ML19290C026
| ML19290C026 | |
| Person / Time | |
|---|---|
| Site: | Trojan File:Portland General Electric icon.png |
| Issue date: | 12/21/1979 |
| From: | Herring K, Trammell C Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19290C023 | List: |
| References | |
| TAC-12369, NUDOCS 8001090049 | |
| Download: ML19290C026 (8) | |
Text
UNITED STATES OF AMERICA NUCLEAR REGULATOP,Y C0fC1ISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD
^
In the MattYr of
)
Docket No. 50-344 O
)
(Control Building)
PORTLAND GENERAL ELECTRIC COMPANY, ET AL. )
)
(Trojan Nuclear Plant)
)
JOINT AFFIDAVIT OF KENNETH S. HERRING AND CHARLES M. TRAMMELL, III ON THE STATUS OF THE TROJAN " WALL PROBLEM" STATE OF MARYLAND
)SS COUNTY OF MONTGOMERY)
I.
Charles M. Tramell, III, being duly sworn, depose and state:
1.
I an a Senior Project Manager, Division of Operating Reactors, U.S.
Nuclear Regulatory Commission, Washington, DC 20555.
2.
I have prepared the statement of Professional Qualifications attached hereto, and, if c?lled upon, would testify as set forth therein.
3.
I am the NRC Staff Project Manager for the Trojan Nuclear Plant.
In this capacity, I have been involved in the NRC Staff's review and evaluation of interim operation of the as-built Control Building since the issuance of the NRC's Order for Modification of License on this matter on May 26, 1978.
I.
Kenneth S. Herring, being duly sworn, depose and state:
4.
I am a senior structural engineer, Engineering Branch, Division of Operating Reactors, U.S. Nuclear Regulatory Comission, Washington DC 20555.
5.
I have prepared the statement of Professional Qualifications attached hereto, and if called upon, would testify as set forth therein.
6.
In ny capacity as senior structural engineer, I have had prinary responsibility for the NRC Staff's evaluation of the structural adequacy of the as-built Control, Auxiliary and Fuel Building Conplex for interin operation since the NRC was i.1 formed of the Control Building design deficiencies in April,1979.
I nave also been involved in the NRC's review and evaluation of the " wall problem" from its inception.
1729 350 80010900 d
^
We the affiants Kenneth S. Herring, and Charles M. Trammell, III, being duly sworn, depose and state:
BLOCK WALL REACTION FORCE PROBLEM Background of Wall Problem On October 19, 1979, the NRC Staff was informed by PGE of a problem with respect to seismic anchor SA-83 in the spent fuel cooling system.
In con-ducting the analyses, inspections and surveys required by IE Bulletins 79-02 and 79-14, PGE found that SA-83 did not meet the acceptance criteria of Bulletin 79-02, and, in addition, the concrete block wall to which SA-83 was attached was structurally inadequate to resist the piping reaction forces that would result from an earthquake.
A detailed chronology of events from October 12, 1979 to Decent er 13, 1979 is attached.
This information led PGE to an investigation as to how this occurred, and the extent of this type of problem throughout the plant.
PGE filed additional information on this subject in License Event Report 79-15 dated November 4,1979, and Supplements 1 and 2 dated November 19 and December 4, 1979. Additional information requested by the NRC staff, resulting from a two-day meeting on December 5 and 6,1979, was furnished in a letter from PGE on December 13, 1979.
Since the wall to which SA-83 was attached was a double wythe concrete block
- wall, PGE focused initial efforts on investigation of similar problems an one-and two-wythe concrete block walls and piping attachments to such walls.
In addition, the adequacy of piping supports attached to other types of support-ing structures, such as composite walls, concrete walls, floor slabs and structural steel have been investigated.
The criteria used in the reevaluation are presented in Supplement No.1 to LER 79-15.
Licensee's Procosed Corrective Action The corrective action resulting from the reevaluation has mostly consisted of removing supports from walls entirely, or bolting supports through the wall in cases where one wythe was by itself inadequate but both wythes together were adequate.
Potential Effects of Wall Problem and Status of Resolution It is essential that the structural adequacy of the walls be assured so that the assumptions made in the analyses and design of structures, equipment and piping i
1729 351 systems are not invalidated (i.e. their predicted behavior is not grossly change.d in the nonconservative direction, e.g., predicted loads increasing.--
substantially and/or strengths decreasing substantially). To determine that these analyses are not adversely affected by the " wall problem," the staff has requested additional information from PGE as set forth specifi-cally below.
Status of Resolution of the " Wall Problem" As a result of the NRC Staff review of the information submitted to date by PGE on this problem (delineated previously), and several conference calls with PGE and Bechtel staffs, additional information has been orally requested, needed by the Staff to satisfactorily resolve this new issue and determine the safety of Trojan for continued operation. This new issue must be satisfactorily resolved before the operation of Trojan is allowed to resume by the NRC per the I&E immediate action letter of October 22, 1979 as supplemented December 20,1979 (copy of December 20 letter attach-ed).
The additional information which is necessary for the NRC Staff to complete its review was transmitted to PGE via conference calls on December 19 and 20, 1979 and is in summary as follows:
- 1) As a result of the review of their response to NRC question 1 of December 6,1979 (contained in their December 13, 1979 submi ttal ),
we find that the information contained therein must be supplemented to include:
a)
In addition to the shear stresses reported, a definition of the tensile stresses at the interface and the basis for such.
b)
The bases for the given shear stresses due to interstory dis-placement must be given, including a description of and basis for their derivation.
c)
Verification that no walls experience in-plane tersion stresses.
If there is any, justify its acceptability.
d) Verification that the vertical shear stresses reported are an upper bound to the resultant horizontal and vertical shear stresses on the interface due to any plate behavior of the wall.
el Verification that the stresses reported are bounding including consideration of all dead and live loads, any thermal loadings on exterior walls, any pipe break loadings and tornado loadings, a
in addition to the seismic loadings.
f)
A description of the displacement profiles being used and the basis for such.
1729 352 g) For the walls reported in Question 1, PGE must provide, along with the in-plane shear stresses (forces) on the gross wall areas, O
the in-plane shear capacities for both flexure and shear controlled failures for these walls and the bases for these capacities. Additionally, PGE nust provide these capacities and the applied in-plane loads, and the interstory displacement induced stresses and information as requested in (b) above for the walls given in Table 1 of Supplement 2 to LER 79-15.
2)
As a result of the review of their response to NRC question 3 of December 6,1979 (contained in their December 13, 1979 submittal) and subsequent telephone discussions on this matter, PGE must docu-ment the results of and conclusions drawn from the tension testing which was performed on the mortared double-block masonry walls, including which walls were tested and the time period in which they were constructed (i.e. are they supposed to be fully mortared).
Al so,
discuss the survey of the drillers and their input as to how much mortar fill can be expected to be found between block wythes.
3)
The available documentation in the PGE response to NRC question 4 of December 6,1979 (contained in their December 13, 1979 submittal) does not adequately substantiate the vertical shear stress levels presented as acceptance criteria.
Besides requiring extrapolation of results from vertical shear stress levels from 9-14.5 psi (at which tension failures occurred in the bed joints) to 18 psi without a firm basis, these results were for wall specimens which did not have masonry ties
" equivalent" to those for the Trojan masonry.
Specifically, the re-ferenced wall specimen had masonry ties which have about three times the steel area per square foot of wall, which are of a truss type and, which is more uniformaly distributed than the masonry ties in the Trojan walls (#3 reinforcing bar ties at a spacing of 4 ft. center-to-center both horizontally and vertically).
Therefore, PGE must provide the direct shear test data discussed in the first portion of the response to substan-tiate the acceptance criteria, including a comparison of the masonry ties in the Trojan walls to any present in these test specimen.
4)
The criteria for substituting 1.5 times the allowable UBC working stresses for the ultimate strength in the Trojan factored load combinations is not in conformance with the historical NRC criteria for dealing with the work-ing stress design of concrete as given in the document referenced in Attach-ment 6 of Supplement 2 to LER 79-15 (" Seismic Design Criteria for Nuclear Power Plants," August 11, 1970).
As compared to this criteria, for example, the Trojan factored load combination for the SSE (1.0 load factor on the SSE) would lead to the application of a factor of 0.89 1729 353
^
(1.33/1.5) on the SSE in the working stress load equations which were accgpted.
Therefore, load combinations consistent with the working stress design method must be supplied by PGE and conformance with these revised criteria documented.
/
Kenneth S. Herring
(
'b /
//i?sNb Charles M. Tramell, III Subscribed and sworn to before me this 21st day of December,1979
^
J.
/,,
- ' rv I729 354 Notary Public
/
My Commission expires: July 1,1982 d
i
Trojan Wall Problem Chronoloay (Inadequacy of Walls to Withstand Support Reactions)
Oct.12,1979 -Trojan shut down to repair steam generator leakage.
2 Oct.17,1979 - Identified potential inadequacy of a thin block wall to withstand reactions from a seismic anchor on the RHR system.
Found while conducting investigations pursuant to IE Bulletins 79-02 and 79-14.
Also, there was a cotential for other supports to be affected.
Oct.19,1979 - Inadequacy of wall was confirmed and reported to the NRC, along with the identification of other supports.
Oct. 22, 1979 - PGE confirmed the existence of the problem in writing.
IE transmitted an immediate action letter to PGE requiring PGE to provide full information as to the nature and extent of the problem prior to the resump'. ion of operation.
Oct. 26, 1979 - Meeting held in Bethesda ameng the NRC, PGE and Bechtel to dis-cuss the wall problem.
The ?xtent of the problem was not fully known.
The cause of the prcolem seemed to stem from a lack of coordination between the structural designer for the wall and the pipe support designers, and inapppropriate design criteria for the walls to withstand the support reactions.
A plan was presented to assure core coeling in cold shutdown in the event of the loss of systems required for decay heat removal outside containment.
Investigations of thin block walls thus far re-vealed 24 of 48 supports requiring modification.
Systems includ-ed RHR, safety injection, charging and boron addition.
It was felt that problem, would be resolved, including completion of required mods., within about 2 wks. from this date.
Nov. 01, 1979 - Oregon DOE transmitted an information request on this problem, IE Bulletins and interim operation to PGE.
Nov. 04, 1979 - PGE supplied an LER on this subject.
Extent of the problem and the exact causes were not yet determined finally.
Appropriate wall evaluation criteria were being finalized.
Supplement to LER would follow and contain more complete details.
Subsequent telephone discussions (the week of Nov. 5) indicated this sucolement would be available on Nov. 13, 1979.
Nov. 12, 1979 - NRC Staff was notified by PGE via telephone that the supplement would be delayed by as much as 3 days.
In olant tests were beino conducted on thin block walls as well as those containing cores (shear walls) to determine their adequacy to withstand the required support reactions.
Date of test completion was not certain.
Nov.19,1979 - PGE transmitted to Oregon DOE the minutes of a Dec. 8,1979 meeting between them addressing the Nov.1 DOE i formation 1729 355 request.
Nov.19,1979 - PGE submitted Supplement 1 to LER 79-15 identifying evaluation criteria and any required modification.
Two attachments were to be provided at a later date.
A total of 87 required modifi-O cations were identified (29 through-bolt and 58 other to reduce wall loading).
One minor mortared double block masonry wall included in the STARDYNE model for interim operation was found to have six (6) attached supports, three (3) of which required modi fica tion.
Two (2) other supports modified for interim operation required modification, one (1) through bolt and the other modified to eliminate an interference on a newly modified support.
Nov. 21,1979 - The additional two attachments to LER 79-15 were submitted.
Nov. 30,1979 - The Licensing Board in the Trojan Control Building hearing proceeding issued an Order, ordering Trojan to remain shut down and requiring information to be supplied by the NRC Staff and Licensee regarding the status of the review for Phase II of the hearing and a schedule for completion, views as to how long in-terim operation should be allowed, and implicit concerns over the latest wall problem.
Dec. 04,1979 - PGE submitted Supplement 2 to LER 79-15.
Changes to evaluation criteria presented in Supolement 1 and results of in-plant wall testing to substantiate the evaluation criteria were supplied.
The total number of required modifications increased to 126 (43 through bolt and 83 other to reduce wall loading).
The results of the evaluation of 3 major mortared double block shear walls was presented.
These contained no significant supports.
Dec. 05, 1979 - A meeting was held in Bethesda, with the NRC, PGE, Bechtel, and PGE consultants.
As a result of this meeting, additional evalua-tions were required (given in 5 NRC questions to PGE) prior to any plant restart, and additional testing (coring) of the in-situ walls and a confirmatory test of a wall test specimen under combined in-plane and out-of-plane gross loadings with an attached support were required which would not have to be completed before start-up.
Dec. 07,1979 - The NRC Staff filed information required immediately by the 11/30/79 Licensing Board Order.
Include in this was an affidavit discussing the " wall problem."'
~
~ ~ ~ " - ' - ~
Dec. 08,1979 - PGE filed information required immediately by the 11/30/79 a '
Licensing Board Order.
In this, the 11/19/79 Supplement 1 to LER 79-15 was corrected in that the minor mortared double block wall included in the STARDYNE model for interin operation contained only 5 supoorts, 2 of which required modification (through bolt).
1729 356
^
Dec.12,1979 - The Trojan Control Building proceeding Licensing Board order
_ a hearing on Dec. 28, 29 to address the items listed in the Order of 11/30/79 and those seismic issues implicit in that Order.
Dec.13,1979 - PGE submitted responses to the 5 f1RC questions requiring response prior to start-up.
1729 357 6