ML20082H492
| ML20082H492 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 11/18/1983 |
| From: | Stokes C GOVERNMENT ACCOUNTABILITY PROJECT |
| To: | |
| Shared Package | |
| ML20082H478 | List: |
| References | |
| NUDOCS 8312010180 | |
| Download: ML20082H492 (37) | |
Text
{{#Wiki_filter:. 1 AFFIDAVIT 6f My name is Charles Stokes. I am submitting this affi-davit freely and voluntarily without any threats, inducerents or coercion to Mr. Thomas Devine, who has identified himself to me as the legal director of the Government Accountak.ility Project of the Institute for Policy Studies. This statement evidences my concern over unreliable and illegal engineering practices at the Diablo Canyon nuclear power plant, Units I and 2. I consistently raised my concerns through proper channels on-site, without success. On October 5, 1983, I disclosed three of the'most signifi-cant problems concerning violations of legal requirements for . the design and quality assurance of pipe supports, welds and other hardware at Diablo Canyon. I believe that these deficiency reports, on top of the other problems I identified, led tn my subsequent layoff, effective October 17, 1983. This lavoff represented an illegal reprisal which I an determined to challenge. On November 14, 1983 I filed a complaint with the U.S. Department of Labor; This affidavit is in support of mv ne l .jpg complaint.
.8no l W@ I have worked as an engineer since 1975 and an a oe gistered 1 o mg Professional Engineer in three states -- Alabama, Florida and 8 Georgia. I also have a l'aw degree. My engineering experience o<
l (), includes work on coal and steam plants, as well as sever nuclear
'88o plants and the Lawrence Livermore National Labora~ory.
t I have never lost a job for performance-related reasons, and received letters of commendation for my work at two of the nuclear lants. ffWf)f /f4nc dan $rsk+Ny $4d od M 1Ajr frArmMc.e.(ddf-After an, extended vacation, in September, 1982 I sent out resumes to obtain a new job and return to work. On Wednesday 1-
, e evening, November 3, 1982, a Bechtel supervisor called to ques-tion me on my knowledge of nuclear steel, design and seismic standards. The conversation in essence was in interview to work at Diablo Canyon as a field engineer. On Thursday morning, Ken Plam, the owner of Code III Associates, Inc., a technical manpower placement agency with a contract for PG&E at Diablo Canyon, called and asked if I could report for work by Monday as a field engineer. My responsibilities would cover pipe supports and pipe stress design on Unit 1.
There was only minimal training; the other contract employees and I received our initial hanger pipe support assignments on the first day. We also received an example of Bechtel calcula-tions to be used as.a model. .We did not immediately receive controlled documents to accompany our assignments, however, as routinely occurred at most other nuclear sites where I had worked. A c,ntrolled document is one whose revisions and updates are l j controlled by the Quality Assurance (OA)/Ouality Control (QC) l Department. The document is either an original or a Xerox that is verified by a control symbol. The documents contain the areas to be checked, procedures to be fo'11 owed, and relevant codes and standards. The document is assigned to an engineer, who has a 1 complete package of all pertinent design material for his area of responsibility. QA/QC insures that the engineer receives the latest revisions as they are issued. A major function of con-l trolled documents is to insure that all work is done to current l requirements. . l
- i I waited around two weeks to inquire why I was not working to controlled documents. By that point I could not continue without the proper paperwork base. Not only were we workina to l
I unverified Xerox copies, but they were incomplete. Beca'use the
f documents did not have.all required information, engineers were working to records of example calculations they had brought with them from other plants, such as Susquehanna, Davis Bessie, Midland, Maguire, Catawba and others. It was as if each engineer was working from his own reference library of resource materials drawn from his or her unique job history. The Final Safety Analysis Reports for these other plants involved different commitments to different issues of various codes than we were committed to at Diablo Canyon. Similarly, they assumed load S OwsT ratings -- allowable design ,_ = for typical conditions -- for other manufacturers' equipment that may not have been appli-cable to Diablo Canyon's design. The controlled documents reduce engineering judgments by spelling out assumptions, based on the nature and location of the hardware. To illustrate, the size of gaps between hardware , controls the definitions, such as whether an item qualifies as a restraint. In the absence of controlled documents, engineers make assumptions based on their analogous experiences at other plants. I know from discussions and reviews of ' engineering cal , culations that the Diablo Canyon engineers used assumptions that were inconsistent with those on the controlled documents. Based . on my reviews, the assumptions used in fact differed from those on controlled documents in 30% of the cases. This is not a comprehensive assessment of the discrepancies, but an indicator that the problem is significant. If the calculations were reviewed now to the latest revision, there probably would not be more than a handful where all the assumptions are correct. Even that estimate may be optimistic. In other words, the engineering calculations at Diablo Canyon were based on conditions and commitments at other plants that may or may not have been identical with ours. The whole point of doing separate individual engineering calculations is to take into' account the unique conditions at each plant.
i . 4'- I alerted my pipe support design project lead, Leo Mangoba, that this flaw could cause the NRC to close down the design office. He supported me and in December sent a memo to Marvin Leppke and Robert Oman. Mr. Oman was the lead of the on-site project engineer group (OPEG) and a Bechtel employee. Mr. Leppke was his assistant and a PG&E employee. l
, To my knowledge, there was no written response to the memo-l randum. In early, January, I inquired about a response. Kevin Waugh, Mr. Mangoba's assistant, informed me of the. answer which had been passed along verbally: we didn't require controlled
! documents and were not going to receive any. This response upset me, because the answer was clearly wrong and assumed we did not care enough to.'do our job properly. l Along with Kevin, I went to Messrs. Oman and Leppke's. joint 2 office, where I notified Mr. Leppke that if the NRC were made aware that we were working to Xerox copies of documents from I other plants, the design office would be closed down. I added i that I would tell the NRC, unless I received the necessary con-l trolled documents. Mr. Leppke informed me that I was not going l l to get them. We left. Within two days, however, before I went to the NRC, I l received the necessary controlled documents, in this case the M-9 design guide for engineering. .While this allowed me to work to current information pertaining to the right plant, not everyone else did. In fact, for an extended period I was the only engineer
.besides Bechtel supervisors who had controlled documents.
