NUREG-1195, Forwards Response to Re NRC Reexam of B&W Reactor Design.Review Efforts Intensified After TMI Accident.Nrr Reorganized Last Yr to Increase Attention to Major Vendor Product Lines.Action Plan Under Development for Rancho Seco
Text
w, f
e, J,
't UNITED STATES C
~
o NUCLEAR REGULATORY COMMISSION j
(E WASHif[GT0N. D. C. 20555 k*..../
/
. April 23, 1986 CHAIRMAN The Honorable Robert T. Matsui United States House of Representatives Washington, D.
C.
20515
Dear Congressman Matsui:
Your' letter of January 29, 1986 asked several questions related to the staff's recently announced reexamination of-the Babcock and Wilcox reactor design.
Each of these questions is addressed in the cnclosure to this letter.
Following the Three Mile Island accident, the staff did intensify its review efforts as you noted.
That effort resulted in a substantial number of improvements in those plants designed by B&W.
The staff is concerned that the number and; complexity of operational transients has not decreased'as expected.
However, there have been improvements in performance; the total number of trips for B&W plants has decreased from an average of about 45 in s
1981, 1982, and 1983 to an aver, age of about 27 over the past two years.
This reduction in cha'lenges to the plants indicates that the actions taken Ly the NRC staff and utilities to improve performance,have begun to take effect.
Complete elimination of trans>ients cannot be-expected to occur.
However, the NRC staff and the utilities have ongoing programs aimed at-furthe'7 reducing the transients as well as improving the effectiveness and ease of the plant operator's response to them.T In addition, the office responsible for licensing activities, Nuclear Reactor Regulation, was reorganized last year to provide increased and specialized attention to the major reactor V.endors' product lines.
This alignment permits staff specialists to focus on specific.yendor's designs, performance and operation.
The staff's reexamination will be directed by the NRR (ivision responsible for B&W reactors.
e Regarding Rancho Seco, an NRC Incident Investigation Team (IIT) was dispatched to the site following the December 26, 1985 loss of integrated control system incident.
Thi-s team was tasked:
(1) to determine the facts surrounding the
'N incident, (2) to identify the probable cause, and (3) to 2
make appropriate findings and conclusions.
The IIT Report, NUREG-1196, identified many weaknesses and vulnerabilities at Rancho Seco.
Given this report the staff is presently developing an action plan which, when completed,:will be forwarded to the Sacramento Municipal Utility District.
(SMUD).
Although SMUD had-originally anticipated being-f w
a.
4 ready to resume operations around March 8, 1986, upon further examination of the situation they decided they are not yet ready to establish a restart date.
The licensee will be required to satisfactorily address the items in the NRC plan prior to NRC restart authorization.
Please note that SMUD has been advised that the Commission would be receptive to a meeting with SMUD management.
The responses to your specific questions are presented in the enclosure to this letter.
We hope this addresses your concerns.
Commissioner Asselstine has the following comment:
In your letter, you note that the pattern of NRC actions regarding B&W plants seems to correspond to the public furor generated by each successive incident.
You go on to state that when public concern is highest there is a flurry of activity at the NRC, but when the furor subsides, NRC activity seems to subside as well.
I believe that your comments are right on the mark.
There is a pattern within the Agency of progressively narrowing the scope of safety reviews that is proportional to the time elapsed after a significant event occurs.
The NRC has been well aware of the vulnerabilities of B&W plants to failures in the Integrated Control System (ICS) and Non-nuclear Instrumentation (NNI) for seven years or more.
In my view, the NRC has not taken effective action to correct these problems.
I believe that view is shared by the Incident Investigative Team (IIT) that conducted a review of the December 26, 1985 accident at Rancho Seco.
The IIT report was highly critical of the NRC staff's failure to heed the lessons of numerous operating events that demonstrated the safety vulnerabilities in this aspect of the B&W plant design.
Based upon the team's report and my own concern regarding the staff's past regulation of the B&W plants, I urged the Commission to form a special review group, including experts from outside the NRC, to conduct a thorough review of the safety vulnerabilities in the B&W design and the adequacy of past NRC efforts to ensure-that the B&W plants meet acceptable safety standards.
This was the approach taken by the Commission in the case of the June 3, 1985 accident at the Davis-Besse plant on the basis of less compelling evidence of failures in the staff's past performance.
