ML13330B565
| ML13330B565 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 10/04/1991 |
| From: | Rosenblum R SOUTHERN CALIFORNIA EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9110100079 | |
| Download: ML13330B565 (7) | |
Text
Southern California Edison Company 23 PARKER STREET IRVINE, CALIFORNIA 92718 R. M. ROSENBLUM TELEPHONE MANAGER OF (714) 454-4505 NUCLEAR REGULATORY AFFAIRS October 4, 1991 U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, D. C. 20555 Gentlemen:
Subject:
Docket No. 50-206 Inservice Testing of Safety Injection Pumps San Onofre Nuclear Generating Station Unit 1 This letter revises and resubmits a previously approved relief request regarding inservice testing (IST) of the SONGS 1 safety injection (SI) pumps.
That relief request was based on an anticipation that modifications allowing substantial flow testing of the SI pumps could be implemented in conjunction with other planned upgrades to this system. However, further review has determined that these modifications are not practical.
We therefore propose to satisfy the IST requirements by adding flowmeters to the SI pumps miniflow lines and testing the -pumps on miniflow.
Background
Section XI of the ASME Code requires that as part of quarterly IST of safety related pumps, the flow rate through the pump be measured and compared with a previously determined reference value. Our two SI pumps, G50A and G50B, cannot presently meet this requirement. The SI system is designed such that the miniflow return lines are the only potential test paths for flow measurement. Since these lines are not instrumented for flow, no flow data can be taken.
Technical position 9 of Generic Letter 89-04 addresses this situation on a generic basis. Position 9 states that where a path exists at cold shutdown or a refueling outage to permit pump testing "under full or substantial flow conditions," this path should be used for flow testing and the increased interval is an acceptable alternative to the quarterly testing required by the ASME Code. In cases where this alternative path is not available, the generic letter requires that flow instrumentation be installed on the pump mini-flow lines. As the result of discussions with the NRC, we evaluated the first option and concluded that the existing plant configuration does not permit full flow or substantial flow testing of the SI pumps at any frequency.
9110100079 911004 PDR ADOCK 05000206 P.
F'DP~R 4
Document Control Desk
-2 However, as part of forthcoming Cycle 12 upgrades to the SI system, we anticipated that changes to the system could be implemented to provide a suitable alternative to miniflow testing.
On October 29, 1990, we submitted PRR No. 10, which presented our evaluation and a proposal to "implement modifications to enable improved flow testing of the SI pumps" by the end of the Cycle 12 refueling outage. In your safety evaluation of PRR No. 10, dated December 19, 1990, you accepted this proposal and granted us interim relief to continue the present testing. You also stated that this interim relief may not be extended.
Additional Considerations and Revision to PRR No. 10 After further review we have determined that the modifications required to achieve substantial flow (by adding new test loops) would be quite extensive, would not provide significant additional benefit, and do not fit in with the planned scope of SI system upgrades.
(For additional details regarding this assessment, see the Enclosure).
By comparison, it is possible to satisfy technical position 9 of Generic Letter 89-04 by installing miniflow measuring instrumentation on the two SI pumps. Although this option is consistent with the generic letter, it represents a deviation from the NRC's safety evaluation dated December 19, 1990, on PRR No. 10.
For this reason, we are submitting a revision to PRR No. 10 for NRC approval.
NRC approval of this revised PRR is requested by January 1, 1992. If approved, the proposed modifications will be implemented during the Cycle 12 refueling outage as required by the January 1990 FTOL order. Quarterly flow measurements will begin using the new instruments by the end of that outage.
If you have any questions, please call me.
