ML20236C829
| ML20236C829 | |
| Person / Time | |
|---|---|
| Site: | Fermi  |
| Issue date: | 06/29/1987 |
| From: | Johnson G, Schultz E LAWRENCE LIVERMORE NATIONAL LABORATORY |
| To: | |
| Shared Package | |
| ML20236C832 | List: |
| References | |
| NUDOCS 8707300209 | |
| Download: ML20236C829 (20) | |
Text
f T
PREIMPLEMENTATION AUDIT REPORT FOR THE DETAILED CONTROL ROOM DESIGN REVIEW FOR DETROIT EDIS0N COMPANY FERMI-2 E. E. Schultz, Jr.
G. L. Johnson Lawrence Livermore National Laboratory June 29, 1987
~
l l
e go g 3,0 0 2 0 k N L
J
(
f 1
l I
PREIMPLEMENTATION AUDIT REPORT FOR THE DETAILED CONTROL RODM DESIGN REVIEW DETROIT EDIS0N COMPANY FERMI-2 1
- 1. BACKGROUND Licensees and applicants for operating licenses 'shall conduct a Detailed Control Room Design Review (DCRDR).
The objective is to " improve the ability of nuclear power plant control room operators to prevent accidents or cope with accidents if they occur by improving the information provided to them,"
NUREG-0660, Item 1.D.I.
(1).
The need to conduct a DCRDR was confinned in NUREG-0737 (2) and Supplement I to NUREG-0737 (3). DCRDR requirements in Supplement I to NUREG-0737 replaced those in earlier documents. Supplement 1 q
to NUREG-0737 requires each applicant or licensee to conduct a DCRDR on a schedule negotiated with the Nuclear Regulatory Commission (NRC).
l NUREG-0700 (4) describes four phases of the DCRDR, and provides applicants and licensees with guidelines for its conduct. The phases are:
~
1.
Planning l
2.
Review i
3.
Assessment and Implementation 4.
Reporting Criteria for evaluating DCRDRs are contained in Appendix A to Standard Review Plan, Section 18.1 of NUREG-0800 (5).
A Program Plan is to be submitted within two months of the start of the DCRDR. Consistent with the requirements of Supplement 1 to NUREG-0737, the Program Plan shall describe how the
'l following elements of the DCRDR will be accomplished:
1 1.
Establishment of a qualified multidisciplinary review team.
2.
Function and task analyses to identify control room operator tasks and information and control requirements during emergency operations.
l 3.
A comparison of display and control requirements with a control room inventory.
4.
A control room survey to identify deviations from accepted human factors principl es.
5.
Assessment of human engineering discrepancies (HEDs) to determine which HEDs are significant and should be corrected.
6.
Selection of design improvements.
DCRDRFermi:7/8/87: dim 1
(
\\
7.
Verification that selected design improvements will provide the necessary correction and do not introduce new HEDs.
8.
Coordination of control room improvements with changes from other programs such as SPDS, operator training, Reg. Guide 1.97 instrumentation, and upgraded emergency operating procedures (EOPs).
Licensees are expected to complete Element l' during the DCRDR's planning phase, Elements 2 through 4 during the DCRDR's review phase, and Elements 5 through 7 during the DCRDR's assessment and implementation phase.
Completion of Element 8 is expected to cut across the planning, review, and assessment and implementation phases.
A Sumary Report is to be submitted at-the end of the DCRDR. At a minimum, it shall:,
1.
Outline proposed control room changes.
'~
2.
Outline proposed schedules for implementation.
~-
3.
Provides sumary justification for HEDs with safety significance to be left uncorrected or partially corrected.
The NRC will evaluate the organization, process and results of the DCRDR.
Evaluation will include review of required documentation (Program Plan and Sumary Report), and may also include reviews of additional documentation, briefings, discussions, and on-site audits.
In-progress Audits may be conducted after submission of the Program. Plan, but prior to submission of the Summary Report. Evaluation will be in accordance with the requirements of Suppisnent I to NUREG-0737.
Additional guidance for the evaluation is provided by NUREG-0700 and NUREG-0800, Appendix A to SRP Section 18.1. Results of the' NRC evaluation of a DCRDR will be documented in a Safety Evaluation Report (SER) dr SER Supplement.
Significant HEDs should be corrected.
Improvements which can be accomplished with an enhancement program should be done promptly. Other control room improvements should be done on a schedule acceptable to the NRC.
2.
