ML17333A910

Forwards Response to Violations Noted in Insp Repts 50-315/97-04 & 50-316/97-04.Corrective actions:post-trip Recovery Procedures Will Be Revised Re Placement of TDAFP in Standby Readiness

ML17333A910
Person / Time
Site:	Cook
Issue date:	06/05/1997
From:	Fitzpatrick E AMERICAN ELECTRIC POWER CO., INC.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
50-315-97-04, 50-315-97-4, 50-316-97-04, 50-316-97-4, AEP:NRC:1260C, NUDOCS 9706090357
Download: ML17333A910 (40)

Text

Indiana Michigan Power Company 500 Circle Drive Buchanan, Ml 491071395 INblANA MICHIGAN POWER June 5, 1997 Docket Nos.: 50-315 50-316 U.S.Nuclear Regulatory Commission ATTN: Document Control Desk Washington,-D.-C.

-20555 Gentlemen:

AEP:NRC:1260C 10 CFR 2.201 Donald C.Cook Nuclear Plant Units 1 and 2 NRC ZNSPECTZON REPORTS NO.50-315/97004 (DRP)AND 50-316/97004 (DRP)REPLY TO NOTZCE OF VZOLATZON This letter is in response to a letter from J.L.Caldwell, dated May 6, 1997, that transmitted a notice of violation and a notice of deviation to Indiana Michigan Power Company.The notice of violation contained a total of eight violations of NRC requirements identified during an NRC inspection conducted from February 16, 1997, through March 29, 1997.The violations pertain to procedures, corrective actions, reportability requirements, and 10 CFR 50.59.issues.Our response to these violations is provided in attachment 1.The notice of deviation involves inoperability of control room power range pen recorders.

Our response to this item is provided in attachment 2.EE+pW E.E.Fitzpatrick

'1ice President SWORN TO AND SUBSCRZBED BEFORE ME~=-" TEZS.~g DAY OF 1997 Notary Public vlb UNDA L BOIlCKE Norory Public, Berrlen Coonly, Ml Attachments My Commr&on Iorpires jonoory 21, 200I 9'706090357 970605 PDR ADOGK 050003i5

1ndiana Michigan Power Company 500 Circle Drive Bvchanan, Ml 491071395 INDIANA NICHIGAH POWER May 5, 1997 Docket Nos.: 56-315 50-316 U.S.Nuclear Regulatory Commission ATTN: 33ocument Control Desk-Washington,--D.--C;-20555 Gentlemen:

AEP:NRC:3.260C 3.0 CFR 2.201 Donald C.Cook Nuclear Plant Units 1 and 2 NRC INSPECTION REPORTS--NO.

50.-3/5/97004

-(DRP)AND 50"316/97004 (DRP)REPLY TO NOTICE.OF VIOLATION This letter is in'response to a letter from J.L.Caldwell, dated May 6, 1997, that transmitted a notice of violation and a notice of deviation to 1ndiana Michigan Power Company.The notice of violation contained a total of eight violations of NRC requirements identified during an NRC inspection conducted from February 16, 1997, through March 29, 1997.The violations pertain to procedures, corrective actions, reportability requirements, and 10 CFR 50.59 issues.Our response to these violations is provided in attachment 1.The notice of deviation involves inoperability of control room power range pen recorders.

Our response to this item is provided in attachment 2.E.E.Fitzpatrick

'1ice President SWORN TO AND SUBSCRIBED BEFORE ME THIS DAY OF 3.997 Notary Public vlb UNDA l SOEt,CKE No&y Pubhc, Bergson Cooniy, Ml Attachmentsg QyCpzmi+~~fQ$

PDR ADQCK 050003i5 8',, PDR;, n'j>QQ5 Illlmllll!

Iillllllllllljlll(lllllll

U.S.Nuclear Regulatory Commission Page 2 AEP: NRC: 1260C c: A.A;Blind A.B.Beach MDEQ-DW&RPD NRC Resident Inspector J.R.Padgett~~l><l

ATTACHMENT 1 TO AEP:NRC:1260C RESPONSE TO NOTICE OF VIOLATIONS

~~

Attachment 1 to AEP:NRC:1260C Page 1 During an NRC inspection conducted from February 17, 1997, to March 29, 1997, four violations of NRC requirements

'ere identified.

