IR 05000445/1986018

IE Insp Repts 50-445/86-18 & 50-446/86-15 on 860512-16 & 27-30.Concerns Noted:Failures to Meet Design Criteria Comments or Specs Not Found by Tera During Scan of Documents for Homogeneity

ML20209G663
Person / Time
Site:	Comanche Peak
Issue date:	08/20/1986
From:	Imbro E, Norkin D NRC OFFICE OF INSPECTION & ENFORCEMENT (IE)
To:
Shared Package
ML20209G653	List: IR 05000445/1986018 ML20209G647
References
50-445-86-18, 50-446-86-15, NUDOCS 8609150046
Download: ML20209G663 (17)

Text

. - - _ _

.

'.

U.S. NUCLEAR REGULATORY COMMISSION OFFICE OF INSPECTION AND ENFORCEMENT Division of Quality Assurance, Vendor, and Technical Training Center Programs Quality Assurance Branch Report No /86-18,50-446/86-15 Docket Nos.: 50-445,50-446 Licensee: Texas Utilities Generating Company Skyway Tower 400 North Olive St., L.B.81 Dallas, Texas 75201 Facility Name: Comanche Peak Steam Electric Station (CPSES)

Inspection At: TERA Corporation, Site office (May 12-16, 1986)

TERA Corporation, Bethesda, Maryland (May 27-30, 1986)

Inspection Team Members:

Team Leader D.P. Norkin, Senior Inspection Specialist, IE Mechanical Systems / J.Blackman, Consultant, WESTEC Services Components S.Klein, Consultant, WESTEC Services J.Nevshemal, Consultant, WESTEC Services

\

'

$-Ih$h

- -

Donald P. Norkin Date Team Leader, IE Approved by:

1. f [& b& Eugene V. Imbro, Chief

.?- 0 -?(

Date Licensing Section Quality Assurance Branch k

O DO

. . ~ _ .- _ _ _ .

__

'

l

'

\

l

.

.

LIST OF ACRONYMS NPSH -

Net Positive Suction Head NPSHa -

Net Positive Suction Head Available ASME -

American Society of Mechanical Engineers MTD -

Mean Temperature Difference SSI -

Safe Shutdown Impoundment LOCA - Loss of Coolant Accident RHR -

Residual Heat Removal i P&ID -

Piping & Instrument Diagram

e e

-2-

- . . - . . . .-. . .- ..-. - - .

. _ _ _ . . - ..

_ _ . _ _ . _ _

,

.

i INTRODUCTION AND SUPEARY

'

1.1 BACKGROUND The NRC's Office of Inspection and Enforcement (IE) conducted an inspection of the Comanche Peak Design Adequacy Program (DAP) during the period from October 28 to November 1, 1985. This inspection focused on DAP review scope, review checklists, and personnel qualifications. The inspection report was issued on January 21, 1986. Since that time, Revision 3 of the Comanche Peak Response Team (CPRT) Progra,m Plan has been issued (January 27,1986),

the Phase 3 DAP scope expansion has been executed in order to encompass all of the design activities for safety-related systems, and TERA has expanded its review team, including the addition of about 60 Gilbert / Commonwealth personnel. IE is performing an in process assessment of (1) TERA inspections of the self-initiated portion of the DAP, and (2) the TERA third party review of Stone and Webster, Ebasco, and Impell reanalysis wor . 2 PURPOSE The IE inspection assessed TERA's self-initiated inspections for the Mechani-cal Systems and Components discipline, which are committed to in Discipline Specific Action Plan'X of the CPRT Program Plan. Inspections for DSAP's VIII, IX, and XI will be reported separatel .3 METHODOLOGY

.

The two key elements of IE's inspection were to evaluate whether (1) desige products sampled by TERA were representative of the design activities being reviewed and (2) TERA performed an adequate review of Gibbs and Hill and other design document With respect to representativeness, Appendix D of the CPRT Program Plan states that homogeneous design activities (HDA's) include items (e.g. calculations)

for which a high degree of correlation (homogeneity) exists in design criteria and design considerations, approach and methodology. Accordingly, evaluation of the adequacy of each HDA can be based on evaluating a representative selec-tion within each HD Procedure DAP-21, " Homogeneous Design Activity Validation

. ' and Selection of Specific Items for Review", established a process for validat-ing the. homogeneity of each HDA by briefly reviewing (or scanning) a minimum of five documents within the HDA population to confirm homogeneity relative to design criteria and design considerations, approach and methodology. Items scanned are required to be " distributed approximately even through the period of performance of the work as indicated by the current revision date of the document."

