A07538, Forwards Addl Info Re Corrective Actions Taken & Plans for Further Corrective Action Re Items Identified During Insp of Plant Emergency Operating Procedures (Eop).Revs Made to EOPs Eliminate Number of Deficiencies Identified

Forwards Addl Info Re Corrective Actions Taken & Plans for Further Corrective Action Re Items Identified During Insp of Plant Emergency Operating Procedures (Eop).Revs Made to EOPs Eliminate Number of Deficiencies Identified

ML20195D026
Person / Time
Site:	Millstone
Issue date:	10/28/1988
From:	Mroczka E NORTHEAST NUCLEAR ENERGY CO., NORTHEAST UTILITIES
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
A07538, A7538, NUDOCS 8811040144
Download: ML20195D026 (28)

Text

.. .

NORTHEAST UTILITIES o.n.r.i On,c.. . seio.n sir i. s.,iin. Conn.ct,cui 1 $l555n5?rm 70_ gog 270 H ARTFORD. CONNECTICUT 06141-0270 b k J C.((Z,7,, (203) 665-5000 October 28, 1988 Docket No. 50-245 A07538 Re: E0Ps U.S. Nuclear Regulatory Comission Attention: Document Control Desk Washington, DC 20555 Gentlemen:

Millstone Nuclear Power Station, Unit No. 1 Emeroency Ooeratinu Procedures The results and conclusions of the special safety team inspection of Millstone Unit No. I's Emergency Operating Procedures (EOPs), which was conducted by the NRCStaffinJuneof1988,wereforwardedtoNorgastNuclearEnergyCompany (NNECO) in a letter dated September 23, 1988. NNECO was requested to submit our plans for corrective actions for each of the items identified in the Executive Sumary of the Inspection Report by October 28, 1988. It was requested that our corrective actions reflect the significance of the human engineering deficiencies identified in the E0Ps.

In a letter dated July 29,1988,(2) NNEC0 submitted information and a correc-tive action plan addressing what we perceived to be the more significant NRC concerns resulting from the audit team's exit meeting on June 30, 1988. Many of the NRC concerns were addressed in the revisions to the E0Ps which were implemented October 15, 1988. We hereby provide additional information addressing the concerns of the NRC Staff, including a discussion of corrective action already taken, and our plans for further corrective action relating to the items identified in the Executive Sumary of the Inspection Report.

In our July 29, 1988 letter, NNEC0 provided the NRC with a historical back-ground of our entire efforts and philosophies on Millstone Unit No. I's E0P development and implementation. Throughout this response to the Executive Sumary, we will refer to that document.

(1) S. A. Varga letter to E. J. Mroczka, "Emergency Operating Procedures inspection (Inspection Report 50-245/88200), dated September 23, 1988.

(2) E. J. Hroczka letter to U.S. Nuclear Regulatory Comission, "Emergency Operating Procedures," dated July 29, 1988.

een040144ADOCK0500g,{

eeic2e s A0[

PDR O I  ;

i

.. .c U.S. Nuclear Regulatory Commission A07538/Page 2 October 28, 1988 Item No. 1:

The revised verification, and validation procedures were not utilized to perform a corplete review of the E0Ps and, the operating procedures referenced by the E0Ps were never verified or validated. In addition, the rev' sed verification and validation procedures were not submitted to the NRC by July 31, 1985, as committed. (Section3.1.1)

Resoonse As described in our July 29, 1988 letter, we opted not to perform a new verification and validation (V&V) on our existing Revision 2 E0Ps as we were anticipating Revision 4 implementation and believed it would be inappropriate to subject the operators to sequential revisions of our E0Ps. For the procedure revisions that went into effect October 15, 1988, a V&V was performed on the sections revised. Also, all significant human factor discrepancies were corrected; however, due to time constraints we opted not to V&V the entire procedures. We are satisfied that our revised E0Ps address the significant concerns identified by the NRC.

With regard to the referenced operating procedures (ops), a V&V was never performed. We did not consider them as E0Ps in the past. As of this letter, all of the ops have been revised except for two and walked down by the operating staff. The two ops are in the process of being revised, and will be in effect by November 18, 1988. Due to time constraints, we decided not to do a complete V&V. For our Revision 4 E0Ps we will incorporate the required OP sections into a common E0P such that this procedure will be processed similarly to other E0Ps. The revised V&V procedures were forwarded in the July 29, 1988 letter to the NRC.

In summary, the revisions made to the existing E0Ps and ops eliminate a number of deficiencies identified and enhance the useability of the E0Ps.

With our commitment to make these changes and proceed immediately into an expeditious implementation of Revision 4 E0Ps, we deemed a full V&V of all procedures to be unnecessary and impractical at this time.