Sporadically, the other employees received controlled' documents with the latest revision. The process was not complete to my knowledge, however, until August, 1983. The significance is that, with exceptions such as my own case, until August, 1983, none
- ._ =-
of the pipe support calculations are reliable. To insure that - the calculations were based on relevant data, they must be compared to the data on the latest revisions of relevant design guides. On February 4, 1983, I challenged a second major problem, involving unreliable load ratings for U-bolts in Unit 1. In lay terms, a load rating specifies the maximum amount of force in a given, direction that the item can safely withstand, based on a 95% confidence level in the rating. I had learned, however, that the supplier's ~(ITT Grinne ratings for U-bolts that we had purchased in some cases we one-third to one-fourth more j stringent as those claimed on Drawing 049243. Drawing 049243 is a PG&E document that specifies acceptable spacing, strengths ! and load ratings for small bore pipe supports, including U-bolts. It is used on-site as the final authority for relevant analysis. ! LTo illustrate, ITT Grinnell only claimed that the load rating [ was 485 pounds tension for 1/4 inch'U-bolts on 1/2 inch pipes i under " design normal and upset conditions". By contrast, Drawing 049243 credited the same U-bolts with a load rating of l 2000 pounds, over. 4 times more than the vendor would vouch for. I have worked with vendor catalogs for years, and the ITT Grintiell catalog is particul'arly common. I had never before seen the custc2.er assert that a part was 4 times stronger than the seller claimed. In my opinion, 049243 would represent a false statement if it were furnished to the NRC. Obviously, PG&E relied on a series of suspect assumptions in order to exaggerate the load ratings. Three examples out . of many should illustrate my point. The first example involves the interaction equation for the combination of tension and side-loading. The sum of the two factors, which always involves (se
. = - -.
decimals,.cannot exceed a value of "one". On Drawing 049243, however, there was a handicap.to unrealistically increase the odds of passing this test: each factor wds squared. Since the factors had values less than one to start with, squaring tnem produced smaller decimals, distorting the values and reducing' , the chances that the combined factors would exceed a valne of .
"one". To illustrate, if two factors each had a value of .'7, their sum would be 1.4 and the bolt would fail. But .7 squared
- is .49, which means their sum would pass with a. 98 total due to rigging the equation. ITT Grinnell does not scuare these,fac-4 tors, nor does any other plant where I have worked in.the nuclear industry. At Diablo Canyon, management has declared its'own version of engineering reality. I do not believe it is justi-fied in this instance. .
Another flaw in the U-bolt load ratings on Drawing 049243 came from the September 28, 1978 PG&E "U-Bolt Test Pro. gram, Revision 1" report. The original test was biased by not reflect-ing actual conditions in the plant. The report concedes dis-crepancies but claims they make the results more conservative. That is inaccurate. To illustrate,.the test report says that stress on the U-bolts was reduced at Diablo Canyon compared to the ITT Grinnell load ratings,due to the absence of shims in certain cases. The report failed to mention what I obssrved in the field, however: in order to' avoid shims, the U-bolts had to be forcibly bent, failing the bolts at installation. In Unit 2, this type of problem was fixed. The U-bolts,had been bent so far that new holes had to be drilled to reinstall them. Management did not want to make these repairs for the same flaws l in Unit 1, however. To my knowledge,'the repairs did not occur during my time at the plant. The report states that'the test apparatus to determine load l ratings had used " extra heavy schedule 160" pipe. Unfortunately, I
4 the pipe in the plant predominantly is schedule 80, which is (@O5) ButKIwv thinner and therefore could create a failure due to e of the pipe itself. As a result, the load ratings for the , bolts should be reduced to compensate. They weren't, and are thereroua unrealistic. The problem is even worse, however, because the U-bolts also are attached to schedule 40 pipe, - which is thinner yet, and copper tubing so thin it can be bent by hand. The effect is that, even if the U-bolt load ratings ere accurate, they would be unrealistic due t BacK/M& of QQ
) possibility o := ;c. ,cnt. by .bhe piping thinner than PG&E assumed.
d ptAod A . third major flaw is that the PG&E tests were sene at ambient, or room, temperature, but the ITT Grinnell ratings were made at 650 Farenheit. Conditions in the plant are hardly room temperature. Well over 50% of the piping lines exceed ambient temperature. Many lines exceed the ITT Grinnell 650 F. assump-tion, with temperatures up to nearly 1000 F. on some lines. On February 4, I informed Kevin Waugh of what I had . learned..; t The flaws described above are just a few of the problems with
- the U-bolt. load ratings. Afterwards, I was permitted to rely l on ITT'Grinnell's load ratings which allowed me.to fail some of l the U-bolts. To the best of my knowledge, however, the other engineers in the group continued to.use the PG&E load ratings.
