ML20058G202

Responds to NRC 820625 Questions Re Use of ITT-Grinnell Pipe Clamps.Several Items Require Testing Programs Which Could Not Be Completed Prior to Response.Commitments Re Completion of Items Encl

ML20058G202
Person / Time
Site:	Satsop
Issue date:	07/30/1982
From:	Bouchey G WASHINGTON PUBLIC POWER SUPPLY SYSTEM
To:	Miraglia F Office of Nuclear Reactor Regulation
References
GO3-82-760, NUDOCS 8208030233
Download: ML20058G202 (11)

Text

O Washington Public Power Supply System P.O. Box 968 3000 GeorgeWashingtonWay Richland, Washington 99352 (509)372-5000 Docket No. 50-508 July 30, 1982 G03-82-760 1

Frank J. Miraglia, Chief Licensing Branch No. 3 Division of Licensing U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Subj ect:

NUCLEAR PROJECT 3 WNP-3 RESPONSES TO NRC QUESTIONS ON PIPE CLAMPS

Reference:

a) Letter, F. J. Miraglia to R. L. Ferguson, dated June 25, 1982 The reference letter transmitted a set of questions regarding the use of ITT-Grinnel Pipe Clamps at WNP-3. Responses to these questions are included in Attachment I.

Several items in the list of thirteen questions require testing programs which could not be completed prior to submittal of this letter. Commitments for completion and reporting of these test programs are included in Attachment I.

If you require additional information or clarification of the responses. in Attachment I, the Supply System point of contact for this action is the WNP-3 Licensing Project Manager, Mr. K. W. Cook (206/482-4428).

0f

/

9 G. D. Bouchey, D puty Director Safety and Security GDB/KWC/ sis cc:

J. A. Adams, NESCO D. Chin, Ebasco - New York Ebasco, Elma D. Smithpeter, BPA WNP-3 File, Richland 8208030233 820730 PDR ADOCK 05000508 A

PDR w

g e

y-

a ATTACHMENT I WNP-3 RESPONSES TO NRC QUESTIONS ON PIPE CLAMPS Question #1 To meet Code case R-249 requirements, there should be requirement on the heat j

treater of a maximum ultimate tensile strength or a range on ultimate to re-1 flect the ASME code requirement and the material specification requirements (i.e., UTS maximum 170 ksi; UTS range 145-170 ksi)

Answer #1 We believe that imposing a limit on measured ultimate tensile strength of 170 ksi for the strap material on the subject clamp (A 693 GR. 630 age hardened at 1075 F) is overly conservative. The Code, in Appendix I, allows the use of SA 584 GR. 83C, in various heat treat conditions, and the H 1075 condition which is chemically and physically identical to ASTM A693 and there is no up-per limit imposed on the maximum measured tensile strength. Further, where the code does impose a maximum tensile strength limit, the limit is set 40 ksi over minimum specified tensile strength. One specific example is SA 354 GR. BD bolting (A1S1 4037, 4340) which has a minimum ult, tensile strength of 150 ksi which is allowed to go to 190 ksi. This is a high strength bolt one would typically expect to see used in an application where assembly stresses would be significant. Therefore, the concern appears to be with stress cor-rosion cracking in materials where tensile strengths far exceed expected min-imums. Since the material supplied to the jobsite was purchased as A 693 and recertified to SA 564, a code approved material, we feel that a more reason-able upper limit to impose at this time would be 185 ksi, which is 40 ksi over the specified minimum.

Question #2 What is the effect of the strap preload on the piping? Specifically address for various pipe diameters and schedules what, the additional stresses in-duced in the pipe, could be as a result of maximum pressure and temperature loadings including thermal transient conditions.

Answer #2 The strap and yoke preloading on the pipe (to provide the required spring rate and to insure no lif t-off during any operational event) will induce pipe wall stresses. The magnitude of such stresses will be determined by test for particular frame sizes that are attached to specific line sizes and sched-ules. An additional torque will be applied during test to account for the pipe diameter expansions due to an increase in temperature. The effects of pressure in terms of pipe diameter expansion are minimal. When the level of stress due to pressure is approximately 15,000 psi the increase in diameter is equivalent to a tgmperature increase of 17 F.