Item No. 2:

The maintenance of the E0Ps and the Plant-Specific Technical Guidelines (PSTGs) was not adequate in that:

a. The licensee did not appropriately control and maintain the PSTGs up-to-date as a design basis document. (Section3.1.2) u s

U.S. Nuclear Regulatory Commission A07538/Page 3 October 28, 1988

Response

The PSTGs and associated calculations have not previously been handled as design basis documents. The PSTG has been maintained current with sepa-rate write-ups, but not as a totally revised document. Henceforth, control of the PSTGs will be enhanced. The calculations will be controlled as QA documents. We have instituted Operations Instruction 1-0PS-3.07 which provides the necessary control and maintenance of all of the E0P support documents. This procedure includes as attachments, the Writer's Guide, PSTG, Technical Basis Document, and the Justification for Deviation Report. As of October 15, 1988, the EPG calculations (Appen-dix C), PSTG, and the justification for deviation documents have been revised and reflect the present E0Ps. Revision 4 of the E0Ps, PSTGs, EPG calculations, and Writer's Guide will be maintained as a design basis document.

b. The licensee had not implemented a formal program for ongoing review and upgrado of the E0Ps and the Quality Assurance Organization was not involved in the development and maintenance of the PSTG. (Sections 3.1.5 and 3.1.6)

Resoonse We are not in full agreement with the audit deficiency with regard to a formal program for review and upgrade of the E0Ps. Our biennial review procedure, OP-260, "Biennial Review of Operations Procedures," identifies the need to review and revise all procedures including the E0Ps. This process is performed by the E0P coordinator or designee who has remained current on all E0P issues. As described in our procedure guidelines, 0)erators are encouraged to submit changes as a result of the use of or o)servations during operations, training, simulator exercises or walk-throughs. In addition, the Training Department has a process that identifies problems or concerns raised during simulator instruction. t Operators are trained yearly on E0Ps and observed by instructors and plant management on their ability to use the E0Ps. This process is '

essentially a continuous V&V activity.

The programs in place are not specific to just E0Ps. We apply this process to all our procedures. To provide some clarification, we will revise 0P 260 or 1-0PS-3.07 by February 1,1989 to reflect some specific E0P review requirements. The Plant Design Change Process includes the requirement to review plant procedures for )otential impact. Again, there is not a specific E0P review requirec ; however, the procedure review requiremer.t does include E0Ps. Step 6.2.2 of ACP-QA-3.10 specifi- ,

cally states "operations and emergency procedures" will be modified as a)propriate and are to be listed on the PDCR form. As such, we plan no clange to ACP-QA-3.10.

U.S. Nuclear Regulatory Commission A07538/Page 4 October 28, 1988 Our Quality Services Department (QSD) was not directly involved in the development or maintenance of the PSTG. Due to the specialization of E0Ps, we do not believe the QSD should participate in the development or maintenance of the E0Ps. We agree that all of the various aspects of the E0P process should be controlled and processed with the necessary QA attributes as recommended in NUREG-0899. Procedures, including E0Ps, are reviewed, processed, and authorized for implementation in accordance with ACP-QA-3.02. This procedure implements the provisions of 10CFR50 Appen-dix B, Regulatory Guide 1.33, and ANSI N18.7.

Although not performed by QA personnel, our PSTG has been extensively reviewed by operators, plant engineers and corporate engineers. However, this process has not been documented and was not credited in the NRC audit. We will revise our E0P implementation process to document the review of the PSTG. We will make these changes by February 1, 1989. An audit of the E0Ps and E0P process will be covered in the audit program.

Revision 4 E0P and PSTG development and maintenance will be done consis-tent with and pursuant to approved Nuclear Engineering and Operation (NE0) and/or station procedures. These procedures are reviewed by the NU Quality Services Department to verify consistency with the Northeast Utilities Quality Assurance Program Topical Report. This process will also be applicable to any additional procedures that may be required to support Revision 4 implementation,

c. The number of interim changes to the E0Ps and the methodology of these changes increased the complexity of the E0Ps. Interim changes were not accomplished in accordance with the human factors' engineering guidance of the Writer's Guide. (Section3.1.3)

Resoonse All of the interim changes have been incorporated into the October 15, 1988 revision of the E0Ps. 1-0PS 3.07, issued October 7, 1988, describes the processing of interim changes for E0Ps. The procedure requires interim changes to be processed through PORC after all the necessary reviews normal to an E0P revision. We agree that the Writer's Guide and ACP-QA-3.02 do not agree. We will evaluate the interim change process relative to E0Ps and modify the ACP, Writer's Guide, and/or 1-0PS-3.07 by February 1, 1989.

d. Insufficient copies of the E0Ps were provided in the Control Room.

