IR 05000397/1985028

Insp Rept 50-397/85-28 on 850724-26.No Violations or Deviations Noted.Major Areas Inspected:Licensee Actions Re Problems Resulting from High Level of Vibration in Reactor Recirculation Pump B & Associated Piping

ML17278A358
Person / Time
Site:	Columbia
Issue date:	08/01/1985
From:	Albert W, Thomas Young NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
To:
Shared Package
ML17278A357	List: IR 05000397/1985028 ML17278A356
References
50-397-85-28, NUDOCS 8508210457
Download: ML17278A358 (22)

Text

U. S.

NUCLEAR REGULATORY COMMISSION

REGION V

Report No.

Docket No.

50-397/85-28 50-397 License No.

NPF-21 Licensee:

Washington Public Power Supply System P.

O. Box 968 Richland, Washington 99352 Facility Name:

WHP-2 Inspection at:

WNP-2 Site Inspection conducted:

July 24-26, 1985 Inspector:

.

G.

A bert, c o nspe or Dat Signed Approved By:

T. Young, Jr.,

C ef, Engin er g Section Date Signed Summary:

Ins ection on Jul 24-26 1985 (Re ort No. 50-397/85-28)

Areas Ins ected:

A special inspection to examine licensee actions with regard to problems resulting from a high level of vibration in reactor recirculation pump "B" and associated piping.

The inspection involved 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> onsite by one NRC inspector.

Inspection procedure 93702 was followed.

Results:

In the area inspected, no violations or deviations of NRC requirements were identified.

I I)

85082i0457 85080b PDR ADQCK 05000397

PDRi (

,>u f

i AI e

I ~'

f J

J

'

I I I it ~

I f,

LJ

II L

f'L I

i I

J

"'"I I

I, L

I If Jf il I'J

JI f

Pl ll J

L, I

I IJ I

li 'l4 I

.JI

0

'i L

J

DETAILS 1.

Persons Contacted The following individuals from the Mashington Public Power Supply System were the inspector's principal contacts.

S. Davidson, Plant, Compliance Engineer (Recently Plant Engineering Supervisor'or Electrical Systems)

L. Martinez, Recirculation System Design Engineer R. Talbert, Nuclear Engineer 2.

Chronolo ical Histor of Pum Problems The problems with the "B" recirculation pump probably started in November 1984.

Listed below in order of occurrence are the significant events:

November 1984 Loss of pressure drop was noted across the second stage of the "B" pump seal.

No vibration problems were n'oted at this time.

However, the single vibration monitor at the top of the motor has not proven sensitive to the vibration which was later observed with more extensive instrumentation.

April 1, 1985 Hydraulic line to actuator on loop "B" Flow Control Valve (FCV) failed.

This caused the FCV to fail in the "as-is" position.

A few days before this failure, personnel reported what appeared to be an increase in vibration on instrument lines which penetrated containment.

During an outage starting on the same day, additional vibration probes were installed on both "A" and "B" pump motors and the broken hydraulic line was repaired.

After startup, monitoring showed the following peak values of vibration;

"A" Pump Motor 0.2"/sec

"B" Pump Motor 0.6"/sec Figure 1 is a plot of vibration observed during this period for both pumps.

April 8, 1985 The breaker for "B" pump motor tripped on high differential over current.

The trip relay was repaired and recalibrated during a one day outage.

At that time WPPSS believed this had solved the problem.

On restart from this outage the vibration levels were observed to be trending up.

"A" pump motor 0.2"/sec (unchanged)

"B" pump motor 0.8"/sec The decision was made to operate both pumps in the 15 Hz mode until the major maintenance outag l

'I

'

(<<

ll a

"i l

'(

I k

~ 1(

<<IE

'

E a

pf W

J I I'I l<<

l I

'I I

V

1 ll l

E E

<<.

EI I<<

<<

I I4 l *

May 1, 1985 Start of major two month maintenance outage.

During this outage a major repair effort was initiated which included the following elements of work:

a)

The motor for recirculation pump "B" was examined and minor problems with trust bearing and thermocouples were corrected.

b)

Pump "B" seals were replaced although condition of these seals showed that they were not source of the vibration problem.

k

Fl F

F E

c)

Pump "B" was disassembled below the seal area where it was found that the 16 cap screws holding the 'bearing housing to the stuffing box had failed.

The bearing surface showed only minor damage but these "graphalloy" surfaces were found displaced rotationally by 120

.

The graphalloy bearing surface is built in two cylindrical sections.