DAP-21 also established a process whereby one or more items are selected for review within each HDA. The emphasis is on selecting items within one of the systems included in the initial self-initiated program vertical slice; e.g.,

the auxiliary feedwater (AFW) system. In cases where this system is not considered to be representative of the HDA relative to comprehensiveness and/or complexity, DAP-21 provides other bases for system selection, starting with a system having direct interfaces to the vertical slice syste IE determined that DAP-21 provided an adequate method for documenting the confirmation of homogeneity of design documents within each HDA with respect-3-

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _

'

to design criteria and design considerations, approach and methodology. DAP-21 also provided an orderly method for documenting the bases of selection for

, review of design documents within specific systems. IE inspected checklists which documented TERA's DAP-21 scanning and selection process, as well as some of the documents scanned by TERA, in order to evaluate whether the design activities were homogeneous and the items selected for detailed review were representative of the HDA in comprehensiveness and/or complexit With respect to evaluating the adequacy of TERA's detailed reviews of selected documents, IE reviewed the checklists, evaluation reports, and Discrepancy / Issue Resolution Reports (DIRs) which documented the reviews and the actual documents which TERA reviewe IE also reviewed some documents within HDAs that were not selected by TERA for review in order to assess the appropriateness of extrapolating TERA findings on the selected documents to others in the populatio As referred to above and under Section 2.0, " scanning" (either by TERA or the team) is a brief review of the document to evaluate methodology for purposes of confirming homogeneity per DAP-21. " Reviewing" is the detailed evaluation of the

, document to determine if it meets design criteria commitment .4 SUMARY The team identified concerns with respect to scanning documents for homogeneit For HDAs M050.0 and M006.0, the scanned documents did not represent equal intervals during the period of design performance (0 pen Items M-15 and M-21).

For HDA 159A, the team scanned 5 documents which had been confirmed by TERA as homogeneous, and found differences in design methodology. This issue has generic implications for HDA confirmation since TERA apparently included Unit 2 documents, which were not actually calculations, in populations of scanned calculations. (0 pen Item M-3). As a related matter, Unit 2 documents stated that Unit 1 calculations could be used for Unit 2 application, but they provided no justification with respect to physical similarity of Units 1 and 2 (0 pen

Item M-22).

'

The team identified some technical concerns not specifically identified by

,

TERA. Generally, these involved additional concerns in areas where TERA

) had already identified concerns through DIRs. Examples we e:

(1) Uncertainties and inconsistencies with respect to correct component

.. cooling water (CCW) temperature (0 pen Items M-6 through M-9). These issues indicate the need for an integrated analysis for several sys-tems in the total heat removal chain. TERA should determine if there are other cases (including other disciplines) where integration of analyses has been lacking, resulting in inconsistencies in design parameters and assumptions.

.

!

(2) Uncertainty as to whether eductor flow is within a range to adequately control containment sump pH (0 pen Items M-12 through M-14).

(3) Pctential adverse interactions between Class 5 (non-safety) piping and the AFW system (0 pen Items M-17 through M-19).

(4) There was no relief valve on a portion of CCW piping which could contain trapped fluid between containment isolation valves (0 pen Item M-11).

,

-4-

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

, .

1.5 PERSONS CONTACTED

,

NAME COMPANY POSITION R. Boyd TERA Reviewer D. Davis TERA President W. Djordjevic TERA Reviewer L. Fernandez TERA Reviewer H. Levin TERA Review Team Leader R. Levline TERA Group Leader R. Newman TERA Site Activities Leader F. Schofer TERA Sys. Eng. Mg R. Sheldon Gilbert Reviewer J. Vance TERA Discipline Coordinator 2.0 INSPECTION DETAILS Review Topic: M02.1-Operating Limits: NPSH Representativeness of Reviewed Documents:

TERA scanned all five calculations in HDA MO 16.0A, NPSH - Rated Conditions -

Mechanical: 233-17 (SSW), 241-1 (Reactor Water Makeup), 311-N (Chilled ,

Water), 229-9 (CCW), and 206-8 (AFW). The team scanned the same calculation Calculation 229-9 includes a factor for velocity head loss which is not in the other calculations, and is not normally included as part of an NPSH cal-culation. Including this factor does not substantially alter the methodolog In other respects, the five calculations are homogeneous with respect to methodology; e.g., determining hydrostatic water pressure, calculating pressure losses due to piping, valves, and fittings, et ,

TERA reviewed calculations 206-8 and 229-9, covering NPSH for AFW and CCW system pumps respectively. The AFW selection was consistent with the DAP-21 requirement to select a review sample from the initial review scope vertical slice (the AFW system). The CCW selection was consistent with the DAP-21 requirement to make an additional selection (that is considered representative) from outside the vertical slice if selection within a vertical slice is not considered to be representative of the HDA relative to comprehensiveness and/or complexit '

Review of Documents:

The team reviewed TERA review checklists and calculations reviewed by TERA

! in order to evaluate the depth of review, and found that TERA review check-

! lists (and attached supplementary comments) adequately covered key points in the calculations. The team also reviewed several DIR's and found them to be appropriate. In general, the team found the TERA review of the NPSH topic to be comprehensive and adequate. TERA informed the team tha.t, under certain -

circumstances, the review will include evaluation of the complete design chain from inputs to outputs, and at other times the review is focused on a particular i aspect of the process; e.g., methodology. The team identified the following l examples where input and assumptions may be significant to NPSH calculations, '

but were not reviewed by TER _ - - - -. . . -

~

Calculation, there was no basis given for the 105 *F op which apparently represents normal condition ,

Higher operating temperatures

-

this system from residual heat removal and plant se Higher temperatures would increase the vapor pressure and decrease availab .

