ML20040F505

Evaluation of Effects of Corrosion Particles on Control Rod Drive Operation at Ja Fitzpatrick Nuclear Power Plant.

ML20040F505
Person / Time
Site:	FitzPatrick
Issue date:	02/04/1982
From:	Brugge R, Zull L GENERAL ELECTRIC CO.
To:
Shared Package
ML20040F500	List: ML20040F499 ML20040F505
References
NUDOCS 8202090276
Download: ML20040F505 (14)

Text

~ .-

EVALUATION OF THE EFFECTS OF CORROSION PARTICLES ON CONTROL R0D DRIVE OPERATION AT THE JAMES A. FITZPATRICK NUCLEAR POWER PLANT Prepared by:

L. M. ull Operating Plant Licensing Approved by: L '*yi "

R. O. Brugge, .anager Operating Licenses II NUCLEAR ENERGY DIVISIONS e GENERAL ELECTRIC COMPANY BAN JOSE.CAUFORNIA 95125 GENERAL h ELECTRIC

, 8202090276 820204 PDR ADOCK 05000333 PDR p

~C - _.

IMPORTANT NOTICE REGARDING CONTENTS OF THIS REPORT PLEASE READ CAREFULLY The only undertakings of General Electric Company with respect to information in this document are contained in the contract between the Power Authority of the State of New York (PASNY) and the General Electric Company (reference GEProposal.N.414-TY67-EP1andPASNYPurchaseOrderNo.W.Y.0-6-77-23) and nothing contained in this document shall be construed as changing the contrac't. The use of this information by anyone other than PASNY, or for any purpose other than that for which it is intended, is no,t authorized; and with respect to any unauthorized use, General Electric Company makes no representation or warranty, and assumes no liability as to.the completeness, accuracy, or usefulness of the information contained in this document.

i

l .-

TABLE OF CONTENTS PJ!2e

1.0 INTRODUCTION

1 2.0

SUMMARY

2 3.0 ANALYSIS 2 3.1 CRD System Flow Changes 2

,, 3.2 Flow Stabilizer Loop Corrosion 3 Particles 3.2.1 Corrosion Particles Below the Drive 4 Piston 3.2.2 Corrosion Particles in the Cooling 5 Water Orifice 3.2.3 Corrosion Particles in the Ball-Check 5 Valve 3.3 Exhaust Water Header Corrosion Particles 6 3.3.1 Corrosion Particles in the.Direc- 7 tional Control Valve Filter 3.3.2 Corrosion Particles in the Direc- 7 tional Control Valve

4.0 CONCLUSION

S 8

5.0 REFERENCES

9 ii

LIST OF ILLUSTRATIONS Figure Title a Pa2e 1 - CRD System Flows Through carbon Steel Piping 10 9

6 iii

1 ..

1 .

1.0 INTRODUCTION

In late 1976, cracks were observed in the vicinity of the Control Rod Drive (CRD) return line nozzle of several BWR reactors. The cause of the cracking was determined to be thermal cycling of the nozzle region due to the relatively cool CRD hydraulic system return line flow. In order to avoid thermal cycling and the possible cracking of the CRD return line nozzle at the James A. FitzPatrick Nuclear Power Plant (JAFNPP), the return line valve was isolated, preventing flow through the nozzle.

With the CRD return line valve is,olated the CRD system flow is changed such that there are now flow paths to'the drives which pass through runs of carbon s' ' teel pipe without subsequent filtering. The NRC has expressed concern (References 1 and 2) that operation under these conditions would increase the likelihood of foreign material being introduced into the drives,in such a fashion and quantity as to potentially impair their ability to properly respond to a scram signal. In order to detect the development of such a condition before it becomes a significant concern, the Commission's Safety Evaluation Report on operation with the CRD return line valve closed (Reference 1) concluded that the frequency of scram time surveillance tests on the control rod system should be increased from ten percent of the operable control rod drives at 16 week intervals to fif teen percent of the operable control rod drives at 8-week intervals.

