Text

Insp Rept 50-440/96-18 on 961221-970203.Violations Noted. Major Areas Inspected:Licensee Operations,Engineering,Maint, & Plant Support
	ML20137K485
Person / Time
Site:	Perry
Issue date:	03/19/1997
From:	; NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
To:	;
Shared Package
ML20137K428	List: IR 05000440/1996018; ML20137K424; ML20137K441;
References
	50-440-96-18, NUDOCS 9704070039
	Download: ML20137K485 (31)
	v • d • e

_.

_ _

.. _._.

.

..

.

..

- _.

.

U. S. NUCLEAR REGULATORY COMMISSION i

.

REGION lli

.

Docket No:

50-440 License No:

NPF-58 -

. Report No:

50-440/96018 Licensee:

Centerior Service Company Facility:

Perry Nuclear Power Plant Location:

P. O. Box 97, A200 Perry, OH 44081 i

Dates:

December 21,1996 - February 3,1997 Inspectors:

D. Kosloff, Senior Resident inspector R. Twigg, Resident inspector M. Holmberg, Reactor Inspector E. Schweibinz, Project Engineer J. Hopkins, Project Manager Approved by:

J. M. Jacobson, Chief Reactor Projects Branch 4

..

9704070039 970319 PDR ADOCK 05000440 G

pm (

w v

~.a

-. - __

_ _ _

. _.. _ __._._ _

__ _

_._

-

_. _ _. _. _ _.. _

_. _. -.. _ _,

. _. _. _ _.

.

EXECUTIVE SUMMARY

.

Perry Nuclear Power Plant, Unit 1 NRC inspection Report 50-440/96018

This inspection included aspects of licensee operations, engineering, maintenance, and plant support. The report covers a 7-week period of resident inspection.

Operations Scram January 7,1997, operator error switching inverter (Section 01.2).

Excessive cooldown after scram, STA error in monitoring cooldown contributed,

violation for failure to follow procedure and inadequate procedure (Sectiori 04.1).

While engineering was performing an operability determination related to an EDG

frequency issue there was poor communications between operations and engineering (Section E1).

Operators observed steam during venting of the residual heat removal system and

made a conservative operability determination. Engineering support of operations was appropriate (Section 02.1).

Review of an earlier transient involving a reactor recirculation flow control valve

(FCV) revealed that ineffective corrective actions for a previous event and lack of

- sensitivity to the need to resolve known equipment problems created an unnecessary challenge to the operators. The ineffective corrective actions were examples of an apparent violation. Even though challenged by the ineffective

,

corrective actions, the operators had opportunities to prevent or reduce the severity

)

of the associated plant transient. However, insensitivity to FCV reactivity additions, insufficient questioning attitudes, and poor coordination of plant evolutions by the operators allowed the transient to occur (Section 08.1).

Prompt attention to a steamline drain leak avoided a plant scram (Section 02.2).

Maintenance Testing activities were conducted appropriately. Although technicians and i

operators approprittely decided to request that a test be changed to correct a o

i problem, the test should have been written more clearly before it was approved for

,

use (Section M1).

]

Generally maintenance was timely and effective. Repair of leaks on a seal pump i

,

suction line and a main steam drain line were examples of such performance.

However, some simple tasks were not completed even though a special

.

maintenance team existed to quickly complete such tasks. Maintenance had i

-

l e

w--,--

-vw

-

,vr rv

--.

wr,,-

r-t w,,-.

-

w ---

w-

-,-w-

--

. _.

_ _

_ _...

. __ _ _ __. __ _____.. __ _

.

I

.

improved material condition in the past, however, during this inspection period no additional overall progress was observed

-

(Section M2).

!

Enaineerina l

Engineering provided appropriate support to operations to evaluate a residual heat

removal steam venting issue (Section 02.1)

!

)

The inspectors identified that although an engineering operability evaluation was

appropriate it was not presented to the shift supervisor on time. An NCV was identified for failure to follow a procedure requirement for requesting an extension

.

of time for the operability determination. This was an example of poor communicaNons between operations and engineering (Section E1).

Engineering p6 'ormed an adequate operability determination following the

January 7,199i. loss of feedwater and excessive cooldown event (Section E1.2).

The inspectors identified a poor quality gage installation that had no direct potential

,

safety consequences. However, the acceptance of the installation indicated that some individuals still did not have an adequate questioning attitude (Section E2.1).

Severalinconsistencies were noted between the USAR and plant practices,

procedures, and parameters. The licensee included the inconsistencies in its corrective action program (Section E2.25.

Plant Suonort Licensee rem 9 dial actions for a significant loss of offsite communications capabi!ity

were prompt and minimized the possibility of occurrence of another such loss (Section P2).

l

,

$

,

-

a

-

.----

-

-.

...

.

-..

.

. _ -...

_

. - - -

.

-- -

-

.

I

'

-

!

Report Details

!

~

Summary of Plant Status j

,

The plant operated at full power until January 7,1997, when the plant scrammed. The reactor was taken critical on-January 9. On January 21 the main turbine was taken off -

line to allow the repair of a steam leak from a main steam line drain. After the leak was repaired, the generator was synchronized with the grid on January 22. For the remainder of the inspection period the plant operated at full power except for short power reductions for testing and control rod realignments.

s I. Operations

Conduct of Operations 01.1 General Comments (717071 Using inspection Procedure 71707, the inspectors conducted frequent reviews of ongoing plant operations. While in general, the conduct of operations continued to be safety-conscious, a scram was caused by a personnel error and was followed by a cooldown that was exacerbated by another personnel error. Those personnel errors were of concern.

01.2 Reactor Scram Caused by Ooerator Error a.

insnaction Scone (71707,92901)

b.

Observations and Findinos At 5i34 a.rn. (EST) on January 7,1997, the plant automatically scrammed from full power. The resident inspectors responded to the site to monitor transient recovery.

The low reactor water level scram resulted from a loss of feedwater control caused by a non-licensed operator error during BOP inverter switching operations. The operator had used the correct operating instruction. However he had incorrectly used a step that should not have been performed. The step removed power from the feedwater control system and various control room instruments. The operators promptly identified the cause of the scram and restored power to the affected circuits. All feedwater was lost and level was recovered by HPCS and RCIC after it decreased to about 83 inches above top of active fuel. All safety equipment functioned as expected during the transient. RWCU isolated and both reactor recirculation pumps tripped so there was only natural circulation mixing in the reactor. Cold water collected in the bottom of the reactor pressure vessel. In the first hour of the transient, temperature dropped to about 230 degrees F from about 540 degrees F in the lower vessel head as measured by a temperature indicator mounted on the outside of the vessel. Centrol rod drive (CRD) water was also injected at about 200 GPM for about 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months before the scram was reset, contributing to the cooldown and stratification of hot and warm water in the vessel.

4 I

-

. _.

___

._ _. __ __

_. _ _.

_.

_ _ __ _ _ _ _._ _ __ _

_

._

The cooldown continued at a lower rate until vessel bottom head temperature l.

stabilized at about 110 degrees F. The stratification was more pronounced then expected and the operators were not aware of the problem until a non-licensed

'

operator informed the STA that the recirculation loop temperature was dropping.

RWCU flow was restored at about 11:30 a.m., which slowly warmed the lower

regions of the reactor. The licensee took the plant to Mode 2 at 3:57 a.m. on i

January 9 in preparation for plant restart. The main generator was synchronized to the grid on January 10,1997.

c.

Conclusions

)

I

The scram could have been avoided by a more thorough pre-job briefing of the non-licensed operator, by more detailed communications during the switching, or by closer supervisico ouring the switching. Operations management identified those weaknesses anc dueloped corrective actions to improve performance. Operators j

appropriately responded to the scram and loss of instrumentation in the control

.

room. However, due to an inadequate procedure an excessive cooldown did occur as described in Sections 04.1 and E1.2.

.

Operational Statqs of Facilities and Equipment

\\

'

.

02.1 RHR Feedwater Isolation Valve Leakaos

,

s.

Innoaction Scoon (37551, 61726. 71707. 92903)

Plant operators unexpectedly encountered steam while conducting Surveillarice Instruction (SVI) E12-T1182, " Residual Heat Removal (RHR) B Low Pressure

Coolant injection (LPCI) Valve Lineup Verification and System Venting." The

,

inspectors observed operations and engineering efforts to identify the source of the steam.

b.

Observations and Findinas

The 'B' RHR LPCI heador was connected to the reactor feedwater (FW) line through

check valve E12-F050B and normally-closed motor operated isolation valve E12-F0538. The 50B valve had no measurable leakage when tested during the last i

refueling outage (RF05) in February 1996. The 538 valve leakage was 3 gpm at that time. The allowable leakage for each valve was 5 gpm, provided both valves

-

in series leaked less than.85 gpm. Testing measured leakage from the hot FW side of the valves to the cooler RHR side of the valves. The leakage limits prevented

.

steam voiding in the RHR LPCI header and was based on engineering calculations that indicated steam voiding could occur if the leak rate was greater than.85 gpm.

j

'

With the plant operating, leakage through valves 50B and 538 was indicated in the

~

control room by increasing RHR heat exchanger pressure. The pressure was logged

,

i three times a day, with an administrative limit of 300 pounds per square inch, gage

(psig). There was also an annunciator alarm at 480 psig. When the administrative

.

limit was approached, the control room operators vented the heat exchanger to the

suppression pool.

_

.

..e..

..r.

..m

-. -.,

m-,,,

-

__.

.

..

_ -.,,

.-...

-

..-..

-. -

- - --- -..- -.-

..--. -..

. -. - _

-. - =.

>

,

_

!

,

i

)

-

i After startup from RF05, the 'B' RHR heat exchanger was vented sporadically. In

.

'.

January 1997, the operators began venting the RHR heat exchanger dai!y as j

j pressure repeatedly reached about 280 psig. This indicated that leakege had

,

increased through the 508 valve. Monthly performance of SVI E12-T1182 indicated no concerns with voiding until after the scram on January 7,1997.

'

When the operators opened the high point vent, steam escaped. The operators

,

j promptly declared LPCI 'B' inoperable and engineering began a high priority '

i (category 2) corrective action investigation. Engineering concluded that additional i

leakage had developed on the P.HR side of the 538 valve, allowing cool RHR water l

to leak out and be displaced by hot FW, increasing water temperatures in the RHR piping. The higher temperatures eventually caused flashing when the high point vent was opened. A minimum pressure of 30 psig was normally maintained by the keeptill pump, but pressure decreased during venting. Voiding in the header pipe

.

was ruled out based on observed changes in header temperatures and pressures indicating that leakage did not approach 0.85 gpm. Therefore, LPCI 'B' had never l

been inoperable, and the operators declared it operable.

,

At the end of the inspection period, the operators were monitoring header pipe temperatures with temporary external thermocouples and conducting a new fill and vent procedure about three times a day. The frequency of the fill and vents was

.

based on observed temperature in the header pipe. The frequent fill and vents kept the RHR water below 200 degrees F, preventing flashing in the header. However, the fill and vents were an additional burden on the operators and increased

'

collective radiation dose for the operators. Engineering was attempting to identify all RHR leakage paths so they could be evaluated for isolation. Engineering was

'

also working on qualifying an additional fill and vent method which would use existing equipment to provide cool water at a higher pressure. This would allow a

larger volume of cool water to displace the warmer water in the RHR piping during each fill and vent and thereby reduce the frequency of the fill and vents.

<

'

c.

Conclusigna

Operations made a conservative operability determinatiun and requested appropriate support from engineering. Engineering support of operations was appropriate.

02.2 Generator Taken Offline Bv Balance of Plant Steam Leak

a.

Insoection Scone (71707. 92901)

Plant operators identified, by remote camera, a small steam leak in the turbine building on January 18,1997. On January 21, the licensee took the main

,

.

generator offline to repair the leak. The inspectors observed operator actions.

!

b.

Observations and Findinas

The steam leak was located on a Main Steam Line (MSL) "D" drain line downstream of the turbine stop valves at the same weld }oint as a leak located on the drain on

.

h d,

" " " '

' ' "

S'

N

+ -

,.

3_,.

_.,,

.- - _-

_ -. - - - - -

.

. -..-

.. -. -

. -

-

. - _ _. - -. -

_

-

.

f*

'

.

.'

L MSL "C" about 8 years ago. Monitoring revealed that the leak size was growing

'

1.

and plans were made to take the generator off line, isolate the steam. supply to MSL I

"D," and repair the leak. Information from the repair of MSL "C" indicated that

'

condenser vacuum was sufficient to handle the 2 inch drain hole during the mpair.

.

The planned start time for the downpower was 10:00 p.m. on January 21.

,

However, plant personnel monitoring the leak reported the leak growing and l

,

L management made the decision to commence the downpower at 3:30 p.m. Power j

!

reduction was well coordinated by the operators with no significant problems until 6:50 p.m. when the drain line separated from MSL "D." The separation was

-

immediately reported to the operators by those monitoring the leak and, with reactor power at 27 percent, the operators tripped the turbine causing five of the steam bypass valves to open. The steam supply to MSL "D" was isolated, stopping

the steam leak. Power reduction continued to 20 percent at 7:53 p.m.

j

<

.

All similar drains were inspected with no problems identified. Repairs were

.

completed, and the turbine was synchronized with the grid on January 22.

i-c.

Conclusions

'9

.

Had the licensee waited until the original downpower start time, a plant scram i

)

would have been necessary. The decision to expedite the downpower was appropriate. However, the downpower could have started even earlier and may 4,

i have avoided the drain separation altogether.

!

02.3 Cold Weather Preoarations a.

insoection Scone (71714)

!

The inspectors conducted walk downs of building spaces during cold weather j

conditions and reviewed the licensee's winter work list.

b.

Observations and Findinas

,

(

The inspectors observed that building spaces were heated, heat traces were

-

functioning, and where appropriate, measures had been taken to minimize cold

+

weather impact on equipment. Work management had begun maintaining a list of winter work items during the summer. Initia'ily the list included all identified work items. As work progressed only items needing additional attention were kept on the list and those items were discussed regularly at rnanagement meetings. One

.

item remaining on the list was a requirement for a space heater for a fire protection

'

water valve in the abandoned Unit 2 offgas building. The fire protection water

system is common to both units. The inspectors verified that a radiant heater was

warming the valve and associated piping. Other items on the list were less l

significant. Emergent items were promptly added to the work list. Work management had also established a similar summer work list and began work in the winter to complete those items. No plant equipment had been damaged by cold i

weather.

_

. -

.

-.

_

_.. _., _ _ _..

.

,_, _.

.,. _ _

.-,._m.,

.__

_. __

__ _ _., _ __ _ _

__ _

_ -_.

_ _ _

_

. _.

. _ _

__

.

..

c.

Conclusions

.

Completion of winter work items was considerably more timely than in previous years. The winter and summer work lists improved awareness of required work

'

and identified responsible organizations. Buildingt and equipment were prepared to sustain extreme cold weather.

02.4 Increase in Reactor Coolant System Unidentified Leakaos

]

'

a.

Insoection Scone (71707)

l On January 8,1997, with the plant in hot shutdown, the inspectors observed that

indicated reactor coolant system unidentified leakage had increased slightly from zero. The inspectors reviewed this condition and the operators' response to it.

'

b.

Observations and Findinas l

After morning shift turnover, the inspectors observed that the drywell floor drain sump flow rate recorder in the control room showed an increase to about 0.2 gpm that had started at about 5:30 a.m. The inspectors also observed that the uppor

drywell cooler drain flow monitor indication was consistent with the leakage indicated on the recorder. These indicators were used to monitor unidentified

-

leakage from the reactor coolant system. For the past 8 months the indicators had

,

i normally indicated zero leakage. The inspectors notified the operators, who

evaluated the leakage indication and noted that it began decreasing at about I

10:00 a.m. and returned to O gpm at about 11:45 a.m. The allowed leakage rate

was 5.0 gpm and the operators recorded the leakage rate three times a day, more i

frequently than required by Technical Specification (TS) Surveillance Requirement , 3.4.5.1 (every 12 hours1.388889e-4 days 0.00333 hours 1.984127e-5 weeks 4.566e-6 months ). The leak rate indications were also observed during shift i

turnover walk downs twice a day. The operators identified some valve operations j

'

that may have contributed to the leak rate change and documented the conditions in PlF 97-033.

'

c.

Conclusions j

The operators' actions were prompt and appropriate after the inspectors informed them of the indicated leak rate. However, although the observed leak rate was well within TS limits, the operators should have identified this issue prior to the t

inspectors' observation.

'

f j

i

)

l i

,...

,.

,

.-

-

- -. - - - -.-,

.. -

-..s-~

-.

,. - -. -. - -

i

.

_ _ _ _. -

_...._-.

-

.-.

_ _ -

. _. - -

. - ~

.,. _.

.

-.

..

.

-

m

..

' 04 Operator Knowledge and Performance 04.1 Technical Soacification Cooldown Rate Exceeded After Reactor Scram a.

Insoection Scone (71707, g2901. 93702)

b.

Observations and Findinos:

'

Surveillance instruction SVI B21-T1176, RCS Heatup and Cooldown Surveillance, allowed the STA to use only one recirculation loop thermocouple to track cooldown.

.

The SVI also allowed use of an ERIS computer point for the thermocouple reading instead of the strip chart, which had thermocouple readings for both loops and a real time visual history of recent conditions. It was also adjacent to another strip

chart recorder which had readings for upper head, RWCU bottom head drain, and

.

bottom head temperatures (temperature element attached by magnet to outside of

'

bottom head). The STA stated that he did not notice when the bottom head and l

bottom head drain temperature started to drop (shortly after the scram). It appears he was focused on the loop recorder and verification that it matched the ERIS data.

