ML20207C819
| ML20207C819 | |
| Person / Time | |
|---|---|
| Site: | Vogtle  |
| Issue date: | 12/23/1986 |
| From: | Bailey J GEORGIA POWER CO. |
| To: | Youngblood B Office of Nuclear Reactor Regulation |
| References | |
| GN-1261, NUDOCS 8612300248 | |
| Download: ML20207C819 (5) | |
Text
r= -
Georgia Fbwer Company
. Post Office Box 28't WaynesboroL Geo'2'a 30830 Telephone 404 554-9961 404 724 8114 Southern Company Services, Inc.
Fbst Office Box 2625 Birmingham, Alabama 35202 reiepnon. 205 87" "
Vogtle Project December 23, 1986 Director of Nuclear Reactor Regulation File: X4BC05 Attention:
Mr. B. J. Youngblood Log:
GN-1261 PWR Project Directorate #4 Division of PWR Licensing A U. S. Nuclear Regulatory Commission Washington, D.C.
20555 NRC DOCKET NUMBERS 50-424 AND 50-425 CONSTRUCTION PERMIT NUMBERS CPPR-108 AND CPPR-109 V0GTLE ELECTRIC GENERATING PIANT - UNITS 1 AND 2 NUCLEAR SERVICE COOLING WATER SYSTEM
Dear Mr. Denton:
A meeting was held December 15, 1986 with your staff in Bethesda to discuss 4
the Nuclear Service Cooling Water (NSCW) system performance. Attachment 1 provides a discussion of the information requested at the conclusion of the meeting.
If your staff requires any additional information, please do not hesitate to contact me.
Sincerely,
.k.
J. A. Bailey Project Licensing Manager JAB /sm rc:
R. E. Conway NRC Regional Administrator R. A. Thomas NRC Resident Inspector J. E. Joiner, Esquire D. Feig B. W. Churchill, Esquire R. W. McManus M. A. Miller (2)
L. T. Gucwa B. Jones, Esquire Vogtle Project File G. Bockhold, Jr.
0949V
{
8612300248 861223 r
j PDR ADOCK 05000424t' i
A PDR i
f- -
C ATTACBENT 1 NSCW WATERHAMMER EVALUATION P
Testing of.the Nuclear Service Cooling Water (NSCW) system.was conducted to determine the hydraulic behavior of the system during a Loss of Offsite Power (LOP) pump restart. Computer modeling (CE-111).of this system provided
- reasonable assessments of relative pressure peaks and locations. However, the computer program predicted pressures were too conservative. By optimizing the
+
system tuning by testing, the pressure peaks are minimized, the short tera duration of the pressure peaks is confirmed -(e.g. approximately 50 milliseconds), and the; piping reactions and system response during the LOP 4
restart are verified (e.g. noise, flows, pipe movements) to be acceptable.
These tests confirmed that with the NSCW system in the final tuned configuration, a LOP or normal restart will not result in a waterhammer.
- During the' testing on train A, two inadvertent waterhammers were experienced.
The testing involved a manual. simulation of the LOP restart sequence for the NSCW system. One event was caused by an incorrect temporary jumper in an
. electrical circuit, and a second more severe event was caused by a premature manual pump start. The second waterhammer was the equivalent of a " multiple" l.
failure event (i.e. both the tower valve and the pump discharge valve were not
.in their required position at pump start). Transient pressures as high as 400 to 500 peig were recorded by temporary test instrumentation. 'Some of the test pressure instruments failed during this event. The peak pressure during the second event at certain locations in the system was conservatively estimated 7
,to be 900 psig based on the following logic:
1 1.
The maximum pump runout flow is expected tc be 40% more than the design flow. -This would give a maximum velocity anticipated in the voided lines of 20 feet per second. Therefore, the total velocity to be dissipated in the middle of the run of pipe would yield a pressure peak of 600 psig i
(Auxiliary Cooler).
2.
The maximum pressure in small lines close to the Auxiliary Cooler L
(i.e. Reactor Cavity Cooler) will be higher than that at the Auxiliary l
Cooler (due to a smaller piping area) and is anticipated to be approximate 1v 900 psig.
3.
These estimated pressures do not include inherent conservatisms in the
. analysis, such as not accounting for dissolved air or pipe movement, which i
would tend to further reduce the peak pressure.
At other locations the estimated peak pressures were in the range of f
400-600 psig. The basis (as discussed in the meeting) for our confidence in determining the areas of maximum pressures is that although the computer
. program may not provide an accurate value of the peak pressures, it does provide a realistic trend of relative peak pressures at the various locations j
in the system.
i i
i l
i
---. +
--,..__-m_m-,--,--.
