ML19329E229
| ML19329E229 | |
| Person / Time | |
|---|---|
| Site: | Midland |
| Issue date: | 12/22/1977 |
| From: | Howell S CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| Shared Package | |
| ML19329E225 | List: |
| References | |
| HOWE-214-77, NUDOCS 8006110649 | |
| Download: ML19329E229 (8) | |
Text
a J
~ ]t ConsumEIS Power s.o...e... ea Campany Voce Presodent Generer Omces. 212 West u cnigen Avenwe. Jackson, u.cn. gen 49201 December 22, 1977 Hove-21k-77 Mr J. G. Keppler, Regional Director Office of Inspection and Enforcement Region III US Nuclear Regulatory Co= mission 799 Roosevelt Road Glen Ellyn, IL 60137 MIDLAND NUCLEAR PLANT -
UNIT NO.1, DOCKEI NO. 50-329 UNIT NO. 2, DOCKET No. 50-330 PIPE SUPPORP FILLET WELDS
Reference:
Letter, S. H. Howell to J. G. Keppler, Midland Nuclear Plant -
Unit No. 1, Docket No. 50-329; Unit No. 2, Docket No. 50-330; Pipe Support Fillet Welds, Serial Howe-197-77, dated Nove=ber 21, 1977 The referenced letter was an interim report which provided corrective action plans relative to the undersized fillet velds on linear pipe supports.
Attachments 1 and 2 provide Bechtel Associates' Second Interim Reports in response to the Management Corrective Action Reports (MCAR-1 Report No.18 and MCAR-1 Report No. 19). These reports cover velds which are undersized to the requirements of the ASFE Code and shop velds which are undersized to drawing requirements. The investigation of the two problems to date has been separated. It is recognized that the two conditions could exist con '
currently on an individual pipe support and this possibility is being investigated. is the ASME response to the Code inquiry (Attachment h to the referenced letter). This response states that the minimum size requirements of Table XVII-2452.1-1 are applicable and the minimum size is independent of calculations indicatin6 that s= aller velds vould be acceptable.
In response to Consumers Power Company's Nonconfor=ance Report Q7-201, which reported four undersized field velds out of a sample of kh, Bechtel has reinspected an additional sample of 112 field velds and all were found to be acceptable. Evidence to date indicates that the nonconfor=ances reported in QF-201 are not representative of a generic problem. The hardware disposition to this Nonconformance Report vill follow the final disposition of the under-sized shop velds reported in MCAR-19 hp 8006 1.10 I~,.
gg
e 2'
s Another interim report vill be sent on or before February 10,15rr7 h....--.
D D
- j t:.s
\\
Attachments: 1) Interim Report No. 2 dated December 16, 1977, MCAR-18
- 2) Interim Report No. 2 dated December 16, 1977, MCAR-19
- 3) Letter, John Millman (ASVI) to W. R. Bird (CPCo),
Subject:
ASME File ENI 77.h08, dated December 8,1977 CC: Dr Ernst Volgenau, USITRC (15)
Director, Office of Management Information and Program Control, W NRC (1)
- -. ~ ~ -
. ~
g
',k Howe-21k-TT
SUBJECT:
HCAR #
18 (Issued in-7n.77 )
INTERI!! REPORT # 2 DATE:
12-16-77 l
PROJECT: Consumers Power Company Midland Plant Units 1 & 2
)
Bechtc1 Job 7220 Description of Discrepancy l
Some fillet weld sizes designated on Crinnc11 hanger shotches are at
)
variance with the minimum weld size requir'enents of AS!!E Section III-l NF 3392, NF-3400, and Appendix XVII-NA 2452.1.
{
3 Potential Safety Implication J
Crinnc11's physical tests of two of the " worst case" wcld size deviant welds conductsd on December 12 and 13, 1977, indicate that these " worst I
case" welds were capabic of load many times their design load requirements.
Bechtel engineering analysis of 80 randomly sampicd hanger sketches having under'specified wcld callouts indicated that all the deviant veld stresses were less than 6S% of Code allowables. The physical tests and the analysis continue to affirm that there is no sifety probicm due to underspecified weld size callouts.
Corrective Action and Investigation i
Bechtel has requested Grinnell, as required by )! CAR-18 Item 1, to a
- provide justification of their position on the deviation from the Code, and for their program and approach for presenting their justification to i
the ASME code committee and other jurisdictional authorities. To date, a response has not been received. Grinnell acknowledges the deviation from the minimum fillet veld size aspect of the Code, but also contends that all those deviant veld callouts are very conservatively designed in i
terms of the allowable veld stress aspcet of the Code, which is the real l
intent of the Code, and which they have met.
