ML20086G307

Forwards HI-91700, Dynarack Valdidation Manual, in Response to NRC 910927 Request for Addl Info Re Rerack of Spent Fuel Pool.Rept Withheld

ML20086G307
Person / Time
Site:	Three Mile Island
Issue date:	11/27/1991
From:	Broughton T GENERAL PUBLIC UTILITIES CORP.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML19353B396	List: ML20086G307
References
C311-91-2139, NUDOCS 9112050034
Download: ML20086G307 (6)

Text

o e o, a

MNUhI [

GPU Nuclear Corporation Post Office Box 480 Route 441 South Middletown, Pennsylvania 17057 0191 717 944 7621 ,

TELEX 84 2366 Writer's Direct Dial Number:

(717) 948-8005 November 27, 1991 C311-91-2139 U. S. Nuclear Regulatory Commission Attention: Document Control Desk Washington, D.C. 20555

Subject:

Three Mile Island Nuclear station, Unit 1 (TMI-1)

Operating 1.icense No. DPR-50 Docket No. SE 289 Response to Request for Additional Information Regarding Spent fuel Pool Perack Enclosed is the GPU Nuclear response to request for additional information contained in the NRC's letter dated September 27, 1991 regarding rerack of the spent fuel pool.

If any additional information is required, please advise.

Sincerely, 1

h. $R T. G. Br ghton Vice Fresident and Director, THI-l DJD/amk

Enclosure:

1) TMI-l Response to NRC Request for Additional Information Regarding Spent fuel Pool Reracking

2) Validation Manual cc: TMI-l Senior Project Manager Region I Administrator TMI Senior Resident Inspector 9112050034 911127 FDR ADOCK OD00026_

p PDR

~

I GPU Nuclear Corporation is a subsidiary of Genera! Put'hc Utihties Corporation

C311-91-2139 Page 1 ENCLOSURE I THI-1 Response to NRC Request for Additional Information Reaardina Spent Fuel Pool Rerack I. Indicate whether the DYNARACK computer program used for nonlinear dynamic structural analysis in the submittal has met the acceptance criteria of NRC as written in Standard Review Plan (SRP), Section 3.8.1-II.4.e, and if affirmative, provide pertinent verification documents including applicable experimental data which demonstrate full compliance to the provisions of the SRP Section.

RESPONSE

The DYNARACK computer program was utilized in the dynamic analysis of the new high density spent fuel racks for THI-l A-Pool in the same manner as in over twenty other fuel rack projects in the past ten years. The validation of DYNARACK is in conformance with the provisions of SRP 3.8.1-II,4.e (page 3.8.1-10), which permits the following two validation options.

(i) The program is demonstrated to give substantially identical results when benchmarked against an independently written public domain program using a series of test problems, or (ii) The program is demonstrated to yield substantially identical results when compared to classical solutions E to analytical results in the

, published literature, g-to accepted experimental results.

DYNARACK has been validated using both of the two allowable validation procedures, as required by SRP 3.8.1. As described in the attached validation manual, the verification has been performed against both a public domain code developed by others and against classical solutions as well as analytical results in the literature. The benchmarking of DYNARACK's model against experimental data compileif by Northeast Utilities /Holtec International is also contained in the vr idation manual.

However, the experimental _ data itself is considered proprietary to its owners.

II. Provide a simplified bounding estimate of a spent fuel and rack dynamic response to SSE. . When a calculation is used, it should be based on a relatively straightforward methodology of general use in open literatures.

The bounding calculation is -such that assumptions made to simplify the analysis are sufficiently conservative so that it is acceptable without

too much effort in way of a justification. Any unconventional assumptions should be explained and justified with credible physical data.

Input function should consider both horizontal and vertical SSE as well as other loadings. Final analysis should address stress, deformation and gaps (minimum) and tip off possibility and compare them with previously established allowables.

1 9311-91-2139 vage 2 Some examples of suggested conservative evaluations are provided below.

1. Demonstrate that impacts between the racks and rack to wall are not likely by evaluating the following:

(a) Select a rack with a least section modulus (b) Assess the maximum inertia force of the rack from SSE based on an appropriate floor response spectra (c) Assure that the rack inertia force is less than the counteracting friction force between the rack and pool floor.

Otherwise, perform an upperbound deformation evaluation based on maximum impact loads and assure that majority of the fuel rods are protected and the basic geometry of the rack is being maintained.

In the upperbound deformation evaluation, assume the two adjacent racks are moving toward each other and one toward the pool wall with velocities of Vmax from floor response spectra (two cases). Add certain portion of the water to the rack to act as an added mass to the rack.

2. Assume the rack supports are prevented from sliding and take a tilted position for SSE input (one side of the rack is lifted while the other side is in contact with the floor). Evaluate maximum impact between the rack and pool floor when the lifted portion of the rack lands on the floor. Calculate corresponding stresses in the pool liner, pool concrete slab and in the rack, compare them with allowables to establish the design adequacy. Demonstrate that the rack will, in no case, turn over.

