W3P87-1063, Forwards Response to NRC 870406 Request for Addl Info Re Analyses Conducted for Pressurizer Safety Valve Piping & Supports,Per 821229 Response to Item II.D.1 of NUREG-0737

Forwards Response to NRC 870406 Request for Addl Info Re Analyses Conducted for Pressurizer Safety Valve Piping & Supports,Per 821229 Response to Item II.D.1 of NUREG-0737

ML20214J834
Person / Time
Site:	Waterford
Issue date:	05/21/1987
From:	Cook K LOUISIANA POWER & LIGHT CO.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
RTR-NUREG-0737, RTR-NUREG-737, TASK-2.D.1, TASK-TM W3P87-1063, NUDOCS 8705280207
Download: ML20214J834 (29)

Text

,

Louisiama / 317 eAnonmeS18eer. e.o.eoxe0340 POWER & LIGHT NEW ORLEANS, LOUISIANA 70160 (504) 5954 100

+

$EONdysE May 21, 1987 W3P87-1063 A4.05 QA U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. 20555

Subject:

Waterford SES Unit 3 Docket No. 50-382 NUREG 0737, Item II.D.1 Response to NRC Questions

References:

(1) W3P82-4011 dated December 29, 1982.

(2) NRC letter dated April 6, 1987.

Gentlemen:

By the Reference (1) letter LP&L submitted its response to Item II.D.1 of NUREG 0737, " Performance of PWR Relief and Safety Valves".

In your Reference (2) response you requested additional information concerning the analyses conducted for the pressurizer safety valve piping and supports.

Enclosed please find the requested information, Should you require additional information, please contact Robert Murillo at (504) 595-2831.

Vey truly yours, 8705280207 870521

'ktualf d

PDR ADOCK 05000382 K.W. Cook P

PDR Nuclear Safety and Regulatory Affairs Manager KWC:MJM:ssf l

Enclosure cc:

R.D. Martin, NRC Region IV J.A. Calvo, NRC-NRR J.H. Wilson, NRC-NRR G. Hammer, NRC-NRR NRC Resident Inspectors Office E.L. Blake Ok W.M. Stevenson

'\\

"AN EQUAL OPPORTUNITY EMPLOYER"

ENCLOSURE to W3P87-1063 Page 1 of 13 RESPONSE TO QUESTIONS ON WATERFORD 3 NUREG-0737, Item II.D.1, SUBMITTAL QUESTION 1:

Additional detail is needed on the RELAPS analysis. Provide the safety valve set pressure that was used, explain how the valve was assumed to open (linearly or otherwise), what valve flow rate was used, and what calculational time step was used. The flow rate used in the analysis was assumed to be 575,371 lbm/hr., which is less than 14% above the ASME rated capacity; the tested valves delivered up to 25% above rated flow rate at 3%

accumulation. The highest expected flow rate should be used or justification of any lower value provided.

In EPRI Report HP-2479-LD, March 1982, it was recommended that the calculational time step used be less than the shortest control volume length divided by twice the sonic velocity.

If this criterion was not adhered to, justify the time step used by demonstrating that bou-ding stresses and loads were calculated or redo the analysis. A sketch of the thermal hydraulic model showing the size and number of fluid control volumes should be provided.

RESPONSE TO QUESTION 1:

The following are the additional input parameters used in the thermal hydraulic analysis:

1)

The node spacing varied from 0.67 ft. to 1.06 ft,

11) The maximum time step used was 2x10~ seconds. This selection satisfies the criteria that no front (whether pressure or fluid) may traverse the length of a control volume in one time step.

ii) The valve flow area used was 0.0235 ft The valve flow area was adjusted to achieve the maximum capacity of the valve.

For this area, the actual steady state flow rate was 576,720 lbm/hr.

iv) The SRV set pressure was 2574.25 psia and it was assumed to open linearly.

v)

The flow rate used (576,720 lbs/hr) in the analysis is more than 14%

1 I

above the ASME rated capacity (504,874 lbs/hr). The SRV rated capacity at 3% accumulation is 460,000 lbm/hr and the maximum rated capacity is 575,371 lbs/hr which is 25% above the rated flow rate.

vi) A sketch of the thermal hydraulic model showing the number and the size of fluid control columns is shown in Figure 1.

l i

4 (6,0.92) i 5 (17,0.98)

~ (4,1.04) 14 15(4,3 06) i 13 (13,1.03)

?

16 (14,1,0; 3 (8,0.95)

(13,1.02)

R1504B R1503A j

' (~4,0.,85).7 97

.iy (3,0.99) 2 N

- (13,1.02) 0.67 4 )

12'(4,0. 98')

4 (2,1.06)

(2,1.06) gg 8 (11,'1.0)

PRESSURIZER g (14,1,o) 10 -(2,0.75)

QUENCH TANK FIGURE.

