ML20133F704
| ML20133F704 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 01/12/1997 |
| From: | Rainsberry J SOUTHERN CALIFORNIA EDISON CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9701140331 | |
| Download: ML20133F704 (6) | |
Text
.
g i s_, _,,,_
JE: EDISON An LD/ SON IVII R V47/OX11. Cunipany January 12, 1997 a
U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, D. C. 20555 Gentlemen:
Subject:
Docket Nos. 50-361 and 50-362 Diesel Generator Testing San Onofre Nuclear Generating Station, Units 2 and 3 This letter addresses Technical Specification 3.8.1, "AC Sources - Operating,"
Surveillance Requirement (SR) 3.8.1.9 of the San Onofre Nuclear Generating Station (SONGS), Units 2 and 3 Technical Specifications. Specifically, this letter discusses the interpretation of this SR and how the current diesel generator (DG) testing satisfies this SR.
The purpose of SR 3.8.1.9 is to demonstrate the DG load response characteristics and capability to reject the largest single load without i
exceeding predetermined voltage and frequency and while maintaining a specified margin to the overspeed trip. The wording of this SR was changed during the preparation of the improved technical specifications for SONGS 2 and 3.
The previous SR merely said " Verifying the generator capability to reject a load...."
This was changed in the new SR to read
" Verify each DG, when operating with design basis kW loading and maximum kVAR loading permitted during testing, rejects a load...."
Edison had a specific concern with including conditions related to power factor as included in the standard specifications. The words in the standard TS Bases imply that the single load rejection test should be performed at maximum design basis kVAR, i0 which, at SONGS, may not be achievable during all possible test conditions.
Although our concern primarily related to the full load rejection test
[M (SR 3.8.1.10), the wording change was also made to the single load rejection test (SR 3.8.1.9). However, the change to 3.8.1.9 inadvertently created a problem with interpretation of the new words regarding the test requirements for the single load rejection test.
Currently, the single load rejection test is conducted at SONGS 2 and 3 using the offsite power system to simulate the rejected load. The test methodology 9701140331 970112 PDR ADOCK 05000361 P
PDR P. O.Ikn 128 San Clemente. CA 92tJ4 0128
g Document Control Desk requires the DG to operate in the droop mode in parallel with the offsite power source. Manually increasing the DG load while synchronized to offsite power requires increasing the speed demand setpoint. The DG is connected to the 4 kV bus with the bus isolated from offsite power. The existing kW loading i
on the bus is recorded.
The bus is then reconnected to offsite power and the DG load is increased to a load equal to the sum of the recorded load plus a load greater than the largest single load.
To ensure that a rejectable load greater than the SR-required 682 kW for the largest single load is available, the test procedure requires an increase of 700 to 750 kW over the recorded l
load. The DG output is then raised to 530 - 560 kVAR. At this point the DG is ready for the test. The breaker to the offsite power system is opened to simulate rejection of the largest single load, with the DG continuing to supply the existing loads on the bus.
Voltage and frequency are recorded and compared to the SR acceptance criteria.
The full load rejection test at SONGS 2 and 3 is also conducted in the droop mode by using the offsite power system to simulate full load.
In this case the DG is connected to the 4 kV bus and synchronized to the offsite power system. The DG is then loaded up to 4450 kW to 4700 kW.
The DG excitation is then increased until one of the following is reached: (1) 3000 to 3200 kVAR, or (2) an exciter field current of 4 amps DC, or (3) 4450 to 4550 Volts on the 4 kV bus.
The DG output breaker is then opened resulting in full load rej ection. DG response is verified to be within the acceptance criteria.
The single load rejection test (SR 3.8.1.9) has three separate acceptance criteria. Two of these criteria (a and c) relate to frequency response which is primarily affected by kW loading. The other criterion (b) is related to voltage response which is primarily affected by kVAR loading.
Each of these three criteria is discussed separately below relative to the interpretation of this SR and the acceptable methods for satisfying this SR at SONGS 2 and 3.
l SR 3.8.1.9.a - Frequency Less Than 66.75 Hz.
" Operating with design basis kW loading" refers to a loading greater than or equal to the kW of the largest single load (682 kW), with rejectable kW greater than or equal to the largest single load (682 kW).
" Maximum KVAR permitted during testing" refers to a kVAR loading greater than or equal to the KVAR associated with the largest single load (369 kVAR) but less than or equal to 3200 KVAR subject to the procedural restrictions that ensure electrical system design limits are not exceeded (i.e., excitation current less than or equal to 4 amps DC, or 4 kV bus voltage less than or equal to 4550 Volts).
Since KVAR has negligible effect on frequency response
Document Control Desk there is no minimum requirement for rejectable KVAR.
At SONGS 2 and 3 this SR acceptance criterion may be satisfied by either the Full Load Rejection Test or the Single Load Rejection Test in Procedure S023-3-3.23.1, " Diesel Generator Refueling Interval Tests," provided the above conditions are met during the test.
SR 3.8.1.9.b - Voltage Within 3294 V to 4796 V Within 4 Seconds
" Operating with design basis kW loading" refers to a loading greater than or equal to the kW of the largest single load (682 kW), with rejectable kW greater than or equal to the largest single load (682 kW).
