ML18033A193
| ML18033A193 | |
| Person / Time | |
|---|---|
| Site: | Browns Ferry  |
| Issue date: | 05/04/1988 |
| From: | Gridley R TENNESSEE VALLEY AUTHORITY |
| To: | NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM) |
| Shared Package | |
| ML18033A195 | List: |
| References | |
| TAC-00299, TAC-299, NUDOCS 8805090158 | |
| Download: ML18033A193 (20) | |
Text
REGULATOR INFORMATION DISTRIBUTION
STEM (RIDS)
ACCESSION NBR: 8805090158 DOC. DATE: 88/05/04 NOTARIZED:
NO DOCKET FACIL: 50-2b0 Browns Ferr g Nuclear Power Stationi Unit 2>
Tennessee 050002bO AUTH. MANE AUTHOR AFFILIATION GRIDLEY> R.
Tennessee Va 1 leg Authoritg RECIP. NAME RECIPIENT AFFILIATION Document Control Branch (Document Control Desk)
SUBJECT:
Describes program for seismic qualification of HVAC ductwork supp ls info provided bg util 870408 5 880310 ltrs 8c 870701 Section IlI.3. 5 of Rev 1 to performance plan. Qualification of seismic Class I
HVAC ductwork 5 procedure encl.
DISTRIBUTION CODE:
00300 COPIES RECEIVED: LTR J ENCL j SIZE:
TITLE:
TVA Facilities Routine Correspondence NOTES: Q. Zech
- 3. cg.
1 cg.
ea to: Ebneteri*xelrad> S.
Richardson'.
D. Llaw'. Barry OI.
050002b0 RECIPIENT ID CODE/NANE JAMERSON> C MORAN> D INTERNAL: ACRS AEOD OGC 15-B-18 COPIES LTTR ENCL 1
1 1
1 1
1 1
0 RECIPIENT ID CODE/NANE PD GEARS' ADN/LFMB NUDOCS-ABSTRACT REG FILE 01 CO< IES LTTR ENCL 1
1 1
1 0
1 1
1 1
EXTERNAL:
LPDR NSIC 1
1 1
NRC PDR 1
1 NOTES:
TOTAL NUMBER OF COPIES REQUIRED:
LTTR 22 ENCL 20
0 r~ +
ACCEIZRATED DISTRIBUTION DEMONSTRATION SYSTEM REGULATORY INFORMATION DISTRIBUTION SYSTEM (RIDS)
D 05000259 05000260 05000296 A
NOTES:G.Zech 3 cy.
1 cy.
ea to: Ebneter,Axelrad,S.Richardson B.D.Liaw,K.Barr, OI.
G.Zech 3 cy.
1 cy.
ea to: Ebneter,Axelrad,S.Richardson, B.D.Liaw,K.Barr, OI.
G.Zech 3 cy.
1 cy.
ea to: Ebneter,Axelrad,S.Richardson, B.D.Liaw,K.Barr, OI.
ACCESSION NBR:8805090158 DOC.DATE: 88/05/04 NOTARIZED: NO DOCKET FACIL:50-259 Browns Ferry Nuclear Power Station, Unit 1, Tennessee 05000259 50-260 Browns Ferry Nuclear Power Station, Unit 2, Tennessee 05000260 50-296 Browns Ferry Nuclear Power Station, Unit 3, Tennessee
. 05000296 AUTH.NAME AUTHOR AFFILIATION GRIDLEY,R.
Tennessee Valley Authority RECIP.NAME RECIPIENT AFFILIATION Document Control Branch (Document Control Desk)
SUBJECT:
Describes program for seismic qualification of HVAC ductwork
& suppls 870408
& 880310 ltrs.
DISTRIBUTION CODE:
D030D COPIES RECEIVED:LTR ENCL SIZE:
TITLE: TVA Facilities Routine Correspondence RECIPIENT ID CODE/NAME JAMERSON,C MORAN,D INTERNAL: ACRS AEOD OGC 15-B-18 EXTERNAL: LPDR NSIC NOTES:
COPIES LTTR ENCL 1
1 1
1 1
1 1
1 1
0 1
1 1
1 9
9 RECIPIENT ID CODE/NAME PD GEARS,G ADM/LFMB NUDOCS-ABSTRACT
~G F NRC PDR COPIES LTTR ENCL 1
1 1
1 1
0 1
1 1
1
- 1 1
Sj A
TOTAL NUMBER OF COPIES REQUIRED:
LTTR 22 ENCL 20
TENNESSEE VALLEYAUTMORlTY CHATTANOOGA, TENNESSEE 37401 5N 157B Lookout Place MAY 04 1988 U.S. Nuclear, Regulatory Commission ATTN:
Document Control Desk Washington, D.C.
20555 Gentlemen:
In the Matter of Tennessee Valley Authority Docket Nos.
50-260 BROWNS FERRY NUCLEAR PLANT (BFN) SEISMIC QUALIFICATION OF HEATING VENTILATION AND AIR CONDITIONING (HVAC) DUCTWORK AND SUPPORTS-(NRC TAC NO. 00299)
This letter describes the BFN program for the seismic qualification of HVAC ductwork.
This. material was requested by R..J. Clark's letter dated July 31, 1986, to S..A.;White.
,=This letter-.supplements the information provided by TVA's letters dated April 8,
- 1987, March 10,
- 1988, and also supplements section III.3.5 of revision 1 to the BFN Performance Plan which was transmitted by S.
A. White's letter dated July 1,
- 1987, and incorporates resolutions to the NRC staff's concerns as discussed in our meeting held on March 18, 1988.
Enclosure 1 to this letter describes the BFN program for resolving this issue.
Enclosure 2 provides the basis for 1.5 factor design allowable bending stress used in the interim operability criteria.
Enclosure 3 is the BFN Class I HVAC duct and duct support seismic qualification interim operability acceptance criteria.
TVA requests your review of this program and the issuance of a written statement documenting the program's acceptability.
Please refer any questions regarding this submittal to M. J.
May, Manager, BFN Site Licensing, (205) 729-3570.
Very truly yours, TENNESSEE VA EY AUTHORITY Enclosures cc:
See page 2
R. Gridley, Di ector Nuclear Licensing and Regulatory Affairs Qo3 0 8805090158 880504 PDR ADOCK 05000259
. P DCD An Equal Opportunity Employer
4r
U.S. Nuclear Regulatory Commission MAY 04 i988 cc (Enclosures):
. Mr. K. P. Barr, Acting Assistant Director for Inspection Programs TVA Projects Division U.S.
Nuclear Regulatory Commission Region II 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Mr. G.
G.
Zech, Assistant Director for Projects TVA Projects Division U.S. Nuclear Regulatory Commission One White Flint, North 11555 Rockville Pike Rockville, Maryland 20852 Browns Ferry Resident Inspector
.Browns Ferry Nuclear:Pl.ant
,.Route '12, P.O.
Box 637
- Athens, Alabama
'35611
ENCLOSURE 1
TENNESSEE VALLEY AUTHORITY BROWNS FERRY NUCLEAR PLANT UNIT 2 QUALIFICATION OF SEISMIC CLASS I HVAC DUCTWORK AND SUPPORTS This report gives TVA's plan to qualify the as-configured seismic Class I HVAC ductwork installation.
Issue Design deficiencies were identified in TVA's Significant Condition Report No.
SCRBFNCEB&603 that was issued during February 1986.
Subsequent walkdowns of'he HVAC ductwork were performed by TVA, and djscrepanci.es were noted between as-constructed installations and the original design.
Backqround Initially, HVAC ducts and duct supports at BFN were fabricated to industry standards without consideration of seismic loads.
In 1970, the need for HVAC ducts to be designed for earthquake loads was identified.
As a result, modification of existing HVAC ducts and supports was initiated.
A HVAC seismic design criteria was issued in July 1970 and transmitted to the BFN Project Manager for implementation.
The duct construction was based on the Sheet Metal and Air-Conditioning National Association (SMACNA) standards, with both the pocket-lock and companion-flange types being used.
Several field evaluations (at least one per unit) were made by design engineers to review the as-built installations against the design criteria.
Recommendations were made as a result of the field evaluations and changes were made accordingly.
In January
- 1986, a significant condition report was written against the design criteria (BFN-50-721) used for installation and qualification of supports for, the HVAC system.
That report questioned whether the, design criteria was adequate to ensure the necessary seismic qualification of the HVAC system.
In addition, field investigations of the HVAC system led to concerns that significant discrepancies might exist between the as-built system and the requirement of the design criteria.
Resolution The scope of this activity involves 11,500 ft of ductwork, ranging in size from 6 in. to 30 in. diameter round ducts, and 5 in.
x 6 in. to 72 in.
x 84 in. rectangular ducts.
There are=935 deadweights, 463 two-way and 137 three-way supports for the duct systems.
(i ~ ~,,q
~g gy
~y y
y~ ~ ~ ~ ~
.g
'ey tQ Q +
tQ t~
g
'C
~
~~
+4 A
As-built sketches are generated as part of the walkdown effort and will document key attributes of the systems including the locations of supports and attachments, as well as their construction details and anchorage.
The key attributes for the ducts which will be documented are routing, size, construction,
- location, and types of attachments.
The technical design criteria has been revised to include the correct weight of the ducts, and the duct system's natural frequency calculation methods have been modified to reflect test results.
Additionally, the revised criteria now addresses cantilevered ducts and DBE (SSE)
- loads, and the allowable stresses are based on the AISC and the SMACNA standards.
The FSAR requires that essential HVAC systems remain functional for all plant conditions.
The design criteria used in the qualification are based on AISC and SMACNA allowables which assure that their HVAC systems remain functional.
The approach used for qualification of the duct systems is to evaluate 100 percent of the ductwork and supports against the design criteria stress levels for DBE (SSE) loads.
Those ducts or supports that do not meet the design criteria will be evaluated against the interim operability criteria.
Those which did not meet the design criteria, but are within the interim operability cri teria, will be modified to the design criteria after restart.
Those which do not meet the interim operability criteria will be modified to the design criteria unless specifically requested and approved by NRC on a case-by-case basis before restart.
A comparison of design and operability criteria is summarized in table 1.
Licensin Issue This program utilizes interim operability acceptance criteria for ducts and duct supports.
Justification The duct stress interim operability criteria are based on test data (see table 1
and enclosure 2),
The duct support interim operability criteria are the same as the pipe support interim operability criteria, which are similar to Sequoyah Nuclear Plant's (SQN's) large bore pipe support operability criteria.
Approval of the Sequoyah operability criteria is documented in NUREG-1252.
A comparison between SQN's and BFN's operability criteria is provided as table 2.
The HVAC qualification is comprehensive and provides assurance that the ducts and supports will remain functional.
Those modifications which are required to meet the design criteria will be completed prior to restart following the next refueling outage.
ENCLOSURE 1
TABLE 1
BROWNS FERRY UNIT 2 HVAC CRITERIA COMPARISON CHART ADDRESSES DESIGN CRITERIA OPERABILITY CRITERIA REMARKS ALLOHABLE DUCT 8,000 PSI STRESS RECTANGULAR PER SMACNA DUCTS 12,000 PSI BASED ON TEST DATA...
REPORTS TVA-CEB-79-7 AND MA 2-79-1 ROUND DUCTS 10,000 PSI PER SMACNA
. 15,000 PSI
. BASED ON TEST DATA REPORTS TVA-CEB-79-7 AND MA 2-79-1 ALLOHABLE SUPPORT PER AISC STRESS TENSION UP TO 0. 9Sy AND BENDING SMALLER OF 1.2Sy OR 0.7Su SAME AS PIPE SUPPORT OPERABILITY CRITERIA ALLOHABLE SUPPORT PER AISC STRESS COMPRES-UP TO 0.9Sy SION, AXIAL AND BENDING 0.9PcR SAME AS PIPE SUPPORT OP ERABILITY CRITERIA SHEAR PER AISC-UP TO 0.52Sy SMALLER OF 0.72Sy OR 0.42Su SAME AS PIPE SUPPORT OPERABILITY CRITERIA ALLOHABLE HELD STRESS SHEAR PER AISC UP TO.525y BASE METAL 0.42Su BASE METAL BASED ON ASME III SUBSECTION NF, APPENDIX F, FOR SUPPORTS ALLOHABLE BOLT STRESS (TENSION)
PER AISC UP TO 0,56Sy OF BOLT Sy HHEN NOT AVAILABLE 0.7Su SAME AS PIPE SUPPORT OPERABILITY CRITERIA
.ALLOHABLE CONCRETE HEDGE TYPE-4 EXPANSION ANCHORS SHELL TYPE FACTOR OF SAFETY 5
FOR TENSION HEDGE AND SHELL 4
FOR SHEAR TYPE ALL TYPES 2'AME AS. PIPE SUPPORT OPERABILITY CRITERIA
ADDRESSES ENCLOSURE 1
TABLE 2 BROWNS FERRY UNIT 2 HVAC CRITERIA COMPARISON CHART BROWNS FERRY UNIT 2 HVAC INTERIM OPERABILITY ACCEPTANCE CRITERIA SEQUOYAH UNIT 2 PIPE SUPPORTS INTERIM OPERABILITY CRITERIA ALLOWABLE SUPPORT SMALLER OF 1.2Sy STRESS TENSION OR 0.7Su AND BENDING SMALLER OF 1.2Sy OR 0.7Su ALLOWABLE SUPPORT 0.9Pcg STRESS COMPRESSION, AXIAL AND BENDING 0'9PcR SHEAR SMALLER OF 0.72Sy OR 0.42Sv SMALLER OF 0.72Sy OR 0.42Su ALLOWABLE WELD STRESS SHEAR 0.42Su 0.42Sv ALLOWABLE CONCRETE ALL TYPES EXPANSION ANCHORS 2
FACTOR OF SAFETY WEDGE AND SHELL TYPE ALL TYPES 2
ALLOWABLE BOLT Sy (WHEN NOT AVAILABLE STRESS (TENSION) 0 7Su)
(SEE NOTE 1)
Sy NOTE 1:
WHEN THE YIELD STRESS OF THE BOLT MATERIAL IS NOT SPECIFIED (e.g.,
ASTMA307), THE ALLOWABLE STRESS SHALL BE 70 PERCENT OF THE MINIMUM SPECIFIED ULTIMATE STRENGTH.
ENCLOSURE 2
BASIS FOR THE 1.5 FACTOR DESIGN ALLOWABLE BENDING STRESS USED FOR INTERIM OPERABILITY CRITERIA For long-term operation of the BFN safety-related.HVAC
- systems, ducts and supports will be analyzed and modified, as necessary, to maintain compliance with applicable code allowable stresses.
However, to permit interim operation before these modifications are
- made, interim operability criteria were established.
For this purpose, a factored allowable stress (equal to 1.5 X design allowable stress) was developed, consistent with an evaluation of duct capacity beyond that predicted by normal code levels.
To determine the value of the multiplying factor, the following steps were taken:
Review TVA's test data to establish seismic accelerations at failure.
Calculate the failure stress of test specimens using the failure seismic accelerations and the analytical procedure used to qualify the ductwork.
Select the lowest failure stress from all test specimens and reduce this failure stress further by subtracting off operating stresses such as pressure and seismic stresses in other directions not simulated in the tests.
Establish that the lowest failure bending stresses is at least two times the code allowable stress.
Select 1.5 as the multiplying factor for interim operability.
Details of the tests and the results of the calculations are described below.
The test program was conducted by TVA (Reference.
1), in which a range of 12 full sized, insulated rectangular ducts were tested to failure when subjected, to single axis, random multifrequency dynamic loading.
The ducts were fabricated to Sheet Metal and Air-Conditioning National Association (SMACNA) standards, identical with those used at BFN, having 20 or 22 gauge thickness and fabricated with either companion angle or pocket lock type joints.
Pocket lock ducts are connected by the use of crimped pockets at the end of each duct segment.
Companion angle ducts have angle iron sections attached to the ends of each duct segment which serve as flanges for bolting the ducts together.
Duct spans were equivalent to those specified in the TVA Design Guide for seismic qualification of HVAC ductwork, which was developed from SMACNA standards and used seismic response spectra peaks for load l
determination.
In order to simulate these conditions, the test fixtures used, variable stiffness supports which enabled tuning of the system frequency to the peak of the Required
Response
Spectrum (RRS), therefore producing the most severe loading condition.
This spectrum had a zero period acceleration of approximately 1.5g-and a peak. of 6.4g from 8 to 12 Hz.
Strain gages and
'ccelerometers were mounted on the shake table and duct specimen to measure input excitation and response.
System frequencies and damping values were
confirmed by sine'sweep testing.
After random multifrequency testing at the RRS level, input was incrementally applied until either duct failure or the shake table limit occurred.
The range of duct size and thickness tested are shown below:
TYPE CONSTRUCTION DUCT SIZE SKIN THICKNESS DUCT LENGTH SPAN LENGTH Companion Angle Companion Angle I
Companion Angle Companion Angle Companion Angle Companion Angle Pocket Lock Pocket Lock Pocket Lock Pocket Lock.
Pocket Lock Pocket Lock 60" x 24" 24" x 60" 48" x 18" 18" x 48" 36" x 24" 24" x 36" 60" x 24" 24" x 60" 48" x 18" 18" x 48" 36" x 24" 24" x 36" 20 ga.
(0.0359")
20 ga.
(0.0359")
22 ga.
(0.0299")
22 ga.
(0.0299")
22 ga.
(0.0299")
22 ga.
(0.0299")
20 ga.
(0.0359")
20 ga.
(0.0359")
22 ga.
22 ga.
(0.0299")
(0.0299")
22 ga.
(0.0299")
22 ga.
(0.0299")
31.3'3.5'7.5'9.5'3.5'3.5'1.3'3.5'7.5'9.5'3.5'9.5'8'6'6'4'2'6'8' 6
I 26'4'2'6'he onset and mode of failure depended upon the type of duct joint'.
Hith companion
- angles, at levels of excitation corresponding to the
- RRS, small tears and buckled regions ranging from small creases to approximately 1
square foot areas originated at duct corners usually near the mid-span with very little propagation until failure levels were experienced.
General failure was a gradual, ductile mode with no complete separation or breaching of the pressure boundary.
Pocket lock type ducts also exhibited very high capacities.
The pocket lock connections are more flexible than the companion angle type and allowed a
certain amount of relative displacements between duct sections.
This effect greatly increased the energy absorption of these ducts.
At RRS levels, no failures occurred, though residual sagging of the duct was present due to the loosening of the pocket lock cohnections.
Ultimate failure of the ducts occurred as either plastic hinge formation within the span or separation of the individual duct sections at the pocket lock connections.
The ultimate capacities, in the form of peak accelerations, for the test specimens are shown below:
FAILURE RUN CALCULATED
~K Companion Angle 35.
41 43 56 64 71 60" x 24" 36" x 24" 48" x 18" 60" x 24" 48" x 18" 36" x 24" Major Major Major Minor Minor Minor 10.2 10.5 11.2 14.0 12.3 13.6 45.0 28.2 38.5 51.7 34.8 29.2 FAILURE RUN CALCULATED Ut tt kt
~(AKA.()AU(( t(k Pocket Lock 83 95 105 112 120 131 36" x 24" 48" x 18" 60" x 24" 36" x 24" 48" x 18" 60" x 24" Majol Major Major Minor Minor Minor 12.0 16;2 14.5 12.0 11.5 16.0 28.5 45.2 48.5 25.2 27.2 45.7 Structural damping was also measured in the testing program.
The companion angle type ducts showed an average of 6.8 percent critical damping, due in part to the energy absorption associated with the gaskets and bolts used to connect the sections.
The pocket lock type ducts showed an average of 9.6 percent damping.
The higher damping can be attributed to the flexibility associated with the crimping used in securing the pockets which connect the sections.
The failure accelerations experienced in the test program greatly exceeded the required values for the appropriate duct suspension elevations at BFN.
The peak BFN amplified response spectrum is 3.6g at 7 percent
- damping, compared to the test 6.4g at this damping.
Furthermore, at the time when failure was actually induced, each specimen had experienced the equivalent of several safe shutdown earthquakes, since the fragility testing occupied several runs above the RRS level, which itself was almost 80 percent higher than the BFN maximum RRS.
It may be concluded that the testing was fully representative of the installed duct construction and test levels and durations were considerably in excess of required response levels'nalysis of the test results, using the acceleration levels sustained at failure and the SMACNA, reference 2, four-corner effective section
- method, indicated a bending stress at failure ranging from 25.2 to 51.7 ksi as shown in the last column of the above table.
However, the testing did not include operational or accident condition parameters such as pressure and seismic excitation in other axes.
These effects were calculated by conventional structural analysis and subtracted from the minimum failure stress of 25.2 ksi, giving a value for the maximum applied load bending capacity of 16.3 ksi.
Since the SMACNA allowable bending stress is 8.0 ksi (Reference
- 2) for normal design
- purposes, then this maximum applied load capacity is 2.04 times the design allowable stress.
A factor of 1.5 was selected for use as the operability criterion resulting in a factor of safety of 1.33 based on conservative methods and tests.
REFERENCES:
1.
Summary Report for HVAC Ducts Seismic Qualification, TVA Report No.
HA2-79-1, June 16, 1979.
2.
Rectangular Industrial Duct Construction Standards, Section 9,
- SMACNA, 1980.