1.

PRESSURIZER RELIEF SYSTEM - Waterford SES Unit 3 (Nunbers refer to CALPIDTF III Piping,Segm6nts)

NOTE: Number of volumes and the volume length in each Piping Segment are shown in enclosed brackets. For Segment No. 7, the reducer volume element length is 0.67. All lengths are in feet.

ENCLOSURE to W3P87-1063 Page 2 of 13 QUESTION 2:

The safety valves ring settings used by the Licensee needs to be specified.

If these are different than the two " qualified" ring settings in CEN-227, test data justifying them needs to be provided.

RESPONSE TO QUESTION 2:

Table 5.2A-1 of the Waterford 3 FSAR specifies the qualified ring settings

(-48, 20, 0 and -48, -60, 0) determined in CEN-227. The pressurizer safety valves use the ring settings of Table 5.2A-1.

i l

1 l

l 2

4

l ENCLOSURE to U3P87-1063 Page 3 of 13 QUESTION 3:

Verification of the CALPLOTF III program is requested.

Show that it predicts correct loading history of the EPRI test data pertinent to Waterford 3.

RESPONSE TO QUESTION 3:

The post-processor CALPLOTF III was used to convert the transient flow conditions (calculated by RELAPS/ MOD 1) into transient forces on the piping system. The derivation of the governing equations are shown in Appendix A of the Ebasco report (Reference 1) submitted in response to Item II.D.1 of NUREG 0737 (W3P82-4011 dated December 29, 1982). The validity of the program coding was verified by comparing hand calculation results against the values computed by the program. The program was further assessed against the GE 4-inch pipe blowdown test results.

Favorable comparisons were obtained between the computed results and the test data.

i CALPLOTF III was also verified by running CE test 1411 for SRV actuation on RELAPS/ MODI using the input from EPRI's RELAP5/ MOD 1 application (Reference 2).

The calculation hydrodynamic conditions were used by CALPLOTF III to determine the transient forces that almost duplicated the forces obtained by EPRI (Reference 2).

The forces calculated for the CE test case 1411 are given in Figures 2 thru 5.

REFERENCES 1.

" Analysis of Pressurizer Safety Valve Discharge Piping - Waterford Steam Electric Station Unit #3", September 1982, 2.

" Application of RELAP5/ MODI for Calculation of Safety and Relief Valves Discharge Piping Hydrodynamic Loads", EPRI NP-2479, December 1982.

l l

m

LOUISIANAPOWER&LIGhirCouPANI

~,.

WATERFORD SES UNIT NO. 3,

{

EPRI TEST 1411 CALPLOTF3 VERIF TEST lull SECMENT 1. FORCE

, o o

o.

cv i

4 o

o

=r.

cv o

o

.b.

I

.cn e

o o.

ev o

T e

o

.4 m es

-i s

m J

i w

aa:

I-oa. o o

o oo r

i

=r l

i i

1 o

F o

1 o

oc

,y

[. 00 0'.02 0',04 0'.06 0'.08 0'.10 0'. I 2 TIMEISEC) 4 FIGURE 2.

LOUISIANA POWER & LIGHT CouPMI nararam sus unir me. s EPRI TEST 1411 CALFLOTF3 VERIF TEST 1411 SECHENT 2 FORCE o

4 o

C.

o.

nn oo

.o2.

oo

~c.

w o

o nt n

N o.

T

...:8 1

nn =r.

. m tv y_

na -

U x

1 oo

u. o

.e.

c o

o J

f o

p.

o I

e.

g l

r Q

.3 i

i e

o i

o I

~

J.00 0'.02 0'.04 0'.06 0'.08 0'.10 O'.12 I

/

TIMEISEC) i

. Figure 3.

l

' LIGHT COMPANY P

gg, g, EPRI TEST 1411 CALFLOTF3 VERIF TEST 1411'SECHENT 3 FORCE o

o i

o.

(d...

1 Ln

~

I o

)

to o.

=r l

o

.o

~

c.

=r

~

.o o

m.

n N

o'

.e

.8 p.

]

m -r.

CD W

.J..

i u.r u

ceao uo e

s oo o

1 I

o

.o ca F

o 1

(l f.00 0'.02 0'.04 0'.06 0108 0'.10

,'12 0.

e TIME (SEC)

~8-Figure 4.

L

Im stas 4 pown a 3,3g,,go,,,,,

858 WIf NO. 3 EPRI TEST '1411 CRLPLOTF3 VERIF TEST 1411 SECHENT 4 FORCE e

O

. w.

w C3 a

O m.

w F

O

.a m

=.

oo

.a.

w N

'o

.e

.o wn mn a_.

IM Uxoo

u. o a.

N l

i I

e o

w.

O

-.o r-

. s.

s

(.'

s.00 0'.02 0'.04 0'.06 0'.08 O'.10 0'.12 e

O TIME (SEC)

W4.se t em 4 m

g_

ENCLOSURE to W3P87-1063 Page 4 of 13 J

l QUESTION 4:

i Verification of the PIPESTRESS 2010 code for use on safety valve opening transient loading conditions is requested. Provide a comparison of i

calculated results with an appropriate benchmark problem (a comparison with EPRI/CE data would be acceptable). The statement is made that PIPESTRESS 2010 performs a generalized response analysis and that this type of i

analysis is known to produce conservative results; verification of this i

statement is requested.

Provide a write up on the PIPESTRESS 2010 code.

It appears that this may be a static pipe stress code being used by a j

dynamic problem.

If this is true, the way the loads are applied will have a major impact on the predicted stresses. The submittal stated PIPESTRESS l

2010 performed a " generalized response analysis" and that a " selected model superposition time history analysis of pipe segments" is performed if i

predicted stresses are too high.

Because the meaning of these statements is unclear, provide a more detailed explanation of how PIPESTRESS 2010 performs the structural analysis.

l RESPONSE TO QUESTION 4:

This question relates to the PIPESTRESS 2010 computer code and its use for analyzing the pressurizer relief piping system subject to a dynamic j

loading transient caused by rapid opening of safety valves. Specifically, the following information is requested:

i a)

A general write up on the PIPESTRESS 2010 code.

b)

Clarification if a static code is being used to analyze a dynamic 1

problem, and if so, what is the justification?

i i

c)

Explanation of selective time history analysis and generalized I

response analysis, and how these analyses are performed with the PIPESTRESS 2010 code, i

j d)

Verification of the PIPESTRESS 2010 code for analyzing a safety valve transient loading condition, e)

Verification of the statement that generalized response analysis l

produces conservative results.

Responses are provided below.

4 Item (a) 1 l

The program PIPESTRESS 2010 is an Ebasco proprietary piping analysis l

computer code.

It performs linear elastic analysis of three-dimensional I

j piping systems which are subject to a variety of prescribed loading

{

conditions. The analyses are performed in accordance with the ASME-II i

classes 1, 2, 3 and ANSI B31.1 Codes as specified by the user. Reports are

}

furnished to provide evidence of design adequacy.

I i

The program constructs a linear finite element model of the piping system j

using load-deflection relationships based on the displacement method.

j Matrix decomposition is used to solve the system of equations for the l

l

,-,-------,-,....-,nm,--n-,-.,--w---,e---.~r,-,-,,---,,,,,n-,.,,n--,,,-,,,,,g-----,

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

i l

ENCLOSURE to W3P87-1063 Page 5 of 13 static problem.

The eigenvalue extraction employed matrix decomposition with matrix iteration and purification. The program includes an advanced dynamic algorithim to extract close eigenvalues and accelerate the rate of convergence.

The major program capabilities are as follows:

1)

Analyses may be performed in conformance with a specific code version, i

l 2)

Static analysis to calculate static loading conditions and/or effects, such as, pressure, applied loads, dead weight, support movements, differential settlement, cold spring, and seismic acceleration etc.,

1 3)

Thermal analysis to calculate thermal expansion and thermal anchor movement loading effects, I

j 4)

Frequency analysis to calculate natural frequencies and mode shapes, j

5)

Response analysis using either enveloped uniform support motion or 1

independent support motion technique, I

6)

Thermal gradient analysis to calculate stress distribution caused by 1

applied thermal transients, 1

I 7)

Time history dynamic analysis using either force time history or displacement time history, 8)

Fatigue analysis in accordance with the ASME B&PV Code,Section III, NB-3600 formulations to compute fatigue usage factors.

I 9)

Combination loading cases to combine components of forces, moments, deflections, stresses, and restraint loads from individual loading conditions to satisfy the load combinations criteria, and i

10)

Extensive user oriented features such as restart capability, file j

manipulations for automatic input data, links to thermal hydraulic codes etc.

j The program has an organized User's Manual.

Its theory and verification i

have been extensively documented. The program resides on the Ebasco mainframe, and is controlled by the Ebasco QA procedures.