In accordance with the'."General Statement of Policy and Procedure for NRC Enforcement Actions", NUREG-1600, the violations are listed below.NRC Violation 1a"10 CFR 50 Appendix B, Criteria V, Inspections, Procedures, and Drawings, requires in part, that activities affecting quality shall be prescribed by procedures of a'type appropriate to the circumstances and shall be accomplished in accordance with these---=---=--procedures;--

Contrary to-the above, The inspectors identified that Procedure 02-OHP 4023.ES-01"Reactor Trip.Response", revision 11, dated 11/21/96, was not appropriate to the circumstances because it did not contain guidance for adequately controlling steam generator (SG)levels while actions were being taken to minimize the reactor coolant system cooldown rate.As a result, on March 11, 1997, a Unit operator reset a turbine driven auxiliary feed pump (TDAFP)too close to the low-low SG level setpoint which resulted in an inadvertent Engineering Safeguard Feature actuation.

This is a Severity Level IV violation (Supplement I)." Res onse to Violation 1a 1.dmission or Denial of the Alle ed Violation Indiana Michigan Power Company admits to the violation as cited in the NRC notice of violation.

2.Reason for Violation This violation resulted from incomplete guidance in procedure 02-OHP 4023.ES-O.l,"Reactor Tri'p or Safety Injection", that allowed the restoration of the TDAFP prior to the unit being in a stable condition.

During the performance of 02-OHP 4023.ES-0.1, the control room team is allowed to remove the TDAFP from service if sufficient feedwater is being supplied to the SGs from the two motor driven auxiliary feedpumps.

This flexibility to remove the TDAFP from service provides the operators with additional reactor coolant system (RCS)temperature control.Technical specifications (T/Ss)3.7.1.2 and 3.3.2.1 require the TDAFP be operable and capable of automatically starting in mode 3.To comply with these requirements, ES-0..1 directs the TDAFP governor to be reset and the valve alignment to meet the standby readiness requirements.

The auto start function is enabled af ter all standing automatic start signals have cleared.During the post-trip scenario the standing automatic start signals are the low-low SG level on.two ef.four SGs,~and,the.mticipated.t ransient without" scram mitigatien'ystem actuation circuitry (AMSAC)signal.The

Attachment 1 to AEP:NRC:1260C Page 2 3~AMSAC signal occurs after all high power trips and is only required above 40%power.The AMSAC signal is then cleared manually during the performance of ES-0.1.The SG low-low level actuation signals are cleared by recovery of SG levels, utilizing the AFW pumps.During the post trip recovery on March 11,'997, the AMSAC signal was reset prior to the complete recovery of all SG levels to above the low-low automatic actuation setpoint.The¹21 SG level lagged the others, as, the loss of main feedwater to that SG was the initiating event which resulted in the reactor trip, and continuous feeding of.the SGs was in progress-to-recover=secondary side inventory levels.While filling the SGs, small.oscillations normally occur in the sensed level.With the¹21 SG level still below the low-low setpoint,,a.small oscillation occurred in¹23 SG that caused the TDAFP auto start signal to clear at its high point, followed by.the engineered safety feature (ESF)actuation when it subsequently dropped and went below the ESF setpoint.Because the setpoint has a 1%reset deadband, it is'extremely sensitive to minor oscillations.

Due to the incomplete guidance provided..:in the emergency.procedure,-emphasis was placed on the restoration of.the TDAFP to standby readiness, rather than on stabilizing SG levels above the ESF actuation setpoint prior to securing the TDAFP and placing it in standby readiness.

Corrective Action Taken and Results Achieved 4~The TDAFP started as designed and performed its desired function.Manual control of th'e SG levels during the post trip recovery continued.

No immediate corrective actions were required.Corrective Actions to Avoid Further Uiolations The post-trip recovery procedures will be revised regarding placement of the TDAFP in standby readiness.

These revisions will allow operators flexibility in equipment management during post trip responses, so that the operator may focus attention on the plant response as post-trip stabilization occurs, while continuing to meet the requirements of the T/Ss for auxiliary feedwater and ESF actuations.

An engineering review of the SG low-low level instrument deadband is being performed.

The purpose of the review is to determine the appropriateness of the 1\reset deadband.This review will be completed prior to the next scheduled calibration surveillance of the associated instruments.

5.Date When Full Co liance Will Be Achieved Full compliance will be achieved by September 1,, 1997, with.the completion of the engineering review of the reset deadband, and the revision of the appropriate post trip recovery procedures.

F we'll 4 d

Attachment 1 to AEP:NRC:1260C Page 3 NRC Violati.on 1b"On March 23, 1997, the inspectors identified that the licensee failed to follow, instructions when personnel woxking adjacent to the refueling cavity in a foreign material exclusion zone, failed to secure light hand tools to themselves by way of a lanyard or tagline, and failed to restrain tools in, the FMEZ when they set the'ools down.These actions were required by Plant Manager's Instruction (PMI)2220,"Foreign Material Exclusion", revision 9, dated 3/26/96.This is a Severity Level IV violation (Supplement I)." Res onse to NRC Violation 1b 1~A Admission-or

'Denial of the Alle ed Violation Indiana Michigan Power Company, admits to the violation as ci.ted in the NRC notice of violation.