Per MCAR-18 Item 2, Bechtel requested an informal code interpretation at j.
the AS!!C Code committee meeting November 1,1977. The code committec chairman stated, if a formal inquiry were presented, he would support a Code interpretation that minimum fillet veld size must be at least the thickness of the thinnest member joined where the Code minimum wold size table calls for a weld equal.to or greater _than the thickness of the thinner member.-
j i
a w-y t
w rv ym v
wy-,
-w r
--'w-,
_R
~
Crinnell submitted the Test Procedure T-S-91990-2 " Full Size Destructive Loading Test of Three llanger Asse=blies, Midland Units 1 & 2 Bechtel 7220-M-106-AC," on December 9, 1977. The tests were performed at Grinnell's test facilities in Providence R.I. on December 12 and 13, 1977, with observers from both Consumers Power Co. and Bechte).
Pro-liminary indications of the tests confirm Grinnell's contention that these " worst case" designs were very conservatively designed.
Grinnell has affirmed that all new designs of fillet welds since April 1977 are in ecmpliance with the minimum weld size Code critoria.
Drawings reviewed by Bechtc1 confirm this.
Bechtel's random sample chech (1% per MCAR-18 Item 3) of 650 earlier hanger sketches yielded 110 fillet veld callouts that did not comply with the Code Appendix XVII-NA 2452.1 criteria for such welds.
The nonecmpliant wcld callouts were underspecified as follows: 59% by 1/16", 43% by 1/8", and 2% by 3/16".
An engineering analysis by Bechtel of 80 of the 110 noncompliant wcld callouts indicated that all weld stresses were 12.3 ksi cr less, well below the IS ksi maximum designated by the Code.
(Completes MCAR-18 Item 3)
To comply with MCAR-18 Item 4, Bechtel QA/QC personnel inspected 23 installed hangers and hangers in the warehouse representing 112 welds.
Tha results of the inspection were:
76% met or exceeded the designated wcld callout, 15% were under the designated wcld size by 1/16 inch or less over Icss than 15% of the wcld length, 5% were under the designated wcld size by 1/16 inch or Icss for more than 15% of the veld length, and 4% were under the designated wcId si:e by 1/S inch for more than 15% of the veld length. The resolution of the "as-built" undersized welds will be addressed in MCAR 19.
(Completes MCAR-18 Item 4)
Forecast Date on Corrective Action The interim reports will be issued on a 30-day basis until the final Grinnell report justifying their positien on the deviation from the Code in February 1978 is issued. Program results will then have to be accepted by the Owner and the Code jurisdictional authority. A final report would then follow.
Submitted by:
dy IFRFn24 Approved by-g_3efp // /A,dA.h Concurrence by:
/
v i
..S. -
Att:chment 2 Howe-214-77 l
SUBJECT:
' MCAR # 19 (Issued 11-7-73 INTERIM REPORT # 2 DATE:
12-16-77 PROJECT: Consumers Power Company Midland Plant Units 1 & 2 Bechtel Job 7220 Description of Discrcoancy A random sample of 32 hangers was inspected at the jobsite, us'ing a weld fillet gauge, to determine the actual shop weld sizes compared to the design weld sizes identified on Grinnell drawings.
22% of the 112 wolds on the sample hangers were found to be under the specified nominal size required by Grinnell drawings.
Potential Safety Innlications To date, Grinnell's analytical analysis of the reported deviant welds continues to reaffirm Grinnell's original position that no safety problem exists.
Status of Corrective Action and Investigation The cause of the deviation was an insufficient icvel of internal weld inspection by Grinnell QA/QC personnel.
(Completes MCAR-19 la)
Grinnell utilized a visual inspection technique per their Quality Control Procedure 02A006-dated 3/14/74, which allows fillet welds in any single continuous weld to underrun the nominal fillet size required by 1/16 inch without correction provided that the uniersized weld does not exceed 10%
of the length of the weld. Grinnell has stated that all shop welds are inspected; however, only welds that appear suspect are actually checked with a gauge.
(Completes MCAR-19 lb)-
Grinnell has rubmitted for' approval their " Visual and Dimensional A,cceptance Criteria for Welds" per their revised Quality Assurance / Quality Control Procedures 02A006 dated 12/8/77 to allow tolerances on fillet welds:
Undersize is permitted, provided it does not exceed 1/16 inch for 10% of the weld length or an equivalent volume.
Oversize welds are permitted as follows:
a)
. Nominal wcld size up to and include 1/4 in.
+1/8 in.
b)
Nominal weld size over 1/4 in. and up to
+3/16 in,
~
and including 1/2 in.
c)
All welds over 1/2 inch
+1/4 in.
c
.'Grinnell is committed to revise the weld inspection procedure to includo both under and over tolerances, reference the inspection procedure on the permanent record QC checklist (one for cach hanger), and provide additional training to the inspection team upon Bechtc1's approval of Grinncil's revised procedures.
Qualification of existing wcld sizes will be by Grinnell's analysis in lieu of MCAR Items 2a, 2b, and"2c.
Bechtel has completed a representative field inspection sample per MIL-STD-105D of "as-built" hanger welds as compared to the wcld size specified on hanger sketches. Grinnell has analyzed 26 of the deviant welds resulting from the Bechtel field inspection. Grinnell analyzed these deviant welds on a " worst case" undersize basis and preliminary shown that all the welds were stressed less than 46% of the Code allowable stress. Grinnell's final report, expected on January 3, 1978, will include the 34 hangers which had deviant welds.
To establish control of hangers not installed or in manufacture, the fol;.owing stcps were taken or conditions established:
Bechtel PSQR shop monitoring program has continued on the esta-blished quality verification daily monitoring basis with satis-factory results. Bechtel PSQR shop monitoring program indicates that Grinnell's internal shop inspection has improved which is evidenced by a smaller number of Bechtel PSQR rejections. The latest field receipt inspection shows no undersized wolds.
The evidence indicates the program is operating satisfactorily.
(Completes MCAR-19 3bl, 5) 3echtel field QA/QC personnel have been performing 100% inspection on all incoming hangers and they have found no welds undersized from the veld sizes designated on the hanger sketches.
Bechtel's incoming hanger inspection will continue until Grinnell's revised inspection procedures have been approved and Bechtel's PSQR verifies that the procedures have been implemented.
Forecast Date on Corrective Action It is expected that the Grinnell summary report will be received by Bechtel in January 1978. A final report would then follow.
Submitted by:
Approved by,:- Cd4d (N Concurrence by:
0h
/
RNT/bkp
n
.~.
Hove-214-77 31' The American Society of Mechanical Engineers
!E'S h/
United Engineering Center / 345 E. 47th St., New York, N.Y. 10017 / 212 644 7815 THE BOILER AND PRESSURE VESSEL COMMITTEE
.s December 8, 1977
'..*,.. *. f. '
' i r,;
LP. zick
.9 V
. s u?
e.
weecwmm f
C.L HARDING Consumers Power Co.
]..,4 212 West Michigan Avenue
"$.a t secretcy Jackson, MI 49201 C.8.HOYT Att:
W. R. Bird C.'.*'. A LLISON 8'
^
ecu Secdon m, MMsion 1, R
ON ER R.1 BOSN AK Table NF-3132.1(b)-1 and NF-3292 P.M. BRISTER H "- ^" AN Ref: Your letter of October 28, 1977 CEP H
LA CHOCKIE ASME File #NI 77-406 YLE. COOPER C.D. DOTY Centlemen:
G.E. F R ATCH ER R.C. GRI F FIN S.F. H A R RISON In response to your inquiry, it is our understanding that you are asking the following questions:
L.
ON E.L KEMMLER E.L KIME QUESTIONS:
1E. LATTAN J. LeCO F F 1R. M ACKAY (1)
If class 1 linear supports are designed by " experimental stress R.H. MO E LLE R analysis" or by " load rating," does paragraph NF-3292 apply?
T.E. NORTHUP C.E. R AWLINS W.R. SuiTH. SR.
(2) When paragraph NF-3292 applies, must all fillet welds comply with W.E. SOM E RS the minimum size requirements of paragraph XVII-2452 and Table XVII 2452.1-1 when the analysis shows a smaller weld size to be satisfactory?
REPLY:
(1) NF-3292 is applicable to linear supports designed by any of the procedures permitted by NF-3132.1(a) and as shown on Table NF-3132.1 (b)-1.
(2) The minimum size requirements of Table XVII-2452.1-1 are applicable when the joints are connected by fillet welds only as stated in XVII-2452.1.
These minumum size requirements are independent of calculations indicating that smaller velds would be acceptable from the computed stresses although la.ger welds shall be provided when necessary to satisfy stress calculations.
l l
l Member of Engineers Council for Professional Development and Engineers Joint Council
p When fillet welds are used in combination with another type of weld (2) such as groeve welds, the minimum size requirements of Table configuration, XVII-2452.1-1 do not apply.'
Very truly yours, ohn Millman Assistant Secretary 1
i a
S
.I l
e y
e