ML20151T627

Enforcement Conference Rept 50-423/88-04 on 880720.Major Areas Discussed:Significance of Concerns Re Environ Qualification of Certain Cable Splices & Containment high-range Radiations Monitors

ML20151T627
Person / Time
Site:	Millstone
Issue date:	07/27/1988
From:	Mccabe E NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20151T618	List: ML20151T615 ML20151T627
References
50-423-88-04-EC, 50-423-88-4-EC, NUDOCS 8808160424
Download: ML20151T627 (39)

See also: IR 05000423/1988004

Text

. ._ .. ,

!

'

. ..

.

.

'

U.S. NUCLEAR REGULATORY COMMISSION

REGION I

' ENFORCEMENT CONFERENCE REPORT

. Report No. 50-423/88-04

Docket No. 50-423

License No. NPF 49

Licensee: Northeast Nuclear Energy Company

Facility: M111 stone 3, Waterford, Connecticut

'

'

Confarence At: NRC Region I, King of Prussia, Pennsylvania

Conference Date: July 20, 1988

Approved by- s M/m8

E. C. McCabe, Chief, Reactor Projects Section 18

7b/88

Date

Conference Summary:

Environmental Qualification (EQ) enforcement conference held to discuss signific-

ance of concerns about EQ of certain cable splices, Litton-Veam connectors, and

containment high range radiation monitors. The licensee provided their evaluation

of the regulatory and safety significance of these matters (Appendix A to this

, report). The NRC will evaluate the licensee input and separately issue appropriate

enforcement action.

!

,

?

A

8808160424 880809

PDR ADOCK 05000423

'

O PNU ..

J

,

. - -

,

o

-

.

..

.

DETAILS

1. Conference Attendees

Northeast Nuclear Energy Company

.

E. Mroczka, Senior Vice President, Nuclear Engineering and Operations

R. Werner, Vice President, Generation Engineering and Construction

P. Blastoli, Supervisor,-Licensing

M. Gentry, Millstone 3 Operations Supervisor

R. Peterson, Senior Engineer, Environmental Qualification

A. Roby, System Manager, Nuclear Engineering and Operations

B. Tuthill, Qualification Engineering

NRC Region I

W. Johnston, Acting Director, Division of Reactor Safety (DRS)

C. Anderson, Chief, Plant Systems Section (PSS), DRS

~

J. Durr, Chief, Engineering Branch, DRS

.

J. Gutierrez, Regional Counsel

R. Mathew, Reactor Engineer, PSS, DRS

E. McCabe, Chief, Reactor Projects Section 1B, Division of. Reactor Projects

R. Paolino, Senior Reactor Engineer, PSS, DRS

2. Conference Scope

r

The Director, Division of Reactor Safety outlined the enforcement conference

considerations as being the following.

--

Environmental qualification (EQ) issues identified during EQ inspection .

"

earlier this year.

.

--

Enforcement criteria are the general criteria applicable to a plant in

the operating license issue process, not the enforcement criteria for ,

plants in operation at the time the NRC environmental qualification rule ;

was issued.

--

The specific concerns are Litton-Veam connectors not potted as required,

the qualification status of Litton-Veam connectors at the time of the

EQ inspection, and high radiation monitor cable qualification under high

temperature conditions.

3. Licensee Presentation

The licensee presented their position on the EQ issues of concern, as identi- t

, fled in the appended presentation outline. '

2

I

3

, -

-

.

.

3

.

4. Raychem Splices Discussion

The licensee stated that their response to the already issued Raychem splice

violation will include justification of their position on the acceptability

of as-installed splice bends.

5. Litton-Veam Connectors ,

The NRC asked whether the Litton-Veam connectors were used to initiate a

safety function or affected the ability to achieve cold shutdown. The lic-

ensee responded that the Litton-Veam connectors are used to initiate safety

. functions but that their review had found that instrument redundancy, instru-

ment diversity (e.g., use of Veritrac instruments in some channels), and the

early time of required instrument actuatinn (in all cases except a slow loss

of feedwater level) all showed that the required safety functions would be i

performed. Also, the loss of the affected indications had been evaluated l

through covering the associated indicators on the plant simulator. In that .

condition, the operators achieved safety shutdown despite the loss of the L

indications. The licensee identified the only safety concern being the un-

likely simultaneous failure of all the affected instruments.

The NRC asked, for the Litton-Veam connectors installed in a configuration .

for which they were not qualified, whether the two valves of concern were

containment isolation valves (CIVs) and if so, were they in the same system.

The licensee responded that the two valves were inside containment CIVs in

different systems, with the outside containment CIVs valves available to

effect isolation not being affected by this problem.

6. Cont.ainment High Range Radiation Monitor (HRRM) Discussion

There was extended discussion of the licensee conclusion that the containment

..' HRRMs met NRC Regulatory Guide 1.97 accuracy criterion even with the identi-

fled cabling problem. The licensee also affirmed that the HRRM post accident i

function was not impeded by this problem. The licensee stated that the mag-

nitude of the error was greater at Millstone 1 than at Millstont 3 and that

an exemption from the Regulatory Guide 1.97 accuracy criteria may be requested.

(Functional capability is not suspect.)

The NRC questioned the timeliness of licensee resolution of the HRRM cabling

when the problem was identified. The licensee stated that their chronology

(Appendix A to this report) showed the timing of their resolution. Resolution

timeliness will receive further NRC evaluation.

The licensee committed to submit the documentation justifying HRRM operability

to the NRC. [

,

)

.

e

.

,

.

ENFORCEMENT CONFERENCE

50-423/88-04 *

APPENDIX A

Northeast Nuclear Energy Company

Millstone Unit No. 3

Docket 50 423

Equipment Qualification

July 20, 1988

. -

+

'

1

-

Meeting Agenda l

'o ,

.

July 20, 1988

Northeast Nuclear Energy Company '

.

EQ Enforcement Conference

for

Millstape Unit No. 3 .

.

I. Introductory Comments - NRC Staff

II. Introduction - E. J. Mroczka

III. Background and Summary of Findings - R. P. Werner

IV. Discussion of Potential Violation and Unresolved items

A. Potential Violations

1. Raychem Splices (50-423/88 04-06) -

B. A. Tuthill

2. Litton-Veam Connectors (IR Concerns) -

A. R. Roby

(50 423/88 04-05)

3. Litton-Veam Connectors (Configuration -

A. R. Roby

Concerns) (50 423/88 04-07)

4. GA High Range Radiation Monitors -

R. S. Peterson

(50-423/88-04-02)

B. Unresolved items

1. GA High Range P.adiation Monitor -

R. S. Peterson

(Potential for a 1.7 rad / hour error)

(50-423/88 04-03)

2. JC0 for Potential EQ Deficiencies -

P. A. Blasioli

(50 423/88-04-01)

3. Target Rock S0Vs (50 423/88 04-04) -

B. A. Tuthill

V. Conclusion - R. P. Werner

._______

.

-

.

'

. .

,

. NORTHEAST NUCLEAR ENERGY COMPANY

N!LLSTONE UNIT NO. 3

DOCKET 50 423 ,

.

EQUIPMENT QUALIFICATION

JULY 20, 1988

Introduction

Northeast Nuclear Energy Company (NNECO) received a full power operating

license for Millstone Unit No. 3 from the NRC on 'anuary 31, 1986. At that

time NNECO had implemented a comprehensive program to address the

environme. ital qualification of electrical equipment (EQ). Both NNECO and the

NRC determined that there was reasonable assurance that the plant w's in

compliance with the requirements of 10 CFR 50.49.

An NRC inspection tecm conducted an inspection at the plant on March 14-18,

1988, to verify NNECO's implementation of the EQ program. In Inspection

Report No. 50 423/88 04 dated July 6, 1988 (EA-88 177), the NRC identified one

Severity Level IV violation, four potential violations, and three unresobed

items with respect to 10 CFR 50.49. In this presentation NNECO addresses

thbe items. In addition, NNECO will respond separately to the Notice of

Violation by August 8, 1980 in accordance with 10 CFR 2.201.

Because Millstone Unit No. 3 received its full-power operating license after

November 30, 1985, the NPC's modified EQ enforcement policy of Generic

Letter 88 07 is inapplicable to the items identified to the NRC's ,

s

[m--

~

l

,' 2-

'

.

., _

,

inspection report. Accordingly, the potential violations noted in the

Inspection Report are addressed below in accordance with the considerations of

F

the NRC's general enforcement policy (10 CFR Part 2, Appendix C).

In summary, NNECO admits the proposed' violation regarding the installation of

1

Raychem splices over braided jacketing. NNEC0 denies the potential violation

with respect to bend radius and seal length of Raychem splices. Further, even

if such deficiencies are found, NNEC0 maintains that testing in March 1987

reflects they would have no safety significance. NNECO admits in part the

potential violations regarding Litton-Veam connectors, but maintains they also

have minimal safety significance given NNECO's demonstration of present

~

qualification, operabi1ity and corrective actions. Finally, NNECO denies the

potential violation regarding General Atomics (GA) High-Range Radiation

Monitors (HRRM). This equipment was qualified when installed. NNECO has

since evaluated vendor information to determine whether qualification status

is impacted, and concluded that 1) qualification has not been undermined and

2) th(. identified concerns have no safety significance. In all these cases,

NNECO has taken prompt measures to address the underlying concerns and to

ensure that qualification is maintained or promptly established.

The E0 Proaram

! NNECO, as the licensee for Millstone Unit No. 3, has implemented an effective

program for the qualification of electrical equipment. NNECO has and

continues to recognize the importance of the EQ program to safe operation of

its nuclear units, and its management is committed to completely addressing

f any EQ deficiencies or unresolved items at Millstone linit No. 3.

l

l

l

l

l

l

.

-3

.

.

'

.

.

NNEC0's EQ program for Millstone Unit No. 3 has been developed and implemented

.

to assure that appropriate electrical equipment will remain functional in a

.

harsh environment during and following a design basis event. As. a result of

the program, equipment should also remain qualified and be maintained in a

qualified condition throughout the life of the plant.

i

NNECO Nuclear Engineering and Operations Procedure Number NEO 2.21, "Nuclear

Plant Environmental Qualification Program," establishes the Millstone Unit

No. 3 EQ program. The program addresses, among other things, the scope of

equipment requiring qualification, the actions to be taken in the event any

equipment becomes unqualified, and the maintenance of the qualified condition

of the equipment. Auditable records of the selection, qualification, and

.

maintenance of the equipment are also maintained. Finally, the program

includes the requirements for EQ training, control of purchases of qualified

equipment, and control of design changes to assure proper EQ consideration.

Overall responsibility for the EQ orogram lies with the Senior Vice President,

Nuclear Engineering and Operations. Specific responsibility 'for overall

coordination, development, and implementation of the program rests with the

Vice President, Generation Engineering and Construction. These individuals

are knowledgeable with respect to the program and are committed to its

continued success.

In the NRC's July 6, 1988 inspection report, the Staff specifically found that

NNECO has implemented its program in compliance with 10 CFR 50.49. The NRC

which classified as Severity Level IV.

identified one violation, it

.

_ _ _ ..

'

.

,

-4-

'., -

In addition, the Staff identified four potential violations and three

.

unresolved items. In terms of both the number of potential deficiencies and

the low safety significance of those items (as discussed below). NNECO

believes that the EQ Program at Millstone Unit No. 3 has been eTfective and

compares favorably to EQ programs throughout the industry. The deficiencies

identified are isolated in nature and do not reflect any programmatic

weaknesses.

The following discussions of each enforcement item separately address the

questions raised by the NRC in the July 6, 1988 inspection report.

.

b

6

1

e

a

,

w

.

.,

,

.

5-

'

.

.,

, ISSUE N0. 1

(RAYCHEMSPLICES)

.

.