2.Reason for the Violation 3.Contract technicians, under I&M supervision, were, making repairs to a dual view camera fixture in a foreign material exclusion zone (FNEZ)when they were observed using hand tools with lanyarda attached to the.tools, but not secured to a person or fixed object.This condition resulted from a misi.nterpretation of the requirements of plant procedure 12 PMP 2220.001.001,"Foreign Material Exclusion" (FNE).Section 5.2.7 of this procedure states, in part,"Light hand tools shall be secured'to the person using them by way of a lanyard or tagline.".However, fuxther on in the same procedure under a section entitled"Securing Tools" (attachment 2, part 6a)it is stated"Tools or equipment which could fall into openings beyond the reach of personnel MUST be secured with a lanyard or tag line, where practical."'he lanyards were felt to be.impractical by the workers~involved in the job.Because attachment 2 did not require lanyards where impractical, the workers did not use them.Additionally, these same contract technicians were observed leaving tools lying loose within an FMEZ.The~persons involved had incorrectly assumed that the"intent" of the FNE procedure was being followed by the compensatory actions they had taken prior to beginning the equipment repair.These actions included: 1)establishing a laydown area within the FMEZ for the specific purpose of repairing this equipment; and 2)assigning an individual to specifi.cally monitor and control loose parts and tools during the repair evolution.

Similar FME practicea had been employed at other nuclear sites.However, the Cook Nuclear Plant procedure that governs activities within an FNEZ (12 PMP 2220~001.001)specif ically mandates the use of lanyards, and does not-.recognize other methods of material control.Corx'ective Actions Taken and Results Achieved Upon notification of the NRC inspectors'oncerns, the project management a installation services (PMRIS)production supervisor contacted.

the contractor's site coordinator,-who reins tructed the te'chnicians on Cook Nuclear Plant FNE

Attachment 1 to AEP:NRC:1260C Page 4 4, requirements.

No additional problems relating to hand tool usage were recorded during the remainder of the project.Corrective Actions To Avoid Further Violations Proce'dure 12 PMP 2220.001 will be revised prior to the fall 1997 unit 2 outage.This revision wilI eliminate th" d screpancies noted within the procedure, and provide the i e" flexibility for using other methods of material control.On May 27, 1997, a plant-wide'-

>>time-out" was held to highlight management'.s expectations in the area of procedure c mpliance.-During this-period, plant and contract employees (including supervision) were brought together to focus on the usage of plant procedures.

PMZ-2011,"Procedure

'se and Adherence", was reviewed.Emphasized topics included the various.levels of procedure usage (continuous use, information

'use, referende use)and the company policy of strict procedural compliance.

Additionally, PM&IS will hold another procedural compliance

>>time-out" prior to the fall 1997 unit 2 outage.Procedural adherence issues will be re-emphasized to both ZaM and contract personnel (including supervision), as well as to individuals brought in specifically for outage support.Within thirty days of the end of the outage, PM&IS will also perform a self-assessment in the area of procedure adherence to determine the effectiveness of our procedural compliance efforts.Date When Full Com liance Will Be.Achieved Full compliance was achieved on March 23, 1997, after all p ysical work had been stopped and the workers'were reschooled on'Cook Nuclear Plant FME requirements (PMZ-2220) and our policy regarding strict procedural compliance.

NRC Violations 1c and 1d"On March 11, 1997, the licensee identified that during refurbishment of 1-QRV-114, the reactor coolant'xcess letdown to excess letdown heat exchanger shutoff valve, in 1994, the valve was reassembled without a cage spacer that was required by maintenance procedure 12 MHP-5021.001.057,"Copes-Vulcan Isolation Valve Maintenance>>'evision 1, dated 3/14/97'his is a Severity Level IV violation (Supplement I).1d.On March 16, 1997, the licensee identified that during the 1995 refurbishment of 1-NRV-163, the pressurizer spray control valve, the valve was reassembled without a cage spacer that was required by maintenance procedure 12 MHP-5021.001;126,"Copes-Vulcan Bellows Seal Control Valve Maintenance", revision 1, dated 3/13/97.This is a Severity Level IV violation (Supplement I)."

Attachment 1 to AEP:NRC:126QC Page 5 Res onse to C Violation 1c and Zd Admission or Denial of the Viol'ations Indiana Michigan Power Company admits to the violation as cited in the NRC notice of violation.