I believe that such a review is particularly appropriate given the NRC staff's expressed' view following the Davis-Besse accident that there was no need for a generic review of the safety of the B&W design.
Unfortunately, I must tell you that all of my colleagues on the Commission rejected my
e proposal.
As a result, any review of the adequacy of the staff's performance will be left to the staff itself.
As to the continued operation of the B&W plants, I believe the B&W plants have two strikes against them and if their performance is not improved quickly, the licensees should bear the burden of demonstrating.that the plants are safe to operate.
I must tell you that I.
have particular concerns regarding Rancho Seco.
In view of the extensive history of operational problems at that plant, I would insist upon a detailed improvement program concerning the management and operation.of that plant together with demonstrated progress in carrying out the essential elements of that program before the plant is allowed to restart.
I would also insist upon a Commission meeting with the-licensee's senior management to review their history.of performance and their detailed plans for improvement.
In regard to Commissioner Asselstine's views, Chairman Palladino has the following comments:
The NRC considers safety to be its fundamental mission.
Any action believed to be necessary to assure that NRC licensed facilities are operated safely is taken independently of whether a public concern is apparent.
By-and-large I have been impressed by the technical competence and dedication of the staff in performing the many safety activities they are called upon to perform.
Comments, as suggested in the IIT report, that the staff was not responsive in as timely a manner as it should have been, are currently being assessed.
It would seem appropriate, therefore,'to await the conclusion of that assessment before suggesting any impropriety.
The industry has accepted the responsibility for reassessing the performance of B&W plants in order to reduce the frequency and complexity of transients.
The staff.is monitoring this effort, and I have no reason to believe that anything short of a thorough and comprehensive reevaluation will be made.
Sincerely,
/~)
/fl(sal <
- ~
Nunzi' J. Palladino
Enclosure:
As Stated
f
- 4-b ENCLOSURE 1
-t i
1.
If the NRC believes that the succession of repeated incidents at B&W plants is serious enough to require a year-long safety review, what is the Commission's position as to the impact of' unresolved safety j
questions on the continued operation of those plants?-
The events that lead the staff to conclude that a reexamination was needed have had little or no offsite consequences.
Changes to the, plants resulting.from the'TMI accident, along with those resulting from other events (at:
}
both B&W and other vendor-designed reactors) are being i
incorporated at the facilities..In order to' ensure.that.
they are done effectively, appropriate implementation schedules are developed to assure proper design, review, purchase, installation, testing and training of operators.
The_ potential unresolved safety questions are being addressed, and in the staff's view the plants can continue to be operated safely while the reassessment is underway, j
+
i 2.
What specific events triggered the public announcement of the current reassessment and how do these events 1
differ from past events?
{
The Davis-Besse event last summer and-the December event at Rancho Seco reinforced the staff's awareness of the
(
sensitivity of B&W plants to operational transients.- ~The l
Davis-Besse event in June involved a partial loss of-lI feedwater while the plant was operating at 907 power.
Following a reactor trip, a loss of all feedwater occurred.
l The Rancho Seco event in December involved a loss of power to the integrated control system (ICS) with.the reactor i
operating at about 75% power.
Subsequent pump and-valve actuations resulted in an overcooling event..0ther, less complex, events have also occurred, such as at the~ Crystal.
j i
River plant in February 1980, the; loss of IOS-power-1 transient at Rancho Seco in January 1979, the' loss of' power.
j to Non-Nuclear Instrumentation (NNI) and ICS at Oconee in-l t
November 1979 and the partial loss of NNI:at Rancho Seco in March 1984.
In the Crystal River event, the reactor ~was at low power and the operators were transferring from the' main feedwater system to a backup system used for-low power-i levels.
Some errors occurred, resulting in a pressure transient occurring in the-plant ~.
A number of these events, although less severe, highlighted the' sensitivity of the design.
1 As a result of previous assessments of operating' experience, o
i and the TMI' accident, utilities have been making I
modifications to the B&W plants, but the number and complexity of such events has not decreased as signif.icantly.
I L
~
4 as expected.
In a number of circumstances, the initial transients were similar to previous events, but subsequent responses of the plants and operators were different.
Complications occurred due to multiple equipment failures or operator actions that previously may not have been evaluated specifically.
In some cases, the responses of the plants and operators to the event and complications were as had been evaluated or were more effective in mitigating the transient than had been previously assumed.
As a result of the aforementioned considerations, the staff determined that a reexamination should be undertaken.
