ML20113C279
| ML20113C279 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 06/21/1996 |
| From: | Jain S DUQUESNE LIGHT CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9607010097 | |
| Download: ML20113C279 (5) | |
Text
...
Be wer VaNey Power Station Shg pingport. PA 15077 0004 SUSHIL C. JAIN (412) 393-5512 DMs6on Vice President Fax (412) 643 8069 DMalon U. S. Nuclear Regulatory Commission June 21, 1996
' Attention: Document Control Desk Washington, DC 20555-0001
Subject:
Beaver Valley Power Station, Unit No.1 and No. 2 BV-1 Docket No. 50-334, License No. DPR-66 i
BV-2 Docket No. 50-412, License No. NPF-73 Response to Request for Additional Information Concerning WCAP-14535; Revised Item 2 Response Attached is a revised response to Item 2 to an NRC staff request for additional information provided by letter dated May 1,1996, concerning WCAP-14535, " Topical Report on Reactor Coolant Pump Flywheel Inspection Elimination." Beaver Valley submitted the subject report by letter dated January 24,1996, as the industry's lead plant on this issue, and submitted a response to the request for additional information on June 14,1996. On June 18, 1996, a teleconference between the NRC staff reviewers, Westinghouse Electric Corporation staff, and members of the Beaver Valley staff discussed the June 14,1996, submittal, in particular Item 2.
This revised response to Item 2 is intended to clarify the interaction and impact of adding the stresses associated with a conservative shrink fit and flaw sizing conservatism associated typically with Section XI acceptance criteria of the ASME. Boiler and Pressure Vessel Code. The response to Item 2 and Table I has been revised to reflect additional conservatism in the allowable crack lengths for reactor coolant pump flywheels.
Please direct questions regarding this submittal to Mr. Roy K. Brosi at (412) 393-5210.
Sincerely, DOKOh00h334 M
0 PDR Sushil C. Jain 0 U A LIT Y 010005 g g7q e
iiiy
1 Beaver Valley Power Station, Unit No.1 and No. 2 Resp,onse to Request for AdditionalInformation Concerning WCAP-14535; Revised Item 2 Response Page 2 c:
w/ enclosure:
Mr. L. W. Rossbach, Sr. Resident Inspector Mr. T. T. Martin, NRC Region I Administrator Mr. D. S. Brinkman, Sr. Project Manager - (3 copies)
Ms. Diane Jackson, Westinghouse Electric Corporation l
w/o enclosure:
Mr. David Haile, South Carolina Electric and Gas Co.
Mr. Pat Naughton, Virginia Power Mr. Don Gulling, Florida Power Corporation l
Mr. Ben Mays, TU Electric Co.
l Mr. Jim Edwards, Georgia Power Co.
Mr. K. J. Voytell, Westinghouse Electric Corporation Mr. S. A. Binger, Jr., Westinghouse Electric Corporation l
l 1
I.
f 1
i
Response to NRC Reauest for Additional Information on WCAP-14535 (Revised Response to Item 2)
Item 2:
Section 1.1 Previous FlywheelIntegrity Evaluations, Page 1 Thefatigue analysis is dependant on thepremise that UT equipment usedfor examinations ofRCPflywheels at thesefacilities is capable of accurately detecting and sizing 0.24 inch long near surfaceflaw. Provideyour basis supporting the probability ofdetection (POD)for the examinations performed Provide details on how the POD values were determined, qualified, and used in concluding the assumed size of the initialflaw.
Response to Item 2:
The initial crack length of 0.24 inch was used in a previous evaluation of RCP flywheel integrity by Babcock and Wilcox (Report BAW-10040, December 1973, " Reactor Coolant Pump Assembly Overspeed Analysis"). This length was assumed to be the largest crack that could be missed in nondestructive testing.
As seen in Table 4-4 of WCAP-14535, crack growth assuming extremely large initial flaw lengths (from 2.04 to 3.28 inches) was found to be insignificantly small over a 60 year extended plant life. In the crack growth evaluation,6000 RCP start /stop cycles were assumed, which is conservative with respect to actual operation. Shrink fit was not included in the WCAP-14535 evaluation. This is conservative, since shrink lit retards crack growth, as discussed in the Response to item 1, above.
This evaluation suggests that very large initial flaws can be structurally tolerated, from a crack growth perspective. As noted later in this discussion, the reflective reference area used for calibration of the j
inspection procedure is nearly an order of magnitude smaller than these structurally stable flaws.
l An alternative method of evaluating this issue is to define an " allowable" flaw size based on the application of margins to the calculated critical flaw size and the calculated stress intensity factor.
The approach used here is to apply the Code pressure boundary margins of ASME Section XI to the flywheel, which is a non-pressure boundary, non-Code component. This application is considered to be extremely conservative. The Section XI criteria are as folicws:
I Criteria based on flaw size:
a,iio, 0.1 a itical (Normal, Upset and Test Conditions)
)
cr a,iiow 0.5 a itical (Emergency and Faulted Conditions) j cr Criteria based on stress intensity factor:
Kj Toughness /10 (Normal, Upset and Test Conditions)
Kg Toughness / 2 (Emergency and Faulted Conditions)
{
The normal condition for the flywheel is the normal operating speed of 1200 rpm. The faulted condition for the flywheel is the overspeed of 1500 rpm.
The results of this approach are provided in Table 1 below. Shrink fit is included in these results, since shrink fit increases the magnitude of the hoop stresses (as shown in Figure 1) and consequently, the stress intensity factor. In Table 1, crack length is measured radially from the keyway, and percentage through the flywheel is the crack length divided by the radial length from the keyway to the flywheel outer radius.
1 L
m
~.
.-.. =..
l Table 1: Allowable Crack Lengths for Flywheels including the Effect of Shrink Fit and Section XI Criteria Flywheel Allowable Crack Lengths in Inches and % through Flywheel Group 1200 rpm (Nomial Speed) 1500 rpm (Overspeed) l RTNDT =
RTNDT =
RT DT =
RTNDT =
RTNDT =
RTNDT" N
00F 30oF 60oF 00F 300F 60oF 1
2.3" (7%)
1.4" (4%)
0.4" (1%)
7.6" (23%)
2.7" (8%)
0.6" (2%)
3 2.3" (7%)
1.5" (4%)
0.4" (1%)
8.0" (24%)
2.8" (8%)
0.5" (2%)
10 1.9" (7%)
1.3" (5%)
0.6" (2%)
8.3" (31%)
3.7" (14%)
1.6" (6%)
14 2.2" (8%)
1.8" (7%)
- 1. I" (4%)
12.0" (43%)
5.4" (20%)
1.2" (4%)
15 1.0" (5%)
0.5" (2%)
0.2" (1%)
4.3" (21%)
1.9" (9%)
0.9" (4%)
16 1.9" (8%)
1.4" (6%)
0.7" (3%)
10.2" (42%)
4.6" (l 9%)
1.8" (7%)
It is important to note that several conservative assumptions were included in the determination of the allowable crack lengths provided in Table 1. These are discussed as follows:
a) The closed form solution for the stress intensity factor was used. This solution assumes that the keyway radial length is included in the crack length, which is conservative for smaller crack lengths.
l This conservatism is evident in Figure 3 for Flywheel Group 1, which indicates that a zero length crack
{
has a stress intensity factor of about 42 ksi Vinch, since a crack of 0.937 inches (the keyway length) is assumed. As shown in Figure 4 of the attached ASME technical paper (Attachment C), finite element analysis shows that the stress intensity factor for emcks less than about one inch long is significantly less than the closed form solution would predict. Therefore, there is significant conservatism in the smaller allowable crack lengths ( l inch and smaller) provided in Table I above.
f b) A conservative shrink fit was assumed, as discussed in the Response to item 1.
l c) A lower bound fracture toughness for ferritic steels was used, as discussed in Section 4.3, page 4-7 of WCAP-14535.
d) The very conservative criteria of Section XI were used. These criteria apply margins of ten (10) to normal, upset and test conditions, and two (2) to emergency and faulted conditions. These margins account for uncertainties in flaw sizing and loading. It should be noted that the loadings associated with I
the flywheel (centrifugal forces and shrink tit) are well defined and were censervatively applied in this evaluation.
i e) The ambient temperature used for the fracture evaluation (70 F) represents a much lower temperature than would be expected in the containment building during normal plant operating conditions (typically 100oF to 120oF).
1 l
f) The stress intensity factor is calculated using the methods oflinear elastic fracture mechanics. This method assumes rapid crack extension in a linear elastic material (i.e., material properties below RTNDT)-
The flywheel material would remain highly ductile since the operating temperature is well above the l
RTNDT of the material. The conservatism using this method is therefore inherent.
t 4
4 4
l Over the past ten years, the examination techniques employed have improved, particularly with the use of the defocused gage hole probe. The detectability of the gage holes at various metal paths displayed in Attachment B indicate that the inspection methods used for flywheels are capable of finding flaws of the sizes identified in Table 1.
In Attachment B, the 1.25 inch diameter gage holes (effectively side drilled holes) were clearly identified at a metal path which is nearly twice the metal path distance involved in the inspection of the keyway area. It should be noted that it is conservatively estimated that the effective reflective surface of a side drilled hole is a 300 arc. The reflective surface from a 1.25 inch gage hole would therefore be 0.33 inch.
This length is smaller than all but the smallest of the allowable crack lengths in Table 1 (0.2 inch, Flywheel Group 15, RTNDT of 600F,1200 rpm). As discussed above, a significant amount of conservatism is inherent in the smaller allowable emck lengths ( l inch and smaller) provided in Table 1.
j in addition, Flywheel Group 15 includes only one plant, Three Mile Island Unit 1. Per Babcock and Wilcox Power Generation Group, Nuclear Power Generation Division Topical Report BAW-10040, 1
December 1973," Reactor Coolant Pump Assembly Overspeed Analysis," the RTNDT value of the flywheel material is estimated to be -100F. Therefore, a crack length of greatet than one inch would be allowed, which is larger than the 0.33 inch reflective surface from a 1.25 inch gage hole discussed above.
Therefore, the inspection methods used for flywheels are capable of finding the flaw sizes shown in Table 1.
l i