ML17310B480
| ML17310B480 | |
| Person / Time | |
|---|---|
| Site: | Palo Verde  |
| Issue date: | 07/21/1994 |
| From: | Stewart W ARIZONA PUBLIC SERVICE CO. (FORMERLY ARIZONA NUCLEAR |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| 102-03044-WLS-J, 102-3044-WLS-J, NUDOCS 9408020111 | |
| Download: ML17310B480 (19) | |
Text
P R.IC3R.IMY'
~(ACCELERATED RIDS PROCESSING)~
REGULA~ INFORMATION DISTRIBUTIO~YSTEM (RIDS)
I'C("ESNIOh1
/BR:9408020111 DOC.DATE: 94/07/21 NOTARIZED:
NO DOCKET I FACIL:STN-50-528 Palo Verde Nuclear Station, Unit 1, Arizona Publi 05000528 STN-50-529 Palo Verde Nuclear Station, Unit 2, Arizona Publi 05000529 STN-50-530 Palo Verde Nuclear Station, Unit 3, Arizona Publi 05000530 AUTH.NAME AUTHOR A'FFILIATION STEWART,W.L.
Arizona Public Service. Co.
(formerly Arizona Nuclear Power RECIP.NAME RECIPIENT AFFILIATION P
Document Control Branch (Document Control Desk)
R
SUBJECT:
Forwards response t6 request for addi info re 930823 proposed amend to licenses, removing license condition from I
units 1
6 3
6 confirmatory order from unit 2.
DISTRIBUTION CODE:
A001D COPIES RECEIVED:LTR ENCL SIZE:
0 TITLE: OR Submittal:
General Distribution NOTES:STANDARDIZED PLANT Standardized plant.
Standardized plant,.
05000528 05000529 05000530 RECIPIENT ID CODE/NAME PDIV-3 LA HOLIANI B INTERNAL: ACRS NRR/DORS/ONDD NRR/DRPW NRR/DSSA/SRXB OC- ~CB REG F~
01 EXTERNAL: NRC PDR COP1ES LTTR ENCL 1
1 2
2 6
6 1,
1 1
1 1
1 1
0 1
1 1
1 RECIPIENT ID CODE/NAME PDIV-3 PD TRAN,L NRR/DE/EELB NRR/DRCH/HICB NRR/DSSA/SPLB NUDOCS-ABSTRACT OGC/HDS2 NSIC COPIES LTTR ENCL 1
1 2
2 1
1 1
1 1
1 1
1 1
0 1
1 D
0 C
U N
NOTE TO ALL'RIDS'ECIPIENTS:
PLEASE HELP US TO REDUCE iVASTE!CONTACTTHE DOCUMENTCONTROL DESK, ROOM PI-37 (EXT. 504-2083 ) TO ELIMINATEYOUR NAMEFROM DISTRIBUTIONLISTS I'OR DOCUMENTS YOU DON'T NEEDI TOTAL NUMBER OF COP1ES REQUIRED:
LTTR 24 ENCL 22
I' C
('/a I
l j
Il II I ~
Arizona Public Service Company P.O. BOX 53999
~
PHOENIX, ARIZONA85022-3999 WILLIAML. STEWART EXECUTIVEVICEPRESIDENT NUCLEAR 102-03044-WLS/J RP July 21, 1994 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Mail Station P1-37 Washington, D.C. 20555
Reference:
1.
Letter 102-02617, dated August 23, 1993, from W. F. Conway, Executive Vice President, Nuclear, APS, to USNRC 2.
Letter 161-01038, dated May 20, 1988, from E. E. VanBrunt, Jr.,
Executive Vice President, Nuclear, APS, to USNRC
Dear Sirs:
Subject:
Palo Verde Nuclear Generating Station (PVNGS)
Units 1, 2, and 3 Docket Nos. STN 50-628/529/630
Response
to Request for Additional Information File: 94-066-026 This letter is being transmitted to provide additional information which was requested during a conference call with the NRC on May 17, 1994 regarding a proposed amendment to the operating license (Reference 1).
The proposed amendment would remove the License Condition from Units 1 and 3, and the Confirmatory Order Modifying the License for Unit 2. The License Condition and Confirmatory Order required Arizona Public Service Company (APS) to implement an augmented vibration monitoring program for each of the'four reactor coolant pumps in Units 1, 2, and 3.