3.
To what extent has the NRC known of problems at B&W reactors in the past and what efforts has the NRC taken specifically and generically to correct those problems?
The staff has been aware of certain sensitivities of B&W plants for a number of years.
Each of the reactor vendor's designs has certain features that result in its responses to events being different from that of other designs.
The staff's approach to any operating event is to examine the cause or causes, assess any plant-specific safety concerns, evaluate the generic implications for that particular model, vendor, and type of design, and see that necessary changes are implemented on a schedule consistent with their safety importance.
The staff's actions with respect to problems at B&W reactors are documented in a number of staff reports and Bulletins and date from before the accident at TMI'until now.
Some of the more major efforts are described below.
1.
NUREG-0560 " Staff Report on the Generic Assessment of Feedwater Transients in-Pressurized Water Reactors Design by Babcock & Wilcox Company" This report documented the efforts of the task group that was appointed shortly after the TMI-2 accident to perform a generic assessment of feedwater transients at B&W plants.
The report presented findings and recommendations for improving the design of several B&W plant systems, and for other actions believed to be-needed following the TMI-2 accident.
These findings and recommendations were factored into the TMI Action Plan, NUREG-0737, which is discussed below.
2.
IE Bulletin No. 79-27 On November 30, 1979, the NRC issued IE Bulletin No. 79-27 regarding an event at the Oconee Power Station, Unit 3, that resulted in loss of power to a non-class IE power panel that supplied power to the Integrated 1
Control System (ICS) and the Non-Nuclear Instrumentation (NNI).
The Bulletin required (1) a number of actions related to class IE and non-class IE busses supplying power to safety and non-safety related instrumentation and contr61 systems; (2) the preparation of procedures or review of existing procedures to be used by control room operators upon loss of power to each of these busses; and (3) re-review IE Circular No. 79-02, Failure of 120 Volt Vital AC Power Supplies, dated January 11, 1979, to include both class IE_and non-class IE safety related power supply inverters.
The written responses were to identify and propose design modifications or administrative controls to be implemented and scheduled for implementing the changes.
3.
March 6, 1980 Generic Letter On March 6, 1980, the NRC sent a letter to each of the operating B&W reactor licensees regarding an event at Crystal River Unit 3 that involved a failure in the power supplies for the non-nuclear instrumentation.
The letter required the licensees to provide certain information including (1) a summary of NNI and ICS power upset events that has previously occurred at each plant; (2) the feasibility of performing a test to verify the reliable information that remains following various NNI/ICS power upsets; and (3) an expansion of the review under IE Bulletin 79-27 to include the implication of the Crystal River Unit 3 event.
4.
NUREG-0667 " Transient Response of Babcock & Wilcox
- Designed Reactors" In May 1980, the NRC published NUREG-0667, " Transient Response of Babcock & Wilcox-Designed Reactors."
This report was the product of a special task force which was established to investigate the apparent sensitivity of the B&W plants to transients involving overcooling and undercooling conditions, small break loss-of-coolant accidents, and the consequences of malfunctions and failures of the ICS and NNI.
NUREG-0667 contains 22 recommendations including system modifications, instrumentation and control improvements, plant procedures and training.. These recommendations were then divided into two groups:
high_ priority items for near term implementation and low priority items for long term reconsideration.
1
5.
NUREG-0737 " Clarification of TMI Action Plan Requirements" By letter dated October 31, 1980, the NRC issued NUREG-0737, " Clarification of TMI Action Plan Requirements."
This document is a compilation of TMI-related items-that were approved for implementation by the Commission as of the date of issuance.
One of the items in NUREG-0737 is the requirement to provide a failure mode and effects analysis of the ICS (Item II.K.2.9).
6.
NUREG-1195 " Loss of Integrated Control' System Power and Overcooling Transient at Rancho Seco on December 2, 1985" In February 1986, the NRC issued NUREG-1195, " Loss of Integrated Contrcl System Power and Overcooling Transient at Rancho Seco on December 16, 1985."
This report includes findings and conclusions of the NRC Incident Investigations Team that was established to investigate this event.
These findings and conclusions are being considered in the staff's B&W generic reassessment.
4.
How effective have the NRC's efforts to correct problems at some B&W plants been in preventing similar incidents at other B&W plants?
The effectiveness of the NRC's corrective measures is difficult to accurately measure since it would rely on quantifying the lack of problems during operational events.
