ML20127H520

Final Deficiency Rept 55-84-19 Re Nelson Studs on Embedment Plates.Initially Reported on 840828.Embedment Plates Repaired to Restore Full Design Load Capacity & Instructions for Welding Structural Steel on Embedments Revised

ML20127H520
Person / Time
Site:	Clinton
Issue date:	04/30/1985
From:	Hall D ILLINOIS POWER CO.
To:	James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
55-84-19, U-10272, NUDOCS 8505210315
Download: ML20127H520 (5)

Text

-

m r.

i

-)M 7 _

a.

-lLLINDIS POWER COMPANY ~1A.120

[3EI U-10272 CLINTON POWER STATION. P.O. box 678. CLINTON ILLINOIS 61727 April 30,'1985

~

Docket No. 50-461 Mr. James G._Keppler Regional Administrator Region III-U.S. Nuclear Regulatory Commission 799 Roosevelt Road Glen ~ Ellym, IL 60137

Subject:

Reportable 10CFR50.55(e) Deficiency 55-84-19:

,-Nelson Studs on Embedment Plates

Dear Mr. Keppler:

On August 28, 1984, Illinois Power Company notified Mr. F.

Jablonski,- NRC Region III, (Ref: IP Memorandum Y-20790, dated August 28, 1984) of a potentially reportable deficiency concern-ing Nelson Studs, which have parted from.the back of the embedment plates. .This initial notification was followed by-two (2) interim reports (ref: IP letter U-10204, D. P. Hall to J. G.

Keppler, dated October 1, 1984; and IP letter U-10242, D. P. Hall to J. G. Keppler, dated January 21, 1985). Illinois Power's investigation of-this matter is complete. Our investigation

' determined that the problem associated with the parted Nelson

' Studs was limited to four (4) embedment plates located on the drywell outer wall. These. plates have eccentrically located welded attachments utilizing large fillet and penetration welds.

In addition, our investigation included a field inspection of other embedment plates for which Nonconformance Reports (NCRs) had been initiated to document occurrences of plate distortion or concrete spalling. No'other occurrences of parted Nelson Studs were. identified. Our investigation into-this matter has determined that this issue represents a reportable deficiency under the provisions-of 10CFR50.55(e). This letter is submitted as a : final report in accordance with the requirements of 10CFR50.55(e). Attachment A provides the details of our investigation.

8505210315 850430 PDR ADOCK 05000461 IO -

S PDR n

I hh

q We trust that this final-report provides you sufficient  :

background information to perform a general assessment of this reportable deficiency and adequately describes our overall approach to resolve the issue.

Sincerely yours,

. P. Hall Vice President RLC/lr (NRC)

Attachment ,

cc: NRC Resident Office, V-690 Director - Office of I&E, US NRC, Washington, DC 20555 Illinois Department of Nuclear Safety INFO Records Center o

,. , , - . . - , - - ,- , ,---.------.---,..i--o- -= ,,,.w. r -- .s -,--,- - e.,a --e r c-

y .g.

i ATTACHMENT A Illinois Power Company Clinton Power Station Docket No. 50-461 Reportable:10CFR50.55(e) Deficiency 55-84-19 Nelson Studs on-Embedment Plates Final Report Statement of-Reportable Deficiency / Background Nonconformance Report (NCR) No. 20031 was. written to docu-ment a condition where.an embedment plate on the outside of the-idrywell wall:had pulled;away from the concrete.(AZ 128* el. 753',-

-S&L dwg. S27-1002-03A). To aid in dispositioning of this condi-tion, the concrete was excavated along the edges of the plate to

-expose the embedded Nelson studs behind the plate. All'.seven (7).

studs inithe outer row were exposed. Six (6) of the seven (7)1 studs were found detached from the-back of the plate. A second

. plate at AZ 140* was also excavated and found with two (2) of

-four-(4)~ exposed studs parted from the back of the plate _(NCR 21516).

Investigation Results There are four (4).embedment plates in the drywell outer-

. wall-(AZ-128*, 140*, 157 , 174*, el.-755') with eccentrically-

. located' beam connections. The end connections'had been field modified due to a design change and required large fillet and penetration welds to be applied to the embedment plates. NCR' 20031-was written after these modifications to document the fact that the plate at-AZ 128 had pulled'away from the concrete. The concrete on-one side of.this-plate was excavated to expose the l Nelson Studs behind the plate. Six (6) of seven (7) inithe outer

' row were found parted from the embedment plate. Although'the .

embedment plates at AZ 140*, 157*, 174* had not shown any obvious signs of distortion or concrete spalling, they were excavated.to determine if any other studs were parted.. The fact that beams-

-were also attached to these embedment plates eccentrically

.(positioned off the centerline of-the plate) gave reason to suspect that the modification welding near to the stud row may 1: have caused thermal stresses and/or unobservable plate distortion

-sufficient: to cause the stud failure.

The embedment plate at AZ 140 was excavated on-both sides.

Two~(2) studs.in the row of seven (7) closest to the welding area

.were found parted-(NCR 21516).- The side away from the' welding areas had six (6) of the seven-(7) studs exposed. All six (6)

Lwere still' attached to the back of~the embedment plate. The embedment plates at AZ 157* and 174* were excavated on the side near.the weld areas. 1k) studs on either plate were found parted.

IPQA has written' Surveillance" Reports to document the results of the excavations.

Page 1 of 3

-: a g

/v l l

j-ATTACHMENT A (continued)

Since:thejbeams were eccentrically attached to the emb'edment

, plate,-Sargent & Lundy1(S&L) was requested to evaluate the

-possibility that-the modification welding using large (7/8")

~

fillets near the outer stud row was a plausible.cause of the

~

. . studs.aarting. S&L provided calculation 8.35.0-4A which shows

-that tiermal stress and/or thermal gradient are plausible.causes forJtheistuds. parting. A copy of calculation 8.35.0-4A was provided to the NRC Region III-(J. Jacobson). Mr. Jacobson expressed concern about.the' possibility that residual stresses remaining in the-plate could reduce the capacity of the remaining

stud connections. In a' meeting February 7, 1985 with Sargent &

Lundy,- NSED, and Mr. Jacobson, a-presentation was,made to

-demonstrate that no' residual stresses would remain in the unbroken 1 studs after'the plate cooled to ambient temperature.

1Mr. Jacobson accepted'this explanation,but requested that stresses due'to. weld shrinkage be considered also.. ~ Sargent &

Lundy-prepared: calculation-no. 8.35.0-5A to account for the

-effects of shrinkage' stresses. 1 Based on this calculation,'the strength:ofJthe' remaining unbroken stud connections on each plate

will not be diminished by residual or shrinkage stresses. .This

' calculation was submitted to'the NRC February 22, 1985, and-accepted March:8, 1985.

The review;of the existing documentatio'n=for the.embedment plates supplied:by Rockwell Engineering has revealed no' obvious discrepancies. 'The. stud welding procedure used by Rockwell (RECO

200-1) had been reviewed and accepted.by S&L and Baldwin-Associates. -The inspection of the stud weld fracture area-behind the: plate indicated the~ studs had been welded to the plates properly. Repair: travelers S-1331, 1332, 1334, 1335 required replacement of missing-studs on-four (4) of the five (5) plates-in question'. .Although it is not precisely known which' studs _were replaced, the travelers S-1334 and 1335 required a visual check of the remaining studs.

Since,.1) the excavations have shown the problem-to be limited'to the-embedment plates at AZ 128' and 140* and only in

<the area where the-fillet welds were applied near the outer row-of studes 2) calculations show welding induced stresses to'be sufficient-to cause the stud failures ands 3) there is'no evidence offimproper fabrications it is concluded-that the stud separations were~ caused by the stresses induced from-the large-weldsLapplied near the plate edge.

Corrective Action.

-The loading on the two (2) embedment plates with parted studs is such that repairs were made to restore the full design load capacity, i.e, the total Design Basis' Accident (DBA) q Page 2 of 3 a

L F .

.l

. ATTACHMENT A (continued)

loa'dingfincluding: Safe Shutdown Earthquake (SSE) and Pool Swell

' loads. The' broken studs cannot be rewelded to the. plates.

Therefore, a repair design was: developed based on,weldingLwing-

. plates to the existing embedments. At AZ 140', a plate ^ tab was tweldedito an' existing, - adj acent embedment plate. At AZ-128' the wing plates,Jtop.and bottom, were secured to the drywell: wall using:Drillco Maxi Bolts. -The excavated concrete was replaced

with grout. ,

't -

In addition to the. specific embedment plates previously Ediscussed,.severalEother-.NCRs were written to describe casesoof embedment plate distortion. There was one known case where two

- -(2). deformed wire anchors, welded to an embedment 11 ate, had parted-(ref: NCR 18764/25462). These were on the'iottom. row of

.an embedment plate that has 3 columns of'5 wire anchors.each;

'.The welding of the beam seat to the embedment plate with a single E < bevel, partial penetration weld, had visibly distorted the plate on1the bottom (ref: NCR 18764). Illinois Power and S&L evaluated '

-these deficiencies and dispositioned them use-as-is. To preclude-anyffuture parting of studs and minimize the distortion of.

. 'embedment plates, Sargent.& Lundy (S&L) has issued Engineering ,

Change Notice (ECN)-4807 with instructions for. welding structural steel on.embedments. A note (ECN 4807) was added to the general structural steel notes on drawing S21-1001 which gives criteria for when special_ caution applies for welding to'embedments.

"Baldwin Associates Technical Services (BATS) -will include these instructions in each traveler that is issued for structural welding to embedments.. These instructions will minimize the buildup of localized thermal stresses by-controlling welding 4

sequences and parameters.

All corrective. action required to correct the identified deficiencies and preclude their recurrence have been completed.

Safety Implications / Significance

? -

Illinois Power has. reviewed the findings of this

investigation, and has determined that the matter of the broken S&L has determined studsLrepresents a-significant safety issue.

that>the loading on the beams, (numbered 104-and 103), attached 4

to the embedment: plates at AZ 128' and 140' respectively is such uo that the capacity of'those plates with broken studs was L ~ insufficient to sustain the total Design Basis Accident loads.

p Repairs were' required and the-use of the special Drillco devices f twas. required on beam 104 due to the.high reaction loads. Beams-t' 103 and 104-are part'of the support for the containment annulus l- . deck at' elevation 755'. Safety related equipment is, in turn,

! supported by this decklarea. Even without a rigorous analysis,

. postulated failure of these embedments and beams is considered unacceptable 1and represents a significant safety issue. This issue is therefore considered reportable under the provisions of 10CFR50.55(e).

Page 3 of 3