RESPONSE

The level of simplification resulting from assumptions such as use of floor response spectra to assess inertia loads, static comparison of inertia loads with available friction, etc. would modify the essence of the non-linear rack dynamics problem. The essential characteristics of the TMI-1 racks are the presence of fluid coupling forces, counteracting friction forces, pedestal lift-offs and impact forces therefrom, fuel assembly rattling forces and seismic inertia forces. Any linearized simplification of such an intrinsically non-linear structure will produce results which have no basis in fact to assess their meaning or validity.

The seismic analyses reported in Section 6 of the licensing submittal are considered bounding values. The assumotions used in the DYNARACK analysis are listed in Section 6.2.1. Sever- key points emphasizing the bounding nature of the solution are summari. ad below:

C311-91-2139 Page 3 (a) All fuel assemblies stored in a rack are assumed to move completely in-phase, thus magnifying the inertia loads due to rattling of the fuel assemblies.

(b) The retarding effect of form drag and viscosity of water are neglected.

(c) While it is a well-known fact that the fluid coupling force term rises rapidly as a rack approaches another rack (or wall), this effect is neglected. The entire rack seismic analysis is performed using linear fluid coupling theory which is based on coupling effects computed using initial gaps and not current gaps. This assumption is demonstrated in reference 6.2.2 to yield uniformly conservative results.

-The conservative character of the above assumptions ensures compliance with the requirement of considering the entire range of input functions to the rack (two horizontal and one vertical seismic input) plus dead load, fluid coupling load, etc. The Licensing Report analysis considers the rack tip-off possibility, and provides values of minimum gaps, as requested in the RAI.

The alternative approach of performing an upper bound deformation evaluation has indeed been performed assuming " opposed rack motion".

liowever, the opposed rack analysis has been performed with proper inclusion of the non-linearity of the system, not by using a response spectrum velocity.

In summary, the bounding calculations reported in our licensing submittal comply with all provisions of the general approach suggested in the RAI.

III. Although references 6.1.5, 6.2.1, 6.2.3, 6.2.4, 6.2.6 and 6.5.1 are quoted in the submittal, no information is given in the reference section.

Provide them.

RESPONSE

Several of the references were incorrectly labelled. Attached is a revised reference list.

1 6.14 REFERNMORB FO SECTION 6 (Revised October 14. 1991) l l 6.0.1 USNRC 8tandard Review Plan, NUREG-0800 (1981).

6.0.2 ABMI Boiler & Pressure vessel Code,Section III, sucesction NF (1983).

6.1.1 USNRC Regulatory Guide 1.29, useismic Design Classification,"

Rev. 3, 1978.

6.1.2 " Friction Coef ficients of Water Lubricated Stainless Steels for a Spent Fuel Raok Facility," Prof. Ernest Rabinowicz, MIT, a report for Boston Edison Company, 1976.

6.1.3 USNRC Regulatory Guide 1.92, " Combining Modal Responses and l Spatial components in seismic Response Analysis," Rev. 1, l February, 1976.

6.1.4 "The component Element Method in Dynamics with Application to Earthquake and Vehicle Engineering," 8. Levy and J.P.O.

l Wilkinson, McGraw Hill, 1976.

6.1.5 " Dynamics of Structures," R.W. Clough and J. Penzien, McGraw Hill (1975).

l 6.2.1 "The Component Element Method in Dynamics with Application to j Earthquake and Vehicle Engineering," 5. Levy and J.P.D. I Wilkinson, McGraw Hill, 1976.

6.2.2 Singh, K.P. and Soler, A.I., " Dynamic Coupling in a closely Spaced Two-Body System vibrating in Liquid Medium: The Case of Fuel Racks," 3rd International Conference on Nuclear Power Safety, Keswick, England, May 1982. l l

l 6.2.3 Singh, K.P. and Soler, A.I., " Dynamic Coupling in a Closely Spaced Two-Body System Vibrating in Liquid Medium: The Case j of Fuel Racks," 3rd Internationaa Conference on Nuclear Power safety, Keewick, England, May 1992.

6.2.4 R.J. Frits, "The Effects of Liquids on the Dynamic Motions of Insersed solids," Journal of Engineering for Industry, Trans. i of the ASME, February 1972, pp 167-172. l l

USNRC Regulatory Guide 1.61, " Damping Values for 5sismic 6.2.5 l Daeign of Nuclear Power Plants," 1973.

6.2.6 "The component Element Method in Dynamics with Application to 1 Earthquake and Vshicle Engineering," 8. Levy ,and J.P.D.

wilkinson, McGraw Hill, 1976.

6.5.1 ASME Boiler and Pressure Vessel Code,Section III, Subsection NF (1986).

1

9 e

w ENCLOSURE 11 l

DYNARACK VAllDATION MANUAL

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __