ML20055G939

From kanterella
Jump to navigation Jump to search

Summary of 900612 Meeting W/Util Re Reactor Protection Sys Walkdown Results & Proposed Mods.List of Attendees & Handouts Encl
ML20055G939
Person / Time
Site: Trojan File:Portland General Electric icon.png
Issue date: 07/19/1990
From: Bevan R
Office of Nuclear Reactor Regulation
To:
Office of Nuclear Reactor Regulation
References
TAC-76907, NUDOCS 9007240376
Download: ML20055G939 (22)
v • d • e


Text

..

D OC.Ab+

o UNITED STATES g

[

NUCLE AR REGULATORY COMMISSION i

6 wAsMiwotow, n. c. 2*6s July 19,1990 Docket No. 50-344 LICENSEE: PortlandGeneralElectricCompany(PGE)

FACILITY: Trojan Nuclear Plant

SUBJECT:

SUMMARY

OF JUNE 12, 1990 MEETING TO DISCUSS RPS WALKDOWN RESULTS AND PROPOSED MODIFICATIONS (TAC NO. 76907)

A meeting was held between representatives of Portland General Electric Company (PGE) and NRC staff on June 12, 1990, to hear a presentation on ac W ns to be taken by PGE as a result of walkdown findings at Trojan on the reactor pro-tectionsystem(RPS). These findings and actions have to do with the under-voltage /underfrequency (UY/UF) input to the RPS reactor trip on loss of reactor coolant flow. A list of attendees is provided in Enclosure 1.

Selected handouts used in the presentation are provided in Enclosure 2.

As a result of the walkdown, the licensee, working with Westinghouse (M) technical support, concluded that changes should be madt to ensure that the RPS will perform as designed under all design conditions.

It was determined that in a severe seismic event or on electrical failure, the non 1E 12.4 KV busses that power the reactor coolant pumps (RCPs) could be lost, along with the UV/UF input to RPS that would give the reactor trip on loss of reactor coolant flow.

Additionally, loss of the non-1E 12.4 KV bus and associated switchgear can cause grounding of the 120 VAC vital busses and consequent common cause failures of the engineered safety features actuation system (ESFAS) that are powered by these busses.

The licensee is addressing the loss of flow without reactor trip by performing an analysis to show that the low flow in>ut to RPS (rather than the loss of flow input) can appropriately serve as tie primary RPS reactor trip input.

The low flow input takes its signal from elbow taps which sense reduction of flow on pump runback when pump power is lost from the bus. This RPS input is not as immediate as loss of flow input from UV/UF, so there is a reduction in minimum DNBR when low flow is the crimary trip signal rather than the loss of flow as signaled by UV/UF input to RPS. When the first tri) signal to RPS is from " low flow" (elbow tap) rather than " loss of flow" (UV/ JF relays), the minimum DNBR can go into the W design margin, but minimum DNBR does not approach the safety margin.

This reanalysis by W to evaluate the use of the low flow input to replace the loss of flow input as the primary RPS input will be followed by a design evaluation of all the non-1E switchgear cabinets and any other non-1E hardware, as well as a design evaluation of the rest of RPS, for similar problems. Appropriate design modifications will be made to upgrade non-1E systems and thus to restore the UV/UF input to RPS as the primary reactor trip input to regain the DNBR margin lost.

9007240376 900719 i

4 DR ADOCK 0500 l

I I

s s.

In the discussion between PGE representatives and W technical support, and NRC staff, there appeared to be some confusion as to wiiether the non-1E switchgear cabinets and non-1E UV/UF reactor trip inputs were meant to be IE classification and if so, why they are not.

It would perhaps not be necessary for UV/UF cabinets to be IE if the UV/UF inputs are not relied on as primary trip input to RpS - as apparently is the case in some other similar W reactors. There appeared to have been some miscomunication between W andlechtel (A/E).

The W representatives at the meeting were reluctant To speculate on this, but theimpressionwasleftthatprobablyanumber(6-10?)iifiguration.

W plants probably have the same design as Trojan, and the same RpS as-built co Aside from the concern regarding the loss of power and loss of RPS input from

' UV/UF relays, resulting from loss of the 12.4 KV busses and switchgear, there was a concern that such loss can cause grounding of the 120 VAC vital busses, and consequent common mode failure of ESFAS's that are powered by these busses.

This ty)e of los$ can be protected against by proper fuse protection of each UV/UF ciannel which is coordinated with existing fuse protection between the 120 VAC power supplies and the 120 VAC busses.

The licensee stated that they are performing a safety evaluation in accordance with 10 CFR 50.59 to determine if the fuse coordination modification can be p(erformed within the requirements / criteria of 10 CFR 50.59.i.e., at the time of the meeting)

At this stage be done under the provisions of 10 CFR 50.59.

Questions were raised regarding the use of fuses as isolators as used in Trojan (and others of that era), and as proposed for use to protect / isolate each UV/UF channel of input to RPS as part of the short term fix. Although Trojan was licensed using fuses for this purpose, current regulatory practices do not permit use of fuses as isolation devices because fuses cannot meet the GDC 18 requirement for testability. However, it was pointed out by the licensee that use of breakers or other isolation devices would not be consistent with the licensed design basis and also would not be practical. Use of fuses in this situation is consistent with the approved design basis.

The question was raised regarding the suitability of making these changes and plant modifications under 10 CFR 50.59. The staff raised the question as to whether use of the non-1E UV/UF inputs to RPS as a primary input for a reactor trip involves an unreviewed safety question, and thus the modifications would need to reviewed by the staff and could not be done under 10 CFR 50.59.

The point was made that the plant had operated for 15 years in the present configuration, and it is improbable that a proposal by the licensee to restore as-built to design should require prior staff review and approval.

The licensee said they intend to proceed to complete an evaluation under 10 CFR 50.59 and to make the described changes provided the outcome of the 50.59 evaluation permits that activity. The 50.59 evaluation will be available for the staff to review.

l 1

- _ _ _ _ _ _ _ _ _7________ _ _ _ - _ - _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

a; +,

t

. July 19. 1990 in' summary, the material presented to the staff ap) ears to raise questions regarding dependability of RPS-inputs to perform t1eir reactor trip functions

'under al design conditions. The licensee is pursuing a course to remedy this perceived deficiency, but the deficiency might be much wider spread and exist at similar facilities.

OrloWSignedBy:

Roby Bevan, Project Manager Project Directorate V Division of Reactor Projects'- 111 IV,.V and Special Projects

' cc: See next page DISTRIBUTION' o' ' $1 M/-

ster Local PDR-OGC-Rockville 4

,PD5 Reading E. Jordan MS MNBB 3701 D. Crutchfield MS 13A2 ACRS(10)(P-315)

B. Boger MS 13E4...

J. Larkins F.Miraglia(12G18)

J.Partlow(12G18)

N. Slosson MS 17G21 i

i 1

o, i

(,

'L 4

la:

6

'DRSP/PD5/lA[y ) )

D5-(

/D:PD5 f/l@/90p')

}/q/90 7///90 DFoster 0 f' v Devan:rc JL s-0FFICIAL RECORD COPY Document Name: MTG

SUMMARY

6/12/90 i

+

s i

- l t '; j.'

n

L !

'l e

s Mr. Jares E. Cross Trojan Nuclear Plant

. Portland General Electric Company CC:

Senior Resident Inspector U.S.' Nuclear Regulatory Commission Trojan Nuclear Plant Post Office Box 0 t

Rainier, Oregon 97048 Mr. Michael J. Sykes Chairman Board of County ComIssioners a

Columbia County' i

St. Helens, Oregon. 97501 Mr. David Stewart-Smith Oregon Department of Energy Salem, Oregon 97310 l

Regional Administrator,~ Region V U.S. Nuclear Regulatory Commission 1450 Maria Lane, Suite 210 Walnut Creek, California 94596 a

t (4) t

ENCLOSURE 1 LIST OF ATTENDEES AT MEETING BETWEEN NRC PGE ON UNDERVOLTAGE/UNDERFREQUENCY REACTOR TRIP ISSUE Roby Bevan NRC T.D.. Walt PGE L.G. Dusek PGE Walton Jensen NRR/00EA Glenn Lang Westinghouse Nuclear Safety Don R. Swanson PGE Scott Bauer PGE Peter J. Kang NRR/ DST /SELB Mark Peery PGE Larry Kopp NRR/SRXB a

i l

TROJAN NUCLEAR PLANT N

s E

s UNDERVOLTAGE / UNDERFREQUENCY i

REACTOR TRIP JUNE 12,1990 l

l

INTRODUCTION OF PERSONNEL TOM WALT.

.. GENERAL MANAGER, TECHNICAL FUNCTIONS SCOTT BAUER.

BRANCH MANAGER, NUCLEAR REGULATION LANNY DUSEK.

. ENGINEER, NUCLEAR REGULATION DON SWANSON.

BRANCH MANAGER, NUCLEAR SAFETY MARK PEERY.

. SUPERVISING ENGINEER, NUCLEAR PLANT ENGINEERING DAVE HUEGEL ENGINEER, W NUCLEAR SAFETY GLENN LANG.

ENGINEER, W NUCLEAR SAFETY O

e

m.

}_

~

PRESENTATION OUTLINE i

SUMMARY

OF UV/UF REACTOR TRIP ISSUES - SCOTT BAUER

=

ENGINEERING REVIEW OF UV/UF REACTOR TRIP ISSUES

=

- MARK PEERY

[

)

  • DESCRIPTION OF THE 7100 SSPS AT TROJAN l
  • FUSE COORDINATION i

=

I ISSUE - DON SVENSON f

{

b l

l s

?

h e

e

y I

l DISCOVERY OF ISSUES i

1 i

i

(

- SEPARATION CONCERN i

- NONCONFORMANCE REPORT INITIATED l

j

= NONCONFORMANCE EVALUATION i

4

- NON-1E UV/UF REACTOR TRIP INPUTS 1

- FUSE COORDINATION d

l l

t i

V NON-1E UV/UF i

REACTOR TRIP INPUTS

= IT IS POSSIBLE THAT A COMMON EVENT AFFECTING TWO NON-1E SWITCHGEAR CABINETS COULD DISABLE THE UV/UF l

REACTOR TRIP INPUTS.

l

  • THE UV/UF REACTOR TRIPS-PROVIDE THE PRIMARY PROTECTION FOR A COMPLETE LOSS OF FORCED REACTOR

{

COOLANT FLOW ACCIDENT.

t I

  • SOLUTION:

- REANALYSIS OF THE COMPLETE LOSS OF FORCED I

REACTOR COOLANT FLOW-ACCIDENT

{

- REVIEW OF THE REMAINDER OF THE REACTOR PROTECTION SYSTEM FOR SIMILAR PROBLEMS i

i DESIGN EVALUATION l

FUSE COORDINATION IT IS POSSIBLE THAT A COMMON MODE EVENT AFFECTING TWO NON-1E SWITCHGEAR CABINETS COULD RESULT IN THE i

POWER INPUTS TO THE SSPS BEING FAULTED TO GROUND.

THE LOSS OF THE POWER INPUTS TO THE SSPS WOULD DISABLE THE SLAVE RELAYS USED TO PROVIDE AUTOMATIC ESF ACTUATION SIGNALS.

SOLUTION:

i FUSE COORDINATION MODIFICATION CONSISTENT WITH ORIGINAL PLANT DESIGN BASIS l'

j 1

i r

I u

e uu 3

e ENGINEERING REVIEW OF UV/UF REACTOR TRIP ISSUES MARK PEERY SUPERVISBNG ENGINEER g

,a 6

--,. o C.

wm=

e,.,.

