ML20080T016
| ML20080T016 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 09/01/1983 |
| From: | BROOKHAVEN NATIONAL LABORATORY |
| To: | |
| Shared Package | |
| ML20080S984 | List: |
| References | |
| FOIA-83-483 NUDOCS 8310180627 | |
| Download: ML20080T016 (5) | |
Text
_
O 5
SUMMARY
AND EVALUATION REPORT Report
Title:
Independent Design Verification Program, HVAC Duct and Supports P.eport Report No.
ITR-15, Revision '0 4~
?
Report Date:
5/20/83
~
Aut hor:
Robert L. Cloud Associates 0310180627 830901 PDR FOIA CALSOYA83-483.PDR t
.m
, INTRODUCTION The Interim Technical Report 15-(ITR 15) was prepared by Robert L. Cloud and Associates (RLCA) for the Diablo Canyon Nuclear Power Plant Independent De-sign Verification Program (DCNPP-IDVP).
The report summarizes the IDVP inde-p'odent seismic analysis and verification of a initial sample of design Class
'l n '!AC ducts and duct supports.
The results showed the samples to be adequate to withstand -the ef fects of the postulated Hosgri event.
Two concerns were noted however reflecting the facts that the span rules used to design HVAC ducts and supports were not documented and that, in certain cases, design Class 1 HVAC duct supports were not evaluateo for Hosgri loadings.
The Diablo Canyon Project has initiated corrective actions to address these two concerns.
These actions will be reviewed by the IDVP.
SUMMARY
OF REPORT RLCA performed independent analyses for two samples of HVAC duct and duct sup-
~
ports with each duct sample consisting of two sections of duct.
Sample No.1
?
was the 36 x 16 inch, 22 gauge rectangular duct loading from the supply fan S-69 to the 4.16 kV switchgear room.
The ducts run vertically from the floor
}.
slab at an elevation of 119', past the supply fan mounted on an elevation of 119', to a 90 elbow at an elevation of 135'.
The duct then passes thru a
^
concrete block wall and runs horizontally 24 feet to the 4.16 kV switch gear room wall. The duct is riveted to built in flanges at the wall and floor pen-etrations and supported by two intermediate supports in the vertical run and one three way support and three rets of vertical rod hangers in the horizon-tal run.
The horizontal run is covered with 7/8 inch thick pyrocrete for f' ire protection.
A short ' horizontal branch joins the vertical run to a flexible coupling at the supply fan.
HVAC sample No. 2 consists of both 12 and 10 inch diameter 24 gauge duct.
The 12-inch diameter duct starts at a wall penetration and extends horizontally and vertically over the primary make up water pumps at an elevation of 100' in the auxilia ry buildi ng.
It is welded to a built in flange at the wall pene-tration and restrained by one bilateral support.
?-
', c,
~ The 10 inch diameter duct is a straight horizontal run extending 18' between two concrete walls in the auxiliary building at an elevation of 100'.
It is welded to built in flanges at the wall penetrations and supported by three sets of 3/8 inch rod hangers.
With the exception of the vertical run of duct from sample No.1, RLCA de-veloped finite elenent models based on field verified dimensions for each duct-support configuration.
In these models, all wall or floor penetrations were considered fixed poirts and supports were modeled in detail.
From these models, the natural frequencies were determined.
For the vertical duct run, a closed form solution for a simple supported beam was used to estimate natural frequency.
Next, seismic accelerations were computed using these natural fre-
.quencies together with the appropriate Hosgri response spectra.
Forces and morents were then calculated for key areas of both the duct and duct supports.
Stresses were detennined from the forces and moments enveloped from the two load cases vertical and East-West and vertical and North-South.
These stresses were compared to allowable streses and the design analysis results.
[,
Standard engineering approaches.were used to calculate loads and stresses.
Since the DCNPP licensing criteria does not explicitly address HVAC ducts and supports, allowabl'es were obtained from the following sources.
General AISC code, 7th edition Anchor Bolts PG&E Spec. 8872 3/4" Anchor Bolts PG&E Drawing 054162, Rev. 3 Local Panel Bucklin3 Formulae for stress & strain, Roarke 5th edition The frequency estimates showed that all configurations exhibited fundamental frequencies in the rigid range.
A comparison of the computed stresses for key areas to allowable levels showed all stresses for each component to be below allowable and in most cases far below allowable.
The closest approach to allowable was a computed duct local panel buckling stress of 70 psi as com-pared to an allowable of 79 psi for the rectangular duct of sample No.1.
Comparisons to PG8E design analyses showed the PG8E stress estimates to exceed 3-
Ai si
~
the IDVP stress estimates owing to the use by PG&E of conservative design pro-
- cedures.
Tabulated comparisons of computed streses to allowable stresses and conputed stresses to PG8E design analysis results are presented.
Three error and open-item reports, 1003,1077 and 1110, were issued as a re-
~ sult of the IDVP studies.
E01's 1003 and 1077 noted the original Hosgri qual-ification documentation for certain supports in the IDVP samples could not be provided.
E0I 1110 noted a difference between the design drawings and as built conditions.
t-The IDVP noted the following two concerns.
1.
In certain cases, design Class 1 HVAC duct supports were not evaluated for Hosgri loadings.
2.
A design methodology for the evaluation of HVAC ducts and supports has not been documented.
r The IDVP made the following recanmendations to address these concerns.
1.
Establish and -implement a program to verify that all design Class 1 HVAC
~
ducts and supports are evaluated for Hosgri loadings.
^
2.
To docunent a design methodology to evaluate design Class 1 HVAC ducts and s uppo rt s.
The Diablo Canyon Project has initiated corrective actions to address these concerns.
These actions will be reviewed by the IDVP.
s i
I y
p
-,n-.
+,
-,,n.-
-,--ee--r-
.vn---
e-.
-mr--e--
n-n-----,
--n.p c
,r
Q.
EVALUAT'I0N In the review field verified infonnation, computer analysis methods and stand-Jard engineering approaches were used to perform a comprehensive independent analysis of a sample of Class 1 HVAC ducts and supports.
The review revealed
- that in general the components exhibited acceptable margins of safety even though a. documented methodology for their design did not exist.
The review seems reasonable and will be coupled with a further review.. The complete review will confinn the adequacy.of the Class 1 methods used to design HVAC ducts.and supports.
e 9
e 4
9 9
4 e
.