ML20204C493

Responds to Retrospective Part of Issue 2a Re Completeness of Staff & Licensee Actions Associated W/Control Sys,Per Stello 860313 Memo to Denton in Response to Investigation of 851226 Incident.Issue 2a Closed

ML20204C493
Person / Time
Site:	Rancho Seco
Issue date:	07/23/1986
From:	Harold Denton Office of Nuclear Reactor Regulation
To:	Stello V NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
References
RTR-NUREG-1195 IEB-72-27, TAC-61635, NUDOCS 8607310246
Download: ML20204C493 (16)

Text

-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

[

July 23,1986 Db b MEMORANDUM FOR: Victor Stello, Jr.

Executive Director for Operations FROM: Harold R. Denton, Director Office of Nuclear Reactor Regulation

SUBJECT:

COMPLETENESS OF VARIOUS STAFF AND LICENSEE ACTIONS ASSOCIATED WITH CONTROL SYSTEMS

REFERENCE:

Memo, Stello to Denton, " Staff Actions Resulting from the Investigation of the December 26, 1985 Incident at Rancho Seco (NUREG-1195)", March 13, 1986.

The purpose of this memorandum is to respond to the retrospective part of Issue 2a from the reference memo and thus to close out issue 2a. The issue and the directed staff action is stated below for convenience.

2. Issue: Completeness of various staff and licensee actions associated with control systems.

Action

a. In light of the ongoing B&W generic review, assess the need to reevaluate the actions taken by the staff and by the licensees in response to the findings, conclusions and recommendations associated with BAW-1564; Bulletin 79-27; NUREG-0667; the February 1980 loss of NNI power at Crystal River; the March 19, 1984 partial loss of NNI at Rancho Seco; and BAW-1791. (Principal Finding #15 and Other Finding #11).

We have assessed the need to reevaluate actions taken by the staff and by licensees in connection with the documents identified above. The staff has reviewed Chapter 7.1 of NUREG-1195 and addressed each subsection and precursor concern. The staff's plans and ongoing activities as well as actions being taken by SMUD and by the BWOG and its member utilities are summarized below, and described in more detail in the enclosure. These activities are expected to be completed later this year.

1. The staff agrees with the IIT's observation that the emphasis on NNI after the 1978 " light bulb incident" may have biased SMUD's subsequent reviews of issues associated with the NNI and ICS. Deficiencies in the NNI and ICS are being addressed by SMUD as part of their recovery / restart program and by BWOG in their design reassessment program.

2. The staff agrees with the IIT's observation that the January 1979 precursor at Rancho Seco (loss of ICS power) did not receive a high level of attention.

The staff is much more sensitive to these events since the TMI-2 accident.

In addition, SMUD and the BWOG are reviewing this event for lessons learned as part of their recovery and design reassessment prcgrams, respectively.

8607310246 860723 PDR ADOCK 05000312 S PDR

a

3. Regarding completeness of actions taken with respect to BAW-1564 (Failure Modes and Effects Analysis of the ICS) and the ORNL review of it, the BWOG has been requested to reevaluate BAW-1564 and to describe its plans to address the ORNL concerns. The staff will assure that the recommendations in BAW-1564 and the ORNL review are reconsidered regarding their applicability, appropriateness, and implementation status at each B&W-designed operating reactor.

4, The staff has requested the BWOG, and they have agreed, to reevaluate IEB 5, 79-27 regarding the consequences of a loss of power to the instrumentation

86. and control systems for all of the B&W-designed operating plants.

With regard to atmospheric dump valves (ADVs) and turbine bypass valves (TBVs) opening on loss of ICS power, the staff has met with the BWOG and determined that only Rancho Seco has the ADV problem and only Rancho Seco and ANO-1 have the TBV problem. Rancho Seco has already redesigned the ADV and TBV controls to eliminate the problem, and the staff will review the modifications prior to their restart. AN0-1 has committed to modify the TBV controls during their next refueling (beginning August 1986). The staff is tracking this commitment.

In retrospect, the staff could have done more in reviewing licensee responses to Bulletin 79-27 by focusing its resources on review of the B&W-designed plants. The staff has learned from experience and is now giving more management attention and resources to problem plants. The staff will thoroughly review the BWOG reevaluation of Bulletin 79-27.

7. The staff has conducted a survey of completeness of actions taken with respect to NUREG-0667 recommendations by the staff and by licensees of each B&W-designed operating reactor The survey shows that 90% of the related staff requirements have been implemented; the rest will be complete by the end of 1987. The staff is planning to review the prioritization of certain lower-priority recommendations which were not required earlier. SMUD and the BWOG are reviewing the recommendations as part of their recovery and B&W-design reassessment programs, respectively.