ASSESSMENT OF DCRDR ACTIVITIES A Preliminary Design Assessment for the Fermi-2 DCRDR was performed by the NRC Divisiun of Human Factors Safety (DHFS) and consultants from Lawrence Livermore National Laboratory (LLNL) and Bio Technology, Inc. on April 27 -
May 1, 1981.
The results of this activity were communicated to the NRC Division of Licensing in a safety Evaluation Report (SER) dated June 22, 1981 (6). Detroit Edison Company (DECO) submitted a Detailed Control Room Design Review (DCRDR) Program Plan for its Fermi Plant Unit 2 to the NRC in a letter dated August 16, 1984 (7). The DCRDR Program Plan was reviewed against the requirements of Supplement I to NUREG-0737 by the NRC Division of Human Factors Safety (DHFS) and consultants from LLNL. Their conynents were DCRDRFermi:7/8/87: dim 2
submitted to the NRC Division of Licensing in a memorandum dated December 5, f
1984 (8).
DECO submitted a DCRDR Summary Report for the Fermi Unit 2 Plant on November 30, 1986. This report covers the following:
- 1) the DCRDR Review Team structure and qualifications, 2) task analysis methodology, 3) task analysis data sheets, and 4) operator interview questionnaires.
A Preimplementation Audit was conducted by NRC and consultants from LLNL on March 16-20, 1987.
This Technical Evaluation Report provides LLNL's evaluation of the degree to which the Fermi-2 DCRDR fulfills the requirements of Supplement I to NUREG-i 0737, and identifies remaining actions required of DECO to achieve compliance. This evaluation is based upon DECO's Program Plan and Summary Report, tne SER based on the Preliminary Design Assessment, and material presented during the Preimplementation Audit.
2.1 DCRDR REVIEW TEAM 2.1.1 Requirement I
Supplement I to NUREG-0737 requires the establishment of a qualified
~
multidisciplinary review team. Guidelines for review team selection are found in NUREG-0700 and NUREG-0800, Appendix A to SRP Section 18.1.
NUREG-0700 guidelines state that support of the applicant's management is needed to provide the DCRDR team with all of the information, equipment, and categories of manpower needed to conduct a control room review, 2.1.2 Discussion Since 1981 the composition of the Fermi-2 DCRDR Team has, with the exception of the Team Leader, been in a state of flux.
A question about the composition of the DCRDR Team was documented in NRC's 1984 review of the Fermi-2 DCRDR Program Plan (9):
"An item of concern is whether the initial and current team members are the same (p. 4)."
During the Preimplementation Audit, the NRC Audit Team learned that various DECO staff and consultants have periodically met to address DCRDR issues, j
j Currently, a reassessment advisory team has been assembled to deal with classification of HEDs.
Deco stated that this team is going to serve as a dedicated team, and will be the decision-making body for future DCRDR
.~
i efforts.
However, the Preimplementation Audit Team was able to confirm neither that from 1981 to the present the Fermi-2 DCRDR activity has been directed by a cohesive, core DCRDR Team, nor that such a team currently exists.
i The reassessment advisory team consists of individuals with backgrounds in l
the disciplines of human factors engineering, operations, instrumentation and i
controls, and system engineering. While several team members have backgrounds 1
DCRDRFermi:7/8/87: dim 3
I a
i l
1 l
in nuclear operations, none has appropriate training and plant experience in j
the area of nuclear engineering.
The NRC Preimplementation Audit Team learned through discussions with operators, system engineers, etc. that access to relevant information, facilities, materials, and personnel is acceptable. For example, a mini-satellite information access area in the plant has been established, and a large set of current plant photos is.readily available to Fermi-2 staff and consultants.
This Audit Team also learned that a process for allowing and documenting dissent during DCRDR activity has been implemented. A space for recording dissent to the classification and resolution of HEDs is provided on HED forms.
j According to DECO, the DCRDR Team Leader gives an informal, verbal orientation of DCRDR activity to personnel who engage in DCRDR related activity. DECO stated that these personnel are also provided with three days of initial training, and some follow-up training on task analysis. Some of the personnel involved in the control room review effort attended two Boiling Water Reactor Owners Group (BWROG) control room survey workshops, one in 1980, and one in 1983. The 1983 workshop covered human factors principles. DECO also stated l
during the Preimplementation Audit that some personnel have received two and i
one-half days of training in the plant simulator and control room. Finally, DECO said that team training is an ongoing effort---copies of NUREG-0700 have been provided to personnel who participate in DCRDR activity, and training on NUREG-0700 provisions will be given in the near future, i
2.1.3 Conclusion DECO has not met the requirements of Supplement I to NUREG-0737 with respect to the composition of the DCRDR team and the provision of orientation to thi.s team. The com' position of the Fermi-2 DCRDR Team is not stable, and there is no nuclear engineer on this team. Most of the personnel who attended the BWROG workshops are no longer part of the DCRDR effort. Human factors orientation for new staff involved in the DCRDR effort has not yet, according to DECO, been provided.
