ML20040D719

Forwards Response to N Chokshi Request for Info Re Cable Tray & Cable Tray Hanger Design for OBE & Safety Relief Valve Load Combinations.W/Revised Writeup of Expected Max Stresses in Buried Pipes & Ducts from Seismic Shear Wave

ML20040D719
Person / Time
Site:	Clinton
Issue date:	01/29/1982
From:	Wuller G ILLINOIS POWER CO.
To:	John Miller Office of Nuclear Reactor Regulation
References
U-0412, U-412, NUDOCS 8202020184
Download: ML20040D719 (9)

Text

'

U-0412

_lLLINOIS POWER 00MPANY ~ -82(01-29)-6 500 SOUTH 27TH STREET, DECATUR, ILLINOIS 62525 -

January 29, 198T

= "N

.Mr. James R. Miller,. Chief 4* EECEJyED '

\

Standardization & Special Projects Branch Division of Licensing- d a FEB2  % 7 ti '

Office of Nuclear Reactor Regulation U.LS. Nuclear Regulatory Commission , s Washington, D.C. 20555 g . , (J

Dear Mr. Miller:

Clinton Power Station Unit 1 Docket No. 50-461 The enclosed report on the. design of-the Clinton Power Station cable trays and-cable tray hangers for the OBE and SRV load combination is submitted in response to a request by Mr. Nilesh Chokshi, SEB and as fulfillment of the information required for NRC's " Confirmatory Issue (6) "

Design adequacy of cable tray system (3.7.2)". If needed, we will be pleased to discuss this information with Mr. Chokski at'our February 3, 1982 meeting with the NRC in Bethesda, Md.

We have also enclosed a revised write-up of the maxi-mum stresses to be expected in buried pipes and ducts from-a' seismic shear wave. The original write-up was transmitted to the NRC by Illinois Power Company's letter No. U-0361 dated December 1, 1981. The two revisions, marked in the margin of the revised write-up are small and do not significantly effect the conclusion of the original write-up.

Sincerely, g/

-1

-[ / /'

G. E. Wuller Supervisor-Licensing Nuclear Station Engineering HBP /lt cc: J. H. Williams , NRC Clinton Project Manager, (w/o attach)

H. H. Livermore, NRC Resident Inspector, (w/o attach)

N. Chokshi, SEB (w/ attach)

^

8202020184 820129 PDR ADOCK 05000461 A PDR

o Evaluation of Clinton Cable Trays and Cable Tray Hanger For the OBE and SRV Load Combination I. Introduction The Clinton cable tray and cable tray hangers were designed to resist the SSE+SRV+LOCA loads with a minimum factor of safety of 1.05 against yield. Stresses for the OBE+SRV load combinations were not checked because operability of cables is not influenced by the stresses in the cable trays and tray hangers provided the maximum deflections are not excessive.

Limiting the stress to 0.95 f y for the SSE+SRV+LOCA loads achieves this goal.

Recently, the NRC Staff has requested information on the factors of safety against yield for the OBE and SRV loading for the Clinton cable trays and cable tray hangers. The requested information is presented in the following paragraph.

4

(

e e

II .-- Cable Trays I The cable trays were designed to resist the SSE+SRV+LOCA loads with a minimum factor of safety of 1.05. For each floor, the inertia forces corresponding to the peak spectral' acce-1eration of the vertical and the horizontal floor response spectra and the dead load were used to determine the maximum allowable unsupported span of-the tray. The cable tray sup-ports were then located so that.the actual spans are less than or equal to the maximum allowable value.

In order to show that the cable trays have a minimum f actor of safety of 1.6 against yield for the OBE+SRV loads, the ratio, r, of the SSE+SRV+LOCA to the OBE+SRV spectra were computed for each floor as follows:

GS G peak acceleration'(EaV+SRV+LOCA) peak acceleration (S RV+OBE)

If the. actual ratio for a floor is equal to or greater than 4

1.60, the trays have a minimum 1.60 factor of safety against

yield for the OBE+SRV loading and no further analysis is

} required. For some locations in the containment and auxiliary buildings, the calculated ratio was less than 1.60. However, it was also determined that the actual unsupported spans are considerably lower than the allowable span length. Lower span length results in a lower moment and a higher factor of safety.

To further examine the implications of these findings, the worst tray system from each building was chosen for a detailed dynamic analysis. In selecting these systems, the following factors were considered: a) the floor with the lowest ratio, r, was chosen in each building, and b) the tray system with the longest unsupported span was selected. The two systems thds selected were analyzed for the OBE+SRV loading and f actors 6i safety were calculated!

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

i The cable tray was modeled as a lumped mass beam system with the supports modeled by springs. The response spectrum method was used to compute the dynamic response using the floor -

response spectrum as the input.

The minimum values of the factors of safety against yield for the OBE+SRV loading condition for each system were computed and are given in the following table:

Minimum Factor Building Tray Width of Safety' Containment 24" 3.08 Auxiliary 18" 3.31 Based on the above evaluation it can be concluded that a minimumfactorofsafety'A~gainstyieldQfthre'gexistsin 5

the Clinton cable tray design for the OBE+SRV loading.

4 S

III. Cable Tray Hangers The Clinton cable tray hangers were designed to resist the SSE+SRV+LOCA loads with a minimum factor of safety of 1.05.

Each hanger is designed uniquely using frame analysis and using the response spectrum method. The widened floor response spectra forms the input for the dynamic analysis.

In order to show that the cable tray hangers have a minimum factor of safety of 1.60 against yield for the OBE+SRV loads, the spectral ratios between the SSE+SRV+LOCA spectrum and the OBE+SRV spectrum were computed of several typical hanger frequencies for horizontal and vertical excitation. If these ratios for a given floor were greater than or equal to 1.60, the hangers will have a minimum 1.60 factor of safety against yield for the OBE+SRV loading and no further analysis is required. However, for some locations and frequencies, the ratio was less than 1.60.

To further evaluate the implications of this finding, 109 hangers were chosen from the floors with the lowest spectra ratios in the Auxiliary, Fuel, Containment and Control Buildings. Each of these 109 hangers were analyzed for the OBE+SRV load combination to determine the actual factor of safety agai'nst yield.

This analysis showed that 103 of the 109 hangers analyzed had a factor of safety against yield in the range of 1.6 ,

to greater than 5. The average value was approximately 4.0.

In computing the average, all factors of safety greater than 5 were set equal to 5 so as not to bias the average. Six hangers had a factor of safety less than 1.6 as indicated below:

Auxilairy Building: 1.34 Fuel Building: l.34, 1.17, 1.17 Containment: . None Control Building: 1.50, 1.54

i

} Because the cable trays are continuous beams supported by multiple hanger supports, the average' factor of safety of 4.0 against yield..for the-OBE+SRV load combination is more accurate measure of the inherent conservatism in Clinton

~

cable tray hanger design.

J 9

i .

d i

I

't i

k i .

e

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

. - . . . - - . - n- -- -. + ,-. -.v. . - -, . -.

f IV. Conclusions The above evaluation.shows that both the cable trays and the tray hangers have an adequate factor of safety against yield for the OBE+SRV' loading condition. It should, how-ever, be noted that the actual margins of safety will be greater than.those shown by the above evaluation because of the conservatism in the evaluation parameters. The yield strength used in the evaluation is the minimum specified value. It is estimated that the actual material yield strength is 20% higher than the minimum specified value providing an additional factor of safety of 1.20. In addition the evaluation is based on the response spectrum method of analysis using a widened floor response spectra as-input. Past' analysis

, of piping and cable tray hanger system have shown that a time history method of analysis for these systems typically yields results 20-50% lower than those from the response spectra method. Similar margins in the Clinton cable tray hanger system are also expected.- Also a 4% damping per RG 1.61 was used for the cable tray and hanger analysis.

Published test results for cable tray support systems indicate that actual damping is considerably higher leading to pro-portionately lower response.

e e

]

, Dl 5 2.

L'cuusus

) ,

CLIth'ON BURIED PIPE A::D D" CTS The maximum strains in pipes and ducts were based on the assumptien that the specified ground particle velocity of 12 in/sec is due ,

to a single inclined shear wave. An apparent shear wave velocity of 2500 ft/sec was used. The single shear wave assumption is con-sistent with the first appronimation suggested in Yeh's paper Inen the relative contribu'tions of the varicus waves to the given maximum particle velocity are not known. The assumption of a single shear wave is also conservative for sites east of the Rockies, where the design earthquake is due to a near-field earthquake and mest of the energy, transmitted to the site is in the form of body Naves.

To evaluate the sensitivity of the buried pipe and duct design to the wave-type assumption, we have performed ,0 w follM im C~#J-wa-tions to show that any increase in stresses is more tk$an compensE s[

for by the margins built into the design. .

If we assume that at the instant the maximum particle velocity of 12 in/sec occurs, 50% is contributed by S-waves, 25% by P-waves, and 25% by Rayleigh waves, then the maximum particle velocities for the S , P, and Rayleigh waves are 9.8 in/sec, 4.9 in/sec, and 4.9 in/sec, respec tively .

Assuming the P-wave velocity to be twice the S-wave velocity and the Rayleigh wave velocity to be the same as the S-wave velccity leads to stresses 22% higher than when the particle velocity is R assumed to be from a single shear uave. ,

--. -- . )

_ _ _ - _ - _ _ _ _ _ _ _ _ _ _ _ _ -1

1HF ,

If we assume that at the instant the maximum particle velocity of 12 in/sec occurs, 3 3 't is contributed by S-wave, 331 by P-wave, and 331 by Rayleigh wave, then the computed maximum particle velocity for each of the three wave types is 6.9 in/sec, and the computed stresses are 4;,*? higher than when the particle velocity is assumed S-

, to be from a single shear wave.

In our design basis calculaticas, we have used an apparent she:r wave velocity of 2500 ft/sec. Clinton is a stiff soil site, and according to Hall & Newmark's paper, a 3000 ft/sec velocity is more appropriate.

This, when combined with a 17% increase in strength due to the actual material streng r.h being larger than the specified minimum, leads to a 40% margin in the Clinton buried ducts. The margins in the buried pipe design are also similar, where the design b sis calculation leads to a maximum stress of 19.4 ksi, compared to the allowable stress of 29.2 kni.

Based on the above discussion, it can be concluded that the Clinton buried duct and pipe design is conservative.

b P

9 e

. g om eum.= em....