ML20236K413
| ML20236K413 | |
| Person / Time | |
|---|---|
| Site: | Calvert Cliffs  |
| Issue date: | 11/02/1987 |
| From: | Lemons J BALTIMORE GAS & ELECTRIC CO. |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| References | |
| NUDOCS 8711090219 | |
| Download: ML20236K413 (5) | |
Text
- - -___ --.._-
i t-
.y BALTIM ORE i
GAS AND l
ELECTRIC CHARLES CENTER P. O. BOX 1475
- BALTIMORE, MARYLAND 21203 I
l JAMES R. LEMONS l
uu.Acen NUCLEAM OPERATIONS DEPARTMENT
. November 2, 1987 1
i U. S. Nuclear Regulatory Commission j
Washington, DC 20555 ATTENTION:
Document Control Desk l
l
SUBJECT:
Calvert Cliffs Nuclear Power Plant j
Unit No. 1; Docket No. 50-317 Justification for Operability: Number 12 Auxiliary Feedwater Pumo i
REFERENCES:
(a) Teleconference between Mr. J. T. Wiggins.(NRC) and j
Mr. J. R. Lemons (BG&E), on October 30, 1987 l
l Gentlemen.
This letter is in response to specific concerns addressed during Reference (a). You expressed concern about the operability of' No.
12 Auxiliary Feedwater (AFW) pump. and our basis for that determination. This letter-justifies the operability of No. 12 AFW pump, describes the action to be taken in the interim to enhance its reliability, and our commitments to improve the reliability of No. 12 AFW pump. provides a summary of the chronology of testing conducted to date, and the results achieved during those tests. The following problems have been corrected and/or identified. The governor valve linkage has been adjusted and aligned and the governor's hydraulic system flushed. The cam plate and linkage plates of the governor valve show wear. This wear has i
affected the clearances and appears to contribute to the ' governor's control oscillations. Excessive condensation in the steam lines' also appears to have contributed to the. governor's control oscillations and pump overspeed trips.
'We believe the two underlying problems which caused the observed performance are water condensation in the steam lines to the turbine-driven AFW pumps, and-the degraded clearances on.the Sovernor valve. linkage ' due to wear. 'A Field
' Representative from Dresser-Rand has been assisting in troubleshooting and -
repair activities to date. ' His ; analysis ' of the - situation concurs with the above conclusions, and is supported by data available in-the-Nuclear Plant Reliability Data System,'and. discussions with both MPR' Associates'and Bechtel representatives. To address these two problems, we arefordering or machining.
the necessary linkage replacement parts (Heim Spherical. Bearing, Servo-lever,
~
and Cam Crank). We also plan;to' add an additional; manual drain line.
p 8711090219 871102 J
PDR. ADOCK 05000317"-
1 I-
.p PDR l-G__-_ _
i l
l I
f y
Documsnt Control Dask
)
November 2, 1987 I
Page 2
]
Tests have shown that the water condensation and governor hunting can be minimized by windmilling the turbine and periodic dreining of the steam lines.
Based on these test results, the Plant Operations and Safety Review Committee has determined that No. 12 AW pump will respond as described in the safety analysis and is, therefore, OPERABLE.
Number 12 AFW pump is currently windmilling at between 1000 and 1200 rpm to ensure it is properly warmed and ready for startup and automatic flow initiation if needed. Number 11 AW pump has been placed in standby status, ready to be placed in service by an operator if needed. Number 13 AW pump is currently aligned for automatic flow initiation. In addition, No. 23 AFW pump is available through the cross-connect line from Unit Two.
The following actions are being taken in the interim to enhance the reliability of No. 12 AW pump. First, we are conducting daily starts of the pump. These tests simulate an automatic demand of No. 12 AW pump without operator action to control speed, etc.'All valves are aligned in their normal position with the exception of the pump's discharge valve, which is shut to prevent feeding the steam generator. These daily tests have been satisfactory and the results are being communicated to the NRC Regional Office.
Second, each shift has been assigned an additional operator above the minimum staffing requirements. This operator will monitor the performance of the pump during tests and transient conditions resulting in AW system demands. These operators are qualified and trained to operate the AW system. Third, procedures have been established for this dedicated operator.
These procedures specify how to correct an overspeed condition or governor hunting problem on the AW pump. Should an AW pump demand be' received, guidance on the response time required for the operator has been provided. Dedicated communications between the control room and the dedicated operator are provided via headset in the AW pump room.
safety assessment, using the process outlined in We have also completed a 10 CFR 50. 59. This assessment verified that the current AW pump configura-tion is not an unreviewed safety question.
As we communicated in Reference (a), we are not satisfied with the current performance of No. 12 AW pump, although we feel it is still OPERABLE, and will aggressively pursue resolution. We informed you that if the AW pump reliability ha, not been sufficiently improved by Friday, November 6,1987, at 0733, we plan to initiate an orderly reactor SHUTDOWN of Unit 1 in accordance with Technical Specification 3.7.1.2 ACTION STATEMENT a.2. (b). We will reinain SHUTDOWN until the situation has been improved to our satisfaction.
i
[.
n Documsnt Control ~Dask
' November 2,~1987.
Page 3
]
1 J
i Should you. have any questions regarding this matter,. we wil1~be pleased to
.i discuss them with you.
Very truly yours,-
/
/
I I
JRL/LES/LSL/ dim Attachment
- _1 cc:
D.'A.