This report addresses the potential impact of foreign material being introduced into the drives in such a manner as to potentially impair their ability to. properly respond to a scram signal. The results of the

~

analysis are then used to determine whether the increased test frequency is required to assure the scram reliability of the CRD system.

1

2.0

SUMMARY

This report evaluates i.he effects of corrosion particles assumed to be generated in the carbon steel piping of the flow stabilizer loop and the exhaust water header on the operation of the CRD system at the James A. FitzPatrick Nuclear Power Plant.

l The possible effects of corrosion particles from the flow stabilizer Icop on the operation of the drive piston, the cooling water orifice, and the ball check valve are analyzed. The possible effects of corrosion .

particles from the exhaust water header on the operation of the No. 121 directional control valves and associated filters are also analyzed.

The results of the evaluations indicate that the presence of corrosion particles does not affect the reliability of the scram function of the CRD system! In addition, scram testing is not effective for detecting potential operating problems postulated to result from corrosion particles.

The increased frequency of scram testing may accelerate drive wear whether or not corrosion particles are present in the drives. Tnereforc, it is recommended that the original CRD scram testing frequency (ten percent of'the operable control rod drives at 16-week intervals) be followed.

3.0 ANALYSIS 3.1 CRD System Flow Changes When the CRD hydraulic system is operated with the CRD return line isolated the CRD system flow is changed. The system flow changes result from the two new flow paths to the drives which pass through runs of carbon steel piping. As shown in Figure 1, these flows are:

1. The flow through the stabilizer loop, which then goes up the cooling water header to the bottom of the drives, and 2

p.-w -

1

2. The reverse flow up through the exhaust water header, through the No. 121 directional control solenoid valve, and then up to the top of the drives.

The possible effects of corrosion particles from the flow stabilizer loop on the operation of the drive seals and cylinders, the cooling water orifice, and the ball-check valve are discussed in Section 3.2.

The possible effects of corrosion particles from the exhaust water header on the operation of the 121 directional control valve, and the

- adjacent filter No.136, are discussed in Section 3.3.

3.2 FIowStabilizerLoopCorrosionParticles With the'CRD return line isolated, flow from the flow stabilizer loop is discharged to the reactor pressure vessel (RPV) via the drive cooling water flowpath. Since the piping downstream of the stabilizer valves (Fi,gure 1) is carbon steel, the flow passes through about 12 inches of l carbon steel piping before it joins the cooling water flow.

The stabilizer loop flow operates continuously betkeen 2 and 6 gpm, and is at relatively low temperatures. Therefore, it is expected that any carbon steel corrosion particulate matter which enters the flow will be small in quantity and particle size.

A quantitative description of the exact size and concentration of the corrosion products developed in the carbon steel piping is difficult to determine. However, the limiting condition is a situation in which l

relatively large sized particles are spalled from the carbon steel piping at a high rate. Assuming that these particles remain entrained in the cooling water header flow, (i.e., the particles do not settle out ,

at low points in the piping), the particles could eventually accumulate in the control rod drives, since the cooling water flow path to the drive is essentially free of flow restrictions and active valves.

3

x- - _ _ . - - _ - - - - - - - - - - - - - . - - - - _ - _ . - - - _ _ ,

1

• Once inside the drives, the corrosion particles could go: (1) into the underside of the drive piston; (2) up through the cooling water orifice; or (3) settle about the scram water ball-check valve.

3.2.1 Corrosion Particles Below the Drive Piston If the corrosion particles were to enter the volume below the drive piston, their presence would probably go unnoticed. During periods of normal plant operation, the corrosion particles would be expected to settle to the bottom of the drive, away from the drive piston above.

These particles would contribute little to the scoring of cylinder walls, or to seal failuras.

Since the pressure and flow to the bottom of the drive is highest during scram operations, the potential for lifting the corrosion particles in suspension and forcing them into the drive seals is the greatest during scram. However, even during scram operations, the corrosion particles would not be expected to cause significant scoring of the cylinder wills, or seal failures, such that the operation of the drive would be affected.