A. STATEMENT OF THE ISSUE

<

NRC Inspection Report Issue 50 423/88-04 06 states, "A review of licensee

walkdown records for Raychem splices indicate (s) additional nonconforming

conditions involving excessive splice bends (less than the recommended

5 X cable outside diameter) and splice seal length areas of less than

2 inches." According to the Staff, these deficiencies constitute a

potential violation of 10 CFR 50.49 in that the splices were not

qualified per the instruction procedures in effect at the time of the

installation. In addition, the Staff has issued a Notice of Violation

proposing a Severity Level IV violation for Raychem splices installed

over braided cable jacketing, contrary to the Raychem installation

procedures.

B. NNECO POSITION REGARDING THE EXISTENCE OF A VIOLATION

NNECO admits the proposed violation regarding installation of Raychem

splices over braided cable jacketing and will provide its written

response in accordance with 10 CFR 2.201.

NNECO does not agree that the walkdown records indicate that the Raychem

installations had splice lengths less than two inches.

.

'

, 6-

.

.

.,

NNECO also does not agree with the Staff's conclusions that installation

.

of Raychem splices with a bend radius less than 5 x the cable outside

diameter was contradictory to the Raychem procedure / instructions existing

'

at that time.

NNECO followed existing procedures with regard to these installation at

the time the plant went into operation,

in the event the Staff concludes that its assertions are valid, NNECO

maintains that the alleged deficiencies should be treated as further

examples of the deficiency already addressed in the Severity Level IV

Notice of Violation (NOV) issued in Inspection Report 50-423/88 04. A

separate violation on these related issues would not be warranted.

Further, should the Staff decide that a separate NOV is appropriate

regarding the potential violation based on the information provid 1

below, NNECO believes that the issue has no safety significance. The

lack of safety significance of this issue and other relevant factors are

discussed below.

C. BASIS FOR DENYING THAT A DEFICIENCY EXISTED

Bend Radius

The original installation of the splices was made in accordance with

Millstone 3 Electrical Installation Specification No. E-350. This

.

.

7

-

.

'

-

.

procedure stated that Raychem installation instructions shall be in

accordance with Raychem's "Product Installation and Inspection Guide -

WCSF-N Heavy Wall, Flar.e Retarded Nuclear Cable Sleeves" '. PII-57100-A

~

and installation instructions provided with each kit. E-350 also states

that QC shall verify (in accordance with specifications) that proper

Raychem components are used, manufacturer's installation instruction have

been followed, and an acceptable splice has been made.

During the time at which Raychem froduct Guide PII-57100-A

(November 1982) was in effect, Raychem did not provide bend radius

acceptance criteria. Accordingly, contrary to the inspection report

findings, the Raychem installations were made in accordance with

procedures in existence at this time. Upon issuance of NRC Information

Notice 86 53 (June 26,1986), NNECO took reasonable and prudent actions

to address the issue. Bend radius criteria were first published in

Pil 57100 0 (November 1987).

Seal Lenath

Contrary to the NRC inspection report, NNECO's review of walkdown records

does not reveal any instances where RUchem splices with less than 2 inch

seal lengths exist. QC witnessed the installation of some of the splices

and did not note any such seal length deficiencies. NNECO has no other

indications that such seal length deficiencies exist. Therefore, NNECO

denies that the alleged seal length deficiency existed. It should be

noted that after issuance of Information Notice 86-53, where this issue

was raised, NNECO took reasonable and prudent actions to address it.

-

-

.

-

.

• -

.. -8-

.

D. SAFETY SIGNIFICANCE FINDINGS

These issues have no safety significance because:

1. March 1987 testing has established that application over braid, a

bend radius much less than 5 x the outside diameter, and seal

lengths much less than 2 inches are qualified configurations.

Accordingly, no replacements or modifications of the components were

required to establish or maintain qualification.

2. Similar enforcement actions at Ginna (J1ne 10,1987), Haddam Neck

(April 1, 1988), Grand Gul f (March 25, 1988). and North Anna

(April 19, 1988) did not have significant safety implications in

that the enforcement actions were categorized at a Severity

Level IV.

E. OTHER FACTORS THAT SHOULD BE CONSIDERED

1. Notwithstanding the final determination regarding the existence of a

violation, craft personnel have received additional training to

emphasize the importance of following procedures during component

installations.

2. In addition to the above, specific Raychem training war provided by

Raychem personnel to ensure that installations were correct.

!L

!;

.

.. _ _ - __ _. ~ _ __

,

'

.

,

.

- 9-

,

• 3. The EQ file was revised prior to the inspection to include test data

,

(dated March 1987) that confirms the qualifiention of Raychem

splices with over braided cable jacket, splices less than a 2-inch

seal -length and less than 5X outside diameter splice bends. The

Staff has concurred with the acceptability of the March 1987

report to further establish qualification. (itt Inspection

! Report 50423/8804,atp.21.)

1

)

!

.

.

-

1

4

I

(

_ - ~ . . . . . . , _ _ , --,__-.-__y .,---_,m --w. , -

-._,

,

.

,

.

-

- 10 -

..

• ISSUE N0. 2

LITTON-VEAM

(INSULATIONRESISTANCE(IR) MEASUREMENTS)

A. STATEMENT OF THE ISSUE

NRC Inspection Report Issue 50-423/88 04-05 states that "[t]he inspector

was concerned that low IR values during LOCA test could be the result of

moisture penetration into the seal. Due to a lack of IR measurements

during the LOCA transient conditions, the worst-case IR measurements and

hence leakage current could not be established, thus making it impossible

to detemine the contribution of the connectors to the instrument loop

accuracy under a LOCA envirnment . . . . Thus, for the Litton Ver.m

multipin connector assemblies installed on instruments located inside

containment, the licensee did not adequately establish the equipment

performance standards in the test acceptance criteria and did not provide

adequate evidence that this equipment we.uld perform its intended function

for all service conditions postulated to occur during the qualified

life."

8. NNECO'S POSITION REGARDING THE EXISTENCE OF A VIOLATION

NNECO agrees that IR measurements were not tahen during LOCA transient

extremes and that inappropriate performance standards were used during

the evaluation of Litton-Veam qualification test results. This

deficiency resulted in insufficient evidence (at the time of the NRC

inspection) to fully support an EQ file conclusion that certain

,

.

.

. .

-

- 11 -

.

Litton-Veam (L-V) connectors were qualified for a 40-year life. NNECO

does not contest this potential violation.

.

.

As discussed further below, NNECO believes that because the IR concerns

do not apply to un-aged connectors, these connectors were in fact

qualified both for their present installed life and for the near term.

Moreover, because the associated components would have performed their

intended function, there was no resultant safety significance.

C. . SAFETY SIGNIFICANCE

This issue is not safety significant because:

1. At a March 21, 1988 meeting, NNECO and the Staff agreed that IR

degradation did not unacceptably affect un-aged connector

performance requirements. Further, until significant aging of the

connector has occurred, NNECO and the Staff agreed that there is

reasonable assurance that existing equipment will perform its

intended function if called upon to operate in a harsh environment,

in that Millstone Unit No. 3 did not receive its full-power

operating license until January 1986, the installed connectors have

to date not been sufficiently aged so as to affect performance

'

requirements for accuracy to a degree that renders them unqualified.

NNECO acknowledges that a definitive long-term qualified life cannot

be determined with current test results. However, NNECO also

4

believes that it is reasonable to conclude that the second refueling

outage is not beyond qualified life. .(Ita Item 2 below). This will

.

.

.

',_ -

- 12 -

l

!

take place when the connectors have experienced 36 months of service

life. At that time the connectors will not have aged more than

~

1.6 % of end of-life condition calculated at normal maximum average

service temperature.

2. The Staff has previously concurred with NNECO's decision to replace

the connectors during the refueling outage currently scheduled for

June 1989. This supports NNECO's conclusion that the connectors are

still qualified and functional, and that there is no safety

significance in using the connectors until replacement.

3. A similar enforcement action at Sequoyah (June 17,1987) resulted in

the aggregation of eight items as a Severity level IV, one of which

was failure to meet loop accuracy requirements for Eaton

multiconductors signal cables, reflecting no significant safety

implications.

D. ROOT CAUSE

The root cause for the L-V testing deficiency is an isolated failure to

sufficiently incorporate IR measurements in the testing program and

adequately consider the effect of IR changes in determining qualification

and qualified life.

E. CORRECTIVE ACTIONS / ACTIONS TO PREVENT RECURRENCE

1. NNECO will take extensive corrective action by replacing the

affected Litton-Veam connectors (i.e., 43) with qualified

{

,

.

'

-

..

13

.

feed-through devices during the June 1989 refueling outage. This is

a conservative action in that we believe that additional testing

?

would likely have justified an extended qualified life. ,

2. NNECO believes that continuing training addressing the evolving

level of knowledge regarding the use of IR data will ensure that

appropriate IR requirements are considered during future testing.

F. OTHER FACTORS THAT SH0VLD BE CONSIDERED

1. As previously stated, NNECO has conservatively agreed to replace

co r'. a i n affected connectors at the next refueling outage, even

though we do not believe that their qualified life will be exhausted

at that time.

2. This deficiency was isolated in that a single root cause affected

only one type of component. The deficiency identified related only

to connectors used in instrument loop applications subject to a LOCA

environment.

3. NNECO did establish acceptance criteria prior to testing, but later

determined that the criteria were not anplicable to the intended use

of the connectors in instrument loop c'rcuits.

4. The issue of loop accuracy is an evolving issue which has only

recently been fully addressed by either the Staff or industry.

.

,

'

e

'

- 14

. .

-

ISSUE N0. 3

LITTON-VEAN(CONFIGURATION)

.

P

A. STATEMENT OF THE ISSUE >

HRC Inspection Report issue 50-286/88-04-07 states that "(djuring the

test plan review (NTS QPP No. 5581657, Revision 1), the inspector

observed that the test specimens were prepared following an assembly

procedure, VAP 241. This procedure required that heat shrink tubing be

applied over the contact wire crimped termination before potting the rear

1

of the connector with epoxy. The licensee's Electrical Installation

Specification No. E-350, which was in place from the initial commencement

of installation of the connectors on March 6,1984, until its cancella-

tion on March 26, 1984, did not incorporate the heat shrink tubing over

the pin / conductor interface . . . . Of the 21 connectors identified

(without heat shrink tubing) only two cases were found that were required

to remain functional to mitigate a LOCA/HELB event. These two cases

involved position indication for two containment isolation valves."

8. NNECO'S POSITION REGARDING THE EXISTENCE OF A VIOLATION

NNECO agrees that 21 titton-Yeam connectors were found in a configuration

for which qualification had not been established. As further discussed

below, the safety significance of this finding is minimal.

- - . ._.

.

,

. - 15 -

'.. '

C. SAFETY SIGNIFICANCE

.

This issue has minimized safety significance because:

1. Analysis reflects that virtually all components associated with

these connectors were qualified for their environments and, as a

practical matter, some could likely be removed from the EQ Master

List. Discussed further below are specific safety significance

findings regarding the 21 connectors. At bottom, only two cases

(four connectors) were found where the equipment (see item e below)

would have been required to remain operable to mitigate the event

causing the local harsh environment. For these two cases, the

failure would not have impacted operation of the associated valves,

but only valve position indication.

a. Several connectors and associated limit switches provide (a)

isolation of component cooling water (non-safety related return

sub-header on a containment isolation Phase A signal (CIA) on a

loss of power signal); (b) isolation of redundant reactor plant

component cooling water flow (to the Train A non safety related

return sub-header on a CIA and/or loss of power signal); and

(c) isolation of Train A/B reactor plant (non-safety related

supply sub-header on a CIA signal and/or a LOP signal). The

connectors are not located inside containment and are not in a

moisture environment. A one-time HELB is postulated for these

connectors. However, the Millstone Unit No. 3 design basis

does not analyze a one-time HELB in the area of the connectors

concurrent with an inside containment HELB. Therefore, the

,

.

-

16 -

• *

..

, connectors would navs remained functional throughout an inside

containment design basis accident or would not have been

required to mitigate an accident while in a moisture

environment.

,

b. Several connectors (located outside containment) involve the

operation of the charging pump cooling pumps. These pumps are

required whenever the charging pumps must operate. A one time

HELB is postulated for the affected connectors. However, the

HELB is net postulated coincident with an inside containment

LOCA or main steam line break (MSLB). Therefore, either the

connector will not be in a moisture environment when required

to respond to an inside containment design basis accident (for

a LOCA and MSLB) or will not be required to mitigate an

accident while in a moisture environment.

c. Certain connectors in conjunction with level switches monitor

reactor plant component cooling water surge tank level, split

,

flow to the reactor coolant pumps and shut the supply / return

valves to the containment air recirculation cooling coils and

neutron shield tank coolers on low surge tank level. The

connectors / switches are located outside containment in a

i

non moisture environment. They are postulated to experience a '

one-time HELB. In that tne HELB is not postulated to occur

concurrent with a requirement to isolate containment (due to a

LOCA), the ability of the connectors / switches to perform

intended safety functions has not been impaired. In addition,

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

.

-

.

'.. -

37

these level indication switches provide a control function

only. Failure does not provide misleading information and,

according to E0Ps, would have resulted in t,he operator

utilizing alternate level indication.

d. Certain connectors in conjunction with pressure switches l

isolate Trains A/B service water system supply to the turbine

plant component cooling water (TPCCW) system (closes TPCCW heat

exchangers tube side inlet isolation valves on low Train A/B

service water header pressure on a Train A/B containment

depressurization activation and/or loss of power signal). The

connectors are not located in a moisture environment (outside

containment) and are postulated to be exposed to a one-time

HELB. However, the one-time HELB is not postulated concurrent

with an inside containment design basis accident. Therefore,

as with several other connectors previously discussed, the

connectors will either be able to perform intended functions or

will fail without being required to mitigate a design basis

accident.

l e. Four connectors are associated with providing nitrogen charge

I

to the pressurizer relief tank, and maintaining a supply of

l

l

primary grade water to the pressurizer relief tank, the reactor

coolant pump level standpipes and the neutron shield tank

cooling system. These connectors are located inside

containment and reasonable assurance that these connectors

.

18 -

,

.

[

would not fail in the event of a LOCA or MSLB was not provided

.

by the qualification file. However, it should be noted that

the associated limit switches impacted by such a failure

provide only valve position indication. The functioning of

associated valves would not be affected should the connectors

fail.

2. In a NOV dated November 19, 1987 agiinst the Maine Yankee Atomic

Power Company (MY), the NRC cited MY for failing to provide a

similarity analysis between the tested specimen and installed

connectors (configuration). This NOV was categorized.as a Severity

Level IV. We believe that the finding at Millstone Unit No. 3 is

similar to the NY NOV in that a similarity analysis should have been

provided to address connectors without heat shrink tubing. We also

believe that the Millstone Unit No. 3 finding is certainly no more

safety significant than the MY enforcement action.

D. ROOT CAUSES

1. An engineering design / Stone & Webster interface failure to ensure

that a design intent was included in an installation procedure.

2. An engineering design failure to include in the test program

adequate requirements to demonstrate operability of connectors s

without heat shrink tubing being applied and subsequent

misapplication of the test report,

,

r

I

.

.

-

,

19 -

.

E. CORRECTIVE ACTIONS / ACTIONS TO PREVENT RECURRENCE i

1. Promptly after it was determined that certain connector

configurations had not been fully analyzed, NNECO performed an

operability evaluation, modified administrative controls, and

"caution tagged" the two affected valves. Position indication for

the two valves is assured by tagging them closed except when needed

to be opened. This provides continuous administrative verification

of valve position.

2. After issuance of a stop work order on March 26, 1984, to terminate

installation without heat shrink tubing, the installation procedure

was promptly revised (by April 5,1984). Prior to implementing the

procedure to require the correct installation configuration, craft

and engineering personnel were retrained.

3. NNECO met on March 27, 1984 with Stone and Webster to discuss

requirements for heat shrink tubing. After significant discussion,

a revised procedure was initiated to require the use of heat shrink

tubing. Additionally, it was decided that connectors without heat

shrink tubing would be included in an upcoming qualification test

program to demonstrate connector qualification without heat shrink

tubing. However, in retrospect, this testing did not resolve the

issue of concern,

4. From that time forward, all subsequent Category I engineering and

design change requests (E&DCRs) were required to be reviewed by E0

personnel prior to implementation. The plant was also walked-down

r , 20 -

. ,

.

100% to verify visually EQ attributes to ensure that they were in

,

accordance with engineering design requirements,

e

'

5. NNECO believes that continuing training regarding the 'use of test  :

data will ensure that appropriate parameters are included in future

<< testing and test results evaluation.

6. The four connectors used for containment valve position indication

will be replaced with new connectors with heat shrink tubing during

the next refueling outage.

F. OTHER ISSUES THAT SHOULD BE CONSIDERED

1. The deficiencies were of minimal or no safety significance.

2. Once the deficiency was discovered, NNECO took prompt actions by

analyzing operational considerations and proposing corrective

actions.

-

.

-

- 21 -

.

. .

,

. '

ISSUE NO. 4

GENERALATOMICS(GA)HRRM

.

.

A. STATEMENT OF THE ISSVE

NRC Inspection Report Issue 50-423/88 04 02 concerns GA HRRMs installed

in containment prior to the licensing of Millstone Unit No. 3. The

Inspection Repcrt states that "there are indications that the GA HRRMs

were qualified when installed and prior to licensing of the Millstone

Unit No. 3 facility." However, ir light of information from the vendor

(dated November 6, 1986) indicating IR problems with the HRRM cable

assembly, the inspection Report states that "it appears that the licensee

was still in the process of evaluating the qualification . . . at the

time of this inspection. The licensee apparently has not followed the

guidelines of the NRC Generic Letter 86-15. The licensee has assumed

that the GA HRRM is operable until the vender information is confirmed."

8. NNECO'S POSITION REGARDING THE EXISTENCE OF A VIOLATION

NNECO's position is that this item does not presently represent a viola-

tion of 10 CFR 50.49. NNECO had adequate documentation in the GA HRRM

file to qualify the components at the time of their installation. NNECO

subsequently analyzed additional vendor information and concluded that

the equipment is still qualified. To the extent that there are still any

open issues regarding these analyses, this item should be more properly

classified as an unresolved item rather than as a potential violatten.

,

- 22 -

. Should the Staff maintain that a violation exists, NNECO believes that

the deficiency has no safety significance for Millstone Unit No. 3.

,

'

C. FACTORS SUPPORTING NNECO'S POSITION

As previously stated, the two GA HRRMs were considered qualified at the

time they were installed. The inspector reviewed NUSCO engineering

assurance audit Finding No. 3-10, contained in Memorandum No., 1103866

(dated July 24,1987), which incorrectly suggested that the GA HRRMs were

being qualified while in use. However, this finding wo addressed in

NUSCO memo GSP 87 387, B.A. Tuthill to B. S. Kaufman (October 29,1987).

NNEC0 has taken and continues to take the position that the GA HRRMs

installed at Millstone Unit No. 3 were qualified. This conclusion was

restated in NNEC0's letter to the NRC, B12671 E. J. Mroczka to

U.S. Nuclear Regulatory Commission, dated September 30, 1987.

The current issue addresses IR uncertainties addressed in post startup

information supplied by the vendor, Sorrento Electronics (Sorrento).

Specifically, the cable used in connecting the detector output to the

containment penetrations (RockbestosRSS6104) exhibited undesirable

characteristics at LOCA temperatures. At these temperatures, the elec-

trical IR decreases, increasing cable leakage currents which in turn

would reduce indicated radiation levels. The issue was first raised in

Northeast Utilities letter GSP 86 570 to Sorrento, J. S. Nicosia/

W. J. Hayes to Zvi Rendel (dated October 14, 1986). In that letter,

NNECO identified IR concerns and sought justification for the factor of 2

accuracy of the HRRM system. Sorrento responded on November 6, 1986,

i

_ _ _ _ _ _ _ _

,- ,

- 23 -

'

.

• This information adequately addressed the issue for Hillstone Unit No. 3

and confirmed qualification of the GA system.

,

Subsequently, in February 1987, Sorrento issued a Part 21 notification of

cable IR deficiencies. Following notification of this potential problem,

NNECO implemented a Sorrento calculational methodology to evaluate the

-

Rockbestos coaxial cables used in the GA HRRMs for Millstone Unit No. 3.

(A detailed chronology of NNECO's efforts in this regard is attached.)

Using that methodology, the calculated overall loop accuracy equated to a

maximum error of <1 R/hr. Thus, the factor' of accuracy for this equip-

ment specified in Regulatory Guide 1.97 is met.

With respect to the issues raised in the July 6,1988 inspection report

(i.e., compliance with 10 CFR 50.49), NNECO considers that the HRRMs are

qualified. To the extent that there are open questions regarding this

conclusion, this item should be regarded as an unresolved issue rather

than a potential violation.

NNECO has followed procedures analogous to those outlined in the Generic

Letter 8615 for addressing the potential EQ deficiency. Specifically,

as indicated by the chronology of events, NNECO has in a timely fashion

addressed the new information and concluded that the equipment is in fact

qualified. Pending this determination, NNECO prepared and maintained a

JC0 only as a conservative approach to be utilized if continued qualifi-

cation could not be shown. However, this JC0 was never implemented.

In this situation, NNECO properly undertook an evaluation to assess the

impact on qualification status and to assess reportability. NNECO

-

.

• Y -

24

.

maintains that it was not required in this instance, upon receipt of

vendor information, to assume that the HRRMs were not qualified (given

that they were already considered qualified). NRC regulations in 10 CFR

Section 50.72 and 50.73 inherently recognize the time allowed for this

evaluation and do not call for interim reports to the NRC of "potentially

reportable" conditions. NNECO ultimately completed its reportability

evaluation and concluded that this event was not reportable under 10 CFR

50.72 or 50.73.

D. SAFETY SIGNIFICANCE

NNECO has concluded that this item has no safety significance. The GA

HRRMs are located inside containment. They are used to detect post-

accident radiation levels between 1 R/hr and 10 million R/hr. Any

unsatisfactory IR at elevated temperatures could lead to a reduction in

the indicated radiation levels at low ranges under certain conditions.

However, NNEC0 has concluded for the following reasons that the HRRMs

remain operable and would not mislead the reactor operator.

1. The principal function of these monitors is to assist in making

emergency assessment determinations, in this regard, NNECO has

performed a conservative evaluation of the HRRM readings under ,

maximum postulated containment temperatures, taking into account

cable and penetration current leakages. The Hillstone Unit No. 3

HRRMs meet Regulatory Guide 1.97 guidelines, even with maximum

potential biases due to reduced cable insulation resistance follow-

ing a LOCA or MSLB.

.

s

o

,

- 25 -

. 2. At the time when assessment of plant conditions would be made in

accordance with established (plant-specific emergency action levels)

guidance, either containment temperatures are not predicted to be

elevated (and, thus, these biases do not exist) or indications

independent of the HRRMs (i.e., elevated temperature, reactor

coolant system level, and leak rate indications) provide appropriate

information to permit valid assessment of plant conditions. Most

importantly, determinations concerning public protective measures

based even in part on containment radiation levels occur at such

high containment radiation levels that the potential biases would be

undetectable.

3. In addition, alternative means are available to determine fuel

conditions and containment curie inventories. The primary alterna-

tive method involves radiation monitors located outside the contain-

ment sensing through penetrations. Also, post-accident sampling

system samples, process monitors outside containment, and thermal

hydraulic parameters provide information that permits assessment of

fuel conditions and/or curie inventory.

4. Finally, although they provide no automatic control function, the

monitors are used at Millstone Unit No. 3 in making certain dect-

sions under the emergency operating procedures. For Millstone Unit

No. 3 there is a decision point at 1 x 105 R/hr for assessing envi-

ronmentally qualified electrical equipment instrument accuracy.

NNECO has concluded that the HRRM biases in the ranges of those

<

t

,

\

'

oj .

- 26 -

decision points are negligible and, thus, the HRRMs fulfill these

functions.

.

In addition, recent enforcement history in a similar case indicates that

this issue has low safety significance. In similar circumstances, the

NRC has classified a violation regarding GA HRRM qualification deficien-

cies as a Severity Level IV. In an enforcement action involving

Tennessee Valley Authority's Sequoyah Units I and 2 (June 17,1987), the

Staff found that the licensee had not adequately established qualifica-

tion for GA Technologies RD 23 radiation detectors and the associated

Brand Rex coaxial cable. This was one of eight qualification deficien-

cies collectively categorized as a Severity Level IV violation. NNECO

concludes that the circumstances in the present case are certainly no

more significant, and probably less significant given that there is less

equipment involved.

E. OTHER CONSIDERATIONS

Inspection Report 50 423/88 04 also includes two unresolved items that

relate to the HRRMs. NNECO believes that these items can be resolved.

1. . Item No. 88 04 03. This item concerns the potential for an error of

about 1.7 rad /hr. over the entire range for the GA HRRMs. The NRC

inspector observed tnat this error "could lead to the detector going

off-scale at the low end and the green ' operate' light going off.

This could result in subsequent high radiation readings which could

confuse the operator even though the readings were accurate."

i

I

-

, . _ _ _ . _ -

.

%

'

.[ .

27 -

NNECO committed to revising the operations procedure number OP 3362,

"Radiation Monitor System Display and Control System," to address

this concern. The revision was completed May 10,198'8 . In addi-

tion, NNECO comitted to incorporate this HRRM alarm response into

the Control Room Operator Requalification Training Program and to

notify control room operators of the potential problems. These

training activities have been accomplished. Therefore, this unre-

solved item should be closed.

2. Item No. 88 04-01. This item concerns NNECO's procedure NEO 2.25,

"Identification and Implementation of NRC Reporting Requirements."

and the need to "proceduralize the licensee's JC0 process." The

Staff was concerned that NEO 2.25 did not contain an express provi-

sion requiring preparation of a JC0 when a "potential EQ deficiency"

is identified. At the exit meeting NNECO stated that it would

consider adding directions in the procedure NEO 2.25 to address the

need for JCOs for EQ deficiencies.

NNECO does have controls in place to assure adequate evaluation and

timely corrective action of identified nonconformances to include

equipment governed by technical specifications that has been deter-

mined to be inoperable for any reason.

NNECO notes that, well prior to the NRC inspection, a program to

strengthen procedural controls regarding EQ was begun. As part of

this program, a 1987 internal audit suggested that NNECO's proce-

dures be clarified regarding preparation of JCOs in the event of

'

.

.' -

28 -

..

9 ,,

, identified EQ problems. In response to this internal audit, NNEC0

has concluded that the need to prepare JCO's are not issue specific

and could relate to concerns other than EQ. Therefore,*this concern

can best be addressed generically in existing Nio 2.25 on

reportability.

NNECO has prepared a draft revision of NE0 2.25 to clarify that a

JC0 will be prepared in appropriate cases. Note, however, that

NNECO does not contemplate preparing a JC0 for equipment simply if

there is information suggesting a possible qualification deficiency.

Rather, where NNECO no longer has reasonable assurance of qua'ifica-

tion cf the equipment, appropriate action, such as preparation of a

JCO, will be initiated. The revised NE0 2.25 is currently under-

~'

going internal NNECO review, and is expected to be issued shortly.

>

,

f

$

i

.

I

f

.

'

,- ', - 29 -

..

.

.

UNRESOLVED ISSME

t

.

, A. ETAT OF THE ISSUE I

Ne ' ion Report unresolved issue 50-423/88-04-04 addrasses the

.

re..c 4 the EQ file for Target Rock solenoid operated valves (S0Vs). )

1

The inspectors were specifically unable to determine if IE Notice l

(IEN) 84-68, "Improperly Rated Field Wiring to Solenoid Valves,' had been

addressed for the Target Rock S0Vs at Millstone Unit No. 3. In addition,

the Staff requested formal documentation of calculations considering the

impact on aging calculations of self-heating.

B. NE C0's RESOLUTION

NNECO has previously completed an investigation to determine the applica-

bility of IEN 84-68 for Millstone Unit No. 3. The results of the

investigation were documented in a letter, NES-38528, dated May 15, 1985.

In part, it stated:

Solenoid valves manufactured by Target Rock are not impacted.

Target Rock has performed a test showing that the internal tempera-

ture rise is less than the rating of the field wiring (Reference

Target Rock Test Report TRP 4207).

t Field wiring at Hillstone Unit No. 3 has been determined to be qualified

fcr forty years when applied within its temperature rating. Therefore,

l

-

i

,

"

l

-

.

-

-

- 30 e

..

, the field wiring which terminates on normally energized Target Rock

solenoids is qualified for forty years at Millstone Unit No. 3.

r

.

NNEC0 will add NES-38528 to EQ File Number 654 to complete the documenta-

tion and resolve this issue. In addition, formal documentation of aging

calculations considering self heating are also being added to the EQ file

to close this issue.

.

b i's --

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

.

'

-

- 31

',.

[ 11)]IMARY

T

.

In summary, from the foregoing, NNEC0 maintains that the Staff inspection

findings associated with Litton-Veam connectors and Raychem installed over

~

braided jacketing constitute violations of NRC requirements. However, these

violations have minimal safety sigificance, and accordingly, do not individu-

ally or collectively warrant escallated enforcement action.

NNEC0 maintains that all EQ equipment at Millstone Unit No. 3 is presently

qualified for the current application, with the exception of four Litton-Veam

connectors associated with two valves. NNEC0 has placed appropriate controls

on those connectors until they can be replaced. NNECO acknowledges that

certain other Litton-Veam connectors are not qualified for a forty year life,

and accordingly, NNECO will replace that equipment at the next refueling

outage. However, that equipment is presently qualified and operable.