I know, because they told me in discussions that I was right, ! but the. group leader had instructed them to use PG&E's ratings. j I also checked the methodology ~of my critique with a Statistics professor at California Polytechnical Institute, who confirmed that I had used the appropriate formulas to develop my analysis proving that the PG&E load ratings for U-bolts were unrealistic and should have been roughly the same as ITT Grinnell's. Even if the load ratings for U-bolts were accurate, the hangers to which they are attached would not meet design require-ments. This is because other relevant facters also were defined M i
i i out of the process. The point of the load rating is to determine how much force will displace the bolt,. and possibly cause .the hanger to move beyond legal commitments. But, as seen below, other parts of the hanger besides U-bolts can lead to excessive movement. For example, the instructions for review of Unit 1 small bore piping direct the engineers to assume that the base plates are. rigid for bolts and sup-ports. The problem is that they are not; many of them move. The plates ! could lead to many times greater disp 3 m:cment than the U-bolts, but this factor was defined out of existence. Also, the maximum displace-ment allowed per M-9 for the support was .025 inches. From the U-bolt report the failure point for the U-bolts was .025 inches. Later, the j support steel (not including the base plate) was checked to a .025 inch displacement, ignoring the .025 inch a'1 ready included in the U-bolt load rating and the displacement at the support point caused by flexi-bility of' base plate. At the end of February, I learned of a third major problem. g I was assi'ned to perform pipe stress calculations on code break sys-tems; many were high-temperature lines. This assignment involved detennining support requirements for code breaks, which are the boun-daries of Class I critical systems,.and other critical seismic systems. We learned that there was not enough off-set, or space between the j valves and the large bore piping, to avoid unacceptable stress on the small bore pipe line that branched off the large bore lines.- The code i break valves were located on these small bore lines. The group leader, _A Mr. Kim, at first did not believe me. But after reviewing my fig-ures, he agreed and supported me. Kevin Waugh a'nd I went to see Mr. Leppke to recommend the
~
necessary corrective action -- additional spans of pipe to increase the distance between code break valves and large. bore piping. Leppke refused to modify the system because the lines had.already been hydro-tested. He' challenged our "M-40" calculations for accuracy, and said ha instead would use the ME 101 computer stress analysis to analyze these lines. He then proceeded, however, to let it slip that the sys-tems probably would not work, and explained why. The vendor had not. received correct instructions when installing the lines. The vendors were told to install the piping at room temperature under normal conditions.
_g_ That would be accurate for a fossil plant. but was wrong for the ! lines in ~ question, which r,equired seismic. supports and had to ~ endure temperature ranges in ex' cess of 650 F. These two fac-tors drastically increased the stress, necessitating greater
- distances between valves and large bore piping.. Leppke even
, admitted that the lines would probably have been installed.right if the vendors had received accurate instructions. Eventually the computer run confirmed my critique, but I am not aware that '
all critical lines were rerouted,.
/
In the' ory, we reached a milestone in March,1983. ~ Supposedly we had finished the cal'culations for review of Unit 1. In fact, they were generally in a preliminary state. Reverifica-tion for accuracy was necessary for final issuance'. At this time we learned that the personnel who had been doing.the calcu-lations would be split. Half would he transferred to a new trailer. P The transfers displayed. a consistent disturbing trait: those who had not questioned suspect assumptions verbally imposed by Bechtel supervisors all went to the new trailer. Those who had been objecting to unreliab'le assumptions stayed in the old trailer. The engineers in the new trailer would continue working . on Unit 1. To review Unit 1 calculations, the revamped. crews were filled out with new recru~its,.many without background on the issues at.Diablo Canyon and without security clearances to enter Unit 1, which is necessary on occasion to check the hardware. Those left behind received new assignments for Unit 2. After this point I kept close tabs on what was" happening ' in the Unit 1 trailer, which was easy due to my. daily association and' friendships with t,he personnel. I learned that they.went beyond routine citation check reviews of the calculations, their official mission. Instead,.the engineers redid the calculations CV _ . _ - - - . - , , . - . - - . . - , . . -- ~.... ..-. ,_..,--- -- -, ..- -.-..-.- , . -- _.
l entirely for nearly all the systems which had failed. In the process, the original calculations demonstrating system failures vanished- It is important to realize.the scop,e of this effort . to rewrite,the results. Prior.to the split of the trailers, the failure' rate for the entire small bore pipe support group had been around 50% swt /MorE, N) Those results would have' meant,that Unit 1 failed the re-evaluation program and recuired complete reanalysis. Bechtel and PG&E management could not accepp.this, so they had the com- ! pliant engineers rewrite our prior failing calculations. One problem is that the ensuing packages did not disclose that'the calculations had been done over. There is no mention of.the
- original failing calculations, which were destroyed.