In addition, DECO has not put into effect any type of systematic orientation program. Thus, it is unclear whether whatever orientation was provided at an earlier point in time corresponds to whatever future orientation is planned. The orientation provided to date has been too informal, and is not adequately integrated into the DCRDR program.
Furthermore, DECO could not produce documentation of the orientation process.
DECO should correct these deficiencies, and should document corrective actions j
in a Supplemental Sumary Report to be submitted to NRC by first refueling.
2.2 FUNCTION AND TASK ANALYSIS 2.2.1 Requirement l
Supplement I to NUREG-0737 requires the applicant to perform a function and task analysis to identify control room operator tasks and to identify control room operator information and control requirements during emergency DCRDRFe.mi:7/8/87: dim 4
/
operations.
Furthermore, Supplement I to NUREG-0737 recomends the use of function and task analyses that have comprised the basis for developing emergency operating procedures (EOPs), technical guidelines, and plant-specific E0Ps to define these requirements.
2.2.2 Discussion General Electric (GE) conducted the Fermi-2 System Function and Task Analysis for DECO.
GE based the task analysis upon Plant Specific Technical Guidelines (PSTG) which were derived from revision 4 of the BWR Emergency Procedure Guidelines. The task analysis lists each function which the PSTGs direct operators to perform.
For each function, specific operator tasks to accomplish the function were detailed.
The task listing was prepared without consideration of whether the task is controlled by E0Ps or normal plant j
i procedures. This approach resulted in analysis of all tasks needed to implement the E0Ps.
The tesk listings were reviewed to identify the information and controls needed for operators to perform each task.
The specific information requirements defined included:
Type of information (numerical value, logical value, alert, trend, o
record, or status indication) o Physical location of information i
o Parameter or variable needed Units in which parameter should be displayed o
l o
Range of display o
Display resolution o
Required alerting setpoint o
Needed calculation or decision tools
~
l Control needs defined included:
Type of control (discrete, automatic-continuous, manual continuous) o o
Physical location of control o
Parameter or equipment to be controlled o
Units of controlled parameter l
l DCRDRFermi:7/8/87: dim 5
J
i o
Control range o
Control resolution i
Communications needs were also defined whenever a task required communication between the control room and a remote location. The requirements specified included source and destination locations and the type of communication (one-or two-way).
The task analysis process did not specify the instrument accuracy required to
~
support operator actions The NRC Audit Team was unable to confirm that another activity had verified that available instrument accuracy is adequate for performance of emergency actions.
3 The NRC Audit Team examined sample task analysis data sheets for portions of the reactivity control PSTG.
This examination showed that, for the most part,
{
the task analysis was conducted as described. The Audit Team, however, noted two deficiencies in the implementation of the process.
Some operator functions were not broken down into specific tasks, and o
thus were not 6nalyzed.
For example, the task analysis did not decompose the function " Determine how many core spray pumps can be shut down" into discrete tasks.
Instead the analysis noted that this task is left to operator judgment. GE did not identify the information or tools needed for the operator to make this judgment.
A few task analysis data sheets were not fully completed.
o 2.2.3 Conclusion The System Function and Task Analysis activities conducted to date are acceptable. However, the task analysis process is incomplete. To completely satisfy the requirements of Supplement I to NUREG-0737 in this area Deco must:
Complete analysis of all operator functions which were omitted from the o
original analysis.
Verify that instrument accuracy will support performance of emergency o
actions. This analysis should consider the suitability of instrument accuracy under plant environmental conditions expected to accompany the emergency conditions in which the instrumentation is to be used.
Review the existing task analysis and complete the analysis for any o
applicable characteristics which were overlooked in the original analysis.
DECO should report the actions taken to address the above items and the results obtained from these actions in a Supplemental Sumary Report to be submitted to the NRC prior to start-up from the first refueling.
DCRDRFerm'.7/8/87: dim 6
2.3 COMPARISON OF CONTROL AND DISPLAY REQUIREMENTS WITH CONTROL ROOM INVENTORY 2.3.1 Requirement Supplement I to NUREG-0737 requires the applicant to make a control room inventory and to compare the operator display and control requirements determined from the task analyses with the control room inventory to determine the availability and suitability of controls and displays. Guidance in NUREG-0700 calls for a review of the human factors suitability of instruments and controls used to satisfy operator information and control requirements.
