ML20092G798
| ML20092G798 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 06/01/1984 |
| From: | Stokes C AFFILIATION NOT ASSIGNED |
| To: | |
| Shared Package | |
| ML20092G713 | List: |
| References | |
| NUDOCS 8406250241 | |
| Download: ML20092G798 (94) | |
Text
b. ., .. a, . . .
14 U% R
/W
^"*****' '84 JR 21 979 ;4 9 I, Charles C,' Stokes, am providing this statement of my own volition in response to various Pacific Gas and ElectriE.- ..~
(PG&E) submittals to the NRC. This is intended to follow up on issues which I have raised in prior statements to the Commission and which I feel must -
still be addressed in detail by PG&E Diablo Project management before Unit 1 is allowed to go beyond 5% power. In some cases my comments -
suggest necessary methodological steps to obtain adequate correction action on my allegations, largely confirmed by Mr. Yin. ,
In other cases, my comments expose f alse or misleading statements by PG&E in their response to Mr. Yin's inspection report.
In -response to PG&E letters on' the subject of Tube steel Radius and Flare-Bevel Welds as addressed in the April 7,1984 and April 11, 1984 to the members of the Appeal Board, and PG&E letter No.: DCL-84-083,
- DCL-84-164, DCL-84-166, and DCL-84-190.
In PG&E letter: DCL-84-164 Enclosure 7 on page 10, PG&E states "The '
adequacy of the DCP criterion was addressed in PG&E letters DCL 083, DCL-8 4-141, and DCL-84-153. As' described in these letters, site inspections confirmed that the tube steel corner radii are, in f act, '
2.0 t (or slightly larger) ."
In PG&E letter: DCL-84-166 on page 47 under the subject " WELDING OVERVIEW WELD SYMBOLS, "PG&E states "The allegations all fail for either a lack of substance, lack of context, technical errors, false or misleading statsments, or a combination of these reasons."
I hope to put an end to the verbal debate about the existence and safety significance of out-of-specification tube steel which is and has been used in constructing f aulty safety-related pipe supports and other safety structures at D'iablo Canyon. This problem is not limited to isolated occurrences, as claimed by PG&E. Until recently, I had only. seen tube steel with radius less than 2 t at Diablo Canyon
' site befcte I was terminated last October 1983. However, that has changed in the last few days. A currently-employed Diablo worker who 8406250241 040621 PDR ADOCK 05000275 #4 0 TPag e_R_o R_3f_2, P@ EPe_S_____ -- - --
. - t, o a- ,
6 was aware of my ongoing debate with PG&E, provided my counsel Thomas Devine a piece of 3 x 3 x 1/4 tube steel which was cut from a piece of tubing being used to constru'ct a Unit 2 class 1 pipe support. The worker informed me' that tubing similar to this example was used during 1983 to construct Unit 1 class 1 pipe supports. The tube steel was given to Mr. Devine for my analysis to forward to the appropriate government authorities. I know that it was intended for use on -
safety-related work, because I examined the relevant support drawing.
It indicated that the support was code class 1 and design class 1.
In PG&E letter no. DCL-84-166 on page fifty under t.opic " WELDING OVERVIEW FLARE BEVEL WELDS TUBE STEEL RADII" PG&E states that " differ-
- ent radii can be inferred for the tube corner dependent upon the measurement method used."
On page 51, PG&E states, "We have termed this the flat side in-tersection method (D dimension). Based on this D dimension, an apparent corner radius of RD is implied. (...) For the 3 x 3 x 1/4 tubes, the apparent radius, RD, is approximately 1.25 t to 1.5 t and the RA is always 2.0 t or slightly larger.
~ ~
For the 3 x 3 x 1/2 tubes, RD is approximately 1.0 t and RA is again 2.0 t."
This tubing (Attachment 1 sketches) not only has a corner radius less than 2t but has only one (1) corner which is 2t. Three corners are less than 2t, of which two (2) are only it as I and others have confirmea. It measures it using the Straight Ruler Method (per c PG&E) and using the Concave Radius Gage Method. An additional flaw exists, which before seeing this piece, not even I was aware of. The tube wall has what appears to me to be a lamination crack running all the way around its' perimeter. This lamination crack is almost exactly at the center line of the tube wall.
On the 24 th of May, I was in a meeting with the NRC staf f in which I was asked how I felt the radius of tube steel should be verified? Should the Straight _ Ruler Method, the Concave Radius Gage Page 2 of 2/ Pages L $$$
. . %L fW Method or some other method be used? I am not going to propose a novel !
method. What is and has been my only concern is that the calculations which PG&E/BECHTEL perform are done with good engineering judgement based on obtaining conservative results. Af ter measuring the sample tube steel referenced above, I recognized that no simple accurate i
method exists for measuring the radii, because this steel is so badly i made that the radii' changes. noticeably along its' length. Depending l
on how you decide to measure the radii you can obtain three (3) !
possible results at each corner. '
As an engineer and in accordance with the established guidelines of relevant professional CODES, I feel the above data ,
should be factored into the methodology. In the case of pipe support i
steel design, I will and would like to see applied at Diablo Canyon the
' definition of tube steel radius as defined by either the (AISC) or the +
(AWS) CODES. Based on my knowledge, and af ter review of page's 4-131 to 4- 149 and Section 1.17 " WELDS" on page 5-44 of the AISC manual, I .
decided that the answer must be within the AWS document. Both documents govern the construction of pipe support steel, but on page 4-149 and 5-44 the AISC references the reader to the AWS code for this -
issue.
.Looking at AWS A2.4-79, the AWS document on SYMBOLS FOR WELDING AND NONDESTRUCTIVE TESTING Section 9.2.9. 5
" Flare-groove welds. The dimension S of flare-groove !
welds is considered as extending only to the tangent points indicated below by dimension lines. (See Fig. 32)" (See Attachment 2, 2 pages)
Based on my own measurements of the sample tube radii and
{
the section 9.2.9, the only conservative method of measurement is the :
.t l
Straight Line Method per PG&E. By using a combination square with a i i i Page3of2/ha s ,
.e , . .
Bevel Protractor Head with a blade graduated in 64 ths in conjunction with a flashlight, a QC inspector or engineer can measure the radius to wittiin 1/64 ' inch. I feel that it is important to state that two (2) values per corner typically can be measured, and it is good engineering practice to use the minimum value in the effective throat calculations to ensure a conservative design. The reason for the Bevel Protractor Head is that the tube steel walls are not necessarily 90'to each other, and the protractor will allow flush contact with the wall of the tube.
I can' t provide PG&E or the NRC with an easy way to resolve the lamination problem. There is one question that should be addressed before any other. Should this defective steel be allowed to remain in the plant, or should a through search be made and all defective materials be removed from the plant? To accept this
~
condition "as is" without compromising public safety, several questions must be satisfactorily addressed. First, Welding-What Joint configuration, ef fective throat, or wall thickness will govern the failure mode which is the most critical? Second, member stresses-What section pr'operly will be used in the analysis of member '
stresses? Third, how will confidence be established that this is the only defective material in the plant?
There is increasing empirical evidence that the answers will confirm a problem. Another Diablo employee told me that some NPS Struts and Snubbers which were for class 1 systems have spherical bearings which are cracked from what appears to be the swageing process when they were installed. Swageing is a process by which material is welded by pressure or hammering. My f riend recalled that Page 4 of 2A }
BagW
e e the spherical bearings are marked "made in Japan." I gave the NRC in December 1983 a large spring-can structural steel at'tachment which also had lamination cracks from cold rolling. I told them that I was aware that three (3) were defective of a random sample of four (4) .
Since. December 1983, I have heard from workers at the plant that 25 defective spring-can attachments had been found in the plant. I should note that the NRC was not concerned with the defective spring-can attachment - it was not installed on a class 1 pipe system (just
- Main Steam!) .
The problem has been confirmed on a suf ficient. scale that to adequately protect the public safety we still must obtain answers before Unit 1 begins power ascension. Results f rom PG&E and the NRC s l
" samples" are sharply in conflict wi'h hardware reports from workers Only a comprehensive review can resolve this dispute of in the field.
fact.
In PG&E letter No. DCL-83-166 on page 48 in the last paragraph, PG&E states "Due to the rapid expansion of the Diablo Canyon plant staf f, specific training programs were conducted regarding AWS A2.4 weld symbols. Three hundred and fif ty (350) engineers and QC inspectors were trained during May, June, and July of 1983. Additional pre-certification training was conducted for the AWS Certified Welding Inspectors Program in June-July and November-December, 1983." c (Emphasis added.).
When you read PG&E's responses to valid problems such as this, their answers are humorous like material for comics performing a night club act. For anyone who has additional f acts to fill in around the flakes of truth, they provide a good laugh. Maybe 350 people sounds like a lot. Let me put the number in perspective. How does 350 ' compare to 7000 relevant employees at the site? Mathematically the ratio is 1. in 20. When you consider that 350 were typically in management, the number left in the field to spread the knowledge Page 5 of 2[ ges ((
. o. , .
cround brings the ratio to more 1ike 1 in 50. When this fact is combined with the time frame, everyone should get a good laugh.
, As of March 1983, according to PG&E Unit 1 modifications were complete.
What good'does it do to train the help af ter the job is finished?
In response to PG&E' letters on Licensing Condition i 1 (Small and Large bore review with technical subjects) per : DCL-8 4-164 , DCL-8 4-16 6 and Transcript between PG&E and NRC Staf f in Bethesda, Maryland on Wed.,
Mny 9, 1984.
- 1. AS-BUILD HANGER DRAWINGS - In summary there are many aspects of the oc-builts which do not supply adequate in forma tio'n , it is PG&E's intention to get any information which is incomplete and required for the review. This appears to be acceptable to the staff. It impossible for the engineer in an of fice to know if the drawing with which he has been provided is an accurate representation of what oxists in the field. There are details which can tip him off to problems. The first place to look is at the concrete pour details chowing embeds, condu.it, ground cable, and drain lines. These items could be the reason the bolt pattern is as shown on the drawing and in the case of drain lines may lower allowable bolt loads. The drawing ucually shows the type and size of anchor bolt. When large bolts (l" cnd 1 " or larger) are 'shown or the bolt embedment length is shown and the embedment requested to be installed is 6" or more. It is important to check the slab thickness because as a guide it should be twice the bolt embedment depth. The concrete drawings are necessary to determine the thickness of the slab or wall. Thin slabs and walls (those not twice the maximum bolt embedment used) may require calculations'under the pipe support loads to determine if they are safe. Three bolt problems exist which have not been addressed in the calculations. 1) No review has been made checking the embedment
}
Page 6 of [ s
o , e length to slab or wall thickness for ef fects on bolt interaction on the opposite f ace of the slab or wall. 2) No review has been made of the effects to the concrete member subjected to a large load which is indicated by the use of the larger bolts and thru-bolts from a single support. 3) Similarly, no review has been made of the effects to the concrete member from multiple support loads.
Members shown intersecting other members at angles other than what appear on the drawing to be square should be a clue that fillet weld may not have full effective throat. This depen'sd on whether the member intersects another at an angle les,s that 60'.
Dimensions or and included angle should be requested from the field in any case where the drawing appears to show members intersecting at other than 90'.
Another item which should be included, was admitted by Mr.
Tressler on page 215 of the May 9 transcript. "I guess another case is I remember seeing a DP where the field asked the engineer of three threads of nut engagement was acceptable on this support. I don' t know for sure what the answer was, but if the answer was yes, we don' t as-built thread engagement." This problem requires that eac" y support be checked for a full nut thread engagement. The engineers performing the review can not verify the adequacy of the bolted connection with out knowing the thread engagement. Will the staff require that thread engagement be as-built knowing that some bolts exist with only three (3) thread engagement?
Other things to watch for are in accurate or insufficient detail. At times items will be shown but not in enough detail for an engineer to perform an analysis or evaluate the ef fects through the Page 7 of 2/WPage &s
use of good engineering judgement. I have seen supports which looked to be quite strong on the drawing but upon a closer examination in the field were found to be completely free to slide in the direction of pipe thermal movement. Another example that comes to mind was a plate attachment near the edge of a wall. The original drawing indicated that the plate was supported under its' full length by concrete but af ter a field trip, it was found to be cantilevered out over the edge so that a large bending stress was placed in the plate. I should note that the anchor bolts were over stressed when considering W=
cantilever. ,
one form of information, which is available and could be used to provide the missing kinds of information on problems which Eust be reviewed can be provided by including the PSTDC forms or Quick Fix sheets with the As-Built Drawings. Hard to find information relevant to the design review process is f requently shown on the PSTDC form as the reason for a modification. It . is true that the information was obtained thru " trial and error," but that process of finding necessary information for review might prove more efficient than trying to locate the applicable concrete pour drawing which would ,
cnly provide approximate information. Construction drawings are
. not very accurate in many ways. For example, the slab dimensions are usually to a quarter inch, but the location of reinforcing steel at best is only good for a rough spacing requirement. Accuracy for the location of items like drain lines and ground wire is none existent.
! Typically drains lines are supposed to be placed near the center of a
( slab but while in the PSTDC group, I had to wr'ite a Quick Fix when a i
construction crew drilled through a drain line when installing anchor j Page 8 of MPages Y
bolts. The location and depth can be determined with more accuracy f
when checking the anchor bolts for reduced allowable load reductions <
.from the . Quick Fix form than from any concrete drawing, other important information might be that the slab or wall is only 10" when it was. supposed to be 12". .
~
Beyond the fact that the Quick Fix sheets show critical information which should be reviewed, there is the necessity to have ,
engineering review these documents and give their approval of the i installed change as is the standard procedure when a design change request is made. The only difference here is the, issuer didn't perform any calculations. Thus requiring both the origination of :
\
some. calculation .or other basis for acceptance from the applicable '
engineering department. Without this act, we will uproot the i foundation which has been laid for a good QA pr'ogram. '
The review of the PSTDC 's and the DP 's has no formal ;
procedure by which to work. PG&E doesn' t appear to be committec to reviewing either of these two critical programs. In the May 9 [
l transcript on page 204 Tressler states that 2,120 DP's were revieweo, P we found 429 related to piping or pipe supports and out of these 55 were .
found to contain design information which applied to pipe or pipe '
supports. What were all the rest written against structural steel, concrete or what? If there was were 55 DP's which contained design - i information, why are' t the others being reviewed by the appropriate '
design departments? In the case o.f PSTDC's , out of approximately i
15,000 only 1100 have been reviewed of which only 20 small and 20 large l bore were selected as having extensive modifications made by the PSTDC l group.
Here the staff was insistent that all be reviewed. The Page 9 of e
M detail of this review is not clear nor is there a clear purpose behinri it. '
All of the examples used are hardware problems that have been confirmed in the field. Our engineering calculations did not take them into account because they weren't on the drawing. The effect on public safety is unknown, but could be devastating.
Problems also loom over the NEW staff corrective action programs to verify the calculations but not the hardware. The drawings are still incomplete, so significant problems won ' t be considere'd. The conditions of the hardware must first be confirmed,,before we can produce reliable engineering conclusions.
- 2. LOAD DETERMINATION -
PG&E's response to the staff has been
- incomplete.
To adequately protect public safety through conservative
- analysis, the following factors must be considered. In order to begin any calculation or review process t' he first step should be to decide how we will determine the loads to be used. This is important to provide uniformity in load detarmination and consistency of application to the structure which is relevant for both small and .
large bore. Due to the fact that less than all possible combinations of loads were actually run in the analysis, I feel that some procedure is required to be followed to ensure that the maximum possible load combination has been analyzed. This problem was one which existed during our review work, which I felt was f ar too vague and thus allowed failing structures to pass. (See Attachment 3a,b,c )
Attachment 3a is the form supplied by management for the pipe support engineers to compute the loads for their calculations.
Page 10 of [ a s
_;e ..
I have added what I believe to be clarification letters to this fo-m which were not part of the original, or were used by only a few other engineers.- I added the (A) or (B) letters to the load cases on form
~
' Though it may not be obvious the dif ference between lA and 1B, 2A
~
3 a'.-
a'nd 2B ," e tc . is quite siinple. All cases with an (A) are the most positive value that can be determined using the formulas and notes on l ~
f the bottom of the form and all cases with a (B) are the most negative I
values, also based on the formulas and notes on the bottom of the form.
Most of the time I only used this form for the most simple support, ie.
one (1) pipe, 2 directional restraint. For this type support and using this form, four (4) combinations are possible for Load case 1,
- not just the tw'o (2) which are on the sheet by the 1A or 1B,but also the .
diagonal cases.
Not all OPEG eng.ineers were aware that pipe loads are u dynamic and each direction should be assumed to be independent of the others unless a very detailed analysis has been performea demonstrat-ing otherwise. At Diablo the practice was to also allow each engineer '
to perform his analysis wit'h whatever he was able to get a checker to also accept. In sum, the new review must have procedures directing .
analysis from all relevant directions. The current methodology does not guarantee this necessary analysis.
Forms 3b and 3c have been included in Attachment 3 along with a discussion of their development. They are intended to be used as an example of the kinds of detail which should be included in any Procedures on load determination, as well as examples of changes to the forms which should be made so that all engineers are consistent in the load calculations. The types of supports addressed in this Page 11 of N $
ages
a , .-
discussion are the most difficult and are the ones which are performed using STRUDL. These are being reviewed because of errors which were found by the staff.
When performing an analysis requiring the use of STRUDL on a simple support as described above, I usually ran four load cases to guarantee that I obtained the maximum stresses. For a gang (a number of individually numbered supports welded together.at one or more points) support, it was not possible to run all possible combinations.
Gang supports are the most difficult supports to analyze. They require the best analysis and the most experienced engineers are assigned to them.
For these suppports, I developed two (2) forms which caused me to spend a little more time in determining the load case I would analyze but reduceci the number of combinations from each pipe to only one (1). Actually if anchor bolts were involved, there would be two (2). However two individual runs could be made, one to check stresses for f rame members, welds, component hardware etc., and one to perform the base plate and anchor bolt calculations. I also wrote (on my own time) and had a co-engineer check a program for an HP41C to perform the -
combinations required in the flow chart on the top of Attachment 3b.
This program provided the same results for load cases (A) or (B) as explained earlier. The reason for my doing all this was to keep up with the production requirements since I was usually given gang supports and multi-pipe supports (support restraining multiple lines but only assigned one support number) to analyze. Production credit was based on support numbers not number of pipe restrained. When I was laid of f, I was and had been working on extremely large gang and Page 12 of 88[
Pages
multi-support structures since approximately the 1st of March 1983 except while in the PSTDC group. Two which I had worked on just before going to the PSTDC group in June were reassigned to me. The structural steel loads had been sent to the city for approval and had
, come back rejected. . The. pipe support loads were causing too much torsional load on some fairly large W14's (structural steel) which were desigr.ed as pined end beams attached with angle clips.
I am not proposing that my forms be used, but I hope the NRC will see the necessity for procedures to ensure that all engineers can
+
consistently obtain the same load values t.o be .used in the calculations. That control still does not exist in the PG&E/ STAFF program. The loophole means that the same deficiencies could repeat themselves, which would disqualify the results of the current corrective action.
- 3. STRUDL MODEL - Several details should be addressed which are critical to obtaining consistent results by all engineers performing the review. To date, they have not been covered by the PG&E/ STAFF responses.
Guidelines should be defined for accuracy of cimensions, ,
both of members and eccentric connections. The use of the word significant as a standard by which to build a nuclear plant has no -
reasonable basis, nor can a definition be given to the word such that all the engineers could apply it consistently - one to another. A standard value of (+ or -) inch would ensure that a consistent level of accuracy is maintained. On other projects which I worked, the standard was (+ or -) h inch. Procedures should be more specific and beyond misinterpretation.
N Page 13 of Pages
e '. .
Second, project specific instructions should be suppliad on the determination of Beta angles. There wers many non-controlled documents on how to determine the Beta angles from all over the country within OPEG. This was not any dif ferent from the San Francisco office, as employees there co'nfirmed to me. Due to the number'of angle members in the small bore supports, I feel it imperative that tables be given to the engineers which have been approved by '
management indicating the correct Beta angle depending on position of the member.
Third, instructions should be given the engineers on when to use joint releases. This should include sketches of the detail and how it should be modeled. There should be a free atmosphere within which questions may be asked both for clarification and to question the validity of the proposed instruction or any technical aspect of the review program.
Finally, guidance on modeling members which have had a part of the cross section cutout for any number of reasons should be provided as examples of good modeling practice. These could range from a hole through the web of a wide flange (WF) to the hole member having been cutout and displaced in one direction or another. This problem with notches and cutouts are stress related. Typically, details are not paid very much attention. When a member has had its' cross section reduced, stresses are intensified in the reduced zone and unless an engineer considers the transferral of stresses f rom one side to the other - an overstress can exist.
On balance the f ailure to provide methodological controls such as described above, means that there is no guarantee that the Page 14 of Pages
e; % .
PG&E/ staff corrective action program will catch all significan*
pro.blems. These factors must be considered.
- 4. STRUDL STRESS SUMMARIES - These represent another problem not adequately covered in the current corrective action plan. A t Diablo, each engineer'was allowed to check as little as his checker would agree to . accept. Even the work at Zimmer was better in this respect.
Enclosed . is a cop'y of the CHECKLIST FOR STRUDL FRAME ANALYSIS (Attachment 4a) which was used at Diablo by OPEG. Also attached is a copy of a form which was used at Quadrex Corporation for checking the stress combinations of members which were analyzed by.Strudl. (See Attachment 4b) I adapted the basic form to the Diablo Title for my use at Diablo, but was not allowed to use it. This form was included as an example of how to properly combine the torsional shear stresses (see line 18) and also how to combine the warping normal bending' stresses with the axial and bending stresses (see line 21) . For the record, I would like to state that this form was part of the calculation package for all small bore supports which were performed by Quadrex on Zimmer Nuclear Plant. I would also like to say that a fortran program based on the same technical literature as the Bethlehem Steel design guide
- on TORSION ANALYSIS of Rolled Steel Sections was used to compute the warping shear and warping normal stresses (lines 7 and 8 of attachment 4b) If there is any remaining question whether my challenge to the loophole is reasonable, in 1980 at the Gaithersburg, MD office of Bechtel, we used this same program. It was on Bechtel's own system.
The sub]ect of warping stresses is not new either to the engineers at Bechtel or PG&E.
Another point should be raised in discussing Strudl stress Page15ofgPages Mk
a , .
summarles, the use of Strudl input forms at OPEG. These forms had all load combinations already written out. When I reviewed tha combinations in detail to make sure all load cases were being combined which ef fected stresses in the Load Combination statements, I found that'the of the structure was not being included in the stress results.
"Self-weight" is the load of the structure itself. In some cases, the stresses on members from "self-weight" load alone will be suf ficient to fail the support, and in all cases this component of. load is important in evaluating the acceptance or f ailure of that support.
This was not obvious since in load case five (5) or so there were three (3) load cases written to provide the self weight in the X, Y, and Z directions. These were being run and combined with the tributary load (also ef fective weight of pipe system in each of the respective directions) for output as the displacement by which we were to check the stif fness of the structure. Apparently, the writer of this form failed to add the necessary load cases in the load combination statements which are combined for stress output. After this discovery, I wrote my own input forms.
I would suggest that the NRC have an in depth look at this -
form if it is being used in the review work. The public record does not demonstrate that this problem has been corrected. If there is any question about the significance of this error, some supports will f ail under their self-weight without any pipe load being applied.
Another point on this subject must be addressed, but is not clear from the public record. Why wasn' t the deflection required to be checked? On all other plants with which I am familiar, a displacement limit was imposed at the point of the pipe attachment.
Page 16 of
$6 Pages
This was a secondary check beyona the stif fness requirement. It is important.to verify the displacement under load for if the support moves in some cases too far the pipe may be overstressed. It is
,possible to have a support which meets the stif fness requirements and still will have large amounts of displacement.
- 5. TORSIONAL WARPING NORMAL AND WARPING SHEAR STRESS - In PG&E's letter: DCL-84-164 Enclosure 7 on page 2 PG&E states "There are three considerations in the pipe support design at Diablo Canyon that tend to minimize the significance of the warping phenomena:
o The predominant use of wide flange sections rather than "I" sections or other sectional shapes having a lesser capacity to restrain torsional loads, o The pipe supports are designed to use standard siz[a members and a stiffness criteria that, in most cases, assure that the member stresses will not be the critical factor in the strength of the support.
o small bore supports typically use angle or square tube section material that are not subject to warping." (Emphasis added.)
The statements by PG&E are ridiculous. Wide flange sec-tions may be used predominantly in large bore, but even if they are that does not preclude many from being overstressed when torsional warping normal and warping shear stresses are added to the existing bending and shear stresses. This is especially true in large bore The end connections supports where very large loads are possible.
, are very important in how much torsion a member can safely carry. It is true that wide flanges have more torsional stif fness than an "I" beam,'but it is equally true that a square, rectangular or circular tube has far more torsional stiffness than a wide flange.
The second PG&E statement concerning the stif fness cri-teria may be somewhat relevant for the most simple support, i.e., a simple cantilever, but even here a counteracting force is at work. As mentioned in the paragraph above, the end condition has a lot to ao Page 17 of $&l$
Pages
with how much torsional loading a member can take. ,
A beam which ia .
Fixed-Fixed will carry more torsional loading than one which is Fixed-Pined, P ined-Pined, or Fixed-Free, providing only end conditions vary. . Again the failure to consider all possible forces could be the differencebetweenpassingandfailingthesuhport.
C The last statement has to be a misprint.
I can' t believe that PG&E would allow such an obvious f alsehood. I believe almost any competent engineer can tell you that any material or shape will warp if loaded. In reply to the angle and tube shape comments of PG&E, I would like to quote several lines from a book published by the Lincoln Arc Welding Foundation " Design of Welded Structures." I selected this document because many engineers have one and use it regularly.
From Section 2.10-3 " Designing for Torsional Loading" "The solid or tubular round closed section is best for
, torsional loading since the shear stresses are uniform around the circumference of the member.
Next to a tubular section, the best section for resisting torsion is a closed square or rectangular tubular section.
(Skip a paragraph)
The poorest sections for torsional loading are open sections, flat plates, angle sections, channel sections, Z-bar sections, T-bar sections, I-beam sections, and tubular sections which have a slot." (Emphasis added.) <
The truth of PG&E's amazing assertion should be tested i
fully, both for the validity of this theory and its effect on the design. I feel that when the evidence is laid out on the table that PG&E will have to admit their incompetence.
- 6. ANGLE MEMBERS - In PG&E letter No: DCL-84-164 Enclosure 7 on page 3 paragraph (b), Dif ferences Between AISC Code and Project Criteria, PG&E state "The so called 'dif ferences' between AISC and the Project criteria using the Australian data, references 1,2, and 3, with respect to allowable stresses of angle sections in bending do not really exist."
While I have not had very much time to fully analysis the 4
_ ~ _ - _ - _ _. - - . _
Page 18 of $$
Page &s
9 o , .
Australian papers, I have studied them suf ficiently to conclude that they are not applicable for type of loading to which they are being applied by PG&E. I would like to quote several lines from the Australian paper " Laterally Unsuppor ted Angles With Equal and Unequal Legs" Section 4. LOADING CASES. " Note that the location of the moment on the cross-section is not critical. However point loadings which are not applied through the shear centre will cause additional twisting (Section 12) ." In Section 12. LOADS NOT THROUGH THE SHEAR CENTRE, additional formulas are given for the additional twist and the additional stresses due to twisting. On the bottom of Section 12, "A more exact and comprehensive solution to this problem can be found in (21) . " Reference 21 is to an article entitled " Deformations of Geometrically Imperfect Beams" written by N. S. Trahair which was printed in Proc. ASCE, 95 (ST7) JULY 1969, PP.1475-1496.
I am sure that the Australian study was not intended to be
~
used by the author in application to the types of end connections which we find in nuclear power plants. In Section 1. INTRODUCTION second paragraph, a discussion is provided as to the use of angles in structures such as transmission towers. The last line of this paragraph says "Even here, however, the underlying research has frequently been highly empirical with strut load capacities given for each member size under practical field conditions." From the qualifiers in the text, these studies were made strictly for expansion of knowledge in the area of transmission tower design.
I don' t believe the Australian papers should be used in the design of nuclear power plants or the construction of any structure which would place the lives of people in danger for the mm._tu s- $2 $
I Y
following reasons: 1) The design of Class 1 nuclear pipe supports requires a much more conservative analysis than do transmission towers. 2) The joint details which exist in constructing a transmis-sica tower are easier to control than they are in pipe s.upports. Thus through a good joint design it is possible to apply loads to the shear centre avoiding the additional twisting stresses inherent to a nuclear plant, discussed in.the first paragraph. 3) The Bending moments to which the angles in the Australian papers are subjected are limited to those caused by wind combined with a small amount of axial load, neither of which require the same degree of evaluation as do the safety-related systems in a nuclear power plant.
On page 33 of PG&E letter: DCL-84-164 Enclosure 7 PG&E states "It should also be pointed out that the 18 pipe supports identified in the DR 83-042-S as discrepant have been reviewed. All of the angle baam spans are found within the Project Design Criteria."
This statement is f alse and misleading. This is not the DR which I signed and submitted to Leo Mangoba on 10/5/83. That document did not contain any pipe support numbers which I felt were discrepant.
It did contain seven (7) pages which I hao copied out of the AISC. The pages which I included were marked up to show the sections which detailed the problem. I did give a list of supports to Jeff Van -
Klompenburg several days af ter I first submitted the DR in preliminary form back in 8/83. This was after I submitten the DR and it was returned with a note that I should provide a support number which was not within criteria as an example. In stead of one support, I submitted a list in rough form which contained approximately 100 Unit 1 supports which were 'not in compliance the AISC code section 1.5.1.4.6b or more specifically 76.0 b g / 6 ywhich for 2" angles = .
25.3", and for 3" angles = 38", and for 4" angles = 50.68". I hav'e no Page20of[Pae
I' .
knowledge as to who added the 18 supports to the DR which I submitted to management on 10/5/83.
- 7. BASE PLATE II - There are several points which should not be over looked in the Base plate calculations beyond those discussed by Mr.
'Hartzman and Mr. Manoly. Base Plate II must as a finite element C
program, in order to give accurate results have certain criteria met.
-These are: 1) The elements must be within a certain aspect ratio, i.e.,
height, leng th, width. Most finite element programs specify that the aspect ratio be not more than 1 to 4. This requirement if applied to a inch thick plate, would require that the length and wi.dth dimensions not exceed 2 inches. I have been informed, that the analyses which were performed were not within the aspect ratio which is required for accurate results. 2) Some plates are not one piece of steel and have been welded together with partial penetration welos. The model should account for the reduced thickness where welded, so that the plate stress calculations will be accurate. Also, in the case of shell type anchor bolts the hole diameter should be used per the manu-f acturers requirements in determining the 10D distance. In the past, this was done incorrectly as the hole diameter was not supplied by management. This information is now supplied to the engineers in the ESD 223 and the calculation should be corrected during the present review. Failure to correct this could result in a safety-related support failing.
On balance the above methodological holes and flaws each must be answered because they incependently identify conditions any which if lef t uncorrected could significantly threaten public safety durin,g full power operation. Even if the plant could operate without mdn$ NN _ --
~~
_ ; e .',. .
. f analysis of some of these issues, the combined effect of the missina factors is decisive. To date the staff has not required these questions to be satisfactorily answereo. Until that occurs, the current corrective action may just amount to another way to endorse portions of the status quo that would be indefensible if publicly scrutinized.
I have read the above -[32 /df
,page statement and it is true and accurate and complete to the best of my knowledge and belief Charles C. S tokes PE.
Subscribed and sworn to before me this l ** st day of June, 1984.
- Y '
~
--.___-= ---- ll , i. . / l lt , . !.
CFFICIAL $EAL Notary Public in and for
- S'3:rt 'r. 'U . f, FO N!A g the County of San Luis I
v.n u23 C21tPO COUffTY Obispo, State of uy cm. bs rei
, Mer 9,19e6 Californla C
e G
9 Page 22 of Pages
,~.,n, . -
A77AC/h1EN7" .f
. I R= >%a" l _
R. a/4a'
,r
' b'%d ef [ R 914s*
R= '4 " R= '$
y)$UAL
' CRAlk on bl%*- Z END- 61 DES
-ENb h
, 4 Aivp 2*//ca 2 ENDS IN R= S/s" g R %"s RADIMS 3 M
'ggggg ,R=*l4s'
\ g= *%s" '
ENb~y~ R.'%E J 5ND R ~ >NS IDS tr16AfuAtb '
If $ = 2 b
- Y
- Y."
M/FisLET gad 6ES Z
= 24/4a n NO' G glDE S bitts)DE fo 148/7 ..
og M 'T3 3 X5 x Y4'
'I2/B E Spy R. *1/48,*
rg ,ENO -A ""
e- }
, R * *%s-1 ,, $
.x r . R = 'llt. R si , p.
V/ s uAL. =, a, 6/AM , ,
visual mAJo os) cF
$I O C.S I , 2. E ENo -A-euo +
sy,. ,, 3 .
z v,m
+z %" on 4 = x*
' }. R = YA ' R'y
$ g, g }gg
- <5 g?'
R:*2/4s" L R= */46 n ,,p t
jgg
- 9g4
..~...
bS W Groove Welds /47
.n .
S(El j (
E]
, l t
S,M '
s Mr '
S
- Radius from point of tangency Des; red weld to too of member SVmD0' E = Effective tnroat (A) Flare-V-groove weld with specified partial joint penetration
--* b ~ .
/ A bo -
h
.- h AM 4 h 8
5 . , .
(
t (if a ! r g b f @ S(El. '(
[
~
- a a Desired weld Symbol (B) Flare bevel-groove weld with specified partial joint penetration and fillet weld S
S(eld [
t
+ .
I i
}+
l i .
l E t . Symbol
- I ; Desired wefd
'h
, . (C) Single flare V-groove weld Fig.32-Application or flare bevel- and flare V groove weld symbols 6 .
3 48/WELDf.sc SYMBOLS
/
+ +
s(E)l ( -- '
f'
- A
)
f -
Q t
~
. + +
- "d*## Symboi (D) Single flare-bevel-groove werd E
- SIEl l (*
S(E) l (
4-l t W-i -
i i .
e 3 - ,
- - Desired weld Symbol ~
('- '
(E) Couble-flare-bevei groove weld i S(E)l (
/
X -
w *
'g S -
Desired weld
- Symbot
- E - EIfective throat (F) Two sing!e flare bevel groove weidS S - Radius of bar "Ellective throat SIE) ) (
\
g \ S i
I t/ t
$e Radius from point For complete of tangency to top - penetratiors weld of member S<E e<, pf *
+f Desired wefd , Symbol f
- e (G) Flare V groove weld with complete penetration
~-
e . -
?:
n 4mgg Pacmc cas ao Lacmic c waav y "" "*
m GENERAL C O M P U T A TI C N. S H E E T "L J
1 LOADING CASES FOR -iGR. No.
~~~ "
OfABIO CANYON UNI -
.< A. - ..
tiANGER No. REV -
n m LINENo ** '
ANALYSIS Naf REV -
DATA POINT > '
LOAD FX FY F4 MK MY MZ CASES (/f.) (/6.) ( /6 ) (IN-/b. ) (IN- /S. ) (IN-l$. )
- l. A 4
l_JS -
- 9. A -
(v
, , ') ~ p8
~ '
9 3A f .
I <4 i h 1
, 4A l
l- c f
5A 5_1 CasEtcAcus um t- k b -
iTH*CLtFL L FRCUCN LCACS REE CCM-M9(4.2H) 2 A r W OL.+ FV+ RNUT+(D EE SAM ) f/2 2 2.THE CIRECUCN CF T)E T}ERMAL FCRCE SMALL NCT SEUSED TO RCUCE THE MAGN.0LCE 9 8 ,m+CL4FW RWT-(O E 2 SN.SI/2 CFM CCM9E m A
-) s 1 8 r%Ct.tFW RWT,(00 EhSAM2)V2 ML FWf6CT-(COE75AM )I/2 2 IFOR tcAD CAEES 3 #c 5 USE THE GRs,, :s CF TH CR THA.
- 4. CASE 5'FCR ANCHCR ECC3 CtLY.
4 A i OL,(HCSESAM 2 )I/2 4 8 e Cts.(HCS, I SAM *)I/2 5 A wot,-(HOS3 S/M)V2 .
- 5. & ?TM-b(dSAM 2)f/2 -
' " * " sHa m PACIFIC GAS AND ELECTRIC COMP AN MEty No.
GEN E R A L COM P UT ATIO N 'f"'~
u.c Arnon A a s s ,-
suesser NW YA ~ Tso.No.- 21.211.*
. DIABLO CANICN UNIT NO. .
~
..... ;... e e3r . . . . . . . . ..........
Lonos l CAO C-A SE DETE4MMM71oM ~~
fe- M-9 D &LobAl SEQ . JEv.,
LINE No. ~
D Loeau. su or ANALYSIS No. '
DATA lb/NT REV. DATE / /
Add 6LE447/oA/ . O PEAK . D ME lol UN/75 -NaRCE"
~~
VALUE.C D 15 H 2 O M-4 o d4fC. Mot??EA/T ?
53f Y iL._.
Fl.ow C. HART :T'Hc!);
=
DL._+,TH = . (A) -
FbR i DL(+)
HANDLINlr J,z,s,4 s No IDE !
= IDL.I
- (6)*
3 l f y (_) ;
DL+ Til v
- g2r y 7
- lDL l - l TH = (6)l.= = (A)
Lonos s
;_f.y[#);
- lTH l '- I DL l ' (Al =
lDLC-) l I' , - l DL I = (8):
l3 2,3,+,5 l TN (-) l - IDL l = (A)*
Y 12X 5,3s .Y Y
Y E F * --
Ml-lDLl = @p LoAo Cases l Fx F'Y FE l MY MY M2 I.a.. ( h) 4 IFLI / / /\
l.b. (8) - IFLI / / fl Za. (A) + IFvl+1Rvor! + (DE) // /l ZA, (8) - l Fyl -lRvor! - (D E) / //l -
l .M. 3.a. (A) + IFV I + IRvoTl + (bDE) f f f e J y 3.b. (6) - !FVl -%4 -(DDE) / ~
/ /
%- %- 4.a. (A) + ( H os) / / /\
L l , .X. .y- 46. (8)- ( Hos) L/ / / !
l # sa. /A3 + ( H os) // fi
- 5.6. (8) - G H-os) / / fl
.s.sresL (DE) = (DE *+ $hthbe )'l** '
FL -(O )oR (&)hAMX .(o7 TEFLM/ )
(DDE) = (_ DDE'+ $hN\cs") N2- IF JM OVEMENT > . ulo25~/N.
2'-
l h05 ) = (. ffoS*t SAAlbs )
Y USE l} OR THA)i?MK JN FIow Ch%e7' 70 08rA/N (A.)l8) '
-X~)(- USE TN = O 5E220 TN/S CASE ON/y
,, e, .. . o * ~^ - ~~
- " ' " SMst? M9. OF SMfET3 PACIFIC GAS AND ELCCTRIC CCMPA Joe -*
GENER AL COMPUT ATION "'E.
~
Loc a tt oN M f d' I sussect HGR . NO. - ELEV.
- DIABLO CANION UNIT NO.
o.or s, 9.ve
/ / enrexeoev_ ....ev ro a v MAI. LOAD CT3INATIM CAIC. PER M-9 3 0 GL0 SAL IDS. Seo asy LINE NO. _~ , ,,
O LOCAL LDS. SH CF ANALYSIS NO. DATA POINT REV. D TE /- /
ACCEIERATIm . O PEAK . .. O ME 101... UNIT F mCE
' ~ -
1 VAIDES O l' HZ5 O M-40 CAIC. McHENr
~
LOAD C m BINATI N FX FY FZ MX MY MZ
- 1. TH+DbFL $
'~2. TH+D hF7+Rvot '
A 2'
~
+(DE2[SAMDE)t~~' 5
- 3. TH or THA +DbF%Rvot A 2
+(DDE2+SAMDDE)Y B 2
- 4. Db(HCS2 + SAMHOS )t s
- 5. ANCHCR BOLT CALC. A TH or THA + CASE 4 8
%- 3 #4 *) D 6.ADJ. FACTCR K= CNB 1 or 2 (LAR.) A IF K > 1.33 ADJ. LOAD =(3 or 4) .
k*
TF K ( 1.33 ADJ. LOAD = li oi332 8
X-Acc* YACC: E-AC(,f 8.IFADJ. DE L ) 1.33FACTm ADJ.ACC. L=ACC. DE(DDE ADT. or AD7.,
HOS)DCr DE D F.
AbI 1.33 Acc: Acc: Acc. one CD =-
IF L ( l'.33 ADJ. ACC.= DE HoS HoS .40s
- 9. STRUCr. W. x ADJ. ACC. (8)
- 10. TRIB. or EFF. W.
( 17. FRAME DESIGX A LOAD = 6 + 9 + 7 -Y ou t y 8
- 12. FREQ. DESIGN A'
< LOAD = 7.+ 10 6 l -
l 13. ANCH3 BOLT DESIGr A LCAD =(5 on 6) mu. 4 9 + 7-Yo.vLy 8 FL = FRICTI M FRIC. CGFF. = .3 STEEL ; .07 TEFLCN USE ALIC4. STRESS FCR CASE 1 or 2 Fb 1-DIR. FL. = 23.9 ksi l AISC STEEL - Fa, Fb, Fb 2-DIR. PL = 19.1 kai Fv = 12.8 ksi. WEID = 18.0 kai t
L
l a.. . !
,, Sheet of PACI FIC GAS AND El.ECTRIC COMPANY DIABLO CANYON NUCLEAR POWER PLANT UNIT HANGER NO. i MADE BY: DATE
. t CALC. SEQ. No.. REY. ;
s -- . ~
CHECEID EY: DATE * ,
- l APPROVED. BT .
DATE k / /W/ MEA /7 7j4. -
i i
.._..~...i CHECELIST FOR STRUDL FRAME ANALYSIS ,
i
~
- 1. MDtBEIS YES No N/A PAGE ,
DOES MIS rRAME INCLUDE ANGtES
~
'O -
- . . i ANGLE PROPERTIES INCORPORATE PRINCIPAL AXES ~ C ' ~
{
ANGLE PROPERTIES INCORPORATE ORIROGONAL AIES O I L/t FOR ANGLES RAS BEEN REVIEWED .
O '
SLENDERNESS RATIOS RAVE BEEN REVIEWED f 1
LENCIES MODIFIID FOR COMPRESSION MEMBERS WIH INTERMEDIATE JOINTS O -
( TORSION ME3ERS ARE ACCEPTABLE ET ENGINEERING JUDCEMENT, C
t OR FOR CALCULATIONS SEE
- 2. NATURAL TREQUEN,CI DIRECTION f ALI4WA31Z n .
RESTRAINED UNRESTRAINED X = a cps eps 0 -
Y = cps eps C
E = cps cps O
i*
, e e
- n .'
- ._
- . g5*..+e 1~ l's . :- .
, ny- . . . ,4., . a ? '3 . _ w , ;
y: [ :- . - l . ,- :y ' 6 *. ; : > ? k , , : j'. , ::. in _ - . y ,
, , .. vo?'s* s P A C: :! O 31 1 Am. ~ '. 2 0T.? ! - - .. : s .'
G E N 2 'l A C C .A ? U T A T C HGR. NO. " .
Rev.
._,,.. SEQ. NO.
DIABLO CANYCN UNIT NO. -
ISO. NO.-
. . - _ . __. e m . / /' cc,.c s, o....u :.__
- s. .
Check comoined stress and deflection -
Uni ts: O Insta11eo support support Loading condition Lbs., In. O Not installed support O A/B C/D OT O c/D OT_
\.33 e **
Ref. Description -
Stress bending MY/SM (1)
Stress cenoing MZ/5M '
(2) .
Stress axial P/A (3)
Force local FY (4)
A rce local FZ (51 I L -
Soan lenoth L (6)
Stress Tors. & warp. sheaf (7) l Shear 1 Waroina normal stress (8)
Column ratio X (9) I
( Radius of gyr. r g(10).
Shear area. Local Ay (11) ,
l Shear area. Local Az ! (12)
(9)x(6) (13)
AISC go) -
Allow. comer. stress Fa (14) 4 Allow bend. stress Fby (15)
Allow cend. stress Fbz (so) c Allow shear stress Fv (17)I Shear + (7) (18)
)+ )
Compr. (3)
Ratio TIT) (IS)
I 9.f 1& to17 < .6, (20) ng (19) (21)
Enter c Allow * .
O.K. if (21,4 o Allow
~
a Max. Max def1ection at joint in the restrained direction is
.< . ' , ok.
If (19)> .15 resolve on separate calc. sneet using Aisc formusas i.o-ia a a.
If (20)> .6, prorate '(15) and (16) per AISC section 1.10.7
1 i 16 cog Ul',
~
To: Rep. Morris Udall, Chairman er House Interior and Insular Affairs Committee Washington, D.C. 20515 Attn: Dr. Henry Myers 4 17.' 21 go :ag From: Harold Hudson .-
Date: 6-9-84
Subject:
Deficiencies in the Nuclear Regulatory Commission Inspection to determine the extent to which Pullman and PG&E had implemented an adequate audit program during startup phase of Pullman work at Diablo Canyon (Report Nos. 50-275/84-16 and 50-323/84-06).
The special, unannounced NRC inspection on April 2-4, 1984, to examine audit records of Pullman Power welding activities, performed during the period of August 1971 through December 1973, does not provide an accurate assessment of the PG&E and Pullman audit programs. There are a number of deficiencies in the report based mainly on the apparent omission of information.
The NRC report concluded: " Based up6n a review of PG&E and Pullman audits it appears that audits of Pullman welding activities were thorough and conducted in accordance with the Pullman QA
-Program during the period of August 1971 through December 1973.
Based upon the above reviews the inspector condludes that:
- a. the Pullman audit program met the intent of Safety Guide 28 (June 1972) and ANSI N45.2-1971, in effect during that time period, and, therefore, the intent of Appendix B,
- b. the audits appeared to be of reasonable competence and quality, and
- c. based upon a sampling of corrective actions it appears that findings were followed up and resolved ina responsible fashion."
There is documented evidence which I believe will show that i the NRC conclusions are not an accurate assessment of the Pullman and PG&E Audit Programs during the 1971 to 1973 period.
- 1. The NRC Report states: "The inspector read through and surveyed the above audits to develop a sense of the audit competency and quality. The audits were performed by Pul-1 man and PG&E to determine compliance with the Pullman Quality Assurance Program (KFP's). The inspector noted that applicable procedures of the QA program related to welding activities were audited:
KFP-8"ProcessPlapningandControl",NON-KFP-12 " Control of Filler Metal", KFP-13 "Poytweld Heat Treatment", and KFP-15 " Welding Qualification". Further, i
a -
i the above audits deomonstrated'that Field Process Sheets (travelers) were audited for compliance to KFP-8 (for each field weld the process sheet lists operations such as fit up, weld completion, NDE, and designated holdpoints for QC l and the Authorized Nuclear Inspector's inspection and ap-
- proval) . "
What the NRC Inspector apparently failed to recognize was that the Pullman QA program related to welding activities, which were audited by Pullman'and PG&E, and surveyed by the NRC Inspector for a sense of audit competency and quality, was not applicable to Pipe Hangers (Supports) and Pipe Rupture Restraints.
- a. The NRC Inspector references a PG&E Audit #73-15 (11-29 and 12-19, 1973) but failed to recognize the significance of its findings in relation to the scope of his work. PG&E Audit #73-15 disclosed that in ;
regard to Pipe Hangers (Supports) and Pipe Rupture Restraints, M.W. Kellogg and PG&E's General l Construction Dept. departed significantly from the requirements of the Specification and PG&E's QA Manual. "MWK's Quality Assurance Program does not ;
comply with Section 4 of Specification 8711 and ;
(PG&E) Procedure PRP-4. The (PG&E) Mechanical ;
Dept's surveillance program does not comply with ,
Procedure PRC-7". PG&E Audit #73-15 reviewed MWK's l Q.A. Manual, with respect to Pipe hangers and restraints, !
for adequacy and compliance to Spec 8711 and PG&E
' QA Procedure PRP-4. Audit #73-15 concluded that MWK's approved QA Manual "does not specifically 1 l
address itself to, nor is it completely applicable to, the control of pipe hangers and restraints." -
Thus MWK wrote an " Engineering Specification,"
ESD 223. The intent of ESD 223 was to set forth procedures and instructions to the field QA inspectors, engineers and foremen implementing the policy stated in the QA Manual. Audit #73-15 concluded ESD 223 ,
established QA policy instead of providing instruc- .
l tions on how to implement the policy stated in the Manual. Audit #73-15 also concluded that "the program set forth in ESD 223 does not meet all of the requirements of Section 4 of the Spec. Deficiencies were noted in the areas of document review and control, qualification of special processes and personnel, work procurement control, receipt l
inspection of material, identification control and status of material, nonconforming material control, inspection and test records and inspection and test
! plans. Consequently the hanger and restrain QA
- l program is in violation of Procedure PRP-4.
Audit #73-15 would review the receipt, storage and
! installation of pipe hangers and rupture restraints -
l and identify numerous major discrepancies (see I attached PG&E Audit #73-15).
I l'
. Audit #73-15 audited PG&E's Resident Mechanical Engineer's surveillance system of the fabricating, furnishin'g and installing of pipe hangers and rupture restraints, which was performed by the-;
Power Plant Piping Group. PG&E Procedure MFI-2 set forth the Resident's written instructions to personnel in this Group. The audit concluded that "MFI-2 Instructions do not specifically address themselves to the surveillance of pipe hangers and restraints and are not applicable to the inspe tion
' gg)4_ '
of pipe hangers and restraints. The inspec is performing other inspections, however, these
- inspections are not documented or described in the MFI.
The bottom line is that sudits performed by Pullman based on its QA Manual (KFP's) and audits by PG&E's Mechanical Department were not applicable to Pipe Hangers and Pipe Rupture Restraints during the August 1971 to December 1973 period. The NRC Report conclusion that "it appears that audits of Pullman welding activities were thorough and conducted in accordance with the Pullman QA Program" is not an accurate assessment in regard to Pullman's Pipe Hanger and Pipe Rupture Restraint Programs. The NRC Report conclusion that " audits appear to be of reasonable competence and quality"-is not accurate in
__ regard to Pipe Hangers and Pipe Rupture Restraints.
- b. A major fallacy in the PG&E audit program in the 1971 to 1973 period was that Pipe Rupture Restraints were audited against PG&E Contract Spec. 8711.
PG&E's C.S #8833XR specified the fabrication, erection and Quality Assurance requirements for Pipe Rupture Restraints. Yet PG&E's Audit #73115 of c Rupture Restraints would be against the requirements of C.S. #8711 not C.S. #8833XR. All PG&E audits performed prior to Audit #73-15 were also against C.S. #8711. This leaves the PG&E audit program for j Pipe Rupture Restraints indeterminate during this >
time period. Again the NRC Inspector has not recop ized this deficiency in the audit program.
j dOk i 5 JA VM4 5 Q. -
- c. The NRC Report does not reference a 10-24-73 Pull-man Internal Audit, which was the first documented
- Internal Audit performed specifically on Pipe Rupture Restraints. Per C.S. #8833XR Pullman was scheduled to begin erection of Pipe Rupture Restraints in Unit I on 7-8-7'. Yet it would be 16 months before l
an Internal Audit would be performed on the program.
! This audit would note discrepancies with weld rod requisition and that some restraints did not have any rod requisitions. Again this audit would reveal
[
1 that C.S. #8711 was being used in conjunction with .
C .S. 48833XR for the installation of restraints. ,
l Yet the most significant aspect of this audit was
-that it revealed there was no Quality Assurance Manual i available for the control of installation of restraints.
Installation was controlled by a " letter approved by :
A. G. Walters on October 19, 1972."
i
~
The omission of this audit from the NRC Report and i the finding revealed by the Audit cast doubts on the l NRC conclusion "that audits of Pullmans welding i activities were thorough and conducted in accordance with the Pullman QA Program." j
- ,3
- d. The first documented Pullman Internal AuditgPipe Hangers (Supports) was performed on 9-18-73 and f is referenced in the NRC Report. The audit report states that the scope of the Audit was " adherence to the Engineering Specifications and Quality Assurance Manual." It would be identified in PG&E Audit
- 73-15 that Pullman QA Manual (KFP's) did not specifically address or was applicable to Pipe Hangers. The PG&E Audit would also identify that ESD 223 established QA policy instead of providing instructions for the installation of hangers, in
-essence, on " alternate QA program, which was not sub-
'mitted to the PG&E Quality Assurance Dept. for review _
i-and approval prior to use.
The NRC Report does not address these discrepancies but concludes that "the audits appeared to be of reasonable competence and quality."
- e. The NRC Report states "the above audits demonstrated that Field Process Sheets (travelers) were audited for compliance to KFP 8". The NRC Report is misleading in regard to Pipe Hangers (Supports) and Pipe Rupture <
Restraints. It has already been established that the l Pullman QA Manual (KFP's) was not applicable to Pipe Hangers and Pipe Rupture Restraints.
PG&E Audit #73-15, dated 11-29 and 12-19, 1973, identified:
- 1. Mangers for pipe 2-1/2" and greater:
- a. Subcontractor supplied hanger assemblies were not inspected by QA personnel.
l b. Hanger assemblies fabricated on site do not receive any in process or final l inspection.
- c. The installation of hangers received a final inspection, but no inprocess inspections.
- - d. Inspectors are using inspection forms which they consider necessary but are not controlled by the QA Program.
- e. The conditions of field welds were generally rough and irregular. Some welds specified on the drawings were not made. No discrepancy reports or other documentation authorizing the acceptance of these discrepancies were e available. !
- f. Inspectors could not furnish any explicit i acceptance criteria. !
- 2. Field run hangers for pipe less than 2-12": ;
- a. All field run pipe hangers are fabricated on site. These hangers do not receive ;
, any inprocess or final inspections.
- b. The installation of hangers receive a -
final inspection but no inprocess in-spections.
' hangers not in the location specified by the drawings, hangers not fabricated in accordance with approved drawings.
- 3. Pipe Rupture Restraints:
- a. The method of recording inspections and acceptance criteria are not set forth in an instruction. All inprocess inspections of workmanship and technique required by the AWS Code are not being performed.
- b. Some we3ders were welding materials of greater thickness than they were qualified for.
- c. Welding was not in complete accordance with the assigned weld procedures.
The Pullman and PG&E Audits of KFP 8 were obviously not applied to Pipe Supports and Pipe Rupture Restraints. Yet the NRC Inspector:does not reveal this even after he had reviewed the '
PG&E Audit #73-15. For the time period being discussed,
. Pulln wr would not identify any of the Hanger and Restraint
~ discrepancies listed above. PG&E's onsite General Construction !
QC' group would not identify any of the discrepancies. Not ,
< until October / November 1973 (the very end of the time frame
- being discussed) , when the PG&E Corporate QA Department per-formed Audit #73-15,' would the discrepancies be identified.
The NRC Report conclusion that " audits of Pullman welding activities were thorough and conducted in accordance with the Pullman QA program during the period of August 1971 through December 1973" is not an accurate assessment regarding Pipe Hangers and Rupture Restraints.
- 2. The NRC Report concludes: "The Pullman audit program met the intent of Safety Guide 28 (June 1972) and ANSI N45.2-1971, in effect during that time period, and therefore, the intent of Appendix B." There is documented evidence which apparently was not reviewed by the FiRC Inspector, which suggest the NRC conclusion about the Pullman Audit Program may not be completely accurate for the August 1971 to December 1973 time period for Piping, Pipe Supports and, Pipe. Rupture Restraints.
.a . A Pullman Corporate Interoffice Correspondence, dated 11-13-78, from the Senior QA Engineer, concerning Corporate Management Addits, states that "The Diablo Canyon project has been audited extensively only in hardware areas. The entire program has not been evaluated." The same IOC states "In the past, Pullman Power Products did not conduct audits or practices to ASME (Pullman's QA Manual is based on ASME Code Section III, 1971 edition) or 10 CFR 50, but I feel it s very essential to do so now".
- b. A PG&E Corporate Management Audit #80422, dated 6-13-78, indicates that the ANSI standards are not applicable to Pullmans QA Program. Audit #80422 states " ANSI N45.2 states in its forward that it is not applicable to work performed in accordance with the Code (ASME Section III).
- c. PG&E's C.S. #8711 and C.S. #8833XR to Pullman makes no reference to or commitment to Safety Guide 28 ,
ANSI N45.2-1971 or 10 CFR 50 Appendix B.
- d. PG&E A6dit #80422 identified that the Pullman Quality Assurance Program is not adequately defined. The ASME Code Section III, 1971, requires that the Quality Assurance Program be documented in detail in a manual consisting of written policies, procedures, and instructions. Pullman Corporate Procedure No. XVIII-I, is presently being used for the performance of manage-ment audits of field activities. This procedure implemented QA requirements of the contract but were not identified as part of the program and revisions were not controlled by the program.
- e. PG&E Audit #80422 indicated that PG&E C.S.#8711 and the 1971 ASME Code required a comprehensive system
t
, of planned and periodic audits to be carried out to !
assure compliance with all aspects of the QA Program. (
~
Audit #80422 identified:
" Procedure KFP-18 states in its scope that it i establishes such a system. However, two types of ;
audits, management audits and internal audits, are described. The procedure does not establish the scope of.either. type of audit and no detailed schedule has been developed to show that all aspects of the program are being audited. Further-more, audit records at the site do not indicate ;
that all aspects of the program are being audited. l Records do not indicate that management audits i have been performed on pipe support work (and pipe .
. rupture restraints) ." An unofficial, unapproved 14c( '
internal audit schedule exists, but it has not been followed consistently and for ESD's appear on th schedule". (
In light of the above listed evidence, which would j have been in effect during the 1971 to 1973 time !
frame, the NRC Report conclusion about the Pullman ;
audit program during the 1971 to 1973 period becomes suspect. [
t The NRC Report reveals that a number of Pullman Internal and Corporate audits were performed as well as FG&E audits, on QA <
. Manual (KFP) procedures related to welding activities. But the [
'QA. Manual was applicable to Piping only, not to Pipe Hangers l (Supports) or Pipe Rupture Restraints.- Therefore the conclusion ;
reached by the NRC Inspection are not applicable to Pipe Hangers "
.and Rupture Rektraints.- In addition, there is evidence documenting l l
serious deficiencies in the Pullman Corporate and Internal audit ;
( program which would suggest noncompliance to 10 CFR 50 Appendix B !
during the 1971 to 1973 time period. l l
This letter has presented evidence which indicates that i Pullman and PG&E did not implement an idequate audit program for all three major areas of construction, Piping, Pipe Hangers t (Supports) and Pipe Rupture Restraints, during the startup phase !
l- of Pullman work at Diablo Canyon. Many areas have since been
! corrected as a result of the Audit Program but only after a j repeated history of discrepancies. Areas of inadequate corrective j
- action include
- ,
i- !
- 1. The QA Program Description for Pipe Hangers (Supports) and Pipe Rupture Restraints. See PG&E Audit #73-15 and
- 80422 and Pullman Internal Audit.of 10-24-73. The Current QA Program Description still does not make a F commitment to 10 CFR 50 Appendix B, ANSI N45.2 series or I
0-ASME Sect. ion III. Per the QA Program Description certain '
sections of the Piping QA Manual apply to Pipe Supports and Rupture Restraints but it does not specify which section ,
for which type of work. The Piping QA Manual is based !
on ASME Section III, 1971 edition. ;
t
- 2. Preheating.of welds. The Pipe Rupture Restraint Crack Repair Program was the result of inadequate corrective action to poor preheating practices.
- 3. The Pullman Audit Program. See PG&E letter dated 9-19-73, ,
Quality Assurance Audits (Failure to follow existing and i and upgrade Quality Assurance prodedures). See PG&E !
Audit #75-2 and PGEE letter dated 3-25-75 concerning !
the need for a more comprehensive and extensive internal audit system. See PG&E Audit #80422. ,
- 4. QA. Documentation?for Ripe Support and Pipe Rupture Restraint {
work. See PG&E Audit #73-15, PG&E NCR # DCl-79-RM-003, >
There are other areas of inadequate corrective action which .
would require more research and time to write about. j I hope this infomation will be of use during your hearings l with the NRC.
Sincerely, Lal o. AM Harold O. Hudson !
Former Pullman, QA/QC Inspector, ,
, QA Program Internal Auditor, Lead Auditor and Pipefitter ,
(w schi. A T k p d b t'b N" " oM h.m,s.s%?4EAAA4 M
& I gc.s.3m 4 M .
d p c.,w LA AdiA 3 2.5 E. h y w w wazb.e%.<5* W ;
- pw se -
j i
(g gg g p adAah.cdidAd h .PoE Q* u M M 9* b bb """'t'Y l
yq,q, A wecmAue A;"'%
M E #~&= W R' 13 ">
b pg3 LA Jaa to am j ,
ma N 3 t-pA W M-w g l
List of Attachments
- 1. U. S. Nuclear Regulatory Commission, Region V, Report Nos.
50-275/84-16 and 50-323/84-06.
- 2. PG&E Audit #73-15, Oct. and Nov. 1973.
- 3. M.W. Kellogg Interoffice Correspondence, 9-18-73, subject:
M.W. Kellogg Internal Audit of Hanger Dept.
- 4. M.W. Kellogg Interoffice Correspondence, 10-24-73,
Subject:
Follow up Audit of Hanger Dept. and Audit of Rupture Restraints QA Program.
~
- 5. Interoffice Correspondence, 11-13-78.
Subject:
Upcoming Audit of Diablo Canyon in December.
- 7. PG&E Letter, 9-19-73, Quality Assurance Audits.
- 8. PG&E, Letter, 3-25-75, Quality Assura.nce Audits. ,
- 9. PG&E Audit #75-2, 2-20,21 and 25, 1975.
p , 9 64 E k; 10 -111 -74 h bMW' bD ^
t4. Fu e W ,12. o w ,h 6 d e -. bp WM EM t5, Pas L A. i q-w, h bhw - b y M J. h = L L
I $(DY U. S. NUCLEAR REGULATORY COMMISSION ,
REGION V
. Report Nos. 50-275/84-16 and 50-323/84-06 Docket Nos. 50-275 and 50-323 I.icense No. DPR-76 Construction Permit No. CPPR-6" Licensee: Pacific Gas and Electric Company 77 Beale Street.
Room 1435 San Francisco, California 94100 Facility Name: Diablo Canyon Units 1 and 2 Inspection at: Diablo Canyon Site, San Luis Obispo County, California Inspection conducted: April 2-4, 1984 Inspectors:
W M f JC [
'aner,RpZetorInspector D4te Sfgned Approvedbh:'
m ~ ML7 Jd D. F?*k'irsch, Chief date Signed Reactor Projects Branch l
l Summary:
Inspection during the period of April 2-4, 1984 (NRC Inspection Report Nos.
L 50-275/84-16 and 50-323/84-06)
Areas Inspected: A special, unannounced inspection by a regional-based inspector to examine audit records of Pullman Power welding activities, performed during the period of August 1971 through December 1973.
The inspection involvad 2n inspartion-hours bv one NRC inspector
- helults: Ne iteins of uencompliance or deviations were identified.
s F
F i
l l
l i
DETAILS
- 1. Individuals Co'ntacted
- a. Pacific Gas and Electric Ccmpany (PG&E) _
~
D. A. Rockwell, Project Field Engineer r J. Arnold, Resident Mechanical Engineer
- b. Pullman Power Products (Pullman)
J. Guyler, Internal Auditor ;
- 2. Purpose The. purpose of the inspection was to determine the extent to which Pullman and PG&E had implemented an adequate audit program during the startup phase of Pullman work at Diablo Canyon.
t
- 3. Inspection Approach The inspectors approach was to locate and review the documentation of !
Pullman self-audits and PG&E audits conducted between August 1971 and l December 1973. The inspector then reviewed these audits to determine i whether: -
t the appropriate criteria of Appendix B were being audited, ,
a.
i
- b. the audits conducted were of reasonable cocpetence and quality, and j
- c. followup action in response to audit findings was reasonable and appropriate. .
~
4 Inspection Results The inspector rev:ewed reports of audits conducted by PG&E and Pullman
- far evidence that Pullman weldir.g activit2es were being audited during the period of August 1971 through December 1973. Pullman audits include
- internal (site organization) and corporate organization audits. The following audits were conducted of Pullman's welding activities
- )
- a. Pullman Self Audits
- Audit Date(s) Welding Areas Audited Dec. 8, 1970 / Corporate Management Audit - Review of QA
. Manual -
Sept. 10, 1971 Weld Filler Material Control (WFMC), Welder l
~
Qualifications a -., ,, , - ~ , . , . . . . .$ . . , - - , . . , , - , . , - . . - - -. , - . , - - ,
2 March 28, 1972 Field Process Sheets (Travelers), Field Weld Records (RT, tff, Heat Chart, Weld Rod Requisition), WDic July 18, 1972 [ Field Process Sheets, Examination (visual) of completed welds, in-process (production) field welding Sept. 29, 1972 Corporate Management Audit - QA Engineering and Administration, purchasing, receiving, material marking, in-process and final welding, material traceability, handling, storage, special processes, liquid penetrant / magnetic particle /
radiography NDE, velding controls, cleaning, beat treatment, records.
Dec. 15, 1972 kTMC Jan.9-10,1973[ WFMCh-March 5,1973[ Fcocess Sheets, WFMC In-Process welding March 26, 1973 Corporate Management Audit - WFMC, Ferrite Control, Fit-Up, In-Process welding, Process Sheets, Welding Equipment May 1, 1973 Welder Stamp Depression riay 21,1973 n- Followup of January 1973 Audit Sept. 6, 1973 Welding Equipment Calibration Sept. 18, 1973 Visual Inspection of Completed Hanger Welds../C Welder Qualifications (weld gages, rod control bend test)
Sept. 19, 1973 [ Welder Performance Qualification and Testing .f Procedures Sept. 28, 1973 / Compliance to Welding Procedure Spe6ifications.
l hTMC (Rod Slips)
October 5, 1973 Followup of Audit of Welder Qualification and f Test Procedures performed September 1973 October 16, 1973 ! ollowup F of Audit on Kellogg Welder Audit of September 28, 1973 Nov. 7, 1973 / Corporate Management Audit - design and i
document control, procurement. control, I
fabrication process control, qualification of personnel and procedures, calibration of l
l measuring and test equipment, nonconformance l
L l
6
3
, reporting and corrective action, QA records, audits
. Audit No. Audit D'te(s)a Welding Area's Audited p..f [
- MA 71-16 Aug. 12, 1971 WDfC
. 6
,.', MA 71-20 Sept. 15, 1971 OMC, Welder Qualifications F .
MA 71-24 O c t'. 20, 1971 Proce's's Sheets, WFMC, Ferrite C6ntrol .
. MA 71-28 Dec. 30, 1971 kTMC, Nondestructive Examination Jan. 4, 1972 (NDE)
MA 72-2 Jan. 20, 1972 hTMC (Filler Metal Traceability), Welder's Symbols MA 72-6 Feb. 25 & 28, 1972 Postweld Heat Treatment (PWHT)
MA 72-8 May 1, 1972 Radiographs (RT) of Class I Weldments, hTMC MA 72-12 ! May 30, 1972 'WFMC (Weld Rod Accountability), M.
'RT of Class I.Weldments MA 72-16 June 27, 1972 WFMC (Storage), In-Process Welding MA 72-18 .Aug. 31 - (Field) Process Sheets, WFMC Sept. 1, 1972 (Rod Control Ovens) l MA 72-23 / Nov. 1, 2, 3 & 6, Rod Oven Control, Portable Rod I
1972 Oven Control (hTMC)
MA 72-27 March 27-28, 1972 kTMC (Compliance to 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> weld .
l ,'r6'd issuhnce)
MA 73-5 Jan. 26,~29-31, kTMC' (Follow-up to MA 72-27),
1973 Fit-Up and Alignment l MA 73-10 March 22-23 & QC Inspector's Qualifications to April 2-3, 1973 Inspect Welds l MA 73-11 April 9 & 12, 1973 kTMC MA 73-12 May 1, 7, 8, 10, Follow-up on Ferrite Checks on 1973 Stainless Steel: Welds MA 73-21 Dec. 11, 1973 Ferrite Control, PWHT, Welding Qualifications i
- c. PG&E Corporate QA Audits Audit No. . Audit Date(s) Welding Area Audited 71-15 . Aug. 23 - WFMC (Receiving Storage, Field Sept. 4, 1971 Use)
. ,, 72-3 Feb. 4, 1972 Piping Installation to Process
' Sheets k.
72-13 June 8-13, 1972 Process Sheets, Welding Procedure Compliance, In-Process Welding, Welder Qualifications, !
WFMC, Welding Equipment (Amps),
Calibration of Equipment (Rod Ovens) Welding Material
~
6 Certifications 73-15 Nov. 29 & Pipe Hangers & Rupture Dec. 19, 1973 Restraints - In-Process Welding, Welder Qualifications, Welding Procedure Compliance The inspector read through and surveyed the above audite to develop a sense of the audit competency and quality. The audits were performed by Pullman and PG&E to determine compliance with the Pullman Quality Assurance Program. The inspector noted that applicable procedures of the QA program related to welding activities were audited: KFP-8 " Process Planning and Control", KFP-12 " Control of Filler Metal", KFP-13 "Postweld Heat Treatment", and KFP- 15 " Welding Qualifications". Further, the above audits demonstrated that Field Process Sheets (travelers) were audited for compliance to KFP-8 (for ecch field weld the process sheet licts operations such as , fit-up, weld completion, NDE, and designated holdpoints for QC and the Authorized Nuclear Inspector's inspection and approval).
In addition, the inspector surveyed the findings and corrective actions generated as a result of the' audits to ass'ss e the appropriateness of the identified corrective actions. The* corrective actions taken, and '
follow-up audits to assure the corrective measures were effective, appeared satisfactory.
- 5. Conclusion Based upon a review of PG&E a'nd Pullman audits it appears that audits of Pullman welding activities were thorough and conducted in accordance with the Pullman QA program during the period of August 1971 through December 1973.
Based upon the abo've reviews the inspector concludes that:
- a. the Pullman audit program met the intent of Safety Guide 28 (June ,
i 1972) and ANSI N45.2-1971, in effect during that time period, and, l
therefore, the intent of Appendix B, l
l
. 5
- b. the audits appeared to be of reasonable competence and quality, and
- c. based uppa a samplinr,of corrective actions it appears that findings were followed up and~ resolved in a responsible fashion.
.1 i'
e 0
1 e *,
- - -u .
Audit Serial No. 73-15 i
,/ .
- Page 1.
PACIFIC CAS AND E!.ECT111C* COMPANY QUALITY ASStHtANCE DEPAPJDIENT
- DIADLO CANYON PROJr.CT PIPE !! ANGERS AND RUPTU!iE RESTPAINTS (Specification 8711) .
Performed by: D. C. Landes Date: Oct. and Nov. ,1973 :
Critiqued: Nov. 29 and Dec. 19, 1973 I SCOPE The audit was conducted to verify that the pipe hangers and rupture restraints are
~ fabricated, furnished and erected in accordance with the Specification and the '
Pacific Gas and Elc,ctric Company and M. W. Kellogg Quality Assurance Manuals.
- r RESULTS AND CONCLUSIONS The resu'lts of this audit were reviewed with the Project Superintendent and his staff in order to (1) discuss the results of this audit, (2) resolve any inadvertent siisrepresentations, and (3) to establish a conpletion date for those items requiring ,
corrective action.
The audit disclosed that H. W. Kellogg (Mi!K) and the General Construction Department
- departed significantly from the requirements of the Specification and P G and E's
, Quality Assurance Manual. MhK!s- Quali,ty_Assuranc,c_prgram c does not _ cc ElY. WEh. '
SCCt.ign 4 of Sp,ce,if e,at,j g 8111_and Pro,cedure gRP-4 The Mechanical Depar_tment's gi,lJaggemgra:._ does not comply w1Eh~ Procedure PRC_-7.
Accordingly, the Project Superintendent has stopped work on the installation of the pipe hangers and rupture restraints. !!c has directed his staff to initiate appro- l
" priate corrective actions to resolve all deficiencies and preclude-recurrence. 'Ihe ;
i Project Superintendent has agreed to foras11y respond to this audit by January 14, j 1974.
t a
- j i h
r s.
, ( .w- >m . e
. . t e.-
Audit Scrini No. 73-15 Page 2 I. I'll:JI A'411TED M. W. Kellogg Company. Quality Assurance Manual.
Revicwed MWK's Qua'lity Assurance Manual, with respect to the pipe hangers and
. restraints, for adequacy and co=p,liance to Specification-8711 and QA Procedure PRP-4.
RESULTS Section 4 of Specification 8711 sets forth the requirements of the standard
. "Supplc=entary Specification for Contractor's Quality Assurance Program" in-cluded in Procedure PRP-4.
MWK's_ approved QA Manual co= plies with Section 4 of the Specification. The NaEual, howcVer, does not 3pecifically address itself to, nor is it completely applicable to, the control of pipe hangers and restraints.
Thus, MWK has written an " Engineering Specification," ESD-223, establishing the QA program applicable to the control of the hangers and restraints. As confir cd by R..G. Fink, MWK's Field Quality Assurance Manager, the intent of the "Engineerihg Specifications" is to set forth procedures and instructions to the field QA inspectors, engineers and foremen implementing the policy stated in the QA Manual. ESD-223 establishes QA policy instead of providing instruc-l tions on how to implement the policy stated in the Manual.
Additionally, the progres set forth in ESD-223 does not meet all of the re-quirements of Section 4 of the Specification. Ocficiencies were noted in the areas of document review nnd control, qualification of special processes and pctsonnc1, work procurenent control, receipt inspection of =aterial, identifitatien centrol and statu: of caterial, nonconforminq catcrial control.
inspe tien :nl test raccrl: a:.d inspection end test plans. Coasequently, the hanger and restraint QA progran is in violation of Procedure PRP-4.
ESD-223 is, in essence, an " alternate QA pro; ram" approved by the Resident .
Mcchanical Engineer. Procedure PRP-4, Paregraphs 3.14 and 3.15, requires that such an " alternate QA program" be sub=itted to the Director, Quality Assurance, for review and approval prior to use. ESD-223 was not sub=itted to the Quality Assurance Depart =ent for review and approval prior to use.
CORRECTIVE ACTION Initiate a separate QA program, which is in accordance with Specification 8711 and Procedure PRP-4, covering the fabrication, rcccipt and installation of the pipe hangers and rentraints.-
Complete discrepancy reports identifying and dispositioning the discrepant itcasand, conditions existing in the work completed to date and initiating steps l to preclude recurrence. ~'
~~s, ,
t
~
. . Audit Scrial No. 73-15 f ,' - - Page 3
/
II. _I_T_E_M AUDITED Altius Corporation and Grinnel Company Quality Assurance Manuals.
Altius and Grinnel, subcontractors to Mh'X, supply fabricated pipe hangers and components. Reviewed those hbnuals for compliance to Specification 8711 and Procedure PRP-4.
RESULTS Altius' QA Manual is in general compliance with Specification 8711 and Procc-
,duro PRP-4.
Grinnel's QA Manual was not availabic on site. Documentation regarding the status of its review and approval was also not available on site.
CORRECTIVE ACTION Obtain a copy of Grinnel's Manual and determine that it is in accordance with the Specification and PRP-4. Obtain or provide objective evidence, documents -
tion 3 of P G and E's review and approval.
III. ITEM AUDITED ,
Receipt, . storage .and installation of pipe hangers and rupture restraints.
Reviewed the receipt, storage and installation operations and documentation of various hangers.
RESULTS Selected several hangers and restraints, determined their status, and reviewed the quality records documenting the fabrication, receipt, storage and instal-lation.
Results of the review.are: .
- 1. Hangers for pipe 2-1/2" and greater: ,
A. Subcontractor-supplied hanger assemblies are inspected by construction j forces aather than QA personnel. These receipt inspections are not i documented and filed in the QA vault. ,
B. Itanger assemblies fabricated on site do not receive any in-process or l
final inspections.
l C. The installation of hangers receive a final inspection, but no in-process inspections.
D. Inspectors are using inspection forms which they consider necessary but
-o -am.
ot controlled by t Ik (\ (<_.
/
f L
i l
[ ..
/ ,
Audit Scrial.No. 73-15 l Page 4 l
- , l' I
E. Tuo hangers which had been in:;pected and accepted by HWK were re- ;
inspected. The condition of the field ucids were generally rough - !
and irregular. Portions of the welds were Icss than the 1/4" fill'et '
specific.f on the drawings; other sections were somewhat oversi e and convci. The condition of the shop welds were uniform and smooth; ;
however, the. weld si:c was generally less than the 1/4" fillet spec- l ified on the drawings. Also, some welds specified on the drawings [
were not made; inaccessibility explained why some were omitted, but i not others. No discrepancy reports or other documentation authorizing * [
the acceptance of these discrcpancies was availabic.
l l
F. When requested, the inspectors could not furnish any explicit acceptance j criteria. This data is not c1carly stated in the ANSI Codes or in any ,
P G and E or NWK document. .
- 2. ' Field run hangers for pipe less than 2-1/2": !
,. A. Receipt inspection and surveillance of raw stock waterials is performed ;
and documented by QA inspectors. -
B. All field run pipe hangers are fabricated on site. These hangers do [
not receive any in-process or final inspections.
C. The installation of hangers receive a, final inspection, but no in- l process inspections. ,
D. No field run hangcrs had been inspected at the ti=e of the audit. Various
- discrepancies were noted by the auditor. Discrepancies noted were: [
rough and irregular welds, undersize and inco=pletc welds, hangers not ;
in the location specified by the drawings, hangers not fabricated in [
accordance with approved drawings and hangers and approved hanger draw ~ }
ings contrary.to the P G and E standard design drawing. No discrepancy I reports or other documentation authorising the installation of hangers !
with these departures was available. .
E. Mcasures providing for the appropriate and timely identification, review, t and dispositioning of hangers when they cannot be installed in accord- i ance with the approved drawings are not evident. Construction crews are
- not required to stop work and obtain appropriate approval when the hanger cannot be installed in accordance with the approved drawings.
Instead they are allowed to proceed with the work and rely on the final [
inspection to detect and resolve any departures. t I
- 3. Pipe rupture restraints:-
A. MWK's rcccipt inspection checks for road damage and completeness of j material only. Surveillance inspections of stored assemblies is performed by MWK. ,
C. The P G and E Civil Department providos the inspection and documentation
- I to assure that the procurement requirements have been met. The rocciv-ing inspection forms, Form C-35, were on file in the QA vault. Several '
forms, however, note contingencies where the inspection, verification that ti g curement requuc.w ni % been met, has not been completed. t
/ Audit Scrial No. 73-15
/ .
Page 5 -
l ,
.. These supports are nc,t placc! on " hold" or withheld from installation.
[ The Resident's Instructions do not require identification and segrega-tion of non-conforming items. Additionally, the auditor could not locate receiving reports for all the restraints.
C. HWK has not determined, or received a written material release from P G and E stating that the procurement requirements have been met. i Unicss exempted in writing, Specification 8711 and Procedure PRP-4 .
require that the contractor inspect company-furnished material for confor=ance to the purchase documents, specification, drawings, etc.
D. Except for the ultrasonic inspection, MWK documents their inspections on " Marked-up" crection drawings. This documentation takes the form
.of sign-offs and color coding. The method of recording inspections and acceptance criteria are not set forth in an instruction and was not clear to the auditor. Upon explanation of the system by the in-spector, it was still difficult to determine the inspection status.
All in-process inspections of workmanship and technique required by the AWS Code are not being performed. -
E. Some welders were welding materials of greater thickness than they ;
were qual'ified. -
F. 1lelding was not in complete accordance with the assigned weld proce- i dures. Several of the non-essential variables had been altered or ,
were not being complied with. .
G. Provisions for the installation and inspection of high strength steel bolts are not in accordance with the AISC, Code.
CORRECTIVE ACTION i Refer to the corrective actions for Audit Item Ho. I.
IV. ITEM AUDITED .
i The Resident Mcchanical Engineer's surveillance system of the fabricating, fur- '
nishing and installing of the pipe hangers and rupture restraints.
Reviewed the Resident's written instructions and surveillance activities for adequacy and compliance to Procedure PRC-7, " Inspection of Materials and Compon-ents During Use and/or Installation."
RESULTS
! Surveillance of the receipt and installation of the pipe hangers and rupture restraints is perfor=cd by the Power Plant Piping Group. MFI-2 sets forth the Resident's written instructions to personnel in this Group.
MFI-2 Instructions do not specifically address themselves to the surveillance of pipe hangers and restraints. The instructions define the inspector's res-ponsibilities ani quimhi::Lto perform and record specific inspections.
. n i
- . ._ .. , ..m ___- _ _ _ .
- *' e i ej Audit Scrial No. 73-15 Page 6 Ilowever, the inspector is not performing any of these inspections; they are not applicabic to the inspection of pipe hangers and restraints. The inspector is performing other inspections; however, these inspections are not documented .
or described in the MFI. ,
CORRECTIVE ACTION Issuc a written instruction describing the responsibilitics and duties of, and the inspections and records required from the hanger inspector.
3 V. ,ITD1 AUDITED Departures from approved drawings. .
Reviewed the Contractor and P G and E system for handling discrepancies and departures from the approved drawings for the 2-1/2" and larger pipe hangers.
RESULTS h' hen a departure is discovered, MWK stops work on the support, completes a dis-crepancy report (DR), and submits the DR to P G and E for approval. Upon approval, they release the support for completion of work.
Upon receiving MWK's DR, General Construction analyzes the departure and writes
..its own DR to disposition the discrepancy or initiates a drawing revision to nullify the discrepancy. In the majority of th'e cases, a drawing revision is initiated. In most instances the concurrence of the Responsible Engineer is documented on a telecon sheet and work proceeds; a revised drawing or mechanical y revision sheet is not obtained before proceeding with the work. In other in-stances, the mechanical revision sheet or revised drawing is received, but all of the signatures required by Procedure PRE-3, Responsible and Supervising Engineers, are not present. -
CORRECTIVE ACTION I
' Document and disposition such departures on a discrepancy report or do not I proceed with the work until a properly approved revised drawing or mechanical revision sheet is received on site.
{
Qd. dwh D. C. 1 ANDES DCL:je cc: PLBussolini RRFriedrichs Gt hrs ( \
MRTresicr m File
t fc.!T .-
. $N?b e ?
(. '
PACIFIC GAS AND E LE C T RI C C O M PANY PGst3 j 245 M ARKET $TREET . SAN FRANCISCO. CALIFOR NIA 941C6 * (415) 7814211
- Twx 910 372 6587 !
P.O. Box 117 Avila Beach, California 93424 I
.' November 29, 1973 M.W. Kellogg Company P.O. ' Box 367 {
. Avila Beach, California 93424 Attention: Mr. J.W. Ryan i Project Manager Diablo Canyon Project Specification 8711 Class I Hangers Gentlemen:
Effective immediately, installation of Class I hange.rs must be discontinued.
This order'is a result of a recent audit performed on installation of Class I hangers. The results of this audit indicate-
- 1. An approved Quality Assurance Manual for installation of Class I 8
hangers does not exist.
- 2. A majority of those Class I hangers installed to date do not confom ;
to drawing require =ents.
l l
Should you have any questions, contact the Mechanical Resident Engineer.
l Sincerely, C .
t ,
C.K. Max ield Project, Superintendent I
8 r
=.;. .
PACIFIC OAS AND E LE C T RI C C O M PANY 3 3*M3 - [ 77 SEALE STREET . S AN FRANCISCO. CALIFoRNI A 94106 * (415)781 4211 . TWX 910 372 6587 M. H. C M A N D LE R mana4 9
.n.........n.....v., -
u... 6 c..m cne.
December 11, 1973 M. W. Kellogg Company P. O. Box 1007 Williamsport, Pennsylvania 17701 Attention: Mr. J. E.* Bowes Manager of Construction
. Diablo Canyon Project Specification 8711 Gentlemen:
~
This will confer approval of the M. W. Kellogg Quality Assurance Manual for Pipe Support Field Procedures dated December 10, 1973 with the ex-ception of rupture restraints. Per discussions with your on-site personnel',
the section on rupture restraints will need further revisions before approval.
Please submit nineteen (19) copies of the revised pages to:
Mr. C. K. Maxfield Pacific Gas and Electric Company P. O. Box 117 Avila Beach, California 93424.
A copy of this letter is to be inserted into each copy of the Quality Assurance Manual.
Sincerely, L
f-Y ' ' RECEIVFD, afbr
- i. POWER P:P NG .
. CONSTRUCT ON ,
! DEC 141973 ;
l RECD .; h"? T(
l c7[tI.
l A. J. M
~~'
- ~
J. R. t. J' '
- K.- E..J ~
__ . . - . - 1 G C, R. ~1
.t
.j ? ,
(l/ - _ . . . . - . . . . . . . . .
~ .....- - . . . . - ...<._. - _ . _ _
(.) tilTEROFFICE CORRESPOMDENCE
.THE M.W.KEtt0GG CCMPANf L DATE SiPTEl'3ER 18,1973 TO f;, G. FI Nii FROM J. h. I'.C SE'.!A A!!D C0f TUTi;0 SUBJECT , ..],I.r...L..li~.h5.0,,C,G G I,UI? l@,(,,64,Q,1,1.,0f,..,(i,GiGE,0,,,,DE,f,A,RT,0,E,UI I -
r N
3r.0:'E OF AUDIT A complete audit was performed as to the cdberence to t'Sc Engineering Specifications end quality Assurance Manual. ,
d BODY _OF AU0iT h o Han'gers were picked ct random in the Turbine Buildir.g for i A.
' traceable tracking and correct installation.
t NOTE: These Hangers havec'cen tagged by Q.A. as.: -
i
- 1. Sys 02 K-503 4240 6-309 li s 42'i2 6-311 f (j 2. sys 02 K-503
- 1. Manger (1.) 6-3G9 va: found to have a Ic.ose cotter oln. The l inspector in the area was notified and ha insediatciy rectified l
2 the area in question.
I 2[ For both Hanger (1. G 2.) 6-309.and 6-311 traceability was fol-lowed through'the Area inspector's Log Deal to the Main Office.
i It was at the Hanger Departet.nt'in that office where tracecbility I
I itopped. I was unable to confirn fron ti.e office recor ds that .
the two.above mentioned Hangers had been checked.
l t
'I 3. The Eng,lneer in charge of the Hanger Department was notified as j to the problers. He has informed into the Arca.Inspectnrs that they the office, for a short period must bring theli Log EcoAs j of tirac, in order to transfer all portinent infornation into
' office records. A follow-up audit will be made to insure com-I pliance.
t dia. cussed with I E. The method of checking for correct I! anger location was l ,
the inspectors.
The acthod wSich they now (mploy consists of r-casur-
't coordinates indicated on the walls by ing with a tz.pc to !! : nearest l P. C. t, E. This appears to be a satisfactory ic.cthed. .
i
} N' C.
It was not ei that sonc hangers have welds which ucre qecstionabic to h VI the Auditor. The inspectors for tN Hanger Ccpartuent were shown l, o ) j., s.N -
j tb ; 4 .' , \
y . eg
4
'3 10: ?. . C. Fit?!: oATE SEPTE!10ER 18,1973 g x+: ret H. V. KELLOGG liiTEP.l:AL AUDIT OF llAMCER DEPARTt:Ef1T e w No. PAGE 2 i
these vicids and a discussion followed, the inspectors agreed that the welds were not of a quality which is cor.patible to good work-
[ manship at this cox. plex. All questionable welds will have a hold
- tag attached to the Hanger and a D.R. will be initiated for final I} disposition.
D. There vicre many discrepancies when comparing the P. G. & E. Drawing Log against the drawings used by the Hanger Department for fabrication.
[ ,
These descrepancies are:
_Drawinn Loc Pevision Copy on File Sheet #
049294 2 1 ---
049303- . 4 1 26 049279 13 12 1 12 _
9 2 NOTE: On 049279 Sheet 1, Chan5c //13 is volded and 049279 sheets #5 ik through /.?19 are not logged in the book. This is only a rendum serple of fifteen drawings. The Field Engineer and the Hanger Department murt combine their infort.ation cod bring their log end prints up to i* date. A future audit will be conducted to, verify cortpliance.
E. Area Ten was examined for proper storage of Hanger material. There is not a definite separation between Class I and Class 11 material, llowever, any storage areas which have Class I material in it are j designated with a Class I sign.
F. This section of the audit pertains to the Hanger Departments com-pliance with ESD-2;t3 Sections 2 and 3:
- 1. H. W. Kellogg is writing the Furchase Orders and material re-quirements. P. G. & E. is not approving the design of an/ 2" and under Hangers. At this time H. W. Ec11::gg's Hanger Department has stopped designing 2" and under Hangers and are in the process of up dating iso's to shuw locntion and design of installed Hangers for F. G. & E.'s approval. When design of 2" and under Hangers are resumed complianwe to ESD-223 will be assured by a follow up audit.
2 Rando.a support .raterial is being order,:d by the Hanger Engineer.
i s . /. . tww.gu r is revic >ing all rurc!.ase C rdert., this is in co.aple:c cc .g liance wi th ECD- '23 Para. J
".2. ,
( T '
rJ z'!
O
/'
// i ' -
t<n R. C. F i t :'. ,
ontt SEPTEMBER 18, 1973 wrurrt-. M. V. I;T.LLGtG I;i1ER;;AL AUDIT G! . llAt:GER DEFARTitEllT re. .! r .- FAGE 3
- 3. Vcndors which cre supplying us with Pipesupports have been certified by P. G. & E. and M. W. Kellogg in the following uinner:
- 13. /!4 Grinnoll: P. G. & E. Engineering release signed by A. G.
Volthers P,-25-71 and approved by v. Linblad 9-3-71.
l$. /!5 Altius. Corporation: Approved per P. G. & E. letter dated
- 3-6-72 with the provision that Altius add the latest revision of AUS D.1.0 to their quality Assurance Manual. Altius re-vised their Q. A. ! nnual with Aa.endraent !!2 which added Sch-Section 13.5 on 4-3-72 according to the above P. G. & E.
recow.cndation.
Ecrgen & Paterson: Hcve been approved by M. W. Kellogg and a letter has beer. sent to P. G. & E. stating this fact and requesting their approval. .
n 4 The M. W. Kellogg Receiving inspector is complying with all aspects s[' Q l-I -
of ESD-223 for the proper receiving of Class l Pipesupport, i,
CoilCLUSIGil .
A follo.,-up audit will be conducted during the raonth of October.
. _t ew J. . hoskua .
j Q.A. ocumentation Supervisor f
Dan Tutko q.A. Inspcctor cc: J.U. F.yan C. Lenzi T. buvanec A. ttacAuley l
JR!!, DT: cc
/
/,'.'
r, s%
-e A 3
0(f) INTCf: OFFICE COR".ESPO!JDENCE
.THE M.W.KELLCCG COMPANY l -
i TO R. G. Flin' DATE OCTOBER 24, 1973
! FROM J. MOS!"'A
.- $UBJECT . ..[.0$LY!~0? 51!A.E W E* 55 ~bIY- I" "- """ ~
. qur.LITY ASSURA!.CE PROGRAM i-FOLLrfg.UP AUDIT OF ffAfiGER DEPARTMENT f(
I-j inspectorr Daily Work Sheet
. During the previous month it was deter..iined that information l ,
was not being transferred from the field to the Hanger Department i Of fi ce. They are r.ow up dating their office logs on a *seekly basis (every l~riday) . This apprears to be of a sufficient occurance to assure reliable records.
~
' Drasinq Revisions l i
- ( M. V. Kc11ogg has completely re.ised its Drawing Log to agree I
with the haster Log kept by P. G. & E. The Hanger Ccpartment is now In the process of co-paring their complete drawing inventory to the l corrected'Orawing Log to insure complete compliance to our Qualitf g.
Assurance Program at this facility.
Bereen & Patcrson Ve still have- not received word from P. G. & E. about our 1ctter of cpproval on ti:e above stated firm.
AUDIT OF RUPTURE RESTRAINT QUALITY ASSURAf1CE PROGRAM
.l i
- 3 Rupture. Restraint Pcicases I Rupture Rcr.traints were released for Unit I with a manager's release. H.V. KcIlo;;g has attempted to receive a tr.aterial reic9se from P. G. & ' E. for Uni t 11 Rupture Restraintr.. As of this date, H.W. Kellogg has been inforrned by F. G. & C. (J. Ifolicy) that a
' release was unnccessery due to the folioaing inforr.tation:
American Bride.c i:.~ an cpproved on site contract vendor and t . thcy, (American !1ridue), uti11zc'. P. C. T. E. cyprove<f dreuings there is, therefore, no need for a ::-cterial relea-c for any h'h Rupture Restraint.,
w
- %s.
j g% ). h A* L .. i.J .
C 2.
o A
ro: R. G. r! *:;; natr OCTOBER 24, 1073
?(. si.n irc t : FCl LUJ-: P .", WIT OF HA!!OER DEpARTl:E iT AtlD AUDIT OF RUPTURE RESTRAINT's (1:ALITY ASSU.*1.!!CE f 10 G'!/.!i v..c no. 2 I. f...nture Ecst raia: Eur: :.i.qu i s i t i on .
< All Rod requir.itions v.cre arranged by Restraint Number and were checked to the actual work in the field by Restraint flumber. A few descrepancice, ocre noted, such as:
Rod requisitions which have been filled out improperly and arc difficult to file. There are some Ecstraints in which I have been unable to trace any Rod requisitions. A further check insured the auditor that all present Rod rcquisitions are being
~
filled out properly cnd this problem should be alleviated in a short time. A future audit will 1 e performed in ilovc=ber to insurc proper steps have been taken to file the remaining re-quisitions.
Adherence to Correct Instc11ation Procedure
[^'T All aspect. of installation were checked to insure compliance
'v' to the letter approved by A. G. Valters on October 19, 1972.
k it appears thct Spec 9933 Xf and 9711 as stated in the body of the letter are being complied trith completely but, it would seem to be beneficial if all the references stated in the letter ucre con-dcnsed into a singic procedure to be u ed at t!ls i complex. This would save considerable time which is normally spent researching the various references.
Vendor's Drawings The Vendor Drawing Lng was correlated to the Paster Log in the P. G. & E. office for Drawing # 663069 (Unit iI) and // 663378 (Unit iI).-
There were no discrt.pancies noted in our Log, the drawings were then ciccked.to the latest copy on the stick in the offico. All drawings were of the correct revision. The foreman'in the field was audited for his cocy of Drawings #663379 (Unit II) and all his revisions were found to be up to date.
Upon checking the quality Assurance Documents which we use for weld documentation, I found that a few drawings were out dated, llcw revisions have been ordered and during a future audit next month all dre sings will be checked for proper revisions, q ' d ^'
Q- ..s. t.os'a..a
-" lQ. A, Dur.onentation Supervisor $
( cc: J.W. Rvan N- f,Y, o a ec [-h'*pL w f;,. (f '
A. MacAulcy // ./.'
,s y //
.. - m , .-, . .
- i
- -- L, qis:: OFFICE COMESPONDENCE
) !
. l
- * ""I"
- DATE November 13, 19Ja~ .
- 10 ,
E.CJ. Manning .
stoM Upcoming Audit of Diablo Cinyon in December CT This audit is very essential and should be conducted for the following reasons: .
- 1. Pullman Power Products has committed itself to audit all areas of the facilities program annually. The Diablo ,
- Canyon project has been audited extensively only in hard-ware areas. The entire program has not been evaluated. ,
- 2. To be in co:pliance with our corporate requirements, an l
ASME Section III, Article NCA-4000, this audit is essential. -
l In the past, Pulltian Power Products did not conduct audits or practices to ASME or 10 CFR 50, but I feel it very essential to do so now.. This type of thinking will satisfy the require ents of the NRC and ASME and, hopefully, the results of the audit will be objective evidence r show the NRC and ASME of compliance of our program.
,- - '- If any'further questions dealing with this matter are necessary, please feel free to contact Central Staff Auditing.
l
'C -
c 1
1 E. J. Manning C .
Senior QA Engineer EJM/kal -
cc: J. E. Bowas A. A. Eck ,
.- P. Runyan
! J. Ryan ,
Pile .
c' .
\ .
\ '.
- \
,C
\.
- i
- As
- .;x ,
. g.% %
4 i
- VM$.3 .
A.__.-_..
. . . . . .a.a ..
. . .I . .. . , , . . . . . . . _ . , , . , . _ . , . _ _ ., _,
. . . , . . . . . . . . . . . . - . . . _ . . ., c . . .. . . .. ..... ......
Audit No. 80422 j.
e' -- -
??6*/]?/a[V/VI <//Y,O f2N?/W/Dd i IC'** U*LC; 0l13l70 Page 1 of 12 h
/f - ~ .- -
- PACIFIC'CAS AND ELECTRIC CCIPANY
- QUALITY ASSURANCE DEPARTIT.NT , ,
4- ,
Tit 13 Pu11aan Power Products Quality Assurance Program .
k dited Organization / :
Facility: Pullman Power Products at Diablo Canyon Power Plant !
Auditors: M. E. Leppke (Ledd Auditor)
C. L. Eldridge '
R. U. Tay]or ,
patra Parformed: April 2 - June 1, 1978 ,, y..
1.0 _ Scope This audit. was performed with three objectives in mind. They verc: .
(t) Verify that the Pullman Power Products Quality Assurance Progran ,
implemented at the site meets contract requirenents and the requirc-ments of cpplicable regulaticns, codes, and standards. .
(b) Review objective evidence to deteruine the _ validity of the-findings g.
of an audit performed by Nuclear Services Corporation (NSC) in 1977 and determine if Pullman's responses were accurate and approprictc.
~
(c). Observe the as-installed con'dition of components a'nd supports fabricated and installed by Pull =sn to verify adherence to applicable .
specifications, design drawings, and quality standards. ,
~
k D.0' conclusions and Exit Interviews
. ~ ' :* .
?
2.1 Conclusions *
,. . .. ,. . .. ' ' ~
,, The Pullcan Power Products Corporate QA Program and the implemen-tation thereof were reviewed in light of the audit perfor cd by .
- Nuclear Services Corporation. Additional audit activities included .
- t. review of the installed hardware. 'The pri ary conclusions are given below with additional details set fo;th in Appendix A. -
(a) Adequacy of the Pu12raan Power Products QA Pro; ram The Quality Assurance Program implemented by Pullman Power Products essentially fulfills contract requircsents and ucces .
requircuents of the ASME 1;oilcr and Prescurc Vessel Code, 1971 cdition. However, three program deficiencies were identified
> and threc deficiencies in the ic:plec;cntation of established ' j, procedural r quirements vere noted. Twc Ecuconformance Reports and four Minor Variat, ion Reperts vere initiated by Cencral Conutruction. Areas were also identified where it nJnurs So be ace {PmeMda @ss sc4 EJWaRRAa @mup's advmRece .
3 ... . . . . . . ,
Audit No. 80422
/. * , .
- hj? / h ft?/ **
l0 Y^ ;4 r ._- *
(b) Evaluation of the Nuclear Servie< n Audit of Pullman Power * .
Products QA Program -
Several apparently generic deficiencies in vori. performed by Pullman were previously identified by the Cencral construction 11cpar tment . As a result, extensive reinspections were directed or performed by the General Construction Department. Additionally, Pullman was asked to perform an overall evaluation of the accept-ability of the installed components and supports. An independent party, Nuclear Services Corporation (NSC) was contracted by Pullman to perform this evaluation. However, the resulting audit did not achicyc its main objective in that NSC concentrated almost entirely on Pullman's Quality Assurance Program and inspected very little of the installed hardware to deterr.ine the quality of the work.
NSC's audit findings allege that major' portions of Pullman's program
- are inadequate. It is essential to understand the requirecents which NSC audited against to place the NSC audit findings in perspective. The audit checklist used by NSC states that require-i ments for the NSC audit were extracted from the following sources: -
c) 18 Criteria (10CFR50 Appendix B) .
~ >
b) Grey Book (UASH.1283 " Guidance on Quality Assurance. .
-~ Pequirements During Design and Procurcacut Phase of Nucicar f Power Piants") , ,
)
c) ANSI Standards . ,
s' .
d) Nuclear Services Corporation (internal guidelines)
\ The,18 criteria of 10CFR50 Appendix B are applicable. Chapter 17.1 of the Diablo Canyon Final Safety Analysis Report co=mits to a quality assurance program meeting the intent of 10CFR50 .
. Appendix B. Pullman's program also commits t'o the ASME Boiler and Pressurc Vessel Code (CODE),1971 edition for quality assurance
,- ' requirements. The 1971 Code is consistent with the requirements of 10CFR50 Appendix B.
.' .The WASH 1283 document (Crcy' Book) is not applicable. Chapter 17 of the Diablo Canyon Final Safety Analysis Kcport makes no coc=itment to WASil 12S3 for the design and construction of DTablo Canyon. WASH 1283
- cndorses ANSI N45.2, 1971 and ANSI N45.2 scrics standards.
MISI N45.2 states in its foreword that it is not applicable .
- to work performed in accordance with the Cede. MISI N45.2 scrics standards state under " Scope" that they are intended " '
for. use in conjunction with NiSI N45.2,1971. .
.s. '. .
The ANSI Standards are not applicabic for the ca=c reasons expressed for UASi! 1283. ,
I Internal NSC corporate guidelines only represent the opinions of the auditors and are not interpreted by PCandE as tequirements.
AuGJt 80. oV94 *
, .e Page.3 of 12 Vlylp / hf}CYAAl
'Hany NSC audit findings state that clements of , Pullman's program arc inadequate but specific deficiencies and references for
. requirements are not identified. Some actual deficiencies werc
. identified by the NSC auditors but.many of their findings appear to represent the auditor's opinions with no bases in applicable -
regulations, codes, or standards. -
, The audit performed by the PGandE QA Department essentially retraced the steps of the NSC auditors. Deficiencies identified are listed in Appendix A. Other alleged deficiencies stated by NSC do not appear to be supported by objective evidence or do
, not appear to be based on applicabic codes, regulations
. standards. ,
~
. i:
.. Pullman's responses to the NSC audit findings in general appear to be correct. Movever, to place the NSC audit findings in I' proper perspective, Pullman should have assessed the applicability ,,
. of requirements which NSC alleged that Pullman violated.
.. j -
(c) Evaluation of the Pullean Power Products Corporate Audit of -
the Unit 2 !!ardware Installed by Pullman Power Products Corporation In February 1978 Pullman's corporate. office performed an audit '
to verify that Unit 2 hardware ite=s were installed in accordance ,
with design drawings and specifications. One hundred twenty-two :
hangers, restraints, and snubbers and seventy-seven isometric l drawing packages were inspected; no discrepancies were noted by
,. the Pullean Power Products auditors.
Approxicately half of the itcas inspected by the Pullman audit team were reinspected by PGandE during this audit; several ,43 discrepancies were noted. In light of the number of discropancies noted, it is apparent that the Pull =an audit did not effectively evaluate the quality of their work. -
Most of the discrepancies noted appear to be minor in nature.
Similar prob 1 cms identified by reinspectiods in other areas '
have generally been " accepted as is" by the PCandE Engineering Department. However, an overall assess =ent of the situation still should be donc to determine whether additional reinspections
- should be performed and the scope thercof. Pe11 man's canagement agreed, during a meeting held on May 25, 1978, to send additional '
qualified staff to the site to perform the required evaluation.
,e -
General Construction plans to direct the performance of the Pullman evaluation. -
HVR H-3725 and M-3726 were initiated by Cencral Construction I
to document and provide for resolution of the noted discrepancies.
l
- = . ,
. .- Audit No. 80422 .
p h j f g () / .//t/f0$MA ,
2.2 Exit Intervicus ,
Two preliminary ecctings and a final exit meeting verc'hcid to discuss the audit findings and to establish the recommended corrective actions.
(a) Preliminary Exit Meetinn (May 10, 1978)
~ '
A meeting was held on !!ay 10, 1978 to discuss the results and preliminary findings of'the audit of the Pullman Quality Assurance Program and of the overall pipe and pipe support inspections. The following personnel attended:
_Cencral Construction Quality. Assurance Engineering -
. C.'K. Maxfield R. P. Wischow J. B. Hoch M. R. Tresler V. L. Killpack l R. Etzler M. E. Lepphe .
G. Arnold .
C. L. Eldridge
. R. W. Taylor , ,
> (b) Preliminary Exit Meeting- (May 25, 1978)
- A meeting was held on May'25, 1978 to discuss corrective actions . ,
with General Construction and Pu11c:an Power Products. The General "
. Construction Department directed Pu11raan Power Products to perform the required corrective actions. Those in attendance
- were:
y .
. Pullman Power Products Quality Assurance General Construction M. Evans .V.'L'. Killpack ~ C. K. Maxfield ".
P. Runyan ' M. E. Leppke 11. R. Tresler J. Ryan
- R. Etzler .
. A. Eck *
(c) Final Exit Meeting (June 1,1978) , ,,.
A final exit interview was held on June 1, 1978. Audit findings
. and agreed-upon correctivo ac.tlons were su:=arized, All deficiencies identified during'the audit had been documented prior to the exit interview by Concral Construction on Nonconformance Reports or ,.
liinor Variation Reports. Those in attendance vere:
- i _Cencral Construction . Quality Assurance C. K. Maxfield .
R. P. Wischov
}!. R. Trcsler -
V. L. Rillpack R. Etz1cr !!. E. Lcppke '
C. Arnold . C. L. Eldridge l
~ , _ . . . . . . , _ . . . . . - - - - . - - " ~ " - ' "
Audit No. 80422 gl /[ ~/ // /? . / /~l/ O /YW YA -
-g
/ _
As a result of this audit the following Nonconformance Reports (NCRs) and Minor Variation Reports (!!Vns) were written by the '
General Construction Department to resolve the prob 1 cms identified:
NCR Description DC-78-131-004 Documents the lack of program
' definition and lack of detailed
, audit schedule. .
. DC-78-R!i-005 . The relative responsibilitics of
- QA and production are not clearly ,
established. c ,, , _ , .
MVR Description
}!-3723 ,
Pullman Corporate Management audits were not performed at the scheduled
~
frequency. -
~
li-N24
- Hoid points were bypassed. -
~ '
H-3725 . Hardware discrepancies were noted. .
If-3726 Discrepancies concerning isometric drawing packages were noted.
. i i
'; ' Corrective actions were agreed upon; the QA Department will verify '
the resolution of these noaconformances and deficiencies. ^
Prepared by: [h 6 l <ft Ke ~
.i f /hh -
H. E. Leppke ! -
[
C. L.ogldridge
': . Av %d- L, T!-f Ed.w2
? ~
. /R. W.jTaylor i
. Approved by: \/* -
t!.< -
V'. L. Ki1] pack '
, . d D_ . bd' -
4
~
' Audit No. 00422 I Page 6 of 12
^ f-)S '?/ 5 bl/
a ,
/ Al h h/? /Yf /I N 0 0
~~
f APPENDIX A ,
IJ this section, the deficiencies which were identified as a result of th audit cra di'scussed. The prob 1cm in cach instance is identified to a specific Noncon-f:rmance Report (NCR) or Minor Variation Report (MVR) that was initiated by the C:n:ral Construction Department. .
s
- 70 cddition, non-mandatory suggestions ' and recom::icadations of program improve-ments are given for consideration. '
LO Pror. ram Dcff ciencies 5 Two Nonconformance Reports 'wcre initiated for the three identified
- d:ficiencies. .
}*
(a) ~ Nonconformance Report No. DC-78-RM-004 .
i,. , . .
f This NCR is comprised of two parts as follows: .'
. (1) The Pullman Power Products Quality Assurance Program is not adequately defined. The ASME. Boiler and Pressure Vessel Code, [ * ~ l'O
( f "'", 7 f'Section III, paragraph NA4140 of the 1971 edition requires that lP t' ?
.' the Quality Assurance Program be documented in detail in a %
manual consisting of written polic.icsuprocedurcs, and instruc- ..
"~,'Y tions. Corporate Procedure No. ."iVIII--1,is presently being used for the performance of managenc$t a6dit's of field activities.
~
- c F d 8 8 Corporate Procedure No. VII-1,is being used for qualifying / 5 . 7 ..I
)
vendors for the , Approved Vendors 1.ist. These procedures ",
impicment Quality Assurance requirements of the contract but
are not identified as part of the program and revisions are '
not controlled by the program. .
, t, .
' ~
~ '
The program is required to be approved by'the ASME, and changes
.- ' to the manual are to be approved by the Authorized Inspection Agency. KFP-1, paragraph 1.13 states that"Iagineering Specifi-c 9
cations (ESDs) shall be part of the program. Most ESDs appear
- to be it:picmenting procedures, but some define actual program '
c1cments. For exampic, ESD-240 establishes the Nonco=pliance Report (NCR) system. No evidence could be found to indicate that ESD-240 has been revicued and approved by the ASME or the Authori:cd Inspection Agency. {
It is not cicar which manuals and procedures are applicabic to specific activitics. The, pipe support manual is considered by
~
sitc personnel to be.a supplemcnt to ihcl pipin'n' manual. .' Thel _ t(*.
' piping manual is approved by Pullman's Vice President 6ut the ~~
support manual is only approved by the field QA Manaccr. Ilowever, "
the front page of cach manual indicates thet it establishes t.hc quality requirements for work performed under that r:anual. The v
defined scope of cach manual indicates that the two apply to - -
..udit liu. LU422
.F Page 7 cf 12
,l, 0 / /5 &/ / NkYAW NO A . '
/
t
/~ . .
e Engine' ring Specifications appear to supplceent one or both .
manuals requirements.or independently establish quality assurance program Special QA instructions are written to supple-ment and clarify EnCincering Specifications or procedures in ,
one or both manuals. .
Recommended Corrective Action (a) Urite a program description which c1carly identifics the documents that are 'to be considered part of the total quality assurance program and establish the hierarchy of the documents (where necessary obtain approval by the proper authority).
(b)
Define approval requirements for the above documents and for revisions and obtain approvals where necessary. (For exampic, approval requirements are not provided for special QA instructions).
(c)
~
Clearly define the scope of work to which the above documents are applicabic. -
(For exampic, do .requirc=ents of the piping
. canual' apply to pipe support work?) .
(d) Revicv the program to insure that suppic=entary procedurec
'do not include requirements which conflict with requirements of the procedures they supple =ent. Several *KFP procedures l require the involvement of the AI. Corresponding KFPS -
procedures allov vark to be done without AI involvement.
- 6,- ,
KFPS procedures cicarly cann6t supplemeqt EFP procedures y
without revising the KFP procedures 'to 111ov vaiving AI involvement on non-Code work. (Exampic: KFP-7 and KF,PS-6).
1 (2) .
PGancE Specification 8711 and the 1971 Code,Section III, paragraph
5
.f NA4700 require a comprehensive system of planned and periodic audits to be carried out to assure compliance with all aspects l of the Quality Assurance Program. j
. Procedure KFP-18~ states in its scope 'that it establishes such a system.
However, two types of audits, manage =ent audits and internal audits , . arc describe 6.
The procedure does not establish the scope of cither type of audit and no detailed schedule has been developed to show that all aspects of the program arc being audited. -
Furthermore, audit records at the site do not indicate that all aspects of the program are heing audited. Records do f
not indicate that management audits have beca performed on pipe supp, ort work. An unofficial, unapproved internal audit schid51c -
. .. Ekists, but it has not been fo11 cued consistently and few ESDs appear on the schedule.
states that KFP-3, -5, -9,A and March 1977 internal audit crroneously
-14 are not to be audited as they , ,
do not apply to Diablo Canyon. ' Internal audit schedules for .
October, i;ovember, and December 1977 and January 1978 were not met.
_ _ _. _ -. s._. _ . - -. - - -
- ~ ~ u-:- " " ' ' ~
> Page 8 ef 12 f
/ / /5V f Y N AY'
{ '
, Recommended Corrective Action
/ .
/ '
' Establisti and impicment a detailed audit schedule to a sure '
compliance with Specification 8711 and the Code.
(b) Nonconformance Report No. DC-78-RM-005 .
PCandE Specification 8711, Section 4, paragraph 3.11 requires that .
Quality Control personnel perform only quality control functions
, and that, they be free of scheduling and production pressures.
A review of procedures and work in progress indicates that Quality
. Control inspectors' independence from scheduling and production -
pressures is not assured by the program as written. Procedures do not clearly indicate that it is the' Production Department's responsibility to read and use the process sheet insuring that steps are performed in the required sequence and hold points "
e arc observed. . .,,
During the course of this audit, it was noted that two hold points were bypassed on FU #362 (see Section 2.2 below). ,
Discussions with" individuals involved indicated that the Quality Control inspector was expect'ed to follow the work and ensure
. ' that inspections were performed at hold points indicated on ,
. The travelcr. The Foreman apparently had not read the traveler A QC inspector should and was unaware that hold points existed.
. not be responsible for directing the course of construction to
~
. ensure that hold points arc observed, particularly if he also I,* ' signs off these hold points.
Pu11 nan's procedures identify the Field QA/QC Manager as respondible "
for ensuring that most Quality Assurance Program functions are .
perforeted. Field QA personnel had already determined that socie procedures needed to be revised to clarify or redc(ine responsibilitics ,
to ensure that production responsibilitics are not assigned to ,,
QA/QC personnel. The Assistant Field QA Manager has drafted revisions to three pipe support manual procedures and is reviewing -
others to determine whether revisions are needed.
Recommended Corrective Action ,
P (a) Revisc KFP-8 and 1* J PS-7 to c1carly state that production is
- . . responsibic for following the traveler and ensuring that hold points are observed. QC should culy be responsibic to inspect or audit. ,
L .
(b) Review procedures and practices to verify that QC is neither procedurally nor functionally placci in situations where their independence may be comproniccd. Rcvisc procedurcs as necessary.
~
Audit No. 80422 I Page 9 of 12 i ,O j)/ l Y/Eff
/ k /~'0E M d7&dd *
.s (c) Perform the training necessary to enn'ure that production and QC personnel fully understand their relationship and the functions
, they are expected to perform. '
,,,,, q ,.,.., e.j. ,. .,
, s y,o peficiencier. in Imp]cmentation of Procedures Four Minor Variation Reports (MVR) (M-3723; M-3724; M-3725, and M-3726) verc initiated by Cencral Construction for the identified deficiencies.
(a) Minor Variation. Report No. M-3723 '
- ~ . '~~.
Recor'ds indicate that management audits have not been performed by ,g((h E #
Pullman Power Products Corporation at the specified frequency. . ~ -~ . '
Management audits are required by KFP-18 to be performed at least "
,[,1
-I -
cycry six months.
Since December 1975, audits have been performed -
at cight to ten month intervals. , , - - - -
Recommended Corrective Action ' s--
Conduct audits at requir' ed intervals or change the requirements. .
(b) Minor Variation ~ Report No. M-3724
- On April 25, 1978, work in progress was inspected to verify that ~~ .
' the Field Process Sheet was being used as required by procedure KFP-8. It was noted that the repair work on FU #362 had procccded to ctop 4 on the Field Process Shect. The Field Process Shoot was
- - // "#
i# in the custody of the area QC Inspector. Inspection of the Field
' Process Sheet indicated that, contrary to RFP-8, paragraph 8.4, '
fir ej d4C
- - work had procccded beyond two hold points and the designated '
inspections had not been performed. '
, , Corrective Action . .
The F3cid QA/QC Manager issued Nonconformance Report #265 and
- agreed to write a procedure requiring the isssanec'of a Field .
Process Sheet to production. . The procedure is to clearly define responsibilitics for using and completing proecss shocts.
(c) Minor Variation Report No. M-3725
- Hinor Variation Report M-3725 was initiated to document the
,n*
,6 following hardware discrepancies, noted by the FCandE QA Depart- ~
~
ment, to facilitate their resolution.
. t, .
, . +
l .
.or av ut TE
,/* Support or
. /. Inometric .
Description of Discrepancy 47-69R Vertical clearance is 3/2" should be 1/16"
[ 47-70R
- Vertical cicarance is 1/2" should be 1/16"
)',46-17R b Clearance is 3/16" should be 1/16"
% Weld Item 2 to 1 not all around I\ 77-12SL Snubber installed on wrong pipe k'
- 77-14SL No torque seal
( 23-7V L 1.ocation Item 7 is 5/8" should be 4"
>
- 23-8V s
- Weld Item 9 to pipe not all around ~ ' .s 23-5R "
Clearance Item 12 is 1/4" should be 1/8" P "," ..23-12R Missinn anchor bolts /
(
23- 16R Crinder Coones 3/32" deep.
- C*
, Loonc. bolt
- Clearance is 1/8" should bc 1/16"- 23-66R .
C)carance is 0" should be 1/16"
\! C 947-3R Weld Item 1'is not all around 90-44R I
' Weld Item 9 is not all around -
C '90-45R
' As-buiJt does not reficct added shim G 90-47R Weld Item 6 only tacked
- I 90-48R As-built does not ref1cet added weld '
e G 96-6V
,,I - 5/8" rod used in lieu of 1/2" rod 90-46A s Weld Item 2 is 5/16" should be 3/8"
, 72-19SL Ueld Item 4 not both sides
, G 6-4R
' Weld on attachment is 1/4" should be 3/8" -
6-28R Fabrication of "t" shoc not to as-built "
g G 6-6V
- f Dinension is 3'-1 1/2" should be 2'-11 11/16" 6-8V No load on support, not tichtened 2730-61 j No Clearance "t" choc to clip
- I 2730-63 Broken stud
- _ 2730-65 No cicarance "t" shoo to clip 2730-66 No cicarance "t" shoc to clip 2730-42 Clamp Joose, tironn location -
2730-21 No cIcarance "t" shoc to clip 935-23 Brace veld not all around '
935-24 .
Brace is 45* should be 55*
935-25 Configuration oppost te to Dtic.
- 935-27 Brace veld not all around ~
=
. .- -..... a . - . -
. , _ - ---- m . __.. _ _ m --
nue2 s. s.o . c,u .. a Y
I P* l' age 11 of 12 l )?o PRt E74'Af /NEDdmisfiop' .
/ . ..
(d) Minor' varintion Report No. M-3726 f
Hinor Variation Report No. M-3726 was initiated to document the following discrepancies noted in isometric drawings to facilitate their resolution:
Support or ifescription of Discrepancy Inometric 2-3-18 ISO shows check valve as Spec 8729 Item 17 (Velan).
Installed valve is Spec 2550 (Ucston Hyd.)
2-3-19 Some as 2-3-18 2-4-418 Line 1058; dimension shown as' 2'-8" is l'-8".
2-9-478 . F.W. 858 is etched on two welds I"
. 2-32-5 7.W. 170 is stamoed 176 2-14-14 Detail for PX263 refers to pump 2-1, should be 2-3.
2-3-418 - . F.U. 1390, 1391, & 1392 are shown by the process sheet to have been performed using stainicss steel
> 309 rod. Joints are all carbon to carbon. '
. Note: Documentation was determi.ned to be
- incorrect. The correct red was verified to have been used.
3.0 Eccommendations
- Several comments and recommendations for program requirements were presented for consideration to the Cencral Construction Dept. during cxit interviews and are summarized as follows: , .
(a) Schedule for Impicmentation of Commitments ,
A schedule for i=plementation of the following cec =itments should be cstablished: .
. _ . . . . . . . .. v . .
.,. /
(Training program - added to KFP-1, '12/23/77. a.4-p .2 Use of internal audit checklists - made in draf t responses to NSC , ,.
y' nudit. .
Icaue a procedure requiring a process sheet to be issued to production. Commitment of 4/25/7S. 4- .
Impicmentation of corrective action resulting from this audit.
- A
s-
.. . . . - - - - - - . . . . . . , , . - . . . . . . xm=~ .ase~~~-'."~~~~.- ~
tsud1L Uu. U0422 f . . = .. Page 12 of 12
, *I,' l g2) /,f.?. V/?/ /k? M f WChl
/7 -
Y (b) , Pullman Prob}cm Reportitu Procedures .
These procedures should be revised to facilitate dctermining and .
verifying corrective actions. ,
/The following changes are recommended: .
Issue internal audit findings as NCRs.
- Expand the use of NCRs to cover all conditions adverse to quality which are not covered by DRs.
P p '.[, Require the cause as well as the corrective action to prevent; "' ~
I p'j. recurrence to be documented. ,
/* Establish a management review system for Df.s and NCRs to identify ,
trends. . .
. y ,
~~
(c) Inspector's Certification _
~*
The Pullman inspec; tor's certification card should be amended to - -
py'eliminatetheclaimthatinspectorsarequalifiedtoA.VSIN45.2.6
.t. p . or inspectors should be qualified in accordance .with its requirements.
(, D p A review of ESD-237 and qualification records indicates that some .
N. Pu11can inspectors are not qualified to ANSI N45.2.6.
(d) _ Description of Supervisorv Responsibilitics .,
'
- h ) The ITPS (pipe support) tnanual gesigns speciIic quality functions to e
the "llanger Engineering Supervisor". The responsibilitics and duties .
,./ /' of this position should be defined in the program.
(^;) Special QA Instruction Index c
/.r'An index for special'QA instructions should be prepared. This index
- s jr'l' should identify the procedures being amplified and the subject being addressed. .
(f) . . Update Pipe Support Procedure KFPS-7
- l The process sheet shown in KFPS-7 is Revision 7 and the process sheet g 4 shown in ESD-223 is Revision 8. The latest revision of the process .
sheet should be placed in KFPS-7 or the process sheet should be removed
.)- 1)fromtheprocedure. - ' ' '
v . .
.I O
.~ _- - .- - -
/
//s _
PAC I1T'I C G- A S AND ELECTR,IC C C M PANY
,. -. 249 M Anr.1 stas.tr . sen rnAncicco, cALIFoR rnA 94!0C = (415) 18:.4211
- Twx 010 372 65S7 l
P. O. Box 117 Avila Beach, California 93424 September 19, 1973 II. W. Kellogg Company P. O. Box 367 Avila Beach, California 93424 Attention: !!r. J. W. Ryan Project 1:anager Diablo Canyon Project Quality Acsuranco Audito Gentlecon:
Past audits conducted both by the Atenic Incrcy Com=ission and our Quality Assurance Lopart=ent have discloced nu=crou Quality Accurance deficiencies.
Theco deficiencies ucually fall into two categorius:
- 1. Failure to follow cxictinc Quality Accurance procedures.
- 2. Failure to up;rado quality Assurance procedures.
Uo requect that you place more ,enphacis on your Quality Ascurance audit program such that cost of the deficicncies noted by the Atenic Incrcy Comuission and our Quality Ascuranco Departcent will be eliminated beforo tho tico of their audits.
Our Quality Control encinocrc till call on you in the near future to acciat you in the upgrading of Quality Ascurance procedures. He vill also audit you periodically to help you to uncover the areas that need improvement.
Your cooperation in tho up;rading of quality Accurance progra=c will be appreciatcd.
Sinccrcly,
( /,<c "*).M
/ l. /(
- C. I". ::an'Jic d
,- Projoc L Ou.rcrint.'ndent O r, ,o.:s.e.c.ar, e2X
= - - - - ---
- .. - -- ~ . - -
- w. .- -. - .
i
/ - .
PACIFIC G-A S AND E LE C T RI C C O M PANY I
- 3OD2r$'!U8) 245 M ARKET STREET SAN FRANCISCO. CAliFOR NI A 94106 * (415) 7814211
- TWX 910 372 6587 l
s P. O. Box 117 Avila Beach, California 93424 March 25, 1975 M. W. Kellogg Company P. O. Box 367 Avila Beach, California 93424 Attention: Mr. J. W. Ryan, Project Manager Mr J. P. Runyan, Field Quality Assurance Manager Diablo Canyon Project Specification 8711 Quality Assurance Audit Gentlemen:
On February 20, 21 and 25, 1975, a Quality Assurance cudit was performed on it. W. Kellogg Company. This audit reviewed piping support and rupture restraint installation procedures to verify compliance to the Pacific Gas and Electric Company and M. W. Kellogg Company Quality Assurance Manuals, Specification 8711 and the FSAR. ThtLdepart#tes which..are noted-are-not -
.infyjdually..cf_ major sigrtif.icance. ,Howevery when considered collectively _
. ,as_a trend the.se departur.as r.ev.ea.1 the _need_,for a _rao.rS_Coplehensive_ and, extens~ive internal audit system.
Please correct the discrepancies noted in the attached audit and reply to the audit, in writing, no later than April 1, 1975. Indicate in the reply both corrective actions being taken and steps to prevent recurrence.
Sincerely,
, ( C.f7'Y K. Maxfield A
Project Superintendent Attachment g -
/
i A % L .c A % C
' .. _.. .. _...... 4 :-, ~ u _ . - ,,. . _ _ . _
b .
Audit Scrial No. 75-2
~
Page 1 of 5
. PACIFIC GAS A!!D ELECTRIC COMPAliY QUALITY ASSURAllCE DEPART!ENT DIABLO CANY0:1 PPOJECT INSTALLATION OF PIPE SUPPORTS AND RUPTURE RESTRAINTS (Specification 8711 and 8833XR)
Performed by: W. J. Tomei
- Date: Tehruary 20, 21 5 25, 197S Critiqued: March 18,1975 SCOPE
.The ' audit was conducted to verify that piping supports and rupture restrai nts
~
are installed in accordance with the Pacific Gas and Electric Company and !!.
W. Kellogg Company Quality Assurance I-lanuals, Specifications 8711 and 8833XR, and the. FSAR.
CONCLUSION l *Ihe results of this audit were reviewed with the Resident !!echanical Engineer to (1) discuss the results. of the audit, (2) resolve any inadvertant misrepre-sentation, and (3) establish a completion date for those items requiring cor-rective action.
The audit disclosed departures from prescribed quality procedures in the areas of drawing control, weld cicctrode control, ultrasonic equip rent calibratien ,
and P G and E surveillance inspection documentation. Individually , the depar-tunts were not of major significance; however, collectively the departures in- -
dicate the need for a more comprehensive internal audit system.
The Ecsident Mechanical' Engineer hcs directed his staff to review the unresolv-ed ite:.s , implement corrective actions, and provide a written response to this audit by April 4, 1975. ,. _.- - - - -
, . N
, t e
_ , - . __. .. rew. m _ _ . z-
, .A -.
Audit Scrial No. 75-2 Page 2 of 5 P
I. ITEM AUDITED
Ref: MWK Pipe Support QA Manual, Procedure KFPS-13 Reviewed INK welder performance qualifications for rupture restraint welds to assure conformance to AS!'I,Section IX. lielder performance qualifications reviewed included tidK welder syrrbols KS, YP, MZ, and LV who were qualified to KJK weld procedure number 7/8.
DISCPIPANCY No discrepancies noted. ,
II. ITEM AUDITED Ref: MWK Pipe Support QA Manual, Procedure KFPS-7 .
Reviewed IfdK Field Process Sheets for hanger numbers 40/24R,These40/25R, hang-40/44R, and 40/21R for conformance to ifdK QA requirements.
ers were also physically examined in the field to assure conformance to the P G and E design drawings and trJK Engineering Specification 1:3D-220.
DISCPIPANCY .
No discrepancies noted.
III. ITEM AUDITED ,
Ref: INK Pipe Support ,QA Manual. Procedure KTPS-6 Reviewed drawings in the field for pipe rupture restraints to assure they are current..
Drawings reviewed included P G and E drawing nu:rbers 438237, 439562, 439566, 439569, 439575, 443250, 447248, and 447255.
DISCREPANCY _
'Ihe KJK QA inspector for the Unit _I pipe rupture restraints was observed Revision to have drawing number 447255, Revision 2, in his possession. 16, 1974 4 is the current revisien and was transmitted to KJK on August Revision 2 of this same drawing was located in the QA office drawing rack.
Further investir.ation revealed that rupture restraint drawings were not being distributed to the MWK QA offilEirby-thqK Office Engineer.
~
. . . . - . . _ ......:;_a.__._____ ..
} ., . . . . . . . . .
Audit Serial No. 75-2
/
- Page 3 of 5 CORRECTIVE ACTION 1hc INK QA Hanager stated that P G and E rupture restraint drawings kill be added to their current drawing control system.
The Resident Mechanical Engineer stated that his personnel will audit drawings in this area to assure that only the latest revisions are used.
IV. ITEM AUDITED Ref: KdK Engineering Specification ESD-223 Reviewed field fabrication of pipe supports in the hanger fabrication Work was reviewed shop located in area "L", Unit 2 Auxiliary Building.
in process to determine conformance to Specification ESD-223.
DISCPIPANCY The following departures were noted:
Centrarf to ESD-223. Paragraph 2.4, Felder syrbol "LH" did not have a Weld Rod Requisition form for electrodes found at his work tabic.
Contrary to ESD-223, Paragraph 2.4, the Weld Pod Requisition form for welder syrbol "VE" referenced the incorrect weld specification.
Contrary to ESD-223, Paragraph 2.16, the temperature of the electrode- A min-holding oven issued to welder syrbol "VE" was noted to be 100 f.
Imum of 225 r is required for low hydrogen electrodes.
Some electrodes were observed which were not properly controlled; Rod Requisition forms were not attached and they were not stored in a rod-oven.
CORPICTIVE ACTION The E4K QA inspector agreed that this area needed immediate attention.
}MK Nonconformance Report No. 204 was written to identify the probicm areas and initiate corrective actions. .
V. ITE_t! AUDITED Ref: INK Pipe Support QA Manual, Procedure KITS-6.
N
,;.. f -
" " Audit Scrial No. 75-2 Page 4 of 5 Ecviewed the IMK storage area for Class I supports and support material for conformance~to KTPS-G. ,
OBSERVATION Procedure KFPS-6, Paragraph 6.'1.7, requires that only color coded struc-tural shapes. and plates may be moved for fabrication purposes. It was observed that the color coding requirement for these items had been dis-continued. Currently, this material is not released until certification is received at the site. This appears to be an acceptable method of con-trolling this material; however, methods which deviate from prescribed plans and procedures should not proceed until the alternate method has been approved. '
CORRECTIVE ACTION .
The }MK Pipe Support QA Manual is in the process of being revised to in-clude the above change. In the future, procedures will not be altered until formal approval is received from P. G and E. . _
VI. ITEM AUDITED Ref: MWK Pipe Support QA Manual, Procedure KFPS-2 The ultrasonic inspection of rupture restraint full penetration welds .
was reviewed for conformance to IMK Procedure KPPS-2, }MK Engineering -
Specification ESD-234, and AUS Dl.0-1969. UT examinations were review--
ed in process and the following items were inspected for conformance to -
procedural requirements: equipment, transducer, weld preparation, tech-nique, basic calibratien blocks , calibration, and inspection documenta-tion.
DISCREPANCY UT inspection work appeared to be in compliance with AWS Dl.0-19G9 re-quirements. One departure was observed in the area of equipment cali-bration. There was no documented evidence that the equipment's gain '
control was checked for correct calibration as required by AWS Dl.0- -
19G9, Appendix C, Paragraph C106a.
COR!U;CTIVE ACTION .
The INK UT inspector stated that a method would be developed to cali-
(I) FI LE (n L. LUewez_,
y ,PowE (n O.A. -
ousi.iy Ais.,, enc. NONCONFORMANCE REPORT b 3..m .:v..n.
ei ,q i . .. . v .., n eu, o.... ~ . .
1' O C 7 9 0 0 1 R M 3 sneci l or 4 TO DE COMPLETEO CY INITIATOR 2 El',,',",",,,,,.. S pe c i fi ca ti on 8831X R. Nonconformance Report CCl-78-RF-009.
3 ,'..%
, Rupture Restraints.
N o.se,*sstion o See page 2.
8 -
N 4b R M hug 3*sted Resolutson A f oo re.ma.1
> / / /*
5 ORIGifJATED.
Uh["l{ Char'iCal "I[2f!79 u S l3B70*
TO BE CC.*.iPLETED BY TECHNICAL RE(IEW GROUP 6 **"" *' "*"**"'*""*"" Pullman Power Products ' Rupture Res traint Precram has had inadeouate design chance control . inspecticn performance, and control .
, ,t
, .. . ]
n w.....m ,'
See paces 2, 3, and 4. ' %
,, 4. . ., c-y
. . - ., . -w
$ .d'-
g 7{ Y Reported to NRC7 CYES O{ BY c.. u . ... a c . n i o . ...n i 8.c o,,,,,. .
.(-3','[ 2;,ne...
. e-.
,, , c,o
, , , ol,o lo.a ) f s Pullnn Pcwer Product! has developed 1nd ?imolemented a procren ivhich ersures.a@qte19 N .
control of design chances. Training and.indectrination programs have been develcne' and i p]en'er.ted which assures adequate perfoNance of inspection personnel.
^
$ ",*,,,,,k b M ,ved. f*" P'f'247/ 7d /'2]1 d b D f *" l / 24 / 79 I! ."l':'.',. .Fr c . Jh- I '" 1 J24 / 79~ h o & #'% i '" ? / 4' / 7 9 5 E."..........fl/d_48'" 14 / 24 / 79 I Y M '/ N , M -4 "" 2 / 5'l / f a seseoutsoasMPLsTeof a.=> ion o.m yAgy 9 -g =*n'1"... p/9/f7 TO DE COMPLETED BY l'.PJLEi.1ErJTING ORG ANIZ ATION
,,,,0..o ou-o
, ...o .. o,, C.,,,......
.. ....3 7 7 E . ,, , *. - on.
letoec tion lif eeque..dl P
.' I C........ ....." C.,,.....o ;'a!N . ' 'ifb' Mfbl. "'# '9/79 y .n.,-..."t....1,... ;aa'T 7 a ..n .. . .
s.V/A l "'
/ /
o .sm.. V
- .s.
. w >
,,; w -
g/ .-
e, s.k .
- TO CE COMPLETLD UY OUALITY ASSUR ANCE 10 VERIFICATION. . D.t.
l The Resciution .ind Co et-ene Action are ca,ao ere
) 11_ OeTwisurios s oie.,, c,.n.. i ...i n io ,ce,...
O Authorired inspector (for ASME ittms O Pl. int Superinterident
.n eo,.. 4 .... e, u e.en . ....n uco . ci e k hao :
O Security
/
/
O Steam Generatico 0, Fnnin.crino Researceh f J Disoc&s . .
- s. -
'~~
',/ -
NONCONFORMANCE REPORT .Page 2 7 .
, DCl - 79 'RM - 003 O N
Description:
O N 1) a. Documentation shows acceptable bolted connections. However, there'are C cases of out of tolerance gaps existing under baseplates, nuts not 0 bearing against splice plates properly and nuts not engaged per require-N ments.
- 4 F 0 .b . D.ocument'ation shows acceptable welded connections. However, there are R cases of material and welds not conforming to the specification.
M A
y
-)2 a. Welds exist which do not have documentation..
C1 b. Modifications have been performed which were not required by the design E
drawings and have not been documented.
. c. There are bolts that have " torque seal" which indicates tensioning and inspection, however, inspection records do not exist. This has led to some ASTM A-490 bolts being tensioned more than one time. Reuse of
' A-490 bolts is not permitted by the current revision of the "Specifica-
. tion for Structural Joints Using ASTM A-490 Bolts" (Revision dated Feb. 4, 1976). Previous editions did not prohibit the reuse of A-490 bolts. ~
D Resolution: ' '
y -
. *S Pullman Power Products shall perform a documented inspection of all bolted P and welded connections and applicable documentation, required by the Specifi-O cation, as set forth in approved contractor's ESD's, in order to: .
7 S
- I 1) identify connections which do not conform to Specification requirgments, and T
- I 2) identify connections which do not have required documentation. ' '
0-
.N Deficient cofiditions shall be corrected as follows: .
~
1-a) Out of tolerance gaps under baseplates have been inspected by the assigned engineer and will be allowed to' remain in. connections not requiring rework. In connections requiring rework the baseplate shall be shimmed tight. New bolts and nuts shall be installed as (. . ,
. required by the Specification. A documented test will be performed -
to determine the minir.ium thread engagement required to develop full designed bolt strength. The test results will be submitted to the assigned engineer for his evaluation. Nuts not having full thread engagement shall be "As-Builted" and the As-Builts submitted to the assigned engineer for review and approval. ,
A )
v/ .
Y fl0NC0tiFORMAf1CE REPORT Page 3 j -
DCl ,79 - RM - 003
'l Resolution (continued) v -
1D) 'Documente'd material and welds which do not conform to Specification requirements shall be removed and replaced with conforming material and welds.
e- -
!. 2-a) Undocumented welds shall be inspected using existing final acceptance
~
criteria as required by the Specification. In addition, all undocu-f -
mented fillet welds shall be magnetic particle inspected. If a magne-F tic: particle inspection is not possible, the contractor shall submit a F.
discrepancy report to P G and E. Welds found acceptable through in-
/ .i spection may remain; welds found u.nacceptable shall be removed and I, . replaced.
- C L !..:. : - :: n - - -
i 2-b) Bedifications not required by the design drawings and not authorized, shall be "As-Builted" and the As-Builts ' submitted to the assigned engineer
- c. for review. . Modifications approved by the assigned engineer need not be removed. Modifications not approved by the assigned engineer shall be removed. -
2-c) .
1 I11'g'h s'trength' ASTM A' 325 and A-490 n'uts and bolts, previously ten- -
sioned in steel to steel connections, shall be replaced with new bolts
- end; puts which conform to the original specifications. New bolts shall be tensioned to current requirements.*
~
All high strength ASTM A-325 and A-490 anchor bo?.ts through concrete
%' I- Fes: .- walls which have been previously tensicned shall have the threaded
.? portion or portions examined using ultrasonic inspection or the nuts
.c::li :- shall be removed and the. exposed threads magnetic particle inspected.
5 F - a r.: .+. alf the. threaded portion of a bolt is inaccessible for magnetic particle 0 czh ., . inspection, a liquid penetrant inspection shall be performed. All 7 i examinations shall be documented and included in the Restraint Docu-
} i; u r mentation package.
C " All anchor bolts through ccncrete walls which have no rejectable indi-C catjons, as defined by the specification, are acceptable and shall be f- Es#'r- tensioned to current requirements.*
i-t . Bolts which have rejectable indichtions shall be discarded and replaced
.with new bolts with new nuts. If bolts are grouted in wall the con-
. nection shall be "As-Builted" and the As-Built submitted to the assigned engineer for review and disposition.
-}il.high strength ASTM A-325 and A-490 anchor bolts embedded in concrete
.which have been tensioned to 30% oft F or less shall be retensioned to
. .. current requirements.*
c :. c.< -
- Bolts.which have been tensioned to more than 30% of Ft shall have the threaded portion examined using ultrasonic or magnetic particle inspec-tion. .If the threaded portion of a bolt is inaccessible for magnetic O'
[~ -
particle inspection, a liquid penetrant irispection shall be performed.
All examinations shall be documented and included in the Restraint
NONCONF0PJMMCE REPORT Page 4 DCl RM - 003
- h' Resolution (continued)
Documentation Package. All anchor bolts embedded in concrete which have no rejectable indications as defined by the specifications, are acceptable and shall be tensioned to current requirements.*
Bolts which have rejectable indications shall be "As-Builted" and the As-Built submi.tted to the assigned engineer for review and disposition.
"'As-Builted" shall mean a complete description of location, condition, and test or inspection performed. .
Ft shall mean specified minimum tensile strength of bolt. .
. J S
- - ~
c
- y 4
4 -
o
. g
- 3. . .
. i .
- dTTAcill'ENT 1 b i .
DCl RM - 003 BOLTING REQUIRE!!ENTS FOR RUPTURE RESTRAINTS Minimum bolt tension per AISC Table 3 for ASTM A-325 and A-490 bolts:
. .70 (mir.,f)
- t Rupture Restraint Steel to St?el Connections:
. .70 (min,f) t Rupture Restraint Anchor Bolts through Concrete flalls (floors): ,
s
.. 55*(min..f)t -
g
Rupture Restraint Anchor Bolts cast in Concrete:
.25 (min. ft) -
~ Rupture Restraint Steel to Steel' Connections which are periodically detensioned and reused: -
3 . . _ . , .
.55 (min. ft) -
S min. f t= specified minimum tensile strengths of bolts s
s
- 9
}. _
^
g.
h!!C GAS ANO di1NiitE'Ui.iI3A$Y
,,,,,,, l COT
~ '
3 .
P. O. Box 117 l
. Avila Beach, California 93424
~
February 22, 1979
- MP J. W.' Ryan '
Project Manager Pullman Power Product.s P. O. Box 367 Airila Seach, California 93424 ,
l
. Diablo Canyon Project Specification 8711 .
s . . Clarification of !!CR ,,
DCl-79-R'4-003, Pcges t
s ' ' '
. 3 and 4, Item 2C,
~ '
Paragraphs 1, 3, 5, -
and 6 -
~ .
Dear Mr. Ryan:
Where reference is made to re-tensioning of embedded bolts or anchor bolts through concrete walls to current requirements, it should be understood this is applicable only if it is necessary to remove nuts ,
for ll.D.E. of the bolt. . . ,
Sincerely,
- . . t I (' Signed) R. D. ET7: cp -
R. D. ETZLER
.. Project Superintendent JAHolley:fr bec: VLKillpack .
IIEleppke ,
JArnold . '
RTorstrem .
ISokoloff -
SHanusiak .
File * -
. ,m . .
h , ,
t -
(*) $ 6 (0 f4 tuct F-3' O *
cu iity Anur nce NONCONFORMANCE REPORT (O FILE-76 ass nev. ans pi.n t rsite v e., mese oeot. N umne.
IDENTIFICATION;
- 1 8 D C 1 7 R H 0 0 9 1 2 Sheet of TO BE COMPLETED BY INITIATOR 2 oo"4,"e-enei.
- r. Ruoture Restraints inside Containment erected Der Scec. So33XR 3 ,",,7;
, Contractor's erection and inspection' of Rupture Restraints documentation.
E "'"*"*" 1) Documentation shows work conplete, correct and inspected. Work is not o
N correct. 2) There is physical evidence of work but inspection records are incomDiete 4j or non-existant. (See page 2 for details).
T '""",'l7 o" *."' " * * " Reinspect all Ruoture Restraints inside containrent. Correct all N
E deficiencies. m , , , ,
'"10/26/ 78 [b7m.,7h 5 oRtGINATED. '6'c} e'chani cal E$'2"."sE ii E1".
TO BE COMPLETED BY TECHNICAL 4EVIEW GROUP 6***""*"'""**"
f ,
nuo<oiion x Cancelled. Refer to Nonconformance Report 50C1-79-RM-003. )
o \ ^
t x -
3 g
. 7E Reported to NRC7 CYES CNO T Co.,ective Action to P,e.ent R ec.e., enc e BY 38',',,2j,n a
n.... c,,o,uo,ta
, y l0.te ) )
o _.-imi f
~
e s i r-n m a I\ T! n N fX PY g
y ; UrOVin i e " ~ -
o,0.o o.ie o
., Cs..,m.n f f I .ie f f it o o.ie o
o..i s t, 4..... nee I.e f f i f f g Other O D .t.
o n e o ..en .. .... i .te / / I / /
8 SCHEDULED COMPLET!ON ""*'""on o.m ,",y,,,
TO BE COMPLETED BY IMPLEMENTING ORGANII EON aoo,o.eo o-Re.olu. son Compieted u
g By 7 j
{e inspec tion lif ,eou.,*d)
/ /
apo....o y corrective Action compieted sy j j s u apo,o,.o inspection (if eeqw eo)
J ey , 7 7 n om .. . .
0 <
N TO BE COMPLETED BY QUALITY ASSURANCE 10 VERIFICATION:
The Resolution .nd Corrective Action .,e cc,nosete *l/p[g;ff '
'", [ C,
//f 11 DISTRIBUTION iOtre, Dem.,imenti io ,ec...e into,,,.i.dn couv oren o,+n.ied - enec= eeio.>
0 Authorized inspector (for ASME items) O Plant Superintendent O Security O Steam Gene,ation O Engineering Research O Materials Engineering g Quality Assurance Contractor 4UMrd /f"M h80-Station Construction O Safety Health and Cl aims Other
~
NONCONFORMANCE REPORT Page 2 DCl RM - 009 N
Description:
(Details)
O.
N 1) a. Documentation shows bolted connections are shirrced, tensioned and C accepted but there are cases of gaps existing under baseplates, O nuts not bearing against splice plates and nuts not having proper N thread engagement.
4 F.
O b. Documentation shows weld process sheet signed off for material R verification and fit up but there are ,:ases of material without M traceability having been installedj material installed in the A wrong place and welds made in the wrong place.
N C 2) a. Welds exist, which although correct,do not have process sheets.
E
- b. Welds exist which are not required and have not been authorized.
- c. Modifications have been performed which were not authorized. ,
- d. There are bolts that have " torque seal" which indicates tensioning e and inspection but inspection records do not ext:t.
(l.
ggEOR M t
e
. g. ,
.. W, . ..
- / . - ,r..-
u u ,., - >
W, - - i. ! w ..y/7 wG , . &g . gC-V M o* b b M
[-[o[A .(DUM
=.-
- . , f_. S k n -
I w .u.,, &b es. - _
-1
. 0 . . .e. s, / -p .? - .
w .n- ec.
'i6
=-.
= =...: 1 'R ei de ... y 6 ,..:.~....'..-. d h:T < !*wd ^* C' .0 h mb.*~ m. .. .- -
- :..=
A/5G du-sd,.., 6-9. a.d'G 1
- ) n xe. u.,,a /, egg.t.feq.c ug . _ _
e y
i _
' 1~-VXa 4 2R_.<p-d m .'Z wwM M Cdp:u; g ar.fi G M 4'V sa dU% &
- I 4,. .6- Zfe_ ! s At ca ? s t-4 a.m & u.12n $~ e. b Y *
. ne a fA> /VSC- 7$ a2v2A en .*8'6 fm %&isexZ6o-dG-nct ,b.
Y i].s$$w.lii'ne/pScs.* '/ , r
- Y -$< ~. o_c.ng .&. pfn.a -. %p'
(. .
e ~gl't., .-C.: du' .)n._s.' ._.a:nn .me.,-?~
g
- h.-h2.y ?..
. % ._ w;ti< ..a....~ .j o. . , .- ?.im. &.
EP:. 6 xn f... . , & .zk e , a .nl' . '6.G ?
' . PL, .C ., v.:./ ~ . %. .-.. _
- , - - ca....,. ___,;. . :,. <f D2.. , << v, i. . .. . . . : .- /
i' r!b;f.1. '! :Gf y:> /1 : '
.S A : *.e '46. s .:jC~~% A' e.1.4;.!r.
4,u-Y.. .h $2l' IY$/?;(f31e..c
~l n . . W) . . . c_%s6 j $.- . c. e...~> /u b - u W_.de.-]2;C.
~, -
\
u:.cc . (g',}'-' .- .-
, A1$
.ft.!1 -~ : w .ats. D ?.6:' ^? L 5 L(d.J.'W$,t, c-- -
./. ' ~.-. .a . .s..:.: f c'e (J!fe. n, ., . /~<
f ).^' t.k. !!--. 6 , </., at/ a .~.7.,/~i 4.~l
-- ~.- i--..
I sV \ . &#
bu r
f: i.C:nt w -
om- -
"c A./ 7%..
EFJa '
A- E' - - - -
- v. A!. .-
A._ iQ. . . . - -_
p, p,,e -
m
.g._
-- - ""e D e .4.e N.
' - e.J - _ em me.e m
- i. -
.'8"D."48--- - . - -*
'M . D 8 8E'."-
ab.
no me ec-o . - .c; >- -_ _.w
""t _ m
i
~* L.~r.._. . .:; .-~~ -~ u.a = : . _w
~ .
l
. /,-nen ic e n, nnu e.t.u. i ,4 u <.umn y
, ,, , , 1 COPY ;
. IGOR 50x0iOF .
I '
i[ ,
,j k, -
- t#)
- t g
October 24, 1974 l'r. E. J. Eichelini .
Vice President -' - - - - ~ - - -
{ Scatro=-Ecrges tietal Products 6 4700 Colisete Way Oak 1c2d, Ca. 9E01 Dicbio Ceyen Project CH 167027 - Spoc. 6831R Pipe T,estraint Q.A. Itatorial TC01223 ,
' bear tir. Micholicit .
f The Qur.lity /.asureco Tcquire=ents for the l!5C dulgued structures ,
which you are fumishing chc11 be cc folleva:
s I - __The
- 1. gener:t1 cyplictble receirenents are those outlined f .
in Epocification $531n cud dettdled in your Qu':11ty 4_ _ -- --
j -
, Aesuraco Esaual prepn:cd for tbct specificatiec.
L - - - - -
_J
- 2. Charpy iepset tests for cr.rbon ateci cre required for infornatica c:dy. Tcats chc11 be cade t.t +20" F. .
I r 3. Charirf frpact testa t,rc not required for type 304
- atninius etcel. (Schnitt Secel chould be so cdvised.) !
- Sincerely, it. H. Chandler c
JWoodward:Is . ,
bect PSDain ,
1
~kP16ff*:.h/VJChio/ISekoloff (2) 8' f
I AFi)cr:sttel (GT4) * *
- . . ^dtN5xUtilI ~
TdSiWa~~
< .A - ._ .
- j. .
' i yp' ,,
p ,
IGDR $0x0ioiF
...r . . .
~
. 4 .
becerbar 23, 1974
.o .
Mr. E. J. Hitiu lini '
Vice President '.
mtm ma tramt. /. ,
4100 Colicot:n t:ay '
Odhnd, CA 94601 .
Diablo C.:nyon Project s C;f 167027 - Spec. On32R :
Pi> Rastrcint Q.A. Hatcrial F001223 -
,i ,.
Nr Mr.1Mndinit -
Tid.n lottor confirns verbal isotructiona civen you by Mr. Art Valenta csd Hr. T. Woldt on Dece2.ber 20,1974. .
. All st41nlent steel rods t:uct be in cecord.uco with the applicablo .'
ASTK epselficatin deaf.r;:ctica as chown on tho deci:n drwhge. /.ny rod not f alling t.% .hin the lie.its et t.Sni cpecif tention dec1Dutica untat be opproved befera proceedin;;. The red b question in chovn. cu
,,, s. your. reference . letter M. pat;4 4 of 4.
l , . ,
b,,c.t -
p cording to lc' . Udente you vare not fully r.*nro that you k-cre to
- fc11ou the general applicch10 Qu.e.lity A.esurcnce requircr.r.cato as out~
_3ps. 11cca in Speci.ficatica 33313 rcthcr tt n cu33n. Tbio vro the datent as centioned previcuely in our October 24, 1974, lotter. j f .
s .
, Sin:croly, ,
- 11. H. MIDLER ,
Wsvoldt:sc bec:. R5. 3.e.in. _.
Annuttc1 (CFA)
- CM..1.:..:.i. i..W c1.d '
Q1St: dan .
Est.'o11ch/VJChio/IScholoff (2)/
MCJFIC GA5 AND ELCCTRIC COMPdNY copv _
50 N
f .
M
./
. N(i ' ( ,
f
' ..n...... . , .. . . . ,, ,,. ,
. . . .v.. .
' ~
- .' [ ' '. , ' January 7,1975
..'~
Fr. E. J. Nichelini '
Vice Presidnt DostrerMicrten iktal Products * -- . - - . . . . .
4700 coLiceu= vay Ock1 cod, CA 94001 Dirblo Canyon Froject *
' Cit 167027 - Spec. EE31F.
Pipo Pcatraint Q. A. I'.ntorial F001223 Dacr Hr. Nichelini:
Hits ictter vill essver questio::s posed in your ruference letter 60, dated Decccher 27, 1774. '
Job directica caquence vill be es fo11cus:
) (1) .
~
Thr.se I Pipo k' ell
. These II Arcs CE/G'J .
- Thasc III Pipcsrcy
- Phase IV Turbine building You have aircady received ::stked up drmingo for Phcsc I choeing our prepo erection seques:ce.
o vill furnfrh crection ccquencco for cach of the re-dn-ing phre:ca, if our 11E.1d porno:mel think it necessary.
Finni relecae daten for Nuc1ccr Scrdce Corpore. tion design (2) drzvings recatu the our.e 4:s stated in their October 2c,1974 1ctter cy. cept for Phr.cc I .thich we received Docenbur 26 cad Decer.b er 31, 1974, cad the cenponcat cool.ics uster Iicot Teu F.y.cher.r.cr Jet Icping,c cat Ecrrier uhich L,. o vill vocks be to deluyed.
these dutc2 vill still need to s.dd a ntninum of for Pccific Cas cnd Elcetric Cenpcay coordication before you vould Tccciva them f,nticf pt.ted tire required to cpprove your shop drsvings vill (3) be eppic: tcr.tely three vecks. .
/g . . ..
.)!C G55"5'ND ELECTRIC COMPANY CO"k_N
.IL IGOR 50 x0LOF7 i
- 2'- Januuy 7,1975 Er. E. J. Nichalini "On Jcounty 2,1975, you ep.in naked for a clcrification cs to vbich cpecifi-
. cation you were to folics for Quality Moursnce, workusaship end cr.teric1. -
As you stat .,
end wo etend corrected, Central Quclity I.csurance should be
+ in accords: tee with Specification 8831R, Msterial cad vork2:enship should be in accord.ccc vith Specificaticu CG33. YR. NSC dealgn dr wings,already reflect ,
this requirerent.
- J j Sincerely, H. H. Chandler Ttvoldt:1v i
bec: J;SAsin
/dT-ansttei (GFA) *
. %fRForbas/f/alents
- CFMaaficid. '
'clG2:=> j EPifo11sk/?JC11o/ISokoloff (2) / :
e
- O G
t e
- e
- k i
- e k
e
17
- r. pp GOVERNMENT ACCOUNTABILITY PROJECT Institute for Policy Studies '84 JUN 21 #10 50 1901 Que Street. N.W.. Washington. D.C. 20009 (202)234 9382 i .--
June 13,1984 The Honorable Leon Panetta U.S. House of Representatives 339 Cannon House Office Building 5lashington, D.C. 20515
Dear Congressman Panetta:
My name is Thomas Devine. I am the legal director of the Government Accountability Project. I am writing on behalf of a Diablo Canyon whistleblower whom I represent. He is fed up after repeated efforts to work within the system at Diablo Canyon, as offered by Pacific Gas and Electric (PG&E), its contractor Pullman Power and the Nuclear Regulatory Commission (NRC) representatives of Region V. He is concerned that tne plant is being licensed on the basis of false statements, while potentially dangerous conditions remain uncorrected. ,
He has taken his evidence to the NRC, which has frustrated him by not responding at all, or by accepting PG&E/ Pullman responses uncritically.
He is angry that the NRC compromised his anonymity, which he believes has led to increased harassment.
He seeks the assistance of your office in persuading the government to respond honestly to legitimate concerns that could affect public health and safety. In order to avoid becoming any more of a public figure than .
the NRC has made him already, he asked me to submit to you the information
- summarized below.
I. MISLEADING OR MATERIAL FALSE STATEMENTS Like many whistleblowers at Diablo Canyon, this employee submitted a sworn statement detailing his charges to the NRC, after the system at the plant did not respond. He believes that PG&E's answers to the allegations contain misleading or material false statements. That is no basis on which to officially bless Diablo Canyon for commercial operations. Illustrative examples are listed below:
- 1. One of the employee's charges was that Pullman's quality assurance (QA) manager did not respond to a 1982 inquiry over a potentially widespread problem with baseplates that are mounted over concrete voids. The employee was concerned that the voids could affect the baseplates' ability to bear heavy structural loads for which they are responsible. (See March 1 2.206 petition, GAP allegation #199, p.12).
s a
, e i L
PG&E answered in part that the employee was at fault for raising the issue through an informal memorandum. In an April 30 letter to the NRC, the utility; stated somewhat stuffily, "Had he documented it on a DR [ Discrepancy Report] or DCN [ Deficient Condition Notice] in accordanct with established procedures, the problem, if indeed there were one, would have been properly addressed." (April 30 letter to the NRC, DCL-84-166, at p.12).
PG&E's response is misleading. It fails to add that management had verbally instructed him nct to write up such problems on formal reports. He was following orders by disclosing QA problems on informal notes. This policy is dangerous, because the reporting system insures that QA problems receive full engineering review by PG&E and are monitored by the NRC for their safety significance.
Unfortunately, the coverups are getting ' cruder. Other witnesses have described 1984 written instructions for inspectors not to write Discrepancy Reports.
This is one of the problems the NRC says is too insignificant or repetitive to interfere with operating the plant. Perhaps that explains why management is so arrogant that it is attempting to close down the reporting system just when it is most needed -- on the eve of operation. Perhaps tha't also explains why we still don't know the condition of Diablo Canyon on the eve of operation. Possible QC violations such as with the baseplates were reported on notes which management now claims it "has not been abla to find, or even recall . . ." (PG&E April 30 letter, DCL-84-166, at p.11). Luckily, the witness kept his copy of the " lost" note in this case.
- 2. .The same witness was also concerned about the use of a copper backing bar for certain welding without first conducting qualification tests on the effects. Cracking occurred in welding done with the copper backing bar.
(March I petition, GAP allegations #176-179, at pp. 6-7).
In partial response, PG&E claimed "there is no documented evidence that the inspector contacted his leadman or the QA/QC Manager regarding his concerns a about the u.se of copper backing." (.PG&E April 30 letter, DCL-84-166, at pp.
20-21 ).
That is a false statement, unless the documented evidence was destroyed.
Esen then~it would be misleading. The whistleblower and another inspector co-authored written findings about the problem on a process sheet. The QA Manager certainly was not ignorant.. He admonished the co-author for writing up the problem report that PG&E now says doesn't exist.
This is another one of those issues the NRC hasn't yet resolved. The agency has not contacted the witness to check the accuracy of PG&E's excuses.
- 3. The witness also had disicosed how un. narked t.als were used on-site for welding on stainless steel pipes. This could lead to metal contamination and cracks in the stainless steel. At the time, the employee wrote up a Deficient Condition Notice (DCN) on the problem and tried to stop work with a F Id tag until the controversy was resolved. Management ordered him to remove i ie hold tag and then waited a year before " resolving" the DCN by rejecting it.
, March 1 2.206 petition, GAP allegations #195-96, pp.11-12).
- _ . ._-._ _ _ u . _ _ _ _ _ _ _
0
~
The NRC's response illustrates why this whistleblower has lost faith in the agency. A few weeks ago the agency informed the employee that witnesses Further, the NRC explained on-site had not backed his recollection of events.
that the dale of the DCN was two days after stainless steel work had stopped, so apparently he had been mistakan.
~
But the whistleblower : L_
On the surface the answer sounded reasonable. He had been an eyewitness to realized that something was seriously amiss.
the stainless work and submitted the DCN two days later. The kRC hadn't Second, the NRC talked with the witnesses bothered to check that detail.
- accused of wrongdoing. They didn't check with the whistleblowers for wit-The employee concluded that the NRC was either nesses who would back him. The staff promised to consider further being duped, or trying to dupe him.
inquiry but has not yet agreed to talk with proposed witnesses.
II. COMPROMISING CONFIDENTIAL SOURCES The whistleblower feels personally betrayed by a new NRC policy to turn Even over all employee affidavits and exhibits to the utility for response.
with his name deleted, the utility was able to identify him as The the utility source of numerous allegations after the NRC turned over his evidence.
then published his name in legal briefs.
This policy puts employees in a Catch-22 dilemma:
either disclose allegations in an abstract manner which the NRC will then dismiss as too vague; or disclose problems with specificity, which will make confidentiality Either way impossible once the statements and exhibits are turned over.
the public is hurt -- through exposing whistleblowers to reprisal, or by drying up the free flow of information on safety issues.
There are That is why this employee is turning to Congress for help.
enough it worse.
coverups at Diable Canyon due to management, wi interested Congresspeople, in an effort to restore Diablo Canyon to legal accountability before it begins commercial operation.
Sincerely, ,
\ AIyrvM?
Thomas Devine
4 u , . . . . . . . . ~ . n .~ n . . n...----,-..~..--r---- -
r.s .g;-y..- mw.wism map.~
n:+Ai.t,.N.,tw?..?w:..
~#
m .-
- . m a n .
M A..m..
.w. . !:.:
-x.%.
..s-ip.s.. .:s.n. : 4pe..k. s mx.L ; .,4.w.
2 r:c,',.*ce.:j.mw .ir,.+r..asu.i.:...L
. . . . .g, ,2 .. . ; .e.=..-
. ... .;.3. c. gr. .:
.,o ;, . . ;.4.m.. s.. . ,
giMW.h.y%..O,@f.dD.if.P..QNO3Fd;4!!OidfiW.Ni-Ti?DJh.7.'.?f.'eliY.25 f./. UY!.'.J9 '
'9 s - M-
~
18 INTEP. OFFICE COi!RESPONDENCE
,84 JD,' 21 A10 :50 DATE MARCH 14, 1984 r
-w" i r , '
To ALL QA/QC AfiD EriGIfiEER! fig PERSOfitiEL FROM HAROLD KARflER/ SKIP CORfilSH suancT RESP 0f4SIBILITY FOR PREPARAT10ft 0F DCf!/DRS Effective immediately, al? eeficient cor.ditic'i: shall ~u e rcpa tcJ on a DCl;, in accordance viitn ESD-268.
. QA and Engineering, during the review of each DCli, will determine whether a DR is warranted and will adhere to the requirements of ESD-240. This directive supersedes all other memos with the exception of arc strikes.
e -on- k / b k N 73 9 Harol( W. Karntf / .
QA/QC Manager
/ .
4 Skip Cornish Chief Field Engineer ,-
HWK:SC:EW: sam 4
. e-. .
1 ^