Further, the calculation logs do not refer to the packages which had been originally prepared. Someone rewrote the calcula- ! tion log to delete the references to the original packages which had failed. The logs have been falsified tal /t/ 0M C 43 , 7e/( co6 MW.AGCLGCTS FIJd o*T 00 *t*A6 THed 100 CAte#Artd utfM [ Pte#4&nb, The only way to reconstruct an accurate history of the , calculations would be to ask each engineer for his or her personal copy of the. assignment. Even that may not work. In many cases, i engineers did not keep copies,. due to Beclitel's policy that the calculations were company property. When Mr. Mangoba told us in I early March that Unit 1 calculations would be redone by others I in Unit 1, most of the affected engineers objected to modifica-tions of prior engineering calculations by a new staffer when the originator was available. This violates normal industry practice. He answered that once we signed off and tiurned in the calculations, they were Bech.tel property and Bechtel"could do what they pleased, including destroying them and having someone else rewrite them. I will share examples of the resulting [
o . - 11 - falsifications of records with the NRC or law enforcement officials. To the best of my knowledge, these abuses were lLmited ' to the systems which',had failed. Those that passed received I the normal checks on assumptions. This system was used to implement engineering judgments that could not withstand scrutiny. Management's first approach to make Unit 1 look good was to reduce code break spans by , relocating the code breaks. .This would have reduced the total number of supports in the safety related systems and omitted many of those which had failed in the. review program. However, management backed off after failing to come up with any plausible explanation for the NRC to justify relocating the code breaks. Next, management decided to use new assumptions that would change the results from " fail" to " pass" on the systems. Typically, engineers would be instructed to assume gaps that did l not exist and vice versa. Another technique was to assume joint releases for rigid connections, which means that' welds p ert. M which were in place whese (assumed to be nonexistent. . To my j knowledge, there was no' attempt to actually remove the welds.
~
l l Reality was only changed on paper,'not in the plant. Those are just a few of the modeling assumptions that were changed. Up , to seven different models were tried, with seven different ' analyses. Some of the hangers still failed, even after the assumptions had been switched repeatedly. At that point, managemen.t request-l l ed the designers of the hang.er supports to do a reverse calcula-t ro l tion tha t -vu(ld etermine the maximum loads that each hanger 1 could support. I now of this practice because I received such assignments. kbbpfeticealmostneveroccursintheindustry,
12 - because it is the equivalent to providing the solution to an /gef) undefined challenge. It is a temptation to juggle the data in CAM.acA9wd L question in order to satisfy an answer that already has been - supplied. . That is exactly what happened, in reverse order. After , the maximum loads were established, results. were'. returned to the stress group. This group norma 11.y supplies the raw data on the loads, so that the support ~ designers can develop ade- - quate structures to reduce the struss to acceptable levels. Now, however, the stress group received the results of what a given support could handle and was asked to change their minds about the models and resulting loads.that supp. ort needed .to handle. I do not know if the original calculations of the stress group . were destroyed and removed from the official history, as in our . group. I do know, however, that the stress group was also suffer-
~
Lg/H3 ing internal' turmoil with three or four changes in the pipe a (Ges) stress engineer over a year. The recently disclosed anonymous 52-page document refers to the manipulations within the stress group. S *# "' " #
"Y ' h N Where this didn't work,' management tried stil'l.another technique. New supports were added within six inches of the' '
existing supports which had failed. The stress group then modeled new support gap assumptions so that the new supports wouldhandlemostoftheloadforwhich.thefailng,. supports had been responsible. This " solved" the problem 3 becau.se the failing support could handle the small load that was left. I doubt that all of the assumptions on gaps are accurate. Even if they were, however, this approach represents a program of
- undocumented modifications. The new supports did not have control numbers or document numbers, so they were not reported to the NRC. The new supports officially did not exist. My
~
net reaction to the developments since we " finished" the calcula-tions in March puts the series of abuses in perspective. I : have never before seen mana.gement at any plant that was so'per-sistent about accepting work that-had been done wrong originally. UNt7~ % W In June, I was transferred to theg" Quick Fix" group,.offici .
. ally known as the Pipe Su port D[esign Tolerance Clarification (PSDT9) group. Quickhhhwasanextensionofthepipesupport group, with responsibility to revise drawings so that hangers could be installed without later rework.
During this assignment, I continued to learn of inaccurate engineering records. For instance, in July, 1983, I learned
. that a pipe was resting on a unistrut -- a light gauge. steel section to which electrical apparatus is attached. This' meant that it was an unintentional restraint which needed to be removed.
