ML20203D538

Forwards EA Solla 860702 Memo Re 860609 Audit of Impell Cable Tray Reanalysis Effort

ML20203D538
Person / Time
Site:	Comanche Peak
Issue date:	07/02/1986
From:	Landers D TELEDYNE ENGINEERING SERVICES
To:	Jeng D NRC
References
6410-37, NUDOCS 8607210239
Download: ML20203D538 (3)

Text

a PTELEDYNE ENGINEERING SERVICES' 3: se:o , u.ru uom wssousensensa -9195 iE ' ', 893-335 ^ '.'. > J' Oi 324 '5:6 July 2, 1986 6410-67 . gG d rt /

t

. 6&r7TE-Mr. David Jeng United States Nuclear Regulatory Commission Phillips Building Mail Stop 316 7920 Norfolk Avenue Bethesda, MD, 20814

Subject:

Audit of Impell Cable Tray Re-Analysis Effort

Dear Mr. Jeng,

Attached is Mr. E. A. Solla's memorandum to me relative to the subject audit. We understand that Mr. Bezler of BNL will write the audit trip report. The attached submittal is essentially for internal TES use but also serves as a " deliverable" under our contractual commitment to the NRC.

Very truly yours, ,

TELEDYNE ENGINEERING SERVICES

^YN Donald F. Landers President DFL:mid attachment cc: V. Noonan (NRC)

C. Tramell (NRC) .,

A. Vietti-Cook (NRC)

G.Bagchi(NRC) 6410 File 0607210239 860702 i PDR ADOCK 0D000445 A PDR L.

gk v.w. a n: .,:,; ,.. ~

_..______._._.__.___w

i s

. SeTA m(NE MEMORANDUM ENGINEERING SERVICES i

T0: D. F. Landers FROM: E. A. Solla '

DATE: July 2, 1986

SUBJECT:

Audit of Impell for Comanche Peak During the week of June 9, I participated in an NRC audit of Impell at their Walnut Creek offices on their cable tray effort for Comanche Peak.

On Monday, June 9, we received a status report and discussed responses to NRC questions from the previous audit in January. On Tuesday, June 10, we had a presentation on cable tray system analysis methods and special studies that are being done. On Wednesday, June 11, we nad a presentation on their SUPERPIPE program and SUPERPOST program. It was decided that the best way to verify that the SUPERPIPE computer code is suitable for cable tray analysis and that the SUPERPOST computer code, which was written specifically for this project, addresses all concerns is to obtain several sample analyses and to reanalyze them using different computer codes. Five analyses were picked: two that EBASCO will do, one for Engineering Analysis Services, one for Brookhaven, and one for Teledyne.

On Thursday, . June 12, an exit meeting was held. Six open items remained which required further action from Impell.

1. In PI-02, Rev. 3, Appendix A, an incorrect stiffness formula was given. Impell must go back to make sure that this formula was not

'used in any analyses.

2. Impell's method is unclear for the modeling of the eccentricities of nonstandard clamp. In some cases their method may yield uncon-servative results. Impell is to provide examples of nonstandard clamp eccentricity modeling.

3. Justification of certain aspects of clamp behavior is required.

A. Given the assumption of positive connection, from friction-type clamps, in the longitudinal direction on a transverse support, Impell must show that this is conservtive for the loads at the longitudinal support.

B. Impell is to prcvide the rationale to model the friction connec-tion at clips with a finite stiffness.

1 "W n ~ =1WNE Memo to D. F. Lancers ENGINEERING SERVCES July 2, 1986 Page Two C. Tray clamps are modeled as a single connection at the center of the tray. This provides a more flexible connection due to the torsionsal stiffness of the tier, which will yield larger deflec-tions and different mode shapes. Impell is to show that this is conservative.

4. In ooing the code check for comoosite channels, the :hannels are separated and tne bending and warping stresses distributec Detween the two members. Impell is to show tnat this method is conservative.

5, In the modeling of a reducer, a rigid link is used to provide the offset. This link transfers all loads from one section to another.

Impell is to provide assurance that the real reducer will transfer these loads.

6. Impell is to show that when the longitudinal support is attached to the top flange that the shear load is shared equally by the two flanges.

Tne audit trip report will be written by Paul Bezler from Brookhaven National Laboratory and will be sent to us when it is complete. .

~

EAS:jej 4

I l

~