ML20238C669

Forwards Revised Pages of Util Re Pipe Support Design Verification Program.Ltr Revised Due to Typos Re Std Deviations in Min Rock Bolt Depth Assumption & Revised Chi- Squared Values

ML20238C669
Person / Time
Site:	Trojan File:Portland General Electric icon.png
Issue date:	09/04/1987
From:	Cockfield D PORTLAND GENERAL ELECTRIC CO.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
TAC-65726, NUDOCS 8709100290
Download: ML20238C669 (3)

Text

..

E M -M pp!- Portland General Electric Coirpinv/

- David W. Cockfield Vice President, Nuclear September 4, 1987 Trojan Nuclear Plant Docket 50-344-License NPF-1 U.S. Nuclear Regulatory Commisaion ATTN: Document Control Desk Washington DC 20555

Dear Sir:

Pipe Support Design Verification our letter of August 18, 1987 [ David W. Cockfield to U.S. Nuclear Regu-latory Commission (NRC) Document Control Desk] described our Pipe Support Design Verification Program and responded to a number of NRC concerns regarding the adequacy of pipe support anchorages. Attachment B of that letter contained a typographical error. Paragraph f on Page 1 of Attach-ment B should indicate that 2.4, vice 2.26, standard deviations were used in the assumption of minimum rock bolt depth.

Also, on Page 2 of Attachment BS, the third paragraph should indicate a valte of 4.06, vice 10.19, for Chi-squared and a 0.05 level of significance for Chi-squared of 16.919 vice 15.507. These numbers result from a final check of the statistics calculation on August 19 and provide an even better fit to a normal distribution.

Revised pages for the August 18 letter are enclosed. These changes were discussed with the NRC on August 20, 1987. They are not significant and do not alter the results or conclusions of the August 18 letter in any way. .

Sinterely, l 344 PDR

[#P Attachments .

c: Mr. John B. Martin Regional Administrator, Region V U.S. Nuclear Regulatory Commission Mr. David Kish, Director State of Oregon Department of Energy Mr. R. C. Bare NRC Resident Inspector / g Trojan Nuclear Plant !l 121 S W Samon Street Portand, Oregon 97204

I

o. {

Trojan' Nuclear Plant Document Control' Desk I Docket 50-344 September 4. 1987- .j License NPF-1 Attachment B' i Page ILof 17 j SAFETY-RELATED PIPING SUPPORT ANCHORAGES A. Program Summary and'Results i i

1.0 Program Summary During.the course of the Pipe Support Design Verification Pro-gram, the following questions arose regarding the use of rock-bolts for anchorage of some of the supports to the concrete:

a. The failure mechanism anticipated for the rock bolts (ie, ,

grout bonding, concrete shear-cone failure,~ or bolt tension )

l failure), i l

b. The significance of variance of the embedment depth from the nominal 8-inch depth. 'j

c. The capability of the as-installed rock bolts to provide sufficient anchorage for tho' safety-related pipe supports with a factor of safety of at least two.

To address these issues, Portland General Electric Company (PGE) has completed a test. program, measurement of accessible rock bolt embedment depths, verification of capacity.versus demand (using a factor of safety of two) and a failure load analysis. Further-more, during the load versus demand comparisons, several very conservative assumptions were used, viz:

a. The shear-cone embedment depth was taken from the top of the expansion shell rather than the bottom. Further reductions were taken to allow for the accuracy of bolt embedment depth-measurement,

b. The presence of reinforcing steel was ignored.

c. A linear interaction ratio was used to measure acceptability.

d. Seismic load demands were conservatively taken as 1.67 times l

the Operating Basis Earthquake (OBE) with onl?/ a 0.5 percent l damping ratio for the piping,

e. Typically, for a given support, the bolt with the minimum capacity was compared to the maximum demand of any rock bolt used in the support.

f. Where rock bolta could not be measured due to inaccessi-bility, a minimum rock bolt depth (using 2.4 standard devi-ations from the mean measured depth) was. assumed, t.

e o

Trojan Nuclear Plant Document Control Desk Docket 50-344 September 4, 1987 License NPF-1 Attachment B5

-Page 2 of 11 In order to allow for measurement error, embedment lengths for measured bolts were determined by, deducting an estimated measurement error for the embedment length from the calculated embedment lengths.

The estimated error was determined by comparing the differences in measurements on 41 rock bolts. These rock bolts were measured inde-pandently by two separate crews using two separate ultrasonic length measuring instruments. The mean value of the measurement differences wasfoundtobe0.16inchandthiswasusedastheestimated~ measure.$

ment error. The embedment lengths with the subtracted measurement' i error were used to determine the tension capacities of the embedment. '4 In order to provide a basis for determining the embedment lengths of tj inaccessible rock bolts, a statistical evaluation was performed usin~g the rock-bolt embedmont length data measured from accessible suppo({s.

The measured embedmont lengths for the 609 rock bolts tested were plotted in the histogram in Figure B5-1. The nornal probability distribution which fits these data has a mean corrected embedmont length (x) of 7.79 inches with a standard deviation (s) of 0.70 inch. A Chi-squared test was performed on the data and resulted in a value of 4.06. Since this value does not exceed the Chi-squared value at the 0.05 level of significance of 16.919 for this data distribution, the normal distribution provides a good fit.

l With the above statistical model, the embedment lengths can bo l estimated for inaccessible rock belts. There is a 95.5 percent probability that the rock bolts in the inaccessible category have embedmont lengths greater than i -2s, which corresponds to an embedmont length of 6.39 inches. This was conservatively rounded down to 6.1 inches which is less than all the measured lengths in Table B5-1. This corresponds to 2.4 standard deviations from the mean. Thus there is a 98.3 percent probability that the embedment length of the rock bolts in the inaccessible category are greater than 6.1 inches.

i 4. Table Legend The columns of Table B5-1 are explained as follows.

Alumn A " Support Number" This colunn lists the support designations of safety-related pipe supports which have 1-inch diameter rock bolts with the minimum nom-inal 8-inch embedmont. The table includes only those supports with rock bolts. The remainder of the supports included in the verifi-cation program used other types of anchorages or longer rock bolts.

I

)

.i e

4 t

~

V u)

Ci ( .

U gy O'

2 I U) G. ,

D N 5 1 1

1 l

1 i

1 h

i I

.1

)

.