8. In connection with the partial loss of NNI at Rancho Seco in 1984, the staff plans to complete in the near future its review of the BWOG submittal dated January 1985. In addition, SMUD and the BWOG are reviewing this event as part of their recovery and design reassessment programs, respectively.

9. The staff has expanded the scope of USI A-47 (Safety Implications of Control Systems) to include the type 820 ICS design as well as the type 721 design for B&W-designed reactors. In addition the staff is reviewing the implementation status of recommendations from NUREG-0667, BAW-1564, and Bulletin 79-27 to assure that any technical issues associated with those recommendations have been resolved or properly coordinated with the scope of A-47.

m The staff will track each of the above items to assure that they are completed in a satisfactory manner.

Another potential retrospective issue was considered, in addition to those identified in NUREG-1195, regarding the implementation of one of the Commission Orders issued after the TMI-2 accident to B&W operating plants. That order required procedures and training to initiate and control AFW independent of the integrated control system (ICS). In addition, at Rancho Seco, the orders included a directive that procedures and training be implemented to provide for control of steam generator level by use of safety grade AFW bypass valves in the event that ICS steam generator level control fails. We conclude that this is not a retrospective concern for the reasons discussed below.

All of the B&W licensees responded that procedures and training were in place to permit operators to manually initiate and control auxiliary feedwater independent of ICS. Also, Rancho Seco provided procedures and training to control steam generator level using safety grade AFU bypass valves upon failure of ICS level control. Licensees performed low power tests to demonstrate their modifications and the NRC staff reviewed and walked through the procedures and, on an audit basis, interviewed operators for training in the procedures. A total loss of ICS power under operating conditions was not simulated. The safety evaluation reports issued by the NRR Bulletins and Orders Task Force (NUREG-0645 Volumes 1 and 2) concluded that the licensees responses were acceptable. The responses and evaluation focused on actions to preclude the loss of feedwater on failure of ICS, but did not specifically evaluate the potential for overcooling resulting from ICS failure.

Nonetheless, all B&W operating plants (including Rancho Seco prior to restart) now have AFW designs which are independent of the ICS or have manual override capability with procedures and training to initiate and control AFW independent of the ICS from the control room.

s q g }lld, G G'E 0 W Harold R. Denton, Director Office of Nuclear Reactor Regulation

Enclosure:

As Stated

• See previous white for concurrences.

DPD/PBD-6 PD/PBD-6 AD/DPLB D/DSR0 D-PW GEdison/cf* JStolz* DCrutchfield* TSpeis* FMira a*

7/7/86 7/7/86 7/8/86 7/7/86 7/8/86 7 E5IMetfT4Q! RVollmer JStolz d DD/ D/ &cket.Eua> FMiraglia TSPeis RW mer HD@R e ton NRC PDR Glainas GEdison 7 /86 7/A /86 Local POR ERossi HDenton F EDO Rdg DCrutchfield TRehm PDB#6 Rdg JHeltemes, AE00 RIngram OQ

9. The staff has expanded the scope of USI A-47 (Safety Implications of Control Systems) to include the type 820 ICS design as well as the type 721 design for B&W-designed reactors. In addition the staff is reviewing the implementation status of recommendations from NUREG-0667, BAW-1564, and Bulletin 79-27 to assure that any technical issues associated with those recommendations have been resolved or properly coordinated with the scope of A-47.

The staff will track each of the above items to assure that they are completed in a satisfactory manner.

Another potential retrospective issue was considered, in addition to those identified in NUREG-1195, regarding the implementation of one of the Commission Orders issued after the TMI-2 accident to B&W operating plants. That order required procedures and trainin the integrated control systemICS). (g toWeinitiate and control conclude that thisAFW is notindependent a retro- of spective concern for the reasons discussed below.

All of the B&W licensees responded that procedures and training were in place to permit operators to manually initiate and control auxiliary feedwater independent of ICS. The safety evaluation reports issued by the NRR Bulletins and Orders Task Force (NUREG-0645 Volumes 1 and 2) concluded that the licensees responses were acceptable. The responses and evaluation focused on actions to preclude the loss of feedwater on failure of ICS, but did not specifically evaluate the potential for overcooling resulting from ICS failure.

Nonetheless, all B&W operating plants (including Rancho Seco prior to restart) now have AFW designs which are independent of the ICS or have manual override capability with procedures and training to initiate and control AFW independent of the ICS, Harold R. Denton, Director Office of Nuclear Reactor Regulation

Enclosure:

As Stated

• See previous white for concurrences.