(Section 3.1.4)

Response

We will keep two official copies of the E0Ps in the Unit 1 Control Room, effective October 15, 1988.

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

U.S. Nuclear Regulatory Commission A07538/Page 5 October 28, 1988 Item No. 3:

The basis of the Primary Containment Control Guidelines of the E0Ps was deficient in that:

a. The licensee did not calculate the primary conta*nment pressure . limit in accordance with the guidance of Revision 2 of the iPGs. This significant deviation was not adequately evaluated, and the E0Ps were not correctly revised to incorporate this deviation. (Section 3.2.1.1.(1))

Response

The Primary Containment Pressure Limit (PCPL) has been revised. The details )rovided in Reference (2) show that using the Revision 4 method-ology, tie limit should be 62 psig. The revised curve also accounts for the instrumentation location.

The PCPL is identical to the Containment Design Pressure Limit. There-fore, E0P-580 has been revised to eliminate the Primary Containment Design Pressure. In addition, the actions in the E0Ps have been resequenced to avoid any confusion. The revised sequence is as follows:

o Before the Drywell Pressure reaches the PCPL, perform RPV flooding, o When the Drywall Pressure reaches the PCPL, initiate Torus and Drywell Sprays if allowed by their spray initiation curves.

o When the Drywell Pressure exceeds the PCPL, initiate venting of the primary containroent.

This hierarchy of action is consistent with the Revision 2 EPGs. These changes were effective October 15, 1988.

b. The licensee revised the original calculational assumptions of the pressure suppression pressure limit and the primar.v containment pressure limit without an appropriate evaluation of the eff $ct on the limits. In addition, the licensee changed the calculation of the minimum number of SRVs required for emergency depressurization w thout an evaluation.

(Section 3.2.3)

Response

The Pressure Suppression Pressure (PSP) curve has been replaced by a number (35 psig). This value repre:;ents the most limiting value of the PSP. Revision 2 of the EPGs allows the use of one value vs. a two dimensional PSP curve.

As discussed in Response 3a, the Primary Containment Design Pressure has been deleted and replaced by the revised Primary Containment Pressure

1 I

l U.S. Nuclear Regulatory Commission A07538/Page 6 October 28, 1988 Limit (PCPL). This revised limit accounts for the location of the Drywell pressure instrument tap.

For Millstone Unit No.1, the minimum number of SRVs required for depres-surization is 4. This requirement comes from the Small Break LOCA analysis performed for Millstone Unit No. I and approved by the NRC Staff. The Appendix C calculation has been appropriately modified to include this reference.

The calculations cpplicable to Pressure Suppression Pressure and Primary Containment Pressure Limit and minimum number of SRVs have been evalua-ted, appropriately documented, and incorporated in the E0P revision on October 15, 1988. The calculation files have been updated to reflect these changes.

c. The licensee did not effectively utilize plant-specific instrumentation to implement the correct primary containment water level limit. As a result, the licensee implemented an unrealistic limit for the maximum primary containment water level and modified the accident mitigation strategy of the EPGs without an adequate justification. (Section 3.2.

1.(2))

Resoonse:

The initial maximum allowable torus water level of 22.2 feet was chosen based upon instrumentation limitations. The scale of this instrumenta-tion has been extended to allow readings up to 27.2 feet. Effective October 15, 1988, the maximum allowable torus water lovel has been increased from 22.2 to 24 feet. A maximum torus water 1r. vel of 24 feet does not allow containment flooding to the extent tk.e core could be submerged. However, this limit does not preclude RP'! flooding. Rather, the procedure only requires the operator to terme. ate all injection from sources external to the containment when torus water level reaches 24 feet. Thus, LPCI and core spray, which take suction from the torus, can still be used in performing RPV flooding. The revised maximum torus water level limit of 24 feet is based upon the concern raised by the NRC in Item 3d below.

Millstone Unit No. I does not have drywell water level indication. Level indication is planned to be added during the 1989 refueling outage. With this design change, the operator will be able to flood the containment to increase the RPV level above the top of active fuel (TAF). Revision 4 of the E0Ps will reflect this change. We have reviewed the use of existing instrumentation to determine drywell level during containment flooding.

We have concluded that containment flooding with the current instrumenta-tion is not feasible for the following reasons:

o The current pressure instrumentation, with taps at the torus bottom and in the drywell, cannot be accurately correlated to the act 'al

e U.S. Nuclear Regulatory Comm usion A07538/Page 7 ,

October 28, 1988 containment level. Such a correlation would be too inaccurate when instrument uncertainties, especially under harsh environment, and containment water temperature are considered, o The drywell pressure tap is below the top of active fuel (TAF).

