ML20135G184
| ML20135G184 | |
| Person / Time | |
|---|---|
| Issue date: | 09/13/1985 |
| From: | Harold Denton Office of Nuclear Reactor Regulation |
| To: | Heltemes C NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD) |
| Shared Package | |
| ML20132C684 | List: |
| References | |
| NUDOCS 8509180257 | |
| Download: ML20135G184 (12) | |
Text
S R 1 U985 MEMORANDUM FOR:
Clcmens J. Haltimes, Jr., Director Office for Analysis and Evaluation of Operational Data FROM:
Harold R. Denton, Director Office cf Nuclear Reactor Regulation
SUBJECT:
REVIEW 0F'AEOD REPORT The Office of Nuclear Reactor Regulation has reviewed the draft case study report " Decay Heat Removal Problems at U.S. Pressurized water Reactors," as requested in your memorandum of July 5,1985.
The draft report appears to be a good and comprehensive survey of operating experience with RHR systems.
We are in agreement with your conclusions concerning the importance of human error in the loss-of-DHR events as well as emphasizing the importance of improved outage planning.
In addition, we believe that the report may serve as an imp 6rtant input to the goals and objectives of the upcoming Maintenance and Surveillance Program by identifying the significance and frequency of human errors and the shortcomings of maintenance operations as precursors to DHR events.
Three divisicns within NRR are providing detailed comments (Enclosure 1) which we believe will provide improved perspective on the safety significance of the problems that have occurred as well as the potential safety value of your recommendations.
In particular, the Division of Human Factors Safety addresses your recommendations concerning "the need for NRC requirements to improve planning, coordination, procedures and personnel training during shutdown to ensure availability of DHR."
Comments from the Division of Safety Technology address the need to perform a quantitative risk and cost analysis to determine the optimum balance between the risks of an interfacing LOCA-Event V and a loss of RHR. The Division of Systems Integration suggests that there is insufficient information to justify the need for more instrumentation, particularly liquid level instrumentation, and notes that the potential for core uncovery in one hour as a result of loss-of-DHR may be overemphasized.
The Division of Systems Integration presents additional support for its position on Auto Closure Interlocks (ACI) in Enclosure 2.
W 2 ped by E E aoma Harold R. Denton, Director Office of Nuclear Reactor Regulation
Enclosures:
DISTRIBUTION YELLOW TICKET 859169 1.
NRR Division Comments Centraldile w/incomingg 2.
Additional Background NRC PDR w/ incoming ^'
GHolahan RWessman NSIC w/ incoming WSwenson cc w/ enclosures:
See next page ORAB Rdg HSmith HDenton/DEisenhut L. Mulley%
CONTACT:
PPAS W. Swenson, NRR H1hompson/FMiraglia X27876 Connie /DCrutchfield
- PREVIOUS CONC E SEE DATE t fy) f D/
D/N ORAB:DL*
B:DL C:0RAB:DL A WSwenson:cl essman GHolahan tchfield h
on D
ut HDe n
f 08/29/85 f/30/85
}C/fp/85
/ o /85 q /i U85
/85 85
/
~
8509180257 850913 PDR MISC 8509180257 PDR
p MEMORANDUM FOR:
C1emens J. Heltemes, Jr.
Director Office for Analysis and Evaluation of Operational Data FROM:
Harold R. Denton, Director Office of Nuclear Reactor Regulation
SUBJECT:
REVIEW 0F AE0D REPORT The Office of Nuclear Reactor Regulation has reviewed the draft case study report " Decay Heat Removal Problems at U.S. Pressurized water Reactors," as requested in your memorandum of July 5,1985. The draft report appears to be a good, as well as comprehensive survey of operating experience with RHR systems.
We are in agreement with your conclusions concerning the importance of human error in the loss-of-DHR events as well as emphasizing the importance of improved outage planning.