DPD/PBD-6 PD/PBD-6 AD/DPLB D/DSR0 D-PWR-B GEdison/cf* JStolz* DCrutchfield* TSpeis* FMiraglia*

. 7/7/86 7/7/86 7/8/86 7/7/86 7/8/86

/86 (Vg% DISTRIBUTION RVollmer JStolz

DD/NRR D/f$ Docket File FMiraglia TSPeis i RVollmer llD4 ton NRC PDR Glainas l 7/ /86 7/(/86 ' cal PDR

_ ERossi i

ED0 Rdg DCrutchfield l PDB#6 Rdg JHeltenes, AE0D l

TRehm RIngram HDenton GEdison l

I

- _ < _ _ - - - , _ . _ - _ . , -_._w- ,_ ___.-,r_. .- - -

9. The staff has expanded the scope of USI A-47 (Safety Implications of Control Systems) to include the type 820 ICS design as well as the type 721 design for B&W-designed reactors. In addition the staff is reviewing the implementation status of recommendations from NUREG-0667, BAW-1564, and Bulletin 79-27 to assure that any technical issues associated with those recommendations have been resolved or properly coordinated with the scope of A-47.

The staff will track each of the above items to assure that they are completed in a satisfactory manner.

Another potential retrospective issue was considered, in addition to those identified in NUREG-1195, regarding the implementation of one of the Commission Orders issued after the TMI-2 accident to B&W operating plants. That order required procedures and training to initiate and control AFW independent of the integrated control system (ICS). We conclude that this is not a retro-spective concern for the reasons discussed below.

All of the B&W licensees responded that procedures and training were in place to permit operators to manually initiate and control auxiliary feedwater independent of ICS. The safety evaluation reports issued by the NRR Bulletins and Orders Task Force (NUREG-0645 Volumes 1 and 2) concluded that the licensees responses were acceptable. The responses and evaluation focused on actions to preclude the loss of feedwater on failure of ICS, but did not specifically evaluate the potential for overcooling resulting from ICS failure.

Nonetheless, all B&W operating plants (including Rancho Seco prior to restart) now have AFW designs which are independent of the ICS or have manual override capability with procedures and training to initiate and control AFW independent of the ICS.

Harold R. Denton, Director Office of Nuclear Reactor Regulation

Enclosure:

As Stated

• See previous white for concurrences.

DPD/PBD-6 PD/PBD-6 AD/DPLB D/DSR0 D-GEdison/cf* JStolz* DCrutchfield* TSpeis* FMir 7/7/86 7/7/86 7/8/86 7/7/86 7/7

/86 u DD/NRR D/NRR RVollmer HDenton 7/ /86 7/ /86

9. The staff has expanded the scope of USI A-47 (Safety Implications of Control Systems) to include the type 820 ICS design as well as the type 721 design for B&W-designed reactors. In addition the staff is reviewing the implementation status of recommendations from NUREG-0667, BAW-1564, and Bulletin 79-P.7 to assure that any technical issues associated with those recommendations have been resolved or properly coordinated with the scope of A-47.

The staff will track each of the above items to assure that they are completed in a satisfactory manner.

Another potential retrospective issue was considered, in addition to those identified in NUREG-1195, regarding the implementation of the Comission Orders issued after the TMI-2 accident to B&W operating plants. That order required procedures and training to initiate and control AFW independent of the integrated control system (ICS). We conclude that this is not a retro-spective concern for the reasons discussed below.

All of the B&W licensees responded that procedures and training were in place to permit operators to manually initiate and control auxiliary feedwater independent of ICS. The safety evaluation reports issued by the NRR Bulletins and Orders Task Force (NUREG-0645 Volumes 1 and 2) concluded that the licensees responses were acceptable. The responses and evaluation focused on actions to preclude the loss of feedwater on failure of ICS, but did not specifically evaluate potential for overcooling resulting from ICS failure.

Nonetheless, all B&W operating plants (including Rancho Seco prior tc restart) now have AFW designs which are independent of the ICS or have manual override capability with procedures and training to initiate and control AFW independent of the ICS.

Harold R. Denton, Director Office of Nuclear Reactor Regulation

Enclosure:

As Stated ,

I j[(/uD Y \

DPD/PBD-6 -6 b D R0 D-PWR-B FMiraglia GEdison/cf JSt D tchfield TSpe' 7/ 7 /86 7/1/86 8 7/ /86 4 7/[/86

DD/NRR D/NRR RVollmer HDenton l 7/ /86 7/ /86 i

l I

l

/

r Enc 1csure Staff Review of Precursors in Chapter 7.1 of IIT Report -

(NUREG-1195) on Rancho Seco Overcooling Incident of 12/26/85

1. The Rancho Seco Lightbulb Incident (Sec. 7.1.1) .

7 On March 20, 1978 Rancho Seco underwent a severe overcooling transient as a result of a loss of power to the NNI. A short circuit in the 24 Vdc NNI-Y power system caused a loss of much of the control room instrumenta-tion and indication, and simultaneously sent erroneous signals to the ICS.

This resulted initially in loss of feedwater and steam generator dryout followed by steam generator flooding and significant reactor overcooling.