Reasons for the Violation This violation was caused by standards and expectations for contract valve technician performance of work to an in-hand procedure being too low.Proper implementation of, the-procedures-by-the-technicians was not verified and reinforced by the first line supervisors.

An additional factor included the valve technician's lack of familiarity with the specific configuration of this style of valve.'U Normal maintenance

'ractice for Copes-Vulcan valve disassembly is to remove the bonnet with the stem intact.This also includes removal of the plug, cage assembly, and cage spacer.During a normal refurbishment the plug and cage assembly are replaced.In these cases, the easiest way to disassemble the internal parts is to cut the stem and let the plug and cage assembly fall into a radwaste container.

This usually means that the cage spacer also falls into the waste container.

The replacement cage, disc, and stem are normally provided together as a"trim assembly".

Because the cage spacer does not see the wear that the plug and cage assembly see, it does not normaLLy need to be replaced during a refurbishment.

Therefore, th'e cage spacer is not included with these parts in a trim assembly.The existing cage spacer must generaLLy be reused when.the valve is reassembled.

Copes Vulcan valves have a unique cage.spacer configuration, which the technicians

~did not commonly work with.Nonetheless, the procedure does specifically call for reinstallation of the cage spacer as part of reassembly of the valve internals.

3.Corrective Action Taken and Results Achieved 4~1-QRV-114 was properly reassembled, with new internals, under JOA R36179-02.

This was completed on March 18, 1997.1-NRV-163 was propeily reassembled, with new internals, under JOA C34692-02.

This was completed on March 27, 1997.Corrective Actions Taken to Avoid Further Violations Two Copes-Vulcan valves have been purchased for training purposes.One valve is configured as a"typical" Copes-Vulcan control valve.The other valve is a duplicate configuration of the pressurizer spray valves.Designation of the cage spacer will be in bold in the reassembly step in Maintenance procedures for Copes-Vulcan valves.A review'f.the maintenance'procedures for Copes-Vulcan valves will be conducted.

Emphasis wilL be on consolidation

Attachment 1 to AEP:NRC:1260C Page 6 5.of the piocedures and implementation of engineering, plannihg, or supervisory identification of applicable procedure information based on the internal conf iguration and application of the valve.This.will be completed b September 1, 1997.e e y Maintenance personnel have been reminded of the need to'roperly implement in-hand procedures.

This means they must read the step, perform the step, document completion of the step, then proceed to the next step.At the time of the original valve work in 1994, contract supervisors..performed-hands-on work=as well's serving as supervisors.

Since 1994, this has been changed and contract supervisors no longer perform hands-on work, but function re l solely in an oversight role.This is reinforced thr h oug 8 gu ar meetings held during the outage.The contr ct n rac bri upervisors are now more involved in preparation and p-'re-jo er'efings, and general expectations for contract p formance, especially regarding procedural adherence,,is or discussed.

with contract management prior to the start of the outage.Date when Full Com liance will be Achieved Full compliance was achieved on March 27, 1997.At that time, both valves were properly reassembled.

NRC Violation 2a"10 CFR 50 Appendix B, Criteria XVZ, Corrective Actions, requires in part, that"Measures shall be established to assur that Zn the case of signifidant conditions adverse to qu 1't th (cor (rective)measures shall assure that the cause of the condition is determined and corrective action taken to preclude repetition." II Contrary to the above, a.On March 11, 1997, in Unit 2, the previous corrective actions to preclude the buildup of electrostatic discharge.from affecting Taylor Mod 30 controllers were ineffective in preventing the failure of the controller for feedwater regulating valve 1-FRV-210.

This controller failure caused the closure of 1-FRV-210 and a subsequent reactor trip." This is a Severity Level ZV.violation (Supplement Z)." Res onse to NRC Violation 2a Admission or Denial of the Alle ed Violation Zndiana Michigan Power Company admits to the violation as cited in the NRC notice of violation.

Reason for the Violation The cause of this violation'is an inadequate root cause determination for the previous controller.

failures ca The iroot cause determin'ation had'identified the static electricity but

Attachment 1 to AEP:NRC:1260C Page 7~0 3.failed.to identify the severity of the problem.Steps had been implemented to reduce the occurrence of static electricity.'owever, not.all", processes that could cause static were identified.

Although measures had been taken to reduce static buildup and to provide a means to safely discharge the static, some day-'o-day practices that could generate static were not identified, nor was it identified that the methods provided to discharge the static were not always effective.