2.3.2 Discussion According to DECO, the comparison of control and display requirements with a l
control room inventory was accomplished through a control room walkthrough procedure. The evaluation team and control room operators reviewed each task analysis data sheet to ensure that every operator action was described correctly.
For every operator action requiring instruments and/or controls, the identification numbers of the equipment which would be used to perform the task were recorded to indicate that the necessary equipment is available.
This availability determination did not include verification that the equipment would be operable under adverse environmental conditions and partial loss of power that may be expected to accompany the emergency event (s) for which the. equipment is needed.
The walkthrough also recorded the actual instrument and control characteristics corresponding to the needs identified by the task an' lysis. Discrepancies between requirements (as identified by task analysis) and this inventory were recorded as HEDs, except when the evaluation team unanimously agreed that a discrepancy was acceptable. A follow-up comparison of control and display requirements with a control room inventory was also performed by human engineering consultants, who used the same methodology.
2.3.3 Conclusion Deco has, for the most part, met the requirements of Supplement 1 to NUREG-0737 in this area. To verify equipment availability, DECO determined whether needed displays or controls were present in the control room. Due to the incomplete state of the task analysis activity, the comparison of display and control requirements with a control room survey does not include a number j
of tasks and associated instruments and controls. After the task analysis activity is completed, Deco should reassess the availability of needed I
displays and controls addressing operability as well as the presence of the operator interface on the control board.
At this time, DECO should also finish the comparison activity for additional tasks identified during the completion of the task analysis. DECO should then describe results (including any resulting HEDs) in a Supplemental Summary Report to be submitted to the NRC prior to start-up from the first refueling.
DCRDRFermi:7/8/87: dim 7
I
---________u
a a
2.4 CONTROL ROOM SURVEY 2.4.1 Requirement Supplement 1 to NUREG-0737 requires that a control room survey be conducted to identify deviations from accepted human factors principles.
NUREG-0700 provides guidelines and criteria for conducting a control room survey.
)
2.4.2 Discussion
)
In 1981, control room surveys covering the following areas were conducted:
1.
Panel layout and design 2.
Instrumentation and hardware t
i
~
3.
Annunciators 4.
Computers 1
5.
Procedures 6.
Control room environment 7.
Maintenance and surveillance procedures 8.
Training and manning The basis for these surveys was checklists developed by BWROG.
Supplemental checklists based on survey items from NUREG-0700 which were not covered in the original BWROG checklists were prepared after the original surveys were conducted in 1981. According to DECO, conditional acceptance was given to a few survey items in which deficiencies were noted, depending upon successful implementation.
Deco also stated that a number of_ survey items deal with non-implemented or non-operational control room design features which were not implemented or were non-operational when the original and supplemental surveys -
were administered several years ago.
DECO plans to evaluate these survey items in the near future. DECO also has checked survey results through supplemental work, including standard operating procedure (SOP) walkthroughs and operator interviews.
Documentation of the results of this supplemental work was not available during the Preimplementation Audit.
1 Checklists were completed by an interdisciplinary team from different utilities.
Team members indicated one of five responses for each survey i
l item:
0 - Not applicable l
1 - Full compliance DCRDRFermi:7/8/87:dlm 8
____=
1 i
2 - Mostly compliant I
3 - Somewhat compliant 4 - Not compliant f
Each response indicated the overall degree of compliance or the relative number of deviations from requirements and guidelines. The degree of deviation of design features conforming least to requirements and guidelines was not necessarily considered.
However, if any deviation on a board was detected during the survey process, it was documented, then reviewed later.
During the Preimplementation Audit, the NRC Audit Team learned that there is no formal mechanism for tracking any of the following survey items:
1.
Items that could not be assessed during the BWROG surveys 2.
Items given conditional acceptance In addition, a number of changes to the control room have been implemented since surveys were originally conducted six years ago.
These changes are the result both of HED-related corrections and other factors.
However, there is no formal process for a human factors verification review of these control room changes. Also, the design process from 1981 to 1986 did not incorporate a formal mechanism to ensure that any design changes resulting from the survey process did not introduce new HEDs.
In short, there has been inadequate planning and management of the process of dealing with changes resulting from the control room survey process.
Finally, the NRC Preimplementation Audit Team performed its own mini-survey of control room HEDs for two panels of the control room.
Fifty-two HEDs relevant to control room display and control design were identified by this Audit Team (see Appendix A). Many of the HEDs fcund durin documented during the original survey process. g the mini-survey were not There appears, in fact, to be little relationship between the survey activity and any other DCRDR-related activity.
2.4.3 Conclusion 1
Deco has not met the requirements of Supplement 1 to NUREG-0737 with respect to the control room survey process.
The BWROG control room survey process is acceptable, but DECO's implementation of this process is not acceptable. The lack of tracking of the survey process, the failure to develop a formal mechanism to ensure that design changes resulting from this process do not result in new HEDs, and the lack of integration of the survey activity with other DCRDR activity all present a major concern about the survey process. To alleviate this concern, a core DCRDR Team should repeat the control room survey. LLNL suggests that the resurvey be conducted using the NUREG-0700 survey criteria instead of the BWROG checklist. DECO should then describe methodology used and results obtained in a Supplemental Sunrnary Report to be submitted to NRC prior to start-up from first refueling.
DCRDRFermi:7/8/87: dim 9
2.5 ASSESSMENT
OF HEDs 2.5.1 Requirement Supplement 1 to NUREG-0737 ' requires that HEDs be assessed to determine which HEDs are significant and should be corrected.
2.5.2 Discussion According to DECO, HEDs will be reviewed to determine that the observations which produced these HEDs are valid.
HEDs based on invalid observations will be eliminated. After the original control room surveys were conducted, HEDs were prioritized according to the product value.resulting from a multiplicative formula based on safety consequences and potential for error of each HED. Enhancements (paint, tape, and labeling) and Priority I HEDs (safety related and/or high error potential) were scheduled for correction by first refueling.
Priority II HEDs (safety related and low error potential) were
~
slated for near-term correction, by second refueling.
Priority III HEDs (non-safety related, high error potential) were scheduled for mediunt-term l
correction, by second refueling. Deco deemed the correctior of Priority IV HEDs (non-safety related, low error potential) optional.
During the Preimplementation Audit, DECO related that current HEDs will be classified according to an algorithm in which numerical products are no longer calculated.
HED priorities are determined according to the following logic:
Priority 1
- high error potential, high safety consequence-immediate fix or interim compensatory action, followed by fix by first refueling Priority II - low error potential, high safety consequence-fix by first refueling Priority III - high error potential, low safety consequence-fix by 1
i second refueling Priority IV - low error potential, low safety consequence-corrective action may be taken, or no action required if adequate rationale exists During the Preimplementation Audit, DECO stated that to assess cumulative and interactive effects, a panel-by-panel review will be undertaken.
Every HED i
(including Priority IV HEDs) related to a particular panel will be sorted.
i The assessment team will carefully examine unresolved HEDs to determine I
whether cumulative and interactive effects might result from the totality of these HEDs. DECO also related that special attention will be paid to panels which are considered more difficult or complex with respect to operator usability.
l Whereas the classification algorithm is acceptable, DECO has not yet provided a description of the actual procedures which will be used in the HED assessment process. The role of human factors personnel in the tracking of the assessment process is yet to be specified. Additionally, the role of a DCRDRFermi:7/8/87: dim 10
DCRDR Team in this process is unclear...This Team should lead the assessment effort, track schedules, etc., to provide the much-needt ontinuity of effort between DCRDR phases.
2.5.3 Conclusion There is insufficient information to determine whether the requirements of Supplement I to NUREG-0737 regarding assessment of HEDs have been fulfilled.
Deco's verbal descriptions of the assessment process are adequate as far as they have gone. However, these descriptions need to be documented in a Supplemental Sumary Report to NRC, and more complete descriptions of the assessment process should be provided.
In this. Report Deco should comit both to human factors tracking of the assessment process and to DCRDR. Team leadership of this activity. The Report should be submitted prior to_ start-up from first refueling.
2.6 SELECTION OF DESIGN IMPROVEMENTS 2.6.1 Requirement Supplement 1 to NUREG-0737 requires the selection of control room improvements that will correct significant HEDs.
It also states that improvements that can be accomplished with an enhancement program should be done promptly.
2.6.2 Discussion According to Deco, a methodology for the selection of design improvements will be developed by a special team. The DCRDR Team will then review this methodology before it is implemented. Although this methodology has yet to be developed, DECO indicated that the selection of design improvements will be a phased activity involving a number of steps.
Each HED has already been reviewed by the DCRDR Team Leader. Guideline checklists will be developed by two DCRDR Team Members, and will be reviewed by the entire team. These checklists will be used to guide the process of identifying potential corrections, including design changes. The potential design changes'will then be scoped on the basis of factors such as priority rating, cost effectiveness, and degree of correction needed. Design changes will then be incorporated
~
into a design package conceptual design to be given to Sargent and Lundy Engineering for implementation. DECO anticipated that the control process over these activities will be " rigid."
A panel-by-panel review is planned to address the requirement to consider cumulative and interactive effects in the selection of design improvements.
Any identified cumulative and interactive effects may, according to Deco, be assigned a higher priority than was originally given.
The schedule for implementing design improvements is satisfactory. Priority I HEDs as well as HEDs requiring enhancements will be corrected by first refueling, and Priority II and III HEDs will be corrected by second refueling.
DCRDRFermi:7/8/87: dim 11
2.6.3 Conclusion Again, there is insufficient information to allow determination of whether the process of selecting design improvements satisfies the requirements of NUREG-0700. Although incomplete, DECO's verbal description of the design improvement process indicates satisfactory commencement of planning for the selection process. DECO should document these verbal descriptions of this planning.
DECO should also commit to having a core DCRDR Team guide the selection process. Documentation of the seler' ion process should be included in a Supplemental Summary Report to be submitted to NRC by first refueling.
~
2.7 VERIFICATION THAT DESIGN IMPROVEMENTS PROVIDE NECESSARY CORRECTION AND 00 i
NOT INTRODUCE NEW HEDs
{
2.7.1 Requirement
~
Supplement 1 to NUREG-0737 requires verification that selected design improvements will provide the necessary correction, and will not introduce new HEDs into the control room.
1 i
~
2.7.2 Discussion DECO indicated during the Preimplementation Audit that a process for verification of design improvements has not yet been developed. Informal, i
undocumented verification reviews by several plant personnel have been conducted in the past, and a formal checklist to guide this process has been proposed.
However, no plans to conduct an independent verification effort have been proposed. Thus, it is possible that the same team which selects design improvements may conduct the verification effort. A more objective, independent verification activity would substantially enhance the quality of DECO's DCRDR effort.
{
2.7.3 Conclusion Due to lack of information, no conclusion can be reached about DECO's compliance with NUREG-0737, Supplement I requirements regarding verification of design improvements. DECO must, in a Supplementary Sunnary Report, document the verification process chosen. The verification process should be I
an independent process; the personnel who select design improvements should not be the same personnel who verify that design improvements result in necessary correction and do not cause additional HEDs.
l 2.8 COORDINATION OF THE DCRDR WITH OTHER PROGRAMS 2.8.1 Requirement Supplement I to NUREG-0737 requires that control room improvements be coordinated with changes from other programs, e.g., Safety Parameter Display System (SPDS), operator training, Regulatory Guide 1.97 (R.G.1.97) instrumentation, and upgraded emergency operating procedures.
DCRDRFermi:7/8/87: dim 12
1 1
2.8.2 Discussion 1
The NRC Preimplementation Audit Team learned that review meetings have been held at Fermi-2 to discuss coordination of DCRDR with other activities.
Personnel involved in the DCRDR effort did not write HEDs for differences in coding conventions between SPDS and the control room boards.
However, DECO maintained that.an effort to keep conventions consistent was attempted. A standard plant nomenclature has been developed and implemented through a i
procedural and labeling review. Discrepancies discovered during this review were recorded as HEDs.
However, the Preimplementation Audit Team noted that other than the area of nomenclature, there has been inadequate coordination between SPDS and DCRDR activities.
The lack of an identifiable, dedicated DCRDR Team has probably contributed to this lack of coordination.
Deco also indicated that an informal feedback loop between training, procedures, and DCRDR has been established.
The Preimplementation Audit Team indicated that a more formal process for coordinating this feedback is needed. Again, the guidance of a core DCRDR Team in this area is essential.
~
A Fermi-2 staff member assisted by operations and engineering personnel reviewed E0Ps for conformance to R.G. 1.97.
According to DECO, GE consultants will informally verify that specific tasks in the task analysis correspond to Revision 4 E0Ps.
The intent of these activities is to ensure that there is a one-to-one correspondence between the task analysis and E0Ps. The NRC Preimplementation Audit Team noted, however, that the task analysis is a very large undertaking. Without a formal process for comparing the task analysis to E0P revisions, it. 's difficult to assure that the task analysis and E0P upgrades will correspond to each other.
2.8.3 Conclusion Deco's process for the coordination of control room design improvements with control room changes resulting from other improvement programs is unacceptable, and does not meet the requirements of Supplement 1 to NUREG-0737.
DECO should address the concerns noted in Section 2.8.2.
Then, in a Supplemental Summary Report, DECO should document the modified coordination 1
process for NRC review.
3.