~

Because of the severity of the incident NRC sent a team of engineers to ,

determine what had occurred. The team reported details of the transient.

However, this was the first of the significant ICS/NNI-related ir.cidents- ~

and no bulletin, information notice, or NRC requirements were issued to 'any -

utility. Following the incident, SMUD made some improvements in the NNI at Rancho Seco. However, it appears that other licensees of B&W-designed plants -

took no action at the time, and the NRC took no further action.

The staff agrees with the IIT's observation that the emphasis on NNI may '

have biased SMUD's subsequent reviews of issues associated with the NNI -

and ICS. Although the improvements listed in NUREG-1195 were made to the NNI system, SMUD did not recognize that the ICS, having a similar -

power supply design, needed similar improvements. This is now being addressed

~by SMUD and by the B&W Owners Group as part of their design reassessment program. The NNI deficiencies are being addressed by the BWOG and the staff as noted in item 8 below.

2. The First Rancho Seco Loss of ICS Incident (Sec. 7.1.2)

On January 5, 1979 at Rancho Seco, a technician inadvertently caused a short circuit in the ICS causing a loss of ICS power. Initially, there was a loss of feedwater and a reactor trip. This was followed by an overcooling transient caused by excessive feedwater. The cause and consequences of this event were similar to those of the later December 26, 1985 incident.

The licensee filed an LER describing the event, and the NRC staff.

reviewed it as part of its routine inspection program. The inspection findings were described in a single brief paragraph in which no items of noncompliance or deviations were identified. No significant changes to the plant design were made or required.

The staff agrees with the IIT's observation that the January 1979 overcooling incident was not as severe as the lightbulb incident and did r'

l t

l not receive the same level of attention. This incident occurred prior to the TMI-2 accident. The NRC and the industry are more sensitive now to these kinds of events than they were prior to the TMI-2 accident and a ,

much higher level of attention is given to them. In addition, the '

January 1979 loss of ICS incident at Rancho Seco is now being reviewed for lessons learned by SMUD as part of its recovery / restart program, and by the BWOG as part of its reassessment program.

3. BAW-1564 "ICS Reliability Analysis" (Sec. 7.1.3)

Because of the TMI-2 accident, orders were issued in mid-1979 to licensees with B&W-designed plants. Included in the orders was a requirement for licensees to prepare a reliability analysis (failure modes and effects) of the ICS. This analysis was performed generically by B&W, and the results and recommendations were documented in BAW-1564 in August 1979.

The recommendations of BAW-1564 are stated below:

"B&W recommends the following areas be reviewed on a plant-specific basis for possible changes to enhance reliability and safety.

1. ICS-Related

a. NNI/ICI power supply reliability .

b. Reliability of input signals from the NI/RPS system to the ICS - specifically, the RC flow signal.

c. ICS/B0P system tuning, particularly feedwater condensate systems and the ICS controls.

2. Balance of Plant

a. Main feedwater pump turbine drive minimum speed control - to prevent loss of main feedwater or indication of main feedwater.

b. A means to prevent or mitigate the consequences of a stuck-open main feedwater startup valve.

c. A means to prevent or mitigate the consequences of a stuck-open turbine bypass valve."

On August 21, 1979, all B&W licensees were requested by NRC to provide their position on BAW-1564 and its applicability to their plant. All responded that they generally endorsed the report as applicable to their plant, and described modifications and plans to implement the recommendations or reasons why modifications were unnecessary. However, questions remain about the depth of licensees' evaluations and the implementation dates of planned modifications. In retrospect, some of the responses appear to have been superficial. Also, the staff did not track the licensees' planned modifications to determine whether they were implemented.

ORNL reviewed BAW-1564 under contract to NRC and provided comments and recommendations to the staff in January 1980. Although the ORNL review generally agreed with the BAW-1564 recommendations, ORNL was critical of the B&W failure modes and effects analysis, primarily because it did not identify the effects of ICS failures on plant equipment controlled by the ICS.

p During 1979 and 1980, the staff cataloged the post-TMI requirements and published them as TMI action items in NUREG-0737. The requirement for a failure modes and effects analysis of the ICS became known as TMI Action Item II.K.2.9. The staff reviewed the material submitted by B&W, B&W licensees and ORNL and prepared SERs stating the licensees met the requirements of TMI action item II.K.2.9. The staff further noted that the adequacy of regulatory requirements for control systems would be determined as part of USI A-47.

The staff does not entirely agree with the implication of the IIT observation that "SMUD concluded that no changes were necessary" in response to BAW-1564 recommendations for ICS changes. SMUD addressed each of the recommendations in BAW-1564 stating it had improved reliability of the 120V inverters (power

- sources for NNI and ICS) through maintenance and minor modifications, and had installed an automatic bus transfer device on NNI and ICS power sources.