The test procedure will in-

clude the use of rosettes of strain gauges at definitive locations around the pipe to yield information regarding stresses in the hoop and longitudinal direction. A representative number of tests will be performed to assure that the significant cases have been considered by measuring the induced stresses caused by step changes in the applied torque up to the maximum. The therraal transient (gradient) conditions need not be separately addressed since the clamp is in intimate contact with the pipe and the strap is sufficiently thin to allow it to increase in temperature in concert with the piping.

It is anticipated that this testing program will be completed within six (6) months. The test results and accompanying analyses will be available by February 1,1983.

Question #3 How does the applicant intend to factor the additional piping stresses due to the stiff clamps into the design process?

Answer #3 The added stresses on the pipe wall due to clamp torquing will be factored in by classifying the test results in appropriate groupings and combining them with the most critical stress analysis results that occur at clamp locations for particular system calculations.

Question #4 The applicant must provide assurance that a tolerance has been established on torque values to ensure that:

a) minimum torque values will result in the required strap preload requirement and b) maximum torque values will not overstress the strap for all operating plant conditions.

Answer #4 The torque values specified in the installation instructions are those deter-mined through test to be required to meet stiffness requirements. These values are readily measurable and tolerances on these values are not neces-sary. Any deviation from the specified values will be solely a function of torque wrench calibration which is normally a minimal amount and should not adversely affect either the spring rate of the clamp or the calculated pre-load stresses or operating stresses.

4

_,.. _. -.. ~.,,,.

A tolerance which would not affect performance of the clamp would be:

Frame Size Tolerance 1-3

-9%, + 5%

4-6

-5%, + 0%

Question #5 What is the margin for lif t-off at service level D limits?

Answer #5 When torqued to the specified values, testing shows that the frames will not lift off the pipe at a load less than the level D load.

" Lift-off" will oc-cur at loads fairly close to the level D service limit.

In the testing pro-gram there was no requirement to record these loads if gapping didn't occur before reaching the faulted load; therefore, specific data is not available.

At WNP-3 the maximum Service limit is C (Emergency). Therefore the maximum operational loads that will be applied are well below the Level D limit.

Question #6 Provide the maximum piping temperature and pressure where these stiff clamps will be used. Address both stainless steel and carbon steel piping conditions.

Answer #6 The maximum temperature and pressure of the piping systems where the stiff clamps will be used are shown below:

PIPE MATERIAL PRESSURE PSI TEMPERATURE F Stainless 550 350 2250*

653*

Carbon Steel 1800 110 1170 454 1143 563

• This condition exists on four clamps only. Modifications or substitu-tions will be made if undue stresses are determined at these locations.

Question #7 The applicant should provide a discussion as to the effect of load rating the i

clamp at maximum piping pressure and temperature.

~

Answer #7 Load rating of the stiff clamps at operating pressure will have no effect on ultimate failure load. This is based on having performed tests on clamps that were purposely overtorqued and found that there was no change in failure load. The additional torque would provide additional strain to the straps simulating any radial expansion of the piping due to pressure. The reason there was no change in the failure load is that the additional strain simply increases the force required to lif t off the pipe, and until this load is ex-ceeded, the straps see no further increase in stress and the ultimate load capacity of the clamp will be attained at essentially the same load.

The difference in testing at operating temperature would only be that the ul-timate test load achieved would be lower by an amount proportional to the ul-timate tensile strength at room temperature and operating temperature.

This is not a code requirement and would still yield the required stiffness.

Question #8 The applicant should address the possibility of preloading relaxation due to thermal and pressure cycling of the piping.

Answer #8 In response to this question, material has been ordered for testing. The clamp will be fastened and torqued to a section of stainless steel piping and heated to 650 F, it will be held at that temperature for an hour and then air cooled to room temperature. This will be repeated for ten cycles and the applied torque checked after each cycle.

Since the radial expansion due to pipe pressure is minimal, as expressed in the answer to question 2, the cycling of the clamp at a higher temperature than is recommended will account for this event and will be conservative.

Material for this testing is due to be in the test lab in August. Testing will begin innediately and take approximately one week; a report will be generated upon completion of testing and will be available October 30, 1982.

Question #9 Provide the materials that were used in the pipe clamp yoke which were tested.

Answer #9 The material used in fabricating these frames was SA 515 GR 65 or 70.

Question #10 Provide the load capacity data sheets and the dimension of the straps and rivet holes for frame size 67 4

~ -. - -

y,,. --

Answer #10 Attached is the Load Capacity Data Sheet for the size 1-6 stiff clamp. Also attached is a copy of a sketch of the frame size 6 strap and rivet holes; the only dimensions not given are those that are a function of pipe diameter, all critical dimensions have been included.

Question #11 What is the effect of the additional piping stresses due to the stiff clamp on the postulation of high energy line breaks?

Answer #11 The induced stresses due to the application of stiff clamps on high energy lines to determine the postulation of high energy line breaks will be evalu-ated. The clamp effects will be combined with the global pipe stresses at 4

the most critical node points in areas of concern to determine the location of a postulated break. This will be compared to the original analysis to un-cover any discrepancies that may result.

Since all clamps are on straight lengths of pipe and since the high stress points in the system are usually at fitting locations it is expected that no impact to the break location will occur due to the induced clamp stress.

Question #12 What is the effect of the local pipe wall stiffness on the clamp assembly?

Answer #12 The local stiffness of the pipe will act as any other spring element placed in series with another spring - i.e., the linear stiffness of the piping would actually reduce the actual stiffness of the clamp. However, the governing stiffness within the complete structural framework of the support /

restraint assembly (pipe, clamp, snubber, steel framing etc.) would be the snubber.

Question #13 Appendix Table 1-13.I has for 17Cr-4Hi-4Cu to SA-564 (footnote 5). The tem-perature of exposure for which these are designed cause embrittlement of this materi al.

Provide a fracture mechanics rationale as to the acceptability of this material in the high humidity and high load conditions in which these clamps might be exposed. The stress levels in the rivet hole area should be specifically addressed.

Answer #13 Our evaluation to substantiate the suitability of this material, in its in-tended use, is still underway. A complete response will be provided by October 1,1982.

1 1

-,,. ~-

...,.-.--.n,.

-n.

ITT GRINNELL Miof' PIPE HANGER Olv S!0N ASME SECTION 111 i

stJBSECTION NF LOAD CAPACITY DATA SHEET

l. GENERAL $ FORMATION SUPPCRT NA.ME >~

EhtihEIR!NG FARTS t.15T(5):

PE 41 Pipe Clca>

APE 4100000

$1s 16 VPP$$ thdu 3 6 5 SUPPORT 7tPE Cocpencat Stadard C001 CLASSIFICATION 41 ass 1,2,36HC OESIGN FROCEDL'RE.Wd utiM. DESIGN TEldPERATURE' 6500 N I

I SERVKE UlillT$i Isis A B.C.D & Desip Loading

2. CERTIFICATION l

ENGINEEni CATI:

THE tcA0 enEITY of Txis CAP 0titut

$1lPFC4T 15 DTED IN ACCCAMNCE *1TH

[0 f.p gj stitsEC701 ttE,----

U;nn t. "':u*. ;':

• s l

i.-

- CHECKED:

CAT Ei l

g, 1 '16. h W l:'

I

~

k 4

[

k

..i:)

LOCC,IDH 0F SUESTANilATION DATA i

r:

imanb, Development & bgiraring

...: ;.,. m

• Pip Bapr Divisien, hev., R.I.

ETE3tD PQ E G'ER

3. REVISION AND RECERTIFICATION LC05

' LOD IN NTE MTE MCERTIFIOTION REY. No.

AEY. NO.

I/14[1#I k1,

__ l 1/30/31 _.

i__

un_;

giig _.1 p 1

l olain l _.

ulnhi LA.1 d-4hivJA.JLi '

i f

)

5 uom i ch,In lAiUL _.___

r i

-i l

a i

i

}

.t i

3. ~

l

' ITT GnlNHEU.

W?

ts.es #

• PlPE HANGER OMSION siirr cue ASME SECTION 111 117.6 I

SUBSCCTION NF I

4. COMPONENT SUPPORT INFORM.iTION ITild WAhENAL SMCIFICAf!CH ANALYSIS SA115 CA 65-70 cr 5136 Tras4 1.e44 h tlag

~

34113 C187 cr SA 556 I,o4d latlng LoadPin(tateptSt.35Tig,g {yg30 Agt krdened i

' SA 564 Typt 630 Age Lead Pin (St. 35 rig. 306/307)

HudHd i 101)#1 Loed tating

$A36 or A!SI 2010 !stapt' Spacer par U 2121 (b)'

Secondary habr-froarc@ixtaptptr Aetalning Rir.g Wr-2121 (b)

Lead lating SA 193 41 47 Lo4d 14 ting

~

AStXI693fypeSJO Straps C:nditloo B 1375H W 4 14 ting SA 193 61 B1 W44 Ratlog SA36 or 01011 Cast hers.

716 Rcd Vasher entd-to Rot.ltwell C.35 444 Rating it.f uset Pee # tift fbi 54307 GR ) or A$75 A301 Nats 4k A*

Jad Rating SA 113 Gk 37 Med Ratlag fie Peds Ae5 $13788 or 444 hting Rivets

$44) Gr. 664 CORD, A M I

).01818.6.1 Secedsry Cc4ter FIA4 EI W t tu MF1121(b)

Beeber 8 Ccde Cast iht H ft2 ding Approval of Ccde Cast # 2 t t. 7.

PAGE 1 W 6 PE41 STIIT CIM m.6 M GUM i

PIPE MCEP. DIVISIC4 QJAl,!T PRONCf3 LCAD U,TIUG nME S12.I MXIMJH LOAD L35.

ken 1 4 ST UiG PATI nnf1n l.

uvtt.-

wat ts u A6B C

D 1

'6000 6610 11320 M3 2.,

2 6000 B610 11520 325 I

3 WO 8610 11320 325 4

20 00 24S40 26100 1000 5

20700 24340 26700 1000 1000 6

20700 24340 26700 arid ke ce stalaltss attfl pins opmths,tWw 35dT not k reviewd bT

!TI Criane!! testarth, Dcetl: pent and h(twering.

/

9 f

*k*g$*I'#

Pg FIC. 21.S!1 STITT CLAP.P TEI6s

'Q yo M ". klO*

a M ---.-

sH% e

=

\\l e

+

,i

A

-7 \\[ X

\\

v

,. - +

t ---

T

_1 -

h A

1-_

-__a 4-i i

e e

e 1

D Y

g s

__~

V

" l [

7 + '*"

2 g

g

/. '

j i

+.

t O

e t

. 1, s

~

e

- p w

g-p g

s

~c+

5 B

'~

e o

e M

t e

o O

i

\\

~

i ii -

j m

m a0 ft al e

J DN,

[

s d

9 z2 ulo Q.l

!:k

{-

-l J

+iste i i

+e o-L.g I I _. 3-0 e F2 W

N.8 7,

y[u M

o l-r l (. -

~

n y-e 4,

1 a-s a.

t j-41~'

I d

E<

ITA. !

r. 4. -

i 1

2 3

4 l

/

- 3. 9 -3 a

3

-[ h

.4J 0;

,i j

U

(

M,r j

9-3-9 /

9

/

S.

l

_Q

@/,

/

z.. Y.

O g

P.

A

-- 6 e g 'a, Ay H

o s' h,j:w r

W W

1. 7

-V d

j 4*g If, ki i

43

7. 0' 9

e O

G

,, _.