~,-,,_
__.,n,,_.,,.,_
of..
r NSCW Waterhammer Evaluation Fase 2
-A program was established to determine the effects of the waterhammer on system integrity and long term operational capability. The program consisted of walkdowns, inspections, evaluations, and component tests as follows:
Piping and pipe support walkdowns were made by trained personnel. Piping o
was inspected for dents; damage to attached vents,. drains and small bore branches; leakage at flanges, valves aud. nozzles; damage to flanges and valves; and pipe deformation or damage at penetrations. Pipe supports were inspected for damaged lugs or stanchions; damaged axial supports; marks on pipe'from supports; pulled or damaged anchor bolts; damaged baseplates; physical damage to support and components; damage to welds and damage-to structural steel.
GPC Nuclear Operations and Engineering prepared a system operability o
review plan. This plan involved a general system flow check of'7 locations in the system for comparison to previous values; inspection of 12 heat exchangers for leakage from the NSCW side; verification of eterability of pump discharge and bypass check valve; and stroking of swtor operated valves.
During the evaluation of the system response where line movements were o
observed, eight snubbers were stroke tested for operability. One snubber (part of a tandom pair) was identified as failed and the parallel snubber was tested and found to be fully operational. The damage to this snubber, whether it could have been due to the waterhammer event cannot be conclusively established. Design changes had already been issued to upgrade these snubbers to a larger capacity system for increased seismic anchor motions. This upgrade has subsequently been implemented. The piping at the broken snubber location was evaluated for local stresses assuming the snubber failed during the waterhammer event. The loads used ware based and snubber rupture loads provided by the vendor. The load piping stresses were calculated to be within acceptable limits.
o Based on the results of the computer hydraulic analyses, ten locations where the peak pressures (900 lbs. for 7 locations and 600 lbs. for 3 locations) determined to occur were evaluated for hoop stress. At each evaluated location, the hoop stress was below the yield strength of the mate rial, o
A review was conducted for the evaluation of components based on the responses received from the vendors and also the nature of the transient and its overall effect on the system.
a) NSCW Pumps - The peak transient pressures at the pumps were approximately 200 psig, and no impact to the pumps is expected.
( - -- y r
. o NSCW Waterhammer Evaluation Fase 3 b) Heat' Exchangers - From discussions with heat exchanger vendors and technical consultants, it was concluded that the type of damage expected from waterhammer would be loosened tube to tube sheet joints and distorted pass partition plates. In addition the piping configuration and the pipe supports in the vicinity of the heat exchangers were reviewed to determine the nozzle welds most susceptible to high waterhammer loads. Consequently, inspections were performed to look for the types of damage expected based on the guidelines given above.
c) Valves - Waterhammer damage to valves would result in binding or unusual movement during valve operation and/or bonnet leaks. Check valves that were subject to high pressure transients could have experienced internal damage. Inspections were performed based on the above guidelines.
d) Instruments - Instrument damage expected would be pressure boundary failures and instrument indication discrepancies. Inspections were performed based on these guidelines.
Based on these examinations and observations no damage to components was identified.
During the testing program (Run 14) a missing temporary jumper prevented the tower valve from opening and a mild waterhammer event occurred. A second waterhammer event (Run 15) occurred to a-premature manual pump start with both the pump discharge and tower valves fully open. The second event resulted from multiple failures in a temporary testing coafiguration and resulted in more severe pressure transient and system response than Run 14.
A single failure analysis of valve operation performed on the NSCW system.
j The results of this analysis indicated five potential pump discharge tower j
valve failures that could lead to a waterhammer event. The pressures that l
could result from these events were determined to be less severe than those i
occurred during the second waterhammer event (a " multiple" failure event) and is expected to be more comparable to the first event (a single failure event).
At other locations the estimated peak pressures were in the range of 400-600 psig. The basis (as discussed in the meeting) for our confidence in determining the areas of maximum pressures is that although the computer l'
program may not provide-an accurate value of the peak pressures, it does provide a realistic trend of relative peak pressures at the various locations in the system.
i i
1 NSCW Waterhammer Evaluation Page 4 l
Therefore, since the actions identified by the NSCW testing program have been successfully completed, the NSCW system has been declared operable and capable of performing its design function for its' design life.
09497 i
i I
f 1
I I
..y.,, - _,.,,,,
_,m,.-
-..,,,m.,__,.,_,,,.-,,,,,....,,,...-.
. -. -, _. _,,,