CONCLUSIONS i
Detroit Edison Company has not achieved an satisfactory degree of p'rogress in the Fermi-2 Detailed Control Room Design Review effort since the Preliminary Design Assessment.
DECO's DCRDR of the Fermi Plant, Unit 2, does not fulfill most of the DCRDR requirements of Supplement I to NUREG-0737.
The lack of a dedicated DCRDR Team with stable composition and a clear mandate from DECO management to lead the DCRDR process has been the greatest obstacle to fulfillment of these requirements.
DCRDRFermi:7/8/87: dim 13
Deco must respond to numerous unresolved specific issues and concerns.
To close out these matters, DECO should complete the following prior to start-up from first refueling for Unit 2:
1.
Take actions necessary to achieve full compliance with the provisions of Supplement 1 to NUREG-0737 concerning a DCRDR Team and proper orientation for this Team.
2.
Complete the function analysis of operator functions not yet analyzed.
3.
Verify that instrument accuracy is sufficient for the performance of E0Ps, including performance under adverse plant conditions.
4.
Review the existing analysis for missing characteristics, and complete the task analysis accordingly.
m 5.
After items 2-4 have been completed, complete the comparison of control and cisplay requirements with a control room inventory.
6.
Conduct control room surveys, using NUREG-0700 as the basis for guidelines.
7.
Completely specify the process to be used in the assessment of HEDs.
Human factors tracking of this activity should be incorporated into the l
proces.s.
8.
Completely specify the process to be used in the selection of design improvements.
9.
Completely specify the process to be used in the verification of design improvements.
Independent verification of design improvements should be f
ensured by whatever process is adopted.
{
Develop a more suitable process for coordination of DCRDR activity with
.10.
other activities, particularly SPDS.
Develop a formal process for coordinating DCRDR with feedback obtained from training, and for ensuring -
consistency between E0Ps and task analysis.
Documentation of actions taken to resolve issues and concerns arising from the I
NRC Preimplementetion Audit should be written in sufficient detail to allow NRC to make a judgment concerning DECO's fulfillment of requirements contained in Supplement 1 to NUREG-0737 concerning DCRDR. This documentation'should be incorporated into a Supplemental Sumary Report to be submitted to NRC prior to start-up from the first refueling.
DCRDRFermi:7/8/87: dim 14
REFERENCES 4.1 GENERAL REFERENCES 10.0.1 10.1 NUREG-0660, "NRC Action Plan Developed as a Result of the TMI-2 Accident," May: 1980, Rev.1, August 1980.
10.2 NUREG-0737, " Clarification of'the TMI Action Plan Requirements," November 1980.
l 10.3 NUREG-0737, " Clarification of TMI Action Plan Requirements," Supplement 1, December 1982.
10.4 NUREG-0700, " Guidelines for Control Room Design Reviews," September 1981.
10.5 NUREG-0800, " Standard Rsview Plan for the Review of Safety Analysis Reports for Nuclear Power Plants,"- Sec.18.1, Appendix A. " Evaluation Criteria for Detailed Control Room Design Reviews," September 1984.
~
10.6 "SER Input for Enrico Fermi-2 Control Room Design Review," letter from S.H. Hanover to R.L. Tedesco, June 22, 1981.
10.7 " Control Room Design Review Program Plan for Fermi-2." Detroit Edison Electric Company,- August 16, 1984.
10.8 "Revied of the Fermi-2 Detailed Control Room Design Review," letter from-W.T. Russell to T.M. Novak, December 5,1984.
10.9 " Comments on the Fermi-2 Power Plant Detailed Control Room Design Review Program Plan," Lawrence Livermore National Laboratory, October 16, 1984.
DCRDRFerini:7/8/87 : dim 15
-4 APPENDIX A j
(Control Room Mini Survey HEDs)
FERMI 2 DCRDR AUDIT NRC CONTROL ROOM. SURVEY FINDINGS 19 MARCH 1987 PANEL P-601 SURFACE C RHR Pump C Discharge Pressure Indicator.
Green pointer does not extend to near scale.
l
~
Hayes vertical ~ panel meters Major and minor graduations not used on meter scales.
Reactor Vessel Pressure ard Reactor Vessel. Water Level Recorder Label position below recorder is obscured for tall users.
I RHR Loop Flow Recorder Scale faded and hard to read.
Recorder paper does not match scale.
Torus Wide Range Water Level Recorder Major and minor graduations not used on scale.
SURFACE B i
Several keylock and one reset switch Functional description not provided on label Examples:
E1150-F610A; E11-F016A; F016 Seal In ADC Countdown Timer Controls not needed or used by operator Examples: ST/SP Reset; Ent Limit; Thumbwheel Switch Reactor Building Leak Detector Controls and indicators not used or needed by operator, system abandoned in place.