NPSH, but it was not clear that these modes of operation had been considered The calculation did not demonstrate whether the surge tank elevation used in .

the analysis represented the lowest anticipated level in the tan The friction factor assumed to calculate piping pressure drops was not sub-stantiate ft." The calculation states, "Specify pump as follows NPSH > 50 This apparent error permits the vendor to provide a pump with NPSH NPSH required should be specified as less than a specific Open Item M-1 For Calculation 232-1, Revision 2, dated 11/22/82, there is no basis given for the "K-factor" which accounts for sumContainment Spray ,

teristics in the computer program used to calculate line losses. p charac-clear howfor sump It was not accounted in thescreen analysi pressure drop, water level, and temperature had been Open Item M-2

,

no basis is given for the 120For Calculation 311-1N, dated 4/12/78, Safety ,

as whether condition modes of operation were evaluated to establish worst cas *

Open Itea M-3 Review Topic: M04 -Water Sources i

Representativeness of Reviewed Documents:

l The team reviewed the following calculations which were scanned (and con as homogeneous) by TERA for HDA 159A, Water Supplies / Volume-Time

.

the calculations).(comments noted below pertain to the team's assessme

'

(1.) Cal t containmcu,lation 528-2, dated 5/12/82, RWST Depletion Time Analysis (with I-due to tank entvolume spray orifice),

depletio iteratively evaluates the effect on pump flow (2.) Calculation 2-215-7 Sizing of Diesel Generator Fuel Oil Storage Tank, is applicable is not a calculation to Unit 2 per,se, (identicaland merely states that Unit 1 Calculation 215-7I engines).

(3.) Calculation 215-7, dated 11/17/75, Diesel Generator Oil ' Storage Tank Sizing, is a trivial calculation of tank dimensions based on fuel consumption'

diesel generator size, and a PSAR capacity requirement of "7 days + 15 ,

.

Condensate Measuring System, is a trivial calculatio used)

unit dimensions It is not clear that for athis level tankinstrument to collect condensate from the coo is safety-relate g-6-

_ ._ _ _ - - - - --

.

(5.) Calculation 229-12, dated 1/23/79, Component Cooling Water Surge Tank Size, addresses system piping and equipment volumes, coolant thermal expansion, system

'

leakage, and other factors pertinent to the required surge tank size and instru-mentation set points for various operating mode The tean concluded that 3 of the 5 scanned calculations, items (2) through (4)

above, differed in design methodology from the remaining two items. Items (2)

and (3) did not fall within the headings of the review topic and HDA; i.e., water sources and water supplies. For the two remaining scanned calculations (528-2 and 229-12), it was not clear from the DAP-21 HDA checklist that the design considerations, approach, and methodology were similar. The team considers that the scanning process should be reaccomplished for HDA 159A. For this HDA, and other HDAs where the initial scanning sample indicates methodology differences, 100% of the documents should be scanned. In addition, other HDAs should be re-examined for cases where Unit 2 documents, which are not actual calculations, are treated as calculations in the scanning proces Open Item M-4 Review of Documents:

Calculation 206-5, Revision 1, dated 10/30/78, AFW/ Condensate Storage Tank Size, was reviewed by TERA. The team found that it adequately documented a change in volume (required to meet design requirements) due to an increase in water temperature-from 100 F (quoted by Westinghouse) to 120 F. The team also reviewed calculations not reviewed by TERA. Calculation 215-11, Revision 2, dated 5/23/83, determined line losses from an air receiver to the diesel generator header and established the volume of the receiver. The num-ber of diesel generator starts assumed was not substantiated. For Calcula-tions 215-7 and 229-12 (scanned by TERA - see above section), no basis was given for input. For 229-12, it was not clear whether operating temperatures represented worst case operating mode Open Item M-5 Review Topic: M13 - Heat Exchangers Representativeness of Reviewed Documents:

The DAP 21 checklist indicated that all three documents within HDA M134.0A (MS49,50, and 51) were scanned. Specification 2323-MS49, Rev.2, 8/20/74, Component Cooling Water Heat Exchanger, which was selected for TERA review was, in the team's opinion, representative of specifications for safety related heat exchanger For HDA M134.08, TERA reviewed the 2 calculations in the populatio Review of Documents:

The TERA review checklists were satisfactory and adequately covered design criteria and requirements for the specification. DIR 00151, Rev. 1, dated 5/9/86 was issued by TERA against Specification MS-49. ASME Code,Section III requires the designer to determine whether impact testing is required, and state in the specification that it is not required, if such is the cas No such statement was found in the specification, as noted in the DIR. The resolution provided by the DIR stated that '; impact testing is not required for this service, but was not included in the design specification."

-7-

. _ . -

S

-

TERA informed the term of its justification for concluding that impact test-ing was n:t required. Service conditions where this option Zay be warranted involve operating modes which expose thicker sections at low temperatures and

. high pressure which could result in brittle fractur For the subject compo-nent, the service is low pressure with a design minimum temperature of 32 F which would be reached gradually. The component consists of predominantly thin segments. The team had no further question The team reviewed Calculation 232-20, Rev.0, dated 4/26/85, Containment Spray heat Exchanger, and found it to be generally adequate. The checklist (DAP-M060, Rev. 1, 5/20/86) adequately covers details of the TERA revie For Calculation 229-14 and Addendum 1, dated 7/19/85, CCW Temperature Profile, the team reviewed the following DIR's:

o DIR D-0115, Rev. 2, 5/9/86 stated that there was "no current calcula-tion which determines the CCW heat exchanger design capacity of 70 E+6 Btu /hr."

o DIR D-0158, Rev. 1, 5/9/86 indicated that an " unstated assumption was used in Calculation 229-14 and addendum that CCW Heat Exchanger transfer rate of 10.9 E+6 Btu /hr. degrees MTD (developed on sheet 3 of calculation) was constant."

Although these DIR's raised important issues concerning the substantiation of the heat exchanger design, the team identified additional concerns, as indicated belo Calculation 229-14 and Addendum 1 determined the CCW outlet temperature pro-file as a function of time, assuming a constant overall heat transfer coef-ficient and accounting for variations in the Safe Shutdown Impoundment (SSI)

temperature. However, only the LOCA and cooldown modes of operation were evaluated. The vendor provided design data for several modes of operation including: regular operation, regular operation with other unit in reactor cooldown, reactor cooldown at four hours, initial recirculation, recirculation with minimum safeguards, and desig There was no evidence that any of these modes of operation were considered with respect to determining the most limiting mode of operation, or that the calculated modes enveloped those not considere In addition, there was no apparent consideration of the effect of (1) a " clean" CCW heat exchanger on containment heat removal rates during a LOCA, and (2) higher heat loads removed from containment during a LOCA on SSI temperatures. Increased SSI temperatures, in turn, result in higher CCW temperature Calculation 229-14 notes that cooling water temperatures are limited by

equipment within the Westinghouse scope of supply, with a maximum allowable CCW temperature of approximately 136*F. Maximum calculated CCW temperature is 133 F. Thus, the above considerations are significant with respect to l potentially exceeding the maximum allowable cooling water temperatures l

imposed by Westinghous Open Item M-6 I

' The team reviewed Calculation 233-16, Rev. 1, 2/7/80, SSI Heat Loads, which served as input to the SSI hydrothermal simulation report, dated 5/80. This

! calculation indicated a CCW temperature of 110* was assumed for the cooling

!

i-8-l l

l t

.

, water to the containment spray and residual heat r:moval heat exchangers, which is lower than predicted in calculation 229-14 (1335F) and implies the

.

use of lower SSI temperature Open Item M-7 The team noted a discrepancy in Calculation 233-16 between the heat loads at one hour after LOCA tabulated for the containment spray and residual heat removal heat exchangers in Table 3, Sht. 4 (325 E+6 Btu /hr), versus Figure 5 (370 E+6 Btu /Hr). If the correct heat load is 370 E+6 Btu /hr, it will affect the results of calculation 229-14 and the above SSI hydrothermal simulation report which are based on 325 E+6 Btu /hr. TERA advised the team that this calculation will be reviewed as part of another HDA and the discrepancy will be investigate Open Item M-8 The above concerns and inconsistencies (0 pen Items M-6 through M-8) indicate the need for an integrated analysis of all systems involved in the total heat removal train from containment to the SSI. As a related matter, the following cases should be evaluated to envelope the total effect of heat exchanger fouling on performance during a postulated LOCA:

(1) To maximize CCW temperature - Clean RHR and spray heat exchangers and fouled CCW heat exchange (2) To maximize containment sump temperature (maximum containment pressure) - All heat exchangers foule ,

(3) To maximize heat load to the SSI - All heat exchangers clea Open Item M-9 Review Topic: M02.3 - Operating Limits: Overpressure Conditions M15.0 - Safety Valves M16.0 - Relief Valves Representativeness of Reviewed Documents:

For the HDA's covered by the review topics (M02.3, M15.0 and M16.0), selection

. ' of representative review items was not necessary because 100% of the occupants (calculations or specifications) were reviewed by TERA, as listed below:

HDA Item (s) Reviewed M01 G&H Calculations 202-3 and 229-16 l M01 TNE Calculation TNE-CA112 M13 G&H Specifications 2323-MS-26 and 2323-MS-77 M13 G&H Specification 2323-MS-23 M17 G&H Calculations 202-16, 229-17 and 550 .