Plans were described for reconfiguring the NNI internal power supplies. No plans were described for modifying ICS internal power supplies. A paragraph from the staff's SER states the following: "The staff asked SMUD to address the B&W recommendation to improve the reliability of the Non-Nuclear Instrumentation (NNI)/ICS power supply. SMUD described modifications already completed and additional modifications to be made to improve the reliability of the power supplies to these systems." The matter appears to be a question of SMUD's preoccupation with the NNI (not recognizing the ICS also needed further improvements).

The IIT report also observes that "the long term issues associated with the B&W report were to be considered in USI A-47." However, there were no specific long term issues identified in the B&W report. The only long term issue associated with the B&W report which was deferred to A-47 was the A-47 issue itself, i.e., the effects of control system failures on plant safety and the adequacy of related regulations.

The retrospective issue raised by the B&W report is whether SMUD and other licensees of B&W-designed plants gave thorough consideration to the i

recommendations of BAW-1564 and the ORNL Review and whether they followed

! through with implementation of planned modifications described in their

! submittals made prior to the staff's SER.

The staff has developed a plan to review completeness of actions taken regarding BAW-1564 and the ORNL review of BAW-1564. As part of the B&W s

reassessment, the BWOG has been requested to reevaluate BAW-1554 to determine

if changes to the ICS design are warranted. The BWOG has also been requested l to describe its plans to address the ORNL concerns. Upon completion of the BWOG assessment, the staff will reconsider the recontendations in BAW-1564 and the ORNL review regarding their applicability, appropriateness and implementation status at each B&W operating reactor.

4. IE Bulletin 79-27, " Loss of Non-Class 1E IEC Power System Bus Durino Operation" (Sec. 7.1.4).

5. Crystal River Event on February 26, 1980 (Sec. 7.1.4.1).

6. NRC review of Responses to Bulletin 79-27 (Sec. 7.1.4.2)

These three subsections of the IIT report are grouped together because of their close relationship.

At Oconee 3 on November 10, 1979, a non-class IE inverter that provides all power to the ICS and to one channel of NNI tripped due to a blown fuse.

The automatic bus transfer (ABT) failed to automatically transfer the loads from the inverter to the alternate regulated AC power source. All valves controlled by the ICS assumed their respective failure positions and the operators lost many indications in the control room. The loss of power lasted approximately 3 minutes.

A senior technical review team was sent to Oconee to review the event.

Bulletin 79-27 was issued on November 30, 1979 to all operating reactors. Bulletin 79-27 required a number of actions as follows:

1. Review the class IE and non-class 1E buses supplying power to safety and nonsafety-related instrumentation and control systems which could affect the ability to achieve a cold shutdown condition using existing procedures or procedures developed under item 2 below:

o Identify and review the alarm and/or indication provided in the control room to alert the operator to the loss of power to the bus, o Identify the instrument and control system loads connected to the bus and evaluate the effects of loss of power to these loads including the ability to achieve a cold shutdown condition.

o Describe any proposed design modifications resulting from these reviews and evaluations, and your proposed schedule for implementing those modifications.

2. Prepare emergency procedures or review existing ones that will be used by control room operators, including procedures required to achieve a cold shutdown condition, upon loss of power to each Class 1E and non-Class 1E bus supplying power to safety and nonsafety-related instrument and control systems.

3. Re-review IE Circular No. 79-01, Failure of 120 Volt Vital AC Power Supplies dated January 11, 1970, to include both Class IE and non-Class IE safety-related power supply inverters. Based on a review of operating experience and your re-review of IE Circular No. 79-01, describe any proposed design. modifications or administrative controls to be implemented as a result of the re-review.

4. Within 90 days of the date of this Bulletin, complete the review and evaluation required by this Bulletin and provide a written response describing your reviews and actions taken in response to each item.

Responses to IE Bulletin 79-27 were due on February 28, 1980.

However, on February 26, 1980 a different loss of NNI event occurred at Crystal River Unit 3 (CR-3). The plant was operating at 100% power when an electrical short caused the loss of the +24 volt power supply to the NNI.

Approximately 70% of NNI became inoperable or inaccurate. The PORV opened and

stayed open as a direct result of the NNI power loss. High Pressure Injection (HPI) initiated as a result of the reactor depressurization, the pressurizer was pumped solid, a primary safety valve lifted, and eventually the RC Drain Tank rupture disk ruptured, spilling approximately forty-three thousand gallons of primary coolant water into containment.

