ML16341D909

Informs of Agreement Made During 860924 Telcon for NRR to Assume Lead Responsibility for Technical Assessment of Resident Inspectors Findings Re Spent Fuel Racks.Nrr Should Make Technical Judgement Re Weld Length Deficiencies

ML16341D909
Person / Time
Site:	Diablo Canyon
Issue date:	09/24/1986
From:	Kirsch D NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
To:	Holahan G Office of Nuclear Reactor Regulation
Shared Package
ML16341D910	List: ML16341D909 ML16342C336
References
NUDOCS 8610150369
Download: ML16341D909 (26)

Text

<>q men>

~c+

Wp

+

0 n "+*++

uNITED STATES NUCLEAR REGULATORY COMMISSlON REGION V 1450 MARIALANE,SUITE 210 WALNUTCREEK, CAI.IFORNIA94596 ENCLOSURE 2

September 24, '1986 MEMORANDUM FOR:

G. Holahan, Director Operating Reactor Assessment Staff FROM:

Dennis F. Kirsch, Director Division of Reactor Safety Projects Region V

SUBJECT:

TECHNICAL ASSISTANCE REQUEST ON DIABLO CANYON UNIT 1

SPENT FUEL RACK FINDINGS NRR has agreed, during a telephone discussion between H. Schierling and P.

arbut on September 24, 1986 to assume lead responisbility for a technical assessment of the resident inspectors findings regarding the spent fuel racks.

Specifically, the resident inspectors have examined the accuracy of the fuel rack as-built drawings generated by the licensee and used in the structural analysis recently reviewed by NRR in response to our September 8,

1986 request.

The results of the resident's inspection have identified a few instances where drawin t e actual length of a weld is somewhat less than that recorded on th

-b il ng.

Generally, the as-built drawings appear accurate within normally e as-u recognized measurement tolerances.

In the first resident.

sample two of 64 welds were short of the as-built recorded e

s y

an Z respectively.

These welds were cases where one of several welds in a single connection were short so that the total connection may not be affected to th o

he same degree.

The licensee has evaluated'hese two cases and has determined that. the shorter lengths are acceptable and within design margins.

Region V is performing further sampling to confirm the as-built deficiency rate but considers the deficiency rate thus far is within normally acceptable rates using Nuclear Construction Issues Group (NCIG) guidance documents.

Since Region V does not have staff with extensive analytical background and since NRR has'ecently reviewed the licensee's techniqu R i U

n ques, eg on requests eb101.503~9 05000275 8b100~

pDR goQCK PDR)

P

0 0

G. Holhan Page 2

that NRR make the technical Judgement as to whether the licensee's positi (that

(

at the identified weld length deficiencies are within normal. design margins) s pos on, is reasonable.

up~

Dennis F. Kirsch, Director Division of Reactor Safety and Pro)ects Region V

cc:

S. Varga, NRR R. Ballard, NRR H. Schierling, NRR ~

J. Martin A. Chaffee M. Mendonca P. Narbut, SRI, Diablo Canyon TIA File (N. Western)

ENCLOSURE 3

SKETCHES OF WELD LOCATIONS FROM PG8IE CALCULATION PACKAGE 52. 15.36. 1.2

4~

4

HO%K-VKLQS f THRU l6 PJK AL

~ GUSSET PLATE.

NKm5 lVTHku 22 ARE ANGLE YO GUSSET H.WTE.

I gl

)

II ql l

.I t

I I

1 I

i I

I I

1 I

I IQ Ii)

I II I

I Ii t

I) 0<

MO t

I t

I I

I

~

~

~j I

~ ~ f QADI L!-

P GETAlL Cq ~ H,ALL FAt.'ES,REF. Okla. +0't6t 1,~ECttOH A j C.

r A

~ rC r

Y

,'fj

i9

~1

/AC:0 5,OEfAlLA, EAST 4 H65T, F El= Ohlb. 4596.17,+ECTW+6, DETAL Z.

~~( g.- <<<

~'-n'-'

9 o

s Q

g 9

~,

I

e n

ENCLOSURE 4 RESE'ENSE K) THE NRC QU1:STION CN DCPP SPAZ-FOEL RACKS AS-BUILT WELD QCNFIGURATION o

NRC reinspection of as-built weld configuration of spent-fuel racks detected deviations in 3 weld segments out of a 170 segment sample size.

The total population of weld segrrents applicable to spent-fuel rack structure connections is over 3600 weld segments (approximately 15 welds/connection x 8 connections/rack side x 4 sides/rack x 8 racks 3840 welds).

Allweld samples were selected by the NRC Inspector.

Since the 170 weld segment samples are less than 10% of the total weld segment population

( = 3840), it is justifiable (Reference

1) to use the bincxnial distribution to ccapute the prcbability of 3 weld segrrent deviations out of the 170 sample size using the following expression:

p(w) -

(i-p)

~l (n-m) 1 where p is the probability of weld segnent deviation; n = sanple size; m = number of deviations in the san@le.

The confidence level of encountering less than or equal to m weld segment deviations, can be ~ted fram Hg.

(1) as follows:

c =

l - Z, P (")

l;=o Using 95% confidence level and the probability of weld aviation of no rrare than 5%, Eq.

(1) and (2) yield number of acceptable weld variations for a sample size.

Table A sutmarizes these results.

It

[32/26]

shears that 3

variations in a sample size o 70 satisfy the acceptance criteria.