Of"'the,recessed cap screws holding the bearing housing in place, two had failed at the head to shank junction of the cap screws and 14 had

. failed at the shear plane formed by the junction

, of'.the bearing housing and the stuffing box.

The'upper register points for the bearing housing were within original specifications (1 to 4 mils).

However, lower register points were outside, specification by 16 to 22 mils.

i d)'nstalied new bearing without repair of lower regi.'ster'Sit on advice of Bingham-Willamette

"(BVC), the pump manufacturer.

BWC informed the

'licensee that, lower register fit has been analyzed as being unnecessary and in fact, recent pump designs have only one register.

e)

Pump "B" was reassembled with new bearings and new seals.

No attempt was made to further disassemble the pump beyond the bearings since this would involve removal of the motor.

However, a fiber optics device was used through a clean out port in adjacent piping to examine a

portion of the main pump impeller.

No anomalies were observed during this examination of about 60 percent of the outside of the impeller.

"A" recirculation pump was disassembled to the same extent as the "B" pump.

Nothing unusual

'as found and the pump was reassembled.

Figure 2 is a view of the cap screw ring holding the bearing housing in place for "A" pum <<I

'I ~

<<I<<

\\

I li

<<

I I

pk IE I'I l

f)-

y, Iil I EE

'lll I,

II p E la Il

\\

I ll

I

'

I I*

I ll-I I/

<<

I I'

g)

A cold test run was performed for "B" pump at 60 H< pump speed, restricted flow of 25,000 gpm, nitrogen pressure on system of 180 psi and temporary vibration instrumentation on motor, pump, valves and discharge piping.

The system was tested for one hour and the results were deemed acceptable; vibrations was:

Pump suction valve 6.0"/sec Flow control valve 1.2"/sec Riser 0.2"/sec Motor 0.5"/sec 4<

'i h)

Tests was made of natural resonant frequency for large valves in the recirculation system.

The frequency was found to be

HZ rather than the 39 H> found by G. E.

when the system was tested during manufacture.

i).

A sti,ffner was installed on pump "B" suction valve in an effort to raise the resonant fre'quency of the system.

') ',Vibration monitors were installed for operation.

,, The licensee stated that a more sophisticated

package is planned during the 1986 refueling

,outage.,

, k)

The pump motor was tested while disconnected from.the pump. It was found to operate without significant vibration.

July 2, 1985 t(

Reactor'tarted up.

During this run, miscellaneous problems restricted operation at higher power.

When these problems were resolved, three attempts were made to put pump "B" into service but it tripped out on high differential

'ver current each time.

Current transformer (CT)

resistances were measured as follows for the phases following each attempt to operate pump "B" at 60 H

Z'un time

~00 tl'l.

Resistance of CTs for Phases B

2 min.

1 min.

10 sec.

2.5 2.5 2.5 2.5 2.6 2.5 4.3 2.5 Open July 7, 1985 The plant was shut down and it was found that wiring for the current transformers had failed because the connection lugs were loosened by vibration.

The plant was restarted after, repairs to these connectors.

On restart no further problems were experienced with the current transformers.

Vibration

-1 f 4

'1 r(

't k

was now measured primarily, by deflection of the pump shaft in mils, which was part of the instrumentation installed during the two month maintenance outage.

The operating characteristics of the refurbished pump were measured as a function of flow (actually flow control valve position).

The results are plotted for July 8 and 9, 1985, in Figure 3.

The initial peak of 10 mils shaft deflection was considered excessive, flow control valve position was reduced in five percent increments until the 8 mil deflection point was reached.

Steady operation at this point showed a slightly upward trend in vibration as shown on Figure 3.

July 9, 1985 On this date another hydraulic line to the FCV broke necessitating a shut down for repair.

At this time a hydraulic damper was installed on the suction valve for "B" pump.

The previously installed stiffener had little affect on the magnitude of the vibration 'in this part of the, system:.

The reactor was restarted after a one day outage.

With the FCV closed, the pump shaft deflection was measured at 9.7 mils.

At the same time, it was found that the vibration of the suction block valve was reduced to about half of previous values.

The licensee considered the shaft deflection values to be excessive and went to single loop operation.

3.

Plant Status At the time of this inspection, tke plant was operating at 68 percent of power with "B" loop shutdown.

Technical Specifications had been changed to permit operation up to 72 percent of rated power with single loop recirculation flow.

4.

Licensee 0 tions At this point in time several options are available to the licensee.

Assuming that the source of vibration is the pump impeller assembly, the principal options include:

a)

Elect to continue in the present mode of operation until the scheduled refueling outage in 1986; at which time, the rotating assembly for Pump "B" would be replaced.