The NRC issued an Information Notice (80-10) to all licensees stating that an addendum to IE Bulletin 79-27 would be issued to reflect the,CR-3 event. In March 1980 letters were sent to the B&W licensees requiring information which included:

1. A summary of NNI and ICS power upset events that had previously occurred at each plant;

2. The feasibility of performing a test to verify the reliable information that remains following various NNI and ICS power upsets; and

3. An expansion of the review under IE Bulletin 79-27 to include the implications of the Crystal River event.

Based on the responses to the above information request, the NRC concluded that implementation of three actions at operating B&W-designed nuclear power plants was necessary to provide continued assurance of public health and safety. In April 1980, Orders were issued to each licensed B&W plant which required the following:

1. Actions which will allow the operator to cope with various combinations of loss of instrumentation and control functions. This includes changes in (a) equipment and control systems to give clear indications of functions which are lost or unreliable; (b) procedures and training to assure positive and safe manual response by the operator in the event that competent instruments are unavailable.

2. Verification of the effects of various combinations of loss of instrumentation and control functions by design review analysis and by test.

l

, 3. Correction of electrical deficiencies which may allow the power-operated relief valve and pressurizer spray valve to open on non-nuclear instrumentation power failures, such as the event which occurred at CR-3 on February 26, 1980.

l t It was not necessary to issue an order to the licensee for TMI-1 as the issue i was incorporated as part of the overall NRC review for restart then in progress.

Appropriate short term corrective actions at the two plants involved in the events (i.e., Oconee 3 and Crystal River 3) were taken before startup was allowed. Specifically at Oconee 3, the licensee installed a redundant automatic transfer switch in parallel with the static transfer switch associated with the inverter. The licensee also relocated some key-instrumentation onto another power supply to ensure that adequate instrumentation would be available to allow the operator to place the plant in a safe shutdown condition if a similar event should occur. The emergency procedures were also

l I

e upgraded to provide more guidance to the operators for events involving loss of NNI. For Oconee 1 and 2, the licensee committed a dedicated person to manually transfer the equivalent inverter to the alternate power sources," if necessary, until the same modifications could be made tc those units during their next scheduled shutdown. In the longer term, the licensee was to re-review emergency procedures and evaluate the possibility of relocation of other control and instrumentation functions.

At Crystal River Unit 3, short term actions (i.e., prior to restart) included (1) a thorough testing of the NNI system to determine the cause of failure, (2) modifying the PORY to fail close on NNI failure, (3) modifying the pressurizer spray valve so that it doesn't open on an NNI failure, (4) providing a positive position indication of all three relief or safety valves, (5) establish procedural control of NNI selector switches, (6) train all operators in response to NNI failures, (7) move 120V ICS "X" power to a vital bus, (8) initiate more extensive program for events recorder system, and (9) provide the operator with redundant indication of main plant parameters.

Additionally, the NRC Order issued as a result of the CR-3 event required the licensee to complete all three actions required by the Order prior to startup.

Longer term actions were also required of the licensee as discussed below.

All licensees responded to IE Bulletin 79-27 and the B&W licensees met the requirements specified in the Confirmatory Orders issued as a result of the Crystal River 3 event. (Note that responses were required from licensees of all operating plants.)

IE reviewed the responses to item 2 of IE Bulletin 79-27 concerning upgrades to emergency procedures. NRR reviewed items 1 (system design modifications) and 3 (inverter power supply design and proposed modifications) of IE Bulletin 79-27.

NRR originally intended to perform an in-depth review comparable to the review that would be conducted as part of the operating license review of a new l

facility. However, by September 1980, it was concluded that very few of the licensees' submittals contained sufficient information to permit the in-depth review anticipated. A lengthy supplement to Bulletin 79-27 was prepared and proposed for issuance. The proposed supplement was never issued for several reasons. First, to obtain some of the information requested, a system test was required. Concerns were raised within NRR that the confirmatory test constituted certain risks to the plants which were unacceptable. Additionally, a significant review effort had already been expended reviewing the initial responses from all licensees. If the supplement had been issued, it would have required a significant NRC staff effort to provide an adequate review during a time of highly competing review priorities (e.g., NUREG-0737, "TMI Action Plan").

It was decided that NRR would not perform an in-depth review but would only ascertain whether the licensees had provided reasonable assurance that the specifically identified concerns of Bulletin 79-27 had been adequately addressed. A memorandum from the Director of Licensing to the Regional

Administrators dated June 22, 1982, documents NRR's closeout for 67 of the 68 licensees' responses to IE Bulletin 79-27. The last plant, ANO-1, was closed out August 4,1982.

Subsequent to the above actions, but as a result of the Crystal River 3 event, the Office for Analysis and Evaluation of Operational Data (AE0D) identified a potential generic concern regarding the failure position of atmospheric dump valves (ADVs) on loss of ICS coupled with the lack of valve position indication. On August 15, 1980, individual letters were sent to each B&W licensee requesting plant specific information concerning the failure position of ADVs and potential corrective actions. All B&W licensees either provided j satisfactory responses or modified the failure position of the ADVs except for Rancho Seco, which comitted to complete the action when EFIC was installed.

With regard to the IIT observations in these three subsections:

1. The staff agrees with the IIT observation that SMUD's response to IEB 79-27 seemed to focus on NNI concerns and did not identify design or procedure modifications to address a loss of ICS power. This reaction by SMUD was similar to their reaction to the Rancho Seco "lightbulb" incident on January 5, 1979. The generic concern is whether other licensees of B&W-designed plants applied a similar narrow focus to Bulletin 79-27. The staff has requested the BWOG, and they have agreed, to reevaluate IEB 79-27 regarding the consequences of a loss of power to

, the instrumentation and control systems for all of the B&W operating i

plants.

2. As the IIT observed, following the Crystal River 1980 event the staff issued orders to licensees of B&W-designed plants requiring three major l

actions to improve plant system responses to the loss of NNI power. The staff also requested information from licensees regarding the AE0D concern about Atmospheric Dump Valves (ADVs) opening on loss of ICS power. None of the licensecs of B&W-designed plants broadened the issue to include other similar valves, such as Turbine Bypass Valves (TBVs), and SMUD deferred resolution of the ADV concern for several years until the EFIC system was to be installed. The generic and retrospective concern is whether other licensees of B&W-designed plants had a design in which ADVs or TBVs would open on loss of ICS power.

The BWOG, following the Rancho Seco December 26, 1985 event, reevaluated the failure modes of the atmospheric dump valves on loss of ICS power.

The BWOG indicated that, except for Rancho Seco, no other B&W-designed plant had a design in which the ADVs would open on loss of ICS DC power.

All except SMUD had fixed the problem following the Crystal River event.

The BWOG also stated that only two plants (AN0-1 and Rancho Seco) had a design in which the TBVs would open on loss of ICS DC power. Rancho Seco has redesigned the ADVs and TBVs to close on loss of ICS power as part of l

their modifications prior to restart. ANO-1 has comitted to modify the TBVs to close on loss of ICS power and to implement this at their next refueling outage (beginning August 1986).

_ .--- c - ,.,o , - , - - - - , - - , , , , - - -

e

3. The staff agrees with the ITT implication that the NPC staff could have

, done a more thorough review of the responses to Bulletin 79-27. In retrospect, the staff could have focused its resources on responses from those plants where the concern was greatest, the B&W plants. The staff has learned from that experience and is more conscious now of the need to focus its resources on specific plants. The staff will conduct a thorough review of the BWOG reevaluation of Bulletin 79-27. In addition, the staff is reviewing the content of USI A-47 to assure full coordination between issues covered by that program and issues to be resolved by the Bulletin (See response to Section 7.1.7).

7. NUREG-0667, " Transient Response of Babcock & Wilcox-Designed Reactors" (Sec. 7.1.5).

In late 1979, a special task force was established within the NRC to assess the apparent sensitivity of B&W-designed plants to transients, including those involving the ICS and NNI. The results of the task force

study were published as NUREG-0667 in May 1980. The coments and concerns of licensees of B&W plants were obtained and appropriately considered in the report. The NRC staff then evaluated the task force recommendations, and prioritized them for implementation. Many of the recommendations were required to be implemented by licensees, some were deferred for 'mplementation in accordance with existing resources and priorities, and a few were not recomended for implementation.

The IIT observed that "The staff initially (i.e., 1979/1980) had concerns about the transient response of B&W-designed reactors and the role of the ICS as an initiator of such transients. NRC perfonned an extensive study which made 22 recommendations on the issue. However, it does not appear that these recommendations were sent to SMUD for action, or that the recomendations relevant to the December 26, 1985 incident were

implemented at Rancho Seco." The staff notes that only four of the 22 recommendations were directly related to the ICS and NNI.

As the IIT observed, the NUREG-0667 recommendations were not formally sent in final form to SMUD (or to other licensees) for action. The reason this was not done was that the NUREG presented task force recomendations, not the staff's

final position on requirements after prioritfzation. Those recomendations I which were required by the staff were sent to licensees for action. The l mechanisms used by the staff to impose the requirements consisted of (a) including the recommendation as a TMI action item, (b) issuing orders to implement the recommendation, (c) establishing a license condition, and 4

(d) issuing a generic letter. The staff provided a near-final draft of the NUREG to licensees for their review and comment and held meetings with licensees to discuss all the recommendations and incorporate appropriate i comments. Thus, the licensees were well aware of the NUREG recommendations and the technical basis for them, and the staff was aware of licensee coments

! and concerns. The retrospective issue is whether the recommendations were thoroughly evaluated by the staff to develop appropriate requirements and whether they were implemented by licensees.

l

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

' The staff surveyed the completeness of actions in this regard for each B&W-designed plant in preparing a response to a recent question from Congressman Markey. This survey shows that 90 percent of the staff requirements have been implemented by the B&W-designed plants; the remaining 10 percent are expected to be implemented by the end of 1987. The staff is planning to review the prioritiration of certain recommendations which were deferred to detennine whether they should be requirements. The BWOG is reviewing the NUREG recommendations as part of their B&W design reassessment program.