In addition, we believe that the report may serve as important input to the goals and objectives of the upcoming Maintenance and Surveillance Program by identifying the significance and frequency of human errors and the shortcomings of maintenance operations as precursors to DHR events.
Three divisions within NRR are providing detailed coments (Enclosure 1) which we believe will provide improved perspective on the safety significance of the problems that have occurred as well as the potential safety value of your recommendations.
In particular, the Division of Human Factors Safety addresses your-recommendations concerning "the need for NRC requirements to improve planning, coordination, procedures and personnel training during shutdown to ensure availability of DHR."
Comments from the Division of Safety Technology address the need to perform a quantitative risk and cost analysis to determine the optimum balance between the risks of an interfacing LOCA-Event V and a loss of RHR. The Division of Systems Integration believes that there is insufficient information to a stify the need for more instrumentation, particularly liquid level instrumentation, and notes that the potential for core uncovery in one hour as a result of loss-of-DHR may be overemphasized. The Division of Systems Integration presents additional support for its position on Auto Closure Interlocks (ACI) in Enclosure 2.
Harold R. Denton, Director Office of Nuclear Reactor Regulation
Enclosures:
DISTRIBUTION YELLOW TICKET 859169 1.
NRR Division Comments Central File w/ incoming GHolahan 2.
Additional Background NRC PDR w/ incoming RWessman NSIC w/ incoming WSwenson cc w/ enclosures: See next page ORAB Rdg HSmith HDenton/DEisenhut CONTACT:
PPAS W. Swenson, NRR Connie /DCrutchfield HThompson/FMiraglia X27876 ORA SL:0RAB:DL C:0RAB:DL AD/SA:DL D/DL WSwdnson:cl RWessman GHolahan DCrutchfield HThompson ogg/85
/ /85
/ /85
/ /85~
_ / /85 DD/NRR D/NRR DEisenhut HDenton
/ /85
/ /85
SEP 1'3'1985 MEMORANDUM FOR:
Clemens J. Halt;mes, Jr., Director Office for Analysis and Evaluation of Operational Data FROM:
Harold R. Denton, Director Office of Nuclear Reactor Regulation
SUBJECT:
REVIEW 0F~AEOD REPORT The Office of Nuclear Reactor Regulation has reviewed the draft case study report " Decay Heat Removal Problems at U.S. Pressurized water Reactors," as requested in your memorandum of July 5, 1985.
The draft report appears to be a good and comprehensive survey of operating experience with RHR systems.
We are in agreement with your conclusions concerning the importance of human error in the loss-cf-DHR events as well as emphasizing the importance of improved outage planning.
In addition, we believe that the report may serve as an important input to the goals and objectives of the upcoming Maintenance and Surveillance Program by identifying the significance and frequency of human errors and the shortcomings of maintenance operations as precursors to DHR events.
Three divisions within NRR are providing detailed comments (Enclosure 1) which we believe will provide improved perspective on the safety significance of the problems that have occurred as well as the potential safety value of your recommendations.
In particular, the Division of Human Factors Safety addresses your recommendations concerning "the need for NRC requirements to improve planning, coordination, procedures and personnel training during shutdown to ensure availability of DHR."
Comments from the Division ( f Safety Technology address the need to perform a quantitative risk and cost analysis to determine the optimum balance between the risks of an interfacing LOCA-Event V and a loss of RHR. The Division of Systems Integration suggests that there is insufficient information to justify the need for more instrumentation, particularly liquid level instrumentation, and notes that the potential for core uncovery in one hour as a result of loss-of-DHR may be overemphasized.
The Division of Systems Integration presents additional support for its position on Auto Closure Interlocks (ACI) in Enclosure 2.
Ori$g send by
, E R.Denton Harold R. Denton, Director Office of Nuclear Reactor Regulation
Enclosures:
DISTRIBUTION YELLOW TICKET 859169 1.
NRR Division Comments Central File w/ incoming GHolahan 2.