This recently licensed (instant SRO) STA was relatively inexperienced as an STA, he had previously been a system engineering supervisor.

The STA stated that he displayed both loop temperatures and the bottom head i

temperature from the RWCU line on the ERIS computer screen. The computer displayed only one current value for each point with no graphic representation. The chart recorders cannot be seen from the STA office, the computer display location, or the horseshoe area.

l The ERIS point read about 536 deg F (except for a relatively small drop and recovery at the time of the scram) until about 7:50 a.m. (scram was at

'

5:34 a.m.).

The RWCU drain line and bottom head chart recorder temperature indications began to drop shortly after the scram.

The STA tocorded his first reading at 6:30 a.m. The STA stated that after he

'

initially verified that the ERIS points were consistent with the chart recorder data, he did not use the chart recorder again until a nonlicensed operator asked him to take a look at the temperature drop on the chart recorder (about 7:50 a.m.).

The thermocouple, attached by a magnet,is not referenced by the SVI and the STA did not use it. The bottom head drain thermocouple was ruled out by the STA because RWCU was not in operation and he did have this ERIS data displayed on the computer monitor that he was using. This was consistent with the system

-

operating instruction for RWCU.

_

.

..

__

-

.

_ _ _. _

_ _.

_

. - - _. -.. -. - -.. - -

-. _ -

.

...

. _. - - -...

.. -

-

'

.

..

.

Once the ERIS point (loop temperature) changed at 7:55 a.m., the crew began i.

Isolating heat loads and restoration of RWCU flow. Had the crew known that the bottom head was cooling rapidly immediately after the scram,it is reasonable to conclude they would have reset the scram well before 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months (7:40 a.m.) after the scram. This would have reduced CRD flow, which was the primary source of the cooler water to the bottom head.

'

On January 9, the inspectors talked with two other STAS. An experienced STA j

stated that he would not want to use the ERIS display for loop temperature indication because he liked to see the graphical history on the chart recorders. It

,

'

appeared that he did not know that the STA had used the ERIS computer display on January 7. Another relatively inexperienced STA was observed taking heatup data

cm b chart recorders. He explained that he preferred to move the chart and

,

record the data directly from the mark that was printed as he watched and that he preferred to look at the data on both recorders. In the past the inspectors had observed the STAS taking the heatup and cooldown data from the chart recorders.

i c.

Conclusions

[

10 CFR 50 Appendix B, Criterion V, instructions, Procedures, and Drawings, states

" Activities affecting quality shall be prescribed by documented instructions,

procedures, or drawings, of a type appropriate to the circumstances and shall be j

accomplished in accordance with the instructions, procedures, or drawings.

instructions, procedures, or drawings shallinclude appropriate quantitative

,

acceptanca criteria for determining that important activities have been satisfactorily

'

accomplished."

i Surveillance Instruction (SVI) B21-T1176, RCS Heatup and Cooldown Surveillance, l

Revision 4, September 13,1996, required that reactor coolant system (RCS)

temperatutes be verified every 30 minutes to be within TS limits during cooldown

.

i (inspector note - the SVI did 091 indicate when the first reading should be taken).

l The SVI also lists the TS limit for cooldown to be 100 degrees F per hour. The l

reactor scram and the start. of the cooldown occurred at 5:34 a.m.

Contrary to the above the first recorded temperature reading was at

'

6:30 a.m. Also, the procedure allowed readings to be taken from temperature

,

!

sensors that did not reflect the actual cooldown rate. This contributed to an

{

inadequate awareness of the cooldown rate which exceeded the TS limits. This is a violation of 10CFR50 Appendix B, Criterion V (50-440/96018-01 (DRP)).

Miscellaneous Operations issues 08.1 Unexoected increase in Reactor Reactivity a.

Insoection Scone (71707,92720,92901)

The inspectors continued their evaluation of an unexpected increase in reactor reactivity which was discussed in Inspection Report 96-17 (URI 50-440/96017-

.

_ _. _ _ _ _ _ _ _ _ - - - _ _ _. _ -,, _ _ _ _ _. -..

..,,

,

, _. -. -. -,

..m

_

_.

,y.__

_ _

m

-

.-.

--

..

.

..-

-

.. --

.-

. - _ - - _. - -...-.._.

.

..

1

..

'

01(DRP). When the operators had attempted to place the 'A' Reactor Recirculation

'.

Flow Control Valve (FCV) back into service it opened unexpectedly. Reactor power increased as a result of increased recirculation flow.

<

i

b.

Observations and Findinas:

i 1.0 Description of the Event On November 9,1996, during recovery from a local power range monitor failure,

'

that resulted in the hydraulic power unit (HPU) being shutdown, the operators were attempting to retum to service the HPU for FCV 'A.' One of the two HPU subloops, 'A2,' had been designated as " limited use only" due to high vibration.

Therefore the operators chose to prepare the other subloop, 'A1,' for startup. The instruction for preparing an HPU for startup required the operators to confirm that output puwer was available from the HPU programmable logic controller. A technician reported that a fuse was blown, thus indicating that the 'A1'subloop

,

operate / isolate solenoid valve had no power. The shift supervisor (SS) consulted with one of the responsible cystem engineers (RSEs) via telephone (the RSEs were not on site) and decided to restart the HPU on subloop 'A1' with the blown fuse in

,

place. The SS had remembered previously running the HPU with a blown fuse.

Restart of the 'A1'subloop required an operator at one of the control room back

]

panels to start the hydraulic pump. As the pump raised system pressure, the FCV

'.

began opening from its position of 49 percent open. Approximately 12 seconds

,

after the pump was started, the shift supervisor recognized the unintended motion

and stopped the FCV at 61 percent open.

'

Reactor power reached 100.2 percent (heat balance). The operators closed FCV 'B'

to reduce power to 98 percent, creating an 8 percent flow imbalance between the recirculation loops. Technical Specification (TS) 3.4.1 was entered due to a greater

than 5 percent flow mismatch between loops. This required a shutdown of one of

'

the recirculation loops (single loop operation) if the flow mismatch could not be reduced to less than 5 percent within 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months . The acting plant manager (engineering director), operations management, reactor engineers, and RSEs

,

responded to the site. At 1 hour1.157407e-5 days 2.777778e-4 hours 1.653439e-6 weeks 3.805e-7 months and 51 minutes into the action statement, the

.

operators had driven rods in to reduce reactor power and increased flow in the 'B'

recirculation loop, exiting TS 3.4.1.

2.0 Hydraulic Power Unit (HPU) Descriotion Each of the two reactor recirculation FCVs (A and B) had one HPU with two subloops (1 and 2). One subloop was required to provide the hydraulic motive force to adjust the position of the associated FCV; thereby adjusting reactor core flow and power. Solenoid valves isolated the non-operating subloops and controlled hydraulic pressure to open or close the FCVs. Perry and other plants had experienced numerous blown fuses caused by sticking solenoid valves.

-

l

.

-

.

- - - -. ~. - -..,

.

..._ -.,.,

,._.r.

-. -,.

-

.

-

- -

.

.-

..

.. -.. --

- -. - -. - -. - -

-

..

.

I On startup of an HPU subloop with the operate / isolate valve mispositioned (not in j -.

the isolate position), hydraulic pressure to the FCV was increased by the pump and was directed by the hydraulic solenoid control valve to the FCV hydraulic operating

.

' piston to either open or close the FCV. On November 9 the hydraulic solenoid

'

control valve caused the FCV to open. In 1994, when a similar change in reactivity

.

had occurred during low power, the FCV moved in the closed direction.

,

3.0 Seauence of Events on Saturday. November 9.1996

~

The sequence of events listed below highlights some of the communication problems that occurred. Note that the SS and STA were talking to two different j

RSEs, and neither of the RSEs are onsite.

i l

"

6:58 a.m.

A local power range monitor failed high, causing one of six Average Power Range Monitors to erroneously indicate a reactor power increase. The erroneous indication caused an Automatic Flow

~

Demand Limit " runback" (partial closure) of the FCVs. The operators verified the runback was due to erroneous indication and p' aced the FCVs in " lockup" (shutdown the HPUs), stopping FCV motion with power at 99 percent. The erroneous indication was corrected at 7:20 a.m. and efforts to return the HPUs to service began.

about Shift supervisor (SS) #2 discussed the blown fuse for HPU 9:00 a.m.

'A1' isolate /operata solenoid valve with the responsible system engineer (RSE) #1 by telephone. RSE #1 was not on site. RSE #1 was uncomfortable with the plan to restart HPU, he said he would call back later with a recommendation.

i 9:15 a.m.

RSE #1 called back SS #2 and recommended restart of 'A1'subloop

.

with the 'A2' subloop in " maintenance."

.

^

10:30 am STA and RSE #2 discussed HPU 'A1' restart on telephone. RSE #2 was also not on site. RSE #2 recommended monitoring FCV during HPU start.

.

10:58 am HPU subloop 'A1' was started and FCV 'A' opened from 49 percent to 61 percent in about 12 seconds. SS #2 shutdown the HPU,

locking up the FCV with reactor power at 100.2 percent thermal

(Neutron power did not exceed 105 percent). Reactor power was

'

reduced to 98 percent using FCV 'B.'

,

J 11:14 am TS 3.4.1 was entei:3d due to > 5 percent flow mismatch between the recirculation loop flows.

12:51 pm Reactor power was reduced to 88 percent by inserting rods and FCV 'B' was opened. TS 3.4.1 was exited with reactor power at 94 percent and flow mismatch <5 percent.

,

i

'

.

...

.

- - -

.

-

- _.

.

...

.-

-. _. -. -. - - -

_

......

.

~

.

$

4.0 Root Causes

.;

'

.

The Perry event investigation team concluded that the root cause of this event was that the level of involvement of the shift supervisor in the decision-making process

'

distracted from his oversight responsibilities. The team also concluded that this

'

action did not meet management's expectations for use of available resources and application of training.

l

'

The team also identified two contributing causes:

1.

The operators did not comply with sol B33 because it directs the operator

<

]

to confirm that output power is available to the solenoid valves. With the i

solenoid valve fuse blown, this power was not available. Startup of the HPU was continued with the requirements of this SOI step not met.

[

2.

Long standing degraded HPU sub-loop material conditions existed. Repeated l

operate / isolate solenoid valve failures had caused reactivity events. Past.

corrective actions had been ineffective in removing those challenges to the

operators.

The inspectors concluded that there were at least three root causes: 1) operator

.

training was imffective related to HPU operate / isolate solenoid valve failure,

'

2) corrective action to cycle HPU subloops was ineffective in minimizing the probability of operate / isolate solenoid valve failure, and 3) corrective action to place a step in the sol to confirm power available to the operate / isolate solenoid valve was ineffective in preventing startup of an HPU without the power available.

t The inspectors also concluded that insensitivity to reactivity additions caused by FCV motion was a contributing cause.

5.0 Safety Slanificance This was the second time in about 2 years that HPU equipment problems had led to unexpected reactivity changes. The potential existed to challenge an automatic scram on high reactor power. The licensee also has a recent history of ineffective corrective actions (seven Severity Level IV violations in the last 2 years). This indicates the potential for a broader concern with the overall implementation of

,

corrective actions.

d 6.0 Licensee Corrective Actions The inspectors verified the following short term corrective actions:

The failed operate / isolate solenoid valve was replaced.

The event shift technical advisor (STA) prepared a management preliminary

review report.

T

-

+----

,.

-,,

,

, -. -.

,

...,..

_- -

-.

.-

-

.-.

... - _-..-. -

.-_ _..

.-_

..-..__

.

..

11..

A' PlF was written and given a high corrective action program priority

(category 2) on the day of the event. A human performance enhancement

'

.

system (HPES) evaluation was also begun. The operations manager

.

upgraded the PlF to a category 1 on November 12,1996. This resulted in

.

the formation of a multidisciplinary HPES investigation team with a written charter. On November 14,1996, the operations, mainti. w-<m and training -

subcommittee of the company nuclear review board held a special meeting to review the event, generate questions, and review the team's draft i

charter. The team report was completed on December 6,1996, incorporated into the PlF investigation, and approved by the plant manager

,

on December 18,1996.

<

Six corrective actions to prevent recurrence (CATPR) were developed as a j

result of the PIF and HPES investigations. Two CATPRs (96-3400-01 and -

'

02) had been completed by the end of the inspection period.

t The inspectors verified that additional long term corrective actions had also been

'

developed. The investigation team recommended accomplishment of four additional CATPRs (96-3400-03 through -06) and 46 PIF remedial actions (PIFRA's). With the

!

exception of two PlFRAs, all these corrective actions or proposed alternates were

'

accepted for completion. The two PIFRAs not accepted were rejected because they involved training of personnel in an organization that will be absorbed into two

existing organizations. The licensee assigned all corrective actions due dates of March 31,1997, or earlier.

i The inspectors reviewed 22 of the PlFRAs that had been completed by the end of the inspection period. Most of them involved training on or discussion of the event

'

and its causes.

,

7.0 Annarent Violations

.

Appendix B, Criterion XVI, of 10 CFR 50 requires, in part, that measures shall be established to assure that conditions adverse to quality, such as deficiencies and nonconformances, are promptly identified and corrected. In the case of sigrcificant

conditions adverse to quality, the measures shall assure that the cause of the i

condition is determined and corrective ection taken to preclude repetition.

Three examples of an apparent violation (eel 50-440/96018-02 (DRP)) of 10 CFR Part 50, Appendix B, Criterion XVI, " Corrective Action" were identified:

1) The shift supervisor's decision to return HPU subloop 'A1' to service with a blown fuse was based on a misunderstanding by the operators that the potentially mispositioned operate / isolate solenoid valve would have no impact on the FCV position even though several past events clearly indicated that this conclusion was

>

wrong. The operatin2 shift's communications with two RSEs did not correct this misunderstanding even though, in later discussions, the RSEs stated that they had

'

expected valve movement. Corrective actions from a similar 1994 FCV event

,

included operator training on the event. Corrective actions related to an almost

,

.

-.

.

m

.-w

.

.,. -

,

,,.m.-i=,

y g.

-e y

-

. -

.. _ _

-.. - -

-.

..

- -. -..-- -. -.. -- _- - -.

.

..

f.

,.

,

identical 1994 event at another' plant also included operator training. The operator

'

training of the impact of a blown fuse for an operate / isolate solenoid valve was

-

ineffective.

i

2) Another corrective action for the 1994 FCV event was to cycle subloops for

each of the HPU units as part of non-Technical Specification Operator Rounds (non-

)

,

TS rounds). This corrective action was developed as a result of a review of an

!

-

almost identical 1994 event at another plant. The cycling exercised the solenoid

<

valves weekly to help prevent sticking valves. In October 1996, the inspectors identified an increased noise level from Subloop 'A2.' Vibration testing then led the

RSE to request " limited use only" of Subloop 'A2.' As a result the 'A1'

'

operate / isolate valve was not exercised for about 6 weeks. HPU 'B' also had one

i subloop designated for " limited use only" and was not being cycled. Even though I

the operators noted in the non-TS rounds logs that cycling was not being completed and the RSE was aware cycling was not being completed, no compensatory actions

-

!

were taken to either ensure the valves were cycled or to review with the operators

!

past lessons learned as a result of sticking valves.

l Discussions with operation's management, work control, and systems engineering

'

revealed no comprehensive impact analysis when non-TS rounds were not

)

completed. Subloop 'A2' repair was scheduled as a priority 4 work order (WO),

normal on-line work. After the November 9 event, the WO was given a priority 3,

repair within 3 weeks. With appropriate consideration by engineering, operations, and work scheduling of the potential consequences of not cycling subloops, repair

of Subloop 'A2' may have occurred prior to the valve sticking on November 9. The

corrective action of cycling the isolate / operate valves weekly was ineffective

,

because of inadequate follow through.

'

3) System Operating Instruction (SOI) B33, Reactor Recirculation System, required, i

during startup of a de-energized HPU (step 4.2.1.a), instrumentation and Control

,

technicians to confirm that all solenoid fuses were functional. The sol provided no

,

.

!

additional guidance on actions if a fuse was found blown. Confirmation of fuse function was a corrective action for the 1994 event. The intent was that blown

fuses would be replaced after any necessary troubleshooting and repairs completed.

The operators decided not to replace the fuse. The corrective action of changing

,

the procedure was inadequate because it did not clearly state the purpose of the added action.

)

8.0 Othatconcema l

,

When Subloop 'A1' was started the operator at the controls was making a plant

announcement concoming the start of the pump and the US, having just returned to the control room, was not aware of the pump start. Perry Administrative Procedure j

'

(PAP) 201, Conduct of Operations, Section 6.4, Reactivity Management, required the operators to maintain " positive control over core reactivity... at all times."

With adoquate oversight and control of the evolution, the duration of the

'

,

inadvertent FCV opening could have been significantly reduced and entry into TS 3.4.1 might not have been necessary.

.

-

.

n.

.

n,

--,

..,-,m m-.-

-.-,,.,,.

, _.

,,,

-., - -

.,,

n

_. _ _ _ _ _ _ _ _ _ _. - _

_ _. _.

. _ _

_ _. _ _..

.

_. _ _ _ _ _ _ _ _ _. _ _ _. -.

. _ _

-

.

..