SMUD is reviewing the recommendations as part of their recovery / restart program for Rancho Seco.

8. March 19, 1984 Partial Loss of NNI at Rancho Seco (Sec. 7.1.6)

On March 19, 1984 a hydrogen explosion and fire occurred in the electrical generator at Rancho Seco while the plant was operating at 85 percent power. During the recovery, with the plant shutdown, a partial loss of NNI power occurred twice within a period of several hours. The loss of NNI was the result of a single failure of an inverter compounded by a separate undetected failure of the NNI power supply monitor due to setpoint drift, which taken together resulted in the loss of redundant NNI power sources.

NRC staff reviewers went to the site to review the circumstances surrounding the hydrogen explosion, fire, and partial loss of NNI. The NRC staff prepared an internal trip report on June 29, 1984 which has not been forwarded to the licensee. Because of the generic aspects of the event, the staff met with the licensees for all B&W-designed plants to discuss the event. Subsequently, on September 4, 1984, the NRC transmitted to the B&W Owners Group a number of questions concerning the NNI system.

l SMUD repaired the imediate damage caused by the series of events and was allowed to restart. This action included repairing damage to the generator caused by the fire, replacing failed components, and calibration and setting of all the NNI and ICS 24V AC power supply overvoltage and undervoltage (power monitor) trip setpoints to the

, manufacturer's recommended setting. Longer term actions by the licensee will include investigating the possibility of adding automatic bus transfers so that loss of a single inverter will not cause the loss of any NNI or ICS 120 V AC supplies, reviewing the surveillance

! requirements and calibration of NNI equipment, and revising operating and training procedures. After discussions between NRR and Region V, the plant was allowed to restart.

l The B&W Owners Group submitted a report dated January 11, 1985 in an attempt to address the NRC's concerns in this area. The NRC staff review of the B&W Owners Group report has not been completed.

l t

e .

i The staff agrees with the IIT observation that, as of December 26, 1985 effective actions had not been implemented by SMUD to resolve concerns about the loss of ICS power. The staff is planning to complete its review of the BWOG submittal dated January 1985 in the near future in conjunction with the B&W plant reassessment. In addition, SMUD and the BWOG are reviewing this event in connection with the Rancho Seco restart from the December 26, 1985 event and the B&W reassessment programs, respectively.

9. USI A-47, " Safety Implications of Control Systems" (Sec. 7.1.7)

Unresolved Safety Issue (USI) A-47 was established to verify generically the adequacy of existing NRC criteria for control systems and to determine the need for additional requirements. Under A-47, analyses are to be performed for one reference plant from each NSSS vendor (for B&W-designed plants, the reference plant is Oconee).

The IIT report noted concerns primarily in two areas:

1. "In the case of B&W-designed plants, the 721 design of the ICS at the reference plant is substantially different than the 820 design used at Rancho Seco."

2. "In a number of cases, the long-tenn implications of various issues concerning the Rancho Seco control system were referred to USI A-47."

"For BAW-1564, the TAP (Task Action Plan) did not recognize that the staff's resolution of the long-term issues had been referred to USI A-47."

"It does not appear that the analysis performed under USI A-47 addresses all of the issues that had been deferred to it by ,

Bulletin 79-27, BAW-1564, or NUREG-0667."

The staff agrees with the IIT that the analysis performed for USI A-47 needs to be broadened, especially in the area of B&W-designed plants.

The staff has revised the A-47 scope of analyses to include the type 820 design of the ICS. Together with the type 721 design already included in A-47, this should cover the range of integrated control systems in use at B&W-designed plants.

With regard to the second IIT concern, the staff notes that no "long-term implications" or "long-term issues" to be referred to USI A-47 were ever specifically identified in BAW-1564, Bulletin 79-27, or NUREG-0667. The phrase "long-term implications" appears to refer to USI A-47 itself, i.e., the issue of adequacy of existing NRC criteria for control systems. However, the staff agrees that the present scope of USI A-47 may not address all of the technical issues that were identified in Bulletin 79-27, BAW-1564, and NUREG-0667. The staff is reviewing the coordination of A-47 with other recommendations made in BAW-1564, Bulletin 79-27, and NUREG-0667 and is also reviewing licensee implementation status of those recommendations to assure that the scope of USI A-47 is adjusted, if necessary, to adequately cover the important technical issues for B&W-designed plants.

L . . . . . , , .

- _ _ _ - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _