ML19259B114

Responds to 781113 Request for Addl Info on Feedwater Line Vibration Monitoring Program Conducted During 1977.Concludes Monitoring Program Has Successfully Verified That Existing Feedwater Sys Will Operate Safely

ML19259B114
Person / Time
Site:	Beaver Valley
Issue date:	01/11/1979
From:	Dunn C DUQUESNE LIGHT CO.
To:	Schwencer A Office of Nuclear Reactor Regulation
References
NSC-1-78-094, NSC-1-78-94, TAC-6505, NUDOCS 7901160272
Download: ML19259B114 (16)

Text

y , ,

<>J I

- h__

(412) 471-4300 435 Sixth Avenue Pittsburgh, Pennsylvania 15219 January ll, 1979 IHIS :9CUMENT CONTAINS Director of Nuclear Reactor Regulation P00R QUAllTY PAGES United States Nuclear Regulatory Commission Attention: A. Schwencer. Chief Operating Reactors Branch No. 1 Division of Operating Reactors Washington, D. C. 20555

Reference:

Beaver Valley Power Station, Unit No. 1 Docket No. 50-334 Response to November 13, 1978 Request for Additional Information on Feedwater Line Vibration Monitoring Gentlemen:

4 Enclosed are four (4) copies of Nuclear Services Corporation Report No. NSC-1-78-094. This report includes the additional information you requested relating to the feedwater line vibration monitoring program conducted at Beaver Valley during 1977. This monitoring program was undertaken for the purpose of obtaining sufficient data, should severe feedwater line vibration be experienced, to determine the initiating mechanism, and to measure the magnitude of the resultant stresses to the piping system.

Prior to initiating the monitoring program, we had determined that the original feedwater regulating valves were dynamically upstable and were capable of undergoing severe oscillations. We had experienced severe feedwater line vibration incidents that were clearly audible and were felt by the operators in the Control Room and other work locations in the plant.

Feedwater flow oscillation of greater than 100% of rated flow were recorded during these events. The valve trim on these regulating valves was changed prior to tne commencement of the monitoring program.

The request for additional information concerns what we consider to be a very minor flow transient, and such low magnitude dyna 2ic forces that the occu'rrence was only detected by a careful review of the magnetic tape data.

This insignificant trar.sient was the single anomaly detected during almost five months of continuous monitoring.

f eh 7901160212_.

._. - t ,~ ,*

b Beaver Valley Power Station, Unit No. 1 Docket No. 50-334 Response to November 13, 1978 Request for Additional Information on Feedwater Line Vibration Monitoring Page We believe that the monitoring program has successfully verified that the existing feedwater system at Beaver Valley will operate in a stable manner within the design rating of the equipment, and that future incidents of feedwater line vibration will not occur.

Very truly yours, fl C. N. Dunn Vice President, Operations Attachment

,~.;_.....nc,.

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

;d[ 3 "

• '.,,',f e* , s.,-

.y,  : ._

.y

.' g N , f ..  ; h1 .

s s

--s.. .

y y -

~

m g ,'

• s 5: .-

4 'e$,. . i .  %

,- +

l

\ . ,., ' % - ,

. , y .

t, . .,-

p ,, ,

7. . x ,

u l p i  !% o  ? g,

• 1 4 .  :,

e 5 p

• . ,9

• **,e, c'

f' * #

i ar - * ,

=.

.p. ,

e a < >g,.,,. '

3' v .';

5 e k,. , -' "

. _. 2( _ . . . . . . _ - . _ - - ..-..-.._x.--

+

~ ~ .m 1700 DELL AVENUE CAMPBELL, CALIFORP!!% 95008

8 NSC-1-78-094 Revision 1 9

j i+ ~ - , -, m ~ gn t= p ra ps u

. - . J _ m L ,2 9 n o. _ } _

S RESPONSE TO NOVEMBER 13, 1978 NRC, REQUEST FOR INFORMATION ON BEAVER VALLEY STATION UNIT N0.1~

S FEEDWATER HYDRAULIC TRANSIENT MONITORING PROGRAM S

g Prepared for DUQUESNE LIGHT COMPANY by O

DUCLERR SERVICES CORPORATIOD CAMP 8 ELL, CALIFORNI A 9 Prepared by: . b . M.24 Approved by: EW1AA R. A. Uffer 5 ll H'. J. Tha r Reviewed by: ,\~ Date:D EGS.,n D g }$ $

. Bhailer H .'