' Once the blown fuse was identified, the sol B33 step for HPU subloop start could

".

not be completed. PAP-201 required that if the decision has been made to continue

.

with a procedure without a procedure step being completed the decision must be

documented and the documentation must include the written concurrence of a

second SRO.

l PAP-201, required in Section 6.9.2 that the US make the decision to return j

affected equipment to sMce following blown fuse events based upon the

following consideration: Trouble-shooting should normally be initiated if a fault

]

(ground or short circuit) is suspected or known to exist. The SS, acting for the US,-

decided not to initiate trouble-shooting. This was a non-conservative decision. The i

1.

final decision to start Subloop 'A1' with a blown fuse was based in part on

perceived pressure felt by the SS to return the plant to 100 percent power for

l economic reasons and other considerations described below, and that the SS had (

remembered previously running the HPU with a blown fuse. Other considerations

!

included a failed power supply for a control system for one of the reactor feedwater

.

i (FW) pumps, scheduled for repair the following week. If the second power supply j

for the FW pump control system were to have failed the FW pump would have

tripped. Without the HPU functioning, an automatic FCV " runback" would not have

been available on loss of the FW pump and the plant would have scrammed.

l

c.

Conclusions l

,

l Ineffective corrective actions for a previous event and lack of sensitivity to the need to resolve known equipment problems created an unnecessary challenge to the

-

,

operators. The ineffective corrective actions were examples of an apparent j

'

violation. Even though challenged by the ineffective corrective actions, the

i operators had opportunities to prevent or reduce the severity of the associated plant transient. However, insensitivity to FCV reactivity additions, insufficient j

questioning attitudes, and poor coordination of plant evolutions by the operators

]

j allowed the transient to occur.

.

'

The licensee was prompt and thorough in responding to the event and developed comprehensive and appropriate corrective actions. However, the corrective actions have not been completed and, in the past, the licensee has been ineffective in

,

implementing appropriate corrective actions.

i i

QL2 (Closed) LER 50-440/94-03-00: " Technical Specification Violation involving Containment Airlocks," NRC Notice of Enforcement Discretion 94-6-002 was

,

'

granted and allowed the licensee to enter containment after the lower airlock outer

door had developed mechanical problems. With the outer door inoperable and the i

upper airlock physically locked due to previous maintenance activities, it was j

l necessary to open the lower airlock inner door for about 1 minute, violating TS i

3.6.1.3. This issue was discussed in inspection Report 93-23. The problem which led to the need for enforcement discretion was not a violation. The inspectors

verified that the door operating mechanisms for all containment access airlock

doors had been modified, thus resolving the mechanical failure problems.

i i

I

.

j

.

'

..

- -

.

- -.

-

. - - -

..

--

-

--

-

, -.

-

.

-

i I

-

'

d.

)

'

.-

q ll. Maintenance

'M1 Conduct of Maintenance j

M1.1 General Comments a.

Inanection Scone (62703, 61726, 92902)

I The inspectors' observed all or portions of the following work and surveillance testing activities.

,

.

Surveillance instruction (SVI) B21-T0368, Safety Relief Valve Pipe Pressure

.

Switch Channel Functional / Calibration j

SVI C11-T2004, Scram Discharge Volume Vent and Drain Valve Operability.

'

Test

.

SVI C51-T0024, Average Power Range Monitor Gain Channel Calibration

i I

SVI E31-T079C Main Steam Line Temperature High Channel C Calibration

Riley Modules E31-604C and E31-605C.

Operations Evolution Order, Post Maintenance Testing of Domineralized

,

l Water Supply Valve for Containment and Drywell Purge (M14) Ventilation j

Penetration Shielding.

l i

b.

Observations and Findings

During the test activities the inspectors observed that personnel correctly used written instructions, that measuring and test equipment used was within its calibration limits, and that test results were appropriately evaluated.

,

The scram discharge volume valve test, SVI C11-T2004, was initially performed to i

set the timing of the air-operated vent and drain valves. After the valve timing was

'

set at about 30 seconds, the technicians waited several hours to perform the SVI as a post-maintenance test. The wait was to minimize the preconditioning effect of the valve cycles that had been used to establish the valve timing. During the SVI one of the valves was stroke timed at 31.99 seconds instead of 30 seconds. As a result of the test failure the SVI was again used to set the timing for the valves.

During that activity the stroke timing was successfully set, but repeated adjustments'of valve timing was required to meet the requirements of a note in the instruction that the inboard and outboard valves operate in sequence at a 5-second

interval with no tolerance. The note had been added to the SVI on January 6, 1997. On January 29,1997, the technicians and the operators determined that

!

the 5-second interval was intended to be a guideline with the valve operating sequence being the important aspect of the test. The technicians and operators

- promptly decided to request that the SVI be changed to make it easier to perform E

as written.

.

,

......

,

, _,,.... - -

.

.. -

-... _ -

-

_ _ _. _ _ _ _., _.

o,

,

i

-

i

.

..

,

'

c.

Conclusions

,

.

Testing activities were conducted appropriately. Although the technicians and operators appropriately decided to request that SVi C11-T2OO4, Scram Discharge Volume Vent and Drain Valve Operability Test, be changed, the SVI should have been written more clearly before it was approved for use.

M2 Maintenance and Matedal Condition of Facilities and Equipment a.

Inanection Scone (71707, 92720)

The inspectors observed plant conditions during plant walk downs. Maintenance activities were also observed.

b.

Observations and Findinas A pin hole leak in a feedwater pump seal pump suction header instrument

isolation valve weld was identified by an operator and promptly repaired using a leak seal device.

A small Main Steam Line (MSL) "D" drain line downstream of the turbine

stop valves developed a steam leak at the same weld joint as another similar drain had about 8 years ago. The turbine was taken off line shortly after the drain line separated from the MSL. Portions of the drain were replaced.

RHR low pressure coolant injection isolation valve leakage in conjunction

with additional unknown RHR leak (s) are discussed in Section 2.1.

Maintenance provided appropriate support for operations and engineering evaluation of the leakage.

The inspectors identified a loose bolt on a standby liquid control system

instrument root valve and missing screws from two safety related electrical junction boxes. Maintenance had not corrected these deficiencies by the end of the inspection period.

Replacement of Reactor Feedwater (FW) Booster Pump "C" was started and

was stillin progress at the end of the inspection period.

c.

Conclusions

'

Generally maintenance was timely and effective. The repair of the leaks on the seal pump suction line and the MSL drain line were examples of such performance.

However, the loose bolt on the standby liquid control valve and the missing junction

box screws were examples of simple tasks that were not completed even though a

.

'

special maintenance team exists to quickly complete such tasks. The work on the reactor FW booster pump had progressed appropriately, but delays were encountered and the licensee has had difficulty keeping all four of these nonsafety

"

related pumps in good working order. Maintenance had improved material condition

2

i -

,

. _ - -., _, - _,

- -.

-

,.1_.---

. - - - --------

- --- - - - - - - - - - - -

-

- _ _. -

_. - _ - _. -_ _... _. _ _ _ _ _ _.. _ _ _ _.

___ _ -. _

_. _ _ _ _ _ _ _ _ - _

.

-

.

.

in the past, however, during this inspection period no additional overall progress

.,j.

was observed.

M8 Miscellaneous Maintenance issues (62707, 61726, 92902)

M8.1 (Closed) Violation 50-440/93023-02: Failure to folicw procedures during

'

maintenance. Corrective actions identified in the licensee response to this violation included a maintenance monitoring program, inspection Report 96-05 identified positive observations of this maintenance improvement program in Section M7.1.

.

Additional management training of employees was also completed. Inspector

!

observations of maintenance activities have indicated no additional procedural

'

compliance problems in the last 2 years. Although sufficient progress has been made to close this item, additional licensee-identified procedure problems indicate

,

'

that personnel still need to be vigilant regarding the proper use and maintenance of procedures.

M8.2 (Closed) Violation 50-440/94014-01: The Division 2 Emergency Diesel Generator (EDG) was inadvertently made inoperable when a lifting rig was left installed on the exhaust testable rupture disc (TRD) due to an inadequate work procedure and

!

personnel error. The licensee's corrective actions included a Human Performance Enhancement System evaluation, additional worker training, and development of a

,

single work procedure for maintenance and testing of the TRDs for all three EDGs.

i

!

The inspectors verified that the procedures were completed and that independent i

verification for removal of the lifting rig was included.

j M8.3 (Closed) Violation 50-440/94014-02: This violation involved falsification of l

information in a work package by a work planner. The licensee conducted a thorough and effective investigation, followed by disciplinary action. However, initial licensee actions were not sensitive to the need to communicate the actions i

taken with regard to the falsification issue and to reiterate and re-emphasize that the practice was unacceptable and would not be tolerated. As part of subsequent

'

corrective actions, the Vice President, Nuclear issued a memorandum to all site

-

managers and directors directing that they address their entire staffs on responsibility for prompt identification of any suspected incidence of falsification of records. The inspectors observed that this issue was appropriately discussed at various meetings. The inspectors verified that workers were aware of the significance of falsification of records.

M8.4 (Closed) IFl 50-440/94010-05: Maintenance workers identified that some wires to the Division 2 remote shutdown safe shutdown switches on motor control conter (MCC) compartment doors had damaged insulation due to rubbing the edges of the compartment doors when opened. All switches were inspected for similar damage,

'

any damaged wiring was replaced, and General Electrical instruction gel-OOO6 was revised to include periodic inspection of the wiring during routine MCC breaker and compartment maintenance.

,

l

J

,

.

. -. - - -.

-

__ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_-

.

d

.

. [

lil. Enaineerina E1 Conduct of Engineering E1.1 Ooerability Determination for EDG Frecuency Conegm a.

Insoection Scone (37551,40500)

The inspectors observed engineering and operations personnel respond to a concern with emergency diesel generator (EDG) frequency variations allowad by the technical specifications.

b.

Observations and Findinas The inspectors asked a compliance engineer if an EDG frequency problem identified at another plant was a problem at Perry. On January 28,1997, design md systems engineering promptly identified related EDG frequency issues f m M three EDGs. These issues were documented in PlF 97-0165 which was provided to the shift supervisor (SS) in accordance with plant administrative procedure (PAP) 1608,

" Corrective Action Program." The SS made a prompt operability determination that the EDGs were operable, but concluded that a more detailed operability determination was required to be performed by engineering. PAP 1608 required that engineering complete the detailed operability determination within 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months .

Five minutes after the 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months had elapsed the inspectors asked the SS if he had the engineering operability determination.

The engineering operability determination was due in the control room at 3:35 p.m.

At about 3:40 p.m. the SS stated that he had not received the operability determination, that he should have, and that engineering had not informed him of the operability determination status. At about 3:55 p.m. the SS informed the inspe<: tors that engineering had responded to his call and told him that engineering had no immediate operability concerns. However, the operability determination had not been completed and was promised in about 30 minutes. The inspectors asked a compliance engineer the specific step location in PAP 1608 that required completion of an operability determination within 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months . The incpectors also reviewed PAP 1608 and noted that Section 6.6.1 described the process for requesting a due date extension for an operability determination. The inspectors then checked again with the SS and at 4:30 p.m. the SS told the inspectors that he had not received the operability determination and had not been contacted by engineering with additional informetion on when the operability determination would be complete. The compliance engineer informed the inspectors that the 24-hour requirement for completing the operability determination was in Section 6.3.2.3 b of PAP 1608. He also told the inspectors about the Saction 6.6.1 requirement for requesting a due date extension and that a PlF would be written because an extension request had not been initiated. The SS received the operability determination about 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months later 6nd concluded that the EDG was operable at 6:07 p.m.

__

._ - _ _

..

.

.

c. Conclusions Due to the complexity of the operability determination, the delay beyond 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months was acceptable. Although the inspectors identified that engineering had failed to provide the SS with an operability determination within 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months as required by PAP 1608; this was identified only a short time after the expiration of the 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months ; there was reasonable assurance that the SS, who was aware of when the operability determination was due, would have contacted engineering; and engineering was still working on the operability determination. The compliance engineer independently identified the failure to follow the PAP 1608 requirement for requesting a due date extension for the operability determination at about the same time as the inspectors. The engineering director stated that the individual who had

failed to request the due date extension had written PlF 97-187 describing his failure. The engineering director concluded that this would help prevent him from making the same error in the future. The engineering director planned to provide additional training on ths PAP 1608 extension request requirement for all engineering personnel involved in operability determinations. The engineering director also planned to enhance the procedure to make the extension request process more apparent. Corrective actions for the PlF will be reviewed by the corrective action review board to determine if any additional corrective actions are required. This failure to follow procedure was a licensee-identified and corrected violation (50-440/96018-03(DRP)) and is being treated as a Non-Cited Veolation, consistent with Section W.B.1 of the NRC Enforcement Policy. This was an example of poor communications between operations and engineering.

l j

E1.2 Onarability Determination Followina January 7,1997. Loss of Feedwater Event

a.

Insoection Scone (37551)

i Inspectors reviewed the corrective actions following loss of feedwater transient and plant scram which occurred on January 7,1997, including interviewing cognizant personnel and reviewing the following related documentation:

Operability determination 97-0017 completed January 8,1997.

GE letter GE-NE-B13-01805-142 TS 3.4.11 RCS Pressure and Temperature (P/T) Limits USAR Section 5.3 i

b.

Observations l

Following the loss of feedwater and scram event on Jarmary 7,1997, a cooldown rate of approximately 300 degrees per hour could be interpreted from data for the

.

vessel lower head (as measured by the RWCU system drain line temperature l

element and the vessel bottom head flange thermocouple). In addition, a cooldown rate of up to 519.4 degrees por hour was recorded for recirculation loop A and a cooldown rate in excess of 300 degrees per hour could be interpreted for recirculation loop B data. These rates exceeded the allowable cooldown rate of 100 degrees por hour specified in TS 3.4.11 and the licensee performed action A.2

.

..

.

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

--.

.

. - -... -

-

.

-. - -

- -

~ -

. _ -. - - - - -

.

... -.. -

.

j

.

<

'

,

u L

to determine RCS acceptability for continued operation, which included performing f..

operability determination 97-0017. This determination concluded that the reactor j

recirculation piping and vessel were operable following the post scram cooldown

.' event based on the following:

)

i The scram was bounded by emergency and faulted cooldown transients.

'

l GE performed a fatigue usage evaluation for the recirculation loops and

'

!

bottom head limiting components and the cumulative usage factor was

"

calculated to be less than one.

)

A fracture mechanics evaluation using the ASME Code Section XI Appendix 1.

E methodology had been performed by GE for the reactor vessel, which demonstrated adequate margin to brittia fracture.

-

$

. c.

Conclusions j

i j

This operability determination appeared to be adequate based on inspectors i

discussions with the GE technical personnel that performed these analysis and reviews of the cooldown transient data, inspectors estimated that the cooldown

-

transient had resulted in plant pressure and temperature ranges that were outsida

,

i that allowed by USAR curve C (Nuclear (Core Critical) Limit) of figures 5.3.2 l

(Minimum Temperature Required vs. Reactor Pressure, Unit 1) for approximately j

6.5 hours5.787037e-5 days 0.00139 hours 8.267196e-6 weeks 1.9025e-6 months .

4.

h E2 Engineering Support of Facilldes and EM z%

.

j.

E2.1 Poor. Pressure Gauge Installation i

a.

Insoection Scone (37551)

l The inspectors observed what appeared to be a temporary local pressure gauge on i

the mixed bed domineralizer (P22) water supply header to the standby liquid control (SLC) suction header from the SLC storage tank. The inspectors asked engineering i

if the quality of the installation was appropriate.

j b.

Observations and Findinos

!

The purpose of the P22 supply header was to provide a slight positive pressure to j

prevent any borated water from the SLC storage tank leaking past the SLC pump

suction isolation valve. The local pressure gauge had been initially established as a

temporary modification to monitor the nonsafety-related P22 header pressure after

'

a concern was identified that the SLC pump suction isolation motor operated valve

(MOV) might not open at higher P22 pressures. MOV calculations resolved the concern. However, the temporary pressure gauge was made permanent. The i

inspectors concern was that the gauge was hung from plastic straps using a valve locking tab and the pressure sensing line was not restrained. The inspectors determined that the gauge and pressure sensing line were normally isolated from

.

. -

_ _,-.

_,,,

,. _.,

,....i y

-.,. -.- -_,..,r..,,,

-

--- --._-. -

..

- _ --

-.

-..

- - _..

--. - - - -..

-

.

-

i

"

.

3 the P22 system. The inspectors also observed that there was no safety related equipment that could be harmed by a failure of the gauge or sensing line.

'

.

Engineering wrote PlF 97-225 to track resolution of concerns with the installation

design and control of the calibration of the gauge. Before the inspectors questioned the gage installation an individual had the gage calibrated, but did not insure that it

.

j.

was placed on a regular calibration frequency.

c.

Conclusions

Although the poor quality gage installation had no direct potential safety

-

consequences, the fact that personnel from various site organizations had accepted the installation indicated that some individuals still did not have an adequate

,

j questioning attitude. The individual who had a questioning attitude sufficient to get i

the gage calibrated did not follow through on getting the gage placed on a regular calibration frequency.

{

E2.2 Review of Uodated Safety Analysis Reoort (USAR) Commitments 1-The inspectors reviewed applicable portions of the USAR that related to the areas inspected; no inconsistencies were identified. The inspectors also reviewed items

,

j that the licensee had identified during its reviews of the USAR. The licensee

included the inconsistencies in its corrective action program. These may be reviewed in a future inspection based on the NRC's policy (61 FR 54461, October 18,1996) for the review of licensee-identified USAR inconsistencies.