Instead of making the necessary repair, however, the pipe stress designermodeledtheunistrutasasupportforthepipeghd The plan backfired. When the data on the new support was fed into the computer, the results incateddi that another support had to be added for the other side of the pipe. You can't have one-directiona(S/5+fC tAsupports under.ME-101, the computer pipe stress . program. Further, upon inspection, it s g ioug g t the unistrut was too small to even support er: c ide . . gglgon" was then t remove the original unistrut and add two full-sized support , -:..i c h 'c a r three time: thc v u. maua w mumi La 52 n l trying to aveid. hW After consulting the stres.s group originators, I deleted the new supports. Bythispgng,pheunistrutalreadyhadbeen j removed in preparation.for the+c installation. There weren't. supposed to be supports at that location anyway. The extra work would have been due to the official fiction created be ced l l l t
Gee) avoid a lesser job /N SWar - In July I v'olunteered 'for nig t quick fix work. I,was one of two such engineers working on Unit 2. Additionally, I was responsible for snubber substitutions at night in both units, and the intake structure which serves both Unit 1 and Unit 2. 4 One night, the Pullman hanger engineer at intake called for help.. When I arrived I learned that while drilling holes to install anchor bolts in the. concrete walls, the crews inexplicably had EM!amseirrte The construction night shift mechanical (& A zum .izzto two pieces of wood. lead reviewed .the matter with me(odd) , as did the 'Foley lead engineer and a Bechtel production superintendent. The concrete appeared to have lacked sufficient cement in the mix, because it was soft and powdery, as if it had too much_ sand. The wood was visible. Foley was responsible for removing the wood. W*sde (C&ll I prep ::d to
- replace the anchor bolts with thru-bolts, which do not rely g g i
g ansch on the concrete for pullout restraint. To support this quick fix modification, as a precondition I required Pullman to write a Deficiency Report (DR). on the wood. and concrete. A Foley crew then removed only one of the two pieces of j wood and grouted the hole.. To.my knowledge, the second piece is still there. I am not sure of the DR's disposition, but dis-cussions on-sit c g e,rg y t g g yf nilmerous deficiwnt
- concrets poursj I am not aware of the solutions, if any. -
In August, 1983, I learned that coverups about defective ateri g ma g veg c g__ r. 3 , a_
, 4 , w __,2 __2 g er
. a from Pul m n Power,
. , _ , . + -
w_ u. . m._
,_1__
.c _----o told me of problems with a eme9e springcany -- the U-shaped member attach-ment bracket -- on a 20 inch line. The springcan had to support l an extremely heavy load, probably over 50,000 pounds. The engineer told me g tg ere were some symptoms of e'xcessive punching on the passe, which had become deformed and concave. l
/ta)
M :
[g y,,Ltr- ifr?'frftfD 76 88 AM M C He ziso had 4the coating removed, which revealed what appeared to be cracks of an unknown depth. As a result, the engineer had this piece and another ultrasonically and magnetically tested. Both pieces failed both tests. He then ordered two more pieces. On visual inspection they appeage g g i g to-the first pieces, so he'had the new springcansg tested also. Out of the two new pieces, one failed for a net 75% failure rate on these random samples.
&VA) iem-m Unfortunately, the engineer's supervisors assmed what he was up to. They told him to stop, reassigned him to another job and replaced him with another engineer who followed instructions and limited himself to visual examinations on the subsequent replacements. Around the time of the transfer, the engineer showed me the hanger and the'four pieces in question. The cracks were there, just as he had described, along with evidence of deformities and multiple punch ssN&, (4'*J
)
My success in obtaining a. Discrepancy Report for the intake concrete was somewhat of a coup. During initial QA training, we were told that any engineer could write Diccrepancy Report's if we found something wrong after QC had finished. DR's are the engineerin egg alent to Nonconformance Reports. A Design Change Notice is the formal document for engineers to initiate modification in response to QC inspections. To my know-M ledge, however, no pipe stress a support engineer besides myself wrote a DR or a Design Change Notice during my employment. In fact, we were not even familiar with the form to use. However, Foley and Pullman regularly,p,repared these documents. At the end of the summer, I realized the necessity to write DR's on several major problems which I had been raising to management without success. The flaws were generic and applied .
to both units. I had been attempting to convince various superiors to' write the DR's in order to minimize my own risk of getting fired. Everyone I talked to supported me but refused to stick their own necks out by signing a DR. They either explained that they were u'p for review, or that they , did not want to jeopardize their positions in the company. As a result, I knew the only way to get anything done was to write my own DR's. I searched out and found the forms one night, made five copies and prepared rough drafts for three DR's. The first involved a subject I had worked on since Ju3y on the inadequate design o.f inaccurate design drawings for d's) and improper installation and OC inspection of certain welds, h) particularly flare bevel, flare-V and other partial penetra-
~
tion groove welds. This DR disclosed generic deficiencies l for pipe support welding throughout the two units:
- 1. The design for flare bevel and flare-V groove welds [4'-4/
l typically used on tube steel, specifies the maximum instead of the minimum radius, which is the relevant indicator of cuality. l , l Site investigations revealed that the welds had up to 25% smaller l-radii than American Welding Society minimums. Similarly, the groove welds did not honor American Welding Society (AWS) standards for support steel work, instead relying on Pullman weld proc,edu g signed for piping. As a result, welds were made togangles which failed to meet minimum industry standards j and were not full penetration weldts as ca.Lled for by the design. Finally, the procedures for' fillet welds were n g with the AWS requirements for control of the weld'sgangle. The effect is that the design. failed to safeguard against slag in the throat j of the weld, which should have solid metal. l 2. b The drawings are inaccurate,because they did not reflect the design inadequacies for the welds. The effects include i inaccurate use of AWS symbols. For example, the partial welds
' 17 -
described above were represented with AWS full penetration symbols. Installation and QC inspection were similarly deficient.
- 3. Pullman's pipe weld procedures were not modified to reflect their use on support steel. The inaccurate procedures compromised QC inspections as well. The inspectors did not look
~
for the effects of using the wrong procedure, since they were inspecting to it.
'The welding controversy was due to, management's insistence that the American Welding Society requirements did not apply to Diablo Canyon. But management's explanation is self-defeating.
Management's alternative standard is the American Insti,tute of Steel Construction (AISC) handbook. Since the AISC section on welded joints references to the AWS code, however,.the distinc-j tion is. irrelevant. I have learned thac a transition is under-way on-site to follow the AWS code for any new work. Unfortunate-ly, the plant already has been built. This reform omits correc-tive action for all the existing deficiencies. My Deficiency Report is enclosed as Exhibit 1. The second DR that I authored concerned the angle members which are part of the piping support structures. Angle members throughout the plant were too long for the allowable bending stress standard used at the plant un AI c The code 1 requires such angles to range from g j -- to 993[ 5 j' 3 , depending on the angle, To research the scope of this discrepancy I walked through Unit 1 for an hour and r g p w g li the angles that exceeded the maximum code lengthg In that short time span, I found around 200 to 300 violations on the angle members within l approximately 100 frames out of approximately 300 frames checked. l The violations were obvious. Angle members were 2 to 3 times l l longer than the maximum permissible lengths. In one case, a member was around nine feet long, when the AISC maximum length was 25.33". d
There has been.some additional bracing, but it is erratic. To illustrate, one oversized member had been repaired, while another next to it remained untouched. A more practical repair would be to cut out and replace the angle frames with tube steel. The effect of this violation is that the supports cannot reliably carry the loads. Theoretically, the angles could buckle under the pressure. The angles support loads up to 1000 pounds or more, such as a bundle of small bore piping. The DR is enclosed as Exhibit 2. The third DR concerned
#ur (M Hith bolts, where the distance-between the center of the holes was not. verified as the same length required and specified on th'e drawing. The DR occurred as a result of a quick fix assignment where I could not determine the cause of the violations. I researched and learned that QC had skipped these inspections and instead measured the distance between the centers of the plates attached to the bolts. The problem is that the location of the bolts was supposed to be the control on the location of the plates. As a result, whole
- packages could be in the wrong location due to failure to measure the bolt holes themselves.
Further, I learned that the design group had made a mistake on the initial calculations for the center to center bolt spacing. Duetothelackofcontrolleddocygggtsg gp had not received neces-sary data on the diameter of the Hith and Phillips shells that precede the bolt when installed. Since the shells are larger than the bolts themselves, the shells determine the hole size. The hole size is controlling for compliance with the M-9 design guide for supports. As a result of the missing data, we had based
.our calculations on the bolt size instead of the accurate, required data from the shell size. -
The impact of this discrepancy skews the reliability of anchor bolt calculations. In close cases, supports judged acceptable should have failed. I cannot state with confidence how often this flaw would be decisive. Again, however, the problem is generic throughout Units 1 and 2, at least for-small bore piping. The DR is enclosed as Exhibit 3. Filing the DRs was the kiss of death for my career at Diablo Canyon. Everyone on-site outside of management who talked with me said that I would be gone shortly. Leo Mangoba said that I wouldbepggppdtoreviewtheresolutionofthediscrepancies, but withga few days a friend told me that he had seen my name on a list of five people in Mr. Leppke's office of those who would be laid off. In the end, I lasted nine days, which is about how long it would take for my DRs to be channeled through the Bechtel and utility chain-of-command and for my lay-off to be processed. On Friday, October 14 at exactly 11:30 AM, Mr. Mangoba announced to me that upper management told him to lay me off due due to decreasing manpower requirements. He assured me that the layoff was not performance-related; he said that if anything, my performance had been above and beyond their requirements. The layoff was effective the next work day, Monday, October 17. I never received written notice, and did not have my security passes removed. l There is no doubt in my mind that the layoff was pretextual. First, the nine-day time lag was suspicious. Second, I was sup-posed to stay ar.ound to review the disposition of my DR's. Even if manpower needs were diminished, someone else should have gone first until I .could follow through on the significant issues I had raised. It was obvious to me that the layoff was designed
/M
~
to prevent me from following up on the issues, just.as our DM trailer had been removed from pipe support reviews in March to get us out of the way. My layoff was an example for every-one else. Third, the idea of reduced manpower needs. empirically. was a sham. The next few weeks the staff was working overtime 3. on the same issuus for which manpower needs supposedly had declined.
'Jul unresolved mystery during my year at Diablo Canyon was the function of the Nuclear R,egulatory Commission (NRC). In my year at the site, I only saw the NRC on one occasion --
when I went to their trailer myself to pick up copies of govern-l ment regulations. On-site they.were invisible. They were a nonfactor. Perhaps that explains why such a massive program j of falsification and manipulation of engineering work has turned the corrective action for the 1981 Diablo Canyon scandal into an even worse scandal in 1983. l The NRC's default is in contrast to their performance at I other plants where I worked, where they conducted walkdowns and spot. checks of the hardware. At other plants the NRC was
, highly feared, despite the absence of even a resident inspector.
I cannot understand how the NRC is seriously considering l licensing Diablo Canyon, because the plant is not close to com-pliance with even its own design requirements. Perhaps it is because the NRC is ignorant of the problems. I hope my state-ment is educational. I will work cooperatively with the NRC to' resolve the violations discussed in this statement, but I will not work with any NRC employee who has been involved with Diablo Canyon. That track record would disqualify any such individual in my eyes. l l l
21 - I have filed a complaint with the Department of Labor to challenge my retaliatory layoff. 'This affidavit is to support my complaint. This paper has been simplified to facilitate review-by the Department of Labor, but I am prepared to provide further details and engineering support for all the issues I raised. I have read the above twenty-one page affidavit and it is true, accurate and complete to the best of my knowledge and belief. [d CIlKkLES STOKES' MTE N U'/I S l + h ' u - L
/ /M .
L ::. - Cou m w NC'/ > __1 - e _ y l i ON MC- / 'T .19M before me, the undemgned, a Notary Public in and for said State, personaDy appnred n- . 2 - T Una w/e & L !. r 3/ en M A t _ _ _ Mary $"
. known to me,
** to la the person __ whose name_ /f cibscribed to the within lastrument, and acknowledged to me that ha __ executed the same.