Additional Background NRC PDR w/ incoming RWessman NSIC w/ incoming WSwenson cc w/ enclosures:
See next page ORAB Rdg HSmith HDenton/DEisenhut L.Mulley%
CONTACT:
PPAS W. Swenson, NRR HThompson/FMiraglia l
X27876 Connie /DCrutchfield
- PREVIOUS CON E SEE DATE f
D/
D/Ng ORAB:DL*
8:DL C:0R :DL A WSwenson:cl essman
'GHolahan tchfield on D
ut HDepon 08/29/85 f/30/85
}c/3p/85 85
/85
/ o/8; q/i(/85
/
l l
f e
O e
ENCLOSURE 1 1
l
DIVISION OF HUMAN FACTORS SAFETY l
COMMENTS ON THE DRAFT REPORT f
" CASE STUDY REPORT--DECAY HEAT REMOVAL l
PROBLEMS AT U.S. PRESSURIZED WATER REACTOR" i
i
)
1.
The report makes several recomendations (e.g., p. 56-61) based upon the l
potential safety significance of loss of decay heat removal (DHR) f events. A number of these reconsnendations concern maintenance. The report states "AE00 recomends that NRR assess the need for NRC requirements to improve planning, coordination, procedures and personnel training during shutdown to ensure the availability of DHR." We agree l
with the conclusions, but recommend that the report emphasize the role of outage planning as an important element of preventing DHR events in j
addition to proper accomplishments of corrective and preventive maintenance activities. The report contains ample information for drawing such conclusions for example:
(1) several events presented in y
the report are caused by poor outage planning and lack of coordination l
in ongoing test, surveillance and maintenance activities (e.g., App. A j
and B); (2) " discussions with plant personnel and inspectors, and the j
review of operating data indicate that the techniques used for planning i
and coordination vary widely from plant to plant and are frequently inadequate" (but inadequate is not defined in the report, p. 50); and f
(3) the DHR losses.at the Crystal River Plant "seem to have stopped at l
about the same time that the plant implemented actions to improve their l
j plantnng, coordination and management of outage and maintenance l
activities"(e.g.,p.26-27). This type of information will provide useful input to the Maintenance and Surveillance Program Plan report at the conclusion of Phase I.
j 2.
The report presents the consequences of the loss of the DHR function j
(p. 16-18, Fig. 4), and it also points out that the time margin i
available for restoring DHR or establishing alternate methods of heat i
removal is frequently short. Down-times are usually governed by the l
availability of parts, assemblies, or components. Therefore, the limiting factor in such situations would be the availability of spare
.{
parts rather than maintenance personnel. We recommend that the report j
consider spare parts management for DHR components.
a
DIVISION OF SAFETY TECHNOLOGY COMMENTS ON THE DRAFT REPORT
" CASE STUDY REPORT--DECAY HEAT REMOVAL PROBLEMS AT U.S. PRESSURIZED WATER REACTORS" i
This report appears to be an excellent comprehensive survey of operating experience with RHR systems. However, the report lacks perspective on the safety significance of the events that have occurred and the potential safety value of the recommendations.
The report correctly identifies the dilemma in RHR system design as a balance between the risks of an interfacing system LOCA and the loss of RHR. The only means of solving this dilemma is to perform a quantitative risk and cost analysis to determine the optimum balance. The reasons given in the report for not doing such an analysis are the large uncertainty associated with such assesgments and the extensive effort required. However, withoutsuchananalysisthereisnobasisotherthanjudgmentforthe recommendations presented. While judgment takes less effort, it does not decrease and more likely increases the uncertainty. Without a quantitative analysis a balance can not be made and there is inadequate assurance that the recommendations are necessary or even worthwhile.
For example, one recommendation is to remove the autoclosure interlocks or the electrical power for the RHR suction isolation valves. We have done a rough quantitative analysis of this issue which indicates that it is only marginally important (i.e., the core-melt frequency due to these interlocks is about SE-6/ plant year and a reduction in risk of about 300 man-rem / plant).