Very truly yours, Enclosure cc:
J. B. Martin, Regional Administrator, NRC Region V George Kalman, NRC Senior Project Manager, San Onofre Unit 1 J. 0. Bradfute', NRC Project Manager, San Onofre Unit 1 C. W. Caldwell, NRC Senior Resident Inspector, San Onofre Units 1, 2&3
ENCLOSURE Page 1 of 4 Revision 1 9/91 PUMP RELIEF REQUEST NO. 10 SYSTEM:
Safety Injection (SI) System COMPONENT:
SI Pumps G50A and G50B CLASS:
2 FUNCTION:
To provide low pressure borated water from the refueling water storage tank to the suction of the feedwater pumps, when operating in the SI mode.
TEST REQUIREMENT:
Article IWP-3000 of ASME Code Section XI, requires pump flow rate to be measured quarterly as part of periodic inservice testing during normal plant operation.
BASIS FOR RELIEF:
The only available flow paths for periodic testing of the SI pumps during normal power operation (Modes 1 through 4) are the recirculation or mini-flow lines designed to provide the required minimum flow for pump cooling. Presently, the design does not provide instrumentation to measure this flow. IST data are obtained while the pump operates in a fixed resistance flow path. Generic Letter 89-04 (Reference A) permits use of minimum flow recirculation lines for inservice testing only if an alternative path does not exist.
In cases where only the mini-flow recirculation line is available for pump testing, regardless of the test interval, the generic letter requires flow instrumentation which meets the requirements of IWP-4110 and IWP-4120 to be installed in the mini-flow line. This instrumentation is necessary to provide flow rate measurements during pump testing, which can be evaluated with the measured pump differential pressure to monitor for pump hydraulic degradation.
This relief request proposes to install miniflow instrumentation during the SONGS 1 Cycle 12 refueling outage. This is the result of evaluating the various available test paths as described below.
EVALUATION OF EXISTING CONFIGURATION Three alternate test paths were identified. These can be seen on P&ID 5178115 (copy attached).
Each of these paths was further examined and found to be unacceptable as explained below.
ENCLOSURE Page 2 of 4 Revision 1 9/91 Paths 1 and 2:
The SI pumps take suction from the refueling water storage tank (RWST) using the normal suction line and discharge via the 2" Feedwater Pump bypass line (line: SIS-5012-2-"CL) to either the reactor coolant system (RCS) directly, or back to the RWST via the Safety Injection Pipe Purge Line (SIS-5009-2"-CL).
The lack of appropriate flow instrumentation in both paths offers no advantage over the current test method. In addition, their large length and small diameter may result in less flow than is currently obtained through the mini-flow recirculation line.
Path 3:
The SI pumps take suction from the RWST from the normal suction line and discharge via the Feedwater Pumps to the RCS using the safety injection flow path (through Safety Injection Discharge Valves HV-851A and HV-851B)
- 1.
System flow using the path through the Feedwater Pumps would be approximately 10,000 to 11,000 gpm. The ASME Code requires a minimum pump run time of 5 minutes under stable conditions prior to taking data. This means that 50,000 gallons of water could be injected into the RCS prior to taking any data. The total RCS water volume is approximately 50,500 gallons. There is no practical way of dealing with this excessive volume as there is no return path from the RCS to the RWST that could handle a 10,000 to 11,000 gpm flow.
Testing during Mode 6 (Refueling) with the Reactor Vessel Head removed and filling the refueling cavity during the inservice test has been evaluated. This test path could result in significant airborne and particulate contamination in the containment and could result in costly delay in the refueling operation to allow for decontamination. It would also increase the risk of personnel exposure and contamination.
- 2.
The Feedwater Pump shaft seals are labyrinth seals. Water passes through very narrow clearances between the labyrinth strips and the shaft. As the water passes from one strip to the other, its pressure and velocity are reduced. During normal operation (feeding the steam generators), the shaft seal water is provided by the condensate pumps and flow is directed along the seal length from the outside to the inside of the pump casing. During Safety Injection, normal seal water supply is unavailable and shaft seal water (pressurized RWST water) is provided from inside the Feedwater Pump casing. Since seal flow is reversed, external leakage is expected.
This will spread contaminated RWST water in the immediate area of the Feedwater Pumps.
