3F0688-19, Ro:On 880527,emergency Feedwater Piping Near Flow Control Valve Found to Be Warmer than Normal.Cause Not Stated.Water Added to Reseat & to Cool Emergency Feedwater Line & Valve Closed to Prevent Inadvertent Manual Addition of Water

Ro:On 880527,emergency Feedwater Piping Near Flow Control Valve Found to Be Warmer than Normal.Cause Not Stated.Water Added to Reseat & to Cool Emergency Feedwater Line & Valve Closed to Prevent Inadvertent Manual Addition of Water

ML20196J231
Person / Time
Site:	Crystal River
Issue date:	06/28/1988
From:	Widell R FLORIDA POWER CORP.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
3F0688-19, 3F688-19, NUDOCS 8807060345
Download: ML20196J231 (12)

Text

I eee b

• e W io ce s'

e.ee Florida Power C ORPOR ATioN June 28, 1988 3F0688-19 Document Control Desk U.S. Nuclear Regulatory Commission Washington, D.C. 20555

Subject:

Crystal River Unit 3 Docket No. 50-302 Operating License No. DPR-72 Emergency Feedwater

Dear Sir:

Enclosed is a report addressing the concerns on the leak at BFV-18.

Additionally, a Licensee Event Report concerning this subject will be submitted by July 21, 1988.

Should there be any questions, please contact this office.

I Sincerely,

) )

i Rolf C. Widell, Director Nuclear Operations Site Support RCW/REF/dhd Bnclosure xc: Regional Administrator, Region II Senior Resident Inspector l

8807060345 880628 ff PDR ADOCK 05000302 gjp S PDC Post Office Box 219

• Crystal River, Florida 32629

• Telephone (904) 795-3802 (

A Florida Progress Company

-t. ,- .

FLORIDA POWER CORPORATION CRYSTAL RIVER UNIT-3

' EMERGENCY FEEDWATER LINE LEAK JUSTIFICATION FOR CONTINUED OPERATION MACEGROUND:

Following the. restart of crystal River-Unit 3 from Refuel VI

'in January 1988, a check valve (EFV-18) in the Emergency Feedwater Pump 3B (EFP-2) injection line to.the "B" Once Through' Steam Generator (OTSG) was identified as having a small bonnet leak. At the time of discovery, the water leaking from EFV-18 was cold, and thought to be from the emergency feedwater tank. A flow diagram is provided as Figure 1.

On May 27, 1988, during a routine. plant walk down, the Shift Operations Technical Advisor (SOTA) discovered the emergency feedwater piping for the "B" OTSG near flow control valve EFV-55 to be warmer than normal. This was an indication that check valve ~FWV-43 wac not preventing back flow into the emergency feedwater line from the OTSG. This was discussed with the' Shift Supervisor on duty at that time and increased awareness of the condition was maintained. By June 19, the temperature'of the line from the reactor building penetration ,

to EFV-18 was hot enough to cause some flashing of the water  ;

-leaking from EFV-18. A Non-Conforming Operations Report (NCOR 88-81) was written on June 19, to evaluate and track the situation, and upon further investigation by the Engineering Department, a one hour report was initiated on

-June 21, based on exceeding the design basis temperature for the emergency feedwater line ( Ref: 10CFR72(b)(1)(ii)(B) ).

The elevated temperatures triggered an Engineering review of potential resolutions to the problem, and during the review it was determined that the original thermal analysis was calculated using a lower temperature than presently measured on' contact with the piping. The emergency feedwater piping and associated reactor building penetration design temperatures were originally specified as 1100F. The actual measured temperatures on contact with the piping and penetration exceeded this design specification. The maximum temperature reading recorded on the pipe at the penetration was 4900F. Attachment 1 provides a summary of temperature data' associated with the elevated temperatures of the piping and penetration. '

A review of operational history revealed several events which could have resulted in the heat up of the emergency feedwater lines and penetrations. Attachment 2 lists these events along with a brief explanation of each.

-,' i ,

INTERIM ACTIONS:

The following activities have been initiated or accomplished in response to the over heating of the emergency feedwater -

line.

1. Added water using EFP-2 to attempt to resent FWV-43 and to cool the emergency feedwater line on May 27, June 6, June 8, June 10 June 12 and June 19, 1988. The attempts to reseat FWV-43 were unsuccessful, and the cool down of the line met ,cith only temporary success.