Zt had been verified that the carpet installed in-the control rooms was a static dissipative carpet, humidity levels in the control~corns

-were being maintained above 40%, and electrostatic discharge (ESD)mats had-been.placed in front of the control panels.However, after the unit trip, it was discovered the controls of the steam generator'level ,controllers were located at a convenient height to make it common practice.for operators to roll.over to the controllers in'-wheeled office chair and adjust the controls.This rendered the static dissipative carpet and ESD mats installed in front of the control panel ineffective at dissipating static electricity.'ngineering had also instructed the operators to discharge their static charge on the control panel prior to.contacting controllers but failed to note the painted surfaces on the control panel-did not provide for proper grounding; Additional grounding methods for the controllers had been developed to reduce the vulnerability of, the controllers to failure during ESD.An'implementation schedule was developed, based on the need to remove a controller from service to perform grounding.

Because of this, a number of controllers could not be done with the unit operating.

This was judged to be acceptable in view of the actions taken to reduce static buildup and providing a means to di.scharge the static prior to an operator interfacing with the controller.

The controller that failed and caused the March 11, 1997, unit trip was scheduled for the grounding enhancement during the next refueling outage.Corrective Ste s Taken and Results Achieved The enhanced grounding methods were installed in unit 2 during the forced, outage from the'ontroller failure and on unit 1 during the refueling outage.Additional in-house testing of the controller confirmed the manufacturer's identification of ESD sensitivity at the right edge of the faceplate.

Testing also showed that sealing the edge of the faceplate prevented static intrusion and doubled the immunity to static discharge.

All panel mounted controller faceplates for both uni.ts were sealed to prevent static intrusion.

Additional'SD readings were taken in the control rooms while operators were performing routine activities, to more thoroughly quantify the static problem.Testing showed an operator could generate 3KV with a simple act of standing up from a chair.Static electricity also failed to,immediately

~drain,while standing;on:>, the anti-static-.-carp'et;-'.and

'took-several=seconds to.drain while standing on the ESD grounding 0

Attachment 1 to AEP:NRC:1260C Page 8 mats due to the insulated shoes worn by most operators.

Following testing, ESD-proof chairs were installed in the control room and operators were'.required to wear commercial shoe grounding straps.Follow-up checks indicated that while operators are wearing the grounding strap, static charge buildup would dissipate immediately on contact with the ESD mats and there was no charge buildup while using the ESD'hair.As a point of information, a design change is being finalized to incorporate a.failover control system design to prevent single point controller failure in critical instrument, loops from=shutting-down-the control loop.Failed controllers will be bypassed with-operator notification and, depending on which controller failed, continue in auto or revert to manual for operator control.4.Corrective Actions To Avoid Further Violations The cause of this violation was failure to properly identify and fully characterize root causes of the failure.A review and revision of Cook Nuclear Plant PMI-7030,"Corrective Action Program," was recently completed and additional training of personnel in proper root cause analysis is being performed.

5.Date When Full Co liance Will Be Achieved Full compliance was achieved on May 9, 1997, with the completion of the grounding modifications during the unit 2 forced outage, and on unit 1 during the refueling outage.PMI-7030, revision 23,"Corrective Action Program", was effective May 19, 1997, and personnel training is ongoing.NRC Violation 2b"10 CFR 50 Appendix B, Criteria XVI, Corrective Actions, requires in part, that"Measures shall be established to assure that In the case of significant conditions adverse to quality, the (corrective) measures shall assure that the cause of the condition is determined and corrective action taken to preclude repetition." Contrary to the above, On March 12, 1997, the inspectors identified that the corrective actions following a repeat gasket failure on l-IRV-311, identified on January 31, 1996, were inadequate to preclude repetition of spiral wound gasket material entering the reactor coolant system, a significant condition'adverse to quality.Specifically, the licensee performed an evaluation to-determine the ef fect of spiral wound gasket material in the residual heat removal system;however, no action was taken to remove this material which resulted in the.re-introduction of spiral wound gasket material in the reactor coolant system on March 12, 1997." This is a Severity Level IV violation (Supplement I)."

Attachment 1 to AEPsNRC:1260C Page 9 onse to Vh.olation 2b Admission or Denial of the Alle ed Violation\Indiana Michigan Power Company admits to the violation as cited in the NRC notice of violation.

2.Reason for Violation This violation is the result of an inaccurate root cause determination for the initial failure of the gasket, which occurred in August 1995.The root cause determination was not accurateMecause

=information--necessary"to make an accurate determination was not available at the time of the initial investigation., A design.change previously installed to improve residual heat removal (RHR).flow control replaced the.original butterfly valves with a V-notched ball valve, model V100-Sin-300lb, manufactured by Fisher Controls.When this design change w l was engineered, it was not known that excessive turbulen ou d develop at the valve's downstream flange when the valve was throttled to an intermediate position.This turbulence can result in hydraulic forces capable of damaging the metallic winding of the spiral wound gasket used to seal-this.bolted connection.