" Maximum kVAR permitted during testing" refers to a kVAR loading greater than or equal to the kVAR associated with the largest single load (369 kVAR) but less than or equal to 3200 kVAR subject to the procedural restrictions that ensure electrical system design limits are not exceeded (i.e., excitation current less than or equal to 4 amps DC, or 4 kV bus voltage less than or equal to 4550 Volts), with rejectable kVAR greater than or equal to the largest single load (369 kVAR.).
At SONGS 2 and 3 this SR acceptance criterion may be satisfied by either the Full Load Rejection Test or the Single Load Rejection Test in Procedure S023-3-3.23.1, provided the above conditions are met during the test.
SR 3.8.1.9.c - Frequency Within 58.8 Hz to 61.2 Hz Within 4 Seconds
" Operating with design basis kW loading" refers to a loading greater than or equal to the kW of the largest single load (682 kW) but less than the kW load that corresponds to an electric governor speed setting of 61.2 Hz (approximately 1700 kW), with rejectable kW greater than or equal to the largest single load (682 kW).
(Note: When using offsite power to simulate the largest single load with the DG in the droop mode, a valid test cannot be performed if the kW loaded onto the DG results in the governor speed demand setting higher that the 61.2 Hertz acceptance criterion in the SR.)
" Maximum kVAR permitted during testing" refers to a kVAR loading greater than or equal to the kVAR associated with the largest single load (369 kVAR) but less than or equal to 3200 kVAR subject to the procedural restrictions that ensure electrical system design limits are not exceeded (i.e., excitation current less than or equal to 4 amps DC, or 4 kV bus voltage less than or equal to 4550 Volts).
Since kVAR has negligible effect on frequency response there -is no minimum requirement for rejectable kVAR.
Document Control Desk At SONGS 2 and 3 this SR acceptance criterion is satisfied by the Single Load Rejection Test in Procedure S023-3-3.23.1, provided the above conditions are met during the test.
The results of the most recent surveillance testing of each of the DGs relevant to SR 3.8.1.9 is as follows:
Single load Rejection Test Results 2G002 2G003 3G002 3G003 Date of Test 1/6/97 12/16/96 8/25/95 8/10/95 kW Loading 710 + 190 740 + 490 710 + 950 740 + 170 (rejectable + existing) kVAR Loading 530 to 560* 530 to 560* 530 to 560* 530 to 560*
kW Rejected 710 740 710 740-Estimated kVAR 413 to 443 220 to 250 -33 to -3 425 to 455 Rejected #
Min /MaxFrequency60.8-61 60.3 - 60.7 60.05-61.13 59.97 - 60.5 Min /MaxVoltage 4460 - 4480 4461 - 4538 4458 - 4514 4429 - 4504 Full load Rejection Test Relevant t.o SR 3.8.1.9 2G002 2G003 3G002 3G003 Date of Test 1/4/97 12/17/96 1/12/97 1/12/97 kW Loading 4450-4700* 4450-4700* 4530 4550 kVAR Loading 532000 532000 532000 532000 kW Rejected 4450-4700* 4450-4700" 4530 4550 kVAR Rejected 532000 53200@
53200@
532000 Max Frequency 63.66 63.25 62.59 63.09 Voltage at 4 Sec 4465 4451 4536 4559
Document Control Desk Notes:
- The procedure requires adjustment to this loading but does not require the specific value to be recorded.
- The kVAR rejected is not recorded in the test procedure.
This value was estimated based on a power factor of.85, which is representative of the SONGS buses during testing.
O The kVAR loading of the DG during this test is obtained by increasing excitation until one of the following is reached: (1) 3000 to 3200 kVAR or (2) excitation current of 4 amps DC, or (3) 4450 to 4550 Volts on the 4 kV bus.
For the full load rejection tests summarized above, not only were the acceptance criteria for the full load test (SR 3.8.1.10) satisfied, but also acceptance criteria a and b for the single load rejection test (SR 3.8.1.9) were satisfied.
For the single load rejection tests summarized above, all of the acceptance criteria of SR 3.8.1.9 were also satisfied.
However, as discussed above, not all of the single load rejection tests represent a rigorous test of the kVAR rejection capability of the DG.
That is because, even though the SR-specified number of kW was rejected during the test, the kVAR rejected was less than the equivalent kVAR of the single largest load for two of the four tests (2G003 and 3G002).
However, the full load rejection tests do provide an. appropriate level of kVAR rejection and therefore represent a satisfactory test of the DG voltage response.
In summary, although the wording in the current single load rejection test, SR 3.8.1.9, is ambiguous, it should be interpreted in the context of the surveillance testing performed at SONGS as discussed above.
Accordingly, the acceptance criteria of SR 3.8.1.9 are satisfied for each DG at SONGS by either the single load rejection test or the full load rejection test.
As a result of this review of SR 3.8.1.9, we have concluded that to avoid confusion on this issue in the future it is appropriate to revise the SR to remove all ambiguity. We will submit a requested change to this SR for your approval by January 24, 1997.
Document Control Desk If you have any questions regarding this information, please let us know.
Sincerely,
\\
\\
- [i J
J.
. Rainsbery Manager, Plant Licensing Enclosure cc:
L. J. Callan, Regional Administrator, NRC Region IV K. E. Perkins, Jr., Director, Walnut Creek Field Office, NRC Region IV J. A. Sloan, NRC Senior Resident Inspector, San Onofre Units 2 & 3 M. B. Fields, NRC Project Manager, San Onofre Units 2 and 3
_