Revision No. Date NSC Job Number _ Jssu Q By ,

Date 0 12/21/78 DUQ 0311 d O. d 17,/9)&

1 1/4/79 DUQ 0311 h

'V

[ V qjpp S

Sheet 1 of 1 e 1 nUCLERR SERYlCES CORPORST10n Rev ho-FIVISION SHEET S IE.{ DESCRIPTION PAGE NO.

1. Added additional pressure data information. 7, 8 9

e 9

e Revised by: . G. . Approved by:

, R. A. Uffer M H.N.Thailer Reviewed by: .

Issued by: /

' H. k Thailer G k.Randall e

Date: um 4, /D9 G

Y e

e i

I Fo= EMI-1.2 9

i' RUCLEAR SERVICES CORPORATION A DIVISION OF QURDREX CQ APQ A ATION 9

1.0 INTRODUCTION

The U. S. Nuclear Regulatory Commission (NRC) requested additional information g concerning feedwater system vibration monitoring program data obtained following the 50% load rejection test of July 17, 1977 at Beaver Valley Power Station Unit 1 (BVPS 1). Response to this request is contained in Reference 1.

O Upon review of Reference 1, the NRC issued a letter, Reference 2, requesting additional information concerning apparent line displacement and feedwater qb control valve (FCV) motion. The responses to these requests are reported herein.

O 9

9 9

0 1-9

• DUCLEAR SERVICES CORPORATIOD A OlVISION OF QURDREX CQ A A Q A ATION 9

2.0 ~ RESPONSES TO: NRC REQUESTS e 2.1 NRC Reauests The following are the requests for additional information that were received from the NRC:

9

1. Following load rejection test from 50% power on July 17,1977, the signals from the linear potenti, meter displacement sensor D-12B, 9 Figure 3, indicated four inches of vertir.al motion of the vertical riser of the loop B feedwater line inside of the containment build-ing. This data was dismissed on the basis that it did not exhibit

1. the characteristics of vibratory pipe motion. However, the rela-tively slow motion indicated by sensor D12-B might be associated with the normal closure of the main feedwater valve and the result-

1. ing change in hydrodynamic forces on the piping system. Please show whether the other data taken corroborate or disagree with the indicated four inches of motion. For example, show whether the

' lateral deflection sensors would have shown any response to this much vertical motion or show whether other vertical displacement sensors would have responded to such motion.

9

2. In Section 3.5*, it is stated that control valve disc movements of approximately 10 percent (of full travel) were recorded as a e

• Nuclear Services Corporation Report, NSC 1-78-012 as transmitted by Duquesne Light Company letter dated March 17, 1978,

Reference:

Beaver Valley Feedwater Station, Unit No.1, Docket No. 50-334, "Feedwater Line Vibration Monitoring Program Results".

9 2-

RUCLEAR SERVICES CORPORATI0n Q(_& DIVISION OFlRDREX CO A AQ AATION O

result of the 50 percent load rejection. Such valve motion could cause significant forces to be imposed on the piping

1. system. The magnitude of these forces would'be more than half of the magnitude of those forces that previously caused damage to piping system appurtenances. This deduction is 8 derived from the analysis in your report entitled " Main Feed-water Vibrations" submitted January 21, 1977. From your January 1977 report, Figure 2 on Page 4 and Figure 4 on Page 6 8 show that 5% motion of the old valve disc at 20% lift with a frequency of 6 Hertz would cause a peak to peak pressure varia-tion of 360 psi. This would correspond to a force variation I of approximately 50,000 pounds. This analysis is applicable to the new ported cylinder valve since the gain of the two valves is the same, 2.3% change in flow for each one percent change in

• valve position when either of the valves is between 25% and 40%

open. Please verify whether the indicated 10% valve motion was due to stem motion or pipe motion or some other phenomenon.

