ML17272A886
| ML17272A886 | |
| Person / Time | |
|---|---|
| Site: | Columbia  |
| Issue date: | 02/06/1980 |
| From: | WASHINGTON PUBLIC POWER SUPPLY SYSTEM |
| To: | |
| Shared Package | |
| ML17272A885 | List: |
| References | |
| NUDOCS 8003310017 | |
| Download: ML17272A886 (162) | |
Text
.UNITED S ATES OF &AMERICA NRC NUCLEAR.REGULATORY COMMISSION Zs t Staf Briefing on Rev'ew and Evaluation of NNP-2 Sacr'ficial Shield Wall, Pipe ~ihip Restraints, C%
and Related Structures Fl Presented by C4 Washington Public Power Supply System QP 6
o U 10 z
CO Fifth Floor Hear'ng Rcom East <lest ower Building 4350 East best Highway 4
12 agL~r,tP ~ -917 Bethesda, Naryland
~gzoo~9',..
13 7-" ~'fDac7imenh RP>gl;,.GEIY GOGKHRLF.
6 February, 1980 9:00 a.m.
teal E
ow PPXSENT: J. B. Henderson, Nuclear Regulatory Commission, presiding 16 cd G; v . Reinmuth, Nuclear Regulatory Commission 17 R. E. Shewmaker, Nuclear Regulatory Commission
Ã.'ishop, Nuclear Regulatory Ccmmission
'4 ~ 44 18 T.
R.
N.
C. Haynes, Nuclear Regulatory Commission J. wagner, Nuclear Regulatory Commission t
C7 R. Gamble, Nuclear Regulatory Comm'ssion S. P. Chan, Nuclea Regulatory Commission 20 H. D. Thornburg, Nuclear Regulatory Commission P. Snaider, Nuclear Regulatory Commission R.
R. P. Georgiev, Nuclear Regulatory Commission Robart, Nuclear Regulatory Commission 22 R. Foley, Washington Public Power Supoly System 23 R. . Johnson, P7ashington Publ'c Power Supply System
- 0. Z. Earle, "ashington Public Power Suoplv System
- 24. E. Nitherspoon, Washington Public Pove Suooly Syster.
- 3. Burns, 4'ashington Public Power Supply System ZR/QoSCkE A
~ ALDERSON REPORTING COMPANY, INC.
POQ Q $ Q Q (7t
PRESENT: (Continued)
J. Parisie, Burns a..d Roe
-L o Akers, Burns and Roe 3
Good, Burns .and Roe O. O'Donnell, Burns and Roe R. Snaith, Burns and Roe Celnik, Burns and Roe 5' Fialkow, Burns and Roe G. H. Hansen, EFSEC 7 E. Cloth, Stone and Webster 8 N. D. Lewis, State of Nashington P. Francisco, Niagara:mohawk . Power Company iC. Nard, Niagara ~mohawk Power Company 10 13 15 16 17 18 19 20
'21 22 23 24 25 ALDERSON REPORTING COMPANY, INC.
P R 0 C E E D I N G S
' HR. HENDERSON: . Okav. I guess we can get I am Jim Henderson from the XE Office here in Bethesda. And I started now.
will be chating this session; md. I will have some, but limited, technical input to it.
These other gentlemen, I think I will let each one of them identify himself. 1t is a little simpler.
?4R. WAGNER: I'm Bill Wagner from Region V.
HAvNES: Ron Haynes, section chief, reactor 10 construction support branch, Region V.
NR. BISHOP: Tom Bishop with Region V.
'IR. SHEWHAZER: Bob Shewmaker, structural e .gineer, headquarters.
14 ~SR. REIELXUTH: G. W. Reinmuth, 'CX, headauarters.
15 "IR. GAiIBLE: Ron Gamble, materials encineering, 16 licensing.
17 HR. CHAN: S. P. Chan, structural engineering.
18 HR. HENDERSON: I think mavbe I have a'ittle bit 19 of an introductory statement to see whether we have this 20 our understanding is the same as yours.
21 The Washington Public Power Supply System is the 22 sponsor and licensee for a nuclear power plant generator 23 identified as !JPPSS-2, an 1100 megawatt boiling water reactor 24 with a ~lark II containment. I understand it was designed bv 25 Burns and Ro'e to functional specifications prov'ded by General ALDERSON REPORTlNG COMPANY, INC.
dsp3 Electric Company.
A vital element of this containment system is an internal structure called the sacrificial shield. .he sacrificial shield is an annular element which surrounds the pressure vessel and serves several functions. Xt restrains llew 5 the reactor vessel from excessive radial and
~oM<i 2 c,'(
IO lA 6
CI C4 7 movement under various design basis occurrences.
C4 ID Cl 8 Zt provides anchorage for various pipe wh'p restraints.
V Zt supports structural elements which in turn support both 0
U 10 safety related and non-safety related 'eauipment. And it z
I lG provides neutror. and gamma shielding for radiation ser.sitive Q 12 equipment and for plant pe sonnel when the reactor i: s shut
'i Cg E
14 down.
ln consideration of the various loads that the o 15 structure may experience, the structure is massively designed CQ I A
16 and fabricated from steel plates.
CO 17 individual compartments within the shield are filled with concrete for biological shielding.
19 Significant problems have been identified in the 20 fabrication and erection of the steel fo" the sacrificial 21 shield. During investigative work on tnis subject, voids were 22 detected in the concrete shielding fill. Fabrication problems 23 have also been identified in pipe whip restraints and concerns 24 have beer raised about fracture toughness of certain parts.
25 As a result, the licensee agreed with Region V that
~ ALDERSO'N REPORTING COMPANY, INC.
0 0
work on these areasshould be halted pending further resolut'on
'I of construction quality problems. l<ork should not restart until the NRC staff h'as been, apprised and has been convinced that proper correciive action has-been described.
tO 5 t'7PPSS feels it is now at the point where it'as IC LQ 6
sufficient information to satisfy thai condition and through CI CC Region V has requested this meeting.
fg CI CI Ol I have cautioned that we cannot absorb and digest the 9 necessary information fast enough to permit an HRC agreement 10 on work restart todav.
Ne need time to eview and understand ihe material. I 12 am advised ihat HPPSS recognizes this and believes that
' 'oday's will nonetheless c
13 meeting be useful.
14 I have had the opportunity for a cursory eview of an E
o rg 15 advance copy of a document which I understand will be I
distributed today. And based on that review, I agree on the 17 benefits of this rueeting. +s f>~ '; W~C9o~ .cc. Mq c.
I+
18 So, I guess with that, I will turn it over to you, Hr. Foley'.
20 .$ R. FOLEY: Okay. bir. Henderson, you have outlined 21 the nature of the session he e today. I am Roger Folev, the deputy prospect manager for engineering at unit 2. I would 23 like to introduce -- without intrcduc'ng all of the people 24 who are here today, though, I would like'o introduce Nr. ?<ickey 25 <<witherspoon who is the corporate quality assurance manager for ALDERSON REPORTING COMPANY, INC.
Supp'y System.
" be have a number you mentioned in you" opening that we would, distribute our report he're todav, and we did not brirg a- distribution set, although I think we have a number of them amongst the team. Ne would be happy to leave l
some with you. And any additional copies we would be happv to forward them to you.
NR. HEVDERSOV.: I think, administratively, since the ultimate technical review is an NRR function, that you might 10 as well forward those -copies to Dave Lynch, perhaps with a copy of the trarsmittal to me, but not necessarily a copy og the report itself.
13 NR. FOLEY: Our purpose here today is to summarize the results of the Supplv System's evaluation of construction on*
15 th sacrificial shield wall, pipe whip restraints, and related 16 structural steel containment in unit 2 and the auality records 17 associated therewith.
18 Our presentation, is in response to some 21'oncerns C~
i CI 19 identified and documented by Region V in October and succeeding C7 weeks of this past year.
C
Ãe have organized the presentat'on 21 into three sectiors: the sacrific'al shield wall, pipe whip 22 restraints, and structu al steel.
Hill will attempt to briefly summarize after each 24 section so as to segregate them clearly. In each 'nstance we will provide a brief introductory review .of the design ALDERSON REPORTING COMPANY, INC.
cons'derations associated with that particular structure.
With respect to the sacri icial shield wall, there are some l2 concerns exp essed; we have divided those concerns into essentially two parts: .
three that we judge to be primary concerns; and nine secondary concerns.
The primary concerns are a correction weld as a result of a construction deficiency between rings two and three of the wall; a repair of shielding voids; and a discussion of general weld quality in the fabrication process. in connection with 10 the third item, the third princ'pal concern on the sac ificial shield, wall,- this report and th"s discuss'n today is a status 12 eport.
As you will see, we have not concluded made any 14 final conclusions with respect to the construction quality. Our 15 evaluation and inquiries into that area have been complicat d, i ~
16 and we have enlarged the scope of that inquiry as a result of the 17 kinds of deficiencies that are coming to light. on the pipe whip 18 restaints. Since the pipe whip restra'nts and the sacrificial 19 shield wall have the same fab icator, we feel it incumbent upon 20 us to consider the interrelationships between the two kinds of 21 problems that were discove ed.
22 And so this again is a status report, particularly 23 on weld quality and some of the documentation deficiencies.
24 The initial presentation on the sacrificial shield wall is by Dr. '<orris Pialkow, who is the -- who is a civil 25 ALDERSON REPORTING COMPANY, INC.
engineering specialist v'th Burns and Roe and the des'gn engineer for the sacrificial shield wall.
DR. FIALZON: Hy talk will cover the structural aspects r ~
of the shield wall as originally designed, and it furnishes background information pertinent to the various concerns and corrective measures which will be discussed separatelv and subseauently.
I have a series of flimsies or transparencies, and I vill proceed with those.
10 (S 3.ide)
In short, this talk will cover methods of analysis 12 and design and a few brief construction features of the vali.
13 And as was noted earlier, this wall is an important component 14 of the BT R ~lark II tvpe containment at, Hanford.
Ne see here the shield wall in relation to other 16 principal structures 'n the dry veil. The wall basica" ly has 17 a 15 foot outside radius, two feet thick, about 48 feet. high, 18 cylindrical shaped structure fil'ed with concrete; it rests on 19 a pedestal below and is supported above by a stabilize= truss, 20 which in turn is supported by the biological shield wall.
21 inside of the shield wall is the reactor vessel vhich 22 is separated from it by the so-called annular space.
23 (Slide) 24- he shield wall is shown here in some detail. Ne 25 see again that it is a cylidrical structure, he framework ALDERSON REPORTING COMPANY, INC.
dsp8 consisting of vertical columns and hori ontal ring beams. But
' the vertical columns are shown in the section below and have either the wide flange shape or are box shaped.
Thgoccur at 15 degrees around the periphery. of the tO 5 wall. he ring beams are shown in the sectional elevation and C
C4 lA IQ 6 are either in channel cross section or box cross section. While CI C4 7 I have this here, I would like to point out a pertinent erection feature of the wall. ln order to accommodate the reactor ".
I U
6 vessel which comes with protruding pipe ends, which then extend 2l oC into the U 10 space occupi'ed bv the shield wall, the shield wall was z
CB erected in two main phases; the lower phase up to elevation U 12 541 was erected in place inside the containment.
built outside
- F 13 14 The upper portion of the wall containment as three 120 degree assem>lies, was each the full height.
o .15 These assemblies were temporarily joined as one unit. outside I I lifted into j
16 containment, the containment and set on top of 17 thi's wall, the lower portion.
18 Then the temporary joining of the assemblies was 19 disconnected, and the three assemblies were jacked out about 20 two feet eight inches to accommodate insertion of tne reactor 21 vessel, which was then lifted from outside containment and 22 placed inside.
23 Subsequent to,lifting in of the vessel, the three 24 assemblies were jacked back into their final position, and 25 the erection was completed. The individual panels formed by t'e
-ALDERSON REPORTING COMPANY, INC.
dsp9 1
network. of verticals and horizontals are each covered on the
' 2 inside and outside by skin plates, except where there are openings to accommodate the pipe passing through.
3 (Slide) ld Certain bas'ic features pertinent to the wall are noted C4 IQ here. In response to requests by NRC, a reporting listing the IA r A
C4 methods of ar alysis and design was submitted and subseauer tly C4 CI approved by NRC by a letter dated October 15, 1975. In short, 0 the report ind'cates that these methods were in conformance and C
10 are in conformance with the Standard Review Plan 3.8.3, which U
z CO is the pertinent document; in particular, the loads, the U
load combinations,and acceptance criteria as used conform with that document.
14 And the general bas's of design was the elastic E'
15 working stress method, Part 1 of the 1969'ISC design 16 specification.
17 As indicated in the report, just referred. to, the shield 18 wall is arQyzed as a space frame, and here we see a flat 4~
C elevation of the cylindrical framework that was adopted for 2
analysis.
21
{Slide)
.22
'.~e see that model, structural model is the fu'1 c'rcle 23 from zero to 345 and back again to zero. It extends from the 24 base at the pedestal to,the biowall and inc'udes the stabilizer 25 truss. To get the legend behir.d what is shown here, the solid
'LDERSON REPORTING COMPANY, INC.
10 dsp10 lines represent. columns and the ring m mbers; the panel having an opening is represented. by a panel with crosslines in it, and 3
all other panels are covered by skin plates. The e are 24 boundLing nodes along the top of the oed stal and eight bounding 5
nodes at the junction with the biowall.
An idea of the size of the model is indicated below C4 Ih lO 6
C4 Ol 7 where the breakdown is listed. Ne see that there are 378 CV 8
members included and 136 elemen s representing the skin plates.
,C4 U
(Slide)
Z U
o C
As I just indicated, the wall was analyzed as .a z
CO space frame; by this method, the analytical model closely U 12 approximates the actual structure. he beams and columns are z
represented as members of this space frame with continuity at
- 13 14 all internal joints.
t he model extends from the pedestal up 15 the stabilizer truss to the containment vessel, and the skin
) plates have been included in the structural model as finite M
17 elements with nodes at the framework joints.
t~%
Pn 18 The actual analysis is then done using the commercially available computer program STRUDL.
20 (Slide) 21 The boundary condit'ons as used ir the source model 22 are listed he e.. Xn short, at the 24. joints along the pedestal, 23 the restraints are in the circumferential direct-'on and of course 24 in the vertical direct'on; at the junction with the stabil'er 25 "russ, the onlv const aint is in the tangent'al direction.
I
. ALDERSON REPORTlNG COMPANY, INC.
dspll he significant loads which were included in the analysis were dead and 1've loads, seismic, including OBE and safe shutdown earthauake, annulus pressurizati'on and reactions du 3
to pipe break.
IO 5 The annulus pressurization was taken as due to recircu-hl lO 6 lation outlet and inlet lines, breaks in the feedwater lines, 7
and breaks in the RHR-LOPCI lines. he pipe break reactions C4 C7 o 8 which were included were those that caused the annulus C4 V
d 9 pressurization and in addition various main breaks in the dry zo 10 well proper.
U z
V) 11 In considering the load combinations listed in the U
z 12 S tandard
\
Review Plan, it was determined that only the factored.
C5 13 load combinations were critical.
CO 14 (Slide)
Cg oLe 15 These are listed he e with the associated permissible lg 16 stress levels; of these, in general, combinations five and rA 17 six were the controlling ones. And we see in each of those
~
Q 18 t:hat the previously called out significant loads were all C
19 included in five and six.
CO CI 20 (Slide) 21 This last transparency shows highlights of the design 22 of the wall. As I'reviously reported, the elastic working 23 stress methods, Part 1 of the 1969 AISC code, was used; as to materials, all plates. except for the top ring, ej 24 25 are A36 steel. The top and members, ring ~ras a hiah st ength steel., A5888.
'LDERSON REPORTING COMPANY, lNC.
12 dsa'12
' The weld member metal was in the E7000 series or equivalent.
design starts with the computer output for these members, The which of course were the ring beams and columrs. And the compute output furnishes each end of the member three components of force 5
and moment. Intermediate values are determined where pertinen+.
6 he wall in general consists of specific tvpes of 7
members, and for each type the controlling stress results were 8
used throughout the design. And connections are full strength welded connections.
10 The skin plate design also starts with the computer out put, which furnishes the membrane stresses as the normal and 12 shear stresses. And the plate thickness and attachment welds 13 were also determined from these controlling stresses.
fg 14 This corcludes by talk. If vou have any auestions, o
Pg we can take them Qp.
)
CO 16 HR. HENDERSON: ves, I will start out with one, sir.
CZ 17
)g P~t Have you done anything about looking at the as-built conditions, 18 the as-built dimensions as against the prescribed dimensions?
DR. PIALKO"/: he as-built dimensions that are the 20 subject of a 'further talk later today are those which charac-21 terize the nature of the welds which were accomplished. The members themselves, I believe, are not, in any auestior, nor are 23; the overall dimensions of the. wall.
, 24
'4R. HENDERSON: Ny question ar'es from the fact that 25 in your original submission on this a great point was made of
'LDERSON REPORTlNG COMPANY, INC.
13 dsp13 citcnln "ity to within 1/8 incb~of olnmbpwitbin I/4 inch f om
.2 top to bottom, within 1/8. 'nch ir. anv 10 feet.
i 3
And looking, at the projection of ho'w a weld might be applied, there is a clear indication that your e'ghths of inches are mor like ten times that much.
Ol lh 6 DR. FIALZON: " Nay I refer this question to someone. I LC C4 7 have no "S. HENDERSON: I have no specifics, but since that 0 degree of d'nensional control was iaentified as a very important
(
zo U 10 thing 'n the original submissior., I was just wondering if z
you had gone back, if vou had been provided the as-built U 12 dimensions, and have gone back through this to find out what
' difference that Q
makes.
tO Pg DR. FIALKON: I have not done that. Ny impression at .
C 15 this moment is tha" these types of deviations would be relatively 16 minor in so far as the structural behav'r of the wall is CO 17 concerned.
tJ
~
ho 'l8 NR. FOLEY: I do not have ?".r. Gioninni he e. Unless c
CI 19 Larry can'peak to.that I suspect that ?h . Gioninni would 20 have been the indiviaual who was ollow'ng that, most closely.
I 21 But I have ro reason to believe at this point that we have any 22 difficulty there, and I know that there were close controls, i
23 i fo example, on settirg the bottom ring beam on the 24 pedestal to get them off. to the right start there.
25 One of ou difficulties in the construction process ALDERSON REPORTlNG COMPANY, INC.
dspl4 arose from the shimming that was recui ed to achieve plumb, and so on and so forth.
i So while I cannot give you absolute confir...ation as to a statement of fact that that has been taken care of, I tO 5 have reason to doubt at this point that it was not pr'operly 6
pursued. 'n the field.
C4 CO C4 7 If there a e no questions, Nor is can just go ahead.
CO C4 IiR. BISHOP: I have one auestion..
U ci 9 NR. JOHNSON: I am Roger Johnson, QA manager for unit Q
10
- 2. And we have no docum nted non-performance based on our U
z M
inspection program as far as the dimensional tolerances for .
plumb on the wall itself.
13 NR. HENDERSON: Do you have any documented record of the as-built dimensions that confirms they are within hl o specification records?
PQ 16
."4R. JOHNSON: Ne do have an as-built Vi 17
~K. SHENNAKER: Are the specs the same as those that
,F 18 t~r. Henderson mentioned that vere in the original submittal, or 1 9 were the specs it was actually built to somehow changed from the 20 j submittal>
NR. HENDERSON: This is 74-2-R2.
22 HR. FOLEY: !!fe would have to go back and examine that.
23 HR. HENDERSON: Appa ently, revision zero.
24 HR. FOLEY: he.would have to go back and look specifi-cally at that.
~
ALDERSON REPORTING COMPANY, INC.
dspl5 D'. FIAL'RON: I do feel that these changes in dimension, the absolute c'rcularity is not something that affects the design to anv grea+ extent that l can see offhand.
So if there are such deviations in the vertical LO 5 direction or in the off-perfec circular'ty, I would not envision 10 4
6 that that is going to change the nature of the design loads or
~ O 7
the design conditions to any extent.
hl C
C 8 HENDERSON: I think it would probably be necessary O
A to speak spec'ficallv to that. One of the greatest areas of O
U 10 staff concern these days is the asynmetric blowdown loads. And
\
z M
if the cylinder is asymmetric also, that further complicates U 12 things.
2,'A 13 DR. FIALKOU: vou had another question?
14
?G. 31SHOP: Dr. Fialkow, concerning the significant o 15 loads used in your analysis, vou did not specifically mention rg i ~
16 loads f om reactor vessel lateral stability and loads from the Vl 17 radial beam system.
18 Here those included in the analysis?
E p>>
19 DR. FIALEOH: The reactor vessel is supported up above, C7 20 not directly to the wall, but it is suppo ted up above, and 21 basically the lateral load from the reactor vessel is car ied by 22 the stabilizer truss out to the shield wall without directly 23 loading the shield wall.
Y~hat was the other?
HR. BISHOP: .he rad'al beam system.
ALDERSON REPORTING COMPANY, INC.
dspl6 DR. FIALKON: he radial beam system; loads that were included under the heading of dead and live loads.
NR. HENDERSON: Is this radial beam system vou are talkirg about something aside f om the trusses that your lO diagram showed?
lO DR. FIALEOW: Yes. The radial beam svs em might 10 show up in one of these.
C4 ID (Slide)
C4 Q
d This is part of the radial beam svstem; there are z --
o 10 various platforms that are and pipe whip restraints that O
z M
are supported off the wall. In general, the junction there. is U 12 free to go radially. So the load imparted to the wall is verti-z A
13 cal or bending moment.
th 14 LAIR. FOLEY: Ne w'll discuss that radial beam matter 0 15 in 'the third section of the presentation,'and we have one slide 16 which will illustrate the radial beams in the containment.
CO
)g 17 NR. BISHOP: All right. hank you.
18 DR. FIALKON: If there are no other questions, then, C
C 19 I will go on to concern nunber one, which involves the inter ace CI 20 at elevation 541. Let me begin by putting in the written 21 cence n as it was written in the NRC memo. Let me read it:
22 "Ring three and ring four of the sacrificial shield wall are 23 not welded together as shown on the design drawings. Numerous 24 welds were mace to shims, between the rings in lieu of actually 25 welding the rings together. -he AE tentatively intends to ALDERSON REPORT1NG COMPANY, INC.
17 install' two inch partial penetration weld around the circum-ference of the shield, wall to structurally join r'ngs three and four."
Let me just show an illustration of these rings. I ld 5 would like to start off by saving that the term'nology, now, Ih lO Pl the designation of the rings which we use in this presentation C)
C4 C4 are those on the Burns and Roe contract drawinas. And as you see, they are designated two and three, and that. is the way ve U
zo vi' be re ferrina to them.
E 10 TÃe have the problem, then, of transmitting shear across interface between two and three, wh'ch was the level of 12 4
the interface between the lower portion of the wall which was 13 constructed in place and the upper portion wh'ich was lifted into M
N C ~
17'his 14 place.
0 15 Just; going back now, repeating in other words, so 16 (0
to speak, what our concern is: we are dealina with the I
A transmission of horizontal sh'ear across this interface at elevation 541.
19 And the concern develops because rings two and tnree 20 have not been welded together as shown on the contract dravinas.
21
~~ ~
Khat we propose is a partial penetration weld along the 22 exterior circumference between the rings. And this represents 23 a change from the oriainal design.
24 In this talk in connection vith this problem, we w'll 25 be covering the original contract reau'ements. and description of
'LDERSON REPORTING COMPANY, INC.
18 dsp18 the correction weld, discuss'on of the design concept used for the cor ection, and a summary of the controlling features of the correction.
(Slide) ld This vu-graph shows or is intended to show that the C) ld 6 original contract requirements in the section below, what was lA Cl C4 7 reauired we should see what was required. And that was constructing slot welds by cutting the slots in a web of C4 0 two above and then joining members two and three through
'umber Q 9 o
10 those slot welds.
U z
CO ll The plan shows a number of these welds; there were U 12 four such welds at, each 24 columns, placed symmetrically around.
z a each of the 24 columns around the wall.
D 13 14 inadvertently rather, to plumb up the upper portion fzj 0 15 of the wall, shim plates were installed between members two and 16 three and then inadvertently in many cases the welding of the CO t~ 17 slot welds was made from number two to the shim plate in between.
1 18 That vitiated. the intent of the design.
E 19 (Slide) 20 he extent and nature of the corrective measure is shown here. This is the outside of the wall,and what is intended "
22 is to build a correction weld along that circumference; the 23 extent will be such that in every panel of the wall the 24 corrective weld correction weld will be done for the width 25 available between splice plates, column splice plates, and this
'LDEQSON REPORTING COMPANY, INC.
19 dsp19 will be done in each of the 24 panels around the wall.
(Slide) 3 his transparency shows the two types of correction welds that will be used; the type that is adopted depends on lO 5 the xvidth of available ledge at the wall. Where that ledge is C4 LC IA 6 at least one inch in width, we use the upper type weld, which only requires cutting of number two as preparation for the weld.
CO C4 7 8 ~
he lower type weld will be used where we do not have U
ci 9 one 'nch available and there cutting of both numbers two and zo U
z 10 three is required.
M ll Both welds, for design purposes, have the same D
R 12 r,.inimum effective tom.
A 13 (Slide) h'ghlight here the design concept that, will be CO r~ 14 We o 15 for th's correction. As has been said, what we are doing used 16 is transmitting or provid'ng for the transmission of horizontal j
Le 7~ 17 shear between channels, the upper ring, ring two, and the
~
0 18 lower ring being three.
CI 19 These shears in the channel result from reactions be-20 tween the skin plate and the channel and reactions between the 21 upper columns and the channel. The direct'ons of these shears 22 are tangential or circumferential from the skin plates and 23 both radial and tangential from the co umns.
- ~4 ~
he analysis is based on a 115 degree panel and is 25 based on that panel which has the largest comb'ed shear. The ALDERSON REPORTING COMPANY, INC.
same correction is then applied uniformly to all panels around the wall.
3 Melding design and weld'ng procedures will be qualified in accordance with the structural welding code, ANS Dl.l, and allowable stresses used for the design are those assoc'ated with partial penetration groove welds.
(Slide) hi's transparency shows the applied forces and the design loads and then the resisting forces which develop in the weld, in the correction veld.
Above we see the design loads. tie have a tangential 12 load which results from the reactions of the sk'n plates and.
13 f om the colmnn, and this total load is taken to be equally 14 d'stributed between the two outside surfaces of the uppe N
15 channel.
16 Each has a magnitude of l63.5 kips. From the column 17 we have a radial'oad which acts in along 'the line of the 18 web of the column. The resisting forces in the weld which 19 developea those design loads are shown below; due to the total 20 tangential loaa, we developed. a uniform tangential weld. force 21 of 9.9 kips per inch.
22 Due to the eccentricity of this portion of the tangen-tial load with respect to the correction weld along the exterior surface, a moment develops which causes a radial load of uniformly varying magnitude. its largest value is 21.6 kips ALDERSON REPORTjNG COMPANY, INC.
dsp21 per inch, and due to the radial load from the column, we developed a radial weld force for this weld which acts only over a portion of this weld and has a magn'tude of 2.7 kips.
(Slide) in this last transparency, we summar'zed the controllin l
features of the proposed co ection. he controlling panel iC'%
Cl CI shear is due to a combination of dead, live, OBE, seismic, and Cl Cl annulus pr ssurization of pipe reaction all caused bv a feed-Cll 0 wate- break.
Q z
o L ~
10 he magnitudes we are talking about are 327 kips in U
z QJ the tangential direction and 27.4 kips radially. he controlling U 12
~
load combination is combination five and has an associated z
13 per. 'ssible stress level of 1.6 times the normal stress, normal 14 allowable st ess. And then the controlling design margin which N
6 oLe results, which is the ratio of the permissible stress to the SQ 16 actual stress developed, is 2.3.
N 17 his concludes my talk.
g 18 Are there any questions?
i CI 19 i'IR. HAYNES: Dr. Fialkow, in all cases, is there a Cl 20 ledge or is there overhang?
DR. FIZZ'COP7: i do not believe there is any ove"hang.
22 ..here is always some kind of ledge. it might be quite small.
23 Someone correct me if I am wrong on that.
- 24 HR. FOLEY: No, that is true.-
25 DR. FIALZOP7: There is always something there.
'LDERSON REPORTING COMPANY, INC.
22 dsp22 Design-w'e,there was supposed to be a ledge.
HR. SHENigQ<ER: . ~i jhat was the magnitude o f that ledge, design-wise? What was the specif'cation?
3 DR. = IALKON: Something on the order of, I think, a IQ little mo e than an inch.
C4 Ld lO NR. S EP7YAKER: ',vas it uniform all the way around?
C4 CO C4 DR. FIALKOi7: It was supposed to be uniform all the o way around.
C4
."<R. SHE$ 8GCKER: I think we ne d to know. That. would z0 10 give us some 'dea of what kind of tolerance that of those U
z M
two pieces when put, together if we know what the design ledge 12 is supposed to be.
a 13 Let, me ask another question: the only force that you CO 14 are transmitting across the interface 's shear?
to the slot welds,. the vertical O DR. FIALKO1/: Due 15 PQ i ~
16 forces are taken care of'onthose column splice plat'es I talked 17 about earlier. There is no problem on those. ,17e are concerned g
only with replacing tl'.ose deficient slot welds.
19
'R. SHE'"/NAKER: I had heard or someone had told me 20 that some of those column splice plates were also shimmed to make 21 Up for this 22 DR. FIALKON: There were adjustments made design-wise.
23 I believe we e.fee+ed the intent of the original design.
24 ALR. SHENI~KKER:, Have there been any stress calculations run on ary "of those splice plates that were not put ir. place the ALDERSON REPORTING COMPANY, INC.
23 dsp23 way they were designed?
e we e based Dp.. FIALZON:
on The adjustment methods calculations.
that were adopted NR. CHAN: How thick are those plates, the vertical plates?
C4 lQ DR. FlALFON: The splice plates?
lO CO HR. CP~: Yes.
C4 C)
DR. FDJZO77: The one on the exter'or and the interio-are different. i cannot g've vou a precise answer. Xt is on 10 the order of an inch, I would venture to say.
NR. CHAN': And the supports of those plates, did you 12 check the buckling strength of them?
resting
' 5 13 14 numb'er DR. FXALKON:
three, in effect, Actually, so as far you have number two as compression is concerned, on o there is a large area for t ansmitting the compressor forces.
16 (Slide)
Vl 17 These are designed there is weld all around this 18 thing, so there is no real unsupported 'length. And then t1 e load.
19 is transmitted from the column and from the channel, which in turn has received certain upward let's talk tension for the 20'1 moment from the steam plate and all through these welds 22 all around.
23 hese go into the splice plate, and. then the weld along 24 th's surface, this is transmit"ed'nto this box beam below. ';low, 25 compression-wise, basically there is member resting on member, ALDERSON REPORTING COMPANY, INC.
24 and this,thenithis splice plate is in effect fully supported all around.
NR. CHER: 'Is i" possible to have tensile strength at those connection welds due to overturning or due to earthquake?
DR. FIALKON: Yes, 't is; that is one of the condition we designed for, especia'ly of course w'th take the annulus pressurization. If we assure the b" ak was right in this panel somewhere pushing the vali this vay, you tend to get tension as a result of overturning.
10 NR. DIPINSKI: Was that based on the as-bu'lt condi-11 tions o original designs?
12 DR. FIALKOÃ: -In general, design was based on the original conditions,. but basically if ve go back to the space 14 frame if we go back to the space frame that vas adopted, we did not crank into the definition of that space frame that i ~
16 it had to be absolutely vertical or it had to be perfectly 17 circular.
18 Ne cranked in the coordinates of the node points, 19 and that is vhat defines the spatial configuration of the vali.
20 And if "hose coordinates vary by a small amount, it vould not 21 the stress on the members. rJe did not in othe words
'ffect 22 study this as a shell; we studied it as a space frame made up 23 of some 300-odd membe s and included the skin plates as finite 24 elements joined to those nodes.
25 So i" is my feeling that whether these deviate from ALDERSON REPORTING COMPAN Y, INC.
25 1
circularity or from perfect vertical conditions, our end stresses in the members would, not be affected to any degree.
.2 MR. DIPIHSKI: But you" conclusion is based on theory.
~ ~~
Do you have something more than just a theory?
lO 5 DR. FIALKOP7: Ne ran no calculations, no.
C4
'0 6
?<R. GA'EDIBLE: Nould you put up your figure that shows C4 C4 7 your ove view. of the vessel on the. wall?
U Is i+ this (Slide)'
d DR. FIALKON: one?
zo 10 MR. GA?<BLZ: Right. The vessel is suppo=ted by the U
z M
shield wall and U 12 DR. FIALKON: Oh, this vessel? This vessel has z
2 13 support down below through this skirt. It is supported N
CO 14 laterally at this level. by a stabilizer; then in effect: it fx1 h
oh is joined to the shield wall. But this passes right through
~
h the stabilizer truss to the bioshield wall.
) A 16 HR. GAK3LE: So you really do 'not have you really'.
CO 4 18 do not have the there is really no load carried from vessel to t: he shield that 20 DR. FIALKO&7: ~Hot in here, not in here.
21 NR. GAMBLE: It is down on the conta'nment or down on 22 the concrete.
23 DR. FIALKON: hat: is right.
24 MR. GAMBLE: Thank you.=
25 DR. FIALOÃ: If 'there are no other auestions, I will ALDERSON REPORTING COMPANY, INC.
26 dsp26 introduce the next speaker who will talk on concern number three, and that is ."Ir. Good.
FR. GOOD: Ny name is Lar y Good. I am the assistant.
site resident project engineer, and my presentation is to address concern number three. Okay, and here is the statement 5
lO 6 of the concern: that is, it deals with the sac shield wall.
IA And the concern is that numerous defic'encies in the structural 7
O4 C)
CI 8 weld auality have been identified in the sac sh'eld wall 0 structure.
o The deficiencies were identified in welds which ~vere U 10 z Deficiences include supposedly inspected and accepted. .
U 12 cracks, undercut, over'ap and..slag on welds, indicating that z
H inspections- could not have been properly performed.
13 (Slide)
P+
14 E'
15 ln response to this concern, the project has developed
~
q rg an action plan that basically started with issuing a stop-work 16 order to offsite contractors who had done any work on the sac 18 shield wall.
19 The second thing the project did was institute an investigation which so far has included review of inspection 20 21 records on the sac shield wall by any contractors in the last 22 two years in which the fabr'cat'on was basically done.
Ãe established a reinspection program and performed
.23 24 sample UT examinations on the wall. hese three things I will 25 cover in more detail.
ALDERSON REPORTING COMPANY, INC.
27 dsp27 And then the last thing I want to cover is the current status, since the investigation is not cor.,pleted at th's time.
Okay. The first thing, as -.1 mentioned, is we. went back to the inspection records, and what we have concentrated on were the inspect'on records that have been generated aga'nst the sac shiel LO 5 lQ 6 wall; as 1 mentioned, for the last two years, there have been lO CO C4 7 various contractors working inside the dry well who have been C4 C)
C) 8 working on the wall and around the wall.
C4 U
4 And "here have been approximately 500 attachments mace 9
o 10 to the wall in the last two years, and these attachments all U
z CO ll generated inspection reports. We then ~rent to the ~~
U 12 reports by the prime cont actor, the mechanical contractor and z
sor+ed out all of the inspection reports that identify defects 13 C6 14 in the wall.
o 15 And I have listed this in the second part which is 16 there have been 31 identified defects on inspection reports fff 17 and nonconformina reports by the prime mechanical contractor.
18 And the breakdown is that three dealt with porosity and slaa; c
19 two with cracks in the base material; eiahteen cracks in the C) 20 weld; two dealt with undercut; and six dealt with lack of 21 fusion or cold lap.
X would like to point out that of these 31 defects that 22 f 23 were identified by the prime contractor, 18 were found using 24 r.";agnetic particle examination, which is a more rigorous exami-25 nation than was applied on the wall during the fabrication, whicn ALDERSON REPORTING COMPANY, INC.
28 dsa28 goes into the third aart of our records investigation, the magne+ic particle examination.
The project':as invoked what we refer to as work arocedure 84 in January of 1978. And this work procedure lO basically recuires that if the contractor makes any attachments C4 lQ to the wall or 'f he removes anything off the wall, he is ecuire LO
\
to do a magnetic particle exam-'nation.
7
Ãe have, therefore, well over lOQO magnetic particle CI U examination reaorts on file, and we arecurrently reviewing those.
0 10 So, like I said, we basically have cone to the insaection reports U
z are looking at the magnetic and the nonconforming reports, and we Q7 aar tie le 'examinations.
12 13 And we have not been able to put it all together. But 14 this, is the current status.
o 15 (Slide)
à 16 The second part of our investigation which is parallel V) 17 to the first aart is that we instituted 'a reinspection program.'.
The reinsaection program consisted of a 100 percent reinsaection 18 all accessible welds in the sac shield wall, and this inspec-H4 c
19 of 20 tion is being performed by an ANS certified weld insaector from our OA deaartment and a welding engineer from our weld'ng groua.
21 22
.hey have been recuested to identify the type of deficiency and 23 the extent of all code deficiencies.
24 The curren+ status is that out of aaa"oximatelv 13,000 welds in the sac shield wall, we eel we can reinspect 1500. Of 25 ALDERSON REPORTING COMPANY, INC.
29 29 the 1500, as of January 31, 1979, we have inspected 1014 and found that 509 conform completely'to the ANS code.
Yes, sir?
NR. ROBART: January 31, '80?
HR. YOLKY: That is correct; '80; we have '79 up lO 5 C4 6
"here. It is 31 January of '80.
C4 7
<>IR..GOOD: Okav. Ne find 509 fu'ly in conformance with the code and 505 that had a deficiency to a code reauirement to some extent.
And I want to clarifv that. when I say we have 505 that have a deficiency, it may be one inch in five feet; i+ mav be tw
'nches in ten feet.
12 Nhat we are doing is we a e writing it down; we are 13 14 making sketches of them, and that is what is taking us so long.
oCe .he type of deficienciesthat we are finding are porosity; 16 incomplete fusion, which is overlap or cold lap on the surface; 17 improper profile; excess convexity; unde'rsized;, crate s; and a c strikes.
It should be noted that our reinspection program has not identified a crack or what we consider to be a serious 20 21 structural defect. F on this reinspection, we feel that the 22 contractor did perform the reauired visual inspection of the 23 wall because we have reviewed his records and found that he has identified defects in his weld'ng.
25
!ve have gone back out there and looked at them, and ALDERSON REPORTING COMPANY, INC.
30 dso30 they have been repaired.
It appears that he had tak n a lenient interpretation of the code in that most of these defects are. workmansh'p type defects; they are defects that are in excess of what the code allows.
lO 5 ICI 6 NR. REINMUTH: Let me see if I unde stand what you are lO CO saying. 500 defects you found bv visual inspect'on and onlv 31 C4 7 CI vou found by mag particle?
C C4 8
0 NR. GOOD: One was a paper review. ice went. back to the 9
zo defects, that when he went to attach 10 contractor, and we found 31 U
z CB ll to the wall, whether it was the base material or over a weld,
)~
U 12 he did an inspection of the wall; z
A And he documented 31 defects on IRs or NCRs. Okay?
13 14 hat's in the paper review.
E o Ye have gone back and wherever we could get to a weld, 15 hl A we did a visual reins@ection to tne same reauirements that the 16 97 17 original contractor was recruired to conform. 'And what we found is we are finding a high percentage of deficient welds in F 18 i
19 our einspect'on program.
20 "4R. HENDERSON: Vione of "hose represented cracks 21 where over 50 percent of the ones reported bv your installation 22 contractor represented cracks.
23 NR. FOLEY: hat is correct.
24 R. GOOD: hat is right, s'. Yhen you say 50 pe cent of the ones that vere documented, he did over 500 attachments, 25 ALDERSON REPORTING COMPANY, INC.
31 d.sp31 and out of 500 attachments, he only found 31 deficiencies in the wall.
3 So when you say 50 percent, it is 50 percent of the documented deficiencies. You are right. Okay?
But he made 500 attachments and MR. HENDERSON: But in your visual exam'nation, zero percent were identified as cracks.
C 7/LR. GOOD: Nell, 50 percent of the I have to go CD A C4 back. 50 percent of the documented deficiencies.
MR. HENDERSON: Yes.
10
""5R..GOOD: here are more than 31 XRs written on the ll wall; there 'are probably several nundred. Every time he went 12 to the wall and made an attachment, he wrote an inspection report 13 14 saying that he looked at the wall and that the wall was accepta-'le.
15 Then we required him to NT it. Then we reaui ed him to 16 butter it up and then perform another M . And then he does another MT on the route pass; another one at 50 percent. So 17 18 there that is why we have thousands of mac particle examina-19 tions.
And we do have hurdreds of inspection reports on the 20 21 wall. Ne went through those several hundred and identified 31 that said, hey, when we looked a the wall, we found a 22
.23 problem.
MR. HENDERSON: X understand.
24 MR. iTOPNSON: Excuse me. Let me make a point her 25 ALDERSON REPORTING COMPANY, INC.
32 dsp32 be want to make it very clear that ii 's not 50 percent of the defects that are cracks that the contractor found; what we are t ying to say is thai. there are numerous deficiencies and out of those we identified 31 that could be called a crack. It is ld not 50 percent.
lO HR. REINNU H: On the other hand, visual inspection, lQ C4 C4 it is almost impossible to detect. a crack. So when you say you hl CI CI C4 are seeing those cracks visually, that, is a little misleading 0
4 in itself.
o NR. GOOD: Okay. There is one other thing, though, U
z M
thai I would say like I sa'd there are well over 1000 U 12 on the wall, and out of the 1000, this is what we found,'CTs z
13 accordir g to our prime contractor. Okay?
And what we rrould like to do is break down ihose F
O 15 AT reports and find out. how many unique locations on the wall 16 like I said, many times we did an AT on top of a previous NT.
CO Okay?
r~
4 18 But we do we feel we have wel" over 500 locations 19 on the wall where ->re have done a magnetic particle examination, 20 and this is all that has been documented. And the reason we 21 wanted to take a good look at this was that there was some 22 concern that every time somebody went +o tne wall to identify what S
.23 a defect our review is pointing out is that that is not 24 true.
25 NR. RZINMUTH: The only report you had on your fabricator was a visual inspection.
ALDERSON REPORTING COMPAN Y, INC.
33 dsp33 ."~R. GOOD: That is riaht.
(Slide) ln addition'o our physical inspect'on, we have also selected ll electroslag groove welds and performed a UT lO 5 examination of these welds. And this was to supplement our visual reinspection program and to gain additional information lO 6 Cl abou+ the weld quality in the wall. Due to some restrictions in O4 7 CI 4
the pipe whip restraints, ice selected eleven locations and C4 8
O A 9 performed an ultrasonic exam'nation and found that all eleven o were acceptable.
U 10 z
CO ll Okav. The last part is the problem status. The status U 12 is at this time that we are performing the reinspection. Our R
Fl 13 metaluraical evaluation group will evaluate the effect. of the CO 14 deficiencies on the effectiveness of the weld joints. And then F
O 15 structural enaineerina will then determine if it has any effect on the existing design margins.
) 16 17 So, in summary, even though our reinspection program'.
has not identified what, we consider to be a serious structural 18 t
E 19 defect, we are planning to investigate and evaluate every C)
C) 20 deficiency that we found. And at that time we will be ready 21 to come up with a conclusion and a report.
22
."K. PZENNUTH: This is an overall tvpe of reassessment 23 24 NR. GOOD: Right.
25 le. RElViMUTH: Late- on you will be talking about ALDERSON REPORTING COMPANY, INC.
dsp34 soecifics?
HR. GOOD: Right,. We vill be talkina about breakina it down later on.
HP,. REINMJ H:~
What kind of inspection are you aoing
. 5 to require on the repair welds that vou are advocating?
NR. GOOD: I cannot answe tha+ ight now.
7 i~JR. FOLEY': We soecified an ir.spection on the reoair P4 C weld. Len?
C4 U
a 9 NR. AKERS- ves. Anythina that is repaired on the O
g 10 sac ificial wall is repaired to work procedure NE4.
Z ll ?~R. REIN."~U H: I ri.ean the main seam weld that y'ou.are 12 advocating as the fix for the shim oroblem.
z Q
13 KLR. AKERS: Okay.
14 NR. REINMUTH: What. kind of inspection .requirements?
hl
?4R. AKERS: ves. The way that is going to work is 16 that the first thing that happens 's the elec+rical resistance cd 17 heaters are put op that wall; it is grounded and it is mag Q
18 particled before anythina is welded. And then -- ar.d ther'eafter 40 19 it is block welded ir and each layer is maa oarticled. And it 20 works out to about four mag particles, the last one beina 72 21 hours2.430556e-4 days <br />0.00583 hours <br />3.472222e-5 weeks <br />7.9905e-6 months <br /> after cooldown.
22
?1R. GOOD: Ar.y other auestions?
23 I HR. BISHOP: Larry, you mentioned that the rev'ew 24 vou have done thus far has included 'nspection reoorts, non-conformance " oorts and maa oarticle reports. Your contractor 25 ALDERSON REPORTING COMPANY, INC.
35 dsp35 also included field- change requests as identifying problems.
1 Do you know if that is included in the scope of your document reviews?
MR. GOOD: Yes.
ld MR. BISHOP: Secondly, may I ask who did the sorting lO of the documents? Nho did the review? Was it the contractor or IA C4 was it the licensee?
MR. GOOD: Okav.'e are looking we have requested OI Q
6 a copy of every one of their magnetic particle examinations.
zQ Okav? And we are going to look at those. Ãhat was happening U 10 z
M was we were doing a parallel review; we asked the contractor U 12 to pull out all his inspect'on reports on the wall and he gave z
13 us the list of all the inspection eoorts that dealt with the wall.
14 E
o Ne went. through every sing'e one of those and broke 15 it down and found that 31 identified a defect in the wall.
ia 17 MR. FOLEY: To respond directly, the prime contractor.
4 18 is conducting a document review. He has essentially completed his document review. How we are conducting are own review, both 19 20 of the work that he has done and, where necessary, of the 21 original document.
22 MR. BIS1:OP: The review that he did, is that +he
'9, I
23 review that was done in the spring of or is there a more 24 recent one.
25 MR. GOOD: Ne just finished one.
ALDERSON REPORTING COMPANY, INC.
36 HR. FOLEY: It is just completed.
NR. BiSHOP: A second auestion: of the 1500 accessible i
welds, do you have a feel for how many of those are field velds and hov many are shop welds?.
NR. GOOD: Right now, ve are running about two to one field wells. I hope to have a breakdown to f'nd out the
. 6 percentaae of .rejectable or deficient field velds as opposed to the percent:age of def'c'ent shop welds.
But riaht now ve have not itemized all 505. Okay?
HR. BIS: OP: Do you know, oughly, again of these 1500 10 accessible welds how many of those vere shielded metal arc or 12 flux core arc or electroslag?
MR. GOOD: Ne had -- we b oke down the electroslaq; 13 14 there is approximately 1300 electroslaa welds. I think only three were made at the site. Three were made at the site, but FOLEY: Your population be said "of the 1500;"
16 and you are talking about a different population.
17 18 PR. GOOD: ."lo, I car.not.
19 NR. BISHOP: ~7hat I was aett.ing at:, of the eleven 20 electroslaa welds that you have ultrasonically tested, what does 21 that represent percertage-vise of the total accessible elect osla 22 welds? is that HR. GOOD: Ye estimate around 40 that ve can get to.
23 24 NR. BiSHOP: 40 elect oslags are accessible.
~!R. GOOD: Yes.
ALDERSON REPORTlNG COMPANY, INC.
37 dsp37 MR. BISHOP: ',";as that UT done in accordance with the ANS code or the original contract. requirements?
MR. GOOD: 'A77S .
MR. REINMUTH: Why. did vou choose to UT them? Normally the electroslag process creates large grains which causes some lO C4 problems when youare doing U.. Normally thev do RT.
lh LO C4 MR..GOOD: Are you asking why we UT-ed instead of ID RT-ed?
C)
Ol V
d MR. REINMUTH: Yes. No mally, it: is not. recommended o
10 to UT electroslag welds because of the large grain size, you QP z
M know, with the process itself. I am wonderinc why you chose U 12 that process.
MR. GOOD: Okay. Our welding engineer might be 13 better able to answer the question.
14 o MR. AKERS: What was your question?
15 A4 16 MR. BISHOP: I am concerned with why you chose to N
17 UT the electroslag welds. Normally, those are pretty urge PQ 18 grains, and the and a lot of times it is not recommended.
19 I will not say it is not recommended, but they choose to use 20 some other process. other than UT.
21 MR. AKERS: Is your question, why on these eleven--
22 MR. BISHOP: No. itchy did you use that technique.
23 MR. GOOD: Ne could not use RT.
24 MR. AKERS: Ne have no other way because we have 25 concrete in back of that. I;e have no other way to get to it.
ALDERSON REPORTING COMPANY, INC.
38 dsp38 Ne have to ao from one side.
l1R. BISHOP: Okay. I understand that. Obviously, you felt it was acceptable. You said that. But there is some cause for concern due to interpretation because of the large grains in the welding process.
ldt 5 lO 6
'<R. HENDERSON: Pursuina this a little bit further, LC CI 7
sometimes I think of UT as something with a deliberate bias CI C) 8 towards success. Zt cannot find anvthing wl.etl.er it is there C4 or not; so i it does not find anything, it is not. the e.
o NR. AKERS: Zt does not work that'ay for me.
Q 10 z
(Laughter)
CO 12 You know what I mean.
MR. GOOD: Anv other questions?
13 14 NR. ERYHES: Kith respect to the eleven electroslaq rg C
we'ds, how many linear inches did you UT?
15 MR. GOOD: Again, I am going to ask--
16 TR. FOLEY: How many linea inches of U ~
on those 17 18 electroslag welds?
HR. AYERS: Xt averaged about two to three feet. They 19 20 were sho t welds.
21 HR. GOOD: Twenty-six inch total on one; 17, 17, 17, 22
- 17. All but two are 17. The other two are 26 inches.
[IR. CYMES: Now, the remaining 29 that are available.-
23 24 thev are the same length, or wna" 'are they?
HR. GOOD: tie esti...ate that there are 29. lfe iust
~
25 ALDERSON REPORTING COMPANY, INC.
39 dsp39 really decided to do the UT reports in the last week two weeks ago. And I cannot give you an exact figure on that; even 3 the 40 is give or take . It is an approximate figure
.'R.
FOLEY: tie spent a considerable amount of time ld discussing the UT, and I do not want to -- since we have spent lO this much time on it, I do not want to mislead you. Ne have CI O4 one UT report .one U exam which we made on a weld which had a v'sual defect that we were concerned about.
Q d And that UT did in fact validate ou- concern about 0
U 10 that visual defect. So there are -- we did make 12 U examina-z M tions, ore on one that we picked, up w'th a problem visually..
MR. FINES: hat did confirm a defect?
MR. FOLEY: Yes, it did. It did confirm the defect.
14 MR.PAYNES:. So, 't looks like you were getting sound 15 to penetrate?
L'iR. FOLEY: Yes. He have no reason to believe in any 17 wav that. those U inspect'ons we e not fair evaluations of the
'8 weld s.
19 MR. BISHOP: Just for background information, one of l
20 the reasons we are asking so many auestions about the electroslag I
process has to do with the 50.55(e) report that the Supply System II
~ ~
sent to Reg'on V ' March of 1979 where tnev reported that the 23 same cont actor experienced a rejection of 34 out o. 45 pipe 24 whip restraint brackets that were. fabricated using the electrcslag 25 process.
ALDERSON REPORTING COMPANY, INC.
40 dsp40 And, as I guess we will talk about later today, we have had elect oslag problems in the contract 90 pipe whip restraints.
And so, obviously, we'are auestioning tne electroslag welds also in the sac shield.
5 One final auestion that deals with the evaluation that 6 is ultimately coina to be performed by Burr s ard Roe Engineering:
CO 7 have cuide'ines been laid down at +his point in time as to what CO Cl 8 that evaluation will consist of, what me+hods they will use in C4 V the evaluation?
A 9 o
E 10 NR. GOOD: No, we do r.'ot -- we cannot commit right now U
z R ll on the method. Right now we are trying to have our metaluray U 12 department cruantify these defects and try to give the structural z
13 people some information on what effect it might have on their 14 joint inauiry.
o 15 So, until our structural people 'nave that information, 16 they have not committed to doing 't one way. There are several 0
CO 17 wavs they could do it.
tn 18 Okay. If there are no more questions, I will introduce E
c CO 19 our next speaker, who is Mr. Celnik, who is our shield'na C
20 specialist.'1 (Slide) 22 NR. CELNIZ: Good morning. I would like to address 23 this mornina some of the shielding asp cts of the saciificial 24 shield wall.
25 As ment'ned earlier, the sacrificial shield wall, in ALDERSON REPORTlNG COMPANY, INC.
dsp41 addition to its structural functions, has two basic shielding type functions: to shield safety related equipment in the dry well and to protect personnel in case of a shutdown.
The basic sacrificial shield wall design is based ld on a GE concept which basically consists two feet of ordinary ld con rete san&riched'etweer quarter inch steel plates. In lO CIl CI C4 addition to the concrete is a two- inch steel plate for shielding C4 Cl CD purposes.
C4 c5 4 Burns and Roe originally ana'yzed this design concept o
10 using the NRN one dimensional removal-diffusion program to U
z ca calculate neutron fluxes throughout the core, p essure vessel, Q 12 sac wall, and bio. wall.
z 4
13 In addition to calculating neutron fluxes, it I
14 calculated thermal flux distributions to generate capture fQ o 15 gamma ray sources.
16
.hese were then input into the QAD gamma ray point-cd kernal computer code.
I The fission products and gamma contribution 18 was also included. A net radiation profile in the dry well was
~ t4 19 then made up.
20 Now, the basic adequacy, shielding adequacy of the 21 sacrificial shield wall has recently been recorfirmed using 4
22 what is today the standard shielding program for such 4
23 inadequacies, namely, the ANISN one-dimensional d:screte-24 ordinates program, which is a coupled neutron-gamma ray program.
25 This program did confirm the adequacv of the shield and ALDERSON REPORTING COMPANY, INC.
42 in add'ion pointed out that the original HHN calculatiors were somewhat conse vative.
he analytical results were then compared with some,'ot.
exact, but typica'P?R plants that have been benchmarked by both A;7lSX AFS, ZPRX work and some NRC work which is presently goira on, and in addition to that, some operating data for typical BNR power plants; Some of the results, dose rate results in the dry well were compared with tne types of numbers we a=e postulating for the 17PPSS design.
10 The conclusion that is based on all this is that the sacrificial shield design 's irdeed, adeauate to carrv out its 12 primarv safety related shield unctions, and in addit'on, that 13
~
the dose rates that we calculated are comparable to those 14 currently being experienced 'n operating power plants.
15 Okay.
16 (Slide) 17 Now that we are confident in our basic design, there 18 are, however, some shieldina concerns due to some cons:truction 19 type deficiencies. These are basically of two types. Ard 20 will present each separately.
21 he first one has to do w'h the shield gap problem, 22 and as mentioned earlier bv D=. Pialkow, the cause "or the sh'eld 23 caps are due to constructior., the fact that the sacrificial 24 shield wall was placed 'r. two sections, and the shim aap problem 25 then arose at elevation 541.
ALDERSON REPORTlNG COMPANY, INC.
43 dsp43 (Slide)
As to the resolution of the problem, some gaps were 3
visually seen, and due to the fact that this is in the active 4 core region right below core.midplane, which has a potential LO radiation problem, which may negate, in e feet, the adequacy of the shield wall -- due to this problem, percent circumferen-C4 a 100 1CI lA CO tia'nspection was performed all the way around the sacrif'cial CO Cl shield wall.
C4 U
d And the results of this inspection are that indeed there P
o U 10 are 40 shim gaps that were seen, of which 25 extend the full radia z
CO depth of the sacrificial shield wall.
U 12 These gaps vary in size from relatively smhll craps to R
H 13 the largest gap, a cross section are approximately 3/8 of a~
N 14 inch.Rp 2-1/2 inches. Obviously, such a gap in this location is 0 15 unacceptable. 'otally fQ 16 Okay. t1hat are we going to do about it? The CO 17 resolution of this problem and the methodology is one that we 8 18 feel will be confirmed by prototype testing. Ne anticipate 19 constructing channels which will exemplify and simulate the types 20 of gaps that actually have been located and have been detected.
21 These channels will then be constructed. For those 22 gaps that go all the way through to the back, a typical 23 compensatory shield material something like a Chemtree 24 product -- will be poured.
25 Now, i should point out that the, particular shielding ALDERSON REPORTlNG COMPANY, INC.
material to be used, compensatory mater'al, has a sh'elaing 8 sp44 effectiveness which is greater than the shielding e fectiveness of the present sacrificial shield wal'esign.
.he purpose of the. prototype testirg is primarily to lO verify the f'll procedure. ~He will go through this until we are C
C4 lO lO assured that we can indeed fill these shield gaps completely.
CI CV They will be tested and looked at.and verified. through this CO prototype testing program.
Ol Q
6 ~his will give us confidence that at least in this 04g U 10 area the shim gaps that have been located and detected by this z
V7 visual inspection have bee'n completely f'ed with a shielding material greater than present sacrificial sh'eld wa13. design.
zU 12 4
Now, we have a backup insurance program, namely, 14 an in-service radiation scan program which is expected to detect E
o 15 any minor gaps at this "evel. Ue expect to have detectors 16 all around the outside of the sac 'ficial shield I'laced wall at N
g 17 this elevation so that if there are any gaps that may have been missed, they will be located at this po'nt and fixed at that i
C0 19 time.
20 MR. GAMBLE: May I ask a question. "$ hy you said 21 the gaps are unacceptable. Why are the gaps unacceptable?
22 MR. CELNIK: Because they are straight-through holes 23 in the active core region. The shield gap elevation i.s slightly I
24 below core midplane. he peak power, in many cases, is at that 25 particular eleva'tion. Streaming will go all the >ray through, and ALDERSON REPORTI NG COMPAN Y, f NC.
45 dsp45 the dose rates at that particular location could be in excess of the design criteria.
MR. GAMBLE: Are you--
MR. HENDERSON: You speak of some Chemtree product 5 MR. CELNIK: Chemtree,- yes.
6 MR. HENDERSON: In my review of the various submitted 7 materials getting up to speed on this subject, I see that the that you at, least at that time i" was proposed that you V
p=.~u.~ii use this ehrem~~'phonetic) o MR. CELNIK: Chromalic was around t1 e gaps.
U 10 R
MR. HENDERSON: ves. Around in pipe penetrations..
MR. C.LNIK: That is cor ect.
13 MR. HENDERSON: Nell, now, chromalic was beirg used 14 on BNRs in Japan, and it has caugnt fire because it just could o
f4 15 not stand the ambient temperature, it would seem.
16 And General Elect ic abandoned use of that several vears lh 17 ago.
4 18 MR. CELNIK: It 's being used at th's project, to my 19 knowledge.
20 MR. HENDERSON: What'2 21 MR. CZLNIK: It is being used. at this project and at 22 other B77R plants.
23 MR. HENDERSON: Okay. Now, have you examined the MR. CELNIK: Chemtree'?
25 MR. HENDERSON: -- Ch mt ee from the point of thermal ALDERSON REPORTlNG COM PAN Y, INC.
46 dsp46 stability?
t4R. CELNIK: No, I have not, because at. this point in time we have not finalized the particular product or material to be used. But we have looked at. some of the aspects of and lO CS properties of fill type materials. Such products have been 10 used at high temperatures, such as we expect.
CO C4 I have not looked at that specific problem, but I C4 C7 C)
C4 do not expect it to be very'iff'ult.
I am looking at a 40 year design life.
h O
A HENDERSON:
o E
O 10 NR. CELNIK: We will look at all the properties, z
H certainly, of the particular material. We are presently concerned U 12 more with the fill, because I do rot think that this particular z
A 13 type of material would pose a fire hazard. I know it has been used in.
15 HENDERSON: Frankly, I am not thinkina so much 16 about fire hazard as I am long term stability.
17 NR. CELNIK: I do not know I carnot answer that auestion for a fact at this point.
19 Okay.
20 (Slide) 21 The next shielding problem, concexn has arisen about 22 the sacrificial shield wall due to the presence of concrete 23 voids. And I should point out that the basic sacrif-"cial shield 24 wall design is a compartmentalized structure. Ard there are 25 three basic types of compartments. One type is the general r
ALDERSON REPORTlNG COMPANY, INC.
47 dsp47 type of comoartment in ~rhich the concrete was poured from the too, so that there is a verv high deg=ee of confidence that
~ ~
indeed the comoartment was adequately filled.
There are, however, at. this elevation 24 compartments lO above elevat'on 541 in which the fill orocedure was as follows:
Ol ICI a hole was made in the skir. plate; concr te was poured from lQ C)
O4 7 the side; and. irdeed voids were found at th s location .
C4 So there is reason to susoect. that other compartL".ents C4 V
4 like this at this elevation all around here may have s'milar Q
0 E
10 type problems due to the techniaue of pouring the concrete, which z
(0 was from the sides, in addition to which, some time ago, a void 12 was located right there.
13 (Indicating) 14 Now, from a shielding poirt of v-'ew, that o esents ro E"
o problem. The radiation levels at that elevation are of no real 16 concern to us. However, it is indicative, oerhaps, of the fact o$
17 that this tyoe of compartment, although the corcrete was ooured 18 from the top, does contain a significant amount of hardware which w
c Cl 19 may restrict the concrete flow.
20 So our program of fix takes that 'nto accourt. Okay.
21 Nhat we expect to do 's do a l00 percent insoectior. of all 22 compartments, tyoe B. How, these comoartments, iype B, are 23 the compartments that were fil'ed from the side, 24 compart..ents, 24 circumferentially around the sacrif'-'al shield wall at 25 elevatior above elevat'r. 541.
ALDERSON REPORTlNG COMPANY, tNC.
48 dsp48 In addition to that, we expect to do a 100 percent inspection of all type C. These are the types of compartments which were filled from the top; however, they have t1.is baffling which may have created a problem.
i@I Again, we will do a 100 percent irspection cf this t
type in the core egion where there is a signi icant radiation C4 IQ IQ C4 CO C4 level. And 'n addi+ion, we expect. to do some random sampling C4 CO of other compartments just to verify the accuracy of the core.
CV C$
4 Again, the resolution and its methodologv will be confirmed by O
E 10 proto+vpe testing.
Q z
CO Here is a two step process: first of all, we riant, D
z 12 to assure ourselves that our method that we will use can indeed 13 detect the voids. @hat we expect to do is, because we expect 1a the wholes to be at the corners, uppermost corners of these o 15 compartments, we expect to,drill a hole in the upper co ne s of 16 the compartments, borescope them.
N 17 If a void is found, we will determ'ne the extent of 18 the void. Ne'hen expect to remove the skin plate and do a 19 visual examination, and compare the results of the visual 20 examination with the boresccpe examination.
21 This will then give us confidence of our ability to use 22 this procedure in detecting those in the irst place. Given 23 that we have all these, how are we assured that we can fill it 24 adeauately?
25 Nell, tnen we can d ill a hole, bor scope it, locate ALDERSON REPORTING COMPANY, INC.
49 dsp49 the void; if there is a void, we w'l fill the hole, ill the shield material, something like
~ ~
crap, the void with a compensatorv the Chemtree product.
Pe will remove the. skin plate after it has been allowed to set: and verify indeed that we have filled the void adequately.
C4 lCI IA Ne will do this until we have assured ourselves that we can CO C4 detect the holes, and once havirg been detected, the voids, C4 ED adequately fill them.
U ci Again, we have an insurance program wh'ch involves 0
U 10 locatirg and fixing voids with the in-service radiat'on scanner z
M program. This program involves using detector foils at the .
U 12 position where any voids mav have been fourd earlier 'n this z
A 13 program, and also placirg some voids some monitors at random 14 locations near the sensitive equipment.
o This, then, gives us assurance that the basic design GQ 16 function of the sacrificial shield wall to sh'eld the safety 17 related equipment has not been compromised.
18 Thank you.
c CO 19 MR. BISHOP: Mr. Celnik, will the fill of the gap 20 between the shims take place before or after the weld repair?
21 VOICE: I had best answer that. It will be done after 22 we prep it for the weld repai and then the circumferential weld 23 will be made. So it is kind of in between.
24 MR. BISHOP: Hill your prototvpe testing for the shim 25 gap affect the space behind the splice plates, your ability to ALDERSON REPORTlNG COMPANY, INC.
50 dsp50 VOICE: Again, we will be conservative. We will assume there is no splice plate back there. So we 'will insert backing.
MR. CELNIK Backing.
MR. BISHOP: hank you.
"TR. GAMBLE: Hhen you did the shield calculations, did vou look to see what kind of neutron radiation you got on the CO structural steel, the steel wall?
C) re. CELNIK: Are'you worr'd about shutdown purposes?
U 4 MR. GArlBLE: No, was just wondering about radiation.
o U 10 MR. CELN K: No, I did not go into that auestion itself, z
CO ll but I believe it will be looked at.
U z 12 MR. GAMBLE: It will be certainly a lot less than the 4
13 vessel.
SD N
14 MR. CELNIK: That is a problem that is being looked Cg o 15 at now by some of the NRC programs.
fd 16 MR. GAMBLE: Well, what I am ask'ng, at--
o$
17 MR. CELNIK: I have not. looked't it at the I
presen+
4 18 time.
c CI 19 MR. GAMBLE: he result of vour calculations will not 20 show, for example, the neutron--
21 MR. CELNIK: Well, we could. rJe have neutron fluent 22 levels for the sacrificial wall.
23 MR. GAMBLE: You just do not know the neutron level 24 for 25 MR. CELNIK: We .know the neutron -- yes, we kro~r the ALDERSON REPORTING COMPANY, INC.
dsp51 reutron levels as a function of energy. Ne have those numbers, yes.
MR. GAMBLE: What 's it, 10 18
, 10 ? The energv level, roughlv lO
~K. CELV~IZ: I have the figure. Just wait a minute.
C4 lO (Pause)
IO O4 C)
C4 Here is the table of numbers that you might consider.
C4 CD There is a neutron flux lev'el at base locations, at the core, C4 U
4 at the irterface vessel, at the sacrificial shield wall. Given o the neutron fluent level as a function of energy, one could E
U 10 z
M then calculate the response functior, the damage criteria. 5:ese U 12 precise criteria are in doubt at, this point in t'me.
z 13 They cou'd be used in a similar manner for potential.
i E oC4 15 damage to the structure of the sacrificial shield wall.
MR. GRILLE: Whoever designed this," has.: has anybody 16 looked 8t the potential .brittle. fracture for whatever accident loads CO g 17 vou might have for this structure with this kind of neutron 18 radiation? Mavbe somebody who already spoke on the design 19 MR. FOLEY: Do you have a auestion?
20 MR. GAMBLE: Has arybody looked at the potential 21 for brittle fractu e 'n the structure under the accident loadings 22 that are considered, consider'g the reut on radiation level you 23 have in the st ucture?
24 VOICE: Not based on that, no. ~here has been some 25 pressure analysis done on the steel itself. Mr. Burns will ALDERSON REPORTING COMPANY. INC.
52 dsp52 address that later on, but not on the sac wall.
MR. GAMBLE: I did not hear the last part of your answer.
HR. FOLEY: Not on. the,sac wall.
HR. GAMBLE: Let me ask a question: what ~ould happen 5
6 if you did have a brittle fractu e of metal, certa'n parts of CO the metal on this wall du=ing an accident? Hhat would be the C4 C) consequence?
C4 c5 A DR. FIALKOÃ: I would like to know the bas' upon wh'ch o
10 you are .postulating brittle fractu e.
U z
GA!4IBLE: I do rot know; I guess you are usinz, some U 12 sort of A36, I guess it is. Ye have a problem identified to us z
in PKRs on similar shields that support the vessel. This does 13 14 not support the vessel, but there some people have done some 15 calculations and found. out that ti ey have cuite an HDT shift 16 because of the neutron radiation you get around structures right 17 outside the eactor vessel, on the order of 120 or 140 degrees.-
18 So the ND of these mate ials can increase signi icantlv 19 and the total .4D after radiation could be up to 220 degrees 20 fahrenheit.
21 That is prettv high. So that was that problem was 22 indicated to us, I think, by Stone and webster on it was -- I 23 think it was a B G N plant.
24 DR. FIALKON: Pell we have -- the relative nature of 25 tne load on that, especially the dynamic load"-, we have included ALDERSON REPORTING COMPANY, INC.
53 dsp53 dynamic factors, and we have 'ncluded uncertainty factors when characteristics such as b'owdown are not known to be precise.
So, the load magnitudes have been taken account of.
Then we went on from there and worked with equivalent static loads based on those factors, and the material itself considered to be A36 steel w'th the usual structural C4 LO was lO C)
O4 properties associated with it.
C4 ED Ne did not postul'ate anv brittle conditions.
V A MR. CAPABLE: That is what I am aettincr at. Shou'd
~ e o you have? Or should you now make that a consideration? You C 10 z
Cll apparently did not consider that this mater'al would have a, U large NDT shif z
A 13 DR. FIALi(077: I cannot really speak to that.
14 MR. GAMBLE: Did vou use a material and assume that o NDT would be something like zero or 20 degrees, and if you did, 16 is that necessary for the integritv of vour design? I do not f6 know. I am asking.
18 DR. FIALft'097: You are raising the quest'on of the E
CI 19 operatinc temperature of the wall, in affect'? ln the wall, as CI 20 affecting the possibilitv of brittle fracture?
21 NR. BURNS: As I understand it, he is ask'ng you, did you do a racture safety design on the sac shield wa'l, and as I understand it, the answer is no.
24 FOLEY: The answer is no.
25 NR. GA?ABLE: And maybe you did not do 't because you ALDERSON REPORTING COMPANY, INC.
54 dsp54 assumed you were picking a material that had and would always have a low value of ND . That is certainly an acceptable way to do it.
Ny cruestion is: should you or did you or should you consider the fact that the NDT may signIficantly increase.
LO NR. BUPZJS: Let me answer the question. - I cannot anwer for the assumptions made in the design of the structure.
LQ'4 o All I can tell, you is what we are doing today.
O4 C)
We are wanting to evaluate the significance / of C4 0
A anv deficienc'es that we find. in the sacrificial wall'tructure o
E 10 and their effect on the performance of this structure;that may U
z lO or may not include a fracture safety design evaluation.
Q 12 We have not addressed to date the effects of the z
A 13 radiation on the material properties of the sac wall structure.
If we do do a fracture safety evaluation of the sac wall struc-15 ture, perhaps this is one of the considerations we will have to 16 look at.
17 I think you are asking you 'are raising, as I under-18 stand it, a fairly recent concern in the context of a structure 19 that was designed some time ago.
20 NR. GAMBLE:
1 That is why I asked that if the structure 21 failed, what does that, mean to you.
22 lIR. BURNS: I think this is -- you know I can 23 answer that.
24 NR. GA."ISLE: I mean, it. may not mean anytning to you in which case you do rot care about it. That 's one possibility.
ALDERSON REPORTING COMPANY, INC.
55 dsp55 On the PWR it was obviously 'mportant because it supported the
..:.~.sp55 reactor vessel; in this case, that is not true. That is 3
what I am asking: what is the consequence of having 'this structu e fail?
1@I NR. BURNS: Let me ask you a quest'on: in the recent lO NUPZG that was put out bv NRR on the fracture safetv evaluation C4 for pressurized water reactor component supports, you categorized 04 the materials into three ca'tegories: categorv one, where vou O
d assume that there was no structural problem; category two, o where you have already evaluated; ai:d category th ee, where U
z CO further information is required.
Q R
12 Zn look'ng through the document, l do not bee a A
13 discussion of the effects of irradiation on ND ~
propert'es. You i 0
~ are categorizing the materials as I understand it from the in terms of three categories, document, on the basis of as-A pl 16 received NDT distributions.
N 17 And, you provided in the report the distributions of 18 NDT and recommended assumptions for the maximum NDT temperature.
CO And I did not see in that document. a discussion of the effects of CI 20 long term irradiation on, these assumed NDT values.
21 ?4R. GANBLK: Well, that is right.
22 i~JR. BURNS: Are vou planning, 'n other words, to 23 require that P~r?R operators assess long term irradiation on the 24 integrity of the operating P'."?R pressure vessel?
25 NR. GAMBLE: The answer to that is no; that this ALDERSON REPORTING COMPANY, INC.
dsp56 report that you are refer ing to deals with component supports, and it was felt that it would be no significant ir'radiation damage to those steels used in component supports.
NR. BURNS: So you lO NR. GAMBLE: The problem that I referred to before is C4 lO the shield that surrounds the reactor vessel on P77Rs and supports lO C4 CO C4 the vessel and it is only in certain designs so it. has 1
C) a high, a much higher level of neutron fluents than component O
4 supports tha" were addressed bv the report you mentioned.
o 10
- his design is very similar to the vessel support U
z V7 type because it does get a lot of apparently a lot of neutron U 12 fluents.
z a ~
here is a distinction between this report on PWR 13 fh component supports and the neutron problem that you might have 9&'
15 in this plant and we see on some on a particular design in
) 16 PNRs. It is a different question. In ore case you do not have any neutron problem; in the other case you do.,
CO 17 18 MR. BURNS: There are, as I understand 't, some 19 support designs that would place regions of support in the areas 20 of high neutron flux.
21 NR. GAMBLE: That is possible.
22 HR.EARLE: If I may interject, my name is Keenan Earle.
23 I an a licensing engineer. Ard to answe- your question, we It is I
24 cannot accept any brittle fra'cture in the sac sh'eld. an 25 impo"tart structure. You have ra'sed the concern that we p obably ALDERSON REPORTING COMPANY, INC.
57 dsp'57 should evaluate it, but as fa as I know, 't has not beer.
previouslv evaluated fo any plant that 1 know of.
PER. GM!BLE:' think that '" riaht; - we had. never looked at it before on this particular design.
NR. EARLE: So, i think it is a
."1R. CACKLE: it is a structure that you cannot 7 cannot to lerate fi1 1 under the accident condition j ust by the locat'on. And i do not know the answer. i, mean you know 1 do not know what the answer is, but -- but apparently othe>>.
10 similar-designs of PNRs do have sign'ficant shifts creater than 100 degrees, i would think you would want to look at 't.
NR. EARLE: Ne will address the concern.
13 NR. GAi'ABLE: T?ell, i guess your plan now is to have a submittal evaluating.
15 HR. EARLE: He did rot currently have a for...al plan of submitting a report to NHR. it was my understand'ng that you were going to caucus and perhaps discuss what action HRR was aoina to formally take.
19 his report that we have, it does not really cover all 20 the issues, and i do not know whether it covers the point NRR 21 is concerned with in detail.
22 NR. T:.ENDERSO~I: it goes outside .the original inter.ded 23 scope of this meeting, Ron, and, i do not think we ought to 24 confuse the issue bv ask'ng a submiss'on on these const=uction 25 aualitv problems, which ra..ge of 'nto the potential for neutron ALDERSON REPORTlNG COMPANY, INC.
58 dsp58 damage.
Zf you think that is of sufficient'current interest to you, T. think you might recuest that DPM address it; perhaps not only to these people, but to other people too.
lO 5 MR. GA&IBLE: All right, fine.
C4 6
"4R. HENDERSO'N: Sir( for the reporter's benefit, what C)
C4 7 's your name?
F4 CO o 8 MR. BUEQIS: Dave Burns. I am a material welding C4 U
A 9 eng'neer witn i'.PPSS eng'nee ing.
o MR. HENDERSON: Thank you.
U 10 z
~ e4 M ll I think it is about time for a break.
U 12 (Brief recess)
R H
13 NR. FOLEY: Ne will start out right where we left. off--
CO 14 MR. HENDERSON:. F ir.e.
o 15 iM. FOLEY: with some record concerns discussed by
Ãe are still on the sac wall.
i ~
16 Roger Johnson.
Ne are still on sac wall?
M 17 MR. HENDERSON:
6 18 FOLEY: Roger Johnson is the equality assurance c
CI 19 manager for the project.
20 le. HENDERSON: Yes.
21 NR. JOHNSON: Nv part of the presentation addresses 22 four, five, six, fifteen, sixteen, and seventeen 'n the items of 23 concern in the NRC memo. We will take these beginning with 24 item four.
25 Ne have the item stated, that NDE records associated ALDERSON REPORTlNG COMPANY, INC.
59 dsp59 with the sacr'ficial wall contained photocopied'nspector's ultrasonic and penetrant test acceptance signatures.
Xtem nine has been identified on .pipe whip 'restraints.
Our review indicates that the specifica ions required ultrasonic testing for checking electroslag weld and tee joints, imited i@I UT testing of base metal thicknesses of 1-1/4 'nch or greater.
C> here was no specification requirement for performing PT testing.
C4 And checking this out with Leckenby, we determined V
that P. was used bv Leckenby for informational pu poses for r
o casing out defects that were identified bv the U process.
U 10 A R
Q) The AhS code requires verification that. once a defect, U 12 has been identified, vou have to ve ifv bv nondestructive R
A examination methods that all defects have indeed been coxrected.
14 Leckenby has stated to us thev used ultrasonic examination to O 15 perform this.
col 16 Xn other words, they performed a primary ultrasonic 17 examination ard c sed out anv defects using PT. And their 18 confirmatory NDE examination was a second UT examination. There c
19 were a total of 27 ultrasonic reports 'ssued by Leckenby. Ne 20 have. identified that 15 of these reports contain photocopied 21 signatures.
22 Those 15 reports a feet a total of 87 welds. Xn other 23 words, the report contained more than one weld on each report.
24 As fa- as the PT reports, there were 30 that Leckenby had indi-25 cated that had been done, ard we have fou- of these in-house at ALDERSON REPORTING COMPANY, INC.
I'
50 dsp60 the present, time, and we have ver'fied that those four, that indeed, the second UT examination was done in all cases after completion of the weld epair.
As far as actions,.we have a sworn statement by Leckenby
'LO inspectors. It was taken by NRC Reg'on V. It indicated that photocopv was used to expedite report processing, and the lQ lO C4 CO C4 NRC is continuing their invest'gation with Leckenby in this CO matter. And we are obtaining this information and we w'll follow 0
d up anv new developments or concerns and factor that into our R
o U
final analysis.
R Some ind'cation that the ult asonic examinat"'on was 12 performed as eauired was the .fact that we nave identified four 13 cases utilizing the PT repo ts that welds were in fact rejected, 14 and we have the repair reports.
E o 15 So we are confident. that the UT was performed as Cil ~
for
~
16 required. We are also confident that PT was performed informational purposes only, as stated by Leckenby.
4 18 We do intend to continue our investigations to determine c
CO 19 that all of the PT reports that 'dentify weld repairs were being CI 20 exam'ned by ult asonic examination after the epairs were made.
21 (Slide) 22 Item five, the conc rn being tnat nondestructive 23 examination qualification records cannot be located for one individual who perfo med ultrasonic test'ng on the sac 'ficial shield wall.'is 'ndividual is no longer employed. by the ALDERSON REPORTING COMPANY, INC.
asp61 subcontractor.
&le performed a complete document r'eview of all sac shield wall NDE records; we have identified that all of the ult asonic examinations vere performed by one Mr. G. Hamilton.
Oualification records for Mr. Hamilton are available, and they C4 lO are under review by the project quality assurance at the present lg C4 time for acceptability.
C4 C) In addit'on to that, we have identified that -a C4 Q
A l1r. Cha les Baldinaer performed ultrason'c examinations on three o
10 weld procedure qualificat'on coupons. Mr. Baldinger's certifica-U z
CO tion papers are not available. Thev apparently have been lost.
U However, ultrasonic testing for weld'rocedure qual"fications z
A 13 is not reauired by the code or by specification.
In reviewing the specific procedures, we note that E
Q W
the reauired tensile and bend, tests were pe formed by W
16 Leckenby and that these procedures are acc ptable. Ne will 17 continue our investigation to assure that i'>>. Hamilton's I 18 certifications are totally adeauate and we vill cont'nue our i
CO 19 investigation on r'ecord reports to assure that any additional 20 concerns are ident'fied.
21 MR. HENDERSON: There is no indication that Mr. Baldin-C 22 ger did arv UT examinations?
23 MR. JOHNSON: Nct on that, tha" is correct.
24 (Slide) 25 Item six and the next item, item fifteen, general y ALDERSON REPORTING COMPANY, INC.
dso62 address the same subject. But item six is that no procedures 2 were generated or records maintaired or. forming of the curved 3 olates used in ti e sac wall.
I might note here that the item of concern here also lO includes an allegation regardincr d'scolo ation of the inside CS lA surface of the curved plate. I will address this further in C4 CI item 15 which. is the next item coming up. he reauirement-C4 C) there is no recui ement in 'the soec'ication for procedure C4 U
A excuse me.
o U 10
'here is no equirement ir the specifications for z
Ch a procedu e or record on forming of the curved plates. In U
R 12 discussing this with Leckenby, it turns out they have sub-H 13 contracted this work to Seattle Bo'ler Norks, and it is common 14 industry practice to heat and/or cold form A36 material.
o 15 Checking this out with our AE metalurg't, he confirms 16 this is indeed an industry accepted practice w'th basically cd 17 no detrimental effects.
GQ 4 18 The Supply System's oosition at this point for this E
c CI 19 item as well as the next item is tl.at the formirg of curved 20 olate in the herding process is not: a soecial p ocess as defined 21 in 10 CPR 50, Apoendix B and at this time plars no additional 22 action on this item.
23 (Slide) 24 In item 15, interviews with Leckenby oersonnel established that sac shield wall segments, 2A, 3A, 3B, and 3C were heat straigl't.ened w'thout the benefit of contrclling ALDERSON REPORTING COMPANY, INC.
63 dsp83 procedures or the maintenance of auality -ecords. Heat straightening, the application of heat or mechanical,was applied to correct weld distortion. As indicated previously, the=e. is no reauirement in the specs for procedures or records on these types of industry practice.
Discussion with Leckenby has indicated that this is industry p actice and basically not detrimental. be have some C4 O
O papers here that discuss the application of heat for berding c$
d large segments, heat straightening them, which is 'dent'fied o as an industry practice.
10 z
Therefore, we are planning no additional act'on at.
U z 12
'his time.
H 13 VOICE: Even though it is not in your spec that they 14 have to have a,procedure for these things, but aren't these 0 things brought to someone's attent'on because they go on some fQ nonconformance report and then the nonconformance repo t goes cd 17 through the cycle and it gets around to someboay and they sav fa 18 they have a corrective action, what they are going to do on it'P 19 Then somebody has to review it. So in a way a procedure is 20 going to be generated, in a sense, say'ng that these are the 21 steps that we are going to take to straighten th's or howeve 22 they are going to do it.
23 So sornebodv will be looking at it. Nill something 24 like this be generated in this case?
25 !TR. JOHNSON: For the part'cular sections we are talking
.AI DERSON REPORTING COMPANY, INC.
dsp64 about here, Leckenby did utilize an in-hoarse document which showed how they applied the heat or how the joints vere welded.
This was not. submitted to us for our review and app oval. They had a basic drawing which established the design regui~ement.
lQ They straightened things as necessary to meet that, specific 40 requirement on the d awings.
C)
C4 VOICE: Are you aware that this even goes on'? Is this CI more or less part of their industrial practice to say, okay, CV U
A this is a minor they call it mino or, whatever. Thev just o go ahead and do this?..
U 10 z
CO NR. JOHNSON: As far as heat st aightening of the .
Q 12 large sections, that is right.
z A
13 VOICE:, You are not even aware it is going on then?
FQ M
14 NR. FOLEY: That. is correct. That is correct. This is E
O 15 an in-process thing for them. I th'nk this became an item of g
concern because of an allegation with respect to some discolora-N tion on plate, and that was pursued back into the fabrication 18 process. And some additional information was developed there.
E CO 19 But it did not. come to us by the standard practice of an 20 inspection report or nonconformance report.
21 VOICE: I was just wondering whether that was part of I 22 the normal, cycle. Anyway, they had a document. In a sense, it 23 was a normal procedure.
It appears that this 's
'4
~LB. HENDERSON: to me a QA 25 problem, and normal commercial 'ndustrv practice 's near the ALDERSO'N REPORTING COMPANY, INC.
65 dsp65 bottom of the scale of auality.
I do not think0 it is consistent with the reauirements o f Appendix B or the recuirement that Append' programmatic requirements be extended, to contractors and subcontractors. So, lO now, obv'ously, Leckenby is providing a self-serving answer, that IO that is just a normal practice.
IA C) I do not share vour conf'dence that that closes the CI issue.
C4 V
4 MR. JOHNSON: Our specification did not reauire them U
o 10 to submit their application of heat for straightening beams z
R or plates.
U z 12 MR. HENDERSON: It should have; 't certainly should H
13 have.
14 MR. JOHNSON: Ple did not feel it was a special process o 15 as defined in the industry.
16 MR. HENDERSON: It is not a special process. Activities u$
17 affecting quality shall be done in accordance with quality 18 control procedures; that is the aeneral requirement.
c CO 19 MR. JOHNSON: They have an in-house procedure which 20 they utilized to do it. Ne have not reauested that that be 21 submitted to us.
22 MR. BISHOP: Thev do have an in-house procedure 23 that defines the temperature limits for heat straiahtenincP 24 MR. JOHNSON: I have not seen the procedure mvself.
25 i1R. BISHOP: You axe not familiar with that.
. ALDERSDN REPORTlNG COMPANY, INC.
66 dsp66 VOlCE: Even though it, 's '..dustry practice, aren' you recruired by Appendix B to review exactly how they do the e practice? And is there not a situation whereby theoretically you could find standard industry practice not acceptable or this application?
HR. JOHNSON: Zf we felt that a particular process was critical enough to the application that we would define 8 it as a special process, then the answer would be yes, we 4 would regui e it. be subm'tted to us for review.
0 U 10 VO CE: But.the auestion is: until you review these z
processes, how can you come to that conclusion?
12 NR. JOHNSON: These processes were known to the A
13 architect engineers who generated the specifications at the 14 time.
o VOICE: Then the architect engineers must have reviewed QQ i ~
16 this and reached a conclusion based on some evaluation that 17 indeed these standard industrv practices are acceptable; is 18 that correct?
E c
CO 19 NR. JOHNSON: No, sir. As I indicated, our specifica-20 tion, as issued, did not reauire that these particular processes 21 VOlCE: l am trying to draw the case that your 22 specif'cation and Mr. Henderson was try'ng to draw the same 23 th'ng -- your specification, as issued, the possibility exists 24 i" could be deficient.
25 b'e are trying to just probe that very br'efly.
,ALDERSON REPORTING COMPANY, INC.
67 dsp67 '4R. JOHNSON: Let me answer that by saying this: we are, due to the pa ticular allegat'on in the two items of concern, we are reviewing the procedures at this ooint, after the fact. ':lhen the specificat'on was o iginally written, there LO were certain industry pract'ces oerformed by a number of Fs C4 cortractors. Ne do not review everyth'..a they do.
lO Ol CO And tnis is basically one of those cases.
C4 CI VOICE: If vou do rot review what thei oract'ces are, 0 how do you know that they are not completely unacceptable A
oF oractices?
Q 10 z
Ql
~IR. JOHNSON: Ne perform audits ir.-l.ouse to determ'ne U
z 12
'f they have control procedures.
Cl 13
- -0 EZ': :he architect ena-'neer preoared the soeci-N hl 14 fication without'he knowledge of what was to be constructed oLe and. the kind of organization or fabricator that was going to do the work. Nor hasi the architect er aineer indicated that if PQ
~
16 Cef
)~ 17 he were to do it .over, in this instanc , that he would do it 6 18 any differently.
E 19 VOICE: Is +his the way you would do it as of now, 20 knowirg any problems that you might have?
21 pOLE>: ~iy oo't is that oe haps if"we were doing iti 22 aaa-'n now and coina i" with Leckenby, we might do some things 23 differentlv. l e have not had any evidence that has been 24 I
presented to us, nor have we urccvered any evidence, which 25 suggests in the formina of these plates that Leckenby followed
. ALDERSO'N REPORTING COMPANY, INC.
68 dsp68 inappropriate prac+ices that would lead. to an unacceptable product.
Ne have rev'ewed the specif'cation, and in 'that light we would not change the specification.
lO VOICE: In your review, did that include a review Ol LQ of Leckenby's "standard industry practice," quote, unquote.
l4 C4 C)
C4 NR. FOLEY: I th'nk we reviewed it in the context of C7 wnat Leckenby did in this 'nstance.
CI O
6 NR. HE;ANDERSON: think we have pursued this far o
10 enough because we do understand your position, and that is what U
'z CQ we are trying to get.
U 12 (Slide) z 4
13 bB.. JOHNSON: All right. Proceeding to item 16, the 14 specific item of concern is that the 23.5 and I should explain E
o that this 215 relates to one of our prime contractors at the 16 site the i ~
quality review of the Leckenby program and CO 17 Leckenby was a subcontractor to 215 did not include verifica-tion t?at all required ultrasonic examinations were performed as 19 required by specification and Leckenby procedure.: the concern CI 20 here being, basically, that the UT exams were performed, as 21 required.
22 l'le have gone back and reviewed the specificat'on, and 23 it required a sample UT program as follows: basically, at this 24 point, for every 16 to 25 welds, we examined six. If no 25 indication of lamellar tearing is found after 24 UT evams, the UT exams can then be reduced to two per 100 welds.
ALDERSON REPORTING COMPANY, INC.
I
~
dsp69 Leckenby made a total of 1270 electroslag welds, and of these, a sample of 48 welds.were to be exam'ned by UT.
Leckenby has performed b
ultrasonic tests on 83 welds. Those test results are contained in the 27 ultrason'c reports that 10 5 we discussed back in item four. his is in excess of the LO 6 spec requirement of 46. ln addition to that, Leckenby lA examined an additional 2 percent of other tvpes of welds', so that C4 CO 7
C4 for the total 27 U reports that we have, they represent 200 O
4 weld areas that we e examined by UT by Leckenby in addition to o
Q 10 electroslag.
z CO
~He will continue our investigations to assure that 0
z these U. examirations were performed bv certified UT examiners.
Q 13 VOlCE: You examined this specific possibility of lamella" tearing?
15 NR. JOHNSON: Yes.
VOZCZ: And tee joints? Are they also taking 'nto 17 consideration that even though there is no indication for lamellar tear, there is always the possibility that vou could 19 have delayed cracking or maybe 20 HR. JOHNSON: Any type 21 VOICE: due to additional loads at a later ti...e.
22 NR. JOHNSON: That, ~~ correct.
23 t
Xtem 17 is a two part item of concern, the 'first part 24 dealing with procedure deficiencies, indicating Leckenby used 25 liquid penetrart testing to examine sac wail structures at ALDERSON REPORTING COMPANY, INC.
70 dsp70 Leckenby shops. Leckenby representatives =eported that there was no approved LP procedure at the time inspect'ons were performed.
I would like to address th's one f'st: as we previously pointed out, LP examinations. were not required in the specs and we e basicallv performed by Leckenby for informational purposes only.
Cl CO C4 Leckenby does nave an approved licuid penetrant C4 C) procedure; it was approved'n the cuality control program for U
4 Leckenby, which was by the 215 cont=acto as part of their QA o
U 10 review.. Only three of. the licuid penetrant reports were z
M submitted to the Supply System for record purposes, since they U 12 were informational.
z A
13 And we have found that at. least four were subseauently 14 examined and. accepted by UT.
0 15 Addressing the second item, the Lec!cenby procedure 16 whichr provides for weld secuence control, entitled "Sacrificial Ql 17 Shield Nail Assembly Procedure," has no procedure number, 18 number irevision number, no date, no evidenc of ever 'beinc
~ tt 19 approved. ANS Dl.l paragraph 3.4.3 recuired submittal to the 20 engineer of weld seauence and distortion control programs.
21 Leckenby has. submitted for review and approval a fabrication 22 and erection procedure that shows the general secuence of 23 fabrication; however, it d'd not show the detail of ireld 24 secuencing or distortion control.
25 ',?e have cone back researching th' item o concern, and ALDERSON REPORTlNG COMPANY, INC.
71 dso71 we have obtained from Leckenby a commercial document, which does depict the sequence of the weld sequenc'ing of restraints, and this document was'made available to the Supply System on Januarv 29, and is presently. under review by our enginee ing lO 5 organization.
Ol lA 6 MR. HENDERSON: eO.at do vou mean bv "comme c al C) 7 document" ?
C4 MR. JOHNSON: It is what they have termed as a C
U 6 guideline. It does not have a procedure neer. It 'simply o
10 shows the sequence by which they welded the beams togethe , the Q
z R sequence of weld.
U 12 MR. HENDERSON: Okay. -Thank you.
z O
13 MR. BISHOP: Is that soecific to the sac ificial 14 shield wall?
O 15 MR. JOHNSON: Yes.
hl 16 Corrective actions: we will continue our investigation rA 4 17 to determine that all repaired areas were re-examined by UT.
18 We have indicated that we will comolete the weld sequencing 19 review 'by engineering.
20 Are there anv questions?
21 MR. BISHOP: The UT examinations that were oerformed 22 were performed basically by one individual. There was a second 23 individual. You only mentioned one.
' 24 25 a The records do indicate.
Mr. Howenstein, who also performed a second a
'ndiv'dual, review.
a ALDERSON REPORTING COMPANY, INC.
72 ds972 liR. JOHNSON: be have no ev'dence from the existinc U inspection reports in-house that he had performed those. Tie 3 have your statement. '7e have identif'ed the problem, and we are presently reviewing that to determ'ne just what he did do.
BISHOP: But, in any event, the crualifications o r
LA the individuals who did the licruid penetrant testing were not IA C4 C4 addressed here. Do you know if vou have reviewed those or intend C4 C) to review those?
IN V
4 NR. JOHNSON: Ne intend to rev'ew those.
o 10 AR. BISHOP: Secondlv, you stated that and correct U
z M me if I am wrong that in the 25 cases where they did licruid U penetrant testing that was not re-perfor .ed by U:, y'ou -re 12 z
A 13 going +o check that aspect.
NR. JOHNSON: That is cor ect. hat is part. of the oCe 15 continuing investigation on the I
U process. Ne w'll ve ify that
ÃAr 16 in all cases where r~
UT was pe formed, that the confi ming UT r
examination was also performed, which would then substantiate cd 4 17 18 their statement that it was used for 'nformation purposes.
a ."LR,. BISHOP: One othe sta+ement that you made was in C) 20 reference to continuing your investigat'on of. the UT examiner's 21 certi.icat'on.: %hat is that continued investigation going to 22 consist of? As I understand 't right now, he had taken a written 23 test, but he had not taken the practical auaiificatiors test.
NR. JOHNSON: Nith discussions with you, we had con-25 firmed that he had taken the general and the specific; at this r
ALDERSON REPORTING COMPANY, INC.
73 dsp7 3 point we have also confirmed that he d'd take the practical.
There is some problem with his eye examination, so I cannot make a fir.. statement that'e is totallv qualified.
In addition to that, we also have some problem with the procedural.
NR. BISHOP: Thank you.
NR. JOHNSON: At this tme, I would like to introduce CI Hr. Jack O'Donnell.
C4 c5 4 NR. O'DON¹LL: I would like to address, basically, Q
10 three of the the last three items on the sac wall of concern U
z M
items number 18, 19, and 20, starting with number 18.
U 12 (Slide) z A
13 This concern is quoted from a letter from a consultant 14 which Leckenby had hired. The original letter was sent out when E
oCe we discovered a crack in the sac wall. The consultant proposed td A
16 the nature of the crack. Ne, in turn, disputed it, and it seems a second letter was brought to light during this ongoing invest'i-18 gation.
19 (Slide) 20 The crack occurs at one of the main intersections of 21 the radial beams to the sac wall. There are approximately 24 of 22 them, and we have in turn reviewed the consultant's letter. he 23 discussion in the letter revolves around why the cracI: happened.
24 And it is still being disputed as .far as contractual conditions between owner and contractor.
AI DERSON REPORTING COMPANY, INC.
74 dsp74 The "whv" has been technically put in the background, since we subseauently removed the two or thr'ee bean'.s that w re currently in place, repaired the wall completed, UT-ed it; using a different weld orocedure, we olaced those beams and remaining lO 5 beams around the wall with complete success. So in cor.tention, IA IO 6 obviously, was the welding secuence, ard I guess we could have a C4 C4 7 discussion about why the weld fa'led ini ially. But we do not C4 CO 8 consider this any longer a 'technical problem.
Cl U
A Anv auestions on that?
R o (No resoonse)
Q 10 z
N (Slide)
U 12 Concern number 19: in some of the past visits with the z
13 NRC, the existence of the possibility of f ee water I .in;agine 14 Region V was under the imoression it was one isolated case. A O 15 review of the IRs during this review of Leckenby's work uncovered
à ~
aooroximatelv seven IRs involv'ng
~
I have to sav again -- "free (0
17 water" quote, uncruote.
18 In these cases, reading tne IRs after the fact, I read 19 into it there was some rust in some areas, and so forth. Now, 20 the only source of free water possibly behind the wall would be 21 concrete because there is no other water that is oresent.
22 (Slide) 23 Ne reviewed the seve locations, and they are all skir.
24 olates in the a ea of 541; in other words, skin olates that were 25 placed on either prior to or after cor crete was placed. In the ALDERSON REPORTING COMPANY, INC.
75 dsa75 the case of these six, we had six of the seven and have records to indicate that the welds that were in fact'ade prior to the concrete being placed'behind it; this would recate a aossibility 4 of free water existir.g when the weld was made.
47 he six welds were made after the alacement of conc ete C4 IA LO in the area discussed by Yw. Celnik and, Dr. Pialkow. l: vill C4 CI point to the a-ea in question. vou have seen this slide before.
C4 CO CI (Indicating)
=his windows< was used by the contractor for corvenience o
10 of a'acing concrete. After the conc=etc vas placed, at some U
z CO 11 aeriod of time later, he fitted up a piece of steel plate, U
z 12 preaared it, ar.d welded it. This subsequent crack was found in A
13 the e.
14 This brought us to the possibility that this may in Q 15 fact have been done too soon prio to the alacement of concrete Pg i
and possibly free water existed in th's one instance.
~
17 So what ve did, we reviewed all the window alates that 18 were susceptible to this tyae of construction; went back, t
CO 19 investigated, reaaired them if necessary for anv c acks. tie 20 are still of the opinion tnat there was no free water existing 21 when we made the weld. Z do not th'nk it would have been possible to have made the weld.
23 So water has seeaed through defic'encies in the weld, 24 in the weld o the skin plates after the skin alates were ir.
place.
ALDERSON REPORTING COMPANY, INC.
dsp76 Any questions on this so far?.
~4R. HENDERSON: Yes. What is the environment in wh:ch these plateswere stored berore thev were finally put "nto their ultimate location? Here they stored in the weather outside?
lO HR. O'DONNELL: No. There was a build'ng adjacent to the contractor, and this C
the reactor building set up by was where EQ lO O4 he prefabricated. the three 120 degree rings.
C) Now, as far as su'bsequent storage" on a month to month ID 8 fairly soon someone that is familiar C4 U
4 basis, I think it was o
10 with the basic storace Nerle Parisie I guess he has stepped U
z Q7 out of the room.'e U 12 prefabrication was done in a closed building until z
4 13 the time it was lifted in place. They took off the roof of the Ql building and then lifted the three l20 degree segments'. into P+
14 E
o 15 place.
16 Any other questions?
th 17 NR. BISHOP: On your last slide you indicated that 18 repairs were performed to the window plates. Nhen you say Q
C CO 19 window plates, you are talking about the small cover plates?
20 NR. O'DONNELL: The small cover plates, yes.
21 (Slide) 22 This is the one case, this one weld that had to be 23 made after concrete was placed in the back of. the skin plate. So 24 our conclusion was possibly they welded too soon after the place-25 ment of the concrete, and with the heating, and what have you, we ALDERSON REPORT1NG COMPANY, INC. 1
77 ds277 may have had a def iciency in the weld.
However, with our preheating orocess, I tnink 8r. Akers will ver'v, we would 'have dr'ed out any free water and el'minated it prior to welding. I do rot think it vas oossible +o make a ld 5 weld of ary consequerce with free water present. Is that correct, lO 6 Ler.?
lA C>
7 :iR. Ai(ERS: Invest"gation shows us there was concrete in back of it. ve feel that the preheat te.;.perature that was O
a involved should have d ied it out. It was not the tyoe of velding o
10 vou wou"d fird if someore had tr'd to weld throug'h vater. Ne U
R CO assume that the moisture was dried out, ard. as they welded, it Q
z 12 in, the heat, after that weld vas being built up, oossibly 0
13 drew water or later or. that moisture occurred because the veld.
CO d'd not have the indication of welding underwater o" in water.
15 looked like it vas just poo- workmanship.
W 16 AR. O'DOHNZLL: Any furthe- questions on that subject?
rd 17 (Ão response)
I would like to go into concern neer 20.
Ce i
CO (Slide) 20 This was brought up by Reg'on V, I imagine, or. the 21 Burr s and Roe drawing S-802, 'Note Number Three, which refers 22 to the spec' ication for the oost-wel'd heat treatment equirements 23 of the stabilize truss. Let me poirt out that at th's time, 24 although the cuestion 'nvolved A5l4, A588, and SA537, the 25 stabilizer truss 's i.. fact A514.
ALDERSON REPORTlNG COMPANY, INC.
'sp78 The 588 is the material that Dr. Fialkow referred to, which is the top of the sac wall. The SA537'is the attachment to the containment vessel. his was built to section th ee, and for those of you that are unawa e of the stabilizer truss, lO spans from the sacrific'al wall to the containment vessel.
4 IO The question was: was post-weld heat t=eatment C4 CO C4 required on the A514 stabilizer truss?
C%
CI C
C4 And, as I said, the drawing refers to the specifications 0
ci 9 and the specifications, the requirements for post-weld heat o
U 10 t eatment is found in section 17D of the specification. And z
(0 that specification, in essence, adopts and supplements Dl.l .where U
z 12 necessary, A$ ?S Dl.l.
13 A review of ANS Dl.l will lead you to the d'sposition that once you have tempered steel it is not recommended for o 15 post-weld heat treatment unless absolutely necessary. For the
/
16 disposition of the other two materials, X have put up here the 40
- 17. vessel attachment; this was SA537, which was 'n fact post.-weld foal 18 heat treated, as requi ed by the code 19 A588, which is the high, strength steel on top of the 1
20 sac wall, ANS Dl;1 is silent on; tne contractor certified his 21 procedure without post-weld heat. treatment and it was accepted on 22 that basis.
23 Any questions?
24 AR. HENDERSON: Have you done any volumetric examination; 25 of those areas?
ALDERSON REPORTING COMPANY, INC.
79 dsp79 O'DONNELL: Excuse me. In these areas?
(Indicating)
MR. HENDERSON: Yes.
MR. O'DONNELL: I would have to say that the exact stabilize truss, I do not have that in the II 5
.examination on the 6 specifications to answer that question. Possiblv when you C)
C4 were reviewing. the'tabilizer truss situation, do vou know if C4 CI CI they did let me see. 215, I do not know. whether it was 0
A mandated that there was -- I would have to look at the inspection.
o Q 10 I would have to z
11 MR. JOHNSON: I would have to assume.
) ticular attention to U 12 MR. HENDERSON: A514. calls for pa z
H 13 the sensitivity of materials for welding. I am wondering just 14 how that procedure for welding A514 was qualified. It assumes 0
o 15 that the qualified procedure will be used.
16 i~1R. O'DONNELL: Absolutely, yep, there is an approved point out that this is not -- for those of cd 17 procedure. Let me 4 18 you who are concerned about distortion here or residual st esses, CO 19 the stabilizer truss is pinned to the vessel; it is a physical C) 20 pin. It is welded to the sac wall, but i is a pin, and it is a trusswith pins a'1 around the circumference.
22 So for distortion control or for, let's say, residual 23 st esses, this..was not our main concern.
24 Any- other questions?
25 I think there has been a change in the agenda. Roger, ALDERSON REPORTING COMPANY, INC.
0
80 dsp80 would you take over.
HR. FOLEY: This concludes the presentation abcut
'I the that is included in your repo t on the sacrificial sh'eld wall.
In the process of our own incui ies on the sacrificial shield LO wall and in inquiries on the pipe whip restraints, we have 7J ld identified these addit'onal concerns'which are not addressed in O4 C)
C4 the report and have not been addressed in our presentation. But C4 C)
C)
C4 they will be addressed in supplements to this report, as required.
U A And I to illustrate well, we do not believe there are o
10 any serious problems here, but there are problems t1 at we do U
ztt
~
Vl have to chase out;to illustrate, or example, concern number, 24, U 12 some 26 deficiencies in with miss'ng documents and missir.g z
13 heat numbers, and so on and so forth.
hat is not on sacrificial sh'eld wall members. Those oLe 15 are on washers and nuts and cotter pins and th'ngs like that.
~
~
16 And, for example, the two weld maps, the contractor: weld maps really quite detailed. It is possible that, these weld maps (0
17 are fd 18 are misnumbered. o say that they cannot be located, we are F
c C)
C) 19 going through an administrative process to determine if they do--
20 if they should exist or if this is a reporting error.
21 hank you, Jack.
22 With respect to these six things, as I said, they 23 largely arose as a result of ou records review. And based on 24 looking at correlation between the kinds of defic'encies we four.d:,'5 in records, Leckenby records on the pipe whip rest a'nts, we ALDERSON REPORTING COMPANY, INC.
dspS1 began to look for s-'milar types of defic'enc'es on the sacrificial shield wall. And we wi"l continue to do that.
Again, though, we think that these are, in terms of
~ ~
4 our meeting here today, matters of secondary concern. Again, we 5
see the three primarv concerns on the sacrificial shield wall as 6 the correction weld for ring two and. ing three, which we believe is a reasonably straight forward kind of correction, one CD CI in which we have the exper'nce to implement.
C4 O
4 I apologize that I did not anticipate your quest'on on O the as-built condition=. Ne have made a call to the site; .I hope U
z CO before the mo ning is over to have an answer for that. But .we U 12 are confident, without. having .the absolute answer, we are confi-13 dent that the design is not "erriblv sensitive to that and that we could make the correction weld.
The solution to the shielding problem, we also believe 16 a practical, pragmatic solution to the problem of construction 17 ove sight and error.
4 18 he weld quality problem, we may have painted a rather bleak picture; we have tried to be frank and forthcoming about I
20 that. In spite of the difficulties that we found, a review of
~ I 21 the -- and our concern about the apparent qualitv of the visual 22 inspection based on our reinspection. Ne are also somewhat 23 optimistic because we have not found in our reinspection major t
24 deficiencies. Ne have found one ,defic'ency which would require a major repair.
ALDERSON REPORTING COMPANY, INC.
82 dsp82 So we are somewhat optimist'c, but we do not want to draw any final conclusions today, as E
we just'ave simplv not followed all the threads to the end of the line.
At vour request, then, we wil'o ahead and : will ask LO Nr. Burns to speak about the structural steel and to do what lO previouslv was the th'rd part of the presentat'on at this time.
IQ C4 CO Cl HR. BISHOP: Roger, mav 1 ask you one thing at this O4 CI CI time. Uhen will the Supply System be in a position to make a C4 N
0 d' statement about. their assessment o the techn'cal 'and structural 10 adequacies of the wal 1'?
U z
9) i'. FOLKv: 5 would like to be able to respond to .that U
z 12 d'rectly, om, but i thirk it would be speculative iT we e to 2 that, because we have not o 1 1 owed a 1 1 the routes down . ~le 1 3 say 14 do not know if 'we are going to find something a week ~
rom now 15 in terms of documert eview which would give us some technical concerns ~
17 Right row our strategy is: ' 'e do not fird any more techn'cal problems,to look at each of the deficiencies we have found 'n some analytical way and bounce that back against the 20 design and go from %ere.
21
>>e think that that will prove to be that w'l prove 22 that the w'all is st ucturallv sourd, even with some of the 23 i de "ic'rcies that we are irding. Z nd that could ta} e several 24 months in the sense of coding each de ic'ercv, descr'b'ng it, 25 rurning 't through an analysis wh'ch says, "Ghat is the 1'.kely ALDERSON REPORTlNG COMPANY, INC.
83 dsp83 discount and joint effic'encv," and then taking that joint efficiency to the various critical members in areas of the wall that we cannot see and unning tnat through the design.
So, as interested as we are in concluding that, we feel ld 5 that a lot oz time has been spent on the sac shield wall in Cl LCI 6 fact, over two years there has been'eneral discuss'on. And l4 Ol C) '7 we are anxious to put it to bed.
C4 So rather than, as anxious as we .are to put it to bed, 0 ve feel that we have to do that in a rathe detailed fashion. So 4
R o
U 10 I -- the short answer is: "I do not know."
z N NR. BURNS: I vill discuss items of concern 13 and l4, 12 both of vh'ch relate to the structural steel in the dry well.
13 This structural steel is steel that vas primarily designed to
'I support pipe whip restraints. Xt is made out of A36 in the main; lt is fabricated to AP'.S Dl.l,. There are three general types of structure. There areplatforms wh'ch are comprised of radial 17 beams.
The e are cantilever members wh'ch are attached to the 19 sac"ificial shield wall.
20 And. there are also a number of other specialized 21 structures, an example being one associated with the main steam 22 reliez valves.
23 Just to br:ieflv give vou some background to the problem, a
1 24 some time aao -- approximately tvo years aao siani. icant I 25 cons"-ruction problem vas i'dentified on some of these welds.
ALDERSON REPORTfNG COMPANY, INC.
84 dsp84 Cracks were found in the some of these welds; subseauent to that time, all field welds that. we e made prior to +his discovery of the problem were inspected by NT. That was a -100 percent
'nspection. And all defic'encies exceeding program criteria were repaired.
C C4 lO All weirds made subsequent to that time were made using C4 Cl C4 new p ocedures. which we cons'de are very conservative; there C4 CI will be NT inspections at a number of intermediate stages of O
4 completion of the weld, as well as a final inspection, and an oF 10 inspection after 72 hours. In other words, we feel tha+ we U
z H have done a very thorough inspection of all welds concerned and U 12 repaired all deficiencies that were identified.
z 0
13 To give you just an example of the kind of structure CO 14 we are talking about, this is a sketch, a plan of the 541 Cl E
oLe 15 elevation.
16 (Slide) u$
17 The inner circle is the exterior of the sacrificial
à tn shield wall. The oute circle is the interior of the containment.
E c
CI 19 vessel. he lines -- the radial lines are the beams we discussed, 20 and the cross members are tie members.
21 These radial beams are heavy members, in general 426 22 pounds per section, and many of the tie members are of similar 23 size at 550 pounds per foot.
24 The radial beams are welded to the sacrif icial shield 25 wall, and the tie members are welded between the radial beams.
ALDERSON REPORTING COMPANY, INC.
85 ds985 That is an examIple of one of the platforms to give you an idea of what the size of the menbe s ar involved..
The first concern, concern 13, let ne ead it out:
"The generic procedure used to reoa'r 1am'naticns in the weld zore of the sac wall related st=uctures requ'res grinding of the lamination s of a maximumi depth or 3/8 'nch followed by CI ewelding. This falls short of the >VIS code which requ'res CI CI grinding to a depth of one inch with supolementary ultrasonic C4 U
A tests, as equired, if laminations are longe thar. one inch.
o 10 he anparent concern 's that we would have to identify U
z laminat'ons and where we have identif'ed them, we t=eated them in 12 a manner which did not conform with the AViS code.
13 The problem is that. the examinations referred to were examinations on rolled surfaces, not on edges o on weld E
O 15 preparations. Ãhere an attachment was being made to an existing SQ 16 memiber, the surrace would be N -ed prior to the attachment being 17 made.
18 2nd then any defects would then be ground to 3/Sths.
CI 19 The 3/Sths was chosen when we discovered the defects we were 20 3.ooking for were rolling defects. This was a simple direction, 21 just grind, to 3/Sths and most grooves would be removed at that 22 time.
23 Zn the main, tne defects do not extend to a great 24 depth. However, the problem really arises because of the term 25 lamirations. ~Pe were not looking for lam'nations. Laminations ALDERSON REPORTING COMPANY, INC.
86 dsp86 that were identified on plate edges or on weld prepa ations were in fact treated per the AHS code. Ne we=e looking for 3 rol ling def ects.
Our conclusion is that. we have no -- we need no further lO action. T'e reasoning behind that is that, as Z said, we were PJ ld looking for the rolling defects, not lamination. The grinding lO CO C4 was adequate in that it allowed us to remove 95 percent of the C)
C) defects without further inspection.
O d However, having g ound the defects to 3/8ths, the R
o 10 excavation was inspected bv !~T and any defect ound in the bottom U
z R of the excavation was evaluated by Dl.l criteria. Further U
z 12 grinding was then resumed if it exceeded the criteri'a. Tf it 13 was within the criteria o ASS Dl.l, then the excavation was 14 rewelded.
0 So, in conclusion, we do not feel that any further PQ action is required on that, concern.
N 17 Anv questions on that concern?'R.
18 BISHOP: Dave, in looking at the records and I
C)
'nterviewing the pe sonnel onsite, we ran across this standard 20 addenda which talks about the lamination repair, and so forth, C
21 and they indicated it was standard procedu e used 'n 35 to 50 22 instances.
23 l think you have just told us that all 35 to 50 24 instances were not laminations, but were in fact rolling de ects ALDE:RSON REPORTING COMPANY, INC.
87 witn the AUS code?
HR. BURNS: It is mv understanding that, having- inter-v'ewed certain people, that the defects which were detected by this examinat'on frere not laminat'ons. They were it is lO 5 a matter of they vere defects found on the rolled surface of C4 t1.e plate, not on the edge. So they'ere defects in an or'enta-LA ICI 6
C%
CO 7 tion that would not normally be exoected to be laminations. Any C)
CI Cl defect tl at was found was in fac" ground to 3/8ths, and then
~- 4 C$
4 the bottom of the excavation vas then 2T -ed, and any defect o
10 rema'ning in the bottom of the excavation was removed. according Q
z lO to the A:$S c"iteria.
U 12 So I am not su=e the terminology 's really relevant.
z C5 13 Anyth'ng t1.at vas found in this inspection vas removed and 14 reinspected accord'g to the criteria.
15 L4R. BISiiOP: This came to us as an allegation, and 16 the allegat'on vas made bv oersonnel involved in the reoair of 17 these, and thei- concern was that they did not even call them laminations; they called them cracks.
19 And they said ' was thei understanding that you 20 were epairing these cracks by cover oasses; grind'g down to 21 3/8 inch and then covering it with welding.
22 It then came to light that these vere not c acks but
.23 that these were 1am'nat'ons. So I guess to close the loop some-24 where along the line, the individuals who or'ginally identified 25 these remaining discontinuities under that 3/8ths gouge, that ALDERSON REPORTlNG COMPANY, INC.
88 dsp88 should be pursued a little bit to find out i there is indeed the remaining laminations under that.
HR. BURNS: 'I thirk Len can explain that.
NR. Ait'ZRS: here has been some edge preparations that lO had laminat'ons in it; thev have been removed. What is involved
à Z here is that to add a nanger to the face of the beam requires, CO C4 rather than visual, an N . S.nd I went back to each and everv hl CD C) welding eng'neer that I had, and he said that when they are in 04 C$
A question, they would g ind them down to 3/8ths ard evaluate by o
10 !1T;if they saw somethirg there that reeded to be emoved, in U
z Ql their opion'on, they did. If not, they welded it.
12 Ne do have records, .'Zom, where ther have been edge 13 laminations, but they have been completely removed.
i'd%. BISHOP- Okay. 'Zhank you.
o 15
'1R. BURNS: If there are not further questions, I will A
16 go on to concern 14. Let me read it: "Steel st ucture bridging 07 17 from the sacrificial shield wall to the 'containment wall have fx7 undergone significant weld repairs in the past two years.
C CI 19 Licensee consultants have determined a need to maintain some CI 20 minimum temperature of weld joints to ensure adequate nil 21 ductility transition temperature characteristics of the 22 structural welds."
23 Just a br'ef ecapp- ng: we did f'nd defects in fie'd A
24 welds. 1'ie did inspect and repair. and did take conservative steps, 25 on all new field welds. The consultant that we used to provide ALDERSON REPORTlNG COMPANY, INC.
89 dsp89 some assistance is setting up this procedure. They made a recommendat on in the'r report that these structures should not be operated below 'that was postulated above 140 degrees fahrenheit.
ld This 140 degrees fahrenheit is supposed to be the Ol lO average containment temperature. it is also above the average O4 C) containment temperature of 135 degrees and it is significantly C) above the minimum containment temperature of 85 degrees.
C4 O
d The source of the 140 deg ees,that was a ma g'n above o the maximum NDT temperature. he consultant recommended a margin U 10 R
CO ll of 90 degrees fahrenheit above th'e max'mum NDT temperature.. He D 12 chose 90 he arrived at 90 by a 60 degree margin plus a 30 z
A 13 degree additional margin tc take into account the thickness of 14 the structures.
o He felt that 60 degrees is adequate for one inch thick ta1 iI A
16 and an additional 30 degrees was adequate for the thicknesses o$
V 17 of inte est.
The maximum NDT temperature identified vas 54 degrees.
19 The source of that was measurements made on two beams, two 20 beams which had been identified has having cracks on them and 21 two beams which brittle propagation of those cracks was identified, 22 The maximum NDT temperature measured from those beams was 54 23 degrees fahrenheit. hat is the source of the consul"ant's 24 recommendation.
"he consultant also indicated that he =elt the e vas ALOERSON REPORTING COMPANY, INC.
90 d sp.90 some conservatism in his recommendation and suggested that we explore that conservatism. h-' is what in fact we did, and we have made an evaluation; the conclus'on of that evaluat'or was that the e is adequate margin aga'nst fracture if we operate over 100 d grees fahrenheit.
7'!e arrived at this marg' by a fracture mechanics evaluation. I should point out .that it is 30 degrees more than our assumed maximum NDT, and it is 50 decrees above the consultant's assumed NDT. And it is 60 degrees above the 1G recommended assumed maximum NDT in the recent NRR-NUHZG document.
l will try and explain the source of our conclusion.
?Ãhat we set out to do was to perform a fracture mechanics evaluation of the fracture safety desicn margin on the structure 14 at the design operating temperatures.
15
~
o do th's, r~
we had three activities: the first one 16 was to do a detailed stress analysis and modeling exercise" on 17 three pipe whip restra't structures.
\
?A?e identified the three 18 pipe whip restrairt structures w'th the highest: design stress/
ps C
CI derlect:ion levels.
l'le then assessed the mate 'al properties of 'nterest in 20 I the st=uctures. The controlling mater'al we determined was A36.
22 Then we also asked Burns and Roe to prov-'de us with a definition 23 24 i".'e used the results of the three steps to calculate r
25 c itical crack sizes witnin these structures in terms of the
. ALDERSON REPORTING COMPANY, INC.
dsp91 design temperatures. in the containment.
In all we did about 30 more than 30 calcu'ations.
Ne analyzed 30 crack-geometry combinations. And I can illustrate 4 the process by just taking one of the geometry conservatisms EO and two c ack assumptions.
C7 C4 LO (Slide)
C%
C)
C4 This is the joint between the radial beam and the Ol C)
CI sacrificial wall. The sacr'ficial wall 's not shown. This is Cl 0
d 9 a radial beam flange. 'his 's a plate. And this is the R
1G attachment is made betireen two cheek plates which are welded to R
ll this web and then welded to the face of the. sacrificial shield g
12 wall.
z 4
' M 13 14 (Indicating)
Nhat we did is we did a final analysis of the structure Cg E
o 15 and calculated the stresses operating at the connections, and
à A then we did, model flaws at these connections to.look at the vf behavio- of these flaws.
18 Again, I will just illustrate two examples of the types of flaws that were introduced. There is the edge crack 20 running across the top of the actual between the cheek plates and the sacrificial wall. Ne, at all time , tr'ed to look at 22 cracks, resulting cracks from the stress.
23 r A further example is a circular defect in a location 24 of maximum stress. During the fracture mechanics analysis, we took into account all design loading, residual stress distr'bution ALDERSON REPORTING COMPANY, INC.
92 and in some cases we took into account st esses which would be introduced du ing the fabrication process.
The results'of the analysis came out on graphs of th's type wh>>ere the stress intensity factor for a particular flaw was plotted against the dimensions of the flaw. he circumlferen- '
t'al b eak and this break indicate two di ferent types of pipe break we analyzed.
( Indicating)
So what happens is the stress intens'y 'factor gets 1G higher as the fracture gets bigger. he stress intensity factor being measured the severity of the crack takes into account 12 the size of the crack and the .stress f'eld within which it lies.
13 Knowing the st ess intensity factors for particular 14 cracks, you are going to have to know the critical volume of 15 the stress intensity factor for the material that 's controlling 16 to assess the crit'cal crack size. As I sa'd, the critical 17 mlaterial was A36 in th's case and it was'ssumed we had a maximum 18 ND7 temperature of 70 degrees fahrenheit. Ne also assumed the 19 behavior of the stress intensity factor was a fur ct'n of 20 temperature, and it. behaved in a standa d fashion w'th a 21 standa d ElD temperature curve index with the NDT temperature.
22 Ne plotted out the cr'ical stress intens'y facto 23 as a function of temperature in the containment. It turns up Gl a l-'ne here at a mark, 70 degrees, in th's case. Go a+ 70 deg ees 25 fahre..heit the critica' ess i tensity ".actor for A36, 'n this ALDERSON REPORTING COMPANY, tNC.
case, is 40 As the temperature increases, the cr'tical stress intensity factor increases, as indicated by these po'nts.
(indicating)
To determine the critical crack size, ' ' then necessarv to go down to the bottom axis of the part'cula temperature of interest; '0 degrees, somewhat less than one-half inch; at about 100 deqrees, we are about half an inch.
l should point, out that this analvsis is or an edge crack, an edge surface crack which extends across the complete 10 width o the member and results in worst-case analvsis. This 12 particular detail represents the" worst-case results of the 13 total number of analvses that we did.
14 So we are talking about an edge crack the complete 15 width of the member
~
~
with a depth of half an inch at 100 degrees 16
'ahrenheit.
17 we look at, other types of c'"ack geometry, we'f'nd 4 18 the critical crack size is much larger. As an illust ation, this 19 crack the crack geometry in this case is a buried circular 20 defect. At 100 degrees we are pretty close to a two inch diamater, 21 crack, which is about the same size as the c oss section of the 22 member.
23 Simil rlv, when we look at surface cracks which were not completely across the width of the ...ember, the aspect ratio, 25 we find again that the critica'rack s'ze 's much large . Ou ALDERSON REPORTING COMPANY, INC.
dsp94 conclusion was that at 100 degrees fahrenheit, we have adequate design margin for these st uctures, given the conservatisms that we have used in the analysis and the wors" case assumptions that we have made all the way through the analysis.
(Slide)
Ne feel that we have a number of further actiors to take. The first one is. that we need to verify the predictions of dry well temperature during test, startup,and operation. Our U
A indication from Burns and Roe predictions -
that a minimum o
E U 1G temperature of 85 degrees is at the discharge of certain heating z
M and ventilating fans in the plume does not impinge upon any. of 12 the structures of interest.
0 13 The structures o" interest, our indication has it at Cl 14 the moment, that the minimum temperature of those structures is A
0Oc 15 95 degrees fahrenheit.i ~
Given the errors in predicting -- making 16 these sorts of predictions, it is pretty close to l00 degrees cd 17 fahrenheit. And those are steady state condit'ons.
fd C>>
to 18 Ne are then going to look at effects of this temperature 19 limitation on the plant startup. Xn other words, do we have any CI 20 conditions during startup where a LOCA would be postulated and 21 the temperature would not be as predicted in the steady state 22 conditions?
23 Now, we intend to look at these things further. Ne 24 already have a tech memo from Burns and Roe which came recently, 25 discussing these factors, discussirg the second factor. isle have ALDERSON REPORT1NG COMPANY, INC.
95 dsp95 not fully evaluated that yet. Iqe 'ntend to provide a final report of this evaluation which summarizes -'- explains 'n detail what. we are doing and'lso .includes any of those .things that we still have remaining.
Cl 5 That is all Z have.
à Z Anv auestions, please?
C4 LR..GAMBLE: l cannot .emember. Did you,say why
~ C4 C)
CO C4 what why did the consultant recommend t.",at why did you hire 0
ci h'm? b'hat was he doing there? 'Zo f'nd c acks or' 0,
U lc NR. BURNS: *Yes. be found cracks in the structures z
R ard we had to repair them. $ 7e hi ed the consultant to provide Q 12 assistance in setting up suitable repai procedures and suitable z
13 correction action. And we set out what we feel are very conser-14 vative inspection and repair procedures.
oW 15 However, the consultant was aware of the function of fil 16 the structures, and he felt that fractu e safety desian was rA 17 reauired for this type of structure and, this type of weldina.
18 He provided what he thought was a reasonable maximum operating c
C) 19 not easonable -- a conservat've minimum operating temperatu e.
C) 20 NR. GAMBLE: So you found some cracks. And " guess the 21 thrust of your -- vou removed them ard repaired HR. BURNS: Yes. o the best of our knowledge 's sound, 23 no cracks. However, we are look'na at the aeneral fracture 24 safety desigr. of the structure.
25 HR. GA."ABLE: And-A you are goina to take all this and put ALDERSON REPORTING COMPANY, INC.
96 dsp96 it in some kind of report?
NR. BURNS: Yes.
MR. GAi~BLE:; And that vill be submitted, vhat, on the docket?
ld MR. FOLEY: Yes.
C4 lO MR. BURNS: Yes.
P4 CD C4 MR..FOLEY: Yes, that is true.
C4 MR. E&YNES: There are no cracks .that extended to the 0 surface because you did surface examinations; is that correct:?
d o i4R. BURNS: Yes.
U 10 z
CO MR. FINES: You did not do UT.
U 12
~IR. BURNS: No, except- in very limited cases for specif '
z 13 reasons; part'ularly, sacrif icial shield wall attachments were 14 UT-ed for specific reasons; that. is, one of the critical P+
E o attachments.
16 HR. GAMBLE: The residual stresses that you used in CQ 17 your analysis, generally, what were the magnitudes?
\ MR.. BURNS: Right. For fillet welds, we used a distri-19 bution which included a maximum magnitude of half the yield; point; 20 other welds, ve use'd a distribution which included a maximum 21 magnitude of yield point.
22 &le feel this is conservative because we are op rating 23 at a significant margin below the NDT. chen you get significantly 24 below the NDT, those stresses do not have a significant ef ect 25 on fractures.
ALDERSON REPORTING COMPANY, INC.
97 dsp97 ~LB,. HENDERSON: this NDTT number cf 54 degrees, that was not generated from specific experimental data from samples on these elements.
NR. BURNS: The consultant's recommendation was based ld on the 54 degrees. That, was generated from two beams that we C4 ld removed from tne containment where ~m actually measured the NDT lO C4 CO C4 temperature --.
~ W
'hl C) NR. HENDERSON: Oh, it was.
U =- by the 4 ?IR. BURNS: AST?0 procedures. Ne measured o max'mum.on the two beams; one was a maximum of 41 degrees; one U 1G z
was a maximum of 54. Ue measured mary locations on each beam to Q
z 12 t y to identify the NDT. c7e d'd not just use the flange H
region which is the nor.:.al AS N posit'on. rle looked underneath 14 the web at the worst case situation. ~He identif'ed the worst hl oLs case of 54.
16 But to take into account possible additional cd g 17 statistical variations, we used a maximun NDT of 70 degrees 4 18 fahrenheit. As I pointed out, we think there is conservatism in that assumption.
20 NR. GA'4IBL'E: How thick are these beams that 21 NR. BURNS: In general, up to about. three inches; there 22 are specific instances where there are heavier members involved, 23 but these are generally located in specif'c locations. I am 24 thir king of one, there is one frame st"ucture which 's associated 25 with the mair steam 'sclation valve -- we cail it the b'rdcage ALDERSON REPORTING COMPANY, INC.
98 structure which has members which are thicker than three inches. That is the exception. ':7e have analyzed one of those situations.
NR. GAMBLE: And tHe 5 MR. HENDERSON: The i unde stand that this you 6 have considered only LOCA loads in def'n'nc the temperature.
7 What is the situation with respect to seismic loads when the plant is shut down and the 'vessel 's open, say, for refueling?
NR. BURNS: We used LOCA, seismic, live, and dead 10 combined. Ãe assumed coincidence of all design loads at one time, adding them all together. So we have the safe shutdown at
?4R. HENDERSON: bell algebraically -'-
A 13
""IR. BURNS: I am not saying the right thing. We took 9)
~ t+
14 the maximum load 15 MR. HENDERSON: That triagered something.
I 16 NR. BURNS: Ne added toaether all the loads and assumed 17 all accidents at the same time, all possible loads.
18 1'4R. HENDERSON: That was only at 135 19 NR. BURNS: We assessed the--
20 NR. HENDERSON: -- operating ambient temperature.
21 MR. BURNS: We have assessed the design over the 22 complete -ange of operating ambient temperatures, yes.
23 MR. HENDERSON: How about the shutdown ambi'ent 24 'emperatures?
25 MR. BURNS: No, we have not analy ed the fracture ALDERSON REPORTING COMPANY, INC.
0 99 dsp99 safety design at the shutdown ambient temperatures because the--
NR. HENDERSON: 'lith, say, simple seismic loads.
NR. BURNS: 'he seism'c loads, as I .understand it from the design engineers, the seismic loads are ve v insignficant ld 5 compared with the LOCA loads. hat is:the stresses generated C
C4 LQ 6 bv the LOCA loads I carnot recall specif-c figures, but they Pl CO 04 7 are significantly larger than the. seismic loads. So, we do not
~ W O
O C4 have, to answer your auesti'on, we 'have not examined the fracture O
A safety design under seismic loading durira shutdown conditions.
o HR. HENDERSON: Okay.
U zH
~
lh NR. GAHBLE: Okav, but what, you sa'd you were going U
R 12 to do was you were going to go hack and see if you vere going A
B to look at the operating conditions I
and see if vou could have the 14 OCA.loads at some lover temperature o 15 NR. BURNS: Yes.
16 I'TR. GA"~BLE: in between, you know, during startup or 0$
17 something like that.
18 NR. BURNS Ou- indication has it at the moment that 19 the thicknesses we are talking about v'th the heatup, normal 2D heatup rates, that k
we should not have a p oblem, a limiting 21 problem at sta tup. Ne should be at, the operating temperatures 22 before we nave postulated 23 5IR. GAMBLE: You mentioned that you did a fracture 24 mechanics analysis; how did you determ'ne the -- vel', you had 25 a critical stress 'ntensity factor of FlC or KlD o whatever it ALDERSON REPORTING COMPANY, INC.
100 dspl00 was. And how did. you determine that particula volume?
SIR. BURNS: To determine the K1D volumei we assumed a characteristic behavior of i(1D as a zuncticn of temperature in 4 a sort of analogous manner to ASIDE section three and section IA eleven. And we indexed that to NDT tempe ature. Ard then we LO used the assumed distribution of HDT temperature to shi t that lO CO C4 7 standa d curve. a long the temperature axis and'hen p lotted the C4 C) rise of K1D as a function of te.,perature based on the maximum CI 4
o 10 NR. GAMBLE: Okay. And the value of '70 degrees Q
z M
fahrenheit for okav, "you said that you had to get. to a U 12 temperature you had to maintain a temperature of roughly R
that.
' hl 1
14 3 1 00 used.
degrees And at or somethinc 100 degrees, 1 ike That was the temperatu what was the value of K1D'?
e you F
O Rl 15
'Nhat I am i ~
getting at is that. you have to be careful 16 because A36 may not have as much toughness as something like 17 what you would find in the AS~IE code. Zt 's going to run out of toughness before the reactor vessel steel rurs out of toughress But if the number is low, it does not make any diz e ence.
'k 20 HR. BURNS': About 60.
21 NR. O'AMBLE: hat is pretty low.
22 LTR. BURNS: Ye employed. a consultant in assisting us 23 in making these assessments or. the properties that we'hould 24 assume for 'A36, and accord'ng to Bill Pell'ni he has analyzed available data and also looked at the available data that, supports ALDERSON REPORTING COMPANY, INC.
101 dspl01 that standard type behavior of K1D versus tempe ature or A36.
MR. GAMBLE: Just for scheduling p'urposes, because 3 we norma 1 ly do not get reports like this, when do you anticipate l ~
4 this to be submitted to us, so that we can plan, you know, so somebody can review it. from our group.
1 CP 40 MR. FOLEY: Ne have a draft of the report included in C%
C7 there, and the final report with respect to the HD matters will
~ W
'C4 CD 8 be a matter of what, four weeks?
O A BURNS: I said in. the report we should. have it o
Q 1G before April '79 April '80.
R tO ll MR. GAMBLE: You are going to submit it on the docket U
R 12 which means we will have to look at it befo e the 1'censing I
A 13 because can go back and find out when the licensing, when we M are scl eduled to finish our licensing ef orts. So I want to know o 15 about when we are going to get the repo t.
16 MR. FOLEY: About April.
cd MR. GAMBLE: Okay.
4 18 ICR. BURNS: Ne have just we have all the information L~
t C7 19 together; it is just tvirg up these two last concerns, reviewing 20 the Burns and Roe technical analysis; and the resulting differences, any questions, incorporating them into our final 22 report.
23 Thank you.
."1R. FOLEY: Any urther: questions about this subject?
25 (No response)
ALDERSON REPORTING COMPANY, INC.
102 dsnl02 Hell, this obviously is 'a subject with sor:.e history; we believe, however, we have taken a comprehensive look at this
. atter in the context of ou= situation. ':Ve have out 'some good talent on it. ':7e have employed some consultants o f both nat'na'nd lO 5 international reputation.
Ol lO 6 ~7e think we have an analysis of the condition that we CI 7 can live with and will be supportive of the project.
C4 CI C) he last and the 'third area of d'scussion is pipe C4 U
ci whip restraints. Yr. O'Dornell will return to the front of the o room and give you a background as to the design considerations U
z Ql associated with the restraints.
U 12
.'4R. O'DOHNELL: Basically, think that the purpose here R
is to just bring people up to speed that may not be familiar with 14 the design reoui ements for pipe whip rest"aints "'n terms of o 15 their func"ion. T. think that the terminology, pipe whip i ~
16 restraint, is self-identifying. Xt is a restraint that is cd 17 placed there in case of the postulated failure of a pipe. The .
18 pipe whip rest"aint is the area within this dark circ'e.
19 (indicating) 20 The pipe .in question and the annular space in between here, which is the maximum distance that we care for this pipe 22 to travel before it impacts on this particular restraint; these 23 are located throughout the dry well. T¹se'rvel8ivents are supported on 24 either structural steel in this manne, imagin'ng this to be a 25 wide flange beam, or another configuration we might have 's ALDERSON REPORTl NG COMPANY, INC.
103 dsp103 supported off the sac wall, and the ones that X think Tom has
. 2 seen more often, because they are in place, 'this type of rest-aint.
Again, the restraint is the p'ece of material within 0 this area.
C4 lO IQ (Xndicating)
C7 he .restraints were fabricated by Leckenbv Corporation CO under a separate contract,'contract 90, as d'fferentiated from C4 the subcontract work that they did or contract 215, 'the O mechanical contract.
U z
I X like to bring that point out because during some U of Nr. Johnson's discussion, "his delineation of cont act will z
' W 13 14 be brought out.
As X said, the material is -- the pipe wh'p supports E
o are procured and fabricated bv contract 90; however, they are N
installed by contract 215; since contract 215 installs all the responsible for the final cd 17 main piping, they in turn are
à h 18 installation here.
19 Thev are also responsible for all the shimmirg which 20 is necessary and the restraint blocks. Xmagine, if you will, the pipe could break in any direction. Go it could impact up or 22 sideways. The end stops take the lateral loads; the horse 23 colla takes the vertical load. Ard the beam itself would take 24 the vertical down impact.
25 And Dave has d'scussed the ramifications of the ALDERSON REPORTING COMPANY, INC.
104 dspl04 structural steel; these restraints are generally high strength mate ial, 516 or 537. Thev have all been post-weld heat t eated, impact tested, and let me also point out that, down here the connection between the collar and the base plate is a full lO penetration weld, which ensures that the attachment of the horse IO 6 collar to the base plat should be stronger than the parent 7 material itself.
Just in-closing, 'the loads were derived based on a I
0 blowdown analysis using the RELAP-3 program. A dynamic load d
o E facto was developed, and a static load used in the design of U
z Ql these with an allowable stress almost up to yield stress.
)
U 12 Are there any general auestions before Yir. Johnson 13 comes up?
14 (No response) h1 o 15
'41r. Johnson will now discuss the remaining seven Cl 16 concerns with relation to pipe whip restraint.
th 17 (Slide) 18 iIP.. JOHNSON: I would like to start with a little bit of history, some of which you just heard Jack discuss. Basically, 20 the subject of my d'iscussion is the seven items of corce n or 21 problems associated w'th QA records for pipe whip restraints.
I would like to repeat that the restraints were made under a 23 contract, contract 90 ~iiith the Leckenby Corporatior.. The restraints 'were insta'd by the 215 site contractor.
25 (Slide}
ALDERSON REPORTING COMPANY, INC.
105 dsp105 The auantity of hardware involved, there are a total of 179 pipe whip restraints. There a e fou= basic tvpes with types one, two, three; four, types three and four having sub-4 categories which have some mino- design changes on them.
Of these, 30 have been installed in place and. inspection is complete and accepted. 21 have been 'nstalled and inspection C4 CI C4 is complete.
CO CD be have 128 that 'are not installed.
C4 c5 A (Slid )
0 Once again, the specific rest aints, types two, three, U
z N and four are basically of this configuration.
U 12 (Indicating) z A
13 (Slide) 14 ype one. is of the othe configuration that you have foal E
o already seen.
16 (Slide) cd 17 Zn mid-1979 the auality assurance department for the 4 18 was asked to review the records associated with contract 'roject 19 90 prefatory to contract closeout. V7e reviewed at that time a 20 sample of the documentation and found some unacceptable items; 21 proceeded on to a 100 percent review; discovered missing 22 documentation and some cases of material substitution.
23 8e contacted the supplier who agreed to provide us with the missing documents, and in earlv November our document 25 review revealed numerous d'sc epancies with the document ackaces.
ALDERSON REPORTlNG COMPANY, INC.
106 dspl06 (Slide)
At that time we felt we had sufficient evidence to analyze this as a potential 50.55(e), and'e did report it to .
NRC Region V as such. A stop work order was issued to the site lO contractor to cease all work on the installation of pipe whip C
V5 restraints in November, and the NPC requested us to p opose a IQ CI C4 corrective action plan before installation was allowed to proceed.
0 C4 CI CI We responded to that in December with our correct'on action U
6 p ogram.
O 10 (Slide)
U z
M Before I go into our program, I would just like to give U 12 vou some indication of the types of documentation problems that z
C 13 we have discovered in the document review. We have broken -these 14 into two categories, the second of which you will see momentarily.
E O 15 These are documentation problems that a e, paper problems that it 16 appear to be items that can be corrected w'thout totally impacting cd 17 the hardware: wrong yea- on UT report; no traceability to to 18 strip chart; welder's ID missing.
E CI 19 And we have identified each specific pipe whip=.,
CI CO 20 restraint affected; 21 (Slide) 22 The other items, the document problems affecting 23 I hardware, once again we have identified the specific 'restra'nts, 24 and these are the numbers of the pipe wh'p restraints: wrong 25 chemical composition on lITR; wrong weld-procedure; weld ALDERSON REPORTING COMPANY, INC.
107 dspl07 procedures not c[ualified for post-weld he t treatment conditions.
(Slide)
Our corrective action plan, as ecuested bv the NRC, we have already i <<plemented parts o. it. Our rev'ew of the lO data for identification of problems, all of the problems LQ associated with the documentation as =elated to each p'pe wh'p C)
C4 restra'n', we completed that in December of "979; also it C4 CO C) included an identificat'on 'of welding and nondestruct've C4 O
d examination conditions; in other words, wnat ue term hardware o type problems.
E U
z M
<7e have identified these on corrective action U 12 recuests which are the meehan'sms we use to transmit to the z
4
' th N
13 14 contractor to get Project a response QA and back fromr h'm.
engineering evaluate: those responses, o 15 and then we issue NCRs Sr unacceptable responses to document PQ 16 the problems. Due to the number of deficiencies that we had and a5 17 the problems with the HDE repo ts and welding procedures, we r~
~ rr 18 felt it necessary to do a nondestructive examination sample C
Cl 19 plan.
20 And we came up with a sampling program to go out and 21 sample the recuired ultrasonic and mag part'cle program that 22 had been implemented bv Leckenby.
23 (Slide) 24 A'nd th's 's our sample program, the bottom line beinc 25 basically that fo" mag pa"t'cle welds we had a o0 percent reject ALDERSON REPORTING COMPANY. INC.
108 dspl08 rate; with ult.rasonic, we came up with a 21 percent re j ect ate, which is well above our total sample program. Ne looked at a total of 29 pipe whip "restraints out of tl.e population of 179.
So, based on the results of our sample program and implementation of it, which was completed in January of 1980, we now feel it C4 LO necessary to do a 100 percent volumetric examination and lO C4 C4 visual examination of all of the pipe whip welds per the original CI CI specifications.
Ol O
What this program, then any defects identif'ed here, o
gW we vill .hire our ovn cont-actor to use qualified personnel, z
CQ qualified procedu es, et cetera, and any defects would be identi-U 12 ied on nonconformance reports, which ve will then tie to NCRs z
13 that we have outstanding against document deficiencies and 14 welding deficiencies so that we have all discrepancies against o 15 any particular pipe whip restraint identified, in essentially one N
16 document for easy traceability.
And we will process this through our 'system.
cd 17 he NCHs 41 w'll be evaluated by our enqineering organization, and deficiencies 19 will be resolved, hardware reworked and/or released as they 20 determine.
21 Okay. Nov, I would like to very briefly address the 22 seven items of concern.
23 (Slide) 24 The first one is pipe whip restraints of the same 25 or similar design we e prov'ded under tvo contracts, one contract, ALDERSON REPORTI NG COMPANY, INC.
109 dsp109 90 I requiring HDE and post-weld heat treatment of welds, the othe",
215, recuiring only visual inspection of welds. 215 PWRs may not have been post-we'ld heat t eated.
Here, the PWRs were all provided by contract 90. You lO heard Nr. O'Donnell say that support steel was provided by hl LO contract 215 and heat t eatment was not requ'red. The spec for C4 CI contract 90 did require AWS Dl.l plus post-weld heat treatment i O4 CI CD and NDE.
0 The 215 required only visual examination.. The NDE A
o Q
was specified on the pipe whip restraints because igid weld z
M inspection will provide addi ioral assurance that the pipe whip Q restraints, the weld metal, o,the welds would withstand a postu-z
' 13 lated pipe weld.
And post-weld heat treatment of the PWRs was specified E
o to assure the dimension tolerance for the PWR for that U
16 critical dia:ension between the pipe wall and the collar on the 17 pipe whip restraint. That is a critical d'mension.
18 (Slide)
Ire ps i
19 Post-weld heat, treatment was not required on the support CI 20 steel because the basic design of the support steel is the shim, and we have plenty of flexibility in that area. So, there ore, 22 we did not need post-weld heat treatment to maintain ever.
23 tolerances ~
24 (Slide) 25 Item eight, NDC reco ds associated with p'pe wnip ALDERSON REPORTING COMPANY, INC.
110 dspl10 restraints conta'n phtocopied inspectors' acceptance signatures.
The next few items basically all deal with record type problems, and our response is basically all of the same .category for these, and this is that the our proposed reinspection program will IO 5 put in place new NDE records for both ultrasonic and mag particle, 04 6 as well as new records fo- performance of those HDL'ctivities lA O4 CI Ol 7 and personnel .qualification.
C4 8 Therefore, that negates the 'tern of concern; however, Ol 0
A 9 we are still concerned wi+h it to the point that we investigated zo U IC each of .these,and as shown here, we have certified statemerts z
VJ ll from Mr. lIoore and Mr. Hamilton saying, yes, they did use U
4 12 photocopy'ng. Zt is a standard practice, and it d'd indeed D
13 affect, basically, all 179 of the pipe whip restraints.
P+
14 Ztem nine states that NDZ qual'ficat'on records cannot o 15 be found for one or i ~
possiblv two individuals who performed 16 ultrasonic and MT on PNRs and that the qualification procedure cd 17 is not in full accordance with SNT- C-lA'.
U 18 Reviewing the inspection eports and the qualifications, i
CI 19 it indicates that there is questionable qualification of level 20 two 'nspections and level three examiners. And 'n fact 21 certification records for one Mr. Charles Baldinger could not 22 be located.
23 Zn support of Mr. Baldinge is a sworn statement that 24 Mr. Baldinger had the experience and >>as indeed probably certif'ed in to SNT-TC-lA level two.- And we have the AE's surveillance ALDERSON REPORTING COMPANY, INC.
dsplll report that was performed on Leckenby which indicated Nr. Baldinger's certif'cation papers were reviewed by the surveillarce engineer:
And a review of the personnel qualif'cat'on procedure 5
revealed that we do have some problems with the procedure as it 6 stands. However, our reinspection p ogram will essentially;;;
7 negate all of these.
C)
C) 8 (Slide)
'I 0
A 9 Ztem ten, the elect"oslag welding procedure used in the o
E 10 welding .of the pipe wh'p restraints was not quali ied using U
R 11 post-weld heat treatment as recuired by the code, ANS Dl.l.,
U 12 This deficiency was identified. by us and has been R
A
- 13 transmitted to Leckenby on one of our corrective act"'on requests.
0) 14 Leckenbv has responded to this indicating that they will re-f4 at this time E
O 15 qualify the electroslag i ~
welding procedure, and 16 we have asked engineering to do an evaluation to determine if post-weld heat treatment is 'specif'cally required for the CO 17 18 electroslag procedure.
c E
C) 19 (Slide) 20 Item eleven states that approximately 90 typical 21 joint configurat"'ons specified on design weld drawings for 22 the pipe whip restraints use fillets which are s'..aller than the 23 min'mum fillet weld size specified in the applicable code.
l:r our review of this, we looked at the majority o 25 the welds in question, and most of these are cont act 215 welds, ALDERSON REPORTING COMPANY, INC.
112 dspl12 rather than cont act 90.
There is a point I would like to bring out here now when I get to where I'am talkina about i..spections. This is an important point because contrac+ 215 inspection records for personnel aua i ications
~
are not in cuestion at this point in time.
The specific concern here on the fillets reaards the engineering rationale for acceptance; that rationale is based on U
A the fact that the fillet does not fulfilla structural reguire-o ment. Therefore, the size is not of prime concern.
z (0
However, we did want an inspec+ion which would irdicate 12 that it was of good aual'tv.:hat inspect'on was pe fo ..e6 by 13 215 ard 213A, another one of our cont actina personnel. And it 14 gives us confidence in the or'inal welds. Our review also 15 indicates engir eer'ng
~
~
has indicated that the design irdicates 16 that fillets are not needed with these strergth reauirements, 0 17 and our inspection has shown us that ro welds are basically 18 undersized by more than 1/16 inch.
19 Our proposed corrective action is that for the welds 20 installed to date we will accept as 's, and for welds rot 21 installed, we will issue a chance rotice to ou= contrac+or, and 22 they will be brought irto line with the code.
23 "4R. B.SHOP: Roger, on that item l am confused a 24 little bit'about the 1/16 inch unde size. The documer t package 25 that Burr s and Roe pxovided me at the site showed about 36 ALDERSON REPORTlNG COMPANY, INC.
113 dspll3 38 of the 90 joint conf'gurations have undersized welds on the order o 1/8 inch to 2/16 inch and one case whe e it was a 1/2 3
inch a 1/4 inch, excuse me, undersized; NR. O'DONNFLL: Tort., I can address that. he original lO head count of 90 was based on the 1969 edition of the AESC code, LO which included plate thicknesses up to six inches. Subsequently, lA C4 CI C4 the ANS let's discuss the 'ntent of the min'mum size fillet Ol CD CI weld.
C4 4 The intent. is to provide enough heat so that when 2:
0 U 10 welding througn large size metal it does not act as a heat svnc and therefore cool the weld of f prematurely. The ANS, recognizing 12 this, no longer even has a reauirement of materials over 3/4 inch
' 13 14 to be more than 5/16 inch.
So the original head count of 90 the majority of E'
15 these fall out of that categorv, the ones that you are referring
) 16 to, and those tnat are not structural welds at all fall out. of cd 17 that category, and there are many welds that are not undersized 18 by more than 1/16 inch.
~ 44 19 Ne recognize that you cannot lay down more than 5/16 20 inch feasibly, and .th's is the maximum heat input; our concern 21 is that to go back and try to reinforce a weld by 1/16 inch to 22 meet the letter of the code r.".ight do more damage than the
'I 23 correction might pose on the structure., So think by the time 24 you filte 'out the discussion in the book ther , you will find 25 that there should not be any that are undersized by more than 1/4, ALDERSON REPORT1NG COMPANY; INC.
114 dspl14 1/16.
MR. BISHOP: The one condition that officially kicked this question off was'the attachment of a pipe whip support bracke+ to the sacrificial shield w'th a 1/4 incn weld. And according to the code, that should have been 1/2 inch weld.
C4 MR. O'DONNELL: That is what 1 am saying. .hat has ld lO C4 C4 been superseded. The maximum size minimum size weld is now C4 CI 5/16 inch.
C4 U MR. BISHOP: Okay.
4 0
10 MR. O'DONNELL: Okay. And tha" is required or plates U
z R up to 3/4 inch. And let me also br'ng up this point: we meet Q 12 the intent of the code because we are basically preheating all our z
13 large size material. So the whole idea of putting the proper 14 heat input into these welds, we have met through the welding in preheating, which is basically what are shooting E
o procedures we i ~
16 for by calling for a large sized weld than the design requires.
Ul MR. HENDERSON: What are you using fo preheat? What
]7 18 temperature for preheat?
E c
19 MR. O'DONNELL: We vary. We vary from approximately O
20 70 degrees up to 250, depending on the thickness of the material.
21 ln other words, there is a m'nimum of 75 degrees; for, materials 22 over l-l/2 inches to 2-1/2 inches, we have been using 175 degrees, 23 plus or minus 25.
24 So this would be the equivalent of the heat input, I am 25 sure, that is intended in this minimum size weld. We have in ALDERSON REPORTING COMPANY, INC.
C I T
ll5 dspl15 essence met the intent of the code, not specifically the letter of the law.
NR. BISHOP: . vou are saying you are meeting the intent of a newer version of the code?
'"IR 0 DONNELL 'o g the intent o f the origi'nal . I C4 LQ you read the commentarv, even on the previous A$;S code, they C>
C4 discuss the rationale for minimum size welds. And in that C4 CI commentary they finally brought it to implementation by revisinc U the ANS code in 1975.
zo (Slide)
U z
NR. JOHNSON: "Item twelve: numerous record rregulari-12 ties and inconsistencies exist between weld maps, manufacturina
' 13 14 orders, welder Inconsistencies and inspection records associated with P"Rs.
include," et cetera, et cete a.
o 15 he answer to this particular item of concern is that r~
16 it is true; basically, all of these th'ngs do exist. Ve have 17 identified those .in our record rev'ew. Ye have identified them hl 18 by pipe whip restraint number. The inspection program will E
c CI resolve many of the NDE and personnel cualification provisions.
CI 20 The balance, as I ind'cated, will be put on NCRs. Any new 21 weldina identified by re'nspection will be identified on NCPs and C
necessary cwork and/or release will then occur.
23 (Slide) 24 I
Item 21, the last item on the PNRs, "The specificat'on i
eauires the use of h'gh st"enath A325 or A490 bolt and use of I
I Ii ALDERSON REPORTING COMPANY, INC.
116 dspll6 hiah st ength bolting specification. Drawinas call out A320 and A540 bolts. These bolts are installed but are not installed in accordance with the high strength boltina specification; that is, plate or bar washers have not been used over long slotted-'
ld holes."
Reviewing the contract, contract 90 did indeed specify C4 C
LQ I C4 C4 A325 and A490 bolts, and the reauirement fo the use of the.
C4 C) high strength bolt specification. The drawings specified A320 C4 0
dl and A540. 'A320 bolts were used. t.
However, they were rot installed o per code eauirements.
U 1G z
0)
Our corrective action is that the insta13,tion U 12 cont'actor, 215, will be instructed to use the strip or plate z
O 13 washers and the spec will be revised to reauire those washers.
a 14 And at the time of the reinspection program, we will have to review the restraints, and 't will be accompl'shed at that t'me.
E o 15 16 Okay, are there anv questions?
cd 17 (No response)
,4 18 NR. FOLEY: That concludes our presentation.
19 In summary, we have attempted to be as frank and forth-20 as we can. We are in none of our evaluations and 21'oming recommendations backing awav from FSAR requirements, nor a e we 22 compromising design.
23 Ãe have two stop work orders. In that light I might 24 say that, mindful of well, it is ou- understanding that these 25 matters are still officially w'thin the purview of Region V and
. ALDERSON REPORTlNG COMPANY, INC.
117 dspll7 that some discussion may occur today as to what extent, these matters will be referred to other organizations.
Mindful of our need to move forward as auickly as possible, and mindful in particular of '!RR's rev'w restraints, lO it is our view that the techrica'ontent of these ma'tters with lO LQ respect to NRR is largely limited to the MD'f structural steel.
C)
Pl Of course, +hat is certainly your decision to make.
C4 C)
CI Ne have two stop work orders in effect: one on the C4 U
4 pipe whip restraints and the other on the sac i icial shield o wall. Ne have with Region V a proaram which would provide ="or U
z Vl incremental release of stop work orders; we believe that 'n U 12 view of our invest'gation to date on the p'pe whip restraints, z
a
' Q) 13 14 thei= accessibility, our determination to repair extent is necessary, that tha+ stop work order them we to whatever seek your lifting that.. stop E
o 15 concurrence in work order.
16 And mind you, I am not asking that todav. T. am 17 "emembering your, introduction. [7e would like to leave you with
.4 18 that today. Ne would like concurrence in lifting the stop work i
C) 19 order on the pipe wh'p restraints and lift the stop wo k order CI 20 in an incremental fashion on the sac"ificial shield wall to perm' us to do whatever additional investigat'on, ar'ndina, and so on and so fortn, and we understand that. om has no objection to 23 that, bu" we wanted to make that clea ; to engage in that work 24 which is necessary to do,the prototype testing and the epair of 25 the shield deficiencies in the wall, which as you may recall, also
. ALDERSON REPORTING COMPANY, lNC.
dspll8 in order to proceed with that, would call for us to prepare the joint fo- the circumferential weld to connect ring two with ring three.
Ne would like to proceed with that work. Ne, at. this 10 time, have no pipe or" add'ional attachments to the wall wh'ch LO lO are pressing in terms of the critical path construction process, CO C4 and we would not seek to make attachments to the wall, further C4 attachments to the wall until we could determin for ourselves 0
4 that it would not in any way obfuscate any additional invest'ga-O U
t'on, make repairs any more difficult.
z M And we believe that as our investigation continueq we U
z may ird a situation'here we come in or additional attachment,
' is not H
1 3 but that our point of view f4 14 To summarize what we would now.'l like to see on the'wall:
O 15 the authority to continue whatever work is necessary, to complete
~
Ql ~
16 our investigation; to do what work is necessary to do the repairs CO to the shielding;, and "to do the work associated with the repair
.tn of the circumferential weld, particularly the surface preps.
~ 44 CI 19
~
Do you have any final questions or comments?
20 NR. BISHOP: One thing X can do, Roger, is to pe haps 21 provide a little background on ou position as far as the immedi-22 i ate action letters we sent. As you explained, it was to prevent 23 any further work on the wall and the pipe whip rest aints that 24 would make the determination of the depth of the problem mo e difficult.
, ALDERSON REPORTING COMPANY, INC.
ll9 dspll9 And because of that, we do not have correct me if I am wrong -- we do not have any objection to you proceeding with any exploratory activity that is needed to define the scope of the
(
problem. And, I thought. we had made that clear in the past, but we have no problems with that at all.
C4 lO 6 It was mainly to prevent urther attachments to the o
C4 7 wall or anything that would make it more difficult to make the C4 C)
C) 8 assessment of the adequacy of the wall.
Ol d MR. HAYNES: If I understand what he is saying, though, o
10 you are also saying the pipe whip work'nstallation could,.
U z
M
'roceed., and that would not affect of the ab'lity of the U 12 inspector or to repair anything on the sac shield wall; is that R
a
' M 13 cor ect?
FOLEY: And I do not mean the installation of the I
E O
LI pipe whip restraints; I am talking about 'the stop wo k which 16 is in effect on the pipe wh'p restraints themselves. We want to cd 17 begin to repai the pipe whip restraints.
g
.4 18 HR. HENDERSON: Continue the examinat'on and repair.
c 19
~
MR. FOLEY: The examination we are not prohibited from CO 20 doing. We are in the process of dragging those pipe whip 21 restraints out of the building; where necessary, we will continue 22 the examination. And as we conclude examinations of the pipe 23 whip restraints and determine that repair is required, we want 24 to proceed with that repair in the most timely and efficient 25 fashion as we can.
. ALDERSON REPORTING COMPANY, INC.
120 dspl20 HR. PAYNES: >Jell, okay. I they were installed, taking it a step further, would that affect the ability to inspect or 3
repair the sac sh'eld .wall'2
- 4R. FOLEY: I would nave to look at that on a case by lO case basis.
Ed NR. FENDERSON: think one thing lO C>
C4 PER. FOLEY: The vast majority of attachments to the C4 CI C7 wall have already been made. The number remaining are nominal.
C4 c5 ci And the amount of space remaining to do inspection is also would be concerned about having the examination N
0 nominal. So we U 1G z
in the area.
U 12 Now, tt ere is no question in our work procedure.,
z 2
13 there is no question about the quality of the material on anv 14 welding associated with the attachments;because of the work f4 F'
15 procedure 84 requi ements, the kinds of testing that, is I
done 16 assures us that that is not a problem.
difficultyvould come -- and we addressed this in N
17 The 18 our incremental work release is what if somebody had in mind 19 to pulkin plate and we were about to make an attachment. Our 20 incremental work release calls for all of the necessary technical 21 and quality reviews on any piece of work that is proposed to be 22 attached to the wall to assure ourselves internally as well as 23 any others of interest that doing that p-'ece of work would not 24 detract from the ove=all. evaluation of the p ogram.
25 The reason i do not seek the lifting of the stop work ALDERSON REPORTING COMPANY, INC.
121 I
dspl21 in the whole sense is because we thirk that the incremental work release procedu e is.a workable th'ng and that we are not 3
at this point pressing to make additional attachments to the wall.
4 Ne would not press that unless we had determined it would not Ih hamper our investigation or repairs as may be reauired, ar d only then if that particular attachment was schedule-cr'ical.
l lQ IQ C4 MR. HENDERSON: But if you do not get them repaired, C)
CO then they will all be schedule-critical.
O 4 MR. FOLEY: Precisely.
o 1G MR. HENDERSON: Are you -- I think I heard you say U
z Ch that you propose to use a separate, specialized contractor U 12 for this repair work.
z
' fg 13 14 MR. FOLEY:
MR. JOHNSON:
I did not say that; That statement was no, I did not.
in the context of 0 15 inspection.
W 16 MR. FOLFY: Inspection, yes.
17 MR. HENDERSON: Oh.
18 MR. FOLEY: And we have not made a decision about who ps will do the repairs to the pipe whip restraints, nor have we final decision about who will per orm t..e ci cumferential C7 20 made a 21 weld on the sacrificial shield wall, attaching ring two to ring 22 three.
23 And I assu e you', whoever does it, we will be monitoring 24 that auite closely.
25 MR. HENDERSON: I guess that 's about all we need to ALDERSON REPORTING COMPANY, INC.
122 dsp12 2 keep you here for todav, then. So you wi11 be hearing from Region V in the nea= future.
.hank you very much. Xt has been a ver~Z informative session.
40 HR. FOLEY: Thank you for you t me and your lO hospk i a13.tye Cl Ol Ol ( hereupon, at 12:45 p.m., the meeting vas adjourned.)
CO CI C4 V
a 0
U 10 R
(0 Q 12
.R oLs N
16 c6 V 17 6 18 c
C 19 20 21 22 1
23 24 25
, ALDERSQN REPORTING COMPANY, INC.
~a7UCLEAR RECiULATORY COMMISSION This is to c . t' y that the attached proc edings befor e the in tne atter o ~; Staff Briefing on Review and Evaluation of blHP-2 Sacrificial Shield Hall, Pipe Whip Restraints & Related Structures Da i e 0 f Proc eec -': pebruarv 6, l98Q Docket l(unber:
Place of Proceeding: 'ethesda, llaryland were he as her 'n that this 's the f'le appears, arid anscr.cz d
or tne of the Connission.
or'"-'na'hereof 4 David S. Parker Offic'l Reoorter {.ypec)
Of.ic'al Reporter (Signature)