Therefore only a relatively inexpensive resolution might be justified (e.g.
'less than $300,000/ plant). While the AEOD recommendation might be inexpensive, the total effect on risk was not determined.
Removing the inter 10cks or power would delay closure of the valves and subject the plant
~
to a higher risk from a leak or break in the RHR system (an intersystem LOCA). Other solutions were not considered.
Improving the reliability of the interlocks is possible. Why is it not recommended? Increasing the l
~
~
)
2-availability of the auxiliary or main feedwater systems or the charging pungs by imposing an LCO during Modes 4, 5 and 6 is another possible solution.
Without a risk and cost analysis there is little basis for selecting one solution (or even any solution) over another.
AE00 appesrs to be relying en NRR to do the necessary risk and cost analyses needed to justify their recommer.dations. However, this results in a delay in resolving these issues and a duplication of effort. AE00 has the detailed knowledge of the issu6 and could do the risk analysis more quickly and easily than NRR. In order to do an analysis, NRR rust first become familiar with the issue which introduces delay and duplication of what AE0D did.
We recommend that AE00 do a rist and cost analysis of this issue before they send their recommendations to NRR If AEOD wishes to rely on NRR to do the risk and cost analyses, the recommendations should be deleted from the report or at least noted as being only possible suggestions needing further evaluation. Issuing reports with recommendations that require extensive further effort to evaluate gives the improper impression that NRR is dilatory with respect to AEOD recomendations.
e M
i I
t
I DIVISION OF SYSTEMS INTEGRATION COMMENTS ON THE DRAFT CASE STUDY REPORT
" DECAY HEAT REMOVAL PROBLEMS AT US PRESSURIZED WATER REACTORS" Comments from Core Performance Branch Section 3.0 - Operational Experience The data on DHR system failures emphasizes the frequency and duration of events.
Subsequent material cites instrumentation and/or surveillance deficiencies as a major contributor to these events.
Although a few examples are given, there is insufficient information presented.to support the hypothesis that the duration of these events implies a need for more instrumentation, particularly liquid level instrumentation. ' Data regarding the means and timing of the initial recognition of loss of DHR function would be more pertinent.
It is probable that many of the events were recognized early but there was no urgency perceived for proceeding to an abnormal backup cooling mode while steps were taken to recover the normal cooling mode.
Section 4.0 4.3.1 - Mode definition - Average coolant temperature is normally a reactor control parameter and is defined as 1/2 the sum of core inlet and outlet j
temperatures.
Since the temperature rise across the core is very small
(<5*F) under decay heat power with forced circulation flow, any of the i
temperature readings cited on page 45 are probably adequate for definition of the " cold shutdown" mode.
It is truth that there may be hotter stagnant regions of coolant in the system.
However, this is of little concern if it is below saturation temperature.
k l
w
~
w m
.n..
p,,
(
' 4.3.2 and 5.1 - need for reliable instmentation and procedures -
Reactor vessel level instrumentation is now required for all PWRs in accordance with item II.F.2 of NUREG-0737. This requirement is addressed by Generic Letter 82-28 which was transmitted to all licensees.
A+. this time (August 1985),
instrumentation has been installed in response to the requirement on at least 29 reactors and is operable and has been approved on 16 of these units.
It is in various stages of completion on the balance of PARS and most are expected to be completed by the end of 1986. The approved systems are a Westinghouse dP design and a Combustion Engineering Heated Junction Thermocouple design. The instrumentation is redundant and fully qualified and is' required by Technical Specifications in operating modes 1, 2 and 3.
Readout and alarms are available in the control Although not required, the systems are being used successfully to track room.
reactor drain and fill operations.
In most cases, this application is probably not incorporated into procedures and is done more as a check of instrument operability with comparison to the tygon tubing " manometer" rigs.
It is suitable for monitoring in modes 4 and 5 but must be disconnected during head removal for refueling operations.