The absence of problems, while possibly due to NRC directed improvements, could be due to the initiating event or initial conditions, plant specific components or design differences, or due to the operators' actions'and timing.
However, it is the NRC's belief that B&W plants have an improved safety performance as a result of NRC measures.
The reduction in scram rate at B&W plants in the last two.
years contributes to this conclusion.
NRC corrective measures cover a wide spectrum of actions from the issuance of relatively routine Information: Notices that inform reactor licensees of ' operational events to the use of formal Orders to modify plant systems, operations or procedures.
The role of the Regional staff, including the resident inspectors, is crucial in evaluating day-by-day.
performance of all reactors, including the B&W plants, and in instituting corrective measures based on close-up observation of plant operations.
While the NRC believes its corrective measures at B&W plants have been effective in improving plant performance and P
4
l safety, we are not satisfied'with the current situation for these plants.
As discussed in the response to questions 1 and 2, we are undertaking a broad based B&W generic reevaluation.
This study will not only assess the B&W plants themselves, but will also determine the adequacy of the NRC's regulatory requirements and programs for the B&W plants.
5.
Are B&W reactors more susceptible to repeated thermal shocks from unusual events than other reactors and is the risk of serious accident compounded by the designs of B&W control systems?
Are B&W control systems inherently unsafe or can they be modified for safe operation?
The B&W overall reactor plant design results in rapid pressure and temperature transients as consequences of secondary plant anomalies.
This susceptibility has long been recognized by B&W licensees.
Part of the generic B&W reassessment currently under consideration will be the evaluation of overcooling and repressurization events to determine whether additional actions should be taken to reduce this susceptibility.
The design objectives for all reactors are to limit the thermal shocks to key plant components.
Allowances in the design have been made should a thermal shock occur.
Added margin in the number of potential thermal cycles has been provided.
Added strength in key components has been provided to withstand such shocks.
A subset of the thermal shock questions deals with pressurized thermal shock.
The NRC recently issued a regulation (10 CFR 50.61) which addresses protection against pressurized thermal shock.
That regulation was developed taking into account the predicted frequency and severity of a wide spectrum of postulated transients and accidents, including the type of overcooling events that have occurred in B&W plants.
With regard to the B&W control systems. while we believe the B&W plant design, including the control systems results in a highly sensitive machine that can undergo rapid transients, we do not believe the control systems are unsafe.
The. staff-and licensees are both involved in examining the
+
interactions between control systems and safety systems.
Because the B&W design is more sensitive'to some transients, the control and safety systems have to respond more quickly 1
than other plants to these same transients.
Part of the basis for the staff reassessment is to determine if the present set of requirements for B&W plants needs to be modified. based on operational history.
An outcome of that reexamination will be the staff's conclusions regarding whether control system modifications are necessary.
Because.
the study is-just underway, a conclusion on what is needed, if anything, is premature.
S
4 6
6.
What are the NRC's specific plans regarding the current reassessment of B&W plants?
How do they differ from past assessments?
Is it possible to complete the reassessment in less than one year?
The staff's specific plans.for reassessment are still being developed.
The staff agreed to wait until the Owner's Group provided their plan before it finalized its assessment of what ought to be included in the effort.
This should be accomplished by about mid-April.
The Owner's Group Chairman, Mr. Tucker, indicated that it is logical for them to serve in a leadership role in resolving the NRC's concern.
He also indicated that the Owner's Group is willing to work with the NRC on this program.
The staff will work with the utilities to the extent possible in this program.
Where possible, the licensees will perform certain evaluations and the staff will review them.
In other cases, the staff will independently do the examinations.
Once the necessary scope is determined and the role of the B&W Owner's Group in supporting the staff's effort is established, resource allocations can be made and schedules developed.
It is the staff's intent to complete its review in CY 1986.
This assessment is expected to rely on operating experience to focus the reexamination.
It will look at the performance of equipment and operators in greater detail.
It will differ from previous assessments in that it will perform a more detailed evaluation of the various failure modes of selected plant systems.
Also, the assessment will evaluate the expected versus the actual responses to design basis events, as well as those events that have actually occurred.
7.
To what extent is Babcock & Wilcox Co. liable for a licensee's monetary losses due to design flaws?
The staff's regulatory authority does not extend to the contractual arrangements between a licensee and its contractor.
As a result, we have no information regarding the liability of B&W for a licensee's losses, u
..