O Show whether th2re were any pressure signals or other indications corresponding to the indicated valve motion."

9 2.2 Response'to Request 1 T,.3 considerations are addressed in Request 1.

O The first is the relation of the apparent line displacement, as indi-cated by sensor D-12B, to normal FCV closure. This apparent line

• RUCLERR SERVICES CORPORATIOR Q(_A Division OFlRDREX CO A A Q A ATION O

displacement could not be caused by normal valve closure for the following reasons: -

9 a) Time Frame Incompatibility The apparent line displacement occurred approximately ten g seconds after the completion of normal closure of the FCV. A wave emanating from the FCV will reach the area of sensor D-128 in approximately .03 seconds.

O o) Magnitude of Forces It has been demonstrated analytically (Reference 3, Attachment D) 9 and by testing (Reference 3, Attachment G) that normal (controlled) valve motion will not create forces large enough to cause the apparent piping displacetents indicated by sensor D-12B vertical.

9 The second consideration addressed in Request I concerned the corroboration by other displacement data of the indications of sensor D-12B vertical 9 data and whether other displacement sensors would show a response to the apparent vertical displacement sensor at D-128.

8 Any hydraulic forces that could cause the apparent displacement indicated by sensor D-12B vertical would cause vertical, lateral, and axial displace-ments throughout the entire piping run. This has been demonstrated both 9 e.nalytically and by field observations of the damage induced by the three vibration incidents of November,1976, December,1976, and January,1977, 9

• DUCT.SAR SERVICES CORPORATION A DivissON OF QLIRDREX COR AC A ATION 9

reported in Reference 3, Attachment A. No such damage was observed following the 50% load rejection test, nor at any time during the operation of the Feec-e water System Hydraulic Transient Monitoring Program. Furthemore, neither the test personnel nor the plant operators observed any visible or audible indica-tion of abnormal pipe motion during the test. The displacement sensors on the 9 main and auxiliary feedwater lines show that the aforementioned vertical, lateral and axial displacements did not occur throughout the piping runs. In fact, most sensors recorded no piping motion at this time. Motion was recorded -

9 on the bypass line sensors at this time; however, as has been explained in the Reference 1 report, bypass line motion of this magnitude was noted during nomal plant operation. As has been stated in the Reference'l report, the indications

1. of sensor D-12B vertical cannot be considered representative of pipe motion due to its frequency and damping characteristics and the lack of corroboration by other data. Piping systems,. subjected to dynamic loading, will exhibit vibratory 8 response of many cycles. The D-12B vertical sensor indicated only a one-half cycle of motion, which is atypical of real piping system responses.

1. 2.3 Response to Request 2 Sensor D-7B was installed to provide an indication of FCV open-closed position and thus measured FCV stem position relative to the FCV body. The reported 10%

• FCV motion was stem motion.

Utilization of the analyses reported in Reference 3, Attachment D, and referred '

O to in Request 2, to obtain the system response to the Reference 1 test data is 9

• RUCLERR SERVICES CORPORATIOD Q(_A OIVISION OFlADREX C O R P O R ATIO N O

not valid for the reasons discussed below:

g, A) The Reference 3 analysis was based upon a +5% FCV motion applied:

1. To the old valve trim,

2. From a 20% open position, and gp 3. In a continuing sinesoidal cycle.

B) The valve movement, indicated in the Reference 1 test data:

as 1. Occurred with the new valve trim,

2. Was initiated from a 0% open position, and

3. Was one half cycle duration only.

O It should be noted that the half cycle of opening and closing recorded in the test data will not provide the reinforcement of pressure waves that could GD result from the continuing sinesoidal. cycle used in the Reference 3 analyses.