NNEC0 notes that while its efforts to implement an effective EQ program at

Millstone Unit No. 3 have generally been successful and demonstrate dedication

to EQ excellence, there are still some areas needing improvement. NNEC0 has

learned that in EQ, as with all important safety programs, success means a

continued effort throughout the life of the plant.

!

,_- -

s -

6 ,

o-

e

'

MILLSTONE UNIT NO. 3

CONTAINMENT AREA

HIGH-RANGE RADIATION MONITORS (HRRMs)

CHRONOLOGY OF EVENTS

-

01/24/86 -

NNECO Special Report MP-8655, W. D. Romberg to

Dr. Thomas Murley--Subject: Containment Area High-Range Radia-

tion Monitors.

04/28/86 -

GA HRRM installation and calibration completed; monitors declared

operational.

10/14/86 -

NUSCO Memo GSP-86-570, J. S. Nicosia/W. J. Haynes to Zvi Rendel

(Sorrento Electronics)--Subject: Request for Resolution of

General Atomic HRRM Qualification Ambiguities. (Note: Identi-

fies cable IR value discrepancy based on Rockbestos

Report QR-6802.)

11/06/86 -

Sorrento Electronics (SE) letter to W. J. Haynes--Subject:

Justification of Environmental Qualification of GA Technologies

HRRM for MNPS Unit 3. (Response to GSP-86-570; includes cable IR

justification and overall accuracy evaluation.)

02/27/87 - SE letter, T. A. Moshenrose to C. H. Clement--Subject: Defective

-

Coaxial Cable. (Note: Initial 10CFR21 notification.)

03/04/87 -

MP3 received Sorrento letter dated February 2, 1987. (Note:

Reasonable assurance of continued qualification provided by SE

letter of November 6,1986.)

03/06/87 -

Millstone Station Controlled Routing 6611 opened to evaluate ,

'

impact on MP1, 2, and 3.

03/10/87 -

MP3 Engineering received CR 6611. Contacted Tom Moshenrose,

Sorrento Electronics for additional information. Began i

collecting drawings, calculations, quality assurance test '

reports, etc.

03/12/87 - SE letter to C. H. Clement--Subject: Update on HRRM Cable

Problem. (Includes technical note with limited cable IR informa- ,

tion; states that "SE has not found a solution.")

03/24/87 -

SE letter to C. H. Clement from T. A. Moshenrose--Subject: Heat

Transfer Methodology for HRRM Cable Acceptability Determination. {

(Includes first model for analysis of cable response.)

03/24/87 -

Opened MP3 Commitment 3 87-0031 to track HRRH cable analysis and

response. [,

i

!

l 04/06/87 -

MP3 received SE letter dated March 24, 1981. Started Calcula-  !

tion 3-ENG-094.

!

'

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

_ _ _ _ _ _ _ _ _

.1

(. *

-2-

I

e

04/09/87 - SE letter to :. H. Clement from T. A. Moshenrose--Subject:

Revised Heat 'ransfer Methodology. (Includes revised heat

transfer modei.)

04/16/87 - MP3 received SE letter dated April 9,1987. Revised

Calculation 3-ENG-094, Rev.1, to incorporate better model.

04/23/87 - Completed review and approval of MP3 Engineering

Calculation 3-ENG-094, Rev. 1--HRRM Cable Operability Determina-

tion. (Note: Assurance of cable qualification conff aed.)

05/18/87 - Letter MP-3-928, C. H. Clement to S. Scace. Closed MP3 response

to CR 6611.

05/19/87- - NUSCO QA Engineering Assurance Audit (A-30123) of EQ Program.

07/20/87

05/21/87 -

MP3 Commitment 3-87-0031 closed.

05/22/87 -

Received SE letter dated May 18, 1987, requesting copies of

calculations for Millstone units.

05/28/87 -

C. H. Clement i.etter HP 3-940 to SE forwarded Calcula-

tion 3-ENG-094, Rev. 1.

06/29/87 - NUSCO Memo GSP-87-198, B. A. Tuthill/W J. Haynes to

M. F. Samek--Subject: Request Loop Accuracy Calculation for MP3

GA HRRMs.

07/24/87 - NUSCO QA Memo 1103866, V. Papadopoli to J. R. Ferraro--Subject:

Report c1 EEQ Program Audit (A-30123). (Note: Finding 3.10

addresses MP3 HRRMs and is source of potential violation.)

9/30/87 -

NNEC0 letter (B12671) to NRC, stating that the two GA HRRMs

satisfied the provisions of Regulatory Guuide 1.97, Rev. 2 and

the requirements of 10 CFR 50.49.

10/29/87 -

NUSCO Memo GSP-87-387, B. A. Tuthill to B. S. Kaufman--Subject:

Response to EEQ Program Audit (A-30123) Findings.

01/13/88 -

NUSCO Memo GIC-88-052, D. Robinson to B. A. Tuthill--Subject:

MP3 General Atomic HRRM Total Probable Error Calculation (Calc.

l No. PA-79-236-851-GE, Rev. 1).

01/14/88 - NUSCO Memo QSD 88-5008, B. S. Kaufman to B. A. Tuthill--Subject:

Status of Engineering Assurance Audit A-30123, EEQ Program.

(Note: Accepted response to Finding 3.10.)

02/02/88 - NUSCO Memo GSP-88-035, P. M. Blanch to

l

R. A. Crandall/E 1. Holloy--Subject: HP1, 2, & 3, and CY HRRM

Uncertainty,

i

i

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ d

y 9

-3- .

I

g ?' , *

'

,

o. j

j 02/09/88 -

hb?,CO Radiological Assessment Branch and Generation Electrical I

, Engineering review / approval of reportability evaluation for l

REF. No. 87-21, Containment High Radiation Monitors. (Note: l

This evaluation covers MP1 and 2 and CY in addition to MP3.) <

03/14/88- - NRC Inspection No. 50-423/88-04, MP3 EQ Audit.

03/18/88 .

i

f

l

i

!

\

l

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