,y' a

os

=

=

,4

.. 4--

s -

a, n, n,.

r--------------,

k<p SOLIO $ TATE LODic l

$s /

y hy e,

on t a 9

um e

  • *t e M

0,0ft U b1 M

i t-4 W 4 9

i:

HF.E u.

70n T*

i s

N IN0ft! U i

h=1 d

i t.4

r m 4i[

-)F.7 E-a.

sg;-

ga or<

un e a+

us e gig a

myg E F"" hQ _ _ _$*=

$a= =c l

um.

a

____6-o.,_ -,

D30 tasvoc cuss is ni D40 - tasvoc ci. ass is

...{

auteur<

Y11 trovac urs etass is a inam neur a

neu, Y13 12 evac ups cuss is a inam msten sus c4%

s s

Y22 taevac ups etass it s inam S

'S Y24 irsvac urs cuss a e raam SSPS UNDERFREQUENCY & UNDERVOLTAGE INTERFACE

n sm.

j n20 pac vliaL DIS 3NaestATION Class E i

sus YM SWE vt3

~(1(1(1(1

=(1(,.(I(1 e

TRAIN A a

o y;

TRAIN 8 i

L-m.unl a

i sen

--ICD -

i i

is.

sen -

TO prUT M

i anars esa M

(58 i

nwa cne e

una um DC J.L O*

re i-O tu.unl j

j esa U

~

t M

e aan 1 i-o

+=.=t ;n i,, i:

cna

=

m e

DC SC D-r._

r_

g um m.au g

cm2

=

D-una n

U M* "b d'

d' ma is.

c

a. ".

e o

2 ve some 1;

i

=M ac. l =c 11 t

FIGURE 2A

  • ~

AC POWER DISTRIBUTION SINGLE LIE DIAGRAM BEFORE MODIFICATION

~

- ~

.?

ai i<r o

l-l now-w =

a j

i i

l i

1 g.

I M

EML l

l

'""9 0

t 4 >

lat >

t >

l 83 4)

(>

H E4i>---r: <@lW j a"1"-#.wil a

  1. '*IE t > 14 4>

t >

I le t >

I O-

~

~

s M ' IdR&

i t

g

(

j,

I' ~

s '~

noi fu l

129V nEAC t

rii l

i i

l l

i l

l:

l' 1

WR l

1. fELAY IS NORMALLY ENOtol2ED WifM CONTACT OPOLLOSS OF 12SV AC CAustl CONTACT TO CLOSE TWS P810Yl0!NO TRIP LOOlc IWT.

& Depto,2E TO TRIP.

1 i

TYPICAL INPUT RELAY SCHEME h

I

i, -.

i

'y)

I' A

i--

m -..... f 4

s l

M f

l\\

3 m

..e i

a

~

s m

m.

m W

=

=

.g rg s9 b

s s

' s, s

s

[<

et g

s s

4e v

t 3.s

~.

l-f N

s s

g

-{

s

q g

s l

q P

s

, a-

>= =

5 4 := =

y g

-j %

q l

s l

w 1 %

+

=-...

=..;

s.: =

I EE A

a O\\\\\\\\M\\\\\\\\\\\\M\\\\\\\\y M ade 4

8 f

1 i

,q.

m F

129 pac vlTAL. DISTIRDEisfiATION CLASS IE -

SUS Yli SUS Y22 eUS Yl3 SUS Y24 (1((.(1

((1(..(..

1 a

[ ~ h.

~

~

~

TRAIN A -

TRAIN 8

% FILTEn j

j 9

OH I

~

y * ^* *:

=

v uze to Drur-cns eELAYS ITYPJ 4ev una m

o

,,3 n

j

-TO ISA D

ira.rsnj j.

jf,y, e

E en N

i-l cn2 4v m

D. ira. ten}- 82,3

?

en 8C

  • C 3-1

,_i_ 1 i

e 85 "

g um E

stave '

cna to 0 %rm.Yun

- 8!

f 5

a 3

G.A B

na en y

e i_ i e

m n

fE VPJ

~

.i FIGURE 28 AC' POWER DISTRIBUTION SINGLE-LINE DIAGRAM i

- AFTER MODIFICATION '(INSTALLED 5 AMP FUSES IN C46A & C47D)

,l ii.

NUCLEAR SAFETY REVIEW OF NON-1E UV/UF ISSUE 9

ACCIDENTS AFFECTED BY THE NON-1E UV/UF ISSUE

- FSAR 15.3.2 COMPLETE LOSS OF FORCED REACTOR COOLANT FLOW

- FREQUENCY DECAY: (ANALYZED, BUT NOT DISCUSSED IN FSAR)-

CURRENT SAFETY ANALYSIS

- PRIMARY ' CONCERN IS. DEPARTURE FROM NUCLEATE BOILING CRITERIA

- ANS CONDITION 111 EVENT

- DNBR > DESIGN LIMIT VALUE DIVERSITY

- UV/UF REACTOR TRIPS PROVIDE THE PRIMARY PROTECTION

- LOW REACTOR COOLANT FLOW PROVIDES BACKUP PROTECTION

- REACTOR COOLANT ~ PUMP PREAKER POSITION PROVIDES

~

ANTICIPATORY SIGNAL ~ AS BACKUP -TO LOW FLOW' O

g

~-s

NUCLEAR SAFETY REVIEW OF NON-1E UV/UF ISSUE NEAR TERM SOLUTION

'l

- REANALYZE THE COMPLETE LOSS OF FORCED REACTOR COOLANT FLOW ACCIDENT WITH THE LOW REACTOR COOLANT LOOP FLOW AS THE PRIMARY MEANS OF PROTECTION.

PREPARE A 10 CFR 50.59' SAFETY EVALUATION e

9

w NUCLEAR SAFETY REVIEW OF NON-1E UV/UF ISSUE

  • EFFECT OF. AN ASSUMED LOSS OF UV/UF REACTOR TRIP CAPABILITY-

- DELAY IN REACTOR TRIP OF APPROXIMATELY 1.5 SECONDS

- DECREASE IN DNBR AS A RESULT OF LOWER RCS FLOW (APPROXIMATE 15% REDUCTION)

-1 1

)

~

WESTIN6M00$E PROPRIETARY CLAS$ 2 f

~

Complete loss of Forced Reactor Coolant Flow Events-(continued):

v, Trojan Cycle 13 DNBR Margins and Penalties for Loss of Flows -

o Typical Thimble Cell Cell Design Basis DNBR Limit 1.38 1.36 Safety Analysis DNBR Limit 1.62 1.59 Loss of. Flow Minimum DNBR.

l.g3 1.83.

4 Accident Specific DNBR Margin

= -13%

(i.e.LOFDNBRversusSAL)

Safety' Analysis DNBR Margin

= -14%

(i.e. SAL versus Design Basis)

-Penalties Allocated Against the: Safety Analysis DNBR Margin:'

Rod Bow DNBR Penalty

= 1.3%

Reduced Design Flow Rate 1.5%

DNBR Sensitivity'to Flow at LOF Conditions:

1% Flow Reduction.

1% DNBR Reduction 4

e

p-q r

WESTINGHOUSE PROPRIETARY C ull't Trojan Loss of Flow i

2.3:

2.1.

J l

1.9 q

i L 1.7 EVENT SPECIFIC MARGIN i l SAFETY ANALYSIS MARGIN p-

.q 7 l.

i i 1.3 MARGIN OF SAFETY jg Ii L

1.0 t'

,O 1

2 3

4 5

6 TINE (Seconds) 4 L

i

Retrieved from "https://kanterella.com/w/index.php?title=ML20055G939&oldid=4485753"