Enclosed with this letter, please find the NRC questions restated and the APS response.
Should you have any questions, please contact Richard A. Bernier at (602) 393-5882.
Sincerely, WLS/JRP/did A~ay~
9408020iii 94072i PDR ADQCK 05000528 P
'Q
US Nuclear Regulatory Commission ATTN: Document Control Desk
Response
to Request for Additional Information Page 2 Enclosure cc:
K. E. Perkins L. J. Callan K. E. Johnston B. E. Holian A. V. Godwin (ARRA)
ENCLOSURE UNITS 1, 2, AND 3 RESPONSE TO REQUEST FOR ADDITIONALINFORMATION PROPOSED AMENDMENTTO OPERATING LICENSE
NRC QUESTION 1'oes APS consider that the proposed alternatives of continued monitoring and evaluation of the RCP's vibration using the installed monitoring systems would provide an acceptable level of quality and safety equivalent to the existing License Condition and Confirmatory Order-requirements?
Are the limits and required actions specified in the Confirmatory Order comparable with the vector acceptance region limits and required actions?
Please provide a discussion.
APS'ESPONSE:
The proposed alternative of using the newly installed computer system provides better quality and safety than the existing License Condition and Confirmatory Order requirements.
These improvements are:
Monitoring the 1xRPM and 2xRPM component every few seconds versus once per day.
Monitoring the phase of the 1xRPM and 2xRPM component, which is not required by the existing License Condition and Confirmatory Order.
Monitoring decreases in 1xRPM and 2xRPM Abilityto quickly access trend data for faster evaluation.
Because of the different designs of the permanently installed system and the portable data collectors used to meet the requirements of the License Condition and Confirmatory Order, the measurements are made differently, and the alarm limits are calculated differently. The data collector takes averaged data, whereas the permanently installed system takes a more instantaneous value.
For this reason, the values measured by a permanently installed system have more "scatter" or randomness than the data collector values.
The alarm values must take this into.account to avoid generating nuisance alarms.
The data collector alarms are 1.6 x the baseline and the average
+3 x the standard deviation.
The maximum alarms set into the permanent system are calculated as the median of several (up to 21) days of data +1.5 x the difference between the maximum and minimum readings seen during the baseline period.
The following table compares the alarm values for the data collector and the installed monitoring system.
l 1
Page 1 of 5
TABLE 1: COMPARISON OF UNIT 1, 1xRPM ALARMVALUES DATACOLLECTOR ALARMS Monitoring System Alarms Pump Probe 1A-X 1A-Y 1B-X 1B-Y 2A-X
~ 2A-Y 2B - X 2B-Y 1.6 x Baseline 3.41 1 2.774 2.970 2.458 2.421 2.315 1.272 1.21 1 Average +3 Sigma 2.292 1.832 1.910 1.612 1.585 1.699 0.894 0.855 Median 1.5 x Range 4.7 4.1 3.7 3.7 3.6 3.6 2.3 2.8 As can be seen in the table, the data collector alarms are very low. This causes many false alarms which must be manually evaluated.
The monitoring system alarms are more realistic, although very low in comparison with most other plants.
The existing License Condition and Confirmatory Order requires NRC notification and pump and plant shutdown at specified conditions.
The vector monitoring system does not require any specific actions, instead good engineering judgement would be used to determine the appropriate actions.
The new computer monitoring system provides for better quality and safety than the existing requirements in the Unit's Operating License and far exceeds the original commitment to monitor and record the vibration data, on each of the four reactor coolant pumps every four hours.
In the future, if it becomes necessary to replace the computer monitoring system, the replacement unit will, at a minimum, meet the requirements to monitor and record the vibration data every four hours.
NRC QUESTION 2:
APS states that since replacement of the original shafts in the three units, no cracks have been detected by UT examination during past refueling outages.
Further, APS reports the replacement shafts were provided with a center bore that allows the UT probe to inspect nearly the entire length of the shaft.
Experience in Europe using center bore UT examination techniques indicates that cracks equivalent to a 0.1 mm deep notch can be reliably detected.