Subsequent f ailures of the gasket.provided information not available't the time of the initial investigation.

This information led us to the conclusion that the valve and flange gasket are incompatible, and the incompatible design resulted in the gasket failures.fl On August 11, 1995, the unit 1 RHR heat exchanger (Hx)bypas ow control valve, 1-IRV-311, downstream flange gasket ass failed with RHR in service during normal cooldown at the end of cycle 14.When 1-IRV-311 was disassembled for repair, it was discovered that the inside diameter of its gasket was smaller than the inside diameter of the corresponding slip-an flange.This.resulted in approximately 0.155 inches of the gasket's metallic spiral windings being exposed to the flow stream, and resulted in gasket failure.The root cause of the initial failure was therefore determined to be an incorrectly sized gasket.~Neither of the other two RHR Hx outlet flow control valves, 1-IRV-310 and 1-IRV-320, have this type of slip-on bolted.flange connection or evidenced a flange leak.Therefore, they were not.inspected at this time.1-IRV-311 was returned to service with new spiral wound gaskets of the correct size.The emergency core cooling system (ECCS)and RHR were flushed of debris, and unit 1 began operation for fuel cycle 15.Shortly after the completion of the unit 1 1995 refueling outage, with the ECCS and RHR.in standby readiness, leakage from the downstream joint of 1-IRV-311 again occurred.When the valve.was removed for repair on January 31,.1996, its downstream flange gasket was found to have experienced damage similar to the previous failure, with a portion of the spiral windings missing.The root, cause of this failure was determined to be incompatibility of the spiral wound gasket with=the V-ball, type:of.control valve.A non>>metallic fibrous gasket was installed in place of the spiral wound

Attachment 1 to AEP:NRC:126QC Page 10 gasket.Once again, 1-IRV-310 and 1-IRV-320 were not opened because they were not exhibiting any evidence of leakage, nor were they suspected of susceptibility to this type of failure as their throttling characteristics.

differ from 1-XRV-311.

As a precautionary measure.in March of 1996, 2-XRV-311, the unit 2 RHR Hx bypass flow control valve, was'opened for had n inspection prior to the unit 2 refueling outage.This al ot evidenced leakage at the downstream joint;however, its spiral wound gasket was found to be damaged upon valve disassembly.

This provided.thefirst evidence that the flange gasket could become damaged without manif esting-.external leakage;--.A-fibrous gasket was installed in place of the spiral, wound gasket.During the refueling outage, the spiral wound gaskets were-removed" from,2-IRV-310 and 2-IRV-320 and replaced with fibrous gaskets.The spiral wound gaskets removed from 2-XRV-310 and 2-IRV-320 were intact, reinforcing, the conclusion that the 1-IRV-310 and 1-IRV-320 were not at risk for this type of failure.During the recent unit 1 refueling outage, a visual inspection of the reactor's lower core plate revealed more spiral wound gasket debris than would have been expected from the failure of 1-IRV-311 discovered in January of 1996.Up to this point, all failures of-the spiral wound gasket were'believed to be isolated to the RHR Hx bypass flow control valve used in the normal cooldown circuit.Although 1-1'RV-310 and 1-IRU-320 had no evidence of leakage, they became suspect as another potential source of debris.When each valve was disassembled for an internal inspection, their downstream spiral wound gaskets were found partially unwound."'.On March 3, 1997, during the unit 1 RCS/ECCS as found pressure isolation valve (PIV)leak test, it was determined that two PIV check valves had failed their leak test du e presence of gasket fragments.

This debris was subsequently removed and an as-left leak test for all PIVS was performed in April 1997 to demonstrate the class I pressure boundary was intact prior to the beginnin of cycle 16.ing o Corrective Action Taken and Results Achieved 4.The spiral wound gaskets were removed from all RHR flow control valves in both units.Corresponding bolted connections are now sealed with fibrous gaskets which are not susceptible to this form of erosion induced by localized turbulent flow.The RHR piping network branches.and ECCS branches in both units 1 and 2 have been flushed to remove foreign material debris, including gasket fragments.

Corrective Actions To Avoid Purther Uiolations ,Xt was confirmed that no other incompatible gasket design of this nature was installed in a system relied upon to achieve safe shutdown or mitigate the consequences of an accident.