Item (b)

I As described in Item (a), the PIPESTRESS 2010 code has extensive capabilities. The loading condition generated due to the rapid opening of the safety valve was analyzed by the time history dynamic analysis capability using the force time history determined by the RELAPS/CALPLOTF III computer codes. Thermal hydraulic analysis of the safety valve transient was performed using the RELAPS computer code. At every node point where piping changes direction, a force time history was calculated

]

which was used as an input in the time history dynamic analysis of the piping system by the PIPESTRESS 2010 computer code.

I i

I

ENCLOSURE to l

W3P87-1063 Page 6 of 13 Item (c) 1 The specific time history dynamic analysis capability incorporated in the PIPESTRESS 2010 computer program is known as the " Generalized Response /

Selective Time History Analysis". The acutal mathematical theory consists of Ebasco proprietary formulations.

[

Theoretically speaking, by performing a dynamic analysis in a time domain, a structural solution of piping responses (i.e. stresses, restraint loads, etc.) can be obtained at every time step of the input loading condition.

However, from the design standpoint, a structural solution at every time step is not required, rather only the bound or maximum design values for the total time interval are required.

i j

The dynamic time history analysis includes two parts; the first part consists of solving the dynamic differential equation of motion, and the 3

second par : consists of obtaining the structural solution by using i

l amplitudes calculated in the first part. While the first part has to be j

executed cc every time step, as mentioned above, the second part is not required at every time step if one is only looking for maximum design valt es.

l The PIPESTRESS 2010 computer program constructs a conservative bound I

cotution known as the generalized response solution which can be used for 1

design purposes. At the same time the data calculated in solving the j

governing dynamic differential equation of motion is stored so that a time history of any response can be obtained, if required by the user.

From a practical standpoint, for a given piping problem, analysis results are severe at only a few node points. The PIPESTRESS 2010 code allows the user to extract the true time history of solution only at these points. The j

ability to extract time history on a selective bnais is important because unnecessary computer time in calculating a structural solution at every time step for each node point is avoided.

1 l

Item (d)

The PIPESTRESS 2010 program theory and verification aru extensively documented in proprietary manuals.

The standard methods of verifying computer program are as follows:

a)

Benchmarking against the published data, which is accepted in the q

public domain, j

b)

Comparing the solutions to sample problems by the program to be l

verified, and another computer program in the public domain.

l c)

Running verification problems which are judiciously formulated so that the characteristic of the solution is known based on the basic theory.

j The computer results are reviewed against theoretient predictions to establish the program validity, i

d)

Comparing solutions against hand calculations.

i 1

i

i" ENCLOSURE to j

W3P87-1063 Page 7 of 13 All four methods of verification, individually or in combination, have been I

adopted for the PIPESTRESS program. The choice of method (s) is based on a review of available options when a specific capability was incorporated in i

the program.

Benchmarking against the published data is a preferred i

method, however, in some cases it was not possible to use this method because no data was available. For a major analysis capability such as the response spectrum analysis, benchmarking was performed against NUREG i

CR-1677.

The time history dynamic analysis capability which is used for analyzing the safety valve dynamic transient was incorporated in November 1980, and in that time frame, no standard benchmarking data was available to verify this particular analysis capability. Therefore, method (a) could not be employed. Between (b), (c), and (d), Ebasco used method (c) because the 3

basic dynamic theory lends itself to judiciously selected problems whose r

solution characteristics can be predicted in advance. Although originally I

method (c) was used to document the program, we have augmented this

]

verification using the approach given in (b).

i f

VERIFICATION TEST CASE 1 1

Problem Definition Problem No. 1111 of the PIPESTRESS 2010 library of verification 2

problems was selected. The piping configuration is as shown in the attached Figure 1.

It consists of an 8" diameter X 0.375" thick pipe layout in three dimensions, with a skewed restraint at node point 5.

I The restraint orientation has all three direction cosines. Node points 1 and 8 are terminal anchors. Three sinusoidal time history forcing functions, with frequency very close to the fundamental period of the piping system, were applied at node points 3, 5, and 6.

The i

]

directions of the dynamic forcing functions are as shown in Figure 1.

i j

Results and Evaluation i

According to the dynamic theory, a set of sinusoidal forcing functions with frequency equal to the first mode frequency of the piping model would produce resonance. The mode shape vector for the first mode

.}

shows maximum excitation in X force at node point 4.

Time history of i

X force at node point 4 as calculated by the program was plotted as

{

shown in Figure 2.

The resonance was observed which indicates that the program is executing results as predicted by the theory.

2 VERIFICATION TEST CASE 2 i

i Problem Definition For the same test problem, instead of applying a sinusoidal forcing function, a step forcing function was applied at node points 3, 5, and 6 as shown in the Figure 1.