M17 TNE Calculation TNE-CA-103 i

Review of Documents:

For HDA M137, the team reviewed the documents reviewed by TERA: Specification 2323-MS-26, Rev. 3, 5/1/85, Nuclear Safety and Relief Valves, and Specification 2323-MS-77, Rev. 2, 8/2/76 Main Steam Safety Valve The team had no concerns with the documents or with TERA's revie _ - _ _ _ _ _ _ _ _ _ _ - - - _ _ _ _ . - . . . __ -

'

The team reviewed Calculation 202-3, dated 7/27/74, Main Steam Relief Valve Sizing and the checklist (DAP-CLC-M-004, Rev. 1, 5/20/86), and found the check-list to be adequate and the DIR's appropriate. The team reviewed Calculation

-

229-16, Rev. O, 11/9/84, Steady State Analysis of Thermal Barrier Cooler Break and the checklist (DAP-CLC-M-003, Rev.1, 5/21/86), and found that the checklist and related DIR's covered a number of items on deficient input documentatio A two phase flow multiplier for pressure drop was not applied to one of the calculated pressure drops. However, application of the multiplier would result in a larger pressure drop and a lower pressure in the line, which would not l change the calculation's conclusion that pressures resulting from the postu-lated thermal barrier tube break will not exceed design pressure for the piping, i Westinghouse letter WPT7524 dated 10/1/84 (Reference 12 of Calculation 229-16)

indicated that the flow used in this calculation was conservativ The team reviewed the checklist (DAP-CLC-M-003, Rev.1, 5/21/86) for Calcula-tion TNE-CA-103, dated 11/19/84, CCW Surge Tank Overpressure Protection Analysis of RHR HX Tube Break, and found that it adequately covered the calculation of the effects of a tube break with reactor coolant flowing through tubes cooled by component cooling water. The worst case condition was identified as a resi-dual heat removal heat exchanger tube break. The team had no concerns with this calculation or with TERA's revie Calculation 550, Rev. O, 5/25/85, Nuclear Class Safety Valves, established the i flow requirements for various safety-related relief valves, including thermal relief valve The basis for sizing thermal relief valves is to relieve the pressure increase resulting from thermal expansion of liquid coolant trapped on the cooling water side of an isolated heat exchanger with flow on the hot water side. Volumetric expansion is a function of the change in fluid tempera-ture, which is related to the heat load input to trapped liqui For all valves sized by Gibbs and Hill (using the methods of Calculation 550) the heat loads used were based on the worst case heat loads transferred by the heat exchanger during operation, such as those tabulated by Westinghouse for equip-ment within the NSSS scope of supply. The sizing analyses therefore involved an implied assumption that the heat transferred on the cooling water side with

no flow (heat exchanger isolated) was the same as with normai cooling water flo DIR 0-0147 identified the absence of documentation to substantiate the sizing of the thermal relief valve on the residual heat removal heat exchange TERA used the above Gibbs and Hill sizing approach to determine whether this valve was properly sized. The TERA calculation indicated that the valve size (26 gpm) was inadequate and that a valve size of 58 gpm would be required. TERA stated that they performed the alternate calculation to determine the signifi-cance of Gibb's and Hill's calculation omission, not to bound all possible operating conditions. The DAP plans to verify that corrective action resulting from this discrepancy will bound all anticipated system conditions to ensure that the component is properly sized for its intended service. The team con-siders that any conclusions in this regard should also be applied to valves not addressed by DIR D-014 Open Item M-10

- 10 -

I

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

._

.. .

.

.

. .

The teaa r2 viewed Component Cooling Water P&ID 2323-M0231, Sheet 3 of 3, -

Revision 12. P&ID 2323 - M0231 is not listed above ss part of an HDA because

'

drawings are not included withih the HDA validation process. TERA considers that drawings do not illustrate changes in either criteria or method 51og However, TERA stated that this P&ID was reviewed under' Review Topic Line 2-CC1-183-152-2 has no relief protection for the~ portion which is normally isolated and penetrates containment. Any coolant trapped between the isolation valves could develop excessive pressures during temperature excursions in containment and compromise the penetration piping. Proper relief protection should be provided for this-portion of the piping, as well as for~li~es n in other systems which can be isolated with a trapped fluid subject to containment'

temperature excursion .

.

Open Item M 11 Review Topic: M18.0 - Eductor

-

Representativeness of Reviewed Documents: -

Since the HDA (M141.0) covered by the review topic (M18.0) contains a single document (G&H Spec. 2323-MS-35 hv.1, entitled Nuclear Eductors), the appli-cation of the DAP-21 procedures for homogeneity validation and selection of a

,

!

representative review item was not necessar '

Review of Documents: '

TERA's review of the Gibbs and Hill eductor purchase specification'(2323-MS-35)

was based on review checklist (DAP-CLC-M-007, dated 4/21/86) criteria that included materials of manufacture, operational loads, industry codes and functional requirements. The TERA review resulted in the following DIR's being written for violations of the following criteri DIR Criteria Violated _

D-0248 Criteria 17 for"opepational loads D-0250 Criteria 28 for material of manufacture D-0223 Criteria 29 for verification of motive fluid D-0249 Criteria 48 for proper industry code

_

The team determined that each DIR apprepriately addressed a discrepancy in the purchase specificatio The ed'uc' tor is required to add NaOH to sustain sump water alkalinity within a design pH range of 8.6 and 10.5. In order to meet this FSAR commitment, Na0H flowrate must be maintained within a range during the injection phase for all possible system configurat. ions. The checklist had en FSAR criterion for minimum Na0H flowrate of 25 9pm, but noie for maximum flowrate. The team could not find a calculation relating the 25 gpm criterion to the pH rang The purchase specification required the eductor to provide an NaOH flow of 45 gpm with given suction pressure and motive fluid flow criaracteristic It was not clear how this 45 gpm requirement (which the vender met) related to the 25 gpm FSAR critorion or the design pH range. Assurance needs to be pro-vided that the purchased eductor ca maintain the desiga'pH range for all system configuration Open Item M-12

- 11 -

__ _-

_ _ . ._ .