Based on the above, PGandE concluded that:

o PGandE's documentation of as-built welds mat the acceptance criteria (95% confidence level with no narc than 5% variations) widely used in the nuclear industry.

o The quality of weld did not show any variations.

o The three deviations in as-built docurantation were determined to be acceptable with ample margins.

'Iherefore, PGandE's as-built documentation program for the weld inspection is adequate.

Reference:

Korn, G. A. and Korn, T. M., "Mathematical Handbook for Scientists and Engineers,"

2nd Ed. ~aw Hill Book Co.

(Page 625).

[32/26]

~

~

~ 0 TABLE A STATISTICAL SAMPLING PROGRAM BY M'HGBUTES Sample size (n)

Accept if the number of'eld variation encountered (m) is less than or equal to 58 0

92 123 152 180 207 233 259 285 310 335 10 360 NClZE:

Confidence level (c) = 0.95 Accept if variations are less t~ (p) 5%

t32/26]

PGandE Letter No.:

DCL-86-285 Septem 26, 1986 TABLE 1

DESIGN SUHMARY OF CHANGED RACKSl

~~<(CG,K 5 r

n m

nn CalcUlated Stresses or Interaction r

R i

FSAR Criteria 1

w 1

Are Criteria M

-1.

Bracing 1.1 Diagonal 1.2 Horizontal 2.

Cell System 2.1 Lacing 2.2 Corner angle 2.3 Interior angle 3.

Diaphragm 4.

Anchor Bracket 4.1 Bending 4.2 Shear 5.

Held Stresses 5.1 Anchor Brackets 16 ksi 0.88 23.4 ksi 0.9 23.3 ksi 0.9 22 ksi 9 ksi 28.3 ksi 1.0 28.3 ksi 1.0 28.3 ksi 1.0 28.3 ksi 16.3 ksi Yes Yes Yes Yes Yes Yes Yes Yes a'.

To embed 31.6 ksi b.

To rack 35.1 ksi 5.2 Intersecting 20.9 ksi diagonal 40.8 ksi 40.8 ksi 40.8 ksi Yes Yes Yes 1096S/0047K 4

PGa Letter No.:

DCL-86-285 September 26, 1986 DESIGN SUHHARY OF CHANGED RACKS (Continued) ri in Calculated Stresses or Interaction r

R FSAR Criteri a Are Criteri a SJII 5.3 Modified diagonal connection at base of Rack 8

5.4 Repaired lacing in Rack 1

6.

Strain in Liner at Embeds 14.8 ksi 16.8 ksi 40.8 ksi 40.8 ksi Yes Yes 6.1 Membrane

.00021 6.2 Membrane and

.00075 bending 7.

Embeds 7.1 Single (8"x8")

.003

.010 Yes Yes a.

Plate bending 0.85 1.0 Yes b.

Stud 0.61 c.

Concrete 3.5 ksi bearing 7.2 Single (8"xl3-1/2")

a.

Plate 0.85 bending b.

Stud 0.85 c.

Concrete 3.5 ksi bearing Connection 8.

As-Built Helds 1.0 9.6 ksi 1.0 1.0 9.6 ksi Connection Yes Yes Yes Yes Yes

~ -Are -Criteri a a.

Rack 2 R/

b.

Rack 5

c.

Rack 6 ~/

1096S/0047K 37 kips 3.2 kips 37 kips 64 kips 45 kips 78 kips Yes Yes Yes

1

PG Letter No.:

DCL-86-285 September 26, 1986 DESIGN

SUMMARY

OF CHANGED RACKS (Continued)

Calculated Stresses or Interaction r

R i

Cr'iteria All w 1

Aie Criteria

~Rm >~rk II.

~l l.

Top Corner 2.

Top Corner 0.38 inch 0.50 inch 2.0 inch 4.1 inch Yes Yes ri i

III'. ~nisi; 1.

Global Rack Model 2.

Individual Cell Model 3.

Diaphragm 3.1 Top Q~mm

~~n 216 nodes, 6 de-grees of freedom each (1296) 411 beam elements 32 truss elements 17 to 19 boundary elements 37, nodes, 6 de-grees of freedom except at the boundary (210) 69 beam elements 140 nodes, 5 de-grees of freedom each (700) 256 beam elements 4 truss elements 4 boundary elements FSAR Criteria 11 w

1 None None None Are Criteria N/A N/A N/A 1096S/0047K i

PGandE Letter No.:

DCL-86-285 Sep mber 26, 1986 DESIGN SUHHARY OF CHANGED RACKS (Continued) ri i

n III. ~nl~i 3.

Diaphragm (Continued)

Calculated Stresses or Interaction r

R Criteria 11 w

1 Are.

Criteria 4.

3.2 Bottom Stiffness 4.1 Diaphragm (in plane bending) 140 nodes, 5 de-None grees of freedom each (700) 256 beam elements 4 truss elements 8 boundary elements 530 kips/inch (EH)

None 700 kips/inch (NS)

N/A N/A 4.2 Fuel Cell Stiffness 13 kips/inch None N/A 4.3 Rotational Stiffness of Anchor Bracket 54,000 kip-inch/

None radian (Accounting for the flexibi 1 ity of the embed)

N/,A NOTES:

1.

The stresses/ratios are based on the Hosgri earthquake; the DE and DDE do not govern the design.

2.

3.

Strength contribution of existing welds was neglected; qualification of this. connection was based on welds for an added bar.

Held is one of several welds contributing to connection. capacity of 78 kips.

1096S/0047K 17-