Therefore, we cannot flood-up containment to the TAF (which is the aim of containment flooding) with the current instrumentation.

d. Insufficient justification existed for the deletion of EPG Caution No. 23 concerning the operation of the suppression chamber vacuum breakers.

(Section 3.2.1.(3)) ,

Response

The intent of EPG Caution #23 applies to plants with internal vacuum breakers only and was not meant to apply to external vacuum breakers.

Since Millstone Unit No. I has external vacuum breakers, this caution was not included in the E0Ps. However, the NRC hai raised a new generic concern that requires further evaluation. The revised maximum torus water level of 24 feet has been derived to ensure proper function of the vacuum breakers, taking into account the NRC concern. A torus level of 24 feet is below the vacuum breuker penetration minus the vacut.m breaker opening pressure. If drywell sprays are initiated with torus level at r 24 feet, torus water will not enter the vacuum breakers. Therefore, proper operation of the vacuum breakers is ensured. We will investigate this concern further to determine if a similar limit needs to be main-tained in Revision 4 of the E0Ps.

Item No. 4:

The human factors engineering of the E0Ps was deficiant in that:

a. The inspection team identified numerous examples of inadequate implemen- ,

tation of the Writer's Guide because the lice nsee had not implemented verification and validation procedures for ECP actions in operational ,

procedures and had not fully implemented the procedures following revi- '

sion. (Section3.2.5)

Response

A V&V was not implemented using the revised Writer's Guide as discussed in the July 29, 1988 letter. As described in that letter, we opted not to rewrite our E0Ps to the latest revised Writer's Guide. This revised Writer's Guide did in fact address the concerns identified during the summer of 1985 NRC audit with Bate 11e Labs. The h te11e report w::s never formally issued or transmitted to NNECO. Expecting Revision 4 EPGs in the near future, we opted to leave our existir; procedures in place and l

p U.S. Nuclear Regulatory Commission A07538/Page 8 Octeher 28, 1988 address all concerns with our new procedures. Tg position was formally documented to the NRC Staff on January 31, 1985 We do not agree that these deficiencies could cause operator confusion.

We consider the procedures useable. Exercising these procedures on tl -

simulator on numerous occasions has proven their useability. We did not want to subject the operators to two revisions of E0Ps within a short time interval. We believe that leaving the Revision 2 procedures in place did not affect the ability of the operators to safely operate the plant or to properly respond to an accident based on our high experience level and exceptionally strong knowledge level.

For our October 15, 1988 commitment, we addressed .nost of the specific items identified in the audit along with many other human factor concerns. We made the appropriate E0P corrections but, due to time constraints, we did not revise the Writer's Guide nor perform a full V&V.

We did walk down each procedure as committed in our July 29, 1988 letter and ensured that the procedure matches each parameter, instrument number, location, alarm, or switch designation. We performed V&V on all sections of the E0Ps where technical changes were made.

b. The Writer's Guide provided inadequate guidance concerning the use of operator information, step highlighting, E0P entry conditions, the format of operator action steps, and E0P identifying information. (Section 3.2.

4).

Resoonse:

The E0Ps contain some inconsistencies with regard to operator informa-tion, step highlighting, E0P entry conditions, the format of operator action steps, and E0P identifying information. These inconsistencies were due to a lack of specific guidance in the Writer's Guide. As stated previously, our training, experience, and knowledge level ensure our ability to implement these procedures. Recent in house final simulator exams have demonstrated effective use of these procedures by our SR0 license candidates. The NRC inspection team acknowledged that they were impressed with the knowledge level of the plant staff relating to accom-plishment of the E0Ps and concluded that the operators could adequately perform the procedures. Thus, the inconsistencies have not hindered the operators' ability to use these procedures. Some of these inconsisten-cies were actually enhancements to the procedures, but were not consis-tently applied. The Writer's Guide for Revision 4 E0Ps will provide specific guidance to eliminate these inconsistencies.

(3) W. G. Counsil letter to J. A. Zwolinski, "Procedures Generation Package, Response to Safety Evaluation," dated January 31, 1985.

r l

U.S. Nuclear Regulatory Commission A07538/Page 9 Oaober 28, 1988 Item No. 5:

a. The inspection team identified three exar..ples in which special tools or equipment were not provided to support the accomplishment of E0P actions.

These examples included: (1) missing safety equipment and inadequate hoses for venting of the hydraulic control units as a method of alternate rod insertion, (2) missing special adapters for the connector of the fire main as an alternate water injection method, and (3) insufficient elec-trical jumpers for use in bypassing the MSIV isolation logic. (Sec-tion 3.3.2)

Response

As discussed in Section 3.3.2 of the NRC Inspection Report, some equip-ment was not properly staged. We have identified the necessary equipment required to carry out the unique actions required by the E0Ps. We expect to have all of the equipment available and staged by February 1, 1989.