However, abrasive corrosion particles may accelerate wear of the drive cylinders and seals. Such wear is not a safety problem, since this degradation in drive performance already occurs during normal rod operations, such as rod position changes and reactor scrams. The minimum performance requirements of the drives during reactor operation is specified in the Technical Specification. If the limiting conditions for operation are not met, the CRD is considered inoperative and the.. subsequent operation of the reactor is adjusted, as required, to account for the inoperative drive.

4

m In addition to accelerating drive wear, increasing the frequency of scram testing provides little information about the conditions of the drive. The drive water pressure and flow available for rod position changes is less than that delivered to the drive during a scram. Therefore, ,

the condition of the drive components is much more* essential to normal rod positioning than to scram operation. Consequently, the drives will continue to meet scram performance criteria even after they have failed to meet the limits for normal rod movement and have, subsequently, been identified as inoperative drives.

3.2.2 Corrosion Particles in the Cooling Water Orifice It is also possible that corrosion particles from the carbon steel piping in the flow stabilizer loop may be carried up toward the cooling water orifice, where some particles may eventually become lodged in the orifice itself.

Theoretically, the orifice could become totally plugged.

Although this is not desirable, it is not a safety, problem. Blocking the cogling water orifice would cause the drive to heat up. Continuing plant operation with such a " hot-drive" would produce a more rapid degradation of the drive piston seals and bushings.

However an increase in the rate of seal and bushing deterioration, as mentioned in the case of corrosion particles in the drives themselves, will first be noted during normal rod operations. Scram testing will be a less precise indicator of the drive condition.

3.2.3 Corrosion Particles About' the Ball-Check Valve

. If corrosion particles were to settle about the drive ball-check valve, their presence would not be expected to affect the operation of the drive. There is sufficient clearance around the check ball and its cage l

l so that corrosion particles would offer little resistance relative to the large upward force exerted on the ball following a scram at elevated reactor vessel pressure. At low vessel pressures, the check valve is l not required to open to complete the scram.

5

? S g t? h W tsn YiW ild S d % W n m @ M k M w  : x m:C ww iww

The corrosion particles would be expected to settle about the ball-check valve, and pile-up like silt. Corrosien particles could get under the check ball (between the check ball and the valve seat), and might prevent ,

the ball from seating properly. However, although operational problems may be encountered during normal rod position changes, no safety problems would result, since the pressures and flows delivered to the drive during a rod scram are sufficient to compensate for possible leakage around the unseated ball-check valve.

3.3 Exhaust Water Header Corrosion Particles The elimination or isolation of the return line results in a rev.erse flow through the No. 121 directional control valves (see Figure 1).

This reverse flow comes from two sources: 1) the reverse flow up through the exhaust water header orificed check valve (No. 100); and 2) the exhaust water flow discharged from adjacent drives.

Consid'ering the limiting condition of relatively large sized corrosion particles spalled from the exhaust water header at a high rate, the potential consequences may be operational problems, but not safety problems. Particles carried by the reverse flow from the exhaust water header up through the 121 directional control valves will affect only two components: the 121 valve itself, and the filter element No. 136 just downstream of the 121 valve in the reverse flow direction.

3.3.1 Corrosion Particles in the Directional Control Valve Filter Fili a element No. 136 is a 50 micron filter. This filter will trap the large corrosion particles which have the greatest potential for causing scoring or accelerated wear in the drives. If the filter became plugged with corrosion particles, flow through the filter in both directions could be restricted. The flow restriction presented by the plugged filter could prevent adequate venting of the fluid above the drive piston and restrict upward rod covement when the rod is to be inserted one or more notches.

6

However, the plugging ef filter 136 would not affect scram performance in any way. Venting of the fluid above the top of the drive piston during a scram is accomplished by the opening of the scram discharge valve located upstream of the 136 filter.

If the filter became plugged to the extent that the limiting condition for drive operation in the Technical Specifications could not be met, maintenance would be performed on the Hydraulic Control Unit (HCU) and the filter would be replaced. If such corrective action were not taken, the drive.would be identified as an inoperative drive and removed from service.

3.3.2 .borrosionParticlesintheDirectionalControlValve The limiting case would be one in which the corrosion particles could theoretically cause the 121 valves to stick in the wide-open or fully-closed position.

If the 121 valves were to stick wide-open, problems would be encountered when attempting to withdraw, or " notch out" the rod. A portion of the drive water flow to the top of the drive would be discharged to the exhaust water header through the open 121 valve.

If the 121 valves were to stick closed, the consequences would be the same as those previously discusse,d for a plugged 136 filter, (i.e.,

problems during rod insertion, or " notch in" operation).

Neither of these two situations would prevent the safe shutdown of the reactor (Reference 3). Failure of the 121 valves in the wide-open or fully-closed position affects only normal rod position cht.nge operations.

If not corrected, failure of the valve in either position would result in the drive being designated as inoperative. However,scramtestinyof a drive would not reveal the existence of a failed 121 valve.

7

, ---sam .-. n.~. a , .

- ~ , n. . -_a . . _ ,

m- .

4.0 CONCLUSION

S This evaluation of the effects of carbon steel corrosion particles on the operation of the CR0 system with the return line isolated supports the following conclusions:

1. The potential presence of corrosion particles will not affect the scram function of the Control Rod Drive System.

e 2. Scram testing is ineffective for detecting potential problems which could result from ter,rosion particles in the CRD system. Although scram testing does provide a means of assessing the condition of the drives, degradation of the drive components is more readily apparent during normal rod position change operations.

3. An. increase in the frequency of scram testing will accelerate drive wear whether or not corrosion particles are present in the drives, 5$dprovideslittleinformationabouttheconditionofthedrive.

Therefore, it is recommended that the original CRD scram testing frequency (10 percent of the operable control rod drives at 16 week intervals) be followed.

4. Long-term operation of the CRD system in its present configuration may accelerate operational and maintenance problems unless appropriate corrective action is taken to replace the subject d.arbon steel piping, or to filter the flows that run through it. However, field experience at FitzPatrick and elsewhere, totaling about 30 reactor years of operation over the past 2 years, has not identified operational problems. The significance of no reported adverse experience is that the evaluation reported herein is conservative, and no performance l

trend changes are anticipated.

l

r

5.0 REFERENCES

1. Amendment No. 30 to Facility Operating License No. DPR-59 for the James A. FitzPatrick Nuclear Power Plant, dated September 16,1977.

2. Letter, G. Lear (NRC) to PASNY, " Safety Evaluation of Proposed Reduction in Frequen.cy of Control Rod Scram Surveillance Requirements, James A. FitzPatrick Nuclear Power Plant, Docket No. 50-333," dated June 1, 1978.

3. "BWR Scram System Reliability Analysis," NEDE-21514, dated December, 1976.

O .

e

+

9

cn .

f. . ' .

.f .... .

s 1 D t =o a a v V I *1 y iGf JL

~

y, U 1 5 bEV  % y D ho W Tw s *

• 2 2 I. d. =. m }

• = N 1

[.

Nuw Q ,.

n . .

v

(= u :h

-- k-. - ,

.;; .;Q _

5 l m

-.. ,s . ,

s s . . 5 rLp t v e

=

.r y t 8 u 5 a

D 2 ,V > X E

< $T b C -

h shig 1 i

, /,

a 2

e e

e H

1 Ah - , e -

e-g)>a--

~

.,. a L

, o

i -

.% 1 (

'I

=s::" - g g

A 0  % E I t *

~

m y e.d y 'Na " - ~ ~

- t h i m G h 2 o

!e i T Y .m a

y  ! d 5

' i j  :

t t O E 5 > Q 7

• y  ;

b ,

3 6 C 3.-

w L W W Q C 4 3

~ $

N Y n l

• 3 i '

i f N . I w$

,a < '

c=. i s T S i l .

& s i 1 g 'J E

N ,- M saveii 2,.s va ro I

s I eu

, g a<j {, --

-- J cp 2 y s  %

• l Jg _V I i U c c, '

c e e Er w

~

2 W. Vg

( d $h ad A

-. Os qh I

10