1 b)

Attempt to 'operate at 100 percent power until the refueling outage by putting Loop "B" back in service with higher vibration limits.

This course of action risks small bore piping damage and an unplanned outage.

c)

Continue operating 'with one loop and replace the rotating assembly

'of the pump at the earliest possible time that a new rotating assembly can be made availabl 'lt I

I4 Ilt I II I

C It I

IJ ht It I

J hl Jt J

lh I

t I

'

lt lt I

J I

( I IP'

p; J't I

~ )

jr PI I

I"

'

p ($ J'I I

I hh1l

5.

Licensee Ex ertise Since the advent of the pump vibration problem with recirculation Loop "B" of WNP-2, the licensee has employed several experts from outside the Washington Public Power Supply System.

The following three individuals were retained during the two month maintenance outage in May and June l985:

L. Laredore

-

Bingham-Willamette Co., with expertise in reactor cooling pumps.

J. Magruther

-

G.E. Marine Electric Division with expertise in vibration analyses for heavy rotating machinery.

H. Tafforodi

-

G.E.

San Jose, with expertise in reactor cooling systems.

Following the startup from the two month maintenance outage a panel of experts was assembled to assess the status of the vibration problem and recommend possible courses of action.

This panel consisted of the following:

W. Dornaus Pump Consultant (independent)

with expertise from experience with the Byron Jackson Division of Borg Warner Corp.

M. Plummer Systems, testing consultant with Nutech Corporation.

M. Torres Principal Engineer with expertise in startup testing and analysis with G.E.

B. Bedrosian

-

Chief Civil Engineer from Burns and Roe.

(Burns 8 Roe was the A.E. for WNP-2)

The individual reports from this panel of experts had yet to be received at the time of this inspection.

At their exit interview, the panel did make these points:

Operation of r'ecirculation pumps with minimum flow should be avoided.

The vibration levels of Loop "B" are excessive at the 60 H pump spec'd and extended operation of the loop at this speed should be

avoided.

There xs no apparent short term fix (e.g.,

dampers).

1[

U I

The'icensee.,shou3,d plan to,completely disassemble and repair pump ",B".,

I e

The recirculation loops-need permanent vibration instrumentation for long term monitorin !

ll I,H l

1 H

"i!I

~

H I

H A

H

\\

• ,

p, r

d

~HH

,1

,H,H HH IH

H'

+

H I'

s" H

II

1

"~,I i H

H H

H

1 I

It ll 1,

!

tl n

c'

I 8 PHP VEHRATXGN METH LGAG A

~

~ ~

~

~

~

~ i ~ 1

~

~ I ~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

< !

~

~

I I

~

~

~

~

~

~

~

~

~

L

~

~

~

~

~

~

~

~

~ I

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~ I

~

~

~

~

~

~

~

'I

~

~

~

~

~

~ I

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~ I

~

~

~

~

~

~

~

~

~ I e

~

~

~ I t

~

~

~ t

~

~

~

~

~ I

~

~ I

~

~

~

~

~

~

~ t

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

. ~

~

~

4

~

~

~

~

~

~

~

~

~

~ l 0

~

~

~

~

~

~

~

~ I

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~

~ I

~

~

~

~ \\

~

~

~

~

~

~

~

~

~

~

~

~

~

~

S

~

0

~

~

Pt<

VIB I t< ~SEC

~

'

~

~

~

~

~

~ I

~

~

i

.

5

9, ii i3, l5 4i5

~ f888 'rB 8 f338 B

PNP VIB APR

!.."D t"WF APR A

PNP V I B APR

L IIg

'

lilt%

t

CI cc I, C.

C. " II I

C.

h l.

II.

f c

I I

I y+r l

hh

+. e

~

.. aaamaamseaaesaaraaaaeaa eraaasaaaaaneaaaaaaaaaaa>

EEEEEEEEh'EEEEEEREEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEI 5505i5595%%5RS55%85%550i5 k855%555SkkkMSQRSAQSkag I

RRRRRRÃORkl%SRRRNRRRSRRRW kNRRNWNRRXRSRRNORRRAI rRaraar iraieraRRaaaaRMRXN aaARRRSRRRSRSaaaaaaraara arear,.

aeki.aaraaaekaaaarerkaaraaeaaaekaaakararaaka nerea'wk<<aaaaaaakaaaneakaanaaaaaaareaaaauaraN

~

XXEEIIIEEia'EEEEEREEXEMEIIEXEEXXXXEXXEXEEEXERXXEXI NNNNNNQNa>.NNSSS aS SSSQQNNSSNSWNNN1 NCNNN15 r

aaaRR larlaai ekaaaaraeaaaXRaaraarkkaaaaaiam S%aaraa RkRRN %5/Nka.~RQRRQNSRRRRRRSiRRRkNRNNRR a k kl

~

5555'555555

"-'EMXXgXXXL EmXXXEEILREI-.-~

Xl

~55l ~ErPEEEirXXie~ii EE

.