Attachment 1 to AEP:NRC:1260C Page 11 5.'Date When Full Com liance Will Be Achieved Full'ompliance was achieved on March 21, 1997, when the last spiral wound gaskets were replaced for 1-IRV-310 and 1-IRV-320.NRC Violation 3"10 CFR Part 50.72, paragraph (b)(2)(i), requires that any event, found while the reactor is shut down, that, had it been found while he reactor was in operation, would hav'e resulted in the nuclear power plant, including its principal safety barriers beings an analyzed-condition that signi.fi.cantly compromises plant safety, be reported to the NRC within four hours of occurrence.

Contrary to the above, the licensee failed to make a timely report in accordance with 10 CFR 50.72(b)(2)(i)when on March 21, 1997, inspection of flood-up tubes in Unit 1 identified cracks in nine tubes and the equipment associated with these flood-up tubes was declared inoperable.

This is a Severity Level IV violation (Supplement I)." Res onse to NRC Violation 3 Admission or Denial of the Violation Indiana Michigan Power Company admits to the violation as cited in the NRC notice of violation.

2.Reasons for the Violation The.primary reason for the violation was the low emphasis placed on resolution of an indeterminate reportability condition.

Environmental qualification (EQ)issues are complex.The personnel who made the initial reportability decision when the degraded condition was identified on unit 1 were unfamiliar with EQ issues as they relate to system and component operability.

It was decided to submit the condition for further reportability evaluation via the process embedded in our corrective action program.The resulting timetable did not appropriately reflect NRC expectations for promptly evaluating and reporting degraded conditions.

The parallel work to inspect, evaluate, and repair tubes in the operating unit 2, took priority over further evaluation of the unit 1 conditions.

This prioritization of resources was appropriate based on the safety significance of the condition in the operating unit versus the shutdown unit;however, it extended an already unacceptable delay in the reporting of the unit 1 condition.

A contributor to the length of the delay in reporting was the completion of the evaluation to confirm all inoperable equipment.'his provided for determination of the complete safety significance prior to making a final reportability determination.

Of the original nine cracked tubes, only seven resulted in declaring equipment inoperable.

Twenty-three devices were serviced by the conduit in the seven floodup tubes, and of these, only thirteen devices were.confirmed to be inoperable.

~~oi Attachment 1 to AEP:NRC:1260C Page 13 determination that the change does not involve an unreviewed safet question.we sa e y Contrary to the above, on March 6, 1997~, the licensee identified that a plexiglass cover was installed below the return air duct to the unit 2 control room without a proper 50'9 safety evaluation.

o e his plexiglass cover had the potential of affect'n th'CREVS).p rability of the unit 2 control room emergency ventilation t sys em This isa Severity Level IV violation (Supplement I)." I Res onse..to-NRC Violation--4'.

Admission or Denial of the'Alle ed Violation i yiqpsr Indiana Michigan Power Company admits to the violation as cited in the NRC notice of violation.

2.Reason for the Violation The cause of this violation's inadequate procedural guidance.Specifically, the procedure regarding the adminis trat'ion'f"Temporary Modif ications",.1'2 PMP 5040.MOD.001,, revision..5, defined a temporary modification (TM)as follows: 3.Any configuration change that exists on plant systems, components, or structures, (hereafter referred to as equipment) which does not conform to approved plant drawings, approved vendor drawings, or other design documents (i.e., ECPs, EDSs, PDSs)and is being used to maintain operation of the plant.A modification on any equipment being returned to service, though not.being used in support of plant operations, where the modification has the potential to adversely affect plant equipment or personriel safety, shall be considered a temporary modification.

At the time of the event, installation of the drip catch basins on the panels near the control room emergency ventilation system (CREVS)intake ducts was not considered a TM per the procedure because it, was not to be installed on an operating system, and the, basins were not required to maintain operation of the plant.Corrective Actions Taken and Results Achieved The drip catch basins were removed from both control rooms on March 6, 1997, eliminating potential impact on the CREVS.*Testing of the~CREVS was conducted in unit 1 on March 13, 1997, to determine system performance with the drip*,catch basin installed below the return air intake grille.The pan was placed in a configuration which mimicked the intermittent position of the unit 2 intake pan during operation of the system for blackout testing.The tests performed verified compliance with T/S 4.7.5.1 and habitability dose calculations.

J Attachment 1 to AEP:NRC:1260C Page 14 The impact on the unit 1 system was used to analyze the status of the unit 2 system, based on data obtained during the last surveillance test for unit 2.The result fell well within the acceptable range required for operability.

Based on the test findings and capability of the unit 2 pressurization system, the un'it 2 control room ventilation system remained operable:with the catch basin partially'bstructing the flow.4~5.Corrective Actions to Avoid Further Violations The TM procedure,'12-PMP 5040.NOD.001, will be revised'to stress-that-any-installation;-regardless of whether installed on can operating system or,not, should be considered a TM if there is reasonable expectation that the potential exists to" adversely impact~the operation of an adjacent system.The pxocedure revision will be completed by June 30, 1997.I As an interim measure until the procedure change can be made, management will communicate this event and their expectations regarding the implementation of the TM process to those-employees that may,be involved in making the decision to invoke the TM process.This will be done by June.10, 1997.Date When Full Com liance Will Be.Achieved Full compliance was achieved on March 6, 1997, whenthe basins were removed.

ATTACHMENT 2 TO AEP:NRC:1260C RESPONSE TO NOTICE OF DEVIATION Attachment 2 to AEP:NRC:1260C Page 1 Notice of Deviation h"During an.NRC inspection conducted February 16 through March 29, 1997, a deviation of your actions committed to in the updated Final the~~G Safety Analysis Report (UFSAR)was identified.

~In accordanc'th eneral Statement of Policy and Procedures for NRC Enforcement Actions, NUREG-1600,'he deviation is listed below.UFSAR Section 7.4.1-stated, in part,"The power range channels are capable of recording overpower excursions up to 200 percent of full power."'ontrary-'-to-the--above,-on-February 25;1997, the NRC inspectors identified three of four recorder pens":inoperable for the power range channels that were capable of recording overpower excursions up to 200 percent of full power.Xn addition, licensee personnel stated that since June of 1991 the-pen's failure rate was such that the percent unavailability average was 14.9 percent.The pens failure rate was such that they were not capable of recording, overpower--

.=excursions." Res onse to NRC Notice af Deviation Reasons for the Deviation The deviation states that the resident inspector identifi d t hat the power range channels capable of recording excursions e up to 200 percent of fu11 power, as described in the UFSAR, were found with three'f the four channels incapable of performing this function.An historical review identified that this particular recording capability has been challenged in the.past including significant periods of recorder unavailability.

The cause for the excessive failures is the relative fragility of the servo-amplifier electronics and overall age.The"fragility" of'he electronics is exacerbated by the original time response specification and by the need for speciali2:ed analog components (state of the art in the late 1960s)to perform this function.The original design philosophy was to capture the span of the Westinghouse Nuclear Instrumentation Power Range channels, 0-200 percent power.In order to capture this range of power, a very fast recorder was believed to be required.The time response requirements have led to a design that has been difficult and expensive to maintain.Very few replacement parts are available from the vendor and these recorders will not be able to be maintained in the near future.The inoperability periods are influenced by the fact these recorders are not qui'ckly corrected when identified as requiring service.Long repair-by dates-are stipulated by the work control process based on the recorders'egulatory significance and the lack of operational usefulness on a daily basis.No surveillance data is required by operators~on these recorders and the normal power level is recorded on different instruments in the control room.This led to the.lack of attention'o these recorders by control room operations personnel.

a~~Attachment 2 to AEP:NRC:1260C Page 22.Corrective Actions Taken and Results Achieved 3.Coxrective action was taken concerning the three failures noted in this deviation.

The unit 2 recorder 2-SG-14 was calibrated and the failed pen returned to service on March 13, 1997.Unit 1 was in a refueling outage and the concerns were addressed in section 3 of this response.Corrective Actions to Avoid Further Deviations The corrective actions to avoid further deviations include improving the control board monitoring to identify substandard=-equipment,--increase importance of all control room instrumentation/recorders in the work control process, and update the specific recorders mentioned in this deviation to allow ease in their maintenance.

4, These actions were accomplished by the following changes: The operations department standard OPP-1,"Control Room Control Board Monitox'ing During Non-emergency Operation Conditions", was revised to stress the importance of control room panel awareness during every day operation.

This issue was discussed at the following shift manager's meeting and communicated to the operator crews.The work control standard that placed time requirements on the repair of critical control room recorders'as revised to include all control room recorders.

Control room recorders requiring maintenance shall be prioritized to be woxked within f ive to f ourteen days as determined by the operations department as per the 1997 AEPNGG site operating and maintenance plan.The Tracor Westronics recorders were removed f rom unit 1 and their points placed on an existing Yokogawa recorder in the control room.Similar changes are planned for the unit 2 control room instrumentation.

These recorders will allow easier maintenance and thus reduce the unavailability.

Date When Corrective Action Will be Co leted The unit 1 corrective actions wexe completed prior to the restart aftex the refueling outage.Unit 2 corrective actions will be completed during the next refueling outage scheduled for.the fall of 1997.