ENCLOSURE Page 3 of 4 Revision 1 9/91 To prevent contamination, an external source of seal water would be required during this test. Providing this external source would entail significant plant modifications.
These evaluations concluded that with the existing configuration the SI pumps could only be tested on miniflow.
EVALUATION OF MODIFICATIONS TO EXISTING CONFIGURATION As the result of additional discussions with the NRC, we evaluated the possibility of achieving improved flow testing by installing new instrumented test loops bypassing the main feedwater pumps. This path could allow substantial flow testing of the SI pumps on a refueling interval basis.
Based on these discussions, the previous revision to this relief request, which was submitted by Reference C, committed to modifications to enable improved flow testing of the SI pumps. In its safety evaluation, dated December 19, 1990, the NRC approved this relief request. However, after further evaluation we have concluded that the modifications are not warranted.
The SONGS SI pumps are single stage, split case horizontal pumps typical for high flow, low head applications. As seen in the attached pump performance curves, a flow of approximately 7,000 gpm out of a rated flow of 10,500 gpm is required to produce a 10% decrease in pump head. Testing at flows less than 7,000 gpm is estimated to provide approximately the same information as our present method using minimum flow. Testing at 7,000 gpm or more would require a new large diameter test loop for each pump and significant modifications to several systems in the immediate area of the SI pumps and the refueling water storage tank. Contrary to what was previously anticipated, we have now determined that the Cycle 12 SI system upgrades will not modify this part of the SI system. Therefore, the benefits associated with this modification do not justify the hardship that is involved.
Once every quarter we take vibration measurements on the pump shaft bearing in accordance with the ASME Code. By trending these vibration measurements along with other quarterly data required by the Code (pump speed, inlet pressure, differential pressure etc.) it is possible to detect pump degradation.
Additional assurance of pump operability is provided by our preventive maintenance program. The lubricating oil of each pump is sampled and analyzed once each year for the presence of foreign material. Should irregularities be found, the pump is overhauled.
ALTERNATE TESTING:
Based on the above considerations, the following alternate testing, which meets the requirements of position 9 of Generic Letter 89-04, is proposed.
ENCLOSURE Page 4 of 4 Revision 1 9/91 Interim:
Test the SI pumps once every quarter* on minimum flow in accordance with ASME Code Section XI, except for the measurement of flow rate.
Long Term:
By the end of the Cycle 12 refueling outage:
a)
Install flow instrumentation which meets the requirements of IWP-4110 and IWP-4120 of Section XI of the ASME Code in the pump miniflow lines.
b)
Begin quarterly* measurements of minimum flow.
APPROVAL STATUS:
This Pump Relief Request (PRR) complies with the technical positions of Reference A. The schedule for modifications complies with that required by the NRC's safety evaluation (Reference D).
NRC approval of this revised PRR is requested by January 1, 1992.
REFERENCES:
A.
NRC Generic Letter 89-04, Guidance on Developing Acceptable Inservice Testing Programs, April 3, 1989 B.
Letter, J. E. Tatum (NRC) to H. B. Ray (SCE), Inservice Testing Pump Relief Request No. 6, May 18, 1990 and its enclosed Safety Evaluation Report (SER).
C.
Letter, F. R. Nandy, (SCE) to NRC, Inservice Testing of Safety Injection Pumps and Refueling Water Pumps, October 29, 1990.
D.
Letter, L. E. Kokajko (NRC) to H. B. Ray (SCE), IST Program Relief Request PRR-10, December 19, 1990.
- One quarter will be equal to 92 days NRC-SIP.AA2
UGtSLASU 0661 JAwSOSQ a-c~g :ax~j SOUBWJO08d dwnd U0130UI APS~w liodwd sisAmu~v 4mmsu Pnxj Pesupdn I ll 'NOI+/-VlS 0NLLV3N3V tfVBOnN MldONO NVS TOTAL HEAD (FT)
EFFICIENCY (%)
00 NPSH (FT)