2. Closed EFV-56 and EFV-57 (normally open "B" OTSG emergency feedwater flow control valves) on June 21 to prevent an inadvertent, manual addition of water to the heated line and penetration, and to possibly aid in the reduction of the leak at EFV-18 (See Figure 1). This action did not prevent the automatic EFIC actuated function of emergency feedwater to the "B" OTSG. The EFIC control logic will automatically open these valves upon initiation of emergency feedwater. These valves were reopened on June 22.

3. Performed a 100FR50 Appendix J Type B pressure test to accident pressure on the annular space of emergency feedwater penetration #109 on June 21. The test demonstrated the penetration remains leak tight.

4. Performed visual inspections of the piping, components and the penetration from EFP-2 to the exterior end cap of penetration F109 at the reactor building wall including the main fiedwater branch cross connection at FWV-34. These inspections were conducted on June 20, 21, and 22. In addition a visual inspection of accessible piping inside -the reactor building was performed. There were no apparent deformation or yielding of the piping, supports or penetration with the exception of EFH-126, a six way pipe restraint. This pipe restraint is a branch line structural support and the damage involved the anchor bolts pulling away from the concrete floor slab by approximately 1/4 to 1/2 inch. There was also surface chipping and spalling of the concrete adjacent to the anchor bolt penetration holes. An analysis of the effects of the EFH-126 damage was commenced along with repair efforts on the pipe restraint.

The restraint was repaired on June 24. The repair included replacement of the three 5 inch and one 8 inch anchor bolts with four one inch in diameter 12 inch long anchor bolts and grouting the space between the base plate and concrete ceiling. The analysis is discussed in the Justification for Continued Operation (JCO) section.

4

5. Initiated analysis on the following:

Manual calculations on the thermal stresses on the penetration steel components and a review of the temperature effects on the concrete-components (initiated June 21)

Thermal analysis on the piping inside the reactor building based on the elevated temperatures (initiated June 21)

Thermal analysis on the piping outside the reactor building in response to the elevated temperatures (initiated June 23)

Ret ..a 1 these analyses are discussed in the JC0 C .toL.

6. Initiated a Non-Destructive Examination (NDE) on the endplate to pipe weld at the penetration outside of the reactor building. The results of the NDE activities are provided in the JC0 section.

7. A radiograph of FWV-43 was performed on June 23. The valve internals appear to be intact and functional.

8. Installed a temporary modification on June 22 to short circuit the leak at EFV-18 to prevent hot water from migrating from the "B" OTSG to EFV-18. A diagram of the flow path of the modification is provided in Figure 2.

This modification, MAR T88-06-ll-01, utilizes a small positive displacement pump to inject water back into the emergency feedwater line at drain valve EFV-65 at a rate that slightly exceeds (approximately 1-2 liters per hour more) the collected leakage at EFV-18. By reinjecting the water into the emergency feedwater line at this point, the leak at EFV-18 is short circuited, thus preventing hot condensate water from the "B" OTSG migrating toward EFV-18, and a resultant heatup of the piping and penetration. This temporary modification cooled the line and has prevented the recurrence of the temperature rise enperienced in the line due to the EFV-18 leak.

9. Initiated a review of the long term environmental qualification issues essociated with components on the emergency feedwater line.

L - . . - _ _ _ _ _ _ _ _ _ _ _ _ -

wh

}.

s JUSTIFICATION-FOR CONTINUED OPERATION The temporary modification installed on June'22 to control the leak at EFV-18 has successfully maintained the

-temperatures of the'energency feedwater line to well below-1500F. .The.present penetration temperatures are being maintained at 95 to 1500F. 'This has reduced the-thermal

-stresses on'the.line,-components, and penetration to within acceptable design conditions,.and has eliminated the concerns Lover' thermal shock and water hammer in the event of emergency feedwater initiation.

The ability to maintain the system configuration, metering pump, makeup, etc., is enhanced by hourly monitoring of system operation and contact temperatures on the emergency feedwater pipe at the reactor building penetration. An additional pump and other components essential to the operation have been obtained and made available in the event that failure of these components should occur. The change out of the major components is estimated to require approximately two hours, and based on an observed 15.to.20 degree per hour temperature rise when the leak is not abated, the repair effort could be accomplished without exceeding 1500F at the penetration.

.There is an additional method to prevent the temperature in the emergency feedwater line from increasing should the above modification fail to accomplish this. Either emergency

'feedwater pump can.be placed into recirculation and the appropriate flow control valve, EFV-55 or EFV-57, opened a small amount to allow water to be injected into-the-line.

