5 g,

_ s E E se .s B u.

[ D-9

,4- -,-s- *> ,,-. Sa - <-u-*- 41 a 4 -- 'b

- ,T, -+44+1 a b J-. -

a 1

E- E

E- E E E - = E -E E E g = 5 d d = = ,

E E E E E E E E . : E - C  ;

= = = = =

Y k g Y kg g *g M g' g Em E ,EE -

EE- -5 s- E Epm I o Erm Ep m- E m I E* w -

'E m- s-E mW W} =S =0 a Wy 2

SW S "M SM mM 5 ' Eg Eg SW W 5-4 s -E -

-a -a sa - -

- - - -E =E s2 gm == mE E=

- . -- .. gm == m h $W- $m $5 d2 8

- - - E2- dmd- $W ~= -=

d-W dW- EW .. -

dW -

mU

- e" as He ES E2 -

-E wa M SI

~m EE

~.

8-

~n Ea

~ Er SE m. Eb~m h' 5 mm E Em am E- -m I *E o g- m e ~e s3. r m

e-i i

2 i W

, E I EE w-J i -

m E WW . . . 2 = 2 ~ ., n . - = = =

E 3

,i a, >

U - - = = = = = = = = - - - = = =

h h k $ k h k k k k k k k $ h h k r w w w w w w w w w w w w. w w w w w

8 2 2 E E E -

2 E E -E R E E E E e i O A

X

    • 2 m m m m m m m m an m m . m m um N b O C O O ~ "

O O O C C L O H* O O O E e E $ = 0 e d "

  • i

%- b b

b b

b Y

b W

b b

b b

b 6 6

$ Y Ek i

H. ~ = = = = = = = = = = . $. . = g .

, z- Iw =

=

n m

3

=

3 m

2 m

5 m

E m

E a

5 m

3 3

m 4

=

4

=

2

=

w w ", 20 f

g M M M M d M M d M M M M E M E E M

M W"

( M-i

  • W

, = = e -

! O .j 2 W b b b b b b b W 2 6 h.

i W L s a e e m a s t t u e n e t .d O

l 4 1

l w w a w w e e e a m e a e m v e

, w O

w O

w O

w O

w O

w O

w O

w w O

w O e ts w

O w

O w

O w

O w

N - - - - - - - O- - - - - - b.

c w z a x z z z a a -a a a a z -r. m

} w 2 d d d a d d d d d d d d d d d ,

J m

~ J J J J J J J' a J J J J J J J J t w . . . . . . . . . . . . . . . .

J 4 5 h. b h D h W 8 8 D D D h h b

' H- S t' "E & E E E "

E E e E t "E

  • t t

l 1

4 -

I h & N 5 5 5 5 U "

  • a 5 5 5 5

', @" .d 2.. .'#

2 .V 7 E .e, 2 2 ./ .'e 2 .M. P g

W Ik

= = i. i i

i i

i w $. $

i i

i i

i w

4 I

! D-10 t

vi --e.-w--o--%e-w 9 m-~v ~r-- -a-- -g- m 9 <-vN--qw1-rg , g. --

yem--wr>-i -- tu mmytea w- .+m-M1---'aewe.p.pgw.y-mew e y--p- - ~ wre- tw  %*- - - -mM- -e =<t-=

L l W W e s -

g -. ..

. m

- E

_ a e a = m_ m =

5 5 E z r es es es c5 == s

g. a.. =.. n.. a.. a.. a

..g a g a

.y aw .. e aeE ..g a.s

.g

- - em e e. en- ew-

- e. e n. e .E e_

y e,_ -

e= -

eg ..

m z we g- w- un w- w. w. g. w-ue _ g ~. u. == !

g us y a8 - ea = ,a s= ,a =e en we we _e

=g es =3

d =e == == =

gE w w r

a Er e k, - - e ma

= e s m_e s --e-- s s am 58 58 58 5mv 58 EWms i x e B- l 1- . u- -= = u- a= == R eE.

. 2 1 :S u wt g= g =-  : - d d

=t

=5agu- u 4W ag ag u- u- =" a- =. =- ml v.g u.g =g

_a e  :- :-

a w s .a er me 3. a. s= es am wE sm um 4 4-

., gg

- . . a = =a e a a w- ww ww =m w v,

we we w-s =~

w-

_E - - w - -- m - m mE wm =E wL m mn wr wM ==

w e. .e -T.

i

=

w 1 G 4 re c=

i 1

) -w 1

ED EEE E I E9 5 E = E _ ~ - . . .

t

= - - - - - _ _ _ _ _ _ _ _ _ _ _ _ _

t T t T t t t t .T t $ $ $  % 4 5 T T T j

- g wy w ewe w=

e - w ew I- y-e w

e y- y- v a a w w

=

w e

w e e u w e

i l -

<' v.

m:

3 M N E e p 5 2 W c p e j

S U g d c M M M M M M f~=: _ g g g g 3 g 5 $ b $ $ $ 5 5 5 5 5

) c 5 4 d M d d d d d W W W W W W W W W W W L 3"u iE gE -

WWW W WW E m

i I E i

i i

m I

m i

m i

m i

w* , 2C

  • 2 E E 2 E E r M N $ W W- W W W- W W W

4

. Nm

! U" M 5 M M M M

( W" = E E E = E

{ J. a A a a a

@I E E 5 5 5 I h' I I I E r- = - - - - -

- - - - - - g m 5 $ $ E E E E E E E I E E E m h h & h I

l 5 5 5 5 5 5.

5 E E E E

= = = = = E Nba

= = = w E

g g e u g

5 g 8 n y

g g y

=

e J J

  • J J J * * "

S M M F B 5 H 2 $ l

- w w

w. w w

w

. w W W W W W W W 'b W r i kt

~

. . . =. . . . d d d a a d a d a J 'd )

i B h b h h g h= g - . ..

E
= e = - u a a d e e
r. r a

a a

i

.I 4 N

! #T

=

IW W .3 W S W.v E m W 53 WI w w

b. g5 E J d a a a a a J $ h $ Y Y Y $ k $ $

-e e a e e e e e e e e e - e e = = = e = =

4 l D-11

J J J J J k i k k 4 4 4 k *

- - - ~ k 1 & r 4 J J E g 7 *

  • uh ~h ~h a w b 1 b o q

~_ -g n; n_ -w -w -w ,, , ,

C C 75 5m gus - 5 25 25 25 z CG A A A s=

-w -eE -E

= W"dh

-w as en -a eg -

m e- as d- a a a a a

a- a.. a d3 d" d" da d . d. d. d &B d d d d d d C d M - te te My-MJ MJ M My0- MW =0 =y Ed M

~

Mw MB

.5 a g 5 vg .

u a g u

g v . . . g u 5

C *'g 5~ 52 SW 52 S F, 5" C C - C- -.CE C4 5 ~W 5 ~W g Ww W W-p:

E W

.o

=5 W

. e. W W mm m$ t h, . W

. =5 as

.o m" r- m-

e. p m.

-5 5 e J

  • eE 8 --

5gCS g'e E-y I OS#& iCJ em er

=t  ? ". ? u.L o. u. ?n.o". L

.W LJ L

. U.

? L*

n. t 6 ~

.c._.". t ~. ?= ".

o o

= n

n. n w. W u. ='

"v wu ou . . = g u g 9 DO wo WE en

  • E m~ W meE.

WE

=~

WCmnWE W mm ei en WE mu W9er " "

E d d U ..8 U mm p, *3 An X *m~ S.

.m M

w O

CC

=

W N

$I ~ - . 2 O U S C . OM = - a a o

= w w w w w w w w w w w u w w w w w 8 E = E n 2 2 2 c 2 2 2 2 E 2 E 2 n

De G" C C C C C C C C C C C C C C C C C 3: S S S U M M d S M S d M M M M U M C 2 g a b b a_ i_ e_ cm e b_ m 2_ m m m b. t_ t_ 2_ 2_ _

C~ = -

5 Y. W.

W.

W W

W W

W W

W.

W ~

U W W

W W

W = . . k U

    • RW m a m m m a m m m . m m m m . .

O

=0 M M M M W W W W W W W W d m M M v.

Nm Q* 5 5 5 5 5 5 5 5 5 W" = = =  % = = = 3 E ,

J mg L  ?  ?  ? L L "

L L L

  • y.

3 5 9 9 M L9 = = ,

- = L k =

W W w m y u E 5 5 5 5

" 5 5 5 5 E E3 I3 E 5

E 5

5 S w 5 ui S

=

t

= a E un a o%n *4 c $ h.

m a u u t

m 5

E 5 5 E E 5 U E O U U 5 5 U 5 F E W W W W W

  • W W W W $ W G

w s

=

s

=

s

=

s r

s A w

e a

s w

W s sW EWs r w w =

x

=

s w

e w

e w

w e i e i i d i a d i a e' d a a i i

= -

r am -

. r - w w g - w w w w w - w

.: d W W y 5 y y y 9 y W V V ES E"- 5 E E E E E E E E E 5 $k E E E E 0-12

J- 4--- 4 -- s.%4_ A , - . u-, s a m .&.. _ m-- - - - _ - . .

f .'

W

.=a W

l . e k

=

j. C - r

- - - - - - - - 3 .g - ME 2 = 3 2 a  !

E E E E e e e g, y Q, g g g g Wg Eg W W W W g

- ., a = = = n x x = = <

v us e u u s x w Em: I e-e-

s= E: E= Es Esm en ee e=

s

=

s .g

= E g . g g rs, = ,

Cm C C 2 tt

  • it' M d W W d ~# g "# d d M ** # d ** nU

~

I ~E s: 5l .

5 "

. ". 5. [9= l l9 .= E@ =-lY EE

= E@ lk =l=@ f=f E$ -

= $h h,=. egE"Y e-

.1

=.

E5

~

I I .=h .=.,= m-Wk,3II IS NI ,

S' N I E IN #h

== sE w e aI Ym- u

= m-Wk me I

a-I a= me a= m= me s. = == m= ==

i E Er A

M r

- D M E W 5 M P e e b P N 4

i .

Y V t= &- V

=-

t 4 = = . .

t t t t t Y t t 4 + 4 w

2 w w w w w y w w w w w w w w

E E 2 2 2 2 = 2 2 m' 2 - 2 2 i

i J

m

! ye

! #8 5 C C C C C C C C C C C C 2 C C C C

s 2 c

M M M S e S e u u e y e e u e

! & e e e t:: 1 E E.,  ; E., w

- - - - m t:. m .

.e., .e., e,., . e

  1. " = W WWWW W t" W ta W W W W W W W W e
u. Im E

. .E - - - I I I I E. E E

I I

. .I E

. .E .I .I v, O= M d 3 W W W W W W IM kN IX lN W M kg M j .*

N en O e.,

  • U e k
s;x a g a m a a si =ya  % =

seea

= xx x r s x & I s s I a s

, 5 aa a s a a r a a r a s a >

i 5 5 5 5 5 5 5 5 5 5 5 5 5

- ,E,E , . . .

5 5

5 5 5 5

s

=

5

=

5 5

h J J J J J J d d a J J J J J J J J J G i iiii ii i i i i i i i i i i i i

, i y5 -

.e l- e k

==

k kN e =

k e

k e hke k

e k

=

k

=

k

=

k e

k

=

D-13

-.;....*e 4 .y@._s ..Am a. - -a. a e4- e, m si m . 1_.4 -eL -

k I

i U # 5 # 5 3 I

- 3 3 a e a a a a j

I- u = e se =e =e s. e e EU E3 E e e 1

.E x .= = .x .m WW a W e. E e. W e.'W e. -. -- -.

!E. *eE gE gE

.sg gE eg gE eg g5_ 58 $3y y

$=

j TE WE WE *I EE ME sg s

- s r e w e r e Er *s I, En Ee Em "

  • j r -5 us g ss M g a zu e. er es i

FE *s de ds *s ds ds e m 5= f m5 w5 Es= mm-s Eai Es ds *s *s *s i

hh

= an I k a= a= a

k h k kh an a= an a- a- a- a- a- a- an a= ==

I L

w r

4 4

I

= f M

el = ,, . = = = E - - - a e n n n

)

J

} Y $ Y $ 5 $ $ T T T T T $ Y Y Y w w w w w w y- w w w y- w

=

w

=

g w

=

E = = = = = = E = = = -

-l 1 -

! Y: . . = . . . . . . _

e . . . . .

I' E h h $ h h b h U h M h h h h h h e = e n m e s_

c- =

e e

e e

m e

e

=

e m

e E

e e E

e e

E e

e e*

e e

e o= ~ * * * * *

  • 3 2 1 2

! U > gg 3 E E  !  ! E g a g g g ,a g a

g a

g a

3

- * == m m a a a a m a a m a a m

d-o- - - - - . . . m 1

3 ei=

-= e i

y ev

, =

=

3 n

i 1

o 5 -

5 o

5 E

5 o

5- m o

U a

o U

s -

o s

i, .E g @

. g g g g E

_ _ _ _ _E h_ _ _ m U.

I i "

1 E E E E "

E E E E 0" E E E E 5

w w $ w *
  • w m @ w e 5 i E E w s s a E E u E e E w E j w w w w w w w w w w w r w w w w y a a a i a d a d d d a J J i a e 4;

3 v 2 E E E E E F t E E R E E R E E 1

b -

b W w i

l d= s 5 I e

E

=

I

=

E

=

W n

=

W W

e E

W e

I e

w N

=

N

=

I= I

=

I

=

I

=

E

=

, D-14 i

l 4

i

~. . . - - . . ~ . . ,.-.-,.,,c, . , , , . , ~ , _ - .

n... . . . - - . - , - , . - - , . , - , - , , , - _ , . . . . . , , ,__,-- -- - - -

w a .

i a w e m - - - g w . 3 a a a y y g g e e es _

r= _e-em

  • w= v- a. E. w

.E .- x w se sa g g E a

a sg sg a , . . . . . . . . . . .

e g w w w w w w w w w w w s eE er e -- -- --

v. m.

E- . _ _g g Eg - -g- -g- _ g_ g g y s: E , su s ~

m: w W

E E E E e E E E E E E g

_m we- wA s- n z

= = c s = s l

-e ER Eg

==

=

x=

) E H E Es wE m, Es e e e s w s Ewm .E w w w w = w w w w w a e e e e e e e e e ee.

I l

,! IE

~ _ _ ~

i i m *l e

n a

s n

9 a a s 9 9 - ~ - . _

i@&

= = m t

i i

4 4 4 4 _

  • _ 4 4 4 4 4_

gew y g w= =w ew5e w*e ?w= 4w w=4_

V 4_ 4 4 4 4 g

e = =w y-w w g w w g g w

- = e - = - - -

l 1 m a m j 84 I m g m m g m g m m 3* O O O u u O u O b i C*

yw & W W y & W t #

1 5-c=

, b b b w b b b E l

om 3 w w w w w w w w w g g g g

u-S I E 3 3 . 5. .3 I

g g g g y* at . m E E . e m- W w

=

2C W W WWWW W W W 5 5 E E

'l N-QW w w w

i 3-

k.  : -

l l -

5 4

5 4

5 4

5 4

r: = ,8 w

5 u = = = c = cg c - _ - _

) d 4 4 4 g g g g y y g 1

w E ~

r E. 3 E 2 *

  • O ~

i - - _E m m . - . . m = . =

1 E M* E E EW E E E E N N E E E E EG E U '- * = w bk w a w wkL w EI i

3 w w w w w w Q E E j _

w w . w r w w m-w w y

g e

a e a e a a

e a

e a e a nmn ,n a g %, yWg g w 1 E E E E E E E E f E E E E 5 E a a E & M j

)

}

[N >

~

,i 5.E, g $  %

e e

$_ y_ - }.  % $

m W..U 2 .

y .5 .E. E.

. .E E.

. E. E.

. E.

. . e . t. a a a a a m.m a D-15 d

1 i

4

? P. ,a. 7 '

! 2nuti TABLE D2. (continued)

.au VEST natal .au j

Ct0tTRf EceTACT FAC!!!TT TRANSIEtf CODE TEST EIPERINENT EIPE41 MENT

]

TYPE NUMOEt if9E MSCRIPTION ,

! i SPA:n 08. JCSE MCARLOS TEAC-PFI ' T3 K MTEMINEB ,

SFA14 H. JCSE MCARLOS itAC-Pfl TO M MTEMINED

)

SFAIN H. JOSE MCARLOS TRAC 4FI TO M CETERPfuEl SFA!4 tt. JOSE MCARLOS TRAC-Pfl TO M ETEM!aEB  !

SPAIE DR. JOSE MCA8tDS TRK-PFI TO M MTEHINED SFAll M. JOSE MCARLOS TRAC-PFI TO M MTEMINED

08. OD98J0pt SAeDEnveG Ps4 pater PLMT TRAC 4FI 6 RID DISCONNECT MO TROINE l Sur.DE4 RUni' ACE TO MUSE LOAFS t*!TED Elu6t"n M. J. FELL LOFT INTEMAL TRAC-PFI SB-l SMAR BREAK LOCA (31 81TM FCPS l

1 HITE3 KI%630M CR. J. FER LDFT INTEGRAL TRAC-PFI SS-2 SMALL BKAt LEA 13*l  ;

WITH PCPS  !

' a j s

l M. J. FEu IuTEGRA T#AC-PFI $9-3 SMALL BREAE tMA WITp0UT WPI; 5 IMITES r!NEDON LOFT '

RECOVfRY U51#S SG FEED M3 KEED i

L11TE3 t!!: set M. J. FELL LFT INTE6AAL TRAC 4F1 L2-3 10 M CETEMINEO

{. '

+ '

HITES tIEli!!I IR. J. FELL LOFT INTESSAL TRAC-PFI L2-5 TO BE DETERMIND 4

58. J. FELL l#TE6EAL TkAC-PFI t t-1 LAPSE BEEAg EDCA AT 504u etT4

. MITER IIMOCM LDFT

! RCPS 81ECouMCTED FRIPI FLYMEELS i i

i i t41YE3 kluGCM D4. J. FELL LFT INTEGRAL TRAC 4FI -L2-6 LARM 9 PEAK LOCA AT 4MW; ,

En TRMSIENT l

I talTE3 t!a530M CR. J. FELL LDFT  !#TEMAL ,

TRAC 4FI FP-1 .TO DE ETEMINES

]

UNITER EIGIM 04. J. FELL LCFT ItTEMAL TPAC4F8 FP-2 TO PE MTEr?'lutt i

tM:TE3 (145P0't ER. J. FELL L0t!-9002 INTE6RR T8AC4F1 A2-81 5%t H' A' LOCA j

l LMITEI stuGUM La. J. FER TRAC 811 TI4 C27tS FRC9 TPAC-PF2 ASSES"ERT TO M MTEPa!NED ,

LTITED #1N500R M. J. FELL TPANFI TO M CETERMTED HITEI tlaSt[M LR. J. FEtt TRAC 4FI TO M CETIRMIWS l L% ITER titGrp te. J. FELL YPAC-PFI T3 EE BETERP!aED ,

_ _ _ _ _ _ __ .-. _- 1_. - - . . , _ . . . _ _ _ , . _

I l

t w E E g

E E

=

. =.

- I I E S -%

E

\

E

=  !

E E

" W-v w

d w w w w w w a x s E E- E en EE - . 8 m"

~

g 5

=E ~_!g ~ E_E= _= .- E E "

m y E E E E E

_E

=-

Eh E

E E E E E E d

a = - -

w w .E w =. w=ua -

E W

'. E" .

~

E -@ E E B $ SM5E M5S E I. . E E w3 .

, "5 E- W E E E W W =g W W W 5 5 5 . w y 1

, wm .

- _" . w W

. .= - E. . e. . W.. ..= - =- =W ee = = s. s-n E

w E

{ EE 4 20 1

1 I "

? ? ? ? ? ?  ? ~

i =g = W  ; W W = m e e {" g . 0 $~ U

= - 2-4

W e e - e a e a e e - n = n n R 5 e

.t 2

  • 4 m . w m m j $ $ t t t t Y t  % .

i y $ t t E .

@w tw Tw tw t w

t T t tu -

a E

y w e =w =w =w =w =w w w

= = = w w y w w w y we aw c,

1 - - = - e = = a a = = _

}

n t

f D

i i

m:

c*

1 'a" . -

h, C" 5 d d d d d d d d d g g g d d d d o" ~

5 5 8 5 3 5 3 S 5 E S E 5 E E 5 i

U- Et W WWWWWWWW W .W W WWW W i =

20 E E E E E E E E E E E E E E E E  !

i 4 N" 1'

a.w

  • a w-

%. w w w w w w w w ww w w w w w w i,

e *q w H- = . M M M M M M M M

'i y E E E E E E E E E 3 3 M M 3 "

i W W W W W W W W W E E E E -E E E i

i i

i

)

W

= = W =W =W=W=W=W=W=W= W= W W

=

T W W W

= =

!' d d W Wd W d =. 2 g dw &

h h h h t t k & & & & h h h h h 1 -

2 d d d Ed d d d d 2 6 ~6 6 d d d d

ya ,. ,. ,. , ,. ,. ,. .E .E m .E

~ ~

E E E E 6E m .E &

~

2

~ ~

& E Em .E >

. . . . . . . . . . . . =. . . . . . . . .

i 8 E E M. E E & E E E E E E E E E F E E E E E- E. E l i.

1 E

g E 5 Eg E g 5 o

~

.n E ; E G = E 2" g W g W W

< a . . . .

1  : _ 9 W W 3

15 E E E E E E E E E E E E E E E E E 1

! D-17 4

E E E a d d u 5 .

a v v a= P4 W w has nas EL O Ge CE is G: Lad en g ng & Er >-

    • O -

b ik Eu= g ga[5 wb su Nw 6na B w E,W WW w w $

a M2

- - 'E' W, gg E

= n E 2.w SEu o a, d 3 WE M a= 4

>=

o s

Q LS .a 2

A d- EER A -

u f9 b w O e W 50 3R 5 *

= md "

6 b " 9. -W w a

  • p =

E "

W 25 a " w h

w m

w w-E E"= WO h5 m- - ~

CS w s do ou g 5E 9; CW CG Et- -

d "d a

m W 5 :t am - CW- -

e a

1 E "E E

  • E E W2 =~ 5 Fs *5 55 QS. W E a. 55 d .M, E_.mS E_. m*E_..5 E,

.S ,,, -. E.

lt

~ 3_, .w m

>=

SOS & Y w-h b _.

P9 em M N 4 W e9 8

en O 4 er ==

h* ut O O E'" Se ** g% =S er e a en O s --* s e e a e me == ** O Lea 4 4 ** m ee 3 P s2 O O an. dp == m'a.

>= em se 4 N.J Na s se == *e me N PS at tr)

N 6 h e me em se se em em m. we e=

en e.e os me me em se aus se e.e e.e s.e w w w w w w w w w w w w w w w w w w w w a w S 2 E 2 2 m = n E 2 2 2 2 2 5 2 2 5  : 2 m

D en w sn w- w G 2. ve nn

- s- .- -

en ne a @

S w

8 W

  • ad LJ na' bp w

bd w

b#

6*

LJ 4.J

  • a' b.4 C
  • 2 2 2

.e &

w k

w

& 2 w w w &.,

. w .&, w y

C **

O d d d d M d d d d d d W W W

=

W= W g

W*

2**

a t**

g g t*'

o= -

g g = 3 g, a g 5 3 3 = ==

3 = g g g u- gg =. =. W. e. e. =. .g. w w. . g =. g 3, 2 g 3 e,

u a m a w a

w u, a a w 4

N "

Cb w

W "

e_3

  • CQ ct 2*

& ~~

e w w w w w m w m

e me EW o

m -

$4 S S S

- - gs g g s g s dB3 04 ee g E c = B d >= -

B M

B B e-- b-b -B b-B h L

>= >.

w w .. e e = w a e a w e e- w e, e E

w W g

=E E E

.dg at

.a JE m.

.E8 e W

m...

g

.M g

9 9 3 m." .E. .w

- == S'* $" 9.=8 s

5.* D'8== $-P m.e O" A8 @" 9e 98 - >* D.h $'S hA S

g h db 6 A b 6 A S & A & & E & e.e

. . . .* . d.im.e. == ee (b.-e >= . dL.

m &.-

. == . . .

.s J

-I - -e.4 - - .. - .. _. - - ee _. - - - we r -

Z 7 K 3 X K K K 2 X T u

et 66 Ga.

X Es w E E sm w A L ab A

nam

[

w K.

(n b b ab & (m 6m an. Eb & io w e e u o e e o e . m , e e o e e .

e e e o e ><

W Em e, $ &

se ** es er

    • *T et ese 2W 6 g de

. . WE et s _.

er *T

$ E $ E C E 2 et *T 4 et at et C

et W W sj se{s 9 0-18

l l

l l-

! =5 r=

==

wu

==

J E

=

rr

=- .

i j E5 = = ==e aaaae=neeneea5EEEE5 i

i - - - - - - - - - - - - - - - - - - - - - - - -

4 w 4 w w 4 w 4 w 4 y4 w4 w5 y5 w5 w4 w4 4 4 44 w w 5 w4 w4 w4 w4 w.* w 4- 4 g ue-4 m

e e e e e z - e e - e e e Ng e = e e e e e e -

3i e e e e e e e e e = e e e e e e = 'e e = e e e e

== e e e e e

" e e" e e y e ew wm b b e " "e e e e e e e *e e e e = e e 1 ,c ,

b w w t tw b &

w w b w w t":i & ew b bw e we w "b e e wb - w

< v- e WWWWWWWWWWWWWWWWW W~ W' W *

  • W W m a a a a a a a m a m m E a a a i.m E.a i.mmi.

8"~

y . E. i. i. i. . i - i i. i. i. .i i. i. i i. E. i.

-; e= a a a

~-

. ow -

w

a

. <=

H "

1 h

2 2 2 b b W W W m a e s a e a u 5 eW e b 4 h b b b b b b e e e e e e e m Ebmm baW mn i

w w a v v v w w w w a w w w w w w wa w w w.w w

= = = = = = = = = = = = = = = = = = = = = =

i k k h h h h h h h h h h h h h h h h h h h h L 1

2 n dv d d 2 d a3 w 2 2 6 d d n y n d d 3 $ 2 2 2 d g

w

. s . . z

s. .

. .a .r g g g r. m .- e g g a.

. g g w g a i

5 m e e a e a a a a n-a e a a a a a a a a s s e e e.

J i .-

Ee h y= y

.5 6 E EEEEE E E E E E E E E E E E EE E E E E E l D-19

i k

I l

'l a z E = =

=

d E E Eg

= = z - - e Eh =~

a a- -W -W m EEv

= s

e e=

-e

. . . .- .- =

s a w w eg ug g -

_= 1 1 = -

(-

g a g. g. g-a- w- n n- ,s

_s - - a e = m g E e.E e = w- -

- r r y . . = , q -

  • g gs E=h g g a m .a g. =E w w E w -r n *=
  • 55 -

E y d "g E h-x w w "t- - - E = k- -- E- E I' Fu-B = m .f" .a sm W g"5 s m

E EE

-- w h- a w w

. v t .?

t t v y d d d y w

-5 E E B B 3 =E= . m n E E E E 5 = sss = d 4 4 4 4 4 4 4 4 4 w w w w w w w w w w 4- 4 4 4 4 4 w w w w w 4 4 4w 4w 4w- 4 w

4 w . w S E = 2 2 2 2 2 = = 2 E E = 2 2 c - E 2 e E 2 E:

  • c n n = = = c c = = c n g n -e e

c O

b b 2 g - O bE bOb Eb 2c b0 b0 E b 2 b b E g-

. E b

e- - a a a a

e -

c

o.  : u a a  : u e. m e, -

.a ,a a

w 3a smWam E u- gg 3 a

s. . s. a. a a a. s- - . .

H

,= s - .a e =

-- =

-- m a a a a a a a a a a a a a e a a N .

o.

=z WW m m a W E

a

E E E E gga

<=

" = y e e 4 4  : E e" -

y y ggg ~

e a e e g

e 4

a u E w y@ . - . .

@ ; I we 5- - -

5 5 5 5 5 m .

f', .6 E' b.

g w w w w w w w w w w

= = = = = = = =

w a w w w w *: w w w

====

w w

= = = =

h h h h h h h h h h h t h h h L h h h L 2 h d d d y

- x. -e z-d .d d y 2 6 d d d d d d d d d d d -

w a v v. .s v. .v

.v 2

e -e .

r. -

z r u .= m r r.

m a e a e . . .

5 e a m a w w a m .m x. / m. r a y. e r

=

E' =

O c hh b $$ $ $0 E O E 5 S 5 0 $ $ $ $$ 6 N E $ 0 0-20

a in .e

$ 0 0 r m -

w W = =

kt g b d EE , = r eM d 6 5

W E G G 0

W E

It' E

==

0 3 9 y

EI N N N w

S w w w h a =

m D

0*

3 =

C M" g C~ 0 d d -d o= -

g g O"* h h'C t"

5

$ t" E

en e

N*

Om au d em J

W

  • _ 5 9 5 p3 p%

w = =

0 m C L h h r d d

- - -d G . .

O E E. E C

5

>~ v < a 22 D. E E D-21

E

m.

. c. .S.-

W 5 E

= m" -

5 E !f 5

5 h 5 C r

W E!"* E =WW E" 8 k = a . . . . . . . . . .

WW W -W &

n W =

=

O .

W E'

_.W B F 5

S E  %

5 W W W WW - W..E -M K

5 5 5 5 5 5 *-

5-E F E W.

.J

.m

.s.

- - 5

.5 C 4 2" 2 5" of .

- w C.

. o - B s- 5.- .- 5.- - - - - - - -

w w . w

'*

  • w"" " " '" " "

- ., w ~ w g g g w - w - w w - .

- h W d. B N Ww *d " " d "

ri.

"5 Eww . E u e.

v .a W S.

s E, W'

w

'4 E.

o W

o Eo E

  • a
  • o* W W Wes E a - o e.

ui

=

t

.eA em a.4 es o.J w

w at.

J o n.

A b .4 p- p. >- p. .n= >= >= >- o. o.

p 6- > > E d K .n

>=

W E k'

.= **

_y O

n m e e 4

< ~

g

.J g gW _ *

. . B. "J m

a u , ,, . _ , . . . ,, ,, ,,

a a a.

4 .a. a.

. a.

. e.

. .n.

o e

. .a. n

a

. .e. a. .a. .a.. . .a.. . e.

. .n. a.

x .. .. M. . .

WW W W W W W WcW WWWWWW WWWWW

& ti W WW g S 2 3 5 = 5 E 2 = = = E E E 2 e 2 2 5 2 2 2 2 O

6 a:

.- 5 5 K G 5 5 .i*

9 "* 5 t t t t t t C 4 d d d 7 C; . . . . . . 3 3 3 m td ** 5 --

1.:' WW W Wo W o p

W '

W a

W a

P' v

P' g .a, E L, . . m E .

M w

M we W w 4

O.

O d - - - -

W W  %  %  %  % WW w w  %.

. .E .C .

x (D

, , = . .

. w m w w -

m X. a c d. .M- a. 6 . . .,

W W 9 %

O .a . . 3 x .3

= w - w

.., m m, -*

i. - - .- ..
w o o u . .- .-

w w w

o. u., . . W L M .

't t t t t . -.. - n -+

,, s w

w w

w w

w w

. W t t t t ~1 m .-

.f

- . o n

o ed u o

o o

a ss u 1

.s 4 *'

b h h h h h h.

p . . . . . . . . . . . .

b c 2 2 2 E 5 K K h a n n n 3 5 3 & 2. E. t 'e A. P 5 > >= b  %

r'a

  • 9 a a-5 5 I 9 ?

.n . a s .i a

f

  • 1 3,,

a a

. :n a

I 5 g g 9 v..

a s.'.

5 S 9a y

~. OW W ".C S" O W W W 3 2 3 4 , . i. P 5. C W w

~

L*.

{-0

" P .

d* 4 4 '* .J

  • a C 5 t ~.:
5 .O* O '-
t. ., ,, ,,

6 b w w w* d u w' *

  • t 3 S f. F. S O L'. 8. f. 8. b h, WU 0-22

g .

u e =

.= ,

n

- . e , a -

a =. = -

W g s s a . n -

. . = s a = ee = a .

WW

uQ Eg 3

E 5 h 8 -

au n" - e e M a 5 5 #, SE = M S*O

==

en w

a d-

, w" s" =

W- r - -e e r

= ne e = g - -

s s 1 55 "T a W Eg my k 5 EM w 5 d . 8O 2y Ew E a M ..E m8 -

u

_ O. . - .-

O. -

.P . . .

w m.

kt 8

E

=

~.5 M "v O. E

e. s s = = = =

=. -=.- =. -=.-

=.

=.

w w w

=. =. =. =.

w w w

=. =.

w w w w w w w w e = = e e = = a = = = = = =

n u.

e: e e

=:

c e e

-- e =

l" o"

3

=

d d d a a =

d a

d d d a a a W d d W d g g = a g= . .

WWW WWW u =

3

=E- r = n W

a z s = a W a W= w EWe m.

Ow w-

a. -

x R

- - . g =.

r :: =

a v

- - - - - . m - =4 ==W s

= m. e = e

. . - e. . m = = .= E 3

. a. a WWW

=

W

=

WWW

=

W WWWWW

= = = = = = )

% t L L L h h k h t b

-d -d d

. .d n =. d

- . - d n d- - .d. d .d 6 L 6 = r n. = r a e r r a e v a.

i .- ,

2 ,2 y E= . . . = g a d 'J d" E E E E E E L E E 3 $ E. $

0-23

APPENDIX E l

ASSESSMENT

SUMMARY

METHODOLOGY The International Code Assessment and Applications Program (ICAP) members assess the capabilities of the USNRC supplied thermal-hydraulic

. codes (TRAC-PWR, RELAP5, and TRAC-BWR) and submit documentation of the

assessment studies to the USNRC, All submitted assessment reports undergo j a USNRC sponsored review process at the Idaho National Engineering l

Laboratory (INEL) or Los Alamos National Laboratory (LANL), Contained in this appendix is a description of the INEL's review process applied to each RELAP5 and TRAC-BWR assessment report. A similar review process is used at LANL for TRAC-PWR assessment reports.

-l i

l j

I -

4 I

i 2

i E-1

CODE ASSESSMENT REPORT

SUMMARY

EVALUATIONS FOR THE TRAC-BWR AND RELAPS CODES--

METHOD, EXAMPLE, AND STATUS

  • Glenn Case and Gary Wilson Idaho National Engineering Laboratory EG&G Idaho, Inc.

Idaho Falls, ID 83415

1. INTRODUCTION The International Code Assessment and Applications Program (ICAP),-

sponsored by the United States Nuclear Regulatory Commission (USNRC),

provides for the exchange of information relevant to the operation and improvement of nuclear reactor systems codes. Under ICAP agreements, the USNRC supplies ICAP members with the frozen version of the RELAPS, TRAC-BWR, TRAC-PWR, and/or COBRA-TF codes. ICAP members using the codes assess the capabilities of the codes and submit documentation of the assessment studies to the USNRC.

This paper describes the USNRC sponsored review process ap9 lied to each RELAP5 and TRAC-BWR assessment report. The review of these reports is performed at the Idaho National Engineering Laboratory (INEL). This paper first relates the objectives of the assessment review. Then, the steps of the review process are presented. An example of the review process as applied to a specific report is presented. Finally, a time table for the review of submitted assessments is discussed.

  • Work supported by the U. S. Nuclear Regulatory Commission, Office of Nuclear Regulatory Research under DOE Contract No. DE-AC07-761001570.

E-2 Y

2. OBJECTIVES OF REVIEW

,The review of assessment-studies submitted by ICAP members is intended to support the use and development of the codes and to assist the USNRC's

' ongoing code uncertainty evaluation. The review process will identify each assessment study's contributions to USNRC code related efforts. During the.

review process, information pertinent to the code development effort and information pertaining to improved user guidelines will be extracted from i

each assessment. The utility of each assessment to the USNRC's code

- uncertainty effort will be considered. Finally, the review process will verify that each assessment contains information in concordance with ICAP i agreements.

3. REVIEW PROCEDURE The review procedure involves a cursory" review followed by a detailed review of the assessment study.

b The cursory review verifies that the assessment study uses a frozen version of the code and conforms to the approved ICAP format. Information n.aded for maintaining a catalog of the assessment reports is provided by the cursory review. The catalog provides monitoring of the types of 5 transients that have been assessed with each code.

The detailed review supports USNRC and ICAP objectives in several I

ways. Qualitative insights into code capability provided by the report are i noted. Such insights include judgements as to what phenomena the code can and cannot calculate. Information useful to the code development effort is extracted from the assessment. Code deficiencies, model shortcomings, and computational efficiency are among the items of interest to code development personnel. Information of general interest to all code users is also extracted from the assessment during the detailed review. This type of i information is. incorporated into user guidelines and distributed for future  !

code use. Another use of the detailed review is to support code uncertainty i.

[.

evaluation. The potential contributions of an assessment to the code i

[ uncertainty effort are identified during the detailed review.

b i

E-3 i

i I

. - .- . ... . ,_ , - .,_.-___,,__.w,-__,-...-.-_ _ . - , - _ _ , ,

r I

The detailed review is performed by knowledgable analysts representing the code assessment, code applications, and code development arenas. In the first stage of the detailed review process, each analyst individually reviews the assessment. .At the end of the review period, the analysts form a task team and meet to discuss and compile their evaluations of the assessment study. The task team identifies the assessment's contributions j to USNRC code related programs. For each assessment, the task tecm produces

! a summary evaluation that documents the findings of the detailed review.

This documentation is then forwarded to the USNRC.

]

4 4

4. EXAMPLE OF AN ASSESSMENT REVIEW I -

To better illustrate the review process, an example of how the review process is applied to a specific assessment report is presented in this section. " Assessment of RELAP5/M002 Against Critical Flow Data From Marviken Tests JIT 11 and CFT 21," submitted to the USNRC by an ICAP member 4

was selected for this example.

The cursory review yielded some simple information. The study was 4

performed using cycle 36.02 of RELAPS/M002 (a version of the frozen' code).

i The code calculatior: were compared with break flow data from a large scale ,

facility (the Marviken facility). The results from a base case and several-sersitivity calculations were included 'in the report. .A description of the experimental facility which supplied the data for the assessment, and a complete description of the RELAP5/ MOD 2 models used for the calculations l were included. The facility and model descriptions are needed for a meaningful evaluation of the assessment to be performed in the detailed review, i

The detailed review has not yet been performed, but a preliminary-list of items from the assessment that the detailed review will address is presented here. For~ saturated steam break flow, the calculated discharge t

flow exceeded the measured discharge rate by 20 to 25 percent. Detailed nodalization of the nozzle region did not significantly improve the-calculation of the break flow. Therefore, the use of~a discharge coefficient of 0.82 was suggested for modeling steam break flow.

E-4

Discontinuities in mass flow rate were observed for some of the steam break flow calculations. An approximation made in the calculation of internal energy in subroutine JCH0KE was identified as the source of the discontinuities. For.subcooled liquid and low quality two-phase break flow, the code was observed to overpredict the measured discharge flow. A discharge coefficient of 0.85 was recommended for modeling subcooled critical flows. Logic in the JCH0KE subroutine was found to preclude a linear relationship between_the specified discharge coefficient and the calculated discharge rate for two-phase flow.

The above findings will be of use to code development personnel and code users. The code development effort was aided by identification of the observed shortcomings in the JCH0KE subroutine. Future code development can be directed to eliminate these shortcomings. The apparent insensitivity to nodalization of the nozzle region is of interest to the code development effort as well as code users attempting to model similar phenomena. The identification of the expected range of discharge coefficients needed to properly model saturated steam, subcooled, and two-phase break flows is of benefit to code users.

The detailed review of this assessment's results and recommendations will result in code improvements and/or enhanced user guidelines. The improvements and guidelines will provide for more accurate and efficient analysis of similar phenomena in the future.

The detailed review will include consideration of how the assessment may be used to support the code uncertainty determination effort. The details of code uncertainty determination are currently being developed.

Thus, it is not possible at this time to indicate the utility of this assessment to the code uncertainty effort.

5. CURRENT STATUS OF THE ASSESSMENT REVIEW PROCESS This section presents a time table for the review of assessment studies at the INEL. The discussion applies to assessments of the RELAPS/ MOD 2 and TRAC-BD1/ MOD 1 codes.

s E-5

To date, seven RELAPS/ MOD 2 studies and one TRAC-BD1/ MODI study have been submitted to the USNRC. Review of the studies will begin February 1, 1987. The detailed review will provide about two weeks for the individual analysts to review the assessment. The task team meeting resulting in the final evaluation will occur after each analyst has reviewed the report. The total time for the review of each report, including the cursory review, detailed review, and preparation of the summary evaluation, will be about four weeks. It is expected that the review of the eight assessment studies currently on file at the INEL will be completed by September 30, 1987.

NOTICE This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, or any of their employees, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for any third .

party's use, or the results of such use, of any information, apparatus, product or process disclosed in this report, or represents that its use by such third party would not infringe privately owned rights. The views expressed in this report arenot necessarily those of the U.S. Nuclear Regulatory Commission.

E-6

Diagram Of Review Process

" Note Report Cursory Review Does the Study

  • Does not satisfy ICAP Satisfy ICAP Agreement Agreement YES 1r Detail Review y

Retain as Code Development Resource V

Code Development input i

V User Guideline input i

1r insight for Code Uncertainty Analysis Additional YES Formal Key Parameter > Request for Data Required Required Data ir Task Team Review

=

1r 5 Final evaluation k transmitted g to NRC E-7

APPENDIX F ICAP REPORT SUMMARIES i

As provided by the bilateral agreements of the International Coda Assessment and Applications Program (ICAP), the USRNC receives code assessment reports and other documents relevant to thermal-hydraulic code development from ICAP member countries. This appendix contains a brief summary of each report the USNRC has received as of January 1987. In all, eighteen summaries are included. The first eleven are summaries of code assessment studies. The final seven are summaries of facility description, thermal-hydraulic phenomena, and code documentation documents.

t 1

i j

l F-1

Piga N2 1

-03/11/87-REPORT

SUMMARY

REFERENCE 2 -l AUTHOR: U.' NEUMANN COUNTRY: GERMANY RESTRICTIONS: NONE.

DATE: ' AUGUST 1986

! PUBLISHER'S IDENTIFIER: NUREG/IA-0001

)'

TITLE: ASSESSMENT OF TRAC-PD2 USING SUPER CANNON AND HDR EXPERIMENTAL j DATA This report assesses the predictive capabilities of the Transient Reactor Analysis Code (TRAC-PD2) using data from the SUPER CANNON and HEISS DAMPF REACTOR (HDR) experimental facilities. The report is divided into three parts. Part I is the TRAC-PD2. assessment using SUPER CANNON data. Part II is the TRAC-PD2 assessment using HDR data.

Part III provides recommendations for the user using the combined assessment results. In general, it is shown that the TRAC-PD2 -

predictions were in good agreement with the actual test pressures and' l- mass flow rates for both these tests. TRAC-PD2 provided considerably l better results than TRAC-P1A. This was particularly true with regard to sound velocity predictions which play a significant role whenever the speed of pressure relief waves must be determined.

e

?

1 I

I F-2

, ,- . - . . . = _ . - . . - _ . - - . -

.. ~- . . _- . .. - .. -

Paga No. 1-03/12/87 REPORT

SUMMARY

REFERENCE 3 AUTHOR: P. CODDINGTON COUNTRY: UNITED KINGDOM. -RESTRICTIONS: DECD-LOFT ONLY.

DATE: JANUARY 19861 PUBLISHER'S IDENTIFIER: AEEW -- R 2039 TITLE: ANALYSIS OF LOFT EXPERIMENT LP-LB-1 USING THE TRAC-PF1 CODE-

This_ DRAFT' report documents the post-test calculation of LOFT Experiment LP-LB-1 using the TRAC-PF1/ MODI computer code. Experiment LP-LB-1 was the second Loss-of-Fluid Test (LOFT) facility large break (200% double-ended cold leg) Loss-of-Coolant Accident (LOCA) experiment to be performed under the' auspices of the Organization of
Economic Cooperation and Development (OECD). The TRAC calculation was performed using version 11.0, and hence does not provide assessment-for the " frozen" 12.1 version. However,_the author has indicated that 4 a similar report format will be-followed after a more recent version of the code is used, and many of the' conclusions of this study may be relevant to proposed code-development activities. The following major findings were reported

4 During the major part of the blowdown period of the'

LP-LB-1 transient (i.e. approximately 0 to 20 seconds) the transient calculated using the code is in very good agreement with the experiment.

- The agreement between the calculation and experiment in the way the vessel refills (i.e. no direct ECCS bypass) is excellent in spite of an apparent different behavior

in the intact loop cold leg.

i - After 40 seconds, the calculated and experimental fuel

] clad temperatures differ significantly (generally related I

to timing not predicted accurately and some difficulties

. with quenching).

J 1

4 h

F-3

. Pag 2 N3. 1

-03/11/87 REPORT

SUMMARY

REFERENCE 4 AUTHOR: S. GUNTAY COUNTRY: SWITZERLAND RESTRICTIONS: OECD-LOFT ONLY DATE: DECEMBER 1985 PUBLISHER'S IDENTIFIER: NONE TITLE: RELAP5/ MOD 2 ASSESSMENT: OECD-LOFT SMALL BREAK EXPERIMENT LP-SB-3.

This report documents the post-test calculation of OECD-LOFT Experiment.LP-SB-3 using the RELAPS/ MOD 2 computer code. Experiment ,

LP-SB-3 simulated a hypothetical loss-of-coolant accident (LOCA) which I resulted from a 4.67 cm diameter single-ended break in the cold leg of a large commercial pressurized water reactor (PWR) with a concurrent 7

loss of high pressure emergency core coolant injection capability.

The RELAP5/M002 computer code was shown to be valuable in understanding the physical phenomena in the experiment. The code generally performed well, predicting all the key events in the correct sequence and with reasonable timing. A discussion of some code shortcomings in predicting core thermal behavior is included.

W F-4

Pags N). 1 03/11/87 REPORT

SUMMARY

REFERtNCE 5 AUTHOR: J. ERIKSSON COUNTRY: SWEDEN RESTRICTIONS: NONE DATE: SEPTEMBER 1986 PUBLISHER'S IDENTIFIER: STUDSVIK/NR-85/99 NUREG/IA-0005 TITLE: ASSESSMENi 0F RELAPS/ MOD 2, CYCLE 36, AGAINST FIX-II SPLIT BREAK EXPERIMENT NO. 3027 This study assesses the frozen version of RELAP5/ MOD 2 using LOCA experiment No. 3027, carried out at the FIX-II test facility at Studsvik. FIX-II is a 1:777 volume scaled nuclear reactor simulator.

Experiment 3027 simulates a 31% area break in the recirculation line between one of two recirculation pumps and the lower plenum. One base case calculation and three sensitivity runs were performed and compared to the data.

RELAPS/ MOD 2 calculations were judged to adequately predict most of the measured data. Some important differences between the calculations

and the data were noted and discussed. The differences were: 1) an i underprediction of core pressure drop, 2) a late prediction of rod dryout, and 3) an underprediction of condenser water content.

I i

F-5

Page No. 1 03/11/87 REPORT

SUMMARY

REFERENCE 6 AUTHOR: E. J. STUBBE COUNTRY: BELGIUM RESTRICTIONS: NONE DATE: JANUARY 1986 PUBLISHER'S IDENTIFIER: TE.NU/EST/ml NUREG/IA-0008 TITLE: ASSESSMENT STUDY OF RELAP5/ MOD 2 CYCLE 36.01 BASED ON THE DOEL-2 STEAM GENERATOR TUBE RUPTURE INCIDENT OF JUNE 1979.

This report presents an assessment study of a pressurized water reactor (PWR) steam generator tube rupture (SGTR) transient that eccured at the DOEL-2 power plant in Belgium on June 25, 1979. The code is judged capable of simulating the observed phenomena reasonable well. Identified weaknesses of the code include:

a. Excessive water level swell in the intact steam generator (attributed to excessive interphase momentum transfer in the riser region)
b. Excessive interphase heat and mass transfer for condensation and evaporation in the affected steam generator.

It was also noted that the predicted behavior of the water-vapor flow slip in the pressurizer and the steady-state break flow exhibited significant improvements over RELAP5/M001/19.

F-6

Pag 2 No. 1-03/11/87 REPORT

SUMMARY

REFERENCE 7 AUTHORS: ANDERS SJ0 BERG, DAVID CARAHER COUNTRY: SWEDEN RESTRICTIONS: NONE DATE: JUNE 1986 PUBLISHER'S IDENTIFIER: STUDSVIK/NP-86/66 NUREG/IA-0009 TITLE: ASSESSMENT OF RELAPS/ MOD 2, AGAINST TWENTY-FIVE DRYOUT EXPERIMENTS CONDUCTED AT THE ROYAL INSTITUTE OF TECHNOLOGY.

RELAP5/ MOD 2 simulations of post-dryout heat transfer in a 7 m long, 1.5 cm diameter heated tube are reported. The Biasi critical heat flux correlation is shown to be inadequate for predicting the experimental dryout. RELAP5 accurately predicted the measured temperatures downstream of the dryout once it was forced to predict dryout at the experimentally measured location.

F-7

p;ga No. 1 03/11/87 -

REPORT

SUMMARY

REFERENCE 8 AUTHORS: G. TH. ANALYTIS AND M. RICHNER COUNTRY: SWITZERLAND RESTRICTIONS: ICAP ONLY DATE: APRIL 1, 1986 PUBLISHER'S IDENTIFIER: TM-32-86-10 TITLE: IMPLEMENTATION AND ASSESSMENT OF A NEW BUBBLY / SLUG FLOW INTERFACIAL FRICTION CORRELATION IN RELAPS/ MOD 2/36.02.

A new bubbly / slug flow interfacial friction correlation is derived and its implementation in RELAPS/ MOD 2 is explained. Comparison of code calculations with rod bundle data from NEPTUN boil-off experiments j show that the new correlation is in excellent agreement with the data, whereas the standard RELAP5/M002 code does a poor job of calculating the data.

F-8

Page N3. 1-03/11/87-REPORT SulHARY REFERENCE 9 AUTHORS: 0. ROSDAHL, D. CARAHER COUNTRY: SWEDEN RESTRICTIONS: NONE DATE: SEPTEMBER 1986 PUBLISHER'S IDENTIFIER: STUDSVIK/NP-86/99 NUREG/IA-0007 TITLE: ASSESSMENT OF RELAPS/ MOD 2 AGAINST CRITICAL FLOW DATA FROM MARVIKEN TESTS JIT11 AND CFT 21 RELAP5/M002 simulations have been conducted to assess the critical flow model in RELAPS. The experiments chosen for the simulations were i Marviken Jet Impingement Test (JIT) 11 (saturated steam flow) and Marviken Critical Flow Test (CFT) 21 (subcooled and two-phase flow).

The experimental facility consisted of a large vessel 5.2 m in diameter and 22 m high having a total volume of 420 m**3. A discharge pipe containing a valve, a nozzle, rupture discs and assorted transducers was attached to the bottom of the vessel.

The experiments used nozzle diameters of 0.3 and 0.5 meters. For all i the RELAPS simulations the experimentally measured fluid conditions in

! the vessel were used as boundary conditions. RELAPS overpredicted the i experimental flow rates by 10 to 25 percent unless discharge coefficients were applied.

F-9

, , ~e-

Paga No. 1-03/11/87 REPORT

SUMMARY

REFERENCE 10 AUTHORS: 0. ROSDAHL, D. CARAHER COUNTRY: SWEDEN RESTRICTIONS: NONE DATE: SEPTEMBER 1986 PUBLISHER'S IDENTIFIER: STUDSVIK/NP-86/97 NUREG/IA-0006 TITLE: ASSESSMENT OF RELAP5/ MOD 2 AGAINST MARVIKEN JET IMPINGEMENT TEST 11 LEVEL SWELL RELAP5/ MOD 2 simulations have been conducted for MARVIKEN Jet Impingement Test 11. The purpose of the simulations was to assess the ability of the code to simulate level swell in a large vessel.

RELAP5/M002 simulations were conducted using 20,40 and 100 nodes to model the vessel.

i Results of the simulations imply that the interfacial drag force in RELAPS/ MOD 2 fell off too rapidly with increasing void fraction. Also, nodalization studies demonstrated that the RELAPS/ MOD 2 solution did not converge with an increased number of nodes. Finally, the computed results were sensitive to time step size. The time step size had to be reduced considerably below the value allowed by the RELAP5/M002 time step control algorithm to obtain converged results during the time when the level was rising.

The report recommends that refinements to the interphase drag model be made, but suggests that the refined model should be geometry dependent.

F-10

Page No. 1 03/12/87 REPORT

SUMMARY

REFERENCE 11 AUTHORS: P. MOEYAERT, E. STUBBE COUNTRY: BELGIUM RESTRICTIONS: ICAP ONLY DATE: OCTOBER 1986 PUBLISHER'S IDENTIFIER: NONE TITLE: ASSESSMENT STUDY OF RELAPS/M002 CYCLE 36.04 BASED ON SPRAY START-UP TEST FOR DOEL-4 This report presents an assessment study of RELAPS/ MOD 2 basedon a pressurizer spray start-up test of the Doel-4 power plant.Doel-4 is a three loop WESTINGHOUSE PWR plant with a nominal power rating of 1000 MWe, This particular start-up test principally involves phenomena in one plant component, the pressurizer. Although the pressurizer spray and heater systems do not strictly have a safety function, they have a large impact on the operational flexibility of the plant (i.e. plant pressure control). Features of the test important code assessment are listed below:

- Full scale Pressurizer thermalhydraulics: which involve the thermal non equilibrium phenomena leading to condensation (by spray) and evaporation (by heaters).

Surgeline thermalhydraulics: form loss factors and counter current flow behavior.

Structural heat from pressurizer shell and heaters.

Asymmetric loop behavior when one or two reactor coolant pumps are tripped, leading to reversed loop flow.

- Single phase pump behavior (coastdown).

F-11

Prg3 No. 1 03/11/87 REPORT

SUMMARY

REFERENCE 12 AUTHOR: G. TH. ANALYTIS COUNTRY: SWITZERLAND RESTRICTIONS: ICAP ONLY DATE: JUNE 1986 PUBLISHER'S IDENTIFIER: NONE TITLE: ASSESSMENT OF TRAC-BD1/ MOD 1 WITH BOIL-OFF AND REFLOODING EXPERIMENTS: MODEL IMPROVEMENTS AND NUMERICAL PROBLEMS An assessment of TRAC-BD1/ MOD 1 was performed using data from NEPTUN boil-off and reflood experiments. The assessment against boil-off experiments identified a number of problems with the code.

Modifications to fix some of the problems were implemented and evaluated.

A new bubbly / slug interfacial friction correlation was implemented in the code to correct the miscalculation of collapsed liquid level and CHF. The new correlation results in a calculation that is in excellent agreement with the experiments. An alteration in the single phase limit of the phasic momentum equations eliminated the calculation of high negative liquid velocities at a vapor fraction of one, without changing the calculation of steam velocity. This alteration reduced the code's computation time. Details of the changes made to the code are documented in the report's appendices.

The assessment of TRAC-BD1/ MOD 1 against reflood experiments was limited by severe numeric problems. It was felt the numerics overshadowed the physics being calculated. The nature of the numeric problems is discussed in the report.

i F-12

Pige No. 1 03/11/87 REPORT

SUMMARY

REFERENCE 13 AUTHORS: E. FREI, F. STIERLI COUNTRY: SWITZERLAND RESTRICTIONS: ICAP ONLY DATE: MAY 1984 PUBLISHER'S IDENTIFIER: TM-24-83-17 TITLE: NEPTUN: INFORMATION ABOUT THE REFLOOD EXPERIMENTS 5012-5056 i

This report presents a collection of information necessary for the evaluation of the measured data collected during NEPTUN reflood experiments 5012-5056. ..

The experiments were performed from September 1981 through May 1983.

The rod bandle consisted of 4 unheated guide-tubes and 33 heated rods, 5 of which (in the center part of the bundle) were fitted with LOFT-thermocouples.

i F-13

Page No. 1 03/11/87 REPORT

SUMMARY

REFERENCE 14 AUTHOR: S. GUNTAY COUNTRY: SWITZERLAND RESTRICTIONS: ICAP ONLY DATE: JULY 1983 PUBLISHER'S IDENTIFIER: TM-32-83-19  !

I TITLE: PRELIMINARY INVESTIGATIONS ON THE REPEATABILITY OF j THERMAL-HYDRAULIC DATA OBTAINED AT THE NEPTUN BUNDLE REFLOODING 1 EXPERIMENTS.

There have been nine NEPTUN bundle reflooding experiments performed which are the repetition of certain experiments conducted before~. The aim of repeating certain experiments is to show how far an experiment can be repeatable in the same test facility, and to investigate the effect of some parameters due to aging and/or power cycling.

This work is conducted to present several comparisons which display the behavior of several sensors obtained at two or three repeated experiments and to draw certain preliminary conclusions from the behaviors for future use. This paper presents conclusions specific to every sensor behavior and also gives some comments on the repeatability.

$t..

o.

~

- F-14

Pags No. I 03/11/87 REPORT

SUMMARY

REFERENCE 15 AUTHORS: W. K0HLER, D. HEIN COUNTRY: GERMANY RESTRICTIONS: NONE DATE: SEPTEMBER 1986 PUBLISHER'S IDENTIFIER: NUREG/IA-0003 TITLE: INFLUENCE OF THE WETTING STATE OF A HEATED SURFACE ON HEAT TRANSFER AND PRESSURE LOSS IN AN EVAPORATOR TUBE The influence of the wetting state of a heated surface on heat transfer and pressure loss in an evaporator tube was investigated for a parameter range occurring.in fossil-fired steam generators. Included in the analysis are quantities which determine the wetting state in steady and transient flow.

The experimental work consists of the following:

I

- occurrence of critical heat flux (CHF) and post-CHF heat transfer in a vertical upflow evaporator tube I - influence of pressure and enthalpy transients on heat transfer in the unwetted region i

- influence of pipe orientation on heat transfer

- two phase flow pressure loss in wetted and unwetted regions ,

Based on these experiments, a method of predicting CHF for a vertical upflow evaporator tube was developed. The heat transfer in the unwetted region was newly formulated taking into account thermal nonequilibrium between the water and steam phases. Wall temperature excursions during pressure and enthalpy transients are interpreted with'the help of the boiling curve and the Leidenfrost phenomenon. A method is developed by means of which it is possible to determine the influence of the pipe orientation on the location of the boiling crisis as well as on the heat transfer in the unwetted region. The influence of the wetting state of the heated surface on the two phase flow pressure loss is interpreted as " Wall effect" and is calculated using a simplified computer model.

F-15

P ge No. 1 03/11/87 REPORT

SUMMARY

REFERENCE 16 AUTHOR: I. V0JTEK COUNTRY: GERMANY RESTRICTIONS: NONE DATE: SEPTEMBER 1986 PUBLISHER'S IDENTIFIER: NUREG/IA-0002 TITLE: HEAT TRANSFER PROCESSES DURING INTERMEDIATE AND LARGE BREAA LOSS-OF-COOLANT ACCIDENTS (LOCAS)

This project investigated the heat transfer regimes that occur during the high pressure portion of blowdown in a vertical flow path. The study focuses on those phenomena which are most important to nuclear reactor safety during postulated loss-of-coolant accidents. In particular, forced convection film boiling heat transfer and maximum and minimum critical heat flux were investigated.

Various correlations, which are used or were developed for the analysis of reactor safety problems, are described in the report.

Experimental results of the 25-rod bundle blowdown heat transfer tests, which were performed at the KWU heat transfer test facility in Karlstein, were used to assess the correlations.

The evaluation of correlations for the prediction of critical heat flux, film boiling heat transfer coefficients and mir.imum film boiling temperature showed that none of the correlations should be used over the entire range of test parameters investigated.

Using results of this investigation a new equilibrium correlation for the calculation of forced film boiling heat transfer coefficients has been developed. This correlation is shown to agree well with the experimental data over-the following range of test parameters: Mass flow rate 300 to 1400 kg/m**2-s, pressure 2 to 12 MPa and quality 0.3 to 1.0.

F-16

pig 3 N). 1 03/11/87 REPORT

SUMMARY

REFERENCE 17 AUTHORS: H. TUOMISTO, P. MUSTONEN

COUNTRY: FINLAND RESTRICTIONS: NONE i

I DATE: OCTOBER 1986 PUBLISHER'S IDENTIFIER: NUREG/IA-0004 TITLE: THERMAL MIXING TESTS IN A SEMIANNULAR DOWNCOMER WITH INTERACTING FLOW FROM COLD LEGS This report describes the test facility and test urogram for studying thermal mixing of high-pressure injection (hpi) water in the two-fifths scale model of three cold legs, semiannular downcomer and lower plenum of a pressurized water reactor. This test series has been carried out by mutual agreement on the pressurized thermal shock (pts) information exchange between the U.S. Nuclear Regulation Commission and Imatran Voima Oy.

The test facility was originally designed to model the Finnish Loviisa plant but it redesigned and modified for this test program. The facility can be operated at atmospheric pressure with loop and HPI flows from different cold legs in the area of interest to PTS.

Transparent materials were used to allow flow visualization during the tests. The choice of transparent materials limit the upper temperature to 75 C. The full buoyancy effect was induced by salt addition and the HPI temperature was used as a tracer.

i' The test matrix consists of 20 tests. The varied parameters were flow rates and the number and configuration of cold legs with HPI and loop flows. Four tests were done with decreasing loop flow temperature to simulate primary flows during steam line breaks.

1 F-17

P:93 N3. 1 03/11/87 REPORT

SUMMARY

REFERENCE 18 AUTHORS: LARS SARDH, KURT M. BECKER COUNTRY: SWEDEN RESTRICTIONS: ICAP ONLY DATE: FEBRUARY 1986 PUBLISHER'S IDENTIFIER: KTH-NEL-36 TITLE: ' ASSESSMENTS OF THE CHF CORRELATIONS BASED ON FULL-SCALE R0D BUNDLE EXPERIMENTS.

The Barnett, the Becker, the Biasi, the CISE-4, the XN-1, the EPRI and the Bezrukov burnout correlations have been compared with burnout measurements obtained with full scale 81, 64, 36 and 37-rod bundles.  ;

The results clearly indicated that the Biasi and the CISE-4 correlations do not predict the burnout conditions in full-scale rod bundles. Since these correlations yield non-conservative results their use in reactor systems computer programs should be avoided.

The Becker and the Bezrukov correlations were in excellent agreement

, with the experimental data. However, it should be pointed out that the Bezrukov correlation only covered the 70 and 90 bar data, while the Becker correlation agreed with the experimental data in the whole pressure range between 30 and 90 bar. The Barnett, the XN-1 and the EPRI correlations were also in satisfactory agreement with the experiments.

It is concluded that for predictions of the burnout conditions in full-scale BWR rod bundles the Becker correlation should be employed.

i F-18

Paga No. 1 03/11/87 REPORT

SUMMARY

l l

REFERENCE 19 AUTHORS: T. AKIM0TO, J. FUKUCHI, S, ZANE, E. KASEI, T. TASHITA, T. AKIBARA, K. OSHIBO, Y. MURAMO COUNTRY: JAPAN RESTRICTIONS: JAERI PROPRIETARY DATE: DECEMBER 1985 PUBLISHER'S IDENTIFIER: NRC 1710 TITLE: OPTIMUM EVALUATION CODE IMPROVEMENT REPORT-6-TRAC-PF1 CODE CONSTITUTIVE EQUATIONS.

A study was performed to document the coding of constitutive equations in selected models employed in TRAC-PF1. The following models were examined:

Wall shear stress model Interface shear stress model Wall heat transfer model Interface heat transfer model Interface sharpener model Subcooled boiling model Horizontally separated flow model The connection between the TRAC-PF1/ MOD 1 manual and actual coding was clarified. Documentation of the coded equations, supplementary to the manual's documentation, was produced.

F-19

l l

APPENDIX G ICAP MEETING MINUTES

! Regular meetings of the International Code Assessment and Applications Program (ICAP) participants provide a forum for the exchange off. technical and administrative information pertinent to the ICAP. Included in this

( appendix are the minutes from four ICAP meetings. First, the minutes from l a Specialist Meeting and a Task Group Meeting, both held in Erlangen, 1 '

F.R. Germany in June 1986, are presented. Next, the minutes of the Second

~

Program Management Group Meeting, held in Bethesda, MD, USA, in November 1986, are included. Finally, minutes of the Specialist Meeting of January 1987, held at Winfrith, United Kingdom, are presented.

I i

4 i

i 1

a 4

i l

G-1

1st Specialist and Task Group Meetings of the International Code Assessment Program June 9-12, KWU, Erlangen, F. R. Germany R. Jenks & T. Knight Los Alamos National Laboratory ICAP First Specialist Meeting of Code Users Monday, June 9 Opening Remarks by F. Winkler, KWU- Welcome was extended to all ICAP partic-ipants from KWU. Mr. Winkler emphasized KWU commitment to ICAP and identified PKL and UPTF as important KWU experiments. These experiments are used as data base for present code assessment activities.

Opening Remarks by D. Ross, USNRC- Several overheads were presented showing how Appendix K revision had evolved.

He stated that a " Technical Basis Report" of approximately 2000 pages that summarizes LOCA research would be available in August.1986. There will be group discussion of this document at the next ICAP meeting. He stressed the importance of code uncertainty quan-tification and noted that an important presentation discussing methodology would be given during the technical group meeting later in the week.

The purpose of ICAP was also identified and some scheduling information was given.

Specifically. he stated that the next frozen TRAC-PWR version provided by NRC (TRAC-PF1/ MOD 2) would be available in June.1987 and would include improved modehng and 3D 2-step numerics. He said approximately $10 to $15M per year would be spent on code development. assessment and providing " user friendlynesr." features to the codes.

Future plans were discussed and NRC's intention to increase emphasis on international cooperation in tiansient and break analysis was reiterated.

RELAP5/ MOD 2 Basic and Separate Effect Test Data - Richner, ElR- This pre-sentation assessed RELAP's performance against NEPTUN Boil-off and Reflood experiments.

Original code overpredicted heat transfer and underpredicted liquid entrainment and hence col-lapsed liquid level.

RELAP nodalization sensitivity generated active discussion and concern from participants following Richner's presentation. This was identified as a high priority issue. There was a perception that nodalization sensitivity increased at low flooding rates. It was stated that sensitivity places greater emphasis on the importance of scalability statements.

RELAP5/ MOD 2 Basic and Separate Effect Tez,t Data - Wilson, INEL- Gary's talk focused on assessment of RELAP's new crossflow model against EPRI single-phase liquid data. A pressure perturbation indicated by data and not predicted by code was attributed to lack of a turbulence model in RELAP.

RELAP5/ MOD 2 Basic and Separate Effect Test Data - Puzalowski, KWU- Pre-sentation showed assessment results for both TRAC-PF1 (version 7.6) and RELAP5/ MOD 2.

Actually. the results were mostly for TRAC. because RELAP calculation failed to run past 0.27s. The simulations were for an ATWS safety valve experiment with two-phase mixture discharge. Experimental facility is located at KWU in Karlstein.

Some high pressure peaks gave version 7.6 problems. Cell lengths were limited to 0.1m.

Using version 12.5 helped. especially with changes to interfacial heat transfer. After talk.

1 G-2

Jenks. R. and Knight. T. Page 2 it was brought up that results from this calculation are useful for evaluating water hammer effect. The question was asked: "What regime should have had changes made to the interface area?" Puzalowski indicated turbulent and bubbly flow.

It was brought up that TRAC version 7.6 was not designed to work at the pressures reached in the experiment. Puzalowski indicated they had similar problems with PF1/ MOD 1 with EC12.5. but did not show any of these results.

RELAP5/ MOD 2 Integral Test Data - Sjoeberg, Studsvik- An assessment of RE-LAP against 25 dryout experiments was presented (no overhead copies were obtained). Com-parison of code to experiment was characterized as "not too bad." There appeared to be little nodalization sensitivity. Finer heater rod nodalization (ten nodes compared with two) did not change CHF (dryout) location.

l RELAP5/ MOD 2 Integral Test Data - Friedmann, KWU- A post test calculation of PKL ll B-2 was presented. The calculation simulated a large break LOCA (cold leg) with both hot and cold leg pumped ECC injection. Although it was stated that "allimportant phenomena were simulated" there were some problems observed. Steam production in the SG tubes was too high. Friedmann also stated they had problems at the core tie plate attributable to CCFL modeling deficiencies. Furthermore. ECC injection port modeling presented them with some difficulties, and he said cold leg injection port configuration of angle-injection has to be taken into account.

RELAPS/ MOD 2 Integral Test Data - Wilson, INEL- Both a Semiscale small break test (S-UT-8) and a GERDA small break test (1605AA) RELAP simulation were presented.

Slides were shown which indicates better agreement to Semiscale core collapsed liquid level was j obtained with an arbitrary reduction in interfacial drag. Gary recommended using crossflow junctions at hot and cold leg connections to the vessel. The conclusions from the GERDA work were that "the physical phenomena observed occurred in the calculation and were generally l, well calculated."

Following the talk questions were raised about the sensitivity of the code to changes in interfacial drag. Gary Johnsen. INEL. indicated that changes to interfacial drag will Le based on separate effects test and not on integral test facility data (like Semiscale or GERDA). There also was some concern of RELAPs inability to calculate superheat conditions.

RELAP5 1D6 was pres / MOD ented. 2 Integralsimulated The calculation Test Data - Nguyen, a small KWU-break LOCA A leg)

(cold RELAP simulation with hot side of PKL upper plenum pumped ECC injection. Although the conclusions stated that there was generally good agreement between code and data, mass flow out the SG primary is not well predicted (see Fig.10 of overheads).

Also, collapsed liquid level in the SG separator was not well predicted. Some sensitivity studies were presented showing results of changes to core model, number of loops, and discharge coefficient. Liquid levels appeared to be most sensitive to discharge coefficient.

RELAP5/ MOD 2 Power Plant Data - Gerth, KWU-- A full plant simulation using RELAP was presented. The calculation was for a reactor trip at fullload, a commissioning test for a PWR. There were some problems getting a steady state calculation with acceptable initial values and a lot of time and computer cost was used in the process. No power comparison was given for time after reactor trip. It was also stated that the initial value for reactor vessel inlet temperature could not be adjusted to the measured value. This was attributed to the heat transfer package in RELAP5/ MOD 2.

l G-3

Jenks. R. and Knight. T. Page 3 RELAP5/ MOD 2 Power Plant Data - Aksan ElR- An overview of the assessment-of thermal hydraulic transient codes at ElR was given. He reported that they had received a stripped version of TRAC.

ElR is giving more emphasis to separate effects work. He said "we are doing assessment.

but most work is directed at development." He identified the steps that ElR would take with regard to their code activities as follows.

. Frozen version would be used.

. An initial base case would be developed.

. Nodalization sensitivity would be investigated.

. Sensitivity to code options would be explored.

. A new code model would be developed.

. Results would be analyzed to determine if code change needed.

If modifications to the code were necessary the following would be important.

. Identification of the regime and phenomena.

. Selection of appropriate correlation.

. Comparison to data (but not temperature).

. Attention focused on numerical oscillations.

Following the EIR talk. there was some discussion of quality assurance. NRC indicated (N. Zuber) that quality assurance documents would be provided by March 1.1987. He stated that quality assurance means "what's in code is what's in manual."

TRAC-PF1/ MOD 1 Basic and Separate Effect Data - Riegle, GRS- Pre-test cal-culations for UPTF hot and cold leg flow pattern test were presented. A three-dimensional VESSEL was employed in the analysis with 13 levels. 4 azimuthal sectors and 2 radial rings.

The pre-test calculations should oscillatory behavior in mass flows at breaks. mass flows at hot / cold legs and core average pressure.

Riegle summarized by saying hot and cold leg oscillations similar to GPWR calculational results were observed. With coarse VESSEL noding, approximately 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> CPU per 100s transient was required.

TRAC-PF1/ MOD 1 Basic and Separate Effect Data - Glaeser, GRS- The results of calculations for UPTF No.12. subcooled tie plate CCF test and for Karlstein Upper Tie Plate CCF were presented. For the UPTF calculations first water break through well predicted.

However, with coarse nodalization local break through areas not well predicted. He indicated that posttest calculations will be done with fine nodalization.

For the Karlstein upper tie plate CCF test calculations total carryover point was reasonably well predicted. However. right flooding behavior not predicted. Also water downflow rate was overpredicted in the CCF (flooding) region. Finer nodalization did not lead to better steady state CCF results. It was recommended that modification or extension of the interfacial shear equations be done.

END OF MEETING, Monday, June 9 l

l l

G-4

Jenks R. and Knight. T. Page 4 ICAP First Specialist Meeting of Code Users (continued)

Tuesday, June 10 TRAC-PF1/ MOD 1 Basic and Separate Effect Data - Richards, UKAEE- Chris presented a calculation which simulated the CISI countercurrent flow limitation tests pressur-izer' flooding). The test facility was described, and the TRAC model outline'd. For model (2 TEES connected by a PIPE with 2 FILLS. 2 BREAKS and a VALVE) the results showed that TRAC over predicted liquid downflow at the lower rates of steam upflow. The test was conducted at 7MPa in a 90mm (i.d.) test section (see overheads for more detail).

TRAC-PF1/ MOD 1 Basic and Separate Effect Data - Richards, UKAEE-The next presentation by Chris Richards gave UK's experience with modeling of stratified countercurrent flow in horizontal pipes and compared TRAC results to an exact solution.

In a horizontal pipe section liquid holdup was predicted very well, but not the pressure drop. He explained to group that interface friction and gravity terms control CCFL.

He also stated that momentum flux terms cannot be neglected (as done in TRAC) for delta-x .It. (rho-sub-k

  • velocity-sub-k ** 2) / (g
  • delta-p)

Richards also presented some work done by P. Black which identified a neglected term in the TRAC momentum equations. He said he had sent us a letter detailing this work.

A " dam break" model was illustrated which showed effect of neglected momentum term.

Results for a " modified TRAC" compared much better. He requested that we provide clarifi-cation of implementation logic for stratified flow model.

He concluded that TRAC has deficiencies in the momentum equation. TRAC should use delta-x instead of delta-p. The results using their modified code looked good.

During discussion after Chris's talk it was stated that in the 3D. terms are not present to allow smoothing of liquid levels.

David Bessette said that the problems identified by UK indicated that more communication with NRC was needed and asked that NRC be on distribution for all correspondence sent to US labs. Fuot Odar stated that funding drives many of the decisions at NRC. and it is very important that NRC be kept informed.

A question was asked as to whether other flow regimes had been examined. Richards indicated they had not.

TRAC-PF1/ MOD 1 Basic and Separate Effect Data - Coddington, UKAEE- Paul presented results of a TRAC analysis of separate effect data from BCL downcomer bypass tests. This work was a carryon from work done at BNL with PD2. He indicated that he had examined input decks received from both BNL and LANL and that they were different. He was asked which one was right. To Paul. it seemed that LANL had used the right HDs. but the wrong FAs whereas BNL had used the right FAs. but the wrong HDs. The LANL VESSEL was 2-r. 6-theta. 8-z. The BNL VESSEL was 2-r. 4-theta. 9-z.

At the junction between vessel and broken cold leg. flow was predicted as stratified by TRAC. NRC (N. Zuber) asked what would the results be if UK had used the correct input.

Paul resporded that they did not know, but was one thing they would like to find out.

Poorer results were evident with higher subcooling. Overall counter current flow was achieved in 3D by downflow balanced in other cells by upflow. He indicated that the momentum flux terms are artificially large which therefore means in 3D it is very difficult to get CCF. Also stated that the condensation rate was too small for high subcooling.

G-5

'Jenks. R. and Knight. T. Page 5 TRAC-PF1/ MOD 1 Basic and Separate Effect Data - Richards, UKAEE- This presentation detailed some aspects of critical flow modeling with TRAC.

Some of the items of concern indicated by Richards are as follows.

. Large nodalization effects evident. We need guidelines. Guidelines should also give reasons for their use. Guidelines especially needed for critical flow modeling for SBLOCAs.

. Two-phase critical flow model differs from HEM.

. Their is a two-phase undocumented feature in the code which indicates that some work has been done in this regard.

. SOUND subroutine has possible error and needs more accuracy.

. Other concerns (see Richards' overheads).

He also noted that in the critical flow model in TRAC an iteration is performed to determine the characteristic of the matrix, but then never changes original conditions calculated by HEM.

Choking depends on alpha and L/D. and the range of L/D determines the method for calculating alpha. Chris said that by increasing length of pipe a small amount (to L/D just over 1.5), flow was doubled in an example problem. This he found intuitively nonphysical.

He also alluded to some superheat problems.

TRAC-PF1/ MOD 1 Basic and Separate EsTect Data - Coddington, UKAEE- Paul's presentation gave results of R. O'Mahoney*s malysis of THETIS reflood hydraulics. The TRAC THETIS model is composed of a VESSEL. two PIPES. a FILL and a BREAK.

It was emphasized that prediction of void profile versus height and time is important to get correct rod temperatures. His results showed comparisons of temperature and void fraction at various heights. The void was derived from delta-p measurements.

He said that using the interface sharpener caused a lot of problems. On this point he elaborated with the following.

. Entrainment correlation is proportional to the sixth power of vapor velocity (over sensitive).

. There is no consistency check between interface friction and entrainment correlation.

. There are problems tracking interface.

. What are shroud contributions.

Other areas of concern were as follows.

. Thermodynamic properties-improve accuracy over different ranges.

. Some concern with 2-phase mixture calculation.

. Calculations where P .gt. P-critical.

. Nucleation delay in all components.

. Collection of all code data statements into a single section of the code.

. Adapting TRAC to specific applications.

Paul stated that letters had been sent to Los Alamos recently documenting some of these concerns. All other errors had either been fixed or are on Los Alamos " unresolved problems" list. He did say that a convergence problem in T-CHF had not yet been communicated to Los Alamos. He indicated that there are no guidelines on use of INVAN which has default =0 now.

However INVAN=1 uses T-CHF which is better than T-SAT for evaluation of condensation heat transfer. He stated that John Dawson at Barclay was doiag work on condensation.

He also said they had received heat structure updates from LANL and were working on implementation in TRAC. >

They had looked at Ishi's recent paper which divided entrainment into several regions ar.d were considering this. They had also looked at COBRA /TF. implemented that correlation and modified interfacial shear package.

! G-6 l

l

Jenks. R. and Knight. T. Page 6 Paul also stated that "without interfacial sharpener and without effect of shroud heat.

TRAC predicts data" for THETIS. In fact, they found that the best approximation was ob-tained when interface sharpener was removed completely.

At this point. NRC (N. Zuber) commented that two options appeared available to address void distribution in core:

1. Revise heat transfer package in code or
2. improve hydro package.

The latter of these two was supported by himself. Katten and Coddington, he implied.

LANL (T. Knight) commented on interface sharpener, saying that there were also calcu-lations where results got worse when interface sharpener was turned off.

TRAC-PF1/ MOD 1 Basic and Separate Effect Data - Coddington, UKAEE-Some user guidelines for TRAC modeling were presented (form losses, momentum at TEES. etc).

He recommends never using NFF option for TEES as he says that TRAC does not calculate loss coefficients accurately. The flow area is not correct for NFF options. Even stronger statement was made about not using NFF less than zero for TEES.

With regard to internal vessel modeling. he said TRAC does not allow for change in direction of fluid - momentum is lost.

NRC (N. Zuber) asked if UK will look at phase changes at TEES and Ys. Paul said they intend to investigate.

TRAC-PF1 was compared a/ MOD gainst 1 Integral PKL-llB3 Testpresentation.

data in this Data - Trambauer, TrambauerGRS- A pretest indicated that TRAC calculation (version 11.1) predicted results very well. He indicated that critical flow limitation depended on the nodalization.

TRAC-PF1/ MOD 1 Integral Test Data - Richards, UKAEE- Another talk was given by Chris Richards. This one presented results of post test calculations for ISP18 LOBl/ MOD 2 integral test facility) with TRAC-PF1. They found that adjustments to condens(ation rate bubble size were necessary to match data. They had problems with intact cold-leg density prediction.

There was some discussion on use of INVAN option with better results obtained with INVAN=1. They were concerned about interfacial condensation heat transfer, critical flow from non-equilibrium stratified conditions and interphase friction in bundle geometries.

As a final note they gave a list of items they would especially like to have in TRAC:

. Code must be fast and reliable:

. There must be no discontinuities: and

. Code must be equipped with suitable "diais" for parametric assessment studies (user needs easier way to get at constants in block data).

TRAC-PF1/ MOD 1 Power Plant Data - Sjoeberg, Studsvik- This presentation described a loss of grid transient calculation for a three loop PWR. A " step-wise method" was employed to obtain the steady state condition. Tube heat transfer areas were adjusted to give steady state heat transfer from primary to secondary. Loss coefficients were adjusted to give correct pressure drops in secondary. initial steam generator inventory was corrected by adjustment of void fractions.

Sjoeberg states that the frozen version simulates the loss of grid transient with acceptable accuracy. He indicated that some difficulties existed in modeling the pressurizer spray, and G-7

Jenks. R. and Knight. T. Page 7 an excessive primary pressure was predicted due to lack of secondary steam flow. He stated, also, that the excessive primary pressure was probably due to upscaling the steam generator heat transfer area to match steady-state conditions. This probably allowed too much heat transfer in the transient and excessively raised secondary pressure thus inhibiting primary-to-secondary heat transfer. Consequently. a higher primary pressure occurred.

TRAC-PF1/ MOD 1 Assessment Overview - Knight, Los Alamos- An overview of independent assessment of TRAC-PF1/ MOD 1 in the United States was presented (see overheads for detail). Only question that was raised related to wetting on U-tube steam generator.

TRAC-BD1/ MOD 1 Basic and Separate Effect Data - Analytis, ElR- Several de-ficiencies were identified in the heat transfer simulation with TRAC-BD1. the code developed at INEL for BWRs. New implementation of interfacial friction correlation for bubbly / slug regime was required to give agreement with data. Heat transfer to the vapor was a little over-predicted. Several numerical oscillations were observed. For low flooding rates large differences existed between data and calculation. Several recommendations were made for TRAC-BD1/ MOD 1 improvement (see ElR report).

END OF MEliTING, Tuesday, June 10 G-8

-~ - - ~

Jenks. R. and Knight. T. Page 8 ICAP Task Group Meeting Wednesday Afternoon, June 11 The ICAP Task Group Meeting began after lunch and began with summary remarks-about the preceding ICAP Specialist Group meeting. David Bessette prompted roundtable discussion by calling upon participants _from member nations to give some re. marks regarding the Specialist Group meeting.

NRC (N. Zuber) stated that it had been the best code assessment meeting in the 13 years he has been in code assessment activities.

!. There was some concern stated (Wolfert. GRS) that we might be_ headed away from original goal. NRC (F. Odar) said that "to quantify code uncertainty" was our first objective. "

However, to model all the important phenomena, even MOD 2 may not be sufficient. He said that.NRC is shooting for final quantification in 1990. He also said that the CS9 matrix forms r basis of ICAP assessment matrix. But, we may need more or different tests, and we should j discuss this.

To the issue of more tests. KWU (F. Winkler) responded that integral tests results are adequate. but that separate effect comparisons showed that all codes had deficiencies. We '

must identify which kind of separate effects tests will provide information needed to assess improvements to these deficiencies. Question was raised (both GRS and KWU) "is_ problem in modeling or models for full-scale simulations?" We need discussion on this.

! NRC (F. Odar) replied that the code developers will read all these reports. combine and

coordinate results, and submit proposal as to what needs to be done. He said NRC wants

! international feedback. particularly with regard to nodalization.

NRC (N. Zuber) said that we need to understand why code responds ~ the way it does i to nodalization changes and also said (F. Odar) that guidelines must be provided as to the minimum number of cells required to model a plant for a given transient.

ElR (Analytis) responded to this with a question: "What if nodalization recommendations suggest that large number of cells are required? How can we justify large run times?" NRC (N. Zuber) replied that we cannot compromise physics for numerics and suggested that this a points to using different quality of model for each scenario.

UKAEE (P. Coddington suggested that phenomena are most important and that first j dominant phenomena must b)e identified, and then use code with models tuned to th dominant phenomena.

ElR (Analytis) suggested that what really needs to be done is to fix interfacial friction ,

and integral correlations and get physical models in the codes that we can support.

INEL (G. Johnsen) said that the codes have geometry-dependent models. How can code f; be expected to model specific effects when it handles calculations with generalized equations?

4 Los Alamos (T. Knight) stated that the correlations in the codes auume quasi-steady.

fully developed flow as well as geometry effects.

] Studsvik said that testing against separate effects must continue, but suggested that we l ask ourselves if maybe we didn't freeze the code version too early for this assessment activity.

Perhaps we need more model development and developmental assessment.

replied that ElRin(Aksan) nothing also We life is complete. stated always that makewe need moreWe improvements. assessment.

exercise aga NRC (N. Zuber available at the time.

! JAERI (Akimoto) said he would like to have LTSS FTN manual from Los Alamos for 1

interpretation of coding. He also said that in future test problem output JAERI would like

[ output for first two time steps added.

G-9

.-u-,, .-- - -- , + -- -, - - - , ,_ , - - - - . - ---.,-,,v.-p , ,r-,, .- ----,---w., .--

Jenks. R. and Knight. T. Page 9 Los- Alamos- (T. Knight) declared there was a need for improvement in the constitutive package. He said that a lot of good work had gone into TRAC code development and the

results of the preceding Specialists Meeting indicated that TRAC-PF1/ MOD 1 predicts very well. -

Review of Procedures for Performing and Documenting Code Assessments - Bes-sette, USNRC- This presentation reviewed the guidelines for performing reporting and

transmitting code assessments. It was stressed again that USNRC must be kept informed.-

He specifically asked that for TRAC-BWR and RELAP5. one copy of assessment report each 1 should go to NRC program manager and INEL program manager. For TRAC-PF1/ MOD 1

) copies should be sent to NRC program manager and the Los Alamos user contact.

i An outline for assessment reports was shown and the "results" section was stressed. -Es-pecially important in the results section is the inclusion of recommendations for user guidelines.

particularly nodalization. Also some feedback regarding code documentation is necessary.

Other requirements pertaining to required information (particularly key parameters) for quantification of code accuracy were presented.

Methodology to Quantify Code Accuracy - Wilson, INEL- In this presentation Gary said that hopefully we can drive " user uncertainty" to a minimallevel by providing better user guidelines and code documentation." Methods developed at INEL and SNL were presented and simple examples of how'these methods would be employed were given.

Both methods provided as a final result, the capability to generate statements about code accuracy. Documentation on both methods was due out in October 1986.

. KWU (F. Winkler) had concerns over averaging techniques for heater rods (in PKL, for example). He wanted to know how averaging would be included. Will error band for data be included. Gary said that arithmetic average is best. This question was prompted by Wilson's i slide number 16. Los Alamos (T. Knight) added that the arithmetic averaging technique does not represent a problem and was an attempt to provide a simple method for data entry.

i KWU (F. Wi'n kler) stated that a general statement of code uncertainty may be too critical of code performance and, in fact, hide where problems exist in the code. The i code may predict some phenomena quite well, but be penalized for deficiencies is specific areas.

! some of which may not be important for safety considerations. It was also stated that there 4

is some danger in the fact that the numbers used in codes are often based on experiments. -

NRC (F. Odar) emphasized that we must continue to look at whether or not code predicts 4

dominant phenomena.

INEL (G. Wilson) responded that we want code to be right before we do uncertainty analysis. We may want to stop at different levels of the AOV tree (see overheads) and make multiple statements of uncertainty.

Gary stated that a frozen version of TRAC-BWR was to be released in about two weeks.

and he strongly encouraged feedback.

Several presentations were deferred to Thursday, as we were running late.

, END OF MEETING, Wednesday, June 11 i

I G-10 c___. - . - . - - _ - . - . . -- - - - - . . . - . . - -

Jenks. R. and Knight. T. Page 10 ICAP Task Group Meeting (continued)

Thursday, June 12 RELAP5/ MOD 2 Frozen Code Maintenance - Johnsen, INEL- A general overview of how INEL handles code maintenance was presented. Specifically, process of problem reso-lution. test problems, resolution history, outstanding problems and ongoing development work were described.

Guidance on problem reporting was also given, directing all correspondence to go to Gary Wilson. ICAP Program Manager. Immediate assistance requests should go to Clay Miller. and users were encouraged to investigate and fix problem if possible, because of limited staff at IN EL.

TRAC-PF1/ MOD 1 Resolving Problems - Jenks, Los Alamos- A general overview of how Los Alamos handles code maintenance was presented. Specifically.

. Configuration Control.

. Problem-tracking process.

. Setting priority.

. Resolved problems, and

. Unresolved problems were discussed.

A flowchart was presented which indicated the depth of the configuration control process at Los Alamos and it was stated that we are always seeking to improve this process and encourage external feedback.

The process for tracking problems was explained, and a flowchart displayed. There was much discussion about how priority was established within the constraints of the problem-resolution flow chart logic.

Of the approximately 600 user requests received between October 1985 and April 1986 (seven months), it was shown that nearly all had been resolved either immediately or after a tracking process was initiated to keep them from " falling through the cracks."

The significant resolved problems were listed. The unresolved errors, model deficiencies and input / output problems were then discussed, giving the source of tne reported problem.

the problem description and the current status of problem resolution. Several problems were flagged as requiring additional information for investigative procedure.

Methodology for Determining Code Scaling Capability and Applicability to Power Plants - Zuber, USNRC- Novak Zuber presented a humorous, yet sobering, overhead (We would like to have copies) depicting Nuclear Safety History in a table. The columns were

" year" " event", " era" and " buzz words." As an example, year 1950 was the start, and the era was characterized as one of " innocence / sin". The buzz word then was " bubble dynamics."

He proceeded to characterize three eras that brought us to our present state. Today we must emphasize " accountability" in the wake of Chernobyl. Challenger disasters and the associated public concern.

The thrust of his talk was regarding how to demonstrate applicability. He indicated that next month (July) there would be a meeting to do some "gedanken" experiments to initiate the code applicability process. Much additional claboration on methodology was delivered.

Talk was well received.

However. GRS (Wolfert) stated that perhaps we should hold off on code uncertainty analysis until code applicability study is complete. Novak replied that it would not take long to make this evaluation. Two or three persons from INEL and two from Los Alamos would spend 3-4 days to work through the "gedanken" experiment to form 2 data points. A report would be available in October at the next ICAP meeting.

G-11

Jenks. R. and Knight. T. Page 11 KWU (F. Winkler) also proposed that we hold off (rethink proposal) of assessment activ-ities until applicability study is complete.

UKAEE (P. Coddington) reiterated that integral tests are needed to form applicability statements. NRC (N. Zuber) responded that we have good integral tests already. To which Coddington replied that it doesn't matter how good integral tests are if they don't address the entire range of applicability.

Code Documentation and Information Transfer - Wilson /Jenks- Because of time constraints. only a brief mention of what was ongoing at INEL and Los Alamos was allowed.

Gary stressed that INEL sought to be responsive to ICAP and RELAP users in general, and requested feedback on their success in that endeavor. He also asked for suggestions on how they could improve the process.

We stated that we had wanted to discuss specifics about the following information transfer tools:

. TRAC User's Guide:

. TRAC Code Manual.

. Models & Correlations document.

. TRAC NEWS. and

. VAX UPDATES software.

However, because of the limitation on time, we just gave some specifics about release of the documents. indicated that handouts for Jenks's presentation were available and asked for feedback on TRAC NEWS. We also sought ideas regarding our VAX UPDATES telecommuni-cations package, and elu'ded to some features that may be available (depending on resources) in the future. Several of these features were welcomed by the participants.

. Applications of Codes to Power Plant Studies - Five Talks- Deferred to next ICAP meeting because we ran out of time.

Misc. information received before end of meeting - GRS (Wolfert) indicated that at the 2D/3D meeting an alternative might be sought for hot-leg modeling by perhaps using 2 3D VESSELS or extending 3D modeling into hot leg.

Some other concerns of KWU were identified during the course of Thursday afternoon.

They are as follows.

. Hot leg ECC injection 3D effects (improve with modeling7 improve with new flow regime maps? improved condensation rates 7

. Want 3D 2-step to improve run times)'for SBLOCA and steady state calculations.

. Want generalized heat slab components.

. Improve symmetry of the TEE component.

. Friction losses.

. Pressure loss due to friction should be function of pipe roughness.

. Single flow regime dependent 2-phase friction multiplier- currently no user option.

. Fuel rod model-currently only one geometry.

Next Meeting Plans - Bessette, USNRC- It was decided that the next meetings would follow October's Water Reactor Safety Information Meeting. There was discussion on switching items between proposed Program Management Group agenda and that of the Tech-nical Group to eliminate unnecessary attendance if possible. Formulation of revised agenda was deferred, and new agenda will be provided by NRC after more informal talks.

END OF MEETING Thursday, June 12 I

G-12

ATTENDANC! LISTS l

l. ICAP Technical Group - June it. 1986 l

NAME AFFILLIATICN BUS. ADDRESS TEL. NO.

l S.A. Naff USNRC/INEL Kraftwerk Union I' HammerDacnerstrasse 12 + 14 8520 Erlangen FRG 09131-18-3905 E.G. Stucce TRACTEBEL 31 Rve Dela Science 1040 Brussels - Belgtue (32)(2)2345208 F. Winkler KWU Kraftwerk Union HammerDacnerstrasse 12 + 14 8520 Erlangen FRG 09737-78-275E K. Wolfert GRS Gesellschaft fur Reaktorsicnernett (GRS)

Forschungsgelance D-8046 Garcntng (089) 32004 104

0. Sancervag Stuasvik Stucavik Energtteknik AS S-61182 Nykoping Swooen

+46 155 21868 A. Sjooerg Studsvik Stuosvik Energiteknik AB S-61182 Nykoping Sweden

+46 155 21828 P. Coccington UKAEA Atomic Energy Estantisnment Winfritn Dorenester. Dorset England inac Q'. Knignt LANL MS K553 Los Alamos National Lacoratory P.O. Box 1663 Los Alamos. NM 87545 USA R. Jenks LANL MS K555 Los Alamos National Lacoratory Los Alamos. NM 87545 USA I. Betttain UKAEA Atomic Energy Estacitsnment Winfettn Dorenester. Dorset 0T2 80H, UK (0) 305-63111 Ext. 2039 Dav1c Bessette USNRC Wasnington O.C. 20555 301-427-4320 F. Cgar USNRC Wasnington. 0.C. 20!55 301-427-4727 Gary Wilson INEL EGG Icano. Inc.

P.O. Box 1625

!cano Falls, !cano 83415 Peter Frieomann KWU KWU 09737/783909 Peter Grucer KWU KWU 09737/783976 Cera Seeoerger KWU KWU 09737/782924 G-13

Hajime Akimoto JAERI Tokat-Mura. Icarakt-Ken Jacan 319-11 Hno-dung Ktm KAERI Nuclear Safety Center. KAERI P.O. Box 7. Daecuk-Danji Cnoong-Nam. Korea 042) 820-2613 Emanuele Negrents ENEA TERM-MSP Cre Casaccia 5.P. Anguillarese KM 1.300 0006C Roma. Italy 0039-6-69484265 L. Utntees ECN ECN 3 Westeroutnweg P.O. Box 7 Petten The Netherlands H. Catzinger KWU KWU j Hammernacnerstrasse 12 + 14  ;

O-8520 Erlangen j FRG l A. ranco GR$ GesellscMaft fuer Reaktorsichernett 8046 Garching FRG 1 089-32004-113 J.J. Pena ENVSA ENVSA l Santiago Rustnol 12 28040 Madria Spain J.H. Izcutereo CSN Consejo Seguricad Nuclear 0034-1-4561062 Ext. 358 Sor Angeth De La Cruz 3 Maarte. Spain Po*cr Hall CE98 GDCD Barnett Way Barnwood. Gloucestee. UK Gary donnsen INEL EGG Ioano. Inc.

P.O. Box 1625 Icano Falls. Icano 83415 USA K1tus Ltesen GRS Gesellsenaft fuer Reaktorsichernett

.i Forschungsgelance 0-8046 Garening (089) 32524-703 S. Nusret Aksan EIR Swtss F9eeral Institute for Reactor Researen 5303 Wuerentingen Switzerlanc 056-992684 M. Rienner E!R Swtss Feoeral Institute for Reaktor Researcn CH-5303 wureulIngen Switzerland 1 056-992677 i

P. Marstli ENEA/ DISP Comitato Nazionale per la Rtcorce a Soteuco dell Energta Nucleare a celle Energietitorettive Vlavttaltano Brancatt AS Roma. Italy 00100 00396/83282174 V. Mancuso ENEA/ TERM Comitato Nazionale Energta Nuclear C.R.E. Casaccia Via Angu111arese Km 1 3 00100 Roma Italy 00396/69484804 G-14

5. Patti Ansaldo Otv. Nira Vta Det Pescatort 16100-Genoa-Italy.

Karner WWU 09737785323 G. Preusene KWU R757 09737 78 464 F. wecer KWU Erlangen 09737 78 2327 G. Gertn KWU Erlangen 09737 1785248 N. Zuber USNRC wasntngton. O.C.

USA 301-427-4440 V. Yrjota U77 Technical Researen Centre of Finland Nuclear Engineering t,acoratory P.O. Box 169 5F-00181 Helsinkt. Finland end I

i s

I i

I i

G-15

- _ . - _ - . - - - . . . - ~ _ . .. -- - _ -

! Minutes of the Second Management Group Meeting of the International Code Assessment and Application Program

[

November 3-5. 1986 R.G. Hanson and G.E. Wilson-EG&G Idaho Inc.

)!

The second management group meeting of the International Code Assessment lr and Applications Program (ICAP) was hosted by the United States Nuclear Regulatory Commission (USNRC) on November 3-5,-1986. The meeting was held

~

j' l

at the Hyatt Regency Bethesda Hotel in Bethesda, Maryland. The meeting j was attended by twenty two representatives from fourteen European and i Asian countries and thirty domestic representatives from the national l

laboratories.-NRC staff, and the nuclear industry. Attachment 1 presents 4 the list of attendees for the meeting.

) The meeting was divided into three distinct categories. Presentations on

}

the first day were made by NRC staff. Presentations by the international j participants were made on the second day followed by presentations by the i

NRC contractors on the third day. The attached appendix contains a copy

{

of each presentation. The meeting was concluded by a panel discussion on the third day in which each member country was represented on the panel.

The meeting was openea by Mr. Philip Ting, NRC Program Manager for the ICAP program, who in turn introduced Dr. Denwood Ross. Dr. Ross reviewed

, the purpose and goals of the ICAP program and the program approach to

! achieve the goals. Dr. Ross stated that the program is driven by the

!- needs of the ECCS rule change effort. The ICAP participants were i identified as Belgium, Finland, France, F.R. Germany, Italy, Korea, Netherlands, Spain, Sweden, Switzerland, Taiwan, and the United Kingdom. ,

Negotiations are in progress with Austria, Japan, ISPRA, and Yugoslavia.

! Or. Ross highlighted significant ICAP findings and regulatory impact of the efforts. Dr. dass concluded with a program status report and i highlighted short term program milestones (annual report, code uncertainty method documentation),

i Dr. Brian Sheron represented the regulatory view of the NRC. The l regulatory significance of the ICAP centers around using world-wide i expertise to obtain acceptable and recognized standard best estimate thermal-hydraulic codes. Dr. Sheron summarized current needs relative to licensing codes and identified a future need as determining when a code is ,

good enough.

i

Dr. Louis Shotkin made a presentation of how the ICAP fits into the i overall NRC code assessment and code development programs. The NRC i adopted plan is to assess the codes against data from small-scale facilities (for the range of scales available) and extrapolate the l resulting code uncertainty statement to full scale code applications. The j point was emphasized that domestic Combustion Engineering and Westinghouse type scaled facilities and BWR type scaled facilities are shut down and the ICAP will be relied on to complete the related code assessment activities. The assessment effort for Babcock & Wilcox type configuration i will use data from currently operational facilities and tne code i assessment activities will be predominantly addressed domestically. The j code assessment studies will directly feed to code applicability and code >

l G-16 i'

i u- _ _ _ _ __. _ ._. _ _. _ _ _. _ _._ .. _ _....._ _ ._., _ _ _

uncertainty quantifi, cation efforts which were oresented in later-discussions. Dr. Shotkin stressed the point that the code applicability and code uncertainty cuantification efforts are part of a single methodology. In conclusion Dr. Shotkin emphasized how the ICAP assessment compliments current NRC activities.

Dr. William Beckner presented an overview of the ECCS rule revision. The point was made that research over the last decade has shown the current i rule to be excessively conservative and the excessive conservatism is i restricting the operation of some reactors. The revised rule is based on l best estimate-licensing calculations with demonstrated accurar.y. The i current rule will serve as a fall-back option for licensees. Dr. Beckner pointed out there is no. direct link between the ICAP and the requirements of the ECCS rule change. However in practice there may be a strong indirect link between the two activities since many licensees make use of NRC codes. Code limitations will fall out of assessment studies thereby directly affecting the future use and development of the NRC codes. Dr.

Beckner stressed that ICAP provides the mechanism for handling code errors

! and resulting corrections and improvements to the code. Dr. Beckner

concluded by pointing out that code management will remain a problem that will require continual attention.

Dr. Novak Zuber presented an overview of a method for characterizing code i scaling capability and code applicability to specific full scale-applications. It was noted that large break loss-of-coolant accident

! (LBLOCA) has been the focus of attention in the past. However, current interest relative to code applications has expanded to include small break

LOCA and operational transients. Scaling is the issue at the center of-

. the code applicability methodology. A key factor in the methodology development is to carefully address the possibility of compensating errors. Dr. Zuber identified the major reouirements that the methodology must address as:

1. Identification and ranking of important phenomena for scenarios;

) 2. Identification and assessment of the effects of scaling distortion;

, 3. Identification and ranking of code capability.

t Included in the presentation (see the appendix) was a flow chart for the

methodology. ,

Dr. Fuat Odar presented an overview of a code uncertainty quantification

, methodology which is part of the overall code applicability methodology.

, Dr. Odar outlined the code uncertainty quantification method as having the following steps:

. 1. Formulation of assessment matrices; i 2. Selection of key parameters; j 3. Performance of assessment studies;

} 4. Performance of a statistically based analysis for each of the

{ assessment studies; j G-17 1

i 1 - - - - - - - - . ,. - ~ , .-- ----.- - - - - - - - - - - -

l 4

5. Evaluation of the statistically based data population and-extrapolation to a full scale application uncertainty.

Relative to selecting key phenomena (and resulting key parameters) Mr.

Gary Wilson (INEL) briefly reviewed the Analytical Hierarchical Procedure (AHP) relative to ranking phenomena. The point was emphasized that the

AHP serves as a check of subjective ranking that has occurred. Dr. Odar

! concluded the presentation by discussing the extrapolation process to obtain a full scale uncertainty statement. The point was made that if a

. scaling dependency is observed, code development is required to eliminate the scale effects.

Dr. Odar then presented a oroposed large break LOCA assessment matrix, including the criteria used to select the test. The foundation of the j matrix was the CSNI large break LOCA assessment matrices.- The point was

! made that the proposed matrix represents a minimum set of tests. However, tests can be added to the matrix or substitutions can be made. The L proprietary nature of some data on the matrix suggested that some studies i would have to be performed by specific ICAP members.

The second day of the meeting was called to order at 8:30 A.M. The agenda for the second day called for' presentation by each partner in the ICAP-

which summarized their respective program plans and status.

I Dr. Stubbe of Belgium indicated their analyses would focus on RELAP5/M002 i applications to plant transients which have occurred or have been j performed in the DOEL power generating stations. It was noted that the ,

. first study (STGR in DOEL 2) has baen completed. A second study involving '

! a pressurizer spray test is nearing completion. Questions were raised l relative to the accuracy of plant data.

4 4

Or. Holmstrom of Finland summarized the assessment studies they intend to i perform. These are summarized in the appendix. The main code problems

! they have observed are the interfacial drag package, discontinuities, the

accumulator model and modeling of horizontal steam generators, and loop
seal behavior. Dr. Holmstrom commented that the LBLOCA matrix presented

! by Dr. Odar should include separate effects tests. Also suggested was the <

l inclusion of REWET II data. The suggestion was made that more emphasis on j qualitative assessment may simplify the quantitative procedures. 1

! Mr. Winkler of F.R. Germany summarized their plans for assessment of the

! RELAP5/M002, TRAC-PF1, and TRAC-801 codes. Plans to assess the TRAC-801 1

{ code could be accelerated to benefit the current NRC needs. BWR

assessments will utilize BWR plant start-up data. _ Germany will perform 15
assessment studies using plant transient data and 35 assessment studies using integral and separate effects test data.

l Or. Negrenti of Italy identified plans to perform 10 assessment studies for tests from the SPES facility and 5 assessment studies from the GEST-GEN facility. In addition, Dr. Negrenti orovided a review discussion  ;

of the SPES facility, GEST-GEN facility, GEST-SEP facility, and the ARAMIS +

facility. A tenable schedule for the delivery of assessment results was l i presented.

l G-18 1

1

Dr. Kim of Korea summarized plans to assess both RELAP5/M002 and TRAC-PFiriOO1. Their planned assessment activities include using plant transient data, plant start-up data, and data from scaled facilities.

Tests from scaled facilities are negotiable with the NRC since Korea has no domestic test facilities.

Dr. Speelman of the Netherlands sumarized plans to assess the RELAP5/M002 code using full-scale plant transient and start-up test data. They do not feel that RELAP5/M002 is appropriate for the analysis of their BWR facility.

Dr. Puga of Spain presented the work currently in progress and also future work plans. The program in Spain is at a very early stage. The assessment activities will include both RELAP5/M002 and TRAC-PF1/M001.

The TRAC-PF1 activities are being performed in cooperation with AEE-Winfrith (United Kingdom).

, Dr. Sandervag of Sweden presented assessment plans for the RELAPS/M002 and TRAC-PF1/ MODI codes. The current status of the Swedish assessment effort is that seven of the ten committed studies are complete. Assessment plans for the TRAC-PF1/ MODI code center around plant transient data.

Dr. Aksan representing Switzerland sumarized plans to assess the RELAP5/M002 and TRAC-801/M001 codes. The Swiss plans call for assessment i studies using both separate effects data, and integral scaled test data as

) well as two start-up tests for a BWR facility.

1 Or. Liao of Taiwan indicated that their assessment effort was restricted i to the RELAP5/M002 code and would utilize plant start-up' data and LOFT

! experimental data. The plant data includes both BWR and PWR configurations. Dr. Liao then presented the results of a study performed I

using the RELAP5/M001 code as a demonstration of their code experience.

l Mr. Brittain of the United Kingdom noted that their assessment effort will

)

! focus on TRAC-PF1/ MOD 1 and RELAPS/M002. The assessment matrices. ,

presented in the appendix, will address LBLOCA. SBLOCA, and separate I' t effects. Mr. Brittain also comented upon the mechanism for code change implementation. The point was raised that ICAP review of changes is appropriate prior to the release of a new code version.

Dr. Reocreux of France reviewed their plans to assess TRAC-PF1/M001 and indicated that a similar effort will likely be undertaken for 1

RELAP5/M002. The majority of their assessment matrix is separate effects

, oriented. Brief facility description and associated matrices are shown in the presentation which appears in the appendix.

4 Or. Murao represented JAERI (Japan) who currently does not have an ICAP l j agreement with the NRC. Dr. Murao made a presentation of JAERI activities in the area of best estimate code applications.

The third day of the meeting was called to order at 8:30 A.M. by Mr. Ting l 1 (USNRC) who then proceeded with a management presentation. It was pointed out that the USNRC codes are intended for use to study reactor safety i issues only. They are not intended for comercial purposes. The NRC does

, not recomend that the codes be used for plant design purposes. Mr. Ting l

, G-19 i

4

. .- -- . - - = , . - . - - - . -. . .- - -

i 4

l also outlined the desirable format for transmittal ' data tapes. These appear.in the presentation in the appendix. In conclusion, the short term publications to be generated by tne USNRC within the ICAP program were i 'aiscussed.

! Mr. Ting then made a short presentation relative to the specialist meeting

' to be held in' January 1987 in Oorset England. The first and second days i of the meeting are scheduled for technical papers that address code-i assessment, code use experience, and code implications of new experimental

! data. The third day has been scheduled for presentations by the i supporting U.S. laboratories on code uncertainty, code error corrections, j code improvements, and the regulatory significance of findings of ICAP assessment efforts. Mr. Ting announced a call for papers for the January specialist meeting and indicated that the title, author, and abstract i should be sent to either Mr. Ting or Mr. Ian Brittian by November 28, 1986. In conclusion, Mr. Ting announced the desire to move the annual specialist meeting to the March / April time frame at a location outside the United States. The program management meeting will continue to be held in conjunction with the. Water Reactor Safety Information Meeting.

! Mr. Gary Johnson of the INEL presented a orief status of the RELAP5/M002 j'

and TRAC-BWR codes. Statistics were given which summarized the number of errors that resulted from ICAP studies and also the quantity of proposed

! code improvements. Each code error correction was reviewed and the status l was given.

i

! Or. Jay Spore of LANL then presented a status discussion for the i TRAC-PF1/M001 code. Dr. Spore reviewed specific code error corrections

! that were of general interest to the ICAP participants.

! Mr. Robert Hanson of the INEL briefly presented a small break LOCA matrix i similar to the LBLOCA matrix presented by Dr. Odar earlier in the

! meeting. The point was made that it simply represented a minimum set of f

tests. However, additions or replacement of tests was' stressed as being i appropriate. The matrix for SBLOCA was derived under the same set of

! guidelines as was the matrix for LBLOCA. The matrix is presented in the

! appendix.

1 i Mr. Rick Jenks of LANL presented a case review in which a study submitted 1 to the NRC for review was evaluated. A summary of the review is presented

{ in the appendix. A point was made that for a study to fulfill part of the i ICAP agreement, the study must be performed using the frozen code j version. The ICAP management document should be consulted to determine j what constitutes an appropriate study.

The final item on the agenda was a panel discussion consisting of a representative from each of the ICAP agreement countries. The format of

the panel discussion was primarily for each member to comment'on the l

overall meeting or any portion thereof.

Dr. Odar (USNRC) opened the panel discussion by inviting connents on the NRC presentations at the meeting.

Dr. Stubbe of Belgium noted that the S8LOCA matrix should include plant transients for steam generator tube rupture. He also noted that it is I

G-20 l

necessary to have similar matrices for operational transients. (Dr. Odar noted that those matrices may be developed in the coming year) Dr. Stubbe ,

noted a lack of interest in including plant data assessments.

Dr. Liao of Taiwan stated that the purpose of the L8LOCA and SBLOCA matrices was not clear relative to who will perform the analyses and suggested that the number of tests be reduced. (Dr. Odar reiterated that the number of tests is decided in part due to requirements of the code uncertainty quantification effort.)

Dr. Reocreux of France noted that the CSNI is also addressing code applicability and uncertainty by a different approach and would like to see documentation on the NRC methods by the January specialist meeting.

In the same light he requested the reaction of the NRC to the CSNI '

matrices. Dr. Reccreux also stated that it was not clear how the ICAP participants were going to contribute to the work outlined in the L8LOCA i and S8LOCA matrices.

Mr. Winkler of F.R. Germany would like to see documentation sumarizing

the objectives of each of the members of ICAP. Mr. Winkler suggested that i

the ICAP members should directly contribute to the code development j activities. He also suggested a need to document an integrated assessment j matrix. In conclusion, he raised the concern of how restricted ,

l experimental data can be best used by all ICAP participants.

t

Dr. Marsili of Italy supports assessment of the same tests with different 1

codes. He expressed the need for the code uncertainty quantification

! method documentation as soon as possible. Dr. Marsili noted the need for f an operational transient matrix, the need for inclusion of plant data in i assessment matrices, and the need for using counterpart tests where available.

Or. Murao of Japan suggested that all the codes should use the same models

) and correlations thereby making assessment statements universal. He

, comented that rather than using i highly developed 8.E. code for i licensing analysis, it may be more appropriate to benchmark a simplified

code for licensing use. Dr. Murao thinks gap conductance is not treated sufficiently relative to licensing. He also suggested that proposed model imorovements should be inserted ed exercised by the users to gain insight i into overall benefit for each competing improvement. Finally the question was raised as to how evaluation of input models and different computers i can be made.

! Or. Kim of Korea reiterated that an operational transient assessment i

matrix needs to be constructed. Part of this effort is to identify

! counterpart tests.

I Or. Puga of Spain indicated a concern relative to restricted experimental data.

Dr. Sandervag of Sweden emphasized the importance of achieving code j uncertainty quantification relative to applying 8.E. codes to licensing i applications. He speculated that the CSNI and NRC efforts are converging. It was noted that separate effects data is needed in the assessment matrices. Dr. Sandervag feels ICAP has been good for technical G-21 i

- _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ - , . _ _ _ _ _ _ _ - . . ~ . _ , . , _ _ , . _ _ _ , _ _ _ , _ _ . _ _ , _ _ _ , _ _ _ _ . . . _ . _ . , _ _ _ _ , . _ , _ -.

i I communication.

Dr. Akson of Switzerland feels the interface with code developers is good. He stated'the need for critical evaluation of code fixes and improvements to ensure they are general in nature. He stated that the j frozen code version needs to be clearly defined. Overlap between the

existing matrices under agreement and the NRC LBLOCA and SBLOCA matrices j needs to be determined. BWR matrices are also needed.

I Mr. Fell-of the Uni,ted Kingdom feels the code revision mechanism is

effective. However, is concerned with the possibility that identified l improvements or fixes may not be implemented in the identified manner.

1 Mr. Fell also feels a need to document the assessment matrices. Open summary documents are required even-for proortetary facilities and data to make the documents useful. Mr. Fell is not convinced that the proposed assessment matrices can be accomplished. However, he feels the NRC is getting a handle on how to quantify code uncertainty, but noted the need for documentation of the method.

Dr. Stadtke of ISPRA commented that it appeared there was not a correct

, balance between code assessment and code improvement efforts. His

impression was that code improvement activities should be increased 1 somewhat. He also questioned how well the member proposed assessment j- matrices matched the proposed matrices for L8LOCA and S8LOCA.

Mr. Fell concluded the panel discussion by welcoming everyone to Winfrith next January.

I i

)

i l

l

l i

l 4

G-22 r-----g-- y- .,qm= wm# %

e9-9 --yr-c-r'r-r+n- T-+--w~+ 1---g5-yg+ Ww---+me,-M-u-pwTt'TTw e - ees wiM-wM e-E- g* m-EMF--yy-Fwc-- p =-s-** M-:---*91~-++N- -

-+m@-"- T ----+'rT'y

[ WT

^

ICAP SPECIALIST MEETING MINUTES NEWS R. Jenks & T. Knight February 1987 Los Alamos National Laboratory The 2nd Specialist Meeting of the in-ternational Thermal-Hydraulic Code As- . Management. Code Development & Error Correc-sessment and Applications Program tions INTRODUCTION Following the welcoming remarks. Mr. Philip Ting. USNRC Program Manager for the ICAP pro-The second specialist meeting of the Interna-gram, officially opened the meeting. He indicated that j tional Code Assessment and Applications Program the published and approved minutes of the Second (ICAP) was hosted by the United Kingdom Atomic Management Group Meeting held November 3-5.1986 Energy Establishment (UK AEE) January 19-21,1987 in Bethesda. Maryland would be distributed during the in Winfrith. Dorset. England.

present meeting. He asked that peop!c in the audi-Three days of technical and informational mect-ence review a mailing list that would be distributed in ings at the Winfrith AEE site brought together over the meeting room and make any changes or additions fifty-three international and domestic thermal-hydrau' needed.

j lic code specialists representing thirty organizations from twelve European and Asian countries and the Mr. l. Vojtek from GRS made the first pre-sentation summarizing work perfctmed when he was l j United States. The meeting was divided into two cat-FRG Resident Engineer at LANL during 1983-1984 as egories. The first and third days were devoted to tech' mcal presentations describing code-assessment activ- well as more recent work. His assessment of TRAC-it_ies carried out by ICAP members. The second day PF1/ MOD 1 heat-transfer correlations led to his con-clusion that TRAC overpredicts maximum critical heat was dedicated to technical presentations by the US-flux and total high pressure wall-to-fluid heat flux. but l

NRC and by USNRC. contractor laboratories.

underpredicts low-pressure wall-to fluid heat flux. In l addition, he found that during a LOCA blowdown the

{ FIRST DAY (Monday, January 19) minimum stable film boiling temperatures are under-l The meeting was opened with welcoming remarks predicted. His work also showed that a Forslund cor-l by Dr. D. Pooley and Mr. Jesse Fell, both of ' ' " C nstant of M gan kst resuRs for waU tend UKAEE. Winfrith. Dr. Pooley said that recent ex- peratures .in the d.ispersed-flow heat transfer regime.

perience with Chcrnobyl icinforces our reactor-safety His recent work had been done using error corrections objectives to " keep it cool and keep it contained? "M Mr. Fell extended a general welcome to all ICAP Dr. M. Analytis from ElR commented that some-member countries. and specifically acknowledged the n needs to tell analysts what interfacial heat-latest country to express interest in joining the ICAP. Ransfer area to use Dt Mngton from AEE Win-Yugoslavia He presented an overhead transparency frith stated that UK results for low-pressure heat that showcd the location of the other cities in the UK transfer were different than those shown by Mr. Vo-I where code-assessment activities were ongoing. An- jtek. Dr. Knight from LANL reiterated that you should 4

other overhead depicted the 5 typical areas that papers n t change a code based on just one set of assess-presented at this meeting fellinto. ment results, but rather based on the whole body of i

. Separate Effects Tests assessment results. Dr. Knight indicated that a post-l . Integral Tests CHF working group is set up to meet this summer to e Plant Studies e Uncertainty Quantification / Test Matrices l

1 G-23

address just this sort of problem and will need to ac- MOD 2 (cycle 36.4). Comparison between pre- and cumulate all results of CHF analyses from UK. GRS post-test results showed that MOD 2 did not crash af-and any others available. ter accumulator injection (as occurred with MOD 1).

Dr. P. Bratby from NNC discussed NNC's plans broken loop seal clearance was predicted, and the cor-for use of TRAC-PF1/ MOD 1 to possibly do licensing rect degree of vessel level depression was calculated.

calculations for the Sizewell 'B' PWR. He presented He noted that the timing of these events was off. In the results of a TRAC analysis (with EC12.3) of the order to match break flow, a 0.85 multiplier was used Marviken Critical Flow tests. It was found that the indicating validity of the contention by Dr. W. Bryce break flow for a stand-alone discharge-nozzle model (AEE Winfrith) that errors existed in the critical flow was very sensitive to noding and did not agree with model causing excessive sub-cooled critical flow. It the break flow calculated using the full model.-In fact, was also shown that draining of the hot legs was too the break flow was underpredicted for a case where rapid once riser level fell below nozzle elevations. It the entire apparatus was modeled, just the opposite was indicated that the excessive draining in the hot of the results for a case where the nozzle alone was legs could be due to the lack of a CCFL modelin the modeled. Furthermore,it was found that a case where code.

2 cells were used gave better results than either 1. 4 Dr. T. Sirkla from VTT delivered the results of or 8 cells. a RELAPS/ MOD 2 plant transient analysis of a Soviet Dr. P. Coddington described LBLOCA simula. VVER-440 type PWR. This plant was somewhat un-tions for a Westinghouse four loop PWR performed usual in that it had a horizontal steam generator. The with TRAC-PF1/ MOD 1 (version 11.0) in various specific transient resulted from a stuck-open turbine stages of modification. A vessel noding study was bypass valve occurring after a reactor-coolant pump f performed by employing first a fine-node model (17- trip and a reactor scram. While the overall transient

z. 4-r 8-9) and then a coarse-node model (11-z,3-r. system response was characterized as excellent, once 4-6). In addition, two parametric calculations were high pressure injection was correct, he did specify that performed for the coarse-node model: additional investigation of the pressurizer response
1. with the code modified to remove the interface was warranted.

sharpener, revise the nozzle loss and correct a Separate work investigating code performance heat-transfer error; and against results from a loop seal behavior test was also

2. with the code modified to reformulate the presented. The code showed more entrainment than 3-D momentum equation along the z axis by exhibited by the experiment. In addition to the plant reweighting the z-direction spatial acceleration calculation and the separate effects test. he presented terms. results of a LOFT LP-LB-1 calculation showing prob-An additional calculation was performed using lems with intact loop hot-leg mass flow prediction and the coarse-node model, but with no accumulator ni_ incorrect prediction of accumulator mass-flow behav-trogen in the primary circuit. ior. Aside from code problems he experienced (related His results indicated sensitivities to noding with to interfacial drag, discontinuities, accumulator mod-the total slab heat flux for fine-vessel noding very cling, and oscillations) he had difhculties modeling much larger than for coarse-vessel noding. the horizontal steam generators.

Dr. L. Rebollo from Union Electrica Fenosa gave Dr. P. Stoop from Netherlands Energy Research the results of a RELAPS/ MOD 2 (cycle 36) plant sim. Foundation presented a RELAP5/ MOD 2 (cycle 36.04) ulation. The objective of the analysis was to use feed-and bleed transient analysis for the Borssele nu-

' lessons learned" from an analysis of the LOFT LP. clear power plant. This study of the system response SB-3 experiment and apply them to a similar transient to loss of all secondary-side feedwater for a two-at the Josd Cabrera nuclear power plant. Their aim loop KWU PWR demonstrated that feed and-bleed was to improve operator guidelines for a small-break will work effectively if started within 10 minutes of LOCA scenario. Based on the results of the simu. the initiating event. He stated that the code provided lation, specific operator actions were recommended a valuable tool for the evaluation of operator guide-depending on the state of accumulator discharge fol. lines.

lowing accident initiation. Mr. F. Winkler from K'NU expressed concern that Dr. A. Scriven from CEGB/CERL presented pre. the operator may not be able to recognite the tran-and post test results for RELAPS analyses of LOBI sient within 10 minutes, and therefore know to take test BL 02. The pre test simulation was performed with MODI and the post test analysis was done with 2

G-24

the recommended actions. He suggested that the po- Dr. F. Odar from USNRC identified progress tential for a TMl-type accident might exist if the oper- that had been made recently with regard to code-ator misinterprets available information. Mr. Winkler uncertainty quantification and scaling issues, inform-also stated that upper-plenum measurements might ing ICAP members of recent meetings held to iden-be required to allow the operator to act correctly. tify and rank the most important phenomena for Mr. G. Johnsen from INEL suggested that control sig- a cold-leg LBLOCA scenario. The meeting results nals identifying steam / feed mismatch might be avail- were presented and projections for follow-up work l

able for that type of plant, and would provide early were given. A preliminary importance-rating table warning of a potential problem. Dr. J. Sursock from for clad temperatures was presented. The proposed EPRI asked if any credit was given for lack of letdown- LBLOCA separate-effects test matrix for PWRs was system modeling. and Dr. Stoop indicated that none also shown. This matrix identified a proposed facil-had. ity and country to be responsible for each specific l

Dr. H. Stadtke from JRC ispra showed re- LBLOCA-phenomenon assessment. He stressed that suits from both the frozen RELAPS/ MOD 2 code and this test matrix addresses only one particular tran-a modified code (RELAPS/ MOD 1-EUR) for calcula- sient (LBLOCA). Dr. Odar further stated that the un-tions of the LOBI test A2-90. The modified code certainty methodology was still under discussion and, version showed better predictive capability than the hopefully, will be presented at the next meeting.

frozen code version for this specific test. Dr. Stsdtke Mr. Winkler stated that UPTF had recently run was asked if his improvements would also help MOD 2. a test which showed ECCS bypass effects. and he and he indicated that because of the differences be- expressed surprise that integral tests were not in-I tween MOD 1 and MOD 2. improvements over MODI- cluded in the proposed matrix. He said that KWU EUR would not be dramatic. However further dis- had performed two-phase pump studies that showed j cussion suggested that MOD 2 might benefit from the work was needed to assess pumps, as well. Dr. Odar smoothing done in MOD 1-EUR to interpolations at responded that ICAP could certainly include PKL in flow-regime transitions. the test matrix, and he would appreciate more sugges-Mr. H. Plank from KWU presented a TRAC- tions along these lines. He acknowledged that pump l PF1/ MODI (with EC12.5) analysis of a double-ended modeling was important to the outcome of reflood.

cold-leg break at a KWU PWR. The main objective Mr. Winkler indicated that we should also be L of the calculation was to obtain initial and boundary concerned about combined injection. He noted that conditions for a similar transient in the UPTF facility. the proposed separate-effects matrix does not repre-This relatively-large TRAC model included a 3-D (15- sent very well the alternative ECCS-injection schemes

z. 4-r. 84) VESSEL. a fine noded steam generator used in countries other than the US. Dr. Odar ac-

) and 121 1-D components for a total of 1168 com- knowledged that the matrix had been groomed for US 3 putational cells. The calculation required 75 hours8.680556e-4 days <br />0.0208 hours <br />1.240079e-4 weeks <br />2.85375e-5 months <br /> of plants. but that NRC was open for suggestions to ex-Cray-1 computer time and 5.84 M Bytes of core mem- pand the matrix.

ory. Dr. Mura'o from JAERI asked what code the work Dr. C. Richards from AEE Winfrith presented a would be done with. Dr. Odar responded that work pre-test calculation of LOBI test BL 02 using TRAC. would be performed with the frozen version of the PF1/ MOD 1 (with EC12.2). A major problem in the code.

calculation was failure of code to predict loop-seal Mr. Fell asked how we intended to use the clearing, possibly due to problems with predict.on of separate-effects matrix inside ICAP. He said that the interfacial drag. He also indicated that the accumu- separate-effects matrix should be used for model de-lator flow rate was over predicted and suggested that velopment. Dr. Odar stated that we do need separate-

this might be due to overprediction of condensation effects assessment results for the next version of the rate. He indicated that except for the loop-scal prob- code but that we can also use results for uncertainty fem, all differences between the experiment and the analysis by putting " dials
  • in correlations and examin-calculation could be explained. but more calculations ing the sensitivity of separate-effects results to these were needed to confirm their explanations. " dials." However. he stressed that " dials" must rep-resent physics. He emphasized that integral tests will be the key to the uncertainty-quantification process.

SECOND DAY (Tuesday. January 20) and that separate-effects tests would primarily be for looking at potential model improvements. Mr. Fell re-sponded that the proposed separate-effects analyses 3

l G-25

represent substantial work, and if done, may extend understood NRC will budget money for these improve-the time until assessed codes are available for uncer- ments. Dr. Odar deferred his response to his third-day tainty analysis. Dr. Odar replied thr.t. in the interim. presentation which would address these issues.

the frozen code could be used. Dr. Odar was asked which models would be ex-Mr. Vojtek expressed concern that assessment amined in each assessment. He stated that spet.ific results will end up at NRC with nothing done, be- models were yet to be determined, but the selection cause they could not use conflicting data (some of of the statistical population was very important and which had been presented at the current meeting). phenomena similar to each assessment will be exam-Dr. Odar indicated that the post-CHF Task Group ined. He was also asked how NRC intends to isolate would meet and recommend improvements to go into specific phenomena to identify capability of current TRAC-PF1/ MOD 3. correlations. He responded that NRC will make lists Mr. Winkler commented that cakulations are of test. phenomena, correlations, etc., that will serve costly. They perform work. and then changing the this purpose.

code causes additional expense. He said that we Dr. R. Pochard from CEA asked how we should should look at code deficiencies first. and then im- account for user effect. Dr. Odar replied that noding prove the code before spending all the money that is sensitivity was recognized as causing trouble, and he available for this program. Dr. Odar responded that had seen noding differences result in a 100K difference the Models and Methods document should be out by in calculated PCT. He said that this problem would be the middle of the year. and that deficiencies identified addressed, stating that we may need to add a bias to in that document should be resolved first. Dr. Odar account for user effect. He emphasized that the bet-also noted that most errors from ICAP are constantly ter the user guidelines were, the less user uncertainty being corrected. but code improvements require plan- would exist.

ning. A question was asked as to whether or not ICAP Mr. Winkler said that Dr. St3dtke's presentation believed the codes could be considered fundamentally had shown that problems needed to be resolved in RE- and theoretically sound. Dr. Odar stated that the LAPS / MOD 2 also. Dr. Odar replied that his presen- Models and Methods document will show basis for tation on the third day would address improvements the methods used.

to RELAP. Mr. G. Johnsen talked about RELAPS/ MOD 2 Mr. Vojtek stated that in the past, at the Salt and TRAC BWR status and plans. Statistics were Lake City heat-transfer meeting presentations indi- given which summarized the number of errors that cated that four new'models should be implemented resulted from ICAP studies. Each code error correc-into the code. However, when the new models were tion was reviewed, and the status was given. Re.

put into the code. they did not give any improvement. cent user-conveniences added to the codes were dis-Dr. Odar responded that the post-CHF Task Group cussed. The RELAP Ouality Assurance Report (due would address this issue. out in June 1987) was also discussed and principal Dr. Murao suggested that a review period was deficiencies in RELAPS/ MOD 2 were identified. He needed for error corrections prior to putting them in asked for feedback from ICAP members to help iden-the code. Dr. Odar agreed, but for code model im- tify where major code deficiencies exist. Objectives provements, only. Dr. Murao proposed that code im- and accomplishments relating to MOD 3 were also provements should be the result of a consensus of presented. Mr. Johnsen briefly outlined numerous the ICAP members. Dr. Odar stressed that USNRC changes made to TRAC-BWR, specifically implemen-needs to know desired code improvements, now, in tation of ElR recommend Bestion correlation imple-order to get them in the code by 1988. He also said mentation of revised annular-flow model, and many that the code newsletter should give ICAP members programming, error correction and user-convenience-information regarding proposed code improvements, related changes made to the code. Development plans Dr. Murao asked if assessment cases would be run for TRAC BWR were also given. He said TRAC BF1 on the selected code model improvements. Dr. Odar would be released January 30,1981.

replied that some developmental assessment would be Dr. O. Sandervag from Studsvik suggested that performed before the code was released. another correlation (Becker) might be more appropri-Mr. Wink!cr asked if interfacial shear. various ate than Biasi for rod bundle dryout. Mr. Johnsen heat transfer problems and other problems with RE- said INEL would investigate.

LAPS / MOD 2 were being addressed. He said that he Regarding problem number 87 04 (in overheads) that addressed critical flow problems with Marviken 4

G-26

tests. Mr. Johnsen was asked if it was possible that Mr. Wilson asked if Sandia's TRAC-PF1/ MOD 1 TRAC-PF1/ MOD 1 would have the same difficulty. He guidelines were also applicable to RELAP. Dr. Buxton indicated that Dr. W. Bryce would provide more in- responded that in some cases. yes. However, not for formation on this to both INEL and LANL. Further flow-area change models, where the differencing equa-discussion on this issue brought forth the comment tions are sufhciently different between the two codes.

that critical-flow results are not attributable to code Dr. Coddington stated that an appendix to a

errors as much as to code models, but results are not recently-released UK LOFT LB-1 assessment report bad enough to justify reformulating the model now. tells why one gets the azimuthal noding dependency It was stated that,in the interim. user guides should exhibited by TRAC-PF1/ MOD 1. He said that this de-include information regarding modeling where critical pendency relates to the differencing scheme. Mr. Brit-flow would be a concern (such as a guide telling users tain from AEE Winfrith was curious as to why SNL to apply discharge coefficients to resolve flow difficul- had obtained different results for the LOFT L2 5 cal-ties). culation. What was different about SNL's model com-Mr. G. Wilson from INEL presented a flowchart pared to models at UK and LANL7 Dr. Buxton indi-depicting the review process INEL would follow to cated. that, for one thing. SNL had used bypass mod-evaluate ICAP assessment reports. Because of time cling. There was also speculation that the ECCS flow constraints. he referred ICAP members to a more- problem (which was indicated by a flow reversalin the detailed paper he provided for their review. He also SNL calculation that was not exhibited in either the indicated that LANL would follow a similar procedure UK or the LANL calculations) was enhanced by the to evaluate TRAC PF1/ MOD 1 assessment reports. pumps running in this particular test. Mr. Brittain Mr. G. Wilson also discussed the results of a expressed interest in what was being done to resolve break-spectrum analysis for a SBLOCA in a RESAR- the problems with differences in the L2 5 calculations.

35 plant. The study was performed to investigate Dr. Murao asked what specific noding-sensitivity steam generator liquid holdup in full-scale systems. studies had been done with L2 5. Dr. Buxton indi-He stated that this work highlighted information that cated that they had looked at 1D versus 3D as well might be important when one models a full-scale as azimuthal variations within the 3D VESSEL model.

plant transient. Both RELAPS/ MOD 2 and TRAC- Mr. R. Jenks from LANL presented recent code PF1/ MODI results were shown and indicated major and documentation changes that had resulted from differences in code-calculated behaviors in core heat- l CAP feedback. He discussed nine specific concerns up and core liquid level predictions. ranging from resolution of UK's concern about a pos-Dr. St3dtke made the point that variations in sible error in the STEST updates for EC12.5 to SNL's flooding behavior can cause large variations in core concern regarding azimuthal noding sensitivity in the level depression. VESSEL component. The source. a "TFORM" iden-Mr. Winkler commented that KWU plant simu- tifier, a description, and the status of resolution were lations have shown similar steam-generator hold up given for each concern. He stated that LANL's objec-behavior. However, a recent UPTF test showed that tive with this presentation was to improve response steam flows are much too low to support CCFL in time for helping code users.

hot-leg / steam generator inlet plenum. Dr. Richards stated in regard to recent stratified-Dr. L. Buxton from SNL talked about a s- flow model changes that the critical gas velocity has sessment and modeling experience using the TRAC- a term that causes V-crit to go to zero at vertical PF1/ MODI code. He gave several suggestions and orientation.

guidelines to TRAC users based on Sandia's model- Dr. T. Knight identified and discussed a subset ing experience. An assessment matrix was shown that of user identihed problems, called "proactive." and in-not only displayed the analyses completed by SNL. vited ICAP members interested in directly addressing but also the analyses underway and those envisioned code problems to come to Los Alamos. Twenty-two for the future. Dr. Buxton briefly discussed the spe- specihc user supplied code corrections, code improve-cific work undertaken for each of the completed anal- ments. user conveniences and code guidelines were l yses. He summarized by saying that much feedback presented. This feedback contained not only a state-had been given to the code developers, that there was ment of the problem, but also a statement of the per- )

a growing need for generalized heat-slab and improved ceived solution to the problem, hence the term "proac-steam separator models, and that noding sensitivities tive." The source. a "TFORM' identifier, a descrip-need to be further studied. tion, and the status of resolution were given for each ,

I problem. When the topic of EXTRACT was brought 5

l G-27 l

\

up, Dr. Knight discussed the value of this TRAC util- e some members of the group had reservations ity. After explaining that a version-independent EX- about the current status of the proposed PMG TRACT would be available soon. he said that EX- meeting in October 1987, and they wanted a pro-TRACT was useful to both ICAP and L ANL. It saved posed agenda and objectives for the meeting be-CPU time when parametrics were required. or when fore concurrence.

input-model errors needed to be fixed. In addition. by Mr. Ting also presented the proposed charter using EXTRACT to provide LANL with an input deck for the post-CHF Task Group to take effect March near the time of a code problem. significant time and 1. 1987. The majcr objectives of the post-CHF money could be saved- Task Group are to to review and recommend specific Regarding the issue of ICAP participation at Los post-CHF interfacial heat-transfer models to be im-Alamos. Dr. Knight stated that this requires coordi- plemented in USNRC best estimate codes.

nation with and approval of USNRC. He also indicated Dr. F. Odar presented USNRC plans regarding that advance notice was required to arrange the nec- RELAPS/ MOD 2 improvements. He stated that US-essary approvals at Los Alamos. NRC uses this code as an audit tool. and for this Dr. Knight explained that larger error-correction purpose,it has sufficient capabilities to calculate the sets would be released in the future in response to required transients. He indicated that further improve-user concerns for more testing before code-release, ments to RELAP will be based on the magnitude of He stated that once LANL's expanded test problem the uncertainty determined under ICAP and the avail-matrix is completed, and the code is tested with that ability of resources. He stated that it is known that matrix, a much larger set of error corrections would some models need to be added or improved, and ICAP be released (in the March 1987 time frame). assessment activities may indicate further deficien-Dr. Buxton asked if we had formalized a proce- cies. Dr. Odar said further, that USNRC welcomes dure to put "TFORM" numbers in comments to up- closer collaboration with the ICAP members to de-date idents. Mr. Jenks replied that LANL would do velop and improve RELAPS/ MOD 2.

it. Mr. Fell asked why there were separate ap-

"Roundtable" discussion: Mr. P. Hall from proaches for each code, and why USNRC did not put CEGB/GDCD asked Dr. Stsdtke if he thought TRAC-PF1/ MOD 1 heat transfer improvements into RELAPS/ MOD 2 should be updated with the same Dr. Odar agreed that codes are RELAPS/ MOD 2.

changes made to RELAPS/ MOD 1 to create the EUR very similar. and hence, there should be little dupli-version. He replied that more study of code changes cation of effort. Dr. Analytis rerr.arked that the logic was needed before that effort would be carried out. in the codes is quite different. As an example, he it was indicated that changes made to MOD 1 to cre- mentioned that there was no Tmin in RELAP. indi-ate EUR version would be considered for implemen- cating that some specific packages are not the same tation in MOD 3. Dr. St3dtke said that a report was in both codes. Mr. Johnsen emphasized that while being prepared that describes RELAPS/ MOD 1 EUR the methods are quite similar, the logic that imple-methodologies. He also gave an example of how os- ments the methods is different. He also stated that cillations were smoothed out, stating that they use an INEL would adopt the recommendations of the post-analytic method to calculate thermodynamic proper- CHF Task Group.

ties and have written new subroutines. Dr. Knight Mr. Winkler suggested that the USNRC add asked that LANL be provided with the report that interfacial. package improvements" to the list of RE-documents their methods. LAP improvements.

Dr. Y. Murao presented the code-assessment work underway at JAERI He stated that JAERI in-THIRD DAY (Wednesday. January 21) tends to assess specihc correlation models in TRAC.

Mr. P. Ting opened the third day with a group P. TRAC B and RELAP by putting them into the J-discussion on future ICAP meetings. The results of TRAC code. He also discussed plans for model as-the discussion were: sessment using a mini two-fluid code developed at a e The group rejected a suggested specialist meet- Japanese university and designed to run on a personal ing in October 1987; computer. An assessment of TRAC-PF1/ MOD 1

. the group agreed that the next specialist meetinK against LOFT L2 5 data was also presented. He rec-should be in the Spring of 1988 with the specific ommended that from the standpoint of accuracy, the site to be determined; and TRAC PF1/ MOD 13D VESSEL should be used how-ever from the standpoint of computational efficiency.

6 G-28

he recommended that the 1D CORE model be im- enhancing core quenching. There were some oscilla-proved and used. He further recommended that more tions that were predicted in core pressure as a result assessment be performed for CCFL at the downcomer, of heat transfer that were deemed to be nonphysi-condensation in the cold leg, and multidimensional cal. Dr. A. Forge from CEA questioned the pressure flow behavior in both the downcomer and the core increase with nitrogen-injection. and Dr. Coddington regions. responded that this was due to the nitrogen expanding Mr. Johnsen asked Dr. Murao to review what after all the liquid was injected.

JAERIintends to do regarding assessment of RELAP Dr. K. Ardron from CEGB/GDCD gave an models in the TRAC P code. Dr. Murao replied that assessment of the interphase drag correlations in l the RELAP correlation set would be made compati- both RELAPS/ MOD 2 and TRAC-PF1/ MOD 1. show-

ble with and then built into the TRAC-P (of JTRAC) ing plots that compared void as a function of flow code. Mr. Johnsen was concerned about JAERl's pro- rate for several pipe sizes. to the Wilson-Rooney cor-posed method for comparing correlations, because re- relation. Among his findings. the difference between suits are also dependent on numerics and logic in each code-predicted void and that given by Wilson-Rooney code, and the proposed approach might give a good increased with decreasing liquid flow. increasing steam assessment of the correlation, but not of the sepa- flow and decreasing pressure. He stated that further rate codes. Mr. Johnsen further stated that using the work needed to be done to test the effects of numerical codes required decisions about noding and other fac- implementation of individual code models.

tors specific to a given code, and this will affect the The comment was made that downflow-void pre-results, too. diction was most important for high void conditions.

Regarding JAERI's L2-5 work. Dr. Coddington and it was believed that TRAC's interpolating scheme was concerned about methods of determining dis- contributed to that prediction.

charge coefhcients. He suggested that adjusting Dr. J. Izquierdo from CSN Spain asked for clari-discharge coefhcients produces meaningless results. fication of what was meant by " analytic calculation."

Dr. Murao replied that their current efforts are aimed Dr. Ardron replied that the calculation employed an-at trying to get information about which kind of mod- alytic equations rather than difference equations like els should be used to get predictions, that they want those used in TRAC or RELAP.

to perturb the boundary conditions to grasp an idea Mr. P. Hall presented a RELAPS/ MOD 2 analy-of the code sensitivities, and that in the future they sis of THETIS boildown experiments, comparing ex-will calculate without making such adjustments. perimental results for void (calculated from AP mea-Dr. Odar asked the ICAP participants if,indeed, surements) with calculated values of void. Using 24 we should be using the 3D VESSEL component in volumes for the heated section. excellent agreement TRAC-PF1/ MOD 1 as suggested by Dr. Murao's con- was obtained with experimental results at 4 and 2 clusions slide. Dr. Coddington stated that their work MPa. Using a 6-volume model did not cause serious with LOFT LB-2 indicated that. clearly. 3D behavior deterioration in code performance. He said that they exists. Mr. Winkler suggested that the 3D VESSEL had to remove the vertical-stratification model from be used for more detail. and that a multi-channel 1D the code in order to climinate oscillations in void frac-model be used, otherwise. because it is more econom- tion during the steady state calculation. He expressed ical. a concern about the vertical stratification modelin the Dr. P. Coddington discussed the results of a code.

TRAC-PF1/ MODI analysis for a LOFT LP-FP-1 ac- Mr. Hall was asked if he intended to repeat the cumulator blowdown calculation. A special feature of analysis with TRAC PF1/ MOD 1. He said that he did this model was that they started with a larger value of not, because he would expect similar results as in gas temperature in the accumulator (600 K) to keep RELAPS/ MOD 2.

the gas from freeting later in the calculation during Mr. Wilson asked if the transient was also run depressuritation. with the vertical stratification model removed, and Dr. Bratby asked if noding-sensitivity studies had Mr. Hall indicated that the model was retained, be-been performed, and Dr. Coddington said that none cause there was no difference when it was on or off had been done yet. for the transient. Mr. Wilson stated that Mr. Hall's Dr. P. Coddington also presented a TRAC- results were confirmatory to work done by Dr. Ana-PF1/ MODI analysis of nitrogen injection tests in the lytis.

Achilles rig. The analysis showed that the presence Dr. Knight asked if they were concerned about of nitrogen-injection produces more liquid in the core, heat losses in the THETIS facility, and Mr. Hall said 1

G-29

that they considered them. A comment was made LOFT LP-FP-2. showing the code correctly predicted that essentially no heat loss occurs in the apparatus, the main milestones of the transient. Code limita-because a heated water jacket surrounds the rod bun- tions were found with regard to radiation heat transfer d!c. and metal-water reaction predictive capability. He said Dr. E. Allen from AEE Winfrith presented pre- that the calculation of their base case indicated the liminary results of a TRAC-PF1/ MOD 1 analysis of main source of uncertainty was the LPIS discharge.

LOFT LP-SB-1. The TRAC input model was de- Because no data was available on LPIS flow, they picted, the steady-state initial conditions were out- performed a sensitivity study to determine what dis-

! lined, and the transient event sequence was shown. charge coefficient gave the best results. Dr. Reventos Dr. Allen indicated that they had found some discrep- stated that they would like to review the RELAP in-ancies in the original deck received from Los Alamos terphase drag model and the two-phase critical flow and had corrected them before proceeding. Prelimi- model further.

nary observation was that the overall agreement was Mr. Ting gave a concluding statement express-1 reasonable, but some discrepancies were noted. She ing gratitude to Winfrith. Mr. Brittain, and the Win-1 indicated that the major discrepancy was with the pre- frith staff. He said papers would be assembled for diction of break mass flow rate, and attributed this to publication as a NUREG-IA, and summarized the pro-failure of the code to properly account for hot-leg flow jected future mecting arrangements by stating that stratihcation near the break. She also indicated that the next Program Management meeting would be in the maximum time step had to be reduced from .5s October 198f. and the next Specialist meeting would to .ls at 1500s in the transient in order to allow the be in the Spring of 1988.

calculation to continue. She said that an improvement Mr. Fell stated that AEE Winfrith enjoyad having j

i in the time step was seen when the bypass flow was the ICAP members at their site and suggested that remodeled using a larger cell volume. "we only make success, if we work together."

r Dr. Knight said that there was some uncertainty ,

i about which pressurizer volume was correct, because sometime in the LOFT program it was reported to be Acknowledgements smaller and then, later. changed back to its original We would like to express our gratitude to Sam

! value. He said that he would check this. Naff and Gary Wilson for their valuable contributions 1 Mr. Brittain indicated that if larger bypass vol- to the compilation of these meeting minutes.

j umes had been chosen to begin with, faster overall

run time would have been obtained.

C Pi es of Papers

Dr. F. Pelayo from CSN Spain delivered the pre.

I liminary results of a TRAC-PF1/ MODI calculation Additional copies of the presentation handouts I of the LOFT SB-2 test indicating reasonably good can be obtained directly from the authors or by spe-agreement with experimental results. He said that cific request from:

the treatment of flow regimes is insufficiently detailed. ICAP Program i and that the code overpredicted system mass Icss. Safety Code Development Dr. Knight asked if he was using default LOFT Los Alamos National Laboratory. MS K555 head degradation curves in code. and Dr Pelayo said Los Alamos. New Mexico 8f 545 they were not and had used data received from INEL 505-667 2021 or 843 2021 (FTS)

] based on L3-6 data, instead. Mr. S. Naff from US-

] NRC/EG&G indicated that this should not be done.

that the LOFT instrumentation was not designed I to measure pump characteristics under very low AP degradation conditions. Dr. Knight also asked how big the hot-leg pipe was relative to the break orifice.

Dr. Pelayo said that the hot !cg pipe was 1.5 inches

'- i.d. relative to 1 inch I.d. for the break orifice.

Dr. Richards reported that his experience showed using the L3 6 head-degradation curves instead of what is in the code did not improve agreement much.

l Dr. F. Reventos from Asociacion Nuclear Asco presented results of a RELAP5/ MOD 2 analysis of i

8 G-30

2nd International Code Assessment and Applications

. Program Specialist Meeting AEE Winfrith, Dorset, England January 19-21, 1987 Participants Organisation Address and Tel No I Brittain AEE Winfrith 201/A32, AEE Winfrith Dorchester Dorset DT2 8DH UK (0305) 63111, Extn 2039 J Fell AEE Winfrith 166/A32, AEE Winfrith Dorchester Dorset DT2 8DH UK (0305) 63111, Extn 3451 P Coddington AEE Winfrith 270/A32, AEE Winfrith Dorchester i

Dorset DT2 SDH UK (0305) 63111, Extn 2345 C G Richards AEE Winfrith 260/A32, AEE Winfrith Dorchester Dorset DT2 8DH

UK (0305) 63111, Extn 3029 R O'Mahoney AEE Winfrith 263/A32, AEE Winfrith Dorchester Dorset DT2 8DH UK i

(0305) 63111, Extn 2375 A J Wickett AEE Winfrith 214/A32, AEE Winfrith Dorchester Dorset DT2 8DH l UK l l

(0305) 63111, Extn 2936 1

G-31

Mico E J A11Gn AEE Winfrith 265/A32, AEE Winfrith Dorchester Dorset DT2 8DH UK-(0305) 63111, Extn 2563 D Pooley AEE Winfrith 372/A32, .AEE Winfrith Dorchester Dorset DT2 8DH UK (0305) 63111, Extn 3468 J C Birchley AEE Winfrith 230/A32, AEE Winfrith Dorchester Dorset DT2 8DH UK I (0305) 63111, Extn 2742

{

i A P Neill AEE Winfrith AEE Winfrith Dorchester Dorset UK (0305) 63111, Extn 2230 P S Black AERE Harwell B392 Harwell Laboratory Didcot Oxon OX11 ORA UK 0235 24141 M W E Coney CEGB/CERL Kelvin Avenue Leatherhead l Surrey KT22 7SE l

UK (0372) 374488, Extn 2238 A H Scriven CEGB/CERL Kelvin Aveune Leatherhead Surrey KT22 7SE UK (0372) 374488, Extn 2308 s

B Chojnowski CEGB/MEL Marchwood Engineering Lab Marchwood Southampton Hampshire UK SO4 4ZB (0703) 865711 G-32

~

K H Ardron CEGB/GDCD Barnwood. j Barnett Way l Gloucester I Gloucestershire UK <

GL4 7RS f I

(0452) 652198 J F M_ Evans CEGB/GDCD Barnwood

  • Barnett Way Gloucester Gloucestershire UK i

GL4 7RS t

(0452) 652958 C Harwood CEGB/GDCD Barnwood Barnett Way Gloucester Gloucestershire UK GL4 7RS (0452) 652268 P Hall CEGB/GDCD Barnwood Barnett Way Gloucester Gloucestershire UK GL4 7RS (0452) 652529 K T Routledge NNC Chelford Road Booths Hall Knutsford Cheshire UK WA16 80J (0565) 3800, Extn 3822 D B Newland NNC Chelford Road Booths Hall Knutsford Cheshire UK WA16 BOJ (0565) 3800, Extn 3816 N Hobson NNC Chelford Road Booths Hall Knutsford Cheshire UK WA16 80J (0565) 3800, Extn 3813 d

G-33

P A Bratby NNC Chelford Road Booths Hall Knuts ford Cheshire UK WA16 80J (0565) 3800, Extn 3812 C Potter Nuclear St Peters House Installations Balliol Road Inspectorate Bootle Merseyside UK 051 951 4392 P Ting USNRC Wilste Building 7915 Eastern Avenue Silver Springs Maryland 20910 USA 301 443 7920 F Odar USNRC Wilste Building 7915 Eastern Avenue Silver Springs Maryland 20910 USA 301 443 7902 R P Jenks LANL MS K555 Los Alamos NM 87545 USA 505 667 2021 T Knight LANL MS K553 Los Alamos NM 87545 USA 505 667 3113 G E Wilson INEL EG&G Idaho Inc PO Box 1625 Idaho Falls ID 83415 USA 208 526 9511 l

l l

G-34

G W Johnsen INEL EG&G Idaho Inc PO Box 1625 Idaho Falls ID 83415 USA 208 529 9854 L D Buxton Sandia Labs Division 6444 Albuquerque NM 87185 USA 505 844 6537 J Sursock Electric Power PO Box 10412 Research Palo Alto Institute CA 94303 USA 415 855 2410 F Pelayo CSN Spain Sol Angela de la Cruz 3 Madrid Spain 456 3401 1812 J M Izquierdo CSN Spain Sol Angela de la Cruz 3 Madrid Spain 456 3401 1812 L Robollo Union Capitan Haya 53 Electrica 28080 Madrid Fenosa Spain 279 2500, Extn 2466 F Reventos Asociacion Girona 64 Nuclear Asco Ba rcelona Spain 343 231 7053 E Negrenti ENEA TERM V Anguillarese 301 00060 Roma Italy 396 3048 4265 G-35 I _ _ _ _ - - -

\

li P Marsili ENEA DISP Via Vialiano Brancati 48

\

00144 Roma Italy 8528 2174 H Stadtke JRC Ispra I-21020 Ispra Varese Italy 0332 789986 P M Stoop Netherlands PO Box 1 Energy Research 1755 ZG Foundation Petten The Netherlands 0 2246 4462 M Analytis EIR 5303 wurenlingen Switzerland 56 992706 I

T Sirkia VTT Technical Research Centre of Finland Nuclear Engineering Lab PO Box 169 00181 Helsinki Finland 358 0 648931 S Petelin Institut Reactor Engineering

" Joseph Stefan" Division 61111 Ljubljana PO Box 100 .