Please provide a discussion of APS'roposed future use of UT examinations, and ifthe areas that cannot be reached via the center bore include any of Page 2 of 5
$ ~
4 4
the areas in which cracks were discovered in similar European RCP shafts, and also include a discussion of the sensitivity and depth of notch the APS UT equipment can reliably detect.
APS'ESPONSE:
Site Engineering intends on performing RCP shaft UT inspections at the following intervals:
The Unit 1 shafts, which are the modified/centerbore design and have chrome plating, will be UT inspected during each refueling outage.
The shafts in Units 2 and 3 (rolled shaft design) will be UT inspected during RCP mechanical seal replacement.
Currently, the seals are replaced during every other outage; however, after PVNGS installs an updated seal design, this replacement will occur every three outages.
~
During any refueling outage as a result of abnormal trends in RCP instrumentation, the trends willbe reviewed on a case-by-case basis to determine ifa UT inspection is warranted.
PVNGS, performs RCP shaft UT inspections during seal replacement, because UT inspections require removal of-all shaft instrumentation, removal of the rigid and Zurn couplings, removal of fire protection equipment, unbolting arid lifting of the main thrust bearing module, and construction of scaffolding walkways and temporary shielding in the RCP pump bays (basically, 3000 man-hour and 10 man-rem evolution) ~ Since" replacement of the RCP seals requires essentially the same initial setup work, PVNGS intends to perform the UT inspections during the RCP seal replacements; and the interval will, therefore, follow RCP seal replacements.
Since replacement of the original RCP shafts in the three PVNGS units, no cracks have been detected by UT examination during each of the units'ast refueling outages (Unit 3 did not UT during U3R4 outage).
The replacement shafts were provided with a 25 mm diameter center bore.
The center bore allows the UT probe to inspect areas where cracks have been found to exist, as well as essentially the entire shaft length.
The scanning sensitivity is at least two times (+6dB) the calibration sensitivity.
The UT cal standard used at PVNGS for shaft keyway examination utilizes the following nominal notch sizes:
Slot No.
1 - 1/2" x 1/8" x 1/1 6" Slot No. 2 - 1/2" x 1/4" x 1/1 6" Slot No. 3 - 1/2" x 3/8" x 1/1 6" (see Figures 1 and 2).
Page 3 of 5
1 l
h
NRC QUESTION 3:
The Confirmatory Order Modification was issued since the root cause of RCP shaft cracking had not been identified.
Has a root cause analysis been performed?
Has the mechanism causing the shaft cracking been identified?
Did the analysis consider the information that cracks in areas where the chrome plating has been removed have been found in similar modified shafts in Europe RCPs?
Please discuss.
APS'ESPONSE:
The root cause analysis was performed and is described in the report attached to letter 161-01 038-EEVB/JRP dated 5/20/88 (Reference 2). The root cause of the shaft cracking was determined to be the reduction in fatigue strength of the shaft material due to the presence/application of chrome plating. Microcracks initiate in the chrome plating during normal operation, due to superimposed mechanical and thermal stresses, and propagate into the base metal by high cycle fatigue.
The replacement shafts installed in Units 2 and 3 are newly designed shafts that have been surface rolled to apply a beneficial, compressive surface stress to the shaft and chromium-carbide coated as opposed to chrome plated.
These design features result in shaft material with significantly improved fatigue strength over the original design.
The replacement shafts installed in Unit 1 are modified, original design shafts. The root cause of the shaft cracking was addressed by removing the chrome plating from around the high 'stressed impeller keyway area and by other modifications to reduce the mechanical and thermal stresses applied to the shaft.
The chrome plating was left in the lower stressed impeller-fit areas.
The root cause analysis was performed prior to the more recent incident of cracking in the non-chrome plated area in a European RCP.
This shaft was not surface rolled; therefore, there was no beneficial compressive surface stress applied to this shaft.
Additionally, the shaft had no coatings at all on'the shaft.
The crack occurred in the upper impeller fit area and has been attributed to vibration wear*at the top edge of the impeller fit area.
This more recent crack incident has no impact on the root cause analysis previously performed by APS.
The original PVNGS shafts had insufficient fatigue strength due to the application of chrome plating, and therefore cracked due to fatigue. The more recent crack incident has no impact on the acceptability of the replacement shafts installed in
~ Units 2 and 3.
The Units 2 and 3 shafts are surface rolled and are chromium-carbide coated with a resulting shaft fatigue strength well above the non-coated European shaft.
The only possible applicability of the recent crack is on the modified shafts installed in Unit 1. Although the potential for shaft cracking in Unit 1 cannot, be entirely ruled out; APS has concluded, due to differences between the European shaft design that cracked and the modified Unit 1 shafts, that cracking in the Unit 1 shafts remains unlikely. The crack initiated at a vibration wear site at the top edge of the impeller fit area.
There have Page 4 of 5
44 1
% ~
0 been 20 shaft replacements performed in the three PVNGS units, and there have been no observed incidences of wear (or fretting) in the impeller fit areas.
The single incident of fretting wear was on an original Unit 1 RCP which experienced fretting wear, and a consequential fatigue crack, on the side wall of the keyway. This particular crack and the corrective actions taken to address it are described in detail in the Reference 2 report.
f NRC QUESTION 4:
How does APS propose to assess low amplitude vibration trends and the need for corrective action using the installed system to continue to monitor and evaluate RCP vibration?
Under. the subject License Condition and Confirmatory Order, were the units ever shutdown or consideration to for shutdown that involved unnecessary reasons based on a low amplitude vibration trend?
Please provide discussion with regard to resolution on the issue.
APS'ESPONSE:
As can be seen in Table 1, the monitoring system is sensitive to fairly low amplitude trends.
In addition, the data collector will continue to be used at least once per month, and could detect even lower trends.
However, since the risk of shaft cracks has been greatly reduced, the value of detecting low amplitude trends is also greatly reduced.
Numerous low amplitude trends have been detected.
The pumps inherently shift balance over starts and over long-term operation.
These trends have all been evaluated as not caused by a crack, and so the unit has never been very close to being shutdown.
However, the evaluations are very time consuming.
Once, during startup from a refueling
- outage, Unit 1 experienced instrument spikes, the Unit was placed in IVlode 5 for evaluation.
It was determined that a loose termination caused the instrument spikes whereby the unit was returned to power.
NRC QUESTION 5:
Please provide a discussion with regard'to evaluation of the 21 days of trend data, the, acceptance limits applied, and corrective actions imposed.
APS'ESPONSE:
The trend data is the same data that is compared to the alarms, so the same acceptance limits are applied before the data is stored in the trend files. The evaluation of the trend data is the same as the current trend data evaluation.
On an alarm, the trend data is evaluated for trends indicating a cracked shaft or other vibration problems.
Good engineering judgment would be used to determine the appropriate actions.
In addition to the 21 days of trend data stored in the installed monitoring system, long-term trend data is collected by using data collectors on at least a monthly basis; and the data from the installed monitoring system is captured and stored in a computerized data base for evaluation of long-term trends.
Page 5 of 5
~I 1'(
631 I
DRILL 25 MM.
¹1
-Fas7 12.598 KEYWAY LENGTH C9 ~
DETAIL C CTYP.) ~/
r LOT i
&arrl 8 ~H pSL
¹3 SLOT
¹2 11,024 I-
.472 DRILL.472 X I
SLOT
¹2 l
~3-DIA.
P.EF.-
394 RADIUS IbREF',85813.464 11220 13.937 12.756 20A)0 15,75
- DEG, (TYP.)
4.472 LOCATING DLMENSIQN REF.
DETAIL C MATERIALMILD STEE NOTEr SLOT ¹1. 1/2 X 1/8 X 1/16 SLOT ¹2, 1/2 X 1/4 X 1/io SLOT ¹3, 1/2 X 3/8 X 1/16 Drcwing g,~ r-ah p,pprcvcd Pev
.'I'u 'i 1N8 HALF IZE UT CAL STANDAPD
.FOR RFCIRC SHAFT KEYWAY EXAM PALO VERDE
'r re le'.le X LMT-386 f3JAPS-001 MV9 vr T.G. LAMBERT 15QPR8+
)
1QF1 Figure 1
I
~
~
~
)
~
~ 'l
~ ~
~ ~
~
I
~
~ '
~
~ t '
~
~
~
~ ~
~
0
~
I
~ ~
~A la~~
1,
~