'. ,

,. .

,

.

As a result of the abov@ concern, the team reviewed relevant calculations,

~.

232-16 and 232-7, which were not part of HDA M141.0 and had not yet been

,

-

reviewed by TER/,.X Calculation 232-16 Rev. 1, dated 5/14/84 is apparently

, an analysis of NaOH flowrates for various system configurations. The team reviewed this calculation and found that it did not account for the Chemical

> Additive Tank (CAT) nitrogen blanket pressure and the effect of the eductor on the system flow; rat s .-

-

-

Open Item M-13 m

.f ,

- The tesm reviewed cdiculation 232-7 Rev. 3, dated 5/21/84 which determined

'

' injection and recirculhtion phase pH based on Na0H flowrates determined by calculation 232-16. This calculation did not substantiate the aforementioned FSAR stipolated pH range'and minimum Na0H flowrate. Since calculation 232-16 may be erroneous, its impact on calculation 232-7 needs to be assessed. In

~

addition, reanalysis of the NaOH flowrates may result in revised flowrates, which could affect the time assumed for fission product removal in the accident

-

analysis and change the size of the CA Open Item M-14 Review Topic: M31.0 - Radiation Exposure

, ,. Representativeness of Reviewed Documents:

The HDA (M050.0 Source Terms - Accident / Normal) originally had twelve cal-culations, from which the following five were selected for scanning:

G&H Calculation'61, dated 4/21/75 G&H Calculation 75, dated 5/8/75 G&H Calculation 76, dated 5/8/75

.G&H Calculation 239-7 Rev. 1, dated 12/28/77 G&H Calculation SH-TMI1G Rev. 1, dated 2/11/82 Thiee of the calculations were performed in 1975, one in 1977, and one in 1982. Since the entire population covered the period from 1973 to 1982, the scanning did not represent equal time intervals, as required by DAP-21. For example, there were calculations in the population from 1973, 1975, 1977, 1981 and 1982 (there were no calculations for the years 1978, 1979, and

- 1980.)

. Open Item M-15

! The original scanning of HDA-M050.0 resulted in identifying the need for a new HDA, M050.0 The team found the new HDA included Calculation 239-7 Rev. 1, a hand calculation, together with four computer calculations (utilizing the HOMES computer code). In this case, the team considered it inappropriate to include the computer and hand calculations within the same HDA since they have different methodologies. TERA agreed, and plans to move the hand calculation to another HDA, M050.0 ,

Open Item M-16 l

- 12 -

,

l

.

.

Review Topic: M08 - Class 5 Piping

,

Representativeness of P,eviewed Documents For HDA-M071.08 Safety Class Interfaces - Class 5 Piping, TERA selected for review the AFW system flow diagrams. Prior to the program cutoff date of 4/1/85, the AFW system was represented on a single flow diagram (2323-MIO206 Rev. CPS, dated 7/25/84). After the cutoff date the AFW system was repre-sented on three flow diagrams. One was produced by Gibbs and Hill (2323-MI-0206

.Rev. CP-6, dated 7/27/85) and the other two by TUGC0 Nuclear Engineering (TNE-MIO206-1 and -2, both Rev. CP-1, dated 7/27/85). TERA reviewed all four documents, i.e. both pre and post-cutoff dat Because of the large amount of Class 5 piping in the AFW system, the team agrees with the selection of the AFW flow diagrams as being representative of HDA-M071.0B for all the safety system Review of Documents:

This review topic evaluates design activities which involve providing satis-factory barriers between safety class (1, 2 and 3) pipe and any interconnect-ing non-nuclear safety (NNS) pipe, which for CPSES is termed Class The code committed to by the applicant for piping classification and barrier requirements is ANSI-N18.2/ANS-51.8, 1973 (FSAR para. 3.2.2). The team reviewed the checklist (DAP-CLC-M-008) used to review the following post-cutoff date documents:

2323-MI-0206 Rev. CP-6 TNE-MI-0206-1 Rev. CP-1 THE-MI-0206-2 Rev. CP-1 The checklist indicated that the reviewed documents were unsatisfactory (UNSAT)

for all checklist attributes, and listed four DIRs that had been written (0-0011, -0014, -0015, and -0020). The team reviewed the DIRs and found that they referenced the pre-cutoff date design document (2323-MI-0206 Rev. CP-5)

as being the source document for each individual discrepancy. The team veri-fied that both the pre and post-cutoff date documents demonstrated the discre-pancies indicated by the DIR On drawi.ng TNE-MI-0206-02 Rev. CP-1, valve 1-LV2478 provides a break between Class 3 and Class 5 pipin This valve is an active component which provides makeup to the Condensate Storage Tank (CST). The valve is not instrumented to close upon the occurrence of a seismic event. If at the time of an event resulting in the loss of the Class 5 piping the valve is open to fill the CST, the single active failure of the downstream check valve (1AF-009) to close might drain the CST to below the minimum operating level needed to support the AFW syste , Open Item M-17 On drawing TNE-MI-0206-01 Rev. CP-1, a normally open high point leakoff. valve (1AF-120) provides the barrier between Class 3 and Class 5 piping; a 3/8 i flow restrictor is in the Class 5 portion of the lin This arrangement apparently violates ANSI-N18.2/ANS-51.8, which requires that, for an open valve barrier, the loss of lower class components due to the seismic event should not prevent the system from performing its safety function. In the current design, the 3/8 in. flow restrictor would be lost as a result of an

- 13 -

,

.

cvent, p ssibly resulting in an unacceptable loss of the AFW backup water supply (service water). The team was unable to find an analysis verifying

, the ability of the AFW system to carry out its safety function with the loss of the 3/8 in, flow restricto The team reviewed the pre-cutoff date flow diagram (2323-MI-0206, Re CP-5), and found that the 3/8 in, flow restrictor was located in the Class 3 portion of the line, thus surviving a seismic event. This design would have met the code requirement if an analysis had verified that the 3/8 in flow restrictor sufficiently minimized loss of service water. The team reviewed the Brown and Root piping isometric drawing team (BRP-AF-1-SB-005 Rev. 11),

which indicated that the 3/8 in. flow restrictor is apparently installed in the Class 5 portion of the lin w Open Item M-18 On drawing TNE-MI-0206-01 Rev. CP-1, a normally open low point leakoff valve (1AF-020) is shown to provide the barrier between Class 3 and Class 5 pipin As in the case for Open Item M-18, a 3/8 in. flow restrictor is in the Class 5 portion of the line, in apparent violation of ANSI-18.2/ANS-51.8. Comments for Open Item M-18 on the pre-cutoff date flow diagram and the Brown and Root isometric apply equally in this cas Open Item M-19 Piping class breaks at both the high and low leakoff valves referred to in Open Items M-18 and M-19 were not indicated on the drawing (TNE-MI-0206-01 Rev. CP-1). This should be resolved along with these open item Review Topic: M03 - Heat Removal Representativeness of Reviewed Documents:

HDA-M006.0 Pressure Drop / Fluid Systems contains 37 calculation TERA scanned five of these: 233-18, 232-15, 229-13, 241-2, and 553. The team scanned the same five and found the basic methodology to be similar in the use of reference documents such as Crane 410, flow diagrams and composite ~

piping drawings. The team scanned three calculations not scanned by TERA, and found the methodology similar to that for the five scanned by TER One of the calculations (241-2) contained NPSHa calculations. This calcu-lation should also be included in an NPSH HDA. TERA should determine whether such inclusion affects the DAP-21 implementation for that HD Open Item M-20 Three of the five scanned calculations were performed in 1979, one in 1980 and one in 1983. The 37 calculations in the HDA were produced between 1975 and 1984 (there were none for 1977). Therefore, the calculations scanned do not satisfy the DAP-21 requirement for equal time interval Open Item M-21

- 14 -

_ _ _ . _ .-

r

.

. Several of thm 37 documents in HDA-M006.0 were for Unit 2. The team reviewed five of these covering the CCW, SSW, CT Spray, RHR and Boron Thermal Recycle

. systems and found that they were not actual calculations. Instead they stated that a Unit 1 calculation can be used for Unit 2 application, without providing any justification with respect to physical similarity between the two units for the systems involved. Physical comparisons need to be made and a technical analysis of any differences should be performe Open Item M-22 Review Topic: M30 - Seismic Qualification Representativeness of Reviewed Documents:

TERA developed a list of mechanical and electrical equipment, which included valves, pumps, tanks, motor control centers, limit switches, electric motors, steam driven turbines, pressure switches, electrical cabinets, relays, and switchgear. From this list, they selected for review 31 items based on the following factors: safety significance; ability to extrapolate; uniqueness; interfaces to other HDAs and Review Topic M43; known industry problems; and design change history. The list of items for review includes 10 I&C components based on TERA's commitment documented in IE inspection report 50-445/85-17, 50-446/85-14 dated January 21, 1986. It also includes the plant's class 1E batteries in response to NRC concerns expressed during the IE inspection of DSAP XI (during the period from July 7-17,1986). The team evaluated the list of equipment selected for review, and determined that it represented a reasonable cross section and that it satisfies the above commitments and NRC concern .

The intent of this HDA was to assess the architect engineer / vendor interface for seismic qualification. While the methodology for qualifying the hardware varies depending upon the type of component and the vendor, the methodology for this interface is relatively homogeneous and not sensitive to specific components and vendors. The team therefore considers the review sample to be represen-tative for purposes of assessing the HD Review of Documents:

The team reviewed the following checklists:

'

. Checklist Equipment DAP-CLC-M-034 Fuel Oil Day Tank DAP-CLC-M-043 Steam Supply Valve in Turbine Driven Pump DAP-CLC-M-044 Steam Generator Isolation Valves DAP-CLC-M-046 Motor Driven Chilled Water Recirculating Pumps DAP-CLC-M-035 Recycle Holdup Tank Gibbs and Hill apparently sent design information to vendors which effectively revised design ~ specifications, but Gibbs and Hill did not actually revise the design specifications. TERA ensured that the seismic requirements imposed by the specification, including the effect of DCAs, were met by the vendor. Separate reviews will be performed by TERA to ensure that the effect of design or field changes not reflected within the specification and interfaces with other