The test data (Figure.1).shows that the FCV moved from an essentially 0% open 45 to a 10% open position and then returned to its original closed position. The relation of CV to valve position is a nonlinear function. In the 0% to 15%

open range the gain of the FCV is very low, producing a 0.3% change in valve

1. CV for each one percent (1%). change in valve position. (Reference 3, Attach-ment B). Thus, the value of CV at the 10% open position is approximately 3% of full valve CV.

9 O

• DUCLEAR SERVICES CORPORATION 4 OlVISION OF Wl.lRDREX C O R P O R ATIO N O

As a limiting worst case, single step motion of the FCV from 0% to 10% has been calculated to result in approximately a 47 psi pressure change. An e instantaneous pressure change of 47 ps'i could result in an imbalance of 7600 lbf on the piping, which is considerably less than the forces obtained from Reference 3,. Attachment D analysis quoted in Request 2. Furthermore, 9 the analysis of Reference 3, Attachments J and K show the piping will be within its design basis limits following the application of forces larger than this force.

O A ramped pressure rise, caused by the actual FCV opening, would result in significantly lower forces as the pressure wave can reach the downstream 8 elbow of a piping segment befo"e it has attained its maximum value at the upstream elbow. Analysis based upon the recorded test data, discussed below, shows a maximum force variation on the piping of 700 lbf.

O Thus, the FCV motion, recorded in the test data, could not have induced forces similar in magnitude to those that previously caused damage to the piping sys-O ten for the reasons discussed 'bove.

a Pressure sensors which provided realistic data, either indicated a single 9

small pressure wave occurring at the time of FCV motion, or provided no discernible indications of an FCV induced pressure transient. An example of the data indicating a single small pressure wave (PT-8A) is presented in 6

Figure 2. Review of Figure 2 shows a steady state value of approximately 7-O

8 AUCLEAR SERVICES CORPORATIOR A OlviSIOlu 07 WURDREX C O R P O R ATIO N 9

1300 psia compared to an expected value of 950 psia. The erroneous reading was caused by drift in gain, bias or both. If the error is solely due to g) e bias, the indicated pressure rise of M to 60 psia would be an accurate interpretation of the data. If the error is solely due to gain, the data would reflect a 35 to 40 ps._ pressure rise. In either case, the minor 9 changes in interpretation due to such instrumentation error would not affect the conclusions drawn above; i.e., a 35 to 60 psia indicated pressure rise only differs by approximately 25% from the calculated 9 worst case condition of 47 psia. Such variations are insignificant when considering these small values of transient pressure increases and the associated force imbalances on the feedwater piping system. Those pressure 8 sensors that did not indicate this wave failed to do so for the following reasons:

a) the presence of the spurious pressure oscillations discussed in 8 Section 3.2 of Reference 1, b) equipment malfunctions, and c) the relatively small magnitude of the pressure wave.

• It should be noted that the ranges of the pressure sensors and their recorders (0-3500 psia) were set to record an incident of greater magni-tude that might have been associated with the old FCV trim.

O 6

9 9

nUCl. ERR SERVICES CORPORATION e A OIVISION OlB

%d.lADAEX CO A AO A ATlCN O

.1;LOSED

-74., . ..g., g

. , .p_7,, _ y y ,-,_

91! ii# M5 d'42-idi;ititili;.iilk;m -ip:d, i.y i; ~, ,'

g , . . - 'u. @..a::-:;p._.;.;.,.:piiC 21

.: t : . .,::

. re. ..lTi r2.,3. . . . . , .aJ;rMa e tr4'ri!)il'rtr: :iGAnf -[rgl-+j;;;iz: Nih H fEyPl'M AM.~iiiiijgi:niinT4tidilltja@lE fr;i:;;.;j :.j pp g,:tpi i