We will also develop a semiannual surveillance to check for proper staging of this equipment by February 1, 1989. Description of these uniy e actions appear in various sections of our normal operating proce-dures. These procedures have all been reviewed, revised, and walked down to ve ify ocr ability to implement them.

With regard to locked valves, we have ordered 600 new locks, all operable with one key type, and will H stall them when they become available.

b. The 'icensee had not administ< ively cnntrolled operator aids to ensure that the aids were available, smained up-to-date and were not incorrect-ly revised or superseded. (Section 3.3.'t)

Resoonse:

As discussed in Sa, a semiannual wrveillance will be initiated by February 1,1989 to verify the avai14bility and proper staging of all necessary equipment. The following applies to specific comments in Section 3.3.3:

o As described in the Inspection Report, the necessary direction for mixhig sodium pentaborate was available for the operators use, however, it was not controlled. This direction has been issued as an operator aid as of September 8,1988 and is audited on a monthly basis, o The Operator Aid 1-87 concerning the Isolation Condenser level meter f.orrection was corrected in the plant and in the Operator Aid Log Book.

o With regard to the other two items identified as be'.ng operator aids, we do not agree with the Staff's Inspection Report and plan no

+

U.S. Nuclear Regulatory Commission A07538/Page 10 October 28, 1988 further action. The vessel level information is controlled by a PORC-approved procedure, OP 316 "Feedwater System," and the posted page is distributed officially at each procedure revision. The location of these posted pages is included in a distribution list.

With regard to the bakelite tags, OP-261, "Centrol of 0)erator Aids," specifically states that the Operations Supervisor will make the judgment as to the need to control any information as an opera-tor aid. Once permanently mounted, regular monitoring of this information is not required.

c. Manually-operated valves in the emergency service water system were potentially not accessible because their access covers were corroded in place. (Section3.3.4)

Resoonse:

The access covers described in Section 3.3.4 of the NRC Inspection Report have been hinged such that the operators can easily access the valves.

The need to perform this evolution would have allowed us more than ample time to gain access to the valves even in the condition found in the audit.

d. Procedures were not provided for complex operator tasks such as jumpering of contacts and bypassing interlocks and isolation signals. (Sec-tion 3.3.1.(1))

Respontg:

Procedural inadequacies identified in Section 3.3.1.(1) of the audit report have been revised and procedures generated as appropriate. All required jumpers have been prepared, labeled, and located in a designated storage area.

e. The operational procedures referenced by the E0Ps provided insufficient information concerning the correct methodology for alternate boron injection. (Section 3.3.1.(2))

Resoonse:

The equi > ment staged for sodium pentaborate injection was available and adequately stored and secured in a locked cage. Although the equipment was not monitored, the equipment would bc available for use in the E0Ps.

For consolidation purposes, an E0P equipment cage will be established containing equipment and materials necessary to carry out actions speci-fled in the E0Ps. A semiannual surveillance will be performed to ensure equipment availability.

E0P 578, "RPV Flooding," has been revised to correct each item identi-fied.

U.S. Nuclear Regulatory Commission A07538/Page 11 October 28, 1988 OP 302, "Control Rod Drive System," will be rev,. .lovember 18, 1988 to maximize CRD flow by closing feedws.ter block vanes.

OP 303, "Reactor Cleanup System," will be revised by Wyember 18, 1988 to correct the discrepancy.

OP 304, "Standby Liquid Control System," will still discuss the use of a hydrostatic pump. The staging of this pump is under investigation with consideration of availability and flow rates.

f. The licensee had not evaluated the accessibility of the secondary con-tainment following a LOCA in the process of developing the E0P's contin-gency actions. (Sectic,n 3.3.1 (3))

Resoonse:

Calculations performed using the source terms provided in NUREG-0737 predict dose rates which preclude access to the reactor building. Since access is precluded, all required actions following a design basis accident in the E0Ps are performed from outside the reactor building and hence survey maps for the re&ctor building were not developed. Thus, for design basis accidents, it is not necessary to perform radiological evaluations of contingency actions.

In regard to postulated dose rates for other conditions which may allow contingency actions in the reactor building, any dose rate between normal operation levels to the calculated NUREG-0737 inaccessible levels are pnssible. The most likely levels would be typical of normal operation as fuel failures are not expected and hence access to all areas of the secondary containment is possible. As such, preparing radiological evaluations for contingencies is not beneficial.

