ML13330A046
| ML13330A046 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 03/31/1980 |
| From: | EG&G, INC. |
| To: | |
| Shared Package | |
| ML13330A044 | List: |
| References | |
| TASK-05-13, TASK-5-13, TASK-RR JAD-80-80, NUDOCS 8005120119 | |
| Download: ML13330A046 (14) | |
Text
JAD-80-80 S80051#ff'j March 31, 1980 Page 1 of 14 ATTACHMENT TO ENCLOSURE.
STEAM GENERATOR WATER HAMMER TECHNICAL EVALUATION SAN ONOFRE UNIT NO. 1 March 1980 EG&G Idaho, Inc.
CONTENTS I. INTRODUCTION.........................
1 II. WATER HAMMER EXPERIENCE..
3 III.
1EANS TO REDUCE THE POTENTIAL FOR STEAM GENERATOR WATER HAMMER...
5 IV. OPERATING EXPERIENCE AND STEAM GENERATOR WATER HAMMER SUSCEPTIBILITY.....
7 V. CONCLUSIONS AND RECOMMENDATIONS.
VI. REFERENCES.......................
12 TABLES I. History of Steam Generator Feedring Uncovery Events.....
8
I.
INTRODUCTION An evaluation was performed for the San Onofre Unit No. 1 feedwater system. The purpose of this evalution was to assess the susceotibility of the feedwater system to water hammer during ooerating transients and situations that could result in conditions conducive to water hammer.
Steam-water slugging resulting in water hammer in the steam generator feedrings and adjacent feedwater piping was considered in this review. This type of water hammer is generally referred to as steam generator water hammer and is characterized by rapid steam condensation causing water slug acceleration and impact. The impact of a water slug in the feedwater system has the potential for overstressing system components and supports. Although.water hammer events have occurred in the San Onofre Unit No. 1 feedwater system, these events are not attributed to steam generator water hammer.
The information for this evaluation was obtained from;
- 1) discussions with the licensee, 2) licensee submittals to NRC of July 14, 19751, December 27, 19772, July 3, 19793, August 31, 19794, and February 14, 19805, 3) the "San Onofre Nuclear Generating Station (Unit No. 1) Final Engineering Report and Safety Analysis" 6, 4) "An Evaluation of PWR Steam Generator Water Hammer",
NUREG-02917, and 5) Westinghouse Technical Bulletin, NSD-TB-75-78.
A review of the water hammer events at San Onofre Unit No. 1 is presented in Section II. The means to reduce the potential for water hammer at this facility are presented in Section III.
Section IV presents descriptions of the feedwater system geometry and operation pertaining to the susceptibility of the system to steam generator water hammer. This section also presents a description and tabulation of operating transients and situations that could result in conditions
conducive to water hammer. Finally, conclusions and recomendations are oresented in Section V concerning the susceptibility of the feedwater system at this facility to steam generator water hamer.
2
II. WATER HAMMER EXPERIENCE During the initial startup of San Onofre Unit No. 1, several minor and sparsely documented water hammer events occurred. The causes and circumstances related to the events are unknown, however, no damage due to the events was found. Since normal power production of the unit began in 1967, three reported and damaging feedwater system water hammers have occurred although none of these events are attributed to steam generator water hammer.
The first event occurred on April 29, 1972 and was assumed to be the result of a failed valve positioner on the main feedwater regulating valve of loop C. During unit startup, a reactor trip and safety injection signal (SIS) occurred resulting in feedwater isolation by closure of the main regulating valves in all three main feedwater loops.
Failure of the valve positioner allowed the main regulating valve to partially open, restoring an unknown feedwater flow rate to the steam generator. The feedring in the loop C steam generator was uncovered and had been about 14 minutes prior to the event that occurred in the feedwater line of loop C. The 'only damage related to this event was the loop C valve positioner. The positioners on all of the main regulating valves were replaced with a tvoe that would reduce the chance for future failures.
It was not known when the second event occurred but the resulting damage was found on January 14, 1974 during a unit outage. The damage was assumed to be the result of a single event although multiple unnoticed and damaging events could have occurred. An in-containment inspection revealed two broken knee supports and one broken snubber on the main feedwater line of loop B. Necessary repair and/or reolacement of all damaged components were made prior to resumption of reactor ooeration.
3
0
.0 The third event occurred on-May 14, 1979.and was assumed to be the result of improoer adjustment of the setpoints for automatically positioning the main feedwater regulating valves following a reactor trio. The valves closed completely instead of to the correct preset oosition that allows about 5% of full feedwater flow to enter the steam generator for residual heat removal. A water.hammer occurred in loop 3 after an unknown feedwater flow was restored to the steam generator in this loop. The feedring had been uncovered about 45 minutes prior to the event. The only damage associated with the event was a sheared locking nut on one of the snubbers on the main feedw.ater line of loop B. The damage was repaired prior to resumption of reactor operation.