Brune,-Esquire J. E. Silberg, Esquire R. A. Capra, NRC-S.'A.'McNeil, NRC W. T. Russell, NRC' T. Foley/D. C. Trimble, rTRC J. T. Wiggins, NRC l
l
(
1 J
.o
. r l
~
au
ATTACHMENT _l u
)
l NUMBER 12 AW PUMP CHRONOLOGY On Friday, October 23, 1987, at 0733, STP 0-9A-1 (AFAS Monthly Lo6 c Test) 1 was completed. Number 12 AW pump trip throttle valve tripped on.overspeed.
The turbine was reset and started successfully with some governor oscillation.
At 1700, No.
12 AW pump turbine 'was started successfully with no trip.
However, the governor was still hunting'.
j On Saturday, October 24, 1987, at 0500, No. 12 AW pump turbine was -started from cold condition and tripped on overspeed. The oil level in the governor was found to be low, and oil was added. The turbine was started successfully, however, the governor still hunted.-T; governor needle valve was adjusted to vent the pump and stabilize the
,overnor.
Governor oscillations were dampened.
On Sunday, October 25, 1967, at 1300, No. 12 AW pump was started from a cold condition and tripped on overspeed. At 1700, the pump was started and again tripped on overspeed. However, we were able to manually start the pump successfully.
On Monday, October 26, 1987, at 0900, a cold start of No. 12 AW pump turbine l
was attempted. The turbine tripped on overspeed. The turbine governor was replaced. A cold start was attempted and the turbine tripped again. We per-formed a slow start and, although the governor hunted, it could be steadied with an adjustment of the governor needle valve. Three turbine starts each exhibited governor hunting which could be dampened. A field representative from Dresser-Rand was called in to assist with troubleshooting.
l On Tuesday, October 27, 1987, at 0800, t.he governor valve linkage was inspected. Wear and excessive play were discovered. A cold start was I
attempted with another trip. The turbine casing relief valve lifted and emitted water. It was suspected that condensate could be contributing to the l
overspeed trip. The turbine was reset and started successfully. However, excessive hunting occurred again.
The oil in the governor was changed out, since it appeared to be old or incorrect oil (dark and heavy). A sample was obtained for analysis. The l
turbine was started locally and governor hunting was still present. Air l
pressure signals to the governor from 1C04 and 1C43 were measured and found to be correct, on Wednesday, October 28, 1987, at 0100, the original governor on the turbine
~
was replaced. The governor was flushed and the oil was replaced. The turbine was started from a cold condition and again. the governor exhibited hunting.
The governor stem and actuator clearances were measured and determined to be incorrect. The governor linkages were re-aligned and the turbine started with no hunting and proper governor response.
A fast start was conducted successfully.
We determined that the source of water in the lines to No. 12 AW pump turbine was steam condensation collecting at the inlet. isolation of No. 12 AW ' pump turbine. This was. due to steam leaking by either 1-CV-4070 or 4071. We developed a strategy to drain the steam lines and to bring up No. 12 AW pump.
ATTACHMENT (1)
NUMBER 12 AW PUMP CHRONOIDGY as the standby pump. At 1230 and 2000, we drained the steam lines to the pump turbine and conducted cold starts successfully.
On Thursday, October 29, 1987, at 1230, a cold start of No. 12 AW pump from the automatic line-up was performed after draining lines upstream of th-trip throttle valve and 1-MS-107 in the AW pump room. The pump ran well for a short period and than an apparent perturbation (which appeared to be initiated by condensate) caused the governor to hunt. The pump was shutdown, restarted, and ran properly.
On Friday, October 30, 1987, at 0115, we attempted to start No. 12 AW pump with it lined up for automatic actuation (on line) and flow initiation to the steam generators. When CV-4070 opened, the pump started and the governor controlled speed at approximately 4700 rpm. Water was emitted from governor valve bushing area, turbine seals. and the turbine casing relief. The pump tripped on overspeed (at 4900 rpm by a hand-held tachometer digital readout) within a few seconds. Prior to the trip, governor control while running was stable and no signs of hunting were evident.
l At 0230, the pump was rolled at 500-600 rpm to ensure water was drained from j
the steam lines. At 0400, we started No. 12 AW pump as the on-line pump to j
supply both steam generators. The pump came up to 4000 rpm, controlled smoothly for a few seconds then began huntiag. The pump did not trip. We continued to roll the pump at approximately 1000 rpm via the warm-up bypass.
l At 0600, we started No. 12 AW pump to supply both steam generators. The pump came up to 4000 rpm approximately five seconds after speed stabilized, it began hunting. Shutting the blocking valves only changed the amplitude of the I
hunting. Throttling down on the trip throttle valve enabled the governor to stabilize control. Smooth operation continued through normal cycling of the remote governor controller even after the trip throttle valve was fully re-opened. Additional attempts to create upset conditions by changing pump speed while in recirculation caused no oscillations.
At 0630, we placed No. 12 AW pump in a vindmilling condition (at about 900 to 1000 rpm) as the on-line pump, with No. 11 AW pump in standby. At 0800, No.
12 AW pump was automatically started successfully. The pump smoothly came up to speed on the governor and exhibited no oscillation. The evaluation was that the condensate and moisture had finally been minimized after approximately six hoen of warming and continuous flow. At 1030, we performed a satisfactory autome::ic start of No. 12 AW pump. It came up to speed with no oscillation and ran properly.
Procedures were revised to establish operator actions to run with No. 12 AW pump as the on-line pump spinning at 1000 - 1200 rpm. Also, procedures were established for operators to start the standby pump.
Number 12 AW pump was automat!.cally started successfully on October 31, 1987, and November 1, 1987, from the windmilling condition, j _ - - - - - - - - -
-