The inspectors also reviewed current safety evaluations for some of the identified

,

i USAR inconsistencies. The safety evaluations were timely and appropriate for the identified issues. It appeared that the licensee had addressed the inconsistencies j

appropriately in accordance with the safety significance.

The NRR project manager (PM) reviewed the submittal of October 8,1996, containing a summary of 281,10 CFR 50.59 safety evaluations performed between

'

September 19,1994, and April 10,1996. The summary descriptions and evaluations contained in that report were understandable and were acceptable.

~

The PM also reviewed nine modification document packages. Each modification checklist was filled out regarding interface reviews, affected documents list, and the 50.59 applicability checklist. The evaluations performed for 50.59 considered the relevant safety issues subject to each change.

,

E8 Miscellaneous Engineering issues (92720,92903)

E8.1 (Closed) Licensee Event Report (LER) 50-440/94004-00:

Containment Penetration

"

-

Leak Rates exceeded TS Limits." This LER is closed based on issuance of a revision, LER 50-440/94004-01.

.

E8.2 (Closed) LER 50-440/94005-01: Loss of Safety Function for Emergency Closed

"

'

Cooling System 'A,'" Revision 1. Inspection of the subject of this LER was '

.

--s

-.--

w

,.,,-

,

,,i-y.

_

.u----_-___~--- - -.- - - -, --- _-,

r--,.,-

- _ -

.

-

.

-

,.

documented in inspection Report 94-08 and the LER is closed based on that

~

inspection.

,

E8.3 1 Closed) LER 50-440/94011-01: " Unexpected Automatic Closure of Containment

'I

,

l Isolation Valves," Revision 1. Inspection of the subject of this LER was documented in inspection Report 96-11 and the LER is closed based on that inspection.

E8.4 (Closed) LER 50-440/94012-00: " Equipment Malfunction Leads to (2) Unexpected j

Annulus Exhaust Gas Treatment System Auto Starts." This LER is closed based on

'

issuance of a revision, LER 50-440/94012-01.

E8.5 (Closed) LER 50-440/94014-00: " Failure to Perform Adequate Leak Rate Test."

This LER is closed based on issuance of a revision, LER 50-440/94014-01.

E8.6 (Closed) LER 50-440/94015-00: " Potential Loss of ESW System due to Loss of Keepfill." This LER is closed based on issuance of a revision, LER 50-440/94015-01.

E8.7 (Closed) LERs 50-440/96006-00 and 01: "MOV Control Circuitry Design Deficiency Results in Fire Protection Violation." Inspection of the subject of this LER was documented in inspection Report 96-16 and the LER is closed based on that inspection.

.

E8.8 (Closed) LER 50-440/9401940: "RCIC secondary containment bypass leakage."

An engineering review found that the reactor core insolation cooling (RCIC) system design relied upon the non-safety, non-seismic gland seal air compressor (GSAC) as a basis for crediting RCIC as a closed system outside containment. The design was not in compliance with the requirements of 10 CFR 50, Appendix A, General Design

,

'

Criteria. The inspectors verified that the licensee had completed corrective actions; updated the USAR to include the identified potential secondary containment bypass leakage paths and that surveillance instructions were in place to test the associated penetrations. This licensee-identified and corrected violation is being treated as a

,

Non-Cited Violation (NCV 50-440/96018-04(DRP)), consistent with Section Vll.B.1

'

of the NRC Enforcement Policy, NUREG-1600.

E8.9 (Closed) IFl 50-440/94004-04: The inspectors questioned the adequacy of fire protection and other system pipe supports for the Unit 1 and Unit 2 interface boundaries, Engineering conducted walk downs and concluded by calculations that the piping interface boundaries were adequately supported. Currently, engineering is conducting a new evaluation of the Unit 1/ Unit 2 interface boundaries since Unit 2 was abandoned. Previous evaluations were conducted with the expectation that Unit 2 would be completed. In August 1996, the inspectors opened Unresolved

Item (50-440/96004-04) when the inspectors identified that a Unit 2 battery room

was not completed, but was being used for an alternate safety related battery.

Given the identified boundary discrepancies identified in 1994, it would appear that an opportunity to identify the problems with the battery room condition was

.

-. -

-.-

_

. -

-

.

-

. -.

. -. -

-

.. -. -

.- --

.

missed. This potential missed opportunity will be considered when Unresolved item f.

(50-440/96004-04) is evaluated.

E8.10 (Closedl IFl 50-440/94010-04: A standby liquid control (SLC) squib valve anomaly

occurred when the SLC 'B' squib valve indicated, by intermittent alarm, a loss of circuit continuity. Troubleshooting at the time did not confirm a loss of continuity or identify any reason for the intermittent alarm. Later, a destructive examination

revealed that a loose screw in the meter module affected the associated relay. All GE type 195 meter modules, manufactured by LFE instruments, had a common

'

failure mode where the screw could back out due to vibration. In January 1995, LFE instruments began using thread locking compound on the screw. The

.

licensee's corrective actions required that for all future purchases of GE type 195 meter modules, the modules would be verified to have a date code of January i

1995 or later. The inspectors verified that this requirement was incorporated in procurement criteria. The inspectors also verified that this defect had been

,

evaluated by GE for safety significance. GE had determined that the defect could

,

}

not have prevented the function of any safety related equipment.

i IV. Plant Support P2 Status of Emergency Preparedness Facilities and Equipment P2.1 Remedi=I Actions for Daemmber 19.1996, I a== of Offsite Communications a.

Insoection Scone (71750)

On December 19,1996, the shift supervisor determined that the plant had a

'

significant loss of offsite communications capability and classified the loss as an Unusual Event. The inspectors reviewed some of the remedial actions that had been completed.

.

b.

Observations and Findinas The inspectors verified that the excavation that had led to the severing of the

,

offsite telephone lines had been completed and restored te its normal grade. The

,

'

. inspectors verified that the licensee had placed six cellular telephorles at two

!

locations in the plant for use if needed. The inspectors also verified that a new

-

tower had been constructed for the offsite microwave telephone link that served as a backup to the underground telephone line.

{

c.

Conclusions F

The licensee remedial actions were prompt and minimized the possibility of occurrence of another significant loss of offsite communications capability.

l

.

,

e..

... -.

m

+

.,, -

,

-

.

-

.

..

.

-

.-

.

,

O

.

V. Management Meetinas

-

,

X1 Exit Meeting Summary l

The inspectors presented the inspection results to members of licensee management at the conclusion of the inspection on February 3,1997. The licensee acknowledged the findings presented.

The inspectors asked the licensee whether any materials examined during the inspection should be considered proprietary. No proprietary information was identified.

_ _ _ _ _

...

-

-.-

-

.. - -.

..

A

i i O

~

PARTIAL LIST OF PERSONS CONTACTED

-

Licensen

!

J. P. Stetz, Senior Vice President, Nuclear L. W. Myers. Vice President, Nuclear

R. D. Brandt, General Manager, Nuclear Power Plant Department i

i W. R. Kanda, Director, Quality and Personnel Development Department N. L. Bonner, Director, Nuclear Maintenance Department

'

J. J. Powers, Director, Nuclear Engineering Department L. W. Worley, Director, Nuclear Services Department J. Messina, Operations Manager

INSPECTION PROCEDURES USED i

IP 37551:

Onsite Engineering

.

IP 40500:

Effectiveness of Licensee Controis in identifying, Resolving, and Preventing Problems

>

l IP 61726:

Surveillance Observations l

IP 62707:

Maintenance Observation IP 71500:

Balance of Plant inspection

,

IP 71707:

Plant Operations lP 71714:

Cold Weather Preparation IP 71750:

Plant Support Activities IP 92700:

Onsite Followup of Written Reports of Nonroutine Events at Power Reactor Facilities

)

'

IP 92720:

Corrective Action IP 92901:

Followup - Operations j

j.

IP 92902:

Followup - Maintenance i

j lP 92903:

Followup - Engineering IP 93702:

Prompt Onsite Response to Events at Operating Power Reactors

.

k

,

I It

27

'

_

-

.-

-

. - -- -

-

_

-- ---

-

.

.- _

.

$

-.

-

ITEMS OPENED, CLOSED, AND DISCUSSED

.

Onened 50-440/96018-01 VIO Inadequate procedure to determine cooldown rate 50-440/96018-02 VIO Apparent corrective action violation, FCV 50-440/96018-03 NCV Extension not requested for operability determination 50-440 96018-04 NCV RCIC gland seal compressor not safety related Closed 50-440/93023-02 VIO Failure to follow maintenance procedures 50-440/94003-00 LER TS violation, NOED, open lower airlock door 50-440/94004-00 LER Containment penetration leaks 50-440/94004-04 IFl Fire protection pipe supports questioned 50-440/94005-01 LER Loss of ECC safety function, cold water 50-440/94010-04 IFl SLC squib valve continuity indication intermittent 50 440/94010 05 IFl Damaged wires on MCC doors 50-440/94011-01 LER Automatic closure of containment isolation valves 50-440/94012-00 LER Unexpected AEGTS automatic starts 50-440/94014-02 VIO False information in work package 50-440/94015-00 LER ESW keepfill design error 50-440/94019-00 LER RCIC gland seal compressor not safety related 50-440/96006-00 LER MOV design affects fire protection 50-440/96006-01 LER MOV design affects fire protection 50-440/96018-03 NCV Extension not requested for operability determination 50-440-96018-04 NCV RCIC gland seal compressor not safety related Discussed 50-440/96004-04 URI Unit 2 battery room not completed (E8.9)

t

,

-

.

_ - _ _ _ _

._.

_

_.

_

.

!

.

LIST OF ACRONYMS USED

-

e BOP BALANCE OF PLANT

CFR CODE OF FEDERAL REGULATIONS CRD CONTROL ROD DRIVE DRP DIVISION OF REACTOR PROJECTS EDG EMERGENCY DIESEL GENERATOR ERIS EMERGENCY RESPONSE INFORMATION SYSTEM ESW EMERGENCY SERVICE WATER FCV FLOW CONTROL VALVE FR-FEDERAL REGISTER FW FEEDWATER GE GENERAL ELECTRIC

,

gel GENERAL ELECTRICAL INSTITUTE

'

GSAC GLAND SEAL AIR COMPRESSOR HPU HYDRAULIC POWER UNIT IFl INSPECTION FOLLOW-UP ITEM LER LICENSEE EVENT REPORT LPCI LOW PRESSURE COOLANT INJECTION MOV MOTOR-OPERATED VALVE MSL MAIN STEAM LINE NPF NUCLEAR POWER FACILITY NRC NUCLEAR REGULATORY COMM.lSSION NRR NUCLEAR REACTOR REGULATION PAP PERRY ADMINISTRATIVE PROCEDURE PDR PUBLIC DOCUMENT POOM PIF POTENTIAL ISSUE FORM PM PROJECT MANAGER PORC PLANT OPERATIONS REVIEW COMMITTEE RCIC REACTOR CORE ISOLATION COOLING RHR RESIDUAL HEAT REMOVAL RSE RESPONSIBLE SYSTEM ENGINEER RWCU REACTOR WATER CLEANUP UNIT SLC STANDBY LIQUID CONTROL sol SYSTEM OPERATING INSTRUCTION SRO SENIOR REACTOR OPERATOR SS SHIFT SUPERVISOR STA SHIFT TECHNICAL ADVISOR SVI SURVEILLANCE INSTRUCTION TS TECHNICAL SPECIFICATION TRD TESTABLE PUPTURE DISC USAR UPDATED SAFETY ANALYSIS REPORT URI UNRESOLVEDITEM VIO VIOLATION WO WORK ORDER

- -

- -..

-..

.

. _

-

. -. - --

- - -

- -.

.

' PARTIAL LIST OF DOCUMENTS REVIEWED DURING THIS INSPECTION (96-17)

$

- ALARA Report - December 1996,1/3/97

,

Control Room standing orders, various dates

Control Room computer printouts, various parameters, various dates Control Room daily instructions, various dates

'

,

Control room daily instructions, supplemental reading, various dates Control room safety tag log, various dates

'

Control room strip charts, various parameters

'

Control room annunciator status books, revisable format various dates i

Control room LCO log, various dates

.

Deficiency tags, various locations, various dates

'

Fire extinguisher inspection tags, various locations, various datos Forced Outages Meeting Minutes (12/19/96)

,

Forced Cutages Meeting Minutes (01/27/97)

-

GEK-63100, Operation and Maintenance instructions, Hydraulic Control Unit 4/80 12/20/96, Audit Report PA 96-22 Access Authorization Program

.

-

. Manager's Communication & Teamwork Meeting, 12/20/96,12/23/96,12/27/96, j

12/30/96, 01/03/97, 01/06/97, 01/08/97, 01/10/97, 01/13/97, 01/15/97,

01/17/97, 01/20/97, 01/22/97, 01/24/97, 01/27/97, 01/29/97, 01/31/97, 02/03/97 Minutes for Company Nuclear Review Board Meeting 122 Perry /259 Davis-Besse i

)

- Minutes for Company Nuclear nova Board Meeting 123 Perry /260 Davis-Besse

!

Monthly Access Level Use Review For December, Dated 01/06/97

'

l

. Monthly Operations Report - December 1996

NEl Concerned About NRC Interpretation of information Notice (01/21/97)

'

NRC INSPECTION 96018 DEBRIEF SUMMARY - 01/29/97 j

,

Operations Administrative Control Tags, various locations, various dates

-

Operations information Tags, various locations, various dates

PAP-0201, Rev. 9, Conduct of Operations,03/28/95 i

PAP-0810, Rev. O, Tool / Equipment Accountability,10/9/95

'

PAP-0205, Rev. 8, Operability of Plant Systems,06/14/96 i

PAP-1608, Rev. 3, Corr 6ctive Action Program, 12/05/96

Perry Lines Weekly, 12/19/96, 01/09/97, 01/30/97 i

Perry Daily Report - Tuesdays and Thursdays, except December 24 and 31,1996, and l

'

January 2,1997.

Perry News Flash,01/09/97

Perry News Flash,01/20/97 Plant Log, Vol. 32,(12/20/96) Page No. 46 - 92 (02/03/97)

i Plant strip charts, various parameters, various dates PNAD Operational Surveillance Report No.96-054 Preparation of Winter Operations (10/23/96)

,

PNAD Operational Surveillance Report No.97-001 Conduct of Maintenance and Operations during the January 1997 Forced Outage (1/15/97)

,

l

>

~

-

.

-

..

-

,. -.

.

- - -

-.

.

-

.

-

.

'$

Plan of the Day

- 12/23/96,12/26/96 and 01/02/97

- 01/06/97 thru 01/10/97

,

- 01/13/97 thru 01/17/97

-

- 01/20/97 thru 01/24/97

- 01/27/97 thru 01/31/97-02/03/97

'

PORC REVIEW ITEM Safety Evaluation #96-0139 (11/20/96)

PORC REVIEW ITEM USAR Change Request 96-169 (11/01/96)

PORC REVIEW ITEM DCN-5423 (11/27/96)

PORC REVIEW ITEM DCP 94-0027 REV. 6 (11/27/96)

PORC REVIEW ITEM SCR #1-96-12144,1145, and 1146 (12/11/96)

PORC REVIEW ITEM DCN 5451 (12/16/96)

PORC REVIEW ITEM DCP 96-00022 (12/10/96)

PORC REVIEW ITEM DCP 96-00079 (12//20/96 PORC REVIEW ITEM DCP 96-00079A (12/20/96)

PORC REVIEW ITEM DCP 95-000798 (12/20/96)

PORC REVIEW ITEM USAR Change Request 96-172 - Site Reorganization (12/23/96)

PORC REVIEW ITEM DCP-5033 Drywell Shield Doors (12/23/96)

PORC REVIEW ITEM DCP 95-0102 (1/10/97)

j

'

POS Performance Indicators - December 1996 PlF 95151,1/24/95, Control Room Cabinet Doors Potential issue Forms No. 96-3769 through 96-3811 Potential Issue Forms No. 96-3400 and 96-3408 Radiation Work Permit 97006 Radiologically Restricted Area Radiation Surveys, various dates

,

Safety Tags, various locations, various dates SS Relief / Turnover Checklist, 12/24/96 System Description Manuals - Various TS Change Request, CR No.97-008

,

Temporary Modification Tracking Report,01/01/97 Unit Log, Unit 1, Vol. 90, (12/20/96) Page No. 106-150 (01/07/97)

Unit Log, Unit 1, Vol. 91, (01/07/96) Page No. 1 -79(02/03/97)

,

'

Updated Safety Analysis Report Weekly Effluent and Release Rate Data Report, about 01/05/97 l

-

-.

..

-

v t e Inspection Report - Perry - 1996018
80 \| 81 \| 82 \| 83 \| 84 \| 85 \| 86 \| 87 \| 88 \| 89 \| 90 \| 91 \| 92 \| 93 \| 94 \| 95 \| 96 \| 97 \| 98 \| 99 00 \| 01 \| 02 \| 03 \| 04 \| 05 \| 06 \| 07 \| 08 \| 09 \| 10 \| 11 \| 12 \| 13 \| 14 \| 15 \| 16 \| 17 \| 18 \| 19 \| 20 \| 21 \| 22 \| 23 \| 24 \|
1996 Quarterly Integrated 1Q (0) 2Q (0) 3Q (0) 4Q (0) Assessments Mid Cycle End of Cycle 7(0) 8(0) 9(0) 10(0) 11(0) 12(0) 13(0) 14(0) 15(0) 16(0) 17(0) 18(0) Security Operator Exams Emergency Planning Other

IR 05000440/1996018

Text

Navigation menu

Search