IR 05000369/1988024

Insp Repts 50-369/88-24 & 50-370/88-24 on 880801-0916. Violations & Unresolved Item Noted.Major Areas Inspected: Past & Current Operability of Hydrogen Skimmer Sys

ML20204J236
Person / Time
Site:	McGuire, Mcguire
Issue date:	10/06/1988
From:	Croteau R, David Nelson, William Orders, Peebles T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
To:
Shared Package
ML20204J225	List: IR 05000369/1988024 ML20204J220
References
50-369-88-24, 50-370-88-24, NUDOCS 8810250048
Download: ML20204J236 (10)

Text

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

.

.

, ..p***%q

'

3* ** UNITED STATES

j NUCLEAR REGULATORY COMMISSION

. o MEGION 11 f

101 MARIETTA ST N.W.

! %o,,,e f ATLANTA. GEORGIA 30323

1 i

q Report Nos.: 50-369/88-24 and 50-370/88-24 Licensee: Duke Power Company i 422 South Church Street

>

Charlotte, NC 28242

) License Nos.: NPF-9 and NPF-17 l Docket Nos.: 50-369 and 50-370 Facility Name: McGuire Nuclear Station 1 and 2 l Inspection Conducted: August 1 - September 16, 1988 Inspectors: A W. Order's, Stin1or Residen l~n spector A ~

J Date-bd Signed VYnYo D. NeTsoh, Resiitfeht insp , t'or l0 * S "U Date signed Y bYs R. Cretefu,' ResicTent Inspec r

/0* $Y date Signed Approved by: dI T. A. Pee 5Ws, SidT.fon Chief

/ /d - g'6 Oate $1gned Division of Reactor Projects SUMMARY Scope: This special inspection was performed to evaluate the past and current operability of the hydrogen skimmer syster Results: In the areas inspected, three violations and one unresolved item were identified. A weakness was noted in the licensees method of writing, controlling, and tracking of Justifications for Continued Operation (JCOs). The violations are being considered for escalated e9forcemen PDR ADOCK0500pg9  ;,

O

____

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

'

.

j .

..

f .

l REPORT DETAILS Persons Contacted Licensee Employees

• J. Boyle, Superintendent of Integrated Scheduling

"B. Hamilton, Superintendent of Technical Services s

1. • T. McConnell, Plant Manager

• N. Rutherford, General Office, Licensing

• M. McIntosh, General Manager, Nuclear Support

• M. Sample, Superintendent of Maintenance

1. • R. Sharpe, Compliance Engineer

1. • J. Snyder, Performance Engineer j *B. Travis, Superintendent of Operations R. White, IAE Engineer Other licensee employees contacted included construction craftsmen, technicians, operators, mechanics, security force members, and office personne ;

• Attended exit interview on August 24, 1988

1. Attended exit interview on September 16, 1988 NRC representatives included:

"C. Hehl, Deputy Division Director, Division of Reactor Projects

• T. Peebles, Section Chief, Division of Reactor Projects

2, Unresolved Items ,

An unres>J1ved item (UNR) is a matter about which more information is  !

required to determine whether it is acceptable or may involve a violation or deviatio One unresolved item was identified and is discussed in  !

paragraph 6 of this repor I Executive Summary During a containment closeout inspection of the Unit I reactor building at the conclusion of the refueling outage in 1987, the resident inspectors ,

noted that a damper in the hydrogen skiemer (VX) system appeared to be ,

shut. The licensee was notified and subsequently visually checked all

'

Unit 1 VX dampers, The licensee stated that the dampers appeared to be in the correct position with the exception of the damper questioned by the {

inspector which was found to be mispositioned slightly ir the closed  ;

direction but not fully closed. The licensee also committed to perform a flow balance on the Unit 2 VX system during the next refueling outag During that outage, the as found flow balance failed to meet the required flowrates. The system flows were adjusted but design basis flows could still not be achieved for some compartments, l l

l . . ___ . - ._ - . - - _ - _ _ ..- -

_ _ _ - - .

_ . _ _ .. __ _ _ _

,

'

.

'

.. >

,

l

'

t i

.

A justification for continued operation (JCO) was written by the '

licensee's design engineering office to justify startup of Unit 2 with ,

VX flow in several compartments below the amount required by the system design basis. The inspectors reviewed the JC0 and brought several concerns to the attention of the licensee relating to past and present operability of the VX syste These concerns are detailed below, i

!

4 System Description l The hydrogen skimmer (VX) system consists of two fans, each with a j capability of 3,000 cfm, with common ductwork leading to various enclosed l areas inside containment. These areas include the steam generator  !

enclosures, pressurizer enclosure, accumulator rooms, instrumentation i room, reactor head area, and fan rooms. The function of the system is to prevent the accumulation of hydrogen in these restricted areas resulting <

from a loss of coolant accident. This is accomplished by drawing air out l cf the compartments at rates which will limit the local hydrogen concen- l trations to four percent by volume as described under the VX design basis '

in the FSAR. Either one of the two redundant trains (one fan per train)  !

is designed to meet the design basis by keeping hydrogen concentration f below 4 percent volum l

!

Another related system, the hydrogen mitigation system (HMS) consists of seventy ignitors located in strategic areas inside containment which serve i to initiate combustion of significant amounts of hydrogen which would be generated af ter a degraded core accident. The system serves to reduce [

hydrogen concentration through controlled burns rather than allowing i random ignition of high concentrations of hydroge ;

r It should be noted that the hydrogen mitigation system provides hydrogen [

control to corrbat the consequent;es of the hydrogen release generated '

during a degraded core accident and to assure that the scenario does not  !

evolve to a breach of containment or an adverse affect upon the [

functioning of essential equipment as specified in the FSAR. In contrast,  !

the VX system is designed to keep hydrogen concentrations below 4 percent  !

volume in various enclosed compartment,s f ollowing a loss-of-coolant I accident. Therefore, the HMS and VX systems are provided for different I accident aseumptions and allowable consequence l Sequence of Events ,

i The following sequence of events was obtained primarily from the licensees  !

incioent investigation report, j i

Date Event I

11/03/87 NRC personnel notice 6 that some Unit 1 VX system

) dampers appeared to be closed. NRC personnel  !

{

I questioned OPS personnel about the positions of I the Unit 1 VX system damper !

I  !

i I I

!

'

I

. .

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

'

.

.

'

. ..

Date El'"t (cont'd)

11/09/87 OPS personnel verified that each Unit 1 VX system i damper was ir; its pre-operations 1 position as evidenced by paint shadows on the damper l actuator ,

NRC personnel were concerned about the potitions of the Unit 2 VX system dampers, and consequently, ,

Operations personnel agreed to do a flow balance test on the Unit 2 VX system dampers during the f 1988 Unit 2 Refueling Out&g /19/88 Performance personnel took "as found" flow measurements using a velometer on Unit 2 Train A and Train B of the VX system and found that some l VX system damper compartment flows did not meet :

the McGuire Final Safety Analysis Report (FSAR) ;

requirements. Also two dampers were found shu ,

7/20/-21/88 Performance personnel retested the Unit 2 VX system using a flew hood and adjusted the dampers in an attempt to achieve flow rates that were l listed in the FSA .

l 7/21/88 Design Engineering personnel reevaluated the i VX system flow rate l Performance personnel tera'nated the flow balance i test on Unit 2 before completing flow measurements :

on Train B of the VX syste !

i Untt 2 entered Mode 4 after a verbal operability ;

determination was m,d l

,

7/22/SB Design Engineering personnel issued an Operability l Determination based in part on the operability of -

the Hydrogen Mitigation sy s t e'n , which they r believed would be a suitable substitute for the l VX syste /1/88 NRC personne) questioned the Unit 2 Operability )

Determination because Design Engineering personnel ;

considered the Hydregen Mitigation system a i substitute for the VX syste I l

,

- . __

'

,

,

,

..

Date Event (cont'd)

8/2 15/88 Design Engineering personnel began reevaluating the VX system flow rates in a manner consistent with the Regulatory Guide 1.7, Revision 2, and 10 CFR 50.44, which reduced the required flow rates for the VX system. Three Unit 2 B train compartment flows (as measured by hand held <

velometer) were still below the required flow rate /16/88 Performance personnel took additional f ' .<

measurements on Unit 2 B train of the VX system using accurate instrumentatio /19/88 Design Engin6 ring personnel issued a new Unit 2 Operability outermination and concluded that the present flow balance condition of the VX system was sufficient and was considered operabl . As Found Condition of VX and System Flow Balance As was cciamitted to, the Unit 2 VX system was tested in accordance with TT/2/A/9100/217 on July 18 through Joly 21, 1988. The as found flowrates were found to be less than those specified in the FSAR in several compart-ments including the reactor head area where valve (dampers) PVX13 and 2VX14 were found fully shu It appears the valves positions had not been verified since preoperational testing. These valves are in an area where they may have been inadvertently mispositioned during a refueling outag The system was subsequently adjusted but flows from some compartments could not be brought into compliance with FSAR requirement The values delineated in FSAR Table 6.6.2-1 were derived to limit post LOCA hydrogen concentrations to less than 3.5 percent volume. Regulatory Guide 1.7, Control of Combustable Gas Concentrations in Containment Following a loss of Coolant Accident, specifies that limiting hydrogen concentration to 4.0 percent volume is acceptable. The design basis text of the FSAR also specifies 4.0 percent. When the flowrates required to maintain concentrations below 3.5 percet, could not be achieved, the licensee's design engineering group calculated reduced flowrates which would limit hydrogen concentration to 4.0 percen The VX system could still not meet those reduced flowrates in two compartments with the A fan running and at least five compartments with the B fan runnin As found testing and initial post adjustment testing was performed using a hand held flow device which has limited accurac The test was reperformed using a mora accurate flow hood device to better determine actual post l adjustment flow Flows in two compartments were still below the design basis using the flow hood with the A train fan operating. While in the process of performing the B train test using the flow hood, management

p l

..

.

.

.. ..

.

decided to terminate the test before cor% 1etion based on a JC0 received from the design engineering group and the fact that the test was impacting unit startu The JC0 pt- "ded by design engineering essentially stated that the degraded Vi / stem in combination with the hydrogen mitigation system (ignitors) tould prevent accumulation of hydrogen, thereby providing adequate protection for continued safe operation of both station unit !

!

The inspectors identified a number of problems with the JC0 as described below:

-

This JC0 applied to both units when the condition of the Unit i VX system was based on an inadequate preoperational test documented in a ,

memo to file in which flows were measured with. both trains running i simultaneousl l

- The JC0 did not contain the accurate train 28 flow data, only data obtained using the less accurate method was availabl The JC0 did not address the current operability of the ignitors in the compartments with inadequate flo The inspectors later deter- i mined that all ignitors in both units were operating properly but this information was assumed by design engineering without initial confirmatio The design basis of the hydrogen skimmer system was not met in that .

'

the system could not maintain various compartments 'below 4 percent volume hydrogen yet the JC0 concluded that the "design basis for applicable hydrogen producing events are met." ,

-

A 10 CFR 50.59 safety evaluation was not performed when the licensee  !

"

- decided to leave the VX system in a candition different that that described in the FSA CFR 50.59 states that the licensee may

, make changes to the facility as described in the safety analysis  !

"

report without prior commission approval unless a T.S. change is i required or the change involves an unreviewed safety question. An l unreviewed sr.fety question is considered to exist if the probability

'

i of an occurrence or the conseqLences of an accider.t or malfuretion of

,

'

equipment important to safety previously evaluated in the stfety i

!

analysis report may be increased or if the margin c.T safety as defined in the basis for any technical specificatten 11 reduce i 10 CFR 50.59 also states that the licensee shall maint3 n records Sf 4 the changes including a written safety evaluation prov' ding the basis  :

for the determination that the change, test, or experi A..it does not i involve an unreviewed safety question. In the case of the VX system -

-

being changed to reduced flow rates a 10 CFR 50.59 safety evaluation ,

was not performed. This item is identified as an apparent vio'iation i (369,370/83-24-02) for failure to perform a 10 CFR 50.59 safety l evaluation,

<

,

i i i

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

__ _ __ __. ,__ _ - . _ . . _ - . _ . _ _ _ - . __ _ , _ . , _ _ _ _ _ _ . , - m_ . _ . , _ . . _ _ _ _ _ _

r

'

.

,

. ..

- The JC0 was based on the belief that the system was not needed since the hydrogen ignitors would provide a controlled burn. This flawed assumption in essence deleted the need for a TS required syste As described in paragraph 4, the hydrogen mitigation (ignitor) system and the hydrogen skimmer system are two separate independent systems each designed to prevent hydrogen concentration from reaching an explosive concentration yet each designed for a differe7t accider,t scenario. The hydrogen mitigation system is designed to mitigate the consequences of a degraded core accident and the hydrogen skimmer system is designed to mitigate the consequences of a design basis loss-of-coolant accident. The existence of one of the systems cannot be used to justify inoperability of the other. The T.5, does not allow the VX system to be inoperable based on the operability of the HMS. The inspectors concluded that the hydrogen skimmer system was inoperable based on the systems inability to reduce post LOCA hydrogen concentration below 4 percent in some compartment This situation was discussed with the licensee management who committed to reevaluate the situation and calculate hydrogen skimmer system flow rates required to maintain hydrogen concentrations below 4 percent based on a 1.5 percent zirconium-water reaction which is currently allowed by the NRC rather than the 5 percent assumed in the FSAR. The licensee believed that the existing VX ficwrates would be greater than the required flow rates using 1.5 percent zirconium reaction therefore the VX system was considered operable while the analysis was reperformed (from August 2,1988 through August 15,1988).

The licensee also believes that the Unit 1 VX system is operable based on preoperational testing with both fans running (although each train was not tested individually) and operations personnel visually checking the damper positions in 198 The licensee has committed to performing a flow balance on the Unit 1 VX system during the next refueling outage scheduled to start on October 11, 1988. Thi> item is identified as Unresolved Item 369/88-24-04 pending results of the as found testin The above issues were discussed with Region II and NRR managemen Several concerns arose from this situatio The Unit 2 VX system was inoperable for an undetermined time in that it could not perform its intended function of maintaining hydrogen concentration below 4 percent volume in the reactor vessel head area and possibly other areas as wel Reactor vessel head area skimmer valves 2VX13 and 2VX14 were found shut during the 1988 refueling outage. In addition as found flow rates for three compartments were zero when the B train fan was running. This constitutes a violation of T.S. 3.6.5.6 which requires two independent Centainment Air Return and Hydrogen Skimmer Systems be OPERABLS in modes 1, 2, 3, or (369/88-24-01)

_ _ _ ._

.

, .,

!

! -

When Unit 2 VX system flows could not be adjusted to obtain FSAR l required flowrates the JC0 dated July 22, 1988, was inadequate as discussed earlie Licensee management decided to not continue to obtain the more accurate B train VX flow balance using the flow hood once the JC0 was received. Although the inspectors do not believe licensee management places schedules above safety, this type of occurrence may foster ;

these beliefs. Better communication is needed down the'line organt-zation to ensure personnel involved are aware of why safety is not being comprised in these situations.

l In a letter dated August 9, 1988, the licensee documented their conclusion I that the VX system was capable of performing its design function and described planned actions. Operability was based on the expectation that additional calculations using a lower metal-water reaction of 1.5 percent would demonstrate that the current flowrates were acceptable. The licensee also stated that tne analysis would be completed by August 15, 1988, and would be made available to the NRC. In addition, a Unit i VX system flow balance would be performed during the upcoming outage scheduled to start on October 11, 1988, flow measurements would be made on the Unit 2 B train l VX system during the next outage involving cold shutdown, and a hardware I modification was being evaluated to allow better flow balancing.

I

! The reanalysis was completed on August 15, 1988, and indicated that flows from three compartments on B train of Unit 2 were below that require Fortunately the three compartments were acenssible at pwe On August 15, 1988, the licensee made a containment entry at 100 percent I power to measure flowrates from the three compartments in question using l the more accurate flow hood method. Flow from only one of the two dampers l in each compartment was measured due to personnel safety consideration These measurements indicated that two compartments were still below the

required flowrates. On August M .988, the licensee measured flowrates l

from the remaining lines in t..a .mpartments in questio The combined

'

flowrates were satisfactory based on the new acceptance criteri The '

licensee's reanalysis of the required compartment flowrates was reviewed l by the NRC and the NRC staff agreed with the reanalysi . Preoperational Testing Preoperational test data was reviewed which revealed that the tests t

were inadequate to demonstrate adequate VX syster. flow balance. Unit 1 l preoperational test TP/1/A/1200/11, Containment Air Return and Hydrogen l l Skimmer System Functional Test, did not perform a system flow balance. A ,

l memorandum to file dated March 23, 1981, however, documented a flow 1

'

balance on the VX system performed on February 26, 1981, with both fans '

operating. No actual completed preoperational flow balance test could be found. This test was inadequate to demonstrate that each train would function individually as designed, therefore, the system may have been

.

inoperable since initial startup in violation of TS 3.6.5.6. This item is l

l l

l l

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

,

. .. ...

,

J identified as an apparent . violation 369,370/88-24-03, for an inadequate

_preoperational test and failure to perform the test in accordance with a written procedure as required by 10 CFR 50 Appendix B, Criterion X Unit 2 preoperational test TP/2/A/1200/030, Containment Air Return and Hydrogen Skimmer System Functional Test, was also reviewed and found to be inadequate in that only the A fan was run to accomplish the flow balaac A train B flow balance was not performed in preoperational tests; therefore, the B train may have been inoperable since initial operation in violation of TS 3.6.5.6. This item constitutes a second example of apparent viola , ,

tion 369,370/88-24-03 for inadequate preoperational testin '

The licensee has stated that the preoperational testinq program of ,

appropriate systems and components will ' be- reviewed to oltermine its adequac . Justification For Continued Operation Program l The licensee initiates JCOs to evaluate the operability of systems and components which are in a degraded conditio The JCOs ire usually written by design engineering. No instructions exist for cont olling the 3 program and furthermore, 10 CFR 50.59 evaluations are not routinely ,

performed as part of the JC !

i In as much as JCOs justify perfornance changes in systems and components from those previously analyzed in the safety analysis report, a 10 CFR -

50.59 evaluation is required to determine if an unreviewed safety question ,

exists. One example was discussed in Inspection Report 369,370/88-20, !

involving degraded performance of the 28 containment spray (NS) heat ;

exchanger. The heat transfer capability of the 2B NS heat exchanger was found to have decreased below the heat transfer capability stated in the safety analysis report. A JC0 was written analyzing the degraded condition and its affect on containment pressure in an acciden The JC0 stated that peak post LOCA containment pressure would stay below the allowable pressure based on the existing heat transfer capability and extra ice inventor However, no 10 CFR 50.59 evaluation was performed to determine ,

I whether the change in heat transfer capability constituted an unreviewed l safety question. The licensee stated that 10 CFR 50.59 safety evaluations '

4 are p ' emed with the processing of a procedure change if a procedure

~

change is needed based on a JCO. In this case a 10 CFR 10.59 safety

'

evaluation was done when the performance test procedure was changed af ter

'

the JC0 was writte The licensee's control and tracking of JCOs is inadequate primarily ,

.

because ne program exists to control the proces Outstanding JCOs were i 4 previously kept in the control room in the Technical Specification Action

'

Item Logbook (TSAIL) in loose form and not serialize During the inspection period the licensee placed all outstanding JCOs in a separate !

book and serialized each JCO. However, several of the JCO's in the I control room were no longer in effect. Two JCOs which were listed as in

l

_ _

_ _ - _ - . ._

,

,

\

.~

.. )

l 9 1 effect limit standby nuclear service water pond temperature to 68 degrees F due to degraded NS heat exchanger performance. These JCOs should have been removed based on heat exchanger cleaning and the JC0 discussed earlier in this section. In addition, at least one JC0 that was in effect was not in the control room. The current "program" in place to administer the writing controlling, and tracking of JCOs is deficient and is indica-tive of a programatic problem. The inspectors have observed that plant personnel including operations appear to be unaware of what JCOs are in effect. The licensees method of writing, controlling and tracking JCOs is considered a weaknes . Exit Interview (30703)

The inspection findings identified below were summarized on August 24, 1988, and again on September 16, 1988 (once the NRC review of the licensee's reanalysis of VX system flows was completed) with those persons indicated in paragraph 1 above. The following items were discussed in detail:

(0 pen) Violation 370/88-24-01, Both Trains of the Unit 2 Hydrogen Skimmer System Inoperable (see paragraph 6).

(0 pen) Violation 369, 370/88-24-02, Failure to Perform a 10 CFR 50.59 Safety Evaluation (see paragraph 6).

(0 pen) Violation 369,370/88-24-03, Inadequate Preoperational Testing of the Hydrogen Skimmer System (see paragraph 7).

(0 pen) Unresolved Item 369/88-24-04, As found Condition of Unit 1 Hydrogen Skimmer System (see paragraph 6).

A weakness was observed with regard to the licensees method of writing, controlling, and tracking JCOs (see paragraph 7).

The licensee representatives present offered no dissenting comments, nor did they identify as proprietary any of the information reviewed by the inspectors during the course of their inspection.

,

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___ _ . . _ _ _ _ _ _ _ _