ML20234B680
| ML20234B680 | |
| Person / Time | |
|---|---|
| Site: | 05000000, North Anna |
| Issue date: | 10/05/1976 |
| From: | Muller R Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| Shared Package | |
| ML20234A777 | List:
|
| References | |
| FOIA-87-40 NUDOCS 8707060160 | |
| Download: ML20234B680 (16) | |
Text
-
1 YRCU L USE 0%'"' \\ Tili
,,,,k UNITED STATES f
y Y, 'c.
' p, NUCLEAR REGULATORY COMMISSION a -l ' '^/-l ADVISORY COMMITTEE ON REACTOR SAFEGUARDS l
t WASHINGTON, D. C. 20555 l
October 5, 1976 1
ACRS MEMBERS l
CERTIFICATION OF THE MINUTES OF THE ACRS NORTH ANNA SUBCOMMITTEE MEETING - JULY 7, 1976 - WASHINGTON, D.C.
The minutes of the subject meeting, issued on September 7,1976, have been certified as the official record of the meeting.
- Please, destroy the working copy and retain the attachments.
O mundhitLfEC~..
Ra wald Muller Senior Staff Assistant i
ATTACHMENT:
cc: ACRS Technical Staff Participating Consultants i
I lin i A,RRI:lil
-,, a.
s,
t 0:F C A_ L S! 0 % _Y i
G707060160 870610 PDR FOIA THOMAS 87-40 PDR a
4} '.
k ( 9 IRCHAL USE LLY g
y 4
l 8
3
~'(
L Meeting Date:
7/7/76 Date Issued: 9/7/76 MINUTES OF THE ACRS NORTH ANNA SUBCOMMITTEE MEETING Room 1046, 1717 H St., N.W., Washington, D. C.
July 7, 1976 The ACRS North Anna Subcommittee met in Washington, D. C., on July 7,1976, to commence its review of the application of the Virginia Electric & Power Company (VEPCO) for a license to operate the North Anna Nuclear Power Plant, l
Units 1 & 2.
Notice of the meeting appeared in the Federal Register, Vol. 41, No. 120, Monday, June 21, 1976 (Attachment A), which contains the proposed agenda for the meeting. A copy of the Tentative Detailed Schedule is included as Attachment B.
A list of attendees is included as Attachment C.
Slides used during the presentations are included as Attachment D.
No requests for oral statements and no written statements from members of the public were received. A list of references available for consideration at this meeting is included as Attachment E.
Executive Session (CLOSED)
The meeting was called to order at 8:30 a.m.
The Chairman asked the consultants about their areas of concern.
Dr. Lipinski expressed an interest in the construction on Unit 2, while Unit I was operating.
He noted that during his visit in January no barriers had been installed.
Mr. Merkle observed that Sunship consultant, Harrison, and VEPCO consultant, Corten, had a similar approach but got differing results. He indicated that the magnitude of residual stresses as affected by peening were uncertain.
Dr. Canonico discussed the properties of the weld metal and the heat effected zone (HAZ).
He observed that if the preheat was as stated the HAZ and weld metal should be superior to the base metal. A572 and A36 don't have much hardenability; probably end with perlite or fine grain ferrite. Keeping down manganese and silicon affects lamellar tearing.
Dr. Okrent was interested in discussing reliability analysis, as well as several AEC inspection reports.
The meeting ended at 9:05.
Open Session The Subcommittee Chairman, Dr. David Okrent, called the meeting to order, and made an opening statement noting the presence of Mr. Ragnwald Muller, the Designated Federal Government Employee in attendance, and noting that a meeting notice.had been published in the Federal Register on Monday, June 21, 1976.
No requests had been received for oral statements at the meeting.
OFRCHAL USE ONLY
4 11FHCHAL USE LLY North Anna Meeting Date:
7/7/76 Dr. Okrent stated that the afternoon would be devoted to the subject of steam generator and reactor coolant pump supports and that it may be necessary to drop some of the topics proposed for the morning discussion. These, he indicated, could be taken up at a later meeting of the Subcommittee.
Staff Re* port Mr. Alexander Dromerick, NRC Staff Licensing Projec t Manager for North Anna, Units 1 & 2, then gave a status report on the project listing in order the 17 outstanding issues in Section 22 of Supplement 1 to the North Anna Safety Evaluation Report.
Mr. Dromerick reported that af the 17 items, 2 had been resolved (1 and 9),
four had not yet received complete review by the Staff (2, 11, 13 & 16),
one was a Westinghouse generic problem (5) and the remainder had not yet been adequately responded to by the Applicant.
Dr. Okrent requested that for the next meeting the Staff discuss any residual differences that might remain within the Staff, on items of safety significance, with their respective pros and cons.
Dr. Okrent next asked the Applicant to cover systems interaction (Agenda item 1.3) including fire protection (Agenda item 3.2).
Applicant Presentation Mr. Samuel C. Brown, Jr., V. P., power station engineering and construction for VEPCO, introduced the members of his staff, and consultants. He indicated Unit 1 is 91% complete with fuel loading scheduled for late 1976 and com-mercial operation in the Spring of 1977. Unit 2 is 70% complete and com-mercial operation follows Unit I by seven months.
Mr. Walter F. Bennett then commented on the Staff report of outstanding issues.
Dr. Okrent next inquired into the nature of the audit and review to assure that no possible systems interactions remain which have potentially un-desirable effects on the public health and safety. At the Applicant's request he cited several examples of such possible interactions.
Mr. Bennett called on Mr. Robert Bracbury of Stone & Webster to respond to the question.
Mr. Bradbury stated that the requirements of Appendix B to 10 CFR 50 were met.
Interdisciplinary review of all systems was conducted to insure compatibility. Changes were similarly reviewed. Vendor supplied equipment is examined to insure compatibility. Safety related systems are separa,ted from potentially catastrophic events to the extent possible and where no4 physical barriers are provided to the extent possibic. Components are qualified to function in accident environments. Non flammable materials are used wherever practical. The fire protection system is designed so as not to interfere with any safety related system.
OFFHCHAL USE ONLY
IFHCHAL USE C LY North Anna Meeting Date:
7/7/76 Dr. Okrent asked about a systematic review of the "as-built" plant, especially in the light of the additional attention given to systems interaction since the plant was designed.
He asked about the Staff review.
Mr. Novak described the interdisciplinary reviews within the NRC Staff, indicating that these are reflected in the Standard Review Plans, and verification is provided by preoperational testing.
Mr. Mcdonald of the NRC Staff stated that the Staff had made a 3 day site visit to verify physical separation of safety and non-safety systems.
John Davis described the North Anna fire protection system. The system meets the (NFPA) National Fire Protection Association Standards and is based on Nuclear Energy Liability Property Insurance Association (NELPIA) recommendations. He used slides 1-6 (attached).
In response to Dr. Okrent's question, Mr. Davis indicated that the seismic class 1 category ended at the building walls.
Mr. Davis described the deluge system, the sprinkler systms, the 2 low pressure CO2 systems, the high pressure CO2 systems, and the Halon systems.
Dr. Okrent asked that, at the next Subcommittee meeting, the Subcommittee be briefed on the principal differences between current, post Browns' Ferry, NELPIA recommendations and what is actually installed at North Anna.
Mr. Philip Matthews, NRC Staff indicated that North Anna was judged I
satisfactory based on criteria existing prior to the Browns Ferry fire.
The Staff is currently preparing an appendix to the branch position to provide guidance on required improvements for plants already licensed.
Next Dr. Lipinski asked about barriers between cables to afford fire protection between different divisions of the safety systems.
Mr. Davis indicated these had been discussed with the Staff. An Amendment to the SAR will describe the proposed barrier sizing and spacing. At present there are no detectors within the containment other than those under the main coolant pumps for fire alarm. This is being reviewed in view of the Browns Ferry information and Regulatory Guide 1.120.
In response to a question frem Dr. Lipinski Mr. Davis indicated that VEPCO had selected a material to seal roughly 6,000 closures, and had also developed procedures not only for installing initial plugs, but also for making repairs.
l l
Mr. Davis also explained that the control consoles could be opened without l
special tools so that Halon could be used on them in case of fire.
l i
OFFECHAL USE ONLY o
1 l
1/MCHAL USE E eLY l
North Anna Meeting Date:
7/7/76 Dr. Okrent asked about water back-up for inert gas in fighting fires.
Mr. Davis pointed out the problem of drainage in certain areas.
Dr. Okrent asked for a discussion at the next meeting, of awkward situations that may exist concerning fires.
i The subject then shifted to agenda item 1.5, the significance to l
North Anna of recent geological observations in Virginia.
Mr. Devine of USGS discussed the geologically recent faulting in the Piedmont.
He stated that the previous theory that all faults in this area had to be ancient was no longer true.
USGS had not completed its review of the microseismic data submitted by the Applicant.
In response to a question by Dr. Wilson, Mr. Devine estimated the nearest post Holocene fault to North Anna was roughly 25-30 miles l
away. Asked if they were " typical" Piedmont faults, Mr. Devine replied that there are no typical faults any more. There has been no seismic activity near these faults.
l Dr. Okrent asked why this recent faulting might be significant at Douglas Point and not at North Anna.
Mr. Devine thought probably because Douglas Point is closer to the fault.
In response to a question from Dr. Okrent, Mr. Devine stated that it would take a larger earthquake on the Stafford fault than one cc.uld reasonably postulate to exceed the.12g SSE for North Anna. He I
i estimated the fault, not a, single break, could extend for between 10 and 40 miles.
Dr. Okrent referred to Diablo Canyon and asked if the fault interlinkage at North Anna was trected as conservatively as that at Diablo Canyon.
Mr. Devine felt they were treated similarly. He pointed out the large difference between faulting in California and faulting in the Piedmont.
The California faults generate earthquakes, whereas the Virginia-Georgia ones do not.
(e.g. Hosgri vs. Stafford and Eclair)
Dr. Okrent asked if USGS had an estimate of probabilities of a SSE for North Anne.
Mr. Devine indicated it would be low and be did not have confidence in techniques that offer probabilities at such low numbers.
He specifically expressed concern with the method used by Robin McGuire with which one might arrive at a recurrence interval of 1,000 years for accelerations of.12g.
OFRCHAL USE ONLY
IRCHAL USE (_ eLY North Anna Meeting Date:
7/7/76 i
l In response to Dr. Wilson's question, Mr. Devine indicated USGS is continuing work on the Stafford fault, but he was not sure about Belair.
Dr. Wilson indicated an interest in finding out if these faults represent creep or slump rather than typical earthquake f a ul t ing,.
He predicted finding additional Eelairs end Staffords and expressed a curiosity about how they fit into the tectonics.
Mr. Devine indicated that as a seismologist he found these faults rather uninteresting. He stated that the Palmdale uplift had greater priority in his office.
Mr. Seth Coplan, NRC Staff, recalled at this point that the microseismic network at North Anna was set up to detect microearthquakes going through the reactor foundations and whether Lake Anna had reactivated these faults.
No evidence of such activity on these faults has been. observed. The Staff is still awaiting the USGS review.
Dr. Okrent asked if there were a maximum MM8 quake on the Stafford fault what intensity would be experienced at the North Anna site.
Mr. Coplan estimated somewhere between MM6 and 7, a little less than the design basis acceleration.
Dr. Devine pointed out that the Stafford fault is not one easily identified break, but a series of short breaks that may hook up or may not.
It is closer to Douglas Point than to North Anna. USGS is continuing its study of the Stafford fault.
This study is not related to reactor siting.
l I
i l
l l
l OFMCHAL USE ONLY l
f
_ IRCHAL USE LLY North Anna Meeting Date:
7/7/76 Af ter a short break the Subcommittee reconvened in CLOSED session to discuss Reactor Pressure Vessel forces during blowdown involving proprietary Westinghouse data.
Mr. Bennett of VEPCO opened the discussion using Table 5A.4-18 from Amendment 43 (Attached) snowing the phases of the analysis.
He called on Mr. Vincent Esposito of Westinghouse who, with the aid of the attached slides, covered the pressure wave at a joint and multi-dimensional pressure wave propagation. He compared the calculated results with tests in the containment system experiment, the Bettis flexible wall experiment, and the Fritz-Kiss experiment, and lastly, covered sensitivity analysis. He showed the wave fromt and reflection of MULTIFLEX vs. theory at 2.3 msec., 4.5 msec., 7.2 msec., and 10.2 msec.
In response to a question by Dr. Catton, Mr. Esposito explained that the frictional factors used were based on Steady State conditions.
Mr. Tom Esselman of !1 westinghouse discussed the structural model, using the core barrel as a beam, to solve the structural problem. The down-comer region was nodalized.
Dr. Theofanous inquired if a sensitivity study hed been done on this. He asked if the results from Keenan and Keyes steam tables had been used in the equation of state. He asked how the one-dimensional solution had been applied to a beam.
Westinghouse stated that the design load was about twice that determined experimentally.
The Subcommittee decided that time would not permit completion of this discussion at this time.
(Subsequently, since this was a generic problem, an Ad Hoc Working Group on Reactor Pressure Vessel Loadings was j
established, and a meeting held on August 5, 1976.)
l l
l OFMCHAL USE ONLY
.IFHCHAL USE C eLY North Anna Meeting Date:
7/7/76 The Subcommittee resumed its meeting in OPEN Session, with a request from the Chairman that at the next meeting the NRC Staff and the Applicant discuss the margin, if any, available in the seismic design.
Next, Mr. John Ahladas, North Anna station manager, discussed the safety measures in operation of Unit I while Unit 2 is still under construction.
Mr.Ahichaspointedout that shared components, such as the control room and its fire protection system, will be complete.
Uncleared workers will be escorted.
Mr. David Speide11 next covered the ability of the plant to withstand lose of A.C. power. North Anna is fed through three redundant station service transformers, and should they be lost, there are four redundant on-site diesel generators.
Mr. Speide11's statement is attached.
In response to a question from Dr. Lipinski, Mr. Speide11 affirmed that the control rods automatically are gravity-lowered into the reactor on loss of A.C. power.
For the auxiliary feedwater system, any needed controls are battery operated.
The battery can function a minimum of two hours. There is no specific procedure to tell an operator what to do if he loses all diesels and all A.C. power.
Mr. Dromerick stated that the Staff had confirmed that there was adequate D.C. power, and that the necessary valves would work. The meeting was then adjourned for lunch.
Dr. Okrent passed the Chairmanship of the meeting to Mr. Harold Etherington for the afternoon session.
Mr. Bennet called on Mr. Sam C. Brown, Jr., Vice President for Power Station Engineering and Construction, to lead off the VEPCO presentation.
Mr. Brown's statement is attached.
Mr. Brown indicated the operating temperatures of the supports would have to be increased over what was thought necessary as recently as a week before.
Next, A] Van Sickle of Stone 6 Webster engineering compared design and fabrication criteria of the North Anna 162 supports to current ASME criteria, using the attached slides.
In response to a question from Mr. Merkle, Mr. Van Sickle stated that l
viscous damping of snubbers is not considered in the dynamic analysis of l
the structure, but is conservatively neglected.
1 I
OFFECHAL USE ONLY l
L1FHCHAL USE C..LY North Anna Meeting Date:
7/7/76 Dr. Okrent asked if, by S&W referring to ASME code requirements, one could '
infer that S&W considered them adequate.
Mr. Van Sickle indicated that sdequacy in terms of brittle fracture was a separate subject.
He felt it was adequate, but not as conservative as some industry uses.
Dr. Okrent asked if he considered it adequate at any temperature.
Mr. Van Sickle indicated that there were no known instances of brittle fracture at room temperature or above.
Next, Norman Goldstein of S&W reviewed the stress and dynamic analysis of the steam generator and reactor coolant pump supports, with the aid of slides (Attached) illustrating the supports.
He mentioned various conservatism in the calculations (see-slide) and presented a table comparing maximum beam stresses to design criteria.
He showed the loading of the beams and indicated,except for the redundancy model (which were within stress limits), none was stressed more than 24 Ks1 taking into account dead weight, design basis earthquake, and LOCA forces.
Mr. Goldstein could not give the uncertainty in his calculations, as for example, the effect if one structural member were unintentionally shorter than it should be.
Mr. Goldstein next showed a table of through transverse stresses at member connection welds, followed by a table evaluating high cycle fatigue.
The maximum cyclic stress on the steam generator support frame was + 1.28 Ksi which was well within limits.
Mr. Goldstein showed a curve of deflection time for typical strain rate time points, and reviewed the equation for vs.
determinin6 disp acement (see slide). Finally, Mr. Goldstein presented a l
summary of SC/RCP support member strain rate due to a LOCA. He indicated the maxima were.14 inches per inch per second for steam generator and
.24 inches per inch per second for the reactor coolant pump.
Next, Mr. James L. Perkins, Director of Quality Assurance for VEPCO, addressed the QA program for the welds.
Mr. Perkins' statement and slides are attached.
Dr. Okrent inquired why post weld heat treatment was listed as " Unit 2 Only."
Mr. Perkins explained that Unit 1 was mechanically stress-relieved by peening.
Mr. Perkins explained that in weld repair there was no spring-back and fit-up problem because the welds were redone about one-half at a time. Asked about I
problems with the original fit, Mr. Perkins indicated that SunShip might be better able to answer, but added that to his knowledge there were no problems in meeting the S&W drawing tolerance requirements.
I OFFECHAL USE ONLY i
LIFHCHAL USE L oY North Anna Meeting Date:
7/7/76 Mr. Etherington indicated he would like to hear more about peening as s substitute for stress relieving.
Mr. Baum pointed out that e.'ery weld originally made by SunShip had been replaced. The heat treatment referred to was on the repairs to the repipcement welds.
Mr. Perkins stated that SunShip was permitted to perform a magnetic particle inspection of one of the supports.
He said that Sunship recorded every powder indication as an indication without any, determination of relevancy.
Later, VEPCO's consultants (Southwest Research Institute) with a level-3 interpreter (the highest legel of NDT examiner) found only four relevant indications out of 119 welds previously inspected by SunShip.
Mr. Perkins stated that the codes do not give a standard whereby relevancy can be measured.
Mr. Etherington asked about indications that showed up only after stress relieving.
Mr. Perkins indicated that VEPCO found during heat treatment some sub-surface slag pockets would cause a crack to open from there.
Dr. Canonico inquired if all repair welds were post-weld heat treated.
Mr. Perkins indicated they were not.
There was one final post-weld heat treatment and for repairs after that, those over 1/4 inch were mechanically peened (except af ter the las t pass).
Mr. Perkins stated that about 18% of the welds had indications of some kind.
Dr. Okrent asked about the efficiency of finding flaws.
Mr. Perkins estimated 95%, qualifying that by adding that all over 3/16 would be discovered.
Dr. Okrent then asked Mr. Goldstein about the concept of redundancy in analyzing the support structures.
Mr. Goldstein reviewed which members had been assumed removed analytically.
The method showed how the frame redistributed stress. Although the load condition chosen was the controlling load case (the corner chosen for removal was the most highly stressed corner of the frame), if you start dieabling too many members the stiffness is compromised.
Mr. Merkle asked about lamellar tears.
Mr. Perkins stated there were none in the final inspection of the completed structures. VEPCO did not record lamellar tears as such in making their magnetic examination for the repair process.
Lamellar tears were identified in SunShip's core sampling, and they identified one or two which VEPC0 was, by that time, in the process of repairing.
In response to a question from Dr. Canonico, Mr. Perkins indicated that lamellar tears might not come to the surface and for that reason ultrasonic examination was done on those welds most susceptible to lamellar tearing.
OFFECHAL USE ONLY
l FFECHAL USE C oY i
North Anna Meeting Date:
7/7/76 l
l Next, Mr. Bennett called on Mr. Joseph McAvoy, VEPCO, to discuss chemical and mechanical test data.
Mr. McAvoy used his slides to indicate how VEPCO had evaluated the materials of the structures.
In showing his slide on Charpy V-Notch test data vs. Temperature, Mr. McAvoy stressed the fact that the materials were bought and met
]
the code in 1970-71, and yet, except for two data points at 80 F, i
the data; met the current ASME Section 3, Boiler Code requirements.
Chemically, between tests by VEPCO and by SunShip, a goodly portion of the beams in the structure were tested.
In showing Table 1, Mr. McAvoy stated it had previously been submitted to NRC on June lith, and that it was intended to convey the drop weight test values.
Figure 4 showed industry data provided to the Special ASME Task Group on Fracture toughness.
Mr. McAvoy pointed out that the Chemical Analysis of A-36 beams (see slide) showed that the Mn to C ratio was about 4.1 which is an indicator of reasonable toughness. Of those listed, only the 605 lb./f t.
beam was silicon-killed.
He also discussed the A-36 plate composition, and then summed up the significance of the VEPCO data on the A-36 material (see slide).
Next, Mr. McAvoy showed where the A-572 (8 beams) beams were used.
Mr. Etherington inquired if Mr. Golds teins's analysis hed included the loss of these beams and was told they were not considered.
Mr. McAvoy next discussed the Charpy V Notch tests on the A-572 material in both the longitudinal rolling and thorough-thickness directions (slides).
He also covered the chemical analysis, pointing out (see slide) that the 605 lb./ft beam had the maximum nitrogen content of.015 (within the spec.
of.018 maximum).
In response to a question by Dr. Bush, he indicated he would provide sulfur content data later. He then provided his conclusions (see slides) ending with the conclusion that the operating temperature of the A-572 nust be increased.
Next, Mr. Van Sickle described how the elevated temperatures would be achieved.
He showed a slide of the lower support frame for the steam generator which illustrated the various heat sources. He illustrated a temp-mat tent and then went through a thermal analysis to show that in-tent temperatures of 180 and 200 F could be maintained at ambient temperatures of 70 and 105 respectively. He explained that yield strengths had been de-termined and showed curves of how they are lowered with increasing tem-perature. Also included in the analysis were bolts, blocks, and other OFFHCHAL USE ONLY
North Anna A 'J Meeting Date: 7/7/76' l miscellaneous items. He then described a proposed hot functional test to check the thermal analysis.
Mr. Bennett then called on Professor H.T. Co'rten, of the University of Illinois', a VEPCO consultant, to discuss fracture mechanics.
Dr. Corten cited his objective as determining the flaw size that would just be large enough to cause fracture. He would take into account static-loads
(< 6Ksi) as well as earthquakes and vibratory loads which might be coupled with a pipe rupture. Fat 1gue is not a problem. He would look at 2 types I
of defects -- defects associated with the weldments, (lack of fusion ~, slag inclusion, lack of penetrations) and lamellar tears.
Dr. Corten showed a slide of Charpy energy vs. temperature for the first four heats of A-36 available to him (see slide). The. specimens were taken from quarter-thickness locations as specified by the code. Data were obtained at a strain rate of 10 inches per inch per second, which is higher than any prevailing in the structure.
He showed that for high yield strength materials, there was only a small shif t due to rate of loading; for lower yield strengths (e.g., 42 and 46 Ksi) there was a rather large shift (up to 160 ).
0 At this point, Dr. Corten explained to Mr. Merkle that, while Barsom may have used a crack tip elastic-plastic boundary strain rate, he (Corten) had not.
He estimated his precision at no more than within i 5%, and felt it representative of something comparable to nominal strain rate at a dis-continuity such as a weld toe.
Dr. Corten next showed curves used to convert Charpy energy to K1C.
(See slides attached).
He indicated that recent SunShip and VEPCO data from several different heats of material fell below his previously used curves.
1 At this point, a decision was made to further heat the A-572 beams. The lowest line gave a value of 1800 at 18-foot-lbs. He reasoned that, based on the lowest SunShip data a 32 foot-lbs, the flaw size would be the same at 1800F* as that previously calculated at a 100 lower operating temperature 0
because the toughness would be identical.
Dr. Corten completed the sketch below:
-/
2C
,r.
a a
=.1 j-2e
- 225 F = Temp. of 32 ft. lbs. impact energy 0
-45 F = Temp. shift due to strain rate difference between Charpy impact specimen and structure l
OFFECEAL USE ONLY
L 1FHCHAL USE 0.. Y North Anna Meeting Date:
7/7/76 Next, Dr. Corten showed a graph of K1 vs. critical flaw sizes, for various residual stresses (0,10,18 Ksi).
He indicated Unit 2 would be in the 10 Ksi category as a result of stress relieving.
Unit 1, he expected to be below 18 Ksi as a result of hammer peening. This graph was based on a suggestion ~
by Harricon and Dolby.
In response to a question from Mr. Merkle, Dr. Corten indicated that a residual stress of 10 Ksi for annealed steel is a fairly standard as-sumptionland probably overly conservative. 18 Ksi for hammer peening is not as easy to defend, but at the toe of the veld, looking transversely it would also be conservative.
Mr. Merkle asked for a reference on the residual stress.
Prof. Corten estimated rero in the direction norm'al to the lamellar tear because it is the direction in which peening is done. Care in peening is critical.
For the lowest value of toughness 32 foot-lbs., corresponding to a charpy impact temperature of 212 F from SunShip, Dr. Corten arrived at a critical crack size of.27 inches deep by 2.7 inches long, a size which, he reasoned, I
could easily be detected.
I l
Dr. Corten stated that without knowing the J-integral he could not do an exact calculation, but by going to strain and dealing with a strain that does not exceed the yield point, he'could do an approximate conservative calculation.
He stated that at 80 F, all the members are in a condition Mr. Pellini and Mr. Harrison would say are crack arrest conditions, yet it was planned 0
to operate at 180, 1000 higher. The added temperature is to take care of several heats of A-572 that cannot be tied down, though several other heats indicate they would perform as well as A-36.
Dr. Eush asked about bending monents on the welds if the snubbers were held up.
Mr. Van Sickle stated they.had hydraulic snubbers, and the supports were in the direction of major expansion.
Mr. Merkle pointed out that it needs to be determined that there is no inconsistency between the definition of strain rate as used by VEPCO and that used by Barsom in his paper on toughness criteria for bridge steel (to which Dr. Corten's slides referred).
Next, Mr. Peter Hepp, Executive Vice President of SunShipbuilding and Dry Dock Company, indicated that the earlier presentation had greatly affected his planned presentation. His prepared presentation is attached.
He used Figure I to show that there is a 5% confidence level that material in one VEPCO heat is in the brittle fracture region.
From there on he summarized the attached presentation.
Dr. Canonico commented that he wondered why two pieces of metal from the same heat should behave so dissimilarly.
Mr. Hepp pointed out that test data showed great variability.
OFFECHAL USE ONLY
...w.
RCHAL USE C.mY North Anna Meeting Date:
7/7/76 i
Mr. Merkle asked VEPCO if they agreed with Mr. Hepp's statement that 90% of the welde have not been volumetricly inspected.
Mr. Perkins explained that VEPCO ultrasonically examined selected high stress welds, those more prone to lamellar tearing.
Mr. Hepp said he recalled 93% as the number, but said in VEPCO's defense,,there was no choice since it would be almost impossible to UT them.
Next, Dr. Pellini pointed out that average shifts of NDTT are meaningless when some steels can vary from a shif t of 160 to 00 This has not been resolved within AASHTO. He discussed aberrations in steels - a 517 Type that should have had a 600 shift had 0 and failed. He cited the Martha Ingram barge where Charpy values did not afford protection.
He added that if a crack tip does not end in base plate (e.g. the heat affected zone, backing bar), one better know the properties of the material before one uses a K curve.
Mr. Pellini added that the y
repair welds are not stress-relieved; peening provides uncertain stress relieving.
Mr. Etherington asked if Mr. Pellini concurred that no failures had occurred at room temperature.
Mr. Pellini cited a World War II ship that failed at 60-650, and a 1954 case of A302B vessels that failed at 70.
Dr. Canonico asked if the repair welds were not preheated.
Mr. Fellini indicated that for a long wold the cooling rates were more easily controllable, but in a small region of repair the analysis gets tricky, Mr. McAvoy (VEPCO) volunteered that the repair welds v3re preheated.
Dr. Canonico asked if good preheat were used if one would not expect good toughness in the heat affected zones.
Dr. Pellini pointed out documenting the control is not clear cut.
Next, Dr. John Harrison, Deputy Head, Welding Research Institute, Cambridge, England, a SunShip consultant, discussed the relative
{
toughness of webs and flanges in sections. The flange, being thicker, cools more slowly resulting in coarser grain size, which will give it lower toughness.
Dr. Harrison then cited additional instances of failures at room temperatures and above -- a Japanese storage tank, 3/4" thick, spherical, similar to A-36 failed at 680F, a 6" Mn steel U.S.
vessel at 690F, and failures in ASTMA 515 steel at 70-80 F.
]
Dr. Canonico asked about the faults in these steels and Mr. Harrison l
indicated some were so small they couldn't be detected.
He indicated there was evidence that low steel toughness will go along with low heat affected zone toughness.
1 1
i l
OFRCHAL USE ONLY
\\s-
\\
OFFHCHAL USE ONLY North Anna-Meeting Date:
7/7/76 Dr. Harrison then stated that he had checked with colleagues in Germany, France, and Japan.and, while they had no code requirements, normal good practice required certain longitudinal and transverse Charpy properties for such structures.
He added that no changes had been made in these requirements since 1970.
Next, he pointed out that structural redundancy was not _always suf-ficient since failure of one element can cause successive failures of others._ He cited examples of Belgian bridge failures.
1 Dr. Harrison continued that residual stresses in welds which were not stress relieved and where the final pass is not peened'could approach the yield point.
He cited work by Nordell and Hall giving transverse l
stresses of 3/4 yield in 1" thick plates conventionally welded.
In response to questions from Dr. Canonico, Dr. Harrison stated that these were rectangular plates with the maximum residual stress usually on the centerline of the weld.
Dr. Harrison indicated there was an error in his first fracture mechanics analysis. Since it was an elastic-plastic situation he should have corrected the stress intensity factor and therefore the crack length for plastic zone size.
Mr. Merkle pointed out that Professor Corten used a factor, Q, for this.
Dr. Harrison pointed out that the use of ptrmanent backing bars introduced a sharp crack-like discontinuity between the bar and the structure.
Dr. Bush expressed a concern for the possibility of low ductility under load conditions with a suffi,ciently low upper shelf (Charpy curve slope) value.
(e.g. about 30 or 35 foot-lb.)
Dr. Harrison made the point that at 180 F the arrest temperature has come off the shelf and, therefore, more reliance can be placed on the redundancy of the structure.
Mr. Merkle asked Dr. Harrison if the relationship between K, the stress I
intensity factor, and 6,_the crack opening displacement, was consistent throughout his analysis.
Dr. Harrison said he would reexamine his analysis in order to answer this question.
At this point the Subcommittee caucused briefly.
Mr. Bennett next called on Dr. Stout, a VEPCO consultant.
Dr. Stout pointed out that the strain aging effect occurs principally below the heat affected zone of a weld, and he *Inew of no failures attributed to the strain aging effect below the heat affected zone.
He also noted that Dr. Harrison's list of failures included none above 16 foot-lbs. and none above 1000F.
OFFHCHAL USE ONLY
/FHCHAL USE C. Y North Anna Meeting Date:
7/7/76 He also noted that the backing bars were A-36 certified material, and that they did not provide a path for propagating fracture, though he conceded that the notch effect must be considered.
Next, Mr. Cavallo, S6W, discussed peening. To achieve uniformity and correctness seven parameters were recognized:
- 1) A specified tool size was used - 1/16" ball.
- 2) A specified pneumatic hammer (Ingersoll-Rand 182) was used.
- 3) Specified air pressure - 80 psi was used.
- 4) All weld layers except root passes, buttering passes, and the final layer were peened.
l
- 5) Peening continued until individual bead slopes were indistinguishable.
- 6) Peening operators instructed and tested by welding engineers.
- 7) Visual samples of properly peened surfaces were prepared for use by quality control personnel.
Mr. Etherington asked if any tests were run in advance to qualify the procedure and show what it would accomplish.
Mr. Covallo stated that iron-workers prepared test coupons, and were then certified as proper peeners.
No determination of residua,1 stress was made.
Critical dimensional tolerances were monitored and kept within 1% to avoid distortion, which was an indicator of residual stresses in the structure.
Mr. Dromerick next asked James Knight, NRC Staff to make a short presentation.
Mr. Knight stated that the Staff review was not yet completed. The Staff has an independent analysis underway at Aerojet. Results are expected in September.
Both Mr. Hepp of Sunship and Mr. McAvoy of VEPCO agreed to provide Dr. Canonico a specimen of the very low shelf A-572 material for him to test at Oak Ridge.
After mention of follow-on meetings on August 5, on Reactor Pressure Vessel Blowdown Forces and August 11, on the North Anna project t.he meeting adjourned at 7 p.m..
OFFHCHAL USE ONLY
_