- 15 -

i . _ _

T

.

disciplines were evaluated by Gibbs and Hil Evidence of these programs is shown in (1) DIR D-0089 End DAP-RFI-M-0052 concerning documentation to verify the as-built condition is qualified, and (2) TERA memo DAP-M-191, March 21,

1986, which outlines the program for the final as-built load evaluatio The team will follow-up on this TERA effor Open Item M-23 A computer program (ANSYS) was used to analyze the fuel oil day tan The com-puter run, which consists of the finite element representation of the compo-nent, the loadings and the results (deflections, forces, moments and stresses),

was referenced in the vendor qualification report, but was not reviewed by TER The team considers that TERA should evaluate the computer modeling, loadings and results in order to reach a conclusion as to whether the vendor qualification satisfied the design specificatio Open Item M-24 TERA identified a generic concern during the review of motor driven chilled water recirculation pumps that all plant valves had been assumed to be oriented along the plant coordinate axes system for purposes of seismic qualificatio (Ref. DAP-M-191 dated 3/21/86). This assumption may be unconservative since there may be different actual orientations of valves that, when seismic design accelerations are applied, result in greater stresses than those based on the assumed orientation.' TERA identified other generic concerns with respect to compatibility of piping analysis as-built loads with the valve qualifications packages: the need to verify consistency between the assumed modeling in the

, vend,r document and the corresponding piping analysis package; consideration of the effects of non-rigid (less than 33hz) valve assemblies; review of actual vs. assumed valve operator supports; and interference assessment (adequacy of interaction space around valve installation). The above concerns are documen-ted in DAP-RFI-0052 dated 5/5/86. TERA plans to review the valve qualification packages after these concerns are addressed. The team will follow-up on this TERA effor Open Item M-25 Seismic qualification checklists did not, for all appropriate cases, address stresses due to operating loads, nozzle load capacity verification, and code i

required load combinations, and verification that manufacturers installation requirements (e.g. bolt torquing requirements for anchorages) are followed or differences resolve .. Open Item M-26 Review Topic: M41 - External Flooding Representativeness of Reviewed Documents:

TERA reviewed all building layout design drawings within HDA M093.0, Flooding-Protection Measures, to assure the commitments established in FSAR Section 2.4.10 were satisfie ,

Review of Documents:

The team reviewed Checklist DAP-CLC-M-015, and found the TERA review of this activity to be complete. DIR D-0231, concerning the Probable Maximum Flood Level (established at Elevation 789.7) questioned whether wind effects, which

- 16 -

._ _ _ _ . _. _

,

raised the ficed level to Evaluation 794.7, would cause any problems. TERA concluded that the increased flood level should cause no problems since no doors or openings are. located below this elevation, all building joints have

'

water stops, and sumps within buildings control leakage. The team confirmed the TERA conclusio The team reviewed the fuel oil storage tank design. This tank, located out-side the plant buildings, is buried and thus subject to possible uplift due to flooding. The team's review confirmed that the tank design had been con-servatively evaluated and found acceptable for uplift as well as other design parameter Review Topic: M43-Balance of Plant Pump and Valve Operability Representativeness of Reviewed Documents:

The pump and valve operability (PVO) assessment was initiated as a result of an incident where stress limits used by a vendor did not comply with those required by the valve design specification. TNE coordinated a review by Westinghouse, Corporate Consulting and Development Co. Limited (CCL) and Gibbs and Hill of active pumps snd valves to ensure that (1) all active pumps and valves were identified and classified and (2) their qualification documentation was in accordance with regulatory criteria. TERA overviewed this PVO assessment -

by using Checklists DAP-CLC-M-009 through 01 TERA issued DIR Nos. D-0187 through 0189 to identify the fact that active pump and valve lists generated in the PVO assessment were not complete, e.g. these lists omitted 8 safety-related active pumps and 9 safety-related active valve CCL, Westinghouse, Gibbs and Hill, and TNE reviewed the technical adequacy of qualification documentation for all active, safety-related pumps and valves on the above list. TERA reviewed three active pumps and three active valves from the lis The team considers the TERA review sample to be representative because it was structured to provide a mix of different vendors and specifications and both mechanical and electrical equipmen Review of Documents:

The demonstration of PVO is enveloped by seismic equipment qualification requirements, and is dependent on operability demonstration for associated electric'al components. TERA's PVO overview focused on assuring that vendor l qualification documentation was consistent with FSAR commitments. More

! detailed TERA review of the adequacy of vendor analyses for pumps, valves, and associated electrical components will be accomplished in TERA's Seismic Qualification Engineering Evaluation (Review Topic M30).

t Technical reviews for the PVO assessment identified no discrepancies for pumps which affected their ability to perform safety-related functions, i

Discrepancies for valves were identified in DIR Nos. D-190'and 191. The team had no further questions.

,

- 17 -

!

l l

l

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