ML20138M974
| ML20138M974 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak, 05000000 |
| Issue date: | 12/20/1983 |
| From: | Abramson L, Lurie D NRC OFFICE OF RESOURCE MANAGEMENT (ORM) |
| To: | Clark J NRC OFFICE OF RESOURCE MANAGEMENT (ORM) |
| Shared Package | |
| ML20138M609 | List:
|
| References | |
| FOIA-85-59 NUDOCS 8511040496 | |
| Download: ML20138M974 (4) | |
Text
Y' h
.T0-381 TEXAS UTILITIES GENERATING CO. IPANY T
OTFICE MEMOR ANDU M July 27,1983 Te R. G. Tolson Dallas. Texas Subject '
Centertive Coatine Adhesion Tests Per your request I have reviewed the results of adhesion tests performed durina backfit inspections (per QI-QP-11.4-23) of miscellanecue steel in Unit 1.
For the purposes of this analysis, misc 5Tlaneous steel was 4
defined as all steel substrate items inspected in the backfit program i
except the containment liner. Of the 3625 adhesinn test. results reviewed,
'25 or 0.695 were unacceptable. A one-sided confidence interval for the binomial parameter p (the proportion of successes) was developed with o =.02.
In sum, we can be 98% confident that greater than 99% of the protective coatings on niscellaneous steel in Unit I have acceptable adhesion.
4 d
i e
J1
~...
L. M. Bielfeldt Special Project Engineer
- , f 4 J.u$
,,, e,'e
.-L,
rC,
<. er
- h =, h S
B511040496 031220 PDR FOIA CARDE85-59 PDR
F l
s s
~
DEC 2 01983 MEMORANDUM FOR: John W. Clark, Chief Management Ahalysis Branch s
Division of Budget and Analysis, RM FROM:
Lee Abramson, Statistical Advisor Dan Lurie, Mathematical Statistician Management Analysis Branch, RM/B
~
SUBJECT:
TRIP REPORT ON COM/,NCHE PEAK BACKFIT INSPECTION On November 29-30, we met with Frank Hawkins (RIII), Claude johnson (RIV) and Lisa Bielfeldt (Texas Utilities Generating Company (TUGC)) in the TUGC offices in Dallas. We also conferred by telephone with TUGC and plant staff located at Comanche Peak. The purpose of the. meeting was to review the statistical analysis of the backfit inspection program at Comanche Peak Units I and 2, with the intention of providing an estimate' of the total amount of coating which might flake off in the event of an accident.
The backfit inspection program is required to verify that the containment coatings were properly applied', both to the steel and concrete liners and to the miscellaneous items (miscellaneous steel, pipe hangers, conduit supports and cable tray supports). The liners were divided into test areas of approxi-100 9 eet and three adnesion tests and five y.trb_le.s.ts.Lto f
r.3,t_el y te Miscellaneous measure coating thickness) were performed in eacr' test area.
items were divided into test areas of between 5 and 100 square feet, whenever possible, and thrae adhesion and five scratch tests were performed in each test area.
If a test area was less than 5 square feet, only one adhesion int _,
~~ ~
and five scratch tests were perfomed.
~
~ ~
If any adhesion or scratch test indicated a defect, the -extent of the defec-tive area was delineated with additional tests and the entire defective araa was repaired'. - However, since the actual area tested 7as a very small f rac-7 ion of the total area of the coating (the adhesion test covers a circle about one inch in dia@ter and the scratch test is only a few inches long), the possibility exists that defective sections of coating, not sampled by the test procedure, still remain.
In order to estimate the total amount of coaAing which might flake off in the event of an accident, it is necessary to estimate the total area of the defective enatinn "ill ramininn aftar the becLfit ins.pectinn has ioentified a'nd~'Ie~ paired the defective. areas _which itToppins to hav'e tested.
The residual
~
defective area can, be estimated from~the observed failure rates for the adhesion and scratch tests as perfomed, alchg with an estimate o'f the average size of the defective areas which were, repaired With this information for l
.the various liners and miscellaneous items, it, is, poss,ible to calculate a i
confidence interval for thefesiduaTyfe-fective area.
l
.w.
....a>.
l
, o *. 4 W S >
~
~
l
...m
~
l 2:. 2::= J_:*
- y=.
2 DEC 2 01933 -
..:," ~..!',TnW. Clark
. --+ m
~-4,j-[;,t the meeting, we recuested Ms. Sielfeldt to provide us with the data described Since the analyses perfomed to date by TUGC were designed to confirm above.,
that the averace thickness of the coating met the technical specifications
,/
_ _.J rather than to estimate the percentage of defective coating, it will be
- /
necessary for TUGC to analyze the backfit inspection records to provide us with the requested data. Furthemore, in order to ascertain that the data we e
will receive is approprir.te for making the required estimates, we asked Ms. Bielfeldt to send us a sample of the data as soon as she collects it. !!e expect to receive this sample data.by the end of the year.
7 Lee Abramson, Statistical Advisor Dan Lurie, Mathematical Statistician
, i'.anagement Analysis Branch, RM/B cc:
F_ Hawkins, RIII C. Johnson, RIY E. Triner, RM/B bec:
L. Abramson, RM/3MA CD' DJr'ib', P?./BPf) ff/Bt's Subj/Chron 1,
I
+, ~ <
.... < > R"./ E %,,j.,,
R"57*,A,,.,,,l,
.l,
"-) @ 9. 7:5 0 M b.b,,,.L u ri.e, I,j, u>
12/ :/E3
. l, IL V/53
- ~....
.70-381 TEXAS UTILITIES GENERATING CO.TIPANY O F F I C E M E M O F. A N D U M July 27,1983 7.,
P. G. Tolson Da!!as, Texas Subject Pr ntactive Coating Adhesion Tests Per your request I have reviewed the results of adhesion tests performed during backfit inspections (per 01-QP-11.4-23) of miscellanenus steel 1
in Unit 1.
For the purposes of this analysis, miscillaneous steel was i,
defined as all steel substrate items inspected in the backfit program q
except the containment liner. Of the 3625 adhesinn test results reviewed,
~
j 25 or 0.69% were unacceptable. A one-sided confidence interval for the binomial parameter p (the proportion of successes) was developed with cv =.02.
1 j
In sum, we can be 98% confident that greater than 99% of the ~ protective coatings on miscellaneous steel in Unit I have acceptable adhesion.
4 i) j Q-4 L. M. Bielfeldt Special Project Engineer y '_ M o c o = o. e. r %
- s s 6 - L, i
I l
.