It is difficult to understand why appropriate use of existing temperature and flow, instrumentation would not be equally suitable for detecting loss of the decay heat removal function. Failure to detect and respond to the loss of decay heat removal function is probably due more to inadequate attention and jnadequate procedures, including improper use of the tygon tubing manometers, rather than lack of suitable monitoring instrumentation.
f
COMMENTS FROM REACTOR SYSTEMS BRANCH W 1.
The DHR systems are not safety grade systems on most operating plants.
Only those plants that have the DHR systems designed and constructed to BTP RSB 5-1 have truly safety grade systems.
2.
The report should clearly differentiate between DHR systems and the DHR system.
I.E. DHR systems = DHR system + ADVs + DTSGs + AFWs (B&W)
=
+ AFWs (W)
=
3.
The point that in about I hour the core will uncover may be overempharsized. There are other design basis events that, if coupled with other failures, can lead to core uncovery within this time.
For example, a total loss of feedwater (LOFW + Loss of AFW) would lead to core uncovery in about this time. However, this doesn't mean that core uncovery due to a loss of RHR system is acceptable. What it means is that the 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> figure should be viewed in the context of all events. While a loss of the RHR system leading to core uncovery is obviously unacceptable, it must be l
pointed out that there are operator actions available.
\\
4.
The author's point regarding the technical specifications required by the p
generic letter seems to imply that these technical specifications actually l
created an operational loophole that has resulted in many losses of the DHR system.
In our view, the technical specifications required by the generic letter (and under the generic MPA B-57) are a significant i
improvement'in that they require redundant DHR systems to be available in all modes of operation. The " loophole" referred to allows all DHR and RCS pumps to be disabled for up to I hour as long as no operations are conducted that could dilute the RCS boron concentration and the fluid at j
t i
i f
(
.g
...I
' i the core exist is kept at least 10'F subcooled. According to CE, this
" loophole" is necessary for some CE designed plants whose SDCS design pressure is below the RCP minimum suction pressure requirements.. For these plants, the RCPs must be stopped and RCS pressure reduced before the SDC6 can be operated.
Also, CE stated that the normal way of conducting a coo)down is to stop the RCPs just prior to aligning the SDCS and starting the SDCS pumps.
This short period of natural circulation is required in the CE plant operating procedures, and is thought to facilitate the transition to the lower flow rate of the SDCS.
B&W stated that there is a pressure overlap between the RCPs and DHR system. The provision of allowing all DHR and RCS pumps to be disabled for up to I hour is not needed for B&W designed plants.
5.
AEOD should be aware of _ RSB's position regarding ACI on the RER system.
See B. Sheron's memo to the Members of RSB dated January 28, 1985 (Enclosure M. Also, T. Dunning had a safety concern regarding the ACI's at Diablo Canyon and this was forwarded to SPEB for prioritization as a potential generic issue. The SPEB has categorized this issue as a HIGH priority issue, and it is referred to as GI-99. The issue addressed was basically whether the ACIs should be removed and the SPEB prioritization seems to imply that they should.
Also, the recent Sandia work for A-45 seems to support the idea that ACI's 4
are not necessary.
RSB has gone on record with Diablo Canyon as encouraging the licensee to look l.ard at the ACI and recommend a design that, on balance, is safest.
NOU RSB had a meeting with W on removing the ACI on March 20, 1985. Wi n performing a PRA that will show an overall improvement in safety with e
removal of the ACI. See W. Jensen's meeting minutes dated April 18, 1985 (Enclosure 2).
6.
The report should discuss decay heat removal in mode 4 and 5 using the steam generators when available as an alternate DHR subsystem should the l
RHR system be inoperable.
l 7.
More of the current symptom oriented procedures adequately treated operation in modes 4, 5 and 6.
I believe this point is worthwhile.
Partial resolution of this issue may take the fonn of extension of the EPG's (TMI, 1.C.I) to shutdown modes.
l l
(
O ENCLOSURE 2
.. -..