4
0 0
III.
MEANS TO REDUCE THE POTENTIAL FOR STEAM GENERATOR WATER HAMMER The following means are employed to reduce the potential for damaging steam generator water hammer at San Onofre Unit No. 1:
- 1. The effective horizontal length of main feedwater piping adjacent to any steam generator is less than 32 inches.
- 2. Administrative controls require operators to maintain steam generator water levels at 50% of narrow range when feedwater is being controlled manually and reactor power is below 20%
of full power.
The main feedwater piping geometry (original plant design) provides a "loop seal" at the entrance to each steam generator. Each seal consists of a downward turning 450 elbow a short-distance from each feedwater nozzle. The horizontal lengths of piping from the feedwater nozzle to the centerline of the elbows are 31 3/4 inches for looos A and C and 19 1/4 inches for loop B. This arrangement limits the length of piping adjacent to each steam generator that could drain through the bottom discharge holes of the feedring and allow steam to enter during periods of feedring uncovery. Thus, limitation of these piping lengths will proportionally limit the energy of potential water hammers due to steam-water slugging.
Based on our understanding of steam generator water hammer, this phenomenon would be most likely during startup, shutdown, and low power situations when feedwater is under manual control and the flow rates are insufficient to maintain uncovered feedrings full of water.
To avoid admission of steam and possible slugging during these ooerational conditions, the feedrings must remain covered with water.
Administrative controls adopted in December of 1977-require that steam 5
generator water levels remain at about 50% of narrow range (the feedrings are uncovered below 26%) during startup, shutdown, and reactor oower levels below 20% of maximum power.
6
IV. OPERATING EXPERIENCE AND STEAM GENERATOR WATER HAMMER SUSCEPTIBILITY The conditions most conducive to steam generator water hammer occur when the steam generator feedrings are uncovered and steam enters the feedrings and attached horizontal feedwater piping.
Stean-water slugging and subsequent water hammer may occur when incoming cold feedwater mixes with the steam in the piping and rapid condensation occurs. The conditions can be avoided by keeping the feedrings and associated piping full of water. This can be accomplished by 1) keeping the water levels in the steam generators above the feedrings or 2) suoplying feedwater at a higher flow rate than the rate at which feedwater drains through the discharge holes on the bottom of uncovered feedrings.
Table I presents the history of feedring uncovery events for San Onofre Unit No. 1 since the unit was started up in 1967.
The table shows that a reactor trip almost always results in feedring uncovery in all steam generators.. The drop or "shrinkage" in water level and subsecuent feedring uncovery is the result of interrupted reactor ocwer oroduction and increased steam generator steam pressure causing the collaose of steam voids within the secondary side of the steam generators. A similar situation would also be experienced during events such as loss of main feedwater, loss of offsite power, steam line break, and loss-of-coolant accident.
ormal startups and shutdowns with feedwater under manual control also resulted in a substantial number of feedring uncovery events.
These events continued to occur in 1978 and 1979 although the administrative steam generator water level controls described in Section III were in effect for all startup and shutdown operations.
7
TABLE I HISTORY OF STEAM GENERATOR FEEDRING UNCOVERY EVENTS NUMBER OF UNCOVERY EVENTS NO. OF OUTAGES SHUTDOWNS REACTOR TRIPS STARTUPS YEAR NO. OF OUTAGES W/NO UNCOVERY EVENTS Aa B
C A
8 C
A B
C (No. caused by TOTAL reactor trips) 1967 5
(3) 2 1
3 0
1 1
1 0
0 0
1968 8
(2) 2 3
2 1
2 1
2 3
3 1
1969 9
(1)b 1
8 3
5 1
1 1
4 10 8
1970 3
(0) 1 1
2 0
0 0
0 1
3 1
1971 15 (9) 2 1
2 0
9 9
9 1
8 7
1972 13 (4) 1 1
6 3
4 4
4 5 20 9
1973 4
(0) 0 3
3 3
0 0
0 2
4 1
1974 8
(2) 3 0
1 0
3 3
3 3
2 5
1975 3
(2) 0 0
0 0
2 2
2 3
6 3
1976 15 (6) 1 6
5 5
6 6
6 13 16 20 1977 7
(4) 0 0
1 0
4 4
4 12 6 18 19785 (4) 0 1
2 1
5 5
4 2
3 2
1979 7
(2) 1 1
2 1
1 2
1 1
2 2
TOTALS 103 39 14 26 32 19 38 38 37 50 83 77 a
Steam generator designation.
b Actually 2 trips occurred, however the charts were available for only one.
8
The only other operational situations resulting in feedring uncovery beside those accounted for in Table.I were.two loss of main feedwater events.