Thereby short circuiting the leak in much the same way as the installed temporary modification. Although this is not intended to be utilized for long durations, it would allow sufficient time to re-establish the temporary metering pump as the primary means of leak control while precluding an unacceptable temperature rise in the emergency feedwater line.

le The "B" OTSG emergency feedwater line check valve, FWV-43, which,is allowing back flow from the OTSG is also of concern.

This valve may not be completely seated due to insufficient differential pressure across the valve needed to overcome the slight frictional forces in the hinge mechanism, or to force the disc into a fully seated position. This valve is listed in CR-3's Technical Specifications Table 3.6-1 as a containment inclation valve but requires only a functional test of its ability to open and allow emergency feedwater injection into the "B" OTSG. There are no requirements to

-leak check this valve or to test its isolation time. Since the steam side piping and OTSG provide the primary containment integrity boundary, the only credible design basis accident involving FWV-43 is a steam generator tube rupture (SGTR). Nith FWV-43 allowing back flow and EFV-18 experiencing a leak to atmosphere, an unmonitored release path could exist to the environment during a SGTR scenario.

While this is an undesirable condition, the small leak at EFV-18 (approximately 0.5 to 1 gpm) does not significantly

[

, a impact the offsite doses during such an event and is bounded by1the FSAR accident analysis which assumes all noble gas and iodine radioisotopes are released through the main steam relief-valves for.the duration of the event. There are no significant-adverse conditions anticipated inside the plant

=which would prevent safe shutdown operations due.to such a

. release.

Preliminary analysis of the one-piping restraint, EFH-126, which was identified as recently damaged from stresses induced by the-heated pipe indicate that dead load capability of-the piping was not_affected, but that the seismic restraint capability was reduced to less than acceptable safety factor's. This support was repaired on June 23. The repair effort was commenced in parallel with the analysis effort and was completed prior to the verification of the analytis il results which indicated problems with the seismic loading en the pipe restraint. Continued operation with the damaged support was deemed acceptable due to the short time -

frame of the-loss of this pipe restraint combined with the low probability of a seismic event at CR-3.

The higher than analyzed for temperatures on the emergency feedwater line, components and penetration are under analysis. Preliminary results have been reviewed and the n following conclusion made.

Penetration M :

The concrete in the innsdiate vicinity of Penetration #109 has not been degraded and is still capable of meeting its original design requirements. The maximum measured concrete surface temperature near the penetration sleeve was 1950F.

The penetration sleeve was measured at 2720F on the exterior surface of the reactor bailding. Concrete temperatures are postulated to be higher on the outside edges and cooler toward the center of the penetration due to higher heat trans#er from contact with the penetration end plates.

Appendix A, titled "Thermal Considerations," of the ACI Code 349. O," Code Requirements for Nuclear Safety Related Conccete Structures" allow long term temperatures in local areu (such as around penetrations) of 2000F maximum.

Temperatures of 3500F for concrete surfaces are permitted for accident or any other short term period. Concrete chemical

, composition breakdown which would significantly affect concrete strength does not typically occur until temperatures (concrete) exceed 4000F.

The conservative thermal stress analysis of penetration #109 steel indicated that the 3/4 inch thick, outside containment, end plate had been stressed to the point where it deflected to relieve the thermal loading. This does not af# met the operability of the penetration or the piping since this end plate is not considered as part of the containment coundary or as a pipe support. The end plate was attached to the

-3.. .

2 penetration sleeve to allow pressure testing during plant

. construction. The containment boundary at the penetration is formed by a two inch thick end plate on the inside containment end of the penetration assembly. The thermal stress analysis indicated that the loading on this plate remained well below code allowable. Visual examinations conducted on the penetration external end pla,es and process pipe revealed no damage. Leak rate testing at containment peuk accident pressure also confirmed that the penetration remains intact.-

A liquid penetrunt examination of the penetration end plate to process pipe weld outside of the reactor building was conducted to reveal whether any damage had resulted from

- stresses induced by the increased temperatures. No indications were identified from the examination.

During the initial review phase following the discovery of the high temperature, continued operation was deemed acceptable based on the following:

Visual inspections of the penetration steel and surrounding concrete both inside and outside of the reactor building which revealed no apparent damage, Acceptable testing of the penetration to accident pressure utilizing a 10CFR50, Appendix J, Type B test, and Review of the design and function of the penetration end plates.

Emergency Feedwater Line Piping and Supports:

The thermal analysis for the piping inside the reactor building indicates that the piping and pipe hungers have not been subjected to stresses excaeding the design allowables and that all supports remained witidn allowable stress values when the thermal loads were combined with dead weight and OBE loading conditions. The thermal analya*,s on the piping outside of the reactor building has nat been coupleted.

Following discovery of the elevated temperatures, continued operation was deemed acceptable based on the following:

Visual inspections of accessible portions of the affected piping and pipe supports both inside and outside of the reactor building identified only one restraint which had partially pulled loose from its point of attachment, and no other pipe or pipe support damage or expansion contact indications.

Review of piping and support configuration which indicated the layout of the line afforded a good measure of flexibility to accommodate the thermal stresses which would be expected under these conditions.

Review of potential problems with water hammer effects in

~

the event-of emergency feedwater initiation indicated that this was an unlikely event based on the line being at the steam generator pressure and temperature, and the fact that the' initial injection would be with heated water which would reduce the likelihood of water hammer. -The once through steam generator emergency feedwater nozzles exit into the steam space, and the nozzle header routinely -

contains.some steam during normal operations.

LONG TERM ACTIONS:

Florida Power Corporation is presently developing options for long term repair efforts. These options include:

Online leak repair of EFV-18

- Repair of EFV-18 during a Mode 5 outage. The next scheduled Mode 5 outage is Refuel VII in September 1989.

However, this valve would be repaired during any unscheduled Mode 5 outage, and FPC is exploring the potential windows of opportunity for a scheduled outage this fall if the online leak repair efforts are unsuccessful.

Upgrade of the emergency feedwater piping, components and penetrations to accommodate the increased temperatures recently experienced.

If the emergency feedwater lite leak increases dramatically, or heating of the pipe or penetration occurs which cannot be controlled and exceeds 1500F, Florida Power Corporation will consider the emergency feedwater line as inoperable and take the specified actions stated in the Technical Specifications.

_ _ _ _ _ _ )

s. <

,s ,

. *s i

.r.

1 4

b

m. .

H@.

O5 d

Y

, *- -a g @

a. O~

2-W L =

(-

L'_

m T $

e +

e 2

a

$[ { r

'A o ms VM (q320su

-w o

%T T

• i' sw m.

W

?-

m b

i LL - L4 lAJ 4 m P s 2- b u _

= I '.o W

= $i t'

t

.ATTACRMENT 1 EMERGENCY FREDW&TER LINE TEMPERATURE PROFILES PEN # PEN #

DATE TIME 109 (SLV) 109 (PIPE) FWV-43 EFV-55 EFV-18 ,.

6/19/88 - - -

490*F 330' F 228 F 6/21/88 -

272 F 480 F 480 F 375 F 255'F 6/22/88 1230 298 F 475 F 460 F 334 F 258 F 1830 224 F 278 F 240 F 122 F ,

122*F 2100 173 F 169 F -

111 F 107 F 6/23/88 0130 120 F 122*F 118 F 98 F 97 F 0600 110 F 106 F 101 F 98 F 96 F 6/24/88 0800 108 F 112 F -

92*F 94*F 1600 -

128 F - - -

6/25/88 - -

97*F - - -

6/26/88 - -

96 F - - -

,,.. e

• a 9

ATTACHMENT 2 HISTORY OF EMERGENCY FREDWATER LINE

, TEMPERATURE EXCURSIONS REFERENCE DESCRIPTION UOBR 82-25 Applied main feedwater through the emergency feedwater line for approximately 17 minutes. No temperature data was available, however, main feedwater temperatures would be expected to be approximately 4750F.

LER 82-76 Failure of EFW ultrasonic flow meters due to excessive pipe temperatures. No times or temperatures were available LER 83-43 Failure of EFW ultrasonic flow meters due to excessive pipe temperatures due to back flow through FWV-43. No time or temperatures were available.

EQ #88-1191 Identified apparent overheating of all four emergency feedwater lines while plant was in Mode 2 in March 1988. The lines were cooled using emergency feedwater and the temperature rise was not experienced again until June.

9

i

' ~

FIGURE 2 0 D1CIC &nott ko. [ $ -O[o-ll0( l OT%'e"

. ,, Aux N

/ s/- z gg%y 4 9'. 3/ "

._ /45'-9 a

I t

d a- '

1. Y"I3 III' 0 N. $l

.L -

o ,,

/02-9 L' i i!

/Oh"O sn-ia j,, ' r4 aov ~.*p' ' '.< pen ios m .2

... . s w.gg use- ..- ~ msxe eme , f g ,, hu -

J ,'. ... .

mm-11 x.f_ a

.i m SFV-GS s

$5Pbl y/y 9 Fhy

\-g 95-0 EfP

' o _ _-S c, 83 m_ _ _

(GV 13't)

POM9)

- .. -