' :!rtTdi teRefitI75,;5uggp;;g:-iegi_.;;, .1,g.; .u.gq.7 e gg;,[- y gg

. if..' !r- d?Ri!

itMFyf a F:.!!gm.6,,g pj:; y;uig:,,;:. 4 giQgJ

'ed i ;ii rg 'f:iJEgi W T:Q 7y Te;  ;:11TiUri

. 7:!: '!11 t W.!Mi,t kr th "ill:in ira :Ig 'dg y ;p: , .;;,9;.;tpTT 7ji' 9 s P,4 !q um*s p-u:. .. --

!.1:. fu. wer6, 7 "if =" @ # ' '

!- ,mm ....h.v. ,!_.i:

iilms.-

-ii

.::s. ~ . ip m J:S

.- .,: ..;.; :d L. a,.,:i;.v

.Wn}.:

h 1.:2:

.:p r:_t ' .2 .q : .i: d.

,. . .p.g.3;p;

.040 sec~ ds

- 125 m- 4c.

e 0.7B - LOOP B FCV P051T10tl

/

O O

O O

FIGURE l e

BEAVER VAlt.Eif POVER STATt0N UNLT.1 LOOP S FCV POSITION FOLLOVING LOAD REJECTION TEST, JULY 17, 1977 e -9_. .

DUCLEAR SERVICES CORPORATIOD Q(.A DIVISION OF.lR m R E X CQ A AQ A ATION O

O 3500 psia # *FWi="IiiWH#M'

• MHENn E # $"4 s r "--4r#t==ig- a dE E;  ;= Pi #iii;isF msitt4 -t:;ntri; -

+ ati.-t

ci3:il;vi;ili+7.! EnMi+_tliisds intile1 - r:: :i palW'Mi; =n-  :- -- N 'i

!!k:

Itifirien #n ati!=HJ4 2,5 an  % r =i co.  !+*t=wf::t-linsM dsi-E i.cfJ;E .

UI4

six imi!Gitin'E=f"Wilif "a -NF4=t - -* ;=#:MER Hi; E=Wrf= W 1E= u12r=ii:Eh=tq-1,yFskH;#f#==4kiiE~iiii4bnte- -#Uri=-!

[N,:- n-_ ---w=ii- =v= ,.r.c= sfw =i E# E =i n w+-- r  :.=,- G

,: E = i&AEE =

ice : -

E i.V u

• E=:=xt= Se :::fiu-F; b 4 ti:Jn=11 :::4 Hi i!.?*i#I.= Niii~@ I5EtiE-?M$5E IEE iE i5 E F.fE 12:# =- iNfi :9f 0 t-:p g=ggge gt- ngvisimi aE- - t==*= =s= N =rE!E: 5 4CM=?

psia .

1. - l l _ .040 seconds 125 nm./sec.

PT 8A Loop A Feedwater Line O

Pressure (Downstream of FCV)

O FIGURE 2 BEAVER VALLEY POWER STATION UNIT FEEDWATER LOOP A PRESSURE FOLLOWING LOAD REJECTION TEST, JULY 17, 1977 O

O O

• RUCLEAR SERVICES CORPORATI0n A DIVISION OP Ql.IRDREX CO A AQ A ATH3N kB

3.0 REFERENCES

1. Nuclear Services Corporation Report Number NSC-1-78-012, " Beaver Valley Power Station Unit No.1 Response to NRC Questions ^3ncerning Reoort 3 'Feedwater System Hydraulic Transient Monitoring Program Results' dated March 13,1978.

2. Letter A. Schwencer, United States Nuclear Regulatory Commission to C. N. Dunn, Duquesne Light Company, dated November 13, 1978, Docket No.

50-334 0

3. " Evaluation of Main Feedwater Piping Vibrations", Beaver Valley Power Station Unit No. 1" Duquesne Light Company, dated January 21, 1977, Docket No. 50-334.

O O

O O

O O

11 -

O