ML20209C117
Text
- _ _ _ _ _ - - _ - - _
.y; AFF / s IE3
[f Cocket "os. 50.'75/323
'tE'80RAtjDUM FGR:
Geurne Lear, C5ief Structural and Geotechnical Enqineerin.1 Prm c:e Division of Engineerino TitRb:
Pao-Tsin Kuo. Section f.eader A Structural Engineering Section B Structural and Geotechnical Engineerina drarch FROM:
Norman D. Romney. Structural Enqineer
~
Structural Engineering Section D Structural and Geotechnical Engineer *r.q Dranch St% JECT:
TRIP REPORT - REVIEW 0F THE GRAND GULF RESOLUTIi 0F CABLE TRAY SUPPORT PROBLEtt On April 25, 1983 P. T. Xuo and Norman Romney of SGER visited the hecht.=1 Gaithersburg offices to review the Grand Gulf Cable Tray Support Calcular. ons.
The Grand Gulf project undertook an extensive testing of the "Unistrut '
material used in cable tray supports as a result of. spot weld failures &rico support fabrication. The Diablo Canyon project is using the results of :he Grand Gulf testing and analysis as the basis for their resolution of an allegation pertaining to "Superstrut" spot welds. The review of the Grand Gulf /Unistrue test program and cable tray calculations are Mrt of the staff investigation of the Diablo Canyon /Superstrut spot welu alle;ation.
The find'ocs from the staff review of the Grand Gulf /Unistrut test orrer an1 calculations are as follows:
1.
The Grand Gulf (GG) program did not include any raterial froo Su:ie~.* r:1.
The test prograc was germane and sensitive to the Unistrut wol.tino equip.ent and manufacturing processes. The Superstrut ven<for oses different welding equipment and manufactarino processes:
2.
The Unistrut spot weld failures at GG were identified wit'i sa cific ins of! material by the vendor wnicn were then isolated fror ne project.
To.date the infomation from Diablo Canyon is that the Su erst rut v : Mar cannot segregate material by lot:
3.
Initial testino of tini,trut material isolatad frr ".G t en:tret o. conu s er":
low and unacceptable spot weld values.
4 Calculations wre parforr'e1 to deter-ino ' un A i t,..
- w. :
u r.
~
i-which was found to be 2400 lbs/wel'.
e 30Mc@9
~ ~ '
- .. - (
- a 1 V, -. ;. l,. ; *..uHio
..o
.--. g.
Georo2 Le:r 5.
The test program established consisted of random samplic:. c' 13%
pieces of Unistrut material installed in GG. A single weld rrre each of the 100+ pieces was isolated and tested in shear.
7t.e eur values determined from tests on the 100+ welds were used to <.tottstically establish a weld value with a specific confidence level.
64 In the test program described above a few welds had test alues 1.,ss than 2400 lbs/ weld. For each weld that had a value of less than 2400 lbs/ weld, adjacent welds were isolated and tested.
These additional tests of adjacent welds indicated values less than 2400 it,s/ weld. These additional tests were not included in the statistical calculations to qualify the Ur.1 strut spot welds.
Norrun D. Romney. Structural Enqineer Structural Engineering Section P.
Structural and Geotechnical Engineering Branch Division of Engineering cc:
R. Vollmer J. Knight P. Kuo B. Buckley i
i p
(#1 N SGES:Df4dE3SGES:DV,.
v...r.
,1 wo gi NRomney/mc j PTKuo '
[ 4/28/83 i
4/
yb O f I i C : /- L l' ' ' O H = '~' ' ~ ' ' r
.u l
1