ML20235D963
Text
i y
UNITED STATES q
h.f 9 E I E'
ff ATOMIC ENERGY COMMISSION
'l ' K*
V.
l' wAssinaros.o.c. mos4s ausi k.;.. -. ' J JUN 2 7 m J Ali S : 1975 A fd II
)
DOCKET NO.
STN 50-447 7 SMIId.12 hW I
3 t
i 1
l APPLICANT:
General Electric Company l
{E APPLICATION:
GESSAR i
.}
MEETING TO DISCUSS DRYWELL LEAK TEST AND STRUCTURAL PROOF TEST On June 11, 1974, we met with GE and many representatives from utilities and AE's to discuss our requirement for structurally proof testing the drywell at 1.15 times the design pressure, as well as leak testing the drywell at about the design pressum.
The meeting was heavily attended'by BWR-6 customers, as well as t
mpresentatives from seven A&E fims.
Presentations wem given by GE, Houston Light and Power, Ebasco, and PEPCO.
In spite of the argunients put forth by industry, we remained firm in our conviction that such tests should be made and we would reflect that position in our SER's. A list of attendees is enclosed as is a copy of the slides presented at the meeting.
Industry Presentations to Why Te'sts Are Not Needed Industry repmsentatives presen'ted their arguments as to why they felt neither test was required to prove the design.
The small break LOCA is the most critical with respect to pool bypass.
For a small break, the bypass area has to be small; otherwise, the fission product barrier (the containment) could be breached due to overpressurization.
It was felt that the low pressure test more realistically simulates the pressure loading from a small break.
It was also stated that a high pressum test won't result in any new leak paths being fortned.
To support this contention, a crack analysis was done to de.temine the extent of cracks due to shrinkage, thermal gradients, pmssure and seismic loads.
The various load combinations wem applied across the 60-inch wall section.
The msults showed that there was no through cracking for any of the load cases examined and the small break was the J
single most significant load the structure would see from a j
d cracking standpoint.
Investigations into the bypass capability for large and small breaks showed that the margin or safety factor as defined by industry is very large.
The safety factor is defined as the bypass capability i
that will not result in exceeding the containment design pressum compared to the estiraated leakage rate.
Even for 11 breaks F:zfmlQ6teTBrp fh'e].*. pressure the ratio is abc 1 y ke i y: i..
testismeaning) hh h dL:s C d
b 8707100362 870623
$MA i M0': '30m0V8 70m A,Me 0.T -n.c 40 PDR
)
, The next reason for not performing these tests wem the test It was stated that the tests would be difficulties and costs.
on the critical path and would result in deferred equipment installation and disrupt crews working in the drywell by banishing At least 4 to 6 them from the drywell during the test period.
weeks time would be lost in the schedule because of preparing for It was estimated and running the test and removing test equipment.
that this stretchout would cost five to eight million dollars Another half for such things as replacement ' power and interest.
A rather elaborate million would be mquired for test equipment.
scheme of flanges welded to the vents sealed with an "o" ring and covered by a hatch was proposed as the test fixture for vent plugging.
A comparison of the maximum steel stresses (meridional and hoop) at two sections of the wall showed that the test stress varied from Therefore, 4 to 73% of the design stmss at the 1.15 test pressure.
it was felt that the high pressure test would not result in appreciable steel stresses.
In conclusion, industry summarized their positions as follms:
The drywell is not a fission product boundary and should not (1) be tested as a containment.
The integrity of the drywell is assured by conservative (2) design and good QA.
The high pressure tests are costly and.have a significant (3) 1 impact on the schedule.
l A low pmssure test more accurately models reality since
]
(4) the bypass is more sensitive to the small break event.
f The staff made the following coments after the presentations.
Our This subject is not new with GE or the BWR-6 applicants.The drywell l
/iew is substantially different than industry's.
must function properly to assure that the containment maintains l
In addition to conservative design, l
its leaktight integrity.
verifying analyses, and good QA, we feel the test is necessary.
Too many times we have seen errors in design, errors in analyses, and QA bmakdowns to assure the drywell has been built correctly.
We feel the high pressum test is necessary to confirm that the j
j If there have been any drywell has been built as intended. design oversights or con f
(
4 i
i i
be indicated by the 1.15 test. We recognize that thermal and seismic loads cannot be simulated but feel that if the reinforcing steel stresses can be raised by the test to about 50-60% of yield and the structure responds as expected, then there is a reasonable basis for believing the building ic acceptable. No single aspect, good design; transient analysis; or a high pressure test will, I
by itself, give absolute _ assurance that the drywell will function pe rfectly. We feel that each part of the design and construction cycle should be done as well as possible, and should be tested and inspected to the extent practical to gain a high overall degree of assurance that the structum will indeed perfom as intended.
l The staff also noted that the lower portions of the drywell wall, l
which forms the suppression pool wall with the horizontal vents emplaced in it, is to be constructed in a fairly novel form with which there is very limited experience.
The large number of 2-foot diameter vent pipes passing through this section of the wall makes the normal reinforced concrete construction method impractical.
Instead, this portion of the pressure-mtaining wall is proposed as an unreinforced composite section, with external steel plates I
carrying part of the circumferential and longitudinal stresses.
The difficulty of analyzing such a composite section, and the lack of experimental failure criteria on which to base structural load, factors and the resulting safety margins, make a proof test under pressure even more important a consideration than would normally be the case for an important safety-related structure.
We mcognize them may be some schedule delays and extra costs associated with such tests, but we feel that these tests am a necessary part of assuring the adequacy of the drywell construction and, themfore essential.
We restated that our position on these tests was firm. We will require a structural proof test at 115% of design pressure and a leak test at about design pmssum for each drywell constructed.
D. M. Crutchfiel., Project Manager Light Water Rea tors Branch 2-1 l
Directorate of Licensing
(
H i
j ATTENDEES
~ JUNE 11, 1974 i
Meeting to Discuss Structural Proof Test and Leak Rate Test i
AEC Northwest Utilities J. Hendrie
. D. Lavoie R. Tedesco R. Maccary 7 Gulf States Utilities weil J. Wright R;
Southem Services J. Stolz T. Robin J. Orndoff W. Rowe D. Lynch A. Niyogi C. Grimes J. Kudrick EBASCO H. Os i k
- C. of ayer gj L._Slegers T. Rebelowski (R0:I)
Stone & Webster W. Paulson T. Su C. Cunningham e
R. Birkel W. Klehm W. Kane A. Toombs C. Anderson S. Burwell Puget Sound Power and Light C. Newkirk L.
e (OGC)
D. Crutchfield Black and Veatch General Electric R. Stippich E. VanZylstra i
A. Levine Public Service of Oklahoma W. McConaghy V. Conrad T. Brown D. Rockwell UE&C G. Wade L. Gifford A. Kalfopulos Pepco Gilbert Associates P. Dragoumis R. Brems J. Scoville TVA L. O'Callghan R. Barnett Houston Light and Power EDS Nuclear R. Klapper S ec J. Sumpter Il n s %er Sargent and Lundy i
S. Rurka H. Krug e_-___-____-