5 E~EiEXEhag~EXEEEEEE l "ylgllII~N-NlmL ll l llg P IIIIISI

~

lEi:RlikF'IIXXILEXEEXENXMSMEXEXXXXEMNEEXMXE

~

<BEEF)EERLXEMEXEE EEENEMEME MXEEXEXEEXEE ~

a Na ieeaaaSN='3EaaaXaaaa Raaaaara5laXr rNXaaaamaaaea N

Na Camsaakaaiaaaaaaaaa RRXemar aa Real aamaarem r l

amr aE ikEXXEEMEMXXla5555555555 55 ERIE EEEEEEEEM

~

L 55 'CEEEEELIEEMEEEREEEÃxaEEEEER Ekhh I%aa+ LEL

~

5.u iEEEIERLEEmE~

eSKSSEES5--,%55&IIXX IREE/

R NVORRRNRL e<

R RRRRNNR SNORER 5RR

~,

XXNEERXXEXXXXEX ~ LEEXEEEMEEEEEEINL EXXE ~

e'-

-~~-.. ariareaararaXRaaaae aaWRRRraaaaaaaaeaeaaaaaera a

~

Ei"qaaEElEEEEEEXEXEEELEELEIEEEEEMEXLIILEEEE

~

a

,.rMaamXSXaraaaaaaeaea RerareerraWeaaamaraaamr R

RR 55%%5kkRERARRRRERRR QRRIRORRRNRiQRMRRNRRQ

%5

~

RXEEEEXEEXXEXEXENEEEXE EL XXXNMIEXEXXXXIEXMXE tl)HR ILk IIINkkmlNNNNlN III SEER RRRRRRRNRRRRSRRRRRRRNiRR 55RRR NM Rk5 SRNRR%

kSRRMOSRQXkkk5%555555kmiMRRSNi55 Rk5kkSSRQkRSiRQki EEEIEEEELEEEEEEEEEEEEEIEEEEEEIERMREEEREEEIEEEEIEEE kiRMSMSRiRNkkQRNS5%%555k5555%MRRRSORSMROkkSRN5%0%8

C Modification of the loops to minimize vibration (e.'g.,

change resonant frequency)

should be seriously considered.

The basis for these recommendations arose, in part, from the following observations or conclusions:

Valves 23 A and B are in resonance.

There have been small bore piping failures due to vibration.

System resonance is close to pump operating speed.

The source of the vibration forcing function is the pump.

Bearing housing to stuffing box wear is excessive.

Xn addition, the panel concluded that wear ring clearances for pump "B" were excessive and poorly matched.

The actions being taken or seriously considered by the licensee generally follow those recommended by the panel (e.g.,

permanent instrumentation)

with one exception; the licensee considers restart of the loop with higher vibration levels a possibility if analysis indicates this will not result in damage to small lines.

Nothing was advanced by the panel along these lines.

The one consultant who agreed with this as a feasible course of action was Bingham-Willamette, the pump manufacturer.

One panel member (Dornaus) strongly urged that a modification be made to system operating characteristics or procedures to preclude operation of the system at less than 19,000 gpm through either pump.

6.

Exit Mana ement Meetin The following understandings of the NRC principal inspector were reviewed with the plant manager at the weekly meeting between the plant manager and Mr. A. Toth, the NRC resident inspector on July 29, 1985:

a)

Observed swings in reactor level wide range level while in single loop operation are instrumentation related noise due to location of wide range pressure taps.

This occurs as flow reverses in the inactive loop.

The matter has been reviewed with General Electric and it has been determined that no flow anomaly exists.

b)

The licensee is proceeding to procure a pump rotating assembly for installation in recirculation loop "B" during the 1986 refueling outage or earlier.

c)

The current transformer lugs will be, examined on the motor for pump "A" and the potential generic significance assessed and reported, as appropriate or necessary, to INPO or the NRC.

(

1

U

d)

The finding that the resonant frequency of major valves in the recirculation loop is 27 Hz rather than the

HZ reported by GE will be analyzed to determine if this implies that unusual stresses are present in the system and whether or not this represents a

generic problem which requires reporting under 10 CFR Part 2 h

, II