Yugoslavia 061 313 022 I V Vojtek GRS GRS mbH Forschungsgelande 8046 Garching '

West Germany 089 32004 112 F Winkler KWU Hammerbacherstrasse 12-14 St D-8520 Erlangen West Germany 09131 18 2156 G-36

H Plank KWU Hammerbacherstrasse 12-14 St D-8520 Erlangen West Germany 09131 18 5634 S Naff USNRC/EG&G KWU Hammerbacherstrasse 12-14 St D-6520 Er1.angen Wcat Germany 09131 18 3905 A B Forge CEA DERS/SEAREL CEN Fontenay aux Roses BP No 6 92260 Fontenay aux Roses Franco 1

33 1 46 54 82 96 R L Pochard CEA CEN FAR DERS/SEAREL BP No 6 92260 Fontenay aux Roses France 33 1 46 45 81 99 B Adroguer CEA Cadarache CEN Cadarache DERS/SEAREL 13108 Saint Paul les Durance France 42 25 23 34 B Spindler CEA Grenoble SETh CENG 85X 38041 Grenoble Cedex France 76 88 46 87 Bub Dong Chung Korea Advanced PO Box 7 Energy Research Daeduk-Danji Institute Choong-Nam Korea 042 829 2613 G-37

9 H'Adachi- JAERI Toka'i-mura Ibaraki-ken 319-11 Japan 0292 82 5272 Y Murao JAERI Tokai-mura Ibaraki-ken 319-11 Japan 0292 82 5272 l

G-38

-APPENDIX H BIBLIOGRAPHY OF PAPERS PRESENTED AT ICAP MEETINGS This appendix presents a list of papers presented at-the meetings held by the ICAP to date.

H LIST OF PAPERS PRESENTED AT THE FIRST MEETING OF THE 'ICAP .

April 23-26, 1985

1. Status of the TRAC-PF1/M001 Code and Documentation, Los Alamos National Laboratory.
2. Status of the RELAPS/ MOD 2 (version 36) Code and Documentation, Idaho National Engineering Laboratory.
3. Status of the TRAC-BD1/ MODI (version 22) Code and Documentation, Idaho National Engineering Laboratory.
4. COBRA-TRAC and COBRA-TF Code Status, Pacific Northwest Laboratory.
5. Management of the ICAP Program, F. Odar, U.S. Nuclear Regulatory Commission.
6. Integration of Assessment Results to Quantify Code Uncertainty Using Statistical Methods and Key Parameters, F. Odar, U.S. Nuclear Regulatory Commission.
7. Guidelines and Procedures for Performing and Documenting Assessment Studies and Discussions, D.E. Bessette, U.S. Nuclear Regulatory Commission.
8. CSNI Assessment Matrix for PWRs, F.R. Germany.
9. U.S. Proposal for BWR Assessment, Idaho National Engineering Laboratory.
10. EPRI Perspective on Code Assessment Requirements to Support Applications Analysis, Electric Power Research Institute.
11. REWET-II Facility; Use of RELAP5, Finland.

H-2

12. Use of RELAP5 and Other Codes in the Netherlands; ECN Opinion on Future Needs, Netherlands.
13. Current Status and Program of KAERI on Code Assessment and Applications, Korea.

l

14. UPTF, PKL, and Karlstein Facility Descriptions, Test Objectives, and Typical Results, Germany, KWU.
15. Code Assessment Matrix Including Description of Proposed Experiment-and Plant Transient Calculations with TRAC and RELAP5, Germany, KWU.
16. Proposed Code Improvements for TRAC-PF1 AND RELAPS Based on GRS and KWU Experience, Germany, GRS.
17. SPES: The Italian Integral Test Facility for PWR Safety Research, Italy.
18. Planned Assessment of RELAPS/ MOD 2 and TRAC-PFI/ MODI, Sweden.
19. Description of the NEPTUN Facility and Survey of Conducted Tests and NEPTUN Working Data Bank, Switzerland.
20. Development of New Post-CHF Heat Transfer Models, Switzerland.
21. Thermal Hydraulic Code Assessment and Applications Program in the United Kingdom, United Kingdom.
22. Status of TRAC-PF1/M001 in the United Kingdom, United Kingdom.
23. Thermal Hydraulic Experiments at the UKAEA Heat Transfer Laboratories, ,

United Kingdom.

24. JRC Ispra Improvements of RELAPS/ MODI, JRC Ispra.

1 i

^

H-3

25.- LOBI /M002 Status and Plans, JRC Ispra.

26. Future Regulatory Needs: Codes,-Experiments, U.S. Nuclear Regulatory Commission.
27. Assessment of COBRA-TF During Reflood with Blockages and Grids, Westinghouse.
28. The OTIS/ MIST Program, Babcock and Wilcox.
29. Current Status of Experimental Data for the Assessment of RELAP5/M002 and TRAC-BWR, Idaho National Engineering Laboratory.

H-4

LIST OF PAPERS PRESENTED AT THE FIRST PROGRAM MANAGEMENT MEETING October 1985

1. Overview of USNRC Plans for TRAC-PWR, RELAP5, and TRAC-BWR for the period 1985-1990, U.S. Nuclear Regulatory Commission.
2. Management of ICAP at the Idaho National Engineering Laboratory (INEL) and Los Alamos National Laboratory (LANL).
3. Status of Assessment Plans for Each Participant.
4. Plans for Program Group, Task Group, and Specialist Meetings for 1986.

l l

H-5

LIST OF PAPERS PRESENTED AT THE SECOND PROGRAM MANAGEMENT MEETING November 1986

1. The Regulatory Significance of the ICAP, Dr.-B. Sheron, USNRC.
2. ICAP- An Integral Part of the NRC Thermal-Hydraulic Code Assessment and Development Program, L.M. Shotkin, USNRC.
3. Revision of the ECCS Rule, W.D. Beckner, USNRC.
4. A Method for Developing a Document on Code Scaling Capability and Applicability to NPP Safety Assessment, N. Zuber, USNRC.
5. Code Uncertainty Methodology, F. Odar, USNRC.
6. Proposed Code Assessment Matrix for LBLOCA, F. Odar, USNRC.
7. Belgian Code Assessment Matrix (Proposal), E.J. Stubbe, Belgium.
8. Finnish Status, Plans and Comments (RELAPS Assessment), H. Holmstrom, Finland.
9. FRG- Code Assessment' Matrix with TRAC and RELAP5,' F. Winkler, F.R.

Germany.

10. Results from Assessment of RELAPS/M002 and TRAC-PF1/M001 in the FRG, F. Winkler, F.R. Germany.

2

11. Italian Activity Plan for the ICAP, E. Negrenti, Italy.
12. ICAP Implementation Plan in Korea, H.J. Kim, Korea.
13. Use of RELAPS/ MOD 2 in the Netherlands, J.E. Speelman, Netherlands.

l H-6

14. Swedish Experience with RELAP5/M002 Assessment, O. Sandervag, Sweden.
15. Implementation Plans for the Code Assessment Matrix and Status at EIR, S.N. Aksan, Switzerland.
16. Thermal-Hydraulic Code Assessment An'd Applications Program in the Republic of China, L.Y. Liao, Institute of Nuclear Energy Research.
17. United Kingdom Code Assessment Program, J. Fell, United Kingdom.
18. French Implementation Plans for ICAP, M. Reaucreaux, France.
19. Plan of Best Estimate Code Assessment and JAERI Activities in the Field of BE Code, Y. Murao, Japan.
20. Management of the ICAP Program, P. Ting, USNRC.
21. RELAP5 Status and Plans, G. Johnsen, INEL.
22. TRAC-BWR Status and Plans, G.-Johnsen, INEL.
23. TRAC Status and Plans, J. Spore, LANL.
24. Small Break Loss-of-Coolant Accident Assessment Matrix, R.G. Hanson, INEL.

L

25. Evaluation of ICAP Results Case Report, R.P. Jenks, LANL.

i l

H-7

LIST OF PAPERS PRESENTED AT THE FIRST ICAP SPECIALIST MEETING June 1986

1. Assessment of RELAP5/M002 in NEPTUN Boiloff and Reflood Experiments -

Richner, EIR

2. Assessment of RELAP5/M002 Cross flow Model with EPRI Single Phase Data

- Wilson,~INEL

3. RELAP5/M002 & TRAC-PF1/M001 Post-Test Calculations of ATWS Safety Valve Experiments With Two-Phase Mixture Discharging - Fuzalowski, KWU
4. Assessment of RELAP5/M002 Against 25 Dryout Experiments Conducted at the Royal Institute of Technology - Sjoeberg, Studsvik
5. Post Test Calculation of PKL-118-2 Experiment With RELAP5/M002 - Dang, KWU
6. Assessment of RELAP5/M002 With Small Break Data From Semiscale and LOBI; Natural Circulation and Boiler-Condenser Heat Transfer Data From GERDA and Steam Generator Data From MB Wilson, INEL
7. PKL-106 (9mm Cold Leg Break With Hot Leg Injection) Post Test Calculation With RELAP5/M002 - Nguyen, KWU
8. RELAP5/M002 Code Assessment Post Test Calculation Results of the Commissioning Test " Reactor Trip At Full Load" at a PWR Power Plant -

Gerth, KWU i

9. Overview of the Assessment of Thermal Hydraulic Transient Codes at EIR

- Aksan, EIR

10. Pre-Test Calculations for UPTF Hot and Cold leg Flow Pattern Test -

Riegle, GRS

11. Karlstein Upper Tie Plate CCF Test Calculations With TRAC-PFl; Pre-Test Calculations for UPTF Subcooled Tie Plate CCF Tests With TRAC-PF1 -

Glaeser, GRS

12. TRAC-PF1 Analysis of the CISI Countercurrent Flow Limitation Tests (Pressurizer Flooding ) - Richards, AEE Winfrith
13. TRAC Modelling of Stratified CCF in Horizontal Pipers (comparison with exact solution) - Richards, AEE Winfrith

! 14. Some Aspects of Critical Flow Modelling With TRAC - Richards, AEE Winfrith

15. TRAC Analysis of the BCL Downcomer Bypass Tests - Coddington, AEE Winfrith
16. TRAC Analysis of the THETIS Reflood Hydraulics - Coddington, AEE Winfrith H-8
17. TRAC User Guidelines for Form Losses, Momentum at T-Junctions, etc. -

Coddington, AEE Winfrith

18. PKL-llB3 Pre-Test Calculation With TRAC-PF1 - Trambauer, GRS
19. Post Test Calculations for ISP18 Using TRAC - Richards, AEE Winfrith
20. Assessment of TRAC-PFl/ MODI Against a Loss-of-Grid Transient in Ringhals Sjoeberg, Studsvik
21. Assessment of TRAC-PF1/ MODI in the United States - Knight, LANL
22. Assessment of TRAC-BDl/ MODI with Boiloff and Reflooding . Experiments:

Model Improvements and Numerical Problems - Analytis, EIR

23. Review of Procedures for Performing and Documenting Code Assessments -

Bessette, NRC

24. Presentation and Discussion of Methodology to Quantify Code Accuracy -

Wilson, INEL

25. Methodology for Determining Code Scaling Capability and Applicability to Power Plants - Zuber, NRC
26. Presentation of Errors Reported and Errors Corrected on RELAP5 and TRAC-BWR; Presentation of Reported Model Deficiencies; Presentation and Discussion on Unresolved Errors and Model Deficiencies and Discussion of Priorities for their Resolution - Johnsen, INEL
27. Presentation of Errors Reported and Errors Corrected on TRAC-PWR:

Presentation of Reported Model Deficiencies; Presentation and Discussion on Unresolved Errors and Model Deficiencies and Discussion of Priorities for their Resolution - Jenks, LANL

28. Discussion & Comments on Documentation of the Code, Including User Guidelines; Discussion and Comments on Documentation Associated with Information Transfer Including Code Newsletters - Johnsen/Jenks, INEL/LANL H-9

LIST OF PAPERS PRESENTED AT THE SECOND ICAP SPECIALIST MEETING January 1987

1. I. Vojtek, " ASSESSMENT OF TRAC-PF1/M001 TWO-PHASE FLOW FORCED CONVECTION HEAT TRANSFER CORRELATIONS," Gesellschauft fur Reaktorsicherheit mbH, Garching, FRG.
2. G. Ahmed, "TRACJ FI/M001 ANALYSIS OF THE MARVIKEN CRITICAL FLOW TESTS,"

(p m re d by P. Bratby), NNC, UK.

3. P. O ddington, "SOME ASPECTS OF LARGE BREAK CALCULATIONS IN A 4-LOOP PWR USING THE TRAC-PF1/ MOD 1 CODE," AEE Winfrith, UK.
4. L. Rebollow Meirano, " APPLICATIONS OF " LESSONS LEARNED" FROM THE ANALYSIS OF OECD-LOFT LP-SB-3 EXPERIMENT TO JOSE CABRERA NUCLEAR POWER l

PLANT," Union Electra Fenosa, S. A., Madrid, Spain.

5. A. H. Scriven, " ANALYSIS OF LOBI TEST BLO2," CEGB, UK.
6. T. Sirkia, "FINNISH ASSESSMENT OF RELAP5/ MOD 2," Technical Research Centre of Finland.
7. P. M. Stoop, J. P. A. v.d. Bogaard, H. Koning, " FEED-AND-BLEED ANALYSIS FOR THE BORSSELE NUCLEAR POWER PLANT," Netherlands Energy Research Foundation ECN.
8. H. Stadtke, B. Worth, C. Addabbo, "RESULTS OF RELAP5 POST-TEST CALCULATIONS FOR LOBI TEST A2-90: LOSS OF 0FFSITE AND NORMAL ONSITE ELECTRICAL POWER - ATWS CASE." Commission of the European Communities, Joint Research Centre, Ispra.
9. H. Plank, " ANALYSIS OF A DOUBLE ENDED COLD LEG BREAK 0F A 1300 MW KWU PLANT WITH COMBINED INJECTION," Kraftwerk Union AG, FRG.
10. C. G. Richards, " PRE-TEST CALCULATION OF L0BI TEST BLO2 USING TRAC-PF1/ MODI," AEE Winfrith, UK.
11. F. Odar, " PROPOSAL FOR LBLOCA SEPARATE EFFECTS TEST MATRIX," USNRC, USA.
12. G. W. John:en, "RELAP5 STATUS & PLANS," EG&G Idaho, Inc., Idaho National Engineering Laboratory, USA.
13. G. W. Johnsen, " TRAC-BWR STATUS & PLANS," EG&G Idaho, Inc., Idaho National Engineering Laboratory, USA.
14. G. Case and G. Wilson, " CODE ASSESSMENT REPORT

SUMMARY

EVALUATIONS FOR THE TRAC-BWR AND RELAP5 CODES - METHOD, EXAMPLE, AND STATUS," EG&G Idaho, Inc., Idaho National Engineering Laboratory, USA.

15. C. D. Fletcher and C. M. Kullberg, " BREAK SPECTRUM ANALYSES FOR SMALL BREAK LOCA IN A RESAR-3S PLANT," (presented by G. Wilson), EG&G Idaho, Inc., Idaho National Engineering Laboratory, USA.

H-10

16. L.D. Buxton, R.K. Byers, M.G. Elrick, L.N. Kmetyk, D. Dobranich, A.C. Peterson, " TRAC-PFl/M001 INDEPENDENT ASSESSMENT AT SANDIA NATIONAL LABORATORIES," Sandia National Laboratories, USA.
17. R.P. Jenks, J.W. Spore, T.D. Knight, D.R. Liles, L. A. Guffee, "RECENT TRAC-PFl/ MODI CODE AND DOCUMENTATION CHANGES RESULTING FROM ICAP," Los Alamos National Laboratory, USA.
18. T.D. Knight, R.P. Jenks, J.W. Spore, D.R. Liles, " MAJOR ICAP CODE CONTRIBUTIONS IN FY86 AND FUTURE ICAP CODE MODIFICAIT0NS: A LOS ALAMOS PERSPECTIVE," Los Alamos National Laboratory, USA.
19. P. Ting, " FUTURE ICAP MEETINGS IN 1987 AND 1988," USNRC, USA.
20. F. Odar, "RELAP5/M002 IMPROVEMENTS," USNRC, USA.
21. Y. Murao, " CODE ASSESSMENT WORK IN JAERI," JAERI, Japan.
22. P. Coddington, "FP1 ACCUMULATOR BLOWDOWN CALCULATION," AEE Winfrith, UK.
23. P. Coddington, " ANALYSIS OF THE ACCUMULATOR NITROGEN PHASE OF A LARGE BREAK TRANSIENT," AEE Winfrith, UK.
24. K.H. Ardon and A.J. Clare, " ASSESSMENT OF INTERPHASE DRAG CORRELATIONS IN RELAPS/ MOD 2 AND TRAC-PFl/ MODI," CEGB, UK.
25. M.G. Croxford and P.C. Hall, " ANALYSIS OF THE THETIS BOILDOWN EXPERIMENTS USING RELAP5/ MOD 2," CEGB, UK.
26. E. Allen, " PRELIMINARY RESULTS OF OECD-LOFT LP-SB-1 ANALYSIS WITH TRAC-PFI/M001 (802A)," AEE Winfrith, UK.
27. F. Pelayo, " PRELIMINARY RESULTS OF OECD-LOFT SB-2 ANALYSIS WITH TRAC-PF1/ MODI," CSN, Spain.
28. F. Reventos, "0 ECD-LOFT LP-FP-2 EXPERIMENT CALCULATION USING RELAP5/ MOD 2," Asociacion Nuclear ASCO, Spain.

l \

H-11

l p eo =

u ucum.ouwoav ca . i . oar ~u u u .,r,oc. ,v., .., ,

'E. '$'- EZLIOGRAPHIC DATA SHEET NUREG-1270 un insta 1 ous o va. . v.as. Vol. 1 2.1'IT LE .No . TITLE 3 LE AVE .LA%E Interna ional Code Assessment and Applications Program Annual Rep rt 4 oAn ae, oar mfuno

s. aur oais' oo 1,.

y ve.a March / 1987

. o.n .v, oar issuae oo.1,. v...

P. Ting, R. Hans n, R. Jenks March /

7l 1987 t ,ea,0....c oac..a.iio. ... . o .,u~o .ooa.ss ,,. i. c , s eaoacriras. nu=1,u..aa Office of Nuclear R qulatory Research . ..s oa ca. .vo..a U.S. Nuclear Regulator Comission Washington, DC 20555 io s,o soa,~o oaa. a ar,0 .... .~o ...u.a .oo .ss ,,, ,, c , V,.o,.., oar Technical Same as 7. above. . ,aaioocoveaeo,,- . ,

12 sue,LE ENT.av IsoTES

\

i3 ..sf aACT (100 werea ., ses.,

The first ICAP Annual Report is devoted to verage of program activities and accomplishments during the period between ril 1985 and March 1987. The ICAP was organized by the ' Office of Nuclear Regu 4 tory Research, United States Nuclear Regulatory Commission in 1985. ThewlCAfisaninternationalcooperativereactor safety research program plannt.d to provide ind't$ pendent assessment of the NRC computer codes developed for analysi 'of reacto transients and loss-of-coolant accidents.

,e 1

is oocu int amatvsis .. asvwonossosscaierons is .v.....iurv

"" l ICAP TRAC-BFI i Annual Report COBRA-TF RELAPS/ MOD 2 Code Assessment Unlimited TRAC-PFl/M Computer Codes '"<cua'tv e'^= cat'o*

rra o ,,,

. iormteeitasio,e nosotsmos Unclassified trs. em, o a.ssifi o, ,.c.es d 1

14 ,ascE

UNITED STATES etassaan ses:aat rouam, rets rano NUCLEAR REGULATORY COMMISSION rosT4ces WASHINGTON, D.C. 20555 WEE"Ec.

PERMIT No. G47 OFFICIAL BUSINESS PENALTY FOR PRIVATE USE,4300

__