~~ u%.s a.m s
- taair== a or as. 3,es WITNESS myhand and officialseat I, '
W h s==t s. % , ,, - _ = _ w% tv-+~.- J L Ju e,
. .., . - . s,
. . m n mw . ~ . +.;varm . m.,, . . g myg,mm _,.,,.,,. - ey,_.-;
^~
. ?, , p ; . ., .
y u...g. :y. , .
*b
- Prscadura 10.1
, Attachment A Page 1 of 1 -
PACIFIC GAS AND ELECIRIC COMPANY ENGINEERING DEPARTMEhT DISCREPANCY REPORT Control Number (h ()- (31)(3)(3)-(3) PROJECI OR PLANT (S): Diablo Canyon Unit No. 1&2 ORGANIZATION AFFECIED: Pipe Support Design Engineering INDIVIDUAL RESPONSIBLE FOR RESOLUTION: C.V.Cranston Proj ect Eng. SUBJECT (ITEM / ACTIVITY):
REFERENCES:
, , , m e r ceedures DISCREPANCY: See attached sheets (5)
PROPOSED, ACTION: Make necessary changes in Design Guides and ESD 223 and Pullman Weld Proceedures to bring them up to AWS requirements for pre qualified partial & full penetration welds. ! SCHEDULED CO:GLETION: /j Initiated by: Date: /0 fh Approved by: Date: ACTIONS TAKEN: CLOSED Approved by: ' Date: Concur
- Date:
Chief, Engineering Quality Control l
(. . ;,. -- .
- a. n. m :v. em , v; : . u. u .- .
1 l . DIABID PROBIINS CDNCERNING hHD DESIGN, DRAWDG DOCGENIATION, INSTALIATION AND QC IN51'fLnON.- C.C. SIOKES,10/4/83-Rewritten frczn 7/5/83 paper.
- 1. FIARE BEVEL AND FIARE-V GROOVE hEDS DESIN Bechtel San Francisco Office-Per Dan Curtis by phone AWS table 2.3.1.4 applied to radius of tube steel obtained frczn table in a paper entitled "A Designers Guide to Welded Joints" written by: Mark Michaels of maximttn outside corner radii ' table 3.3'.
Note: The word max 2mtra radii. This is not good engineering practice. The conservative approach dictates that the minimum radil be uwl to I i ensure the safety of the weld joint. It should also be noted that Fark Michaels paper on the design of welded joints has, to my knowledge never been approved by the Engineering Depart 2nent and issued as a control doc-ument to engineering for use on Diablo Canyon. Bechtel Site-Per a handbook supplied by the tube manufactures institute, all tubu.g manufactured in the U.S.A. is made or rolled with a radius of 2t to 3t for all sizes. Having assuned 2t to be the miminun, . all calculations were made to AWS. table 2.3.1.4 using 2t = R. Per site investigations, Jeff Van Klapenburg, Ken Palmer and myself discovered scme tubing on site (Diablo Canyon) has a minimum radius of 135 t. Therefore, all welds of this type per this design group are not conservative. Westinghouse-Through review of drawings issued, their designs are also in question.
- 2. PARTIAL AND HIIL PEN ~rTRATION GROOVE HEIDS DESIGN Bechtel San Franciscx>-Per drawings supplied to field, very few if any are designed correctly. Symbols indicate emplete joint welds.
This is true for all joints requiring preparation. No angle for prep-aration has been indicated and it is not obvious that the designer is aware of the mimimtra joint requirements per AFS 2.3.12 and Fig. 2.10.1. However, on joints created by natural intersection of 2 hede.rs, it is obvious that the joint in many cases is a partial per AWS and rot a full penetration weld due to angle created by intersecting marbers being to small for a full penetration weld to be made.
-~ r -
.. :,n .-:. . .. , n
- m. : m. - . v.:. . .
~
Page 2 Bechtel Site-one group has tried to emply with AWS. requirements in designating both S and E per 2.1.3 and 2.10.3.1 table 2.10.3 and Fig. 2.10.1 and the included dihedral angle either on preparation or by natural creation of m mbers inta.s cting. However, Pullman QC have continually refused to check weld per call out because ESD does not provide th e with a procedure for performing verification. They have required the (E) effective throat call out to be reoved. Westinghouse-Sea cmments for Bechtel, San Francisco.
- 3. SKDED JOINT FTTM WEIDS DESIGT Bechtel San Francisco-Per drawings supplied to field and continued use of fillet all around call out instead of a specific call out adju ding the leg size for the dihedral angle adjustment. It's obvious that the dihedral angle has not been considered in the joint design or if it was, it was done incorrectly.
Bechtel Site-one group has tried to be consistent in adjusting the inner and outer fillet leg size hae.M on dihedral angle so that the effective throat on all sides is the same. This allows the joint to be analyzed as though it is an equal leg fillet all around with only a leg adjustment to obtain this when welded. This is shown en drawing, so construction only has to make what is shown and not interpret what is shown using a table which is not usually at hand. Westinhouse-See Bechtel San Francisco.
- 1. FIARE BEVEL AND FIARE-V GROOVE WEEDS DIGWING REPRESENI7CION Bechtel San Francisco-No partial welds nave been shown since S(E) have been cmitted. AWS 2.1.3 and 2.10.3.1 and 2.10.3 state that (S) groove weld depth and (E) effective throat shall be specified on shop or wcrrking drawings. The hanger drawings sent to the field are both shop and working drawings. Also, M-9 states that only pre-q=14fiM joints should be used on-Diablo Canyon. Many inmanagment contend that this job is not covered by AWS code. However, M-9 states that it is governed by AISC 7th, Ed.. In AISC section on welded joints, page 4-131, paragraph 4, AISC states that small deviations are possible per AWS code and other joint forms and welding pro-cedures nay be employed provided they are tested and qualified in accordance with AWS Dl.1-72. Therefore, Diablo is governed by AWS Dl.1-72.
.v. - . , ...- 7._,...;.,,,.
.. y .,, . ,, ,
,,y,. y .,. . ._.. .,
-i Page 3 CORRECT DW6. REPRESEMTATION s
L L a swyq S(E)l( FLARE - BENEL FLARE-V S= Radius of tube E= Effective Throat per table 2.3.1.4 AWS Bechtel Site-one group has otr: plied with call outs above. However, Pullman OC per ESD 223 have required the remval of S (E) frczn all drsrings. This was because the ESD does not provide a procedure for them to use to verify the welds above. 'Ibe symbols left, after remval of S (E) indicate full penetration welds even though, QC states this is not the case. Westinghouse-See Bechtel San Francisco.
- 2. PARTIAL AND FUIL PENETRATION GROOVE WEIDS DRAWING REPRESERTATION Bechtel San Francisoc>-No partial welds are shown since S(E) call outs have been crnitted along with preparation angle. Primary weld used is bevel.
Now indicated ' Per AWS this is a full penetration N weld. L Should show SCE) =<l\ - l With consideration given to AWS section 2.3.1.3 in specifing S(E) and Bechtel Site-One group on site has tried to ocruply with the above call out per AWS 2.3.1.3. Ibwever, they have meet aantinued resistance frcru QC in that the weld specs used for installation were written for piping per AWS chapter 10, Fig.10.13.1.1A is 37
- differs frun the prep angles for structural steel specified in Fig. 2.9.1 and 2.10.1, which usually indicate a minimum angle l of 45*.
2
-. - -- "-~
--. .r -c.-r ,; ; - r-vP Page 4 t
Westinghouse-See Bechtel San Frarciam Ocunents.
^
- 3. SIGHED JOINT FILIET WEIN DRAWING REPRESENTATION Bechtel San Francisco-Per Drawings are shown as fillet all around i
for all angles. Per AWS fillets are limited to a mininun dihedral angle of 60' and a maxinun external dihedral angle of 135'. All fillet made beyond these angles are considered partial penetration welds since AWS requires a reduction of effective throat of 1/8" when angle is less than 60* and greater than 45' and k" r*+Mn when less than 45' but greater than'. 30'. 30* is the mininun dihedral angle for structural steel except tube steel which has a mininum angle of 15*. Per ESD 223 the leg is adjusted for dihedral angle but no increase has been aMM to account for throat re-duction required by design resulting in adequate effective throats per AISC and AWS. Bechtel Site-one group has considered dihedral angle. To facilitate design, joint was sized assuming a constant effective throat size. After sizing effective throat requirements then using dihedral angle the 2 non 90* sides were adjusted so that the installed effective throats would be correct { per design. Fillets are not called out when dihedral" angles are less than 60*. ! Partial penetration welds are shown.
- 1) 2)3) INSTAIIATION AND QC INSPIrrION l
All insh11ation has been made per Pullman's weld procedures. Per a copy obtained frun Pullman QC of these procedures. 'Ihese procedures as written state that they are for the installation of pipe and pipe attachments. No mention is made of their use in installation of pipe supports. All prep angles and joint details are written for pipe, no notes or nodifications are indicated for their use in installing pipe support steel. OC has been supplied ! with these procedures and ESD 223 to inspect pipe support welds. They have been supplied no information as to the correct insh11ation of pre-qualified joints per AISC or AWS. Furtimmure, per ESD 223, they have been instructed to check sczne weld joints and not others. A joint formed by flare-bevel or flare-v welds on all sides such as 2 tubes crossing does not require checking and per ESD 223 when the size of a flare-bevel or flare-v is shown, the method of weld measurement per ESD 223 does not supply the effective throat or any dimension which can be used to determine the effective throat. l l we tw - -,*r g r a.g-w w^re-wwg---+w7-a--g- -' sfear w *--'-vsaweTv-3G7--v '
=wv'W W-'"V'7-*av *'*D'N'-"?WT"- " ' - * - ' " " ' - " ' " ^ ^ ~ - " ^ ^ " - - " ' --
., . .y . . n_,.
Page 5 Per attachnent I of ESD 223, no limitation are indicated for structural or tube steel dihedral angles. AWS gives the minimum dihedral angle 30' for structural and 15' for tube steel. Also ro throat increase is included to ocznpensate for required throat reductions ha W on dihedral angle per AWS 2.3.1.3. Per attachment J of ESD 223, of what use is the measurement of S ? Per Ah3 unless S=R. R-Radius of tube, E can not be deteImined through use of table 2.3.1.4. Also upon review of this attachnent and table of maximum radii of tube steel in paper by Mark Michaels, it can be shown that table J was and is based upon the maximtzn radius and not the minimtzn resulting in a non-conservative design. I I i I l l t i e -
......,...,.....,,.v,.
. . ..........wm., ,
. ~ ..-~..
, Proc 3 dura 10.1 Attachment A Page 1 of 1 -
PACIFIC GAS AND ELECIRIC COMPAhT , ENGINEERING DEPARTMEhT DISCREPANCY REPORT Control Number 0 0 - C9 E> O _e