The events occurred on April 29, 1972 and Octoter 21, 1973. Each event was the result of an SIS which caused isolation of the feedwater system and termination of feedwater flow to the steam cenerators. The first water hammer event described in Section II occurred in conjunction with the April 29, 1972 loss of feedwater event.
It is apparent from the review of the plant operating history that it is not always possible to avoid drainage of the feedrings and adjacent piping by keeping the feedrings covered with water. Although the drainage time (or uncovery time) varied among the uncovery events, complete drainage of the feedrings and adjacent piping has been frequent since the time required for complete drainage is less than one minute and the uncovery times were generally of longer duration.
The alternative to continous coverage of the feedrings with water is to maintain sufficient feedwater flow through uncovered feedrings to keeo the feedrings full of water. The feedwater flow required for this at San Onofre Unit No. 1 is approximately 1500 gpm per steam generator based on static pressure drainage calculations. Based on the maximum design main feedwater flow of about 14,000 gpm or 4750 gpm oer steam generator, the flow required to keep the feedrings full of water is about 30% of maximum flow. Since the maximum design main feedwater flow is the approximateflow required for heat removal at full reactor power, the feedrings, if uncovered, would not be kept full of water with feedwater flow rates corresponding to power levels below about 30% of full power. Thus, during startup, shutdown, and low oower operation, feedwater flow requirements are insufficient to keeo the feedrings full of water during periods of feedring uncovery.
This situation would also be expected during recovery from events resulting in reactor trios from power operation such as loss of main feedwater, loss of offsite power, steam line break, and loss of coolant accident.
9
Fiedwater can also be supplied to the steam generators via the auxilia y feedwater system. Auxiliary feedwater, when required, is usually pumped into the main feedwater header upstream of the location where the header solits into three main feedwater lines to supply each of the steam generators. Since the auxiliary feedwater system can be actuated only by manual local control, main feedwater is utilized whenever possible. Normally, auxiliary feedwater is used only for heat removal during subcritical reactor operation (startup and shutdown). The motor driven auxiliary feedwater pump, with a 235 gpm flow ca:acity, was reported to be used for these situations. The 300 gpm turbine driven auxiliary feedwater pump is used only as a backup system. The auxiliary feedwater system cannot, however, supply sufficient flow to keep the feedrings full at any time, since the maximum flow capacity of the-system is only about 535 gpm.
The review of the operating history at this facility indicates that the conditions conducive to steam generator water hammer exist during iormal operating situations. Water hammer would be most likely
-during startup, shutdown, and low power operation since feedring uncovery events are frequent due to manual feedwater control and feedwater flow requirements are insufficient to keep the feedrings full of water. However, the available water hammer experience reveals that this facility has not experienced steam generator water hammer although other types of water hammer have occurred in the feedwater system.
10
V.
CONCLUSIONS AND RECOMMENDATIONS We have reviewed the operating history of San Onofre Unit No. 1 oertinent to steam generator water hammer and the related operational and procedural characteristics of the feedwater system. This review has shown that conditions conducive to steam generator water hammer have been encountered during normal operating transients and startup and shutdown ooerations. However, steam generator water hammer has not occurred at this facility although a few feedwater system water hammer events have occurred. It is concluded that the potential for steam generator water hamner is sufficiently low to permit continued ooeration of this facility.
11I
VI.
REFERENCES
- 1. K. P. Baskin, Southern California Edison Company (SCEC), letter to R. A. Purole, NRC, Subject - "Secondary System Fluid Flow Instability", July 14, 1975.
- 2. K. P. Baskin, SCEC, letter to A. Schwencer, NRC, Subject "Susceptibility to Steam Generator Feedwater Line Waterhammer Events", December 27, 1977.
- 3. K. P. Baskin, SCEC, letter to D. L. Ziemann, NRC, Subject "Information Requested on Feedwater Lines", July 3, 1979.
- 4. J. G. Haynes, SCEC, letter to 0. L. Ziemann, NRC, Subject "Information Requested Concerning Steam Generator Water Hammer",
August 31, 1979.
- 5. K. P. Baskin, SCEC, letter to D. L. Ziemann, NRC, Subject "Steam Generator Water Hammer", February 14, 1980.
- 6. Final Engineering Reoort and Safety Analysis, San Onofre Nuclear Generating Station (Unit No. 1), SCEC and San Diego Gas and Electric Company, NRC Docket No. 50-206.
- 7. J. A. Block, et al, An Evaluation of PWR Steam Generator Water Hammer, Creare, Inc. NUREG-0291 (December 1976).
- 8. W. E. Bennett, Waterhammer in Steam Generator Feedwater Lines, Westinghouse Technical Bulletin, NSD-TB-75-7 (June 10, 1975).
12