DCRDRFermi:7/8/87: dim 16
Surface A ECCW Temperature Controller Three indicator scales not labeled -- appear to be valve position, demand setpoint, and error signal.
Position indication does not correspond to valve position.
Operator stated that valve ist fully closed when indicator reads about 15%
open. Only way to verify full closed is to drive valve closed manually until movement stops.
~
RCIS Keylock Switches Poor contrast on switch position ' labels, i
Motor Current Panel Meters Meter zero adjustment exposed and prone to inadvertent operation.
e t
Left Side of Panel Unfilled holes in panel.
RCIC Turbine Test Mode Signal Generator Potentiometer I
Control not needed or used by operators.
Standby Feedwater Pump A & B Control Switchts Handwritten enhancements to labels.
RCIC Turbine Pump Discharge Flow Controller Handwritten temporary operator aid label.
RHR Pump C Control Switch Flashing red light indicates switch not in normal position rather than an alarm condition.
GENERAL COMMENT
One operator commented that non-glare glass on recorder faces inhibited accurate reading of recorder scales and recorder trace. Audit team noted numerous recorders open during operation apparently to aid reading, j
Same key fits all control room keylock switches; therefore, keylocks do not
)
protect against inadvertent operation of incorrect device.
i 1
DCRDRFermi:7/8/87: dim 17
c i
PANEL P-805 SURFACE A Units are not indicated on Reactor Feed Pump Turbine Oil Coolers Temperature.
Control Gauge Units are not indicated on Hotwell Coil Steam Supply Control Gauge.
Two gauges above heater Drains Start-up LCV lack labels and units (there are also handwritten enhancements to labels).
No set point label for Extraction STM from HP Turbine.
Inadequate identification of buttons marked with X'd circle.
Too many temperature tags.
It is not clear what the reset button does.
Banding of Normal Hotwell Supply N20-F407 Gauge makes scale gradations impossible to read.
SURFACE B
{
Inappropriate color coding for System No.1 Controlling (red) and System No.1 Failure (white) (RFP Turbine N&S).
Tach A Verify Mushroom Knob Controls lack position indicators (RFP Turbine N& S).
Inadequate. labeling for reset and trip buttons (RFP Turbine N&S).
The auto setting on switches has inconsistent color coding (sometimes red, sometimes white).
It is unclear which switch's open/ closed status is represented by each pair of indicators (RFP Turbine N&S).
No banding on Valve. Position Diff. Pressure Control P43-F405, Low Pressure Stop Valve Position, Turbine N&S Demand Valve Position, Reactor Feed Pump Recirculation H21-F400A, and many others.
Inadequate labeling in TBCCW from Cndr. Vacuum Pumps Throttle P43-F605.
Incomplete mimics, Extraction STM from HP Turbine (4th demarcation from left).
Handwritten enhancements to mimics.
Two test buttons for autostart capability in RFP Turbine C have identical labels.
Inadequate identification of buttons marked with X'd circle.
Ambiguous mimics for n RFP Recirculation, Line 1--does arrow mean end of path?
Identical labels used for Block Status Indicators (RFP Turbine S).
SURFACE C Unit label (degrees F) on east outlet of N&S Hotwell Temperature Indicators is difficult to read.
Poor readability of output for N&S RFP Turbine and Pump Bearing Metal Temperatures.
Positive and negative scale readings for Hotwell and Condenser Instrument are easy to confuse.
DCRDRFermi:7/8/87: dim 18
's ~
d Inappropriate control-instrument arrangement for N&S Cond.
Hotwell Narrow and-Wide Range Level Gauges and Selectors.
Demarcation-needed for N&S Cond. Hotwell Narrow and Wide Range Level Gauges and Selectors.
Recorder labels are partially occluded by bottom of recorder boxes.
Three-pen recorder plots overlap too much.
Green banding is temporary; is falling off (re. Reheater Seal Tank Level, Separator Seal Tank N Level, Separator Seal Tank S Level, Heater 2C, Heater IC, and Seal Water Return Tank Level).
Green banding appears blue, rather than green.
Inadequate labels for RFP Turbine N&S 011 Coolers Temperature Recorders (handwritten enhancements to labels).
Many circular gauge pointers are red, but there are red-banded areas on gauges.
OTHER Phones and plugs protrude from front edge of boards--could be inadvertently disconnected by operators who walk by.
Lighting above boards causes blinding glare.
DCRDRFermi:7/8/87: dim 19 u_