The E0P contingency actions address multiple, alternate methods for controlling the required parameters. These include actions from the control room, secondary containment and outside the secondary contain-ment. The required action (s) to control parameters are performed consis-tent with the plant operating environmen'. and equipment accessibility and operability using the existing evaluation procedures.

g. The inspection team identified numerous referencing errors between the E0Ps and supporting procedures and several examples of incorrect or inadequate lat,eling. (Section 3.3.1.(4) and 3.3.1.(5))

Re.toonse:

We have evaluated our approach to referencing of other procedures in the E0Ps. For abnormal operation of any system, we will reference the proce-dure number and title. The specific section will not appear in the E0P.

U.S. Nuclear Regulatory Commission A07538/Page 12 October 28, 1988 The index of the referenced procedure will have the section identified.

For normal system operation (i.e., normal feedwater flow to vessel), no procedure will be referenced in the E0P, as each system would be operated in its normal configuration. By referencing the abnormal system opera-tion procedures, the operators will recognize immediately that some unique evolution must be necessary.

The present Writer's Guide was not revised to reflect this change. As of the October 15,1988 E0? revision, all references have been revised to reflect the preceding paragraph. For our Revision 4 procedure develop-ment, we intend to use a similar approach except that the abnormal system operations will all be incorporated into one E0P procedure. This way, we will only reference one procedure rather than branching to a number of procedures. We believe this approach will enhance our ability to control these actions with all of the requirements of an E0P including the necessary V&V.

With regard to labeling, we have corrected each of the specific items identified in the NRC audit as of the October 15, 1988 revision to the E0Ps. We also verified each item listed in all of the E0Ps is labeled

correctly and is consistent with the procedure wording by performing a walkdown. Therefore, a full VaV was not deemed necessary.

l h. The inspection team identified discrepancies with respect to the adequacy of control room instrumentation because a detailed control room design review had not been completed. (Section3.3.1.(6))

Resoonse:

)

i As you are aware, the control room design review (CRDR) is presently

( underway but has not yet been completed for Millstone Unit No.1. Of the l three items identified in the Inspection Report, Section 3.3.1(6),

incorrect identification of drywell temperature instrument points were identified as a technical deficiency in our July 29, 1988 letter. E0P 580 has been revised to address this concern. The other two are minor deficiencies and will be evaluated along with the other human engineering discrepancies identified during the CRDR.

General Response to 3.3.1 We agree that most, if not all, of the items discussed in your Section 3.3.1 would have been identified with performance of a thorough V&V. Our July 29, 1988 letter discussed the history of our process describing what we did and why. Due to time constraints, we again opted not to perform a full V&V on our current E0Ps. We have spent the last three months expending resources in a manner that resulted in the implementation of the most beneficial changes to the E0Ps. That effort was intense and afforded us an end product that is an improvement.

l

i U.S. Nuclear Regulatory Comission A07538/Page 13 October 28, 1988 Our next evolution of E0Ps has already begun. To implement Revision 4 E0Ps, a majur resource commitment is required. This effort will incluae a full, thorough V&V addressing the concerns raised by the audit. We do not believe that expending any additional resources now doing a full V&V on our current E0Ps will measurably enhance our procedures, especially in light of our near-term implementation of Revision 4 E0Ps. As recognized on page 1 of your audit, "the operators could implement the required "

actions of the E0Ps...". The enhancements to these same procedures further strengthen our position that our E0Ps are useable by our opera- [

tors and would not cause them any confusion. We believe that our opera- '

tors can safely mitigate the consequences of an accident situation with use of the existing E0Ps.

l Item No. 6:

Several deficiencies were noted during the E0P accident scenarios. These included:

a. The E0P simulation demonstrated that the shift manning described by the Technical Specifications was adequate to accomplish the required actions of the E0P's, however the inspection team concluded that insufficient l personnel were available to accomplish all the actions required in an  !

emergency. Specifically, effective implementation of the Emergency Plan or activation of the Fire Brigade coincidental with implementation of the E0Ps could not be performed if only the minimum shift crew described by Technical Specifications were available in the control room. The duties  ;

of the Shift Supervisor Staff Assistant (SSSA) were essential to the satisfactory performance of the E0Ps; however the SSSA was not included j in all administrative procedures or in the Technical Specifications as a '

i required member of the minimum shift crew. (Section3.4.2.(5))

Resoonse:

l

We have reviewed the shift manning concerns and believe that our existing -

manning levels are adequate. The Shift Supervisor Staff Assistant (SSSA) {

is an important asset to the minimum suft complement. He plays an t important role during implementation of the emergency plan; however, he I has no responsibilities relative to the satisfa: tory performance of the ,

E0Ps. The SSSA position need not be incorporated into Technical Specifi-cations. However, by April 1,1989, we will revise Operations Instruc-tion 1-0PS 1.09 to reflect the responsibilities of the SSSA. As discus-sed in our operations instruction, we always schedule one extra operator  ;

on every shift, exceeding the minimum required shift complement. The ,

fire brigade would 'se supported by operators from Hillstone Unit Nos. 2 and 3, especially if Unit No. I was in an E0P situation. Our emergency plan also directs additional personnis1 to be called in to provide the

) necessary manpower. As presently manned, we believe our crews can safely and effectively mitigate the consequences of any accident scenario. No staffing changes are currently being considered.