Separately, for the same piping configuration, a static solution was produced.

l 1

i i

f

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

m_

. -. ~. -. - -

. =.

ENCLOSURE to W3P87-1063 Page 8 of 13 i

l Results and Evaluation It is obvious that after some time, the dynamic time history solution should damp down to the static solution due to the damping factor.

l Time history of force X at node point 1001 was plotted as shown in Figure 3.

As expected, the final value of the force in member 1 is identical to the static solution.

VERIFICATION TEST CASE 3 Problem Definition l

The dynamic piping problem for the verification test case 2 was analyzed using the ADLPIPE computer code.

J Results and Evaluation l

Table I lists maximum responses at representative node points as calculated by both programs. Resultant forces and moments are within j

1%.

Based upon test cases 1, 2, and 3, it can be concluded that the dynamic time history analysis capability incorporated in the program executes correctly and produces results consistent with the dynamic theory.

l 1

Item (e) 1 As explained in Item (c), the generalized response solution is not the true time history solution.

It represents a cost effective way to avoid

]

unnecessary computer time to produce time histories at each mass point.

The program includes two methods for producing the generalized response solution: absolute bound and optimum bound. The absolute bound method I

generates a very conservative solution which is normally not used. The optimum bound method produces a solution where the resultant values of forces, moments, stresses etc. are within the engineering design range of the true time history results. Table 2 provides values of resultant forces f

and moments for the same sample problem as Item (d), verification test case j

3, calculated by the two methods as well as the true time history. The 1

table shows that generalized response optimum bound solutions are within the acceptable design range of the true time history solutions.

l 1

)

l I

i i

i

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

FIGURE 1 VERIFICATION CASES 1. 2 and 3 PROBLEM No. 1111 8

i M

1007 7

F3 o 10 01 1006 SKEW 6

o2 RESTRAINT 5

l002 F2

1009, F,

3 o

9 fs

(

PROB lill 2

ioooo siw wt ioooo -.

icooo g

2, F3 g

2, F3 FORCING FUNCTIONS F F

FORCING FUNCTIONS F F

VERIFICATION CASE 1 VERIFICATION CASLS 2 & 3

-__-~

e lr son casti

(-

x_g,~,._>____,_

3 h

o "i limaggag r n

~

FICURE 3 VERIFICATION CASE 2 MODAL SOLUTION FOR STEP LOAD h-tutor nosses

(

i i

i X FORCE AT POINT 1001 IN MEMBER 1 FOR STEP LOADING l

g DANPING/CRITICR. =.05 l

~O l

A

-9 l

5 m

l h.

to E-[

l'.

2'.

3'.

4 s'.

s'. T I ME SEC.

,o.

Y l

J I

j TEST VERIFICATION CASE 3 TMH 1 COMPARISON OF THE RESULTS CALCULATED BY PIPESTRESS2010 & ADLPIPE PROGRAMS FORCES IN Lbs, M0tGNTS IN Ft Lbs, STRESS IN PSI l

4 i

I NODE POINT 1 NODE POINT 8 NODE POINT 4 REMARKS COMPONENT.

PIPESTRESS ADLPIPE PIPESTRESS ADLPIPE PIPESTRESS ADLPIPE l

l F

9613 9620 9364 9465 6959 7037

% Diff. in Resultant x

'*' d Force F 11273 11266 3060 3198 1271 1224 Y

Stresses are within 1%

l F

3819 3616 11984 11813 1959 1%8 z

Fr 15299 15249 15514 15471 7340 7409' t

Mx 36775 36093 30778 31445 1428 840 M

9426 9335 85665 86443 11348 11303 y

Moment Mz 87858 87542 8188 8200 17540 17558 i

j Mr 95710 95150 91393 92343 20940 20898

{

Stress 57970 57511 55355 55712 28212 28221 l

1 I

i i

I i

i I

i 1

j

~. -

- - ~.

j 1

i i

j i

TABLE 2 l

I COMPARISON OF TRUE TIME HISTORY VS. CENERALIZED RESPONSE SOLUTIONS I

i FORCES IN Lbs, MOMENTS IN Ft Lbs 1

i 1

i 7

]

RESULTANT PORCES RESULTANT DEBENTS i

TRUE GER. RES GEN. RES TRUE GM.RES GEN. RES REMARKS i

DATA POINT TDE NISTORY OFTDEM ABSOLUTE TDE MISTORY OFTDEM ABSOLUTE i

1 15178 14974 34351 94886 100778 123882 J

i 1

4 6101 5359 17457 20812 20462 28273 i

i i

j 8

13462 15266 36071 91245 95044 117256 i

i

?

]

I 4

I 1

s l

i j

4 r

.,m.._..

y m_

_ =. - - - - -

i ENCLOSURE to W3P87-1063 Page 9 of 13 l

QUESTION 5:

t l

Provide detailed information necessary to evaluate the PIPESTRESS 2010 i

l piping model used in the analysis.

If this is a dynamic analysis code, this would include the spacing of the lumped masses, computational time i

step, damping, how the loads were applied, support restraint information (including what gaps and tolerances were considered, etc.), and the frequencies considered in the analysis. The cutoff frequency or range of frequencies considered should include frequencies up to 10011 or justification provided for a lower value. The lumped mass spacing and calculational time step used should be consistent with the 100 Hz cutoff frequency or justification provided.

The damping factor used should be i

1% for normal through upset conditions and 2% for the faulted condition; if t

higher damping was assumed justification should be given.

RESPONSE TO QUESTION 5:

The mathematical model of the pressurizer relief piping nystem is shown on i

the attached pipe streso isometric no. RC-204-4.

The stress isometric provides the definition of the piping configuration, rentraint locations and functions. Responnes to various questions are provided below Lumped Mann Spacing j

Each critical node point on the phynieni piping layout is specified an a i

lumped mass point in the analysin. Generally these points include all fittings, welds, restraints, nozzles, etc.

In impicmenting this requirement, the spacing is automatically created nuch that frequency 7

between two node points as a beam element in larger than the cut-off l

frequency.

In the case of the subject piping model, the maximum spacing on the 8" diameter pipe is shown as 6'-6".

For a simply supported pipe beam, this amounts to a frequency of 1140 liertz which is much larger than the cut-off frequency.

The mass point spacings for 6" and 12" diameter pipes generate even larger frequencies.

Computation Time Step i

Computational time steps are the name as the RELAP model.

Damping Factor i

Damping factor used in 1% of crittent damping, j

1.oad Application Thermal hydraulic forcing functions were applied at the same locations an the RELAp/CALp!.0TF III model.

Support /Rentraint Information Support /Rentraint locations and functions are shown on the isometric. The gaps are annumed to be zero, consintent with the linear elastic analysin.

ENCLOSURE to W3P87-1063 Page 10 of 13

\\

Cut-off Frequency l

A cut-off frequency of 156 IIertz was used in the analysis. Contribution of the rigid modes was included.

Tolerance Critoria Being a safety class I calculation, the as-built strena analysis was i

performed with analynia configuration identical to the as-installed configuration.

i i

P i

l t

k i

4 i

i

-R'

.... _ m -

FIG. Al-l 21 5

a 3

• <,4,

r'ab 2

I s

m I"

a

i 2 3

0 L

y

,i i 1 5 +

e u

ug A,

46 v

• +

e

. e.;

o o:s u s e w[ ra a a n bk [ b 8+

/

~~<a r.55 j '. 13 1 E d u.

na o o

a ;

sy a

u.. ze a

e a-sw h

w., n e

S S s.t ud:

m::

=: g e,

i 8 $

2 "q

eE e

P 2*

V u3 t

,t

.o h

$ $ d 48

[

hi h

+g g 4

?

x o B

U

. q,.\\4

,f g

1. '#/,

S

+9 15 5

EE 4

T

'9 "g

a e s g

'#- g 9f.j, 8 fg 5 0. pn=--

se a t n*

o~ ~o d i 8

a g

e

=

n.

e y \\c,p,j ee n

• g G

L* '

e q, t

s e,.,.4 s

+

3 s.

s u

o

, a,, j,

\\ lr h.

be l

E y

r c9 p/.0/

,_ t f-t 4

e e

,i,*

E y

M', 4 p;f

1. 79?y ?Q:

f

\\

I j

s

?.'

s'.

y.

c

@ap J

'A c

.n a

fg A 'o g

o

.p%

0.>./ V y 9g 7w a.

pd e,.

s..e s,

h, k, %, N g*,,

=

2 4

-p i' 4 3

14 0 M

'^

o 4

,\\

d4 3

/ ' ~md i :s

• N

,D%

j -I,

\\S#

s fn.,.s o

i g

.. y-

%, %. a s,% h-- A

,A

+

g.

q s>' \\ s,g J.c#. W f.

~ 2 e

'V 0 3 ---

q, ' '

dE 9

r e

r~

1, e

c, + '\\%

9 f*

t

.<,\\ f

.c

.\\

/

g^

y.,

u g

4.%.

\\ k

• )

.f!

. 4,.7,.h +,, s exaw"'g 3.a, 4-p t

x a

numur?

_~.q...,,w.....- s,.y

. y w

s.

,c g

9,;ge

• v:ptm,A

'g s$e e~% o %

i c

4

%m a

i

'\\rQ-N i

.o e s ci m

e, n

g.,9 n y>x,.. n.; y, c-n a s s s em

.o, w,

,. Wig 15s

%r,y. g ' J A M, Y... g 6 l

=

1. d% * %'

dg" R

. 'an $

=

jh m m h

? ds "

?hl H "\\

0

!!s Q i s

e b

n.. = -....

n..e. a.

~

a

__. _.... _ ~_-

=

4 ENCLOSURE to W3P87-1063 l

i Page 11 of 13 QUESTION 6:

t i

Provide a table showing the highest support loads (or stresses) predicted j

compared to the support rated or allowable capacities for the most critical or highly loaded support locations.

Identify what criteria or standard was used to determine support allowable 1cada.

RISPONSE TO QUESTION 6:

j The attached table provides a comparison of the calculated support loads for the various loading conditions with the maximum allowable loads during 1

[

the normal and upset and faulted condition.

Each of the supports shown on Isometric Drawing No. RC-204-4 is shown on the table.

I l

1 1

i i

1 1

1 j

i i

l i

I l

l i

A i

i i

l I

-w

=

w we--v----=*mme-eew p-w w e--w+-e,g-r--

ye---.,--%

e-,.e

--+,_.,.-sy,-igwp

--w~g+,--.w-er--e--g gwverw 9

-w*s:---,--wy.y-+-. -

y-4 y

-,y----%+

e--

0231y 5/8/87 Page 1 WATERFORD SES-3 NUREG-0737 ITEM II.D.1 SUBMITTAL TO QUESTION NO.6 SA. CALC.11163 RUN DATE 9-13-82 S.NO.

h\\RK NODE LOAD CASE NOS.

SUPPORT CRITICAL PART & MAX. ALLOW. LOAD NO.

Pr.

21 115/118 503 504 710/715 720/725 NORMAL 6 UPSET FAULTED 4 DIR.

1 RCSR-257 4FX 0

0

-+9060

-+9753

-+9060

-+9753 2540-15 15000#

22100#

Snub (1) 2 RCSR-255 6FY 0

0

-+9266

-+9735

-+9266

-+9735 2540-15 15000#

22100#

SNUB (1) 3 RCRR-271 14FX 146/+5

-+9346

-+%93

-9544

-9891 2200-10 10,000#

14600#

-9392

-9739 STRUT (2) 4 RCSR-270 15FZ 0

0

-+9247

-+9776

-+9247

--+9776 SNUB (2) 15000#

22100#

1. 2540-15 5

RCSR-269 16FY 0

0

-+2398

-+2967

-+2398

-+2967 SNUB (1) 15000#

22100#

2540-15 6

RCSR-268 1800FZ 0

0

+7707

+8641

+7707

+8641 SNUB (1) 15000#

22100#

2540-15 7

RCSR-266 2101FY 0

0

-+3731

-+5537

--+3731

-+5537 SNUB (1) 15000#

22100#

2540-15

-8 RCSR-265 21FR 0

0

-+10241

-+14589

-+10241

--+14589 WPA 3149#/ LUG 6298#/ LUG, X-Z (4 LUGS) 9 RCRR-280 31FX

+33 -399/0

-+12097

-+12538

-12466

-12907 2200-10 10000#

146000#

+12126

+12569 STRiff(2)

l 0231y x

5/8/87 Page 2 WATERFORD SES-3 NUREG-0737 ITEM II.D.1 SUBMITTAL TO QUESTI(N NO.6 SA. CALC.21163 RUN DATE 9-13-82 S.NO.

MARK NODE LOAD CASE NOS.

SUFFWcT CRITICAL PART 4 MAX. ALLOW. LOAD NO.

PT.

21 115/118 503 504 710/715 720/725 NORMAL 4 UPSET FAULTED -

4 DIR.

10 RCSR-279 32FZ 0

0

+8641

-+9227

+8641

-+9227 SNUB (2) 15000#

22100#

2540-15 RCSR-278 331Y 0

0

-+3445

-+3929

-+3445

-+3929 SNUB (1) 15000#

22100#

2540-15 12 RCSR-276 37FI O

O

+6031

+6327

-+6031

-+6327 SNUB (1)

-15000#

22100#

2540-15 13 RCSR-275 39FY 0

0

+6676

-+7500

-+6676

-+7500 SNUB (1) 15000#

22100#

2540-15 14 RCSR-274 41FR 0

0

+6249

+8409

+6249

+8409 WPA/4 LUGS 5816f/ LUG 11632#/ LUG X-Z 15 RCSH-256 555FY

-1725 VS2C#12 2240#

2240#

SPRING 16 RCSH-254 3001FY

-447 CSHf5 445#

445#

00NST(2)

SPRING 17 RCSH-267 18FY

-451 CSHf7 755#

755#

CONST SPRING u

g,

- i

.j

.,; i

.y

.,

• d

~

.y c J= 4 0231y 5/8/87 Page 3 WATERFORD SES-3 NUREG-0737 ITEM II.D.1 SUBMITTAL TO QUESTION NO.6 SA. CALC.31163 RUN DATE 9-13-82 S.NO, MARK NODE LOAD CASE NOS.

SUPPORT CRITICAL PART 4 MAX. ALLOW. LOAD NO.

Pr.

21 115/118 503 504 710/715 720/725 NORMAL 6 UPSET FAUL1ED ~

6 DIR.

t 18 RCSH-264 2102FY

-686 CSH#12 1295#

1295#

^

CONSr.

SPRING 19 RCSH-272 1300FY

-557 C5HIS

' 905#

905#

CONST.

SPRING 20 RCSH-277 34FY

-604 CSHf6(2) 600#

600#

SPRING CONST.

21 RCSH-273 4101FY

-1534 s.CSHf20 2990#.

2990#

SPRING

, CONST.

ENCLOSURE to W3P87-1063 Page 12 of 13 NOTES TO QUESTION 6:

1)

The Tabulated Load Case No.

21 Operating Weight, 115 Thermal Max. Positive Components.

118 Thermal Max, Negative Components.

503 Seismic OBE(T+A) SV Discharge Forces.

504 Seismic DBE (I+A) SV Discharge Forces.

710 S/R Normal & Upset Load Max. +

Ther.

715 S/R Normal & Upset Load Max. -

Ther.

720 S/R Faulted Load Max. + Ther.

725 S/R Faulted Load Max. - Ther.

2)

As-building of Pipe Supports based on Ebasco Specification of General Power Piping.

A.

Support Deflection Criteria Deflection of pipe supports on all Safety Class 1, 2, and 3 and Non-Safety Seismic I Systems plus Main Steam shall be limited to a maximum of 1/16 inch.

Deflection criteria shall not apply to non-safety non-seismic systems (except Main Steam). Rather, the allowable stress of the component material will be used to determine acceptability.

B.

Allowable Material Stress Limits Safety Class and/or Seismic I Supports Normal and Upset Conditions Use material stress allowables of ASME Section III, 1971 thru Winter 1972 or, if not acceptable, ASME Section III, Appendix XVII-2200 of the 1974 edition (which is the same as AISC manual - 7th Edition).

Equations to develop these allowables are:

Eb = 0.66 Fy Ft = 0.60 Fy Fv = 0.40 Fy Fp = 0.90 Fy Where Fy is the yield stress of the naterial at the appropriate design temperature.

Faulted Conditions When faulted loads are evaluated (i.e. D"E, Tornado) the faulted allowables given in the B-P LCD Shts for standard parts may be used.

For non-standard items like structural steel, tube steel, plates, welds, etc.,

the allowable stress is 0.9 x yield stress at temperature of the base material.

O ENCLOSURE to W3P87-1063 Page 13 of 13 QUESTION 7:

Only the upset condition was analyzed for ASME code acceptance; this included loads from operating pressure, deadweight, operating basis earthquake (OBE) and SV opening.

The faulted condition includes the safe shutdown earthquake in place of OBE.

It is recognized that the allowable stresses are higher for faulted conditions than for upset conditions; however, provide justification or analysis that shows the upset condition to be the limiting condition. Provide a table showing the maximum stresses compared to allowable stresses for all load combinations for the most highly stressed locations.

RESPONSE TO QUESTION 7:

The attached Table 4 presents maximum pipe stresses, code allowables, and stress ratios at the highly stressed locations for the upset and faulted load combinations.

4 A

i 1

a TABLE 4 MAXIMUM PIPE STRESS RESULTS ALL STRESSES IN PSI UPSET CONDITION FAULTED CONDITION NODE

,__ PRESS. + DW t OBE + SRV PRESS. + DW + SSE + SRV POINT ALLOWABLE STRESS ACTUAL ALLOWABLE STRESS ACTUAL STRESS STRESS RATIO STRESS STRESS RATIO 45 Branch 15402 23830

.646 16654 47660

.349 28 Elbow 23498 23830

.986 29169 47660

.612 27 Elbow 19933 23830

.836 22902 47660.

.480 48 Elbow 23035 23830

.966 26172 47660

.549 46 Elbow 19671 23830

.825 22253 47660

.467 O

i 1

-