3

U.S. Nuclear Regulatory Commission A07538/Page 14 October 28, 1988

b. The control room responsibilities assumed by the operators for E0P implementatior. differed from the method identified in the administrative procedures. (Section 3.4.2.(2))

Response

We do not fully agree with this item. During each scenario, the Shift Supervisor (SS) maintains the "big picture" at all times. This does not >

mean the SS cannot silence an alarm or read an indication while monitor-ing the panels if other operators are busy. He must remain in control and knowledgeable of the status of the entire plant at all times. The supervising control operator (SCO) did read the procedures and provide the necessary direction to the control operators (COs). At certain steps throughout the E0Ps, the SCO at times will just verify a particular action was taken or a parameter being within an expected range without giving specific direction to a CO. Many of the specific actions may have already been taken by direction of the SS before the SCO proceeds in the procedures. He must ensure that each step of the E0P is accomplished.

The audit team's conclusion that a particular step was misir.terpreted because the E0Ps were being used as a verification method is not correct.

The operators have indicated that they were aware of the need to termi-nate feedwater injection; however, they did not terminate because sodium pentaborate was not available for injection. The SS and SCO made that decision based on the conditions at the time. Their actions were not related to improper use of the E0Ps. We will review our administrative procedures to determine if any clarification is required with regard to control room responsibilities. This review will be completed by April 1, 1989.

c. The licensee failed to effectively train and implement a method for placekeeping during the performance of the E0Ps. (Section 3.4.2.(1))

Response

We recognize that placekeeping while using the E0Ps can be cumbersome.

We have advised the operators that the line on the left of each step is specifically for placekeeping and it should be used as such. This will be further stressed during licensed operator training. With the im)1e-mentation of Revision 4 procedures, improvements in placekeeping will be considered.

d. The immediate operator actions required by the abnormal procedures in response to the loss of feedwater were more conservative than specified

, in tha E0Ps because alternate high pressure feedwater systems were not avail:ble. This difference in the plant-specific equipment was not l addressed as a significant deviation from the EPGs. (Section3.4.2.(3))

l l

. O U.S. Nuclear Regulatory Commission A07538/Page 15 October 28, 1988

Response

ONP procedures are event-specific and written to provide an optimum response to that specific event. Often, the training provided on ONPs will assure a more serious event requiring E0P entry does not occur. A transient being addressed in an ONP may or may not lead to an E0P entry.

If E0P entry does occur, subsequent actions specified in the E0Ps do not provide any conflicts, as shown through our simulator experience.

For example, during the simulator exercise of a loss of feedwater tran-sient, the operators closed the MSIVs. If this does not occur rapidly, the HSIVs would close a short time later, once reactor level reached the low low level setpoint. Our approach has been to phimize the inventory loss through training on ONPs. By isolating the MS; before the isola-tion signal is reached, we eliminate the cha11ange to many safety systems. If the isolation point is reached, drywell cooling is lost which will exacerbate containment conditions, emergency power sources start, cleanup system isolates, all ECCS pumps start and automatic depressurization logic may be initiated. By responding to this event early, normal system availability is maximized while challenges to safety systems are minimized. We expect the operators to follow the ONPs until the entry conditions of the E0Ps are met at which time the operator will proceed in the E0Ps. Knowing the capabilities of our plant, we believe this is the proper way to operate.

We do not consider this a deviation from the EPGs in that our PSTGs and E0Ps follow the EPGs accordingly. If our off normal procedure gu. Nance did not mitigate the loss of feedwater transient, then the E0P would be entered and the guidance of the EPG/EOP would be adhered to. The ',econd action of that E0P would be to confirm that any isolations did occur such as the HSIVs closing. In our Technical basis document, used primarily for information and training, we have specifically advised the operator that "manual isolation may be advisable in some circumstances even when no automatic signal is generated." NUREG-0899 specifically states "In some cases, the event may appear to be obvious, in which case the opera-tor i.iay wish to use a procedure that deals with tne event immediately and directly. In these cases, safety function should be monitored continu-ously and concurrently with the event-oriented procedures." Our approach to a loss of feedwater is well within this guidance.

Based upon this discussion, we do not believe it is a deviation from the EPG or PSTG. In general, our primary objective is to conserve inventory.

If the operator is alert enough to recognize the loss of feedwater and isolates the reactor before the isolation signal is generated, a more conservative approach results. We do not believe m operator violates any E0P step by performing this action as an ONP prior to entering an E0P. We believe that our 0NPs and E0Ps are well coordinated. We believe that our 0NPs sbuld direct the operators through any normally expected transient without allowing the conditions to deteriorate to the E0P entry

U.S. Nuclear Regulatory Commission A07538/Page 16 October 28, 1988 level. The operators are highly skilled and kno::ledgeat:1e and are er.pected to mitigate any transient to the extent possible. Our procedur-al approach provides that guidance. ,

e. The licensee incorporated the secondary containment guidelines concerning depressurization of the reactor pressure vessel upon high seconda y temperatures and radiation levels as an abnormal operating procedure differently than specified in Revision 3 of the EPGs without justifica-tion as a significant deviation. (Section 3.4.2.(4))

Response

As discussed in response 6d above, we have developed procedures to ,

provide the operators with as much gudance as possible to mitigate the consequences of a transient. The potential for a high pressure leak existed long before Revision 3 of the EPGs. ONP 516. "High Energy Pipe .

Ru)ture," provides the operator with the necessary guidance to respond ;o a 11gh energy leak in the reactor building. Our approach is conservative -

and provides for an adequate safety margin.

As you are aw:tre, we have not implemented Revision 3 of the EPGs, there-fore, our existing ONP 516 is not considered a deviation from the EPGs.

The existence of this ONP, considering that we have not implemented Rovision 3 of the EPGs, is indicative of a strong approach to providing as much sound and relevant guidance to the operators as possible. Our actions are conservative and provide the operators with appropriate guidance to evaluate the situation and make a judgment before emergency >

depressurization of the reactor. When our Revision 4 procedures are implemented, ONP 516 will be evaluated and revised or cancelled accord-ingly.

Summary In our July 29, 1988 letter, we indicated that we would pursue implementation of Revision 4 E0Ps by September 1, 1989, or three months after the 1989 refueling outage, whichever occurs later. As a result of discussions witn the NRC Staff and their desire for earlier implementation of Revision 4 E0Ps, we ,

will pursue a gaal of implementing Revision 4 E0Ps by start-up from the 1989 L refueling outage. However; we emphasize that several factors could delt.y the implementation of our Revision 4 E0Ps, including the following:

o The appropriateness of implementing some or all of these procedure l changes under the provisions of 10CFR50.59, even with the issuance of the .

NRC's Sr.R on Revision 4 of the EPGs, has yet to be established.

l o Problems with the implementation of plant design changes required to [

support implementation of Revision 4 of the EPGs could impact our sched-  !

ule. A number of hardware modifications are being evaluated for possible [

implementation prior to revising the E0Ps. i e

r _- - - _.- ,_ m.__m - - -

U.S. Nuclear Regulatory Connission A07538/Page 17 October 28, 1988 o The Control Room Design Review (CROR) could result. in significant plant design changes and have corresponding effects on the E0Ps.

o Optimum use of the Isolation Condenser, especially during a station blackout, needs to be evaluated prior to implementation of Revison 4 E0Ps.

Near-term E0P fixes were implemented on October 15, 1988. Good faith efforts will be made to implement Revision 4 of the EPGs upon start-up after 1989 refueling outage. However, if the NRC Staff intends to issue a Cor firmatory Action Letter, we request that the date of September 1, 1989, or three months after the 1989 refaling outage, whichever occurs later, be used.

As stated in our July 29, 1988 letter, many items identified by the audit were not due to oversights by NNECO. Conscious decisions were made to eliminate any lass of procedure credibility while assuring that technical adequacy was not compromised. Further improvements are necessary. The corrective action plan contained in this submittal represents a major step to accomplishing these improvements. We believe that a meeting to discuss the Inspection Report, this letter, and our action plan would be prudent. We will discuss this with the NRC Staff.

In conclusion, we believe that our operators maintain a high level of experi-ence and knowledge of Millstone Unit No.1 operation, and have been effective-ly trained on existing E0Ps. Thus, public health and safety continues to be protected.

Very truly yours, NORTHEAST NUCLEAR ENERGY COMPANY A 4-E . 7. /

Senio/r5czka r Vice President cc: W. T. Russell, Region I Administrator M. L. Boyle, NRC Project Manager, Millstone Unit No. 1 W. J. Raymond, Senior Resident inspector, Millstone Unit Nos.1, 2, and 3 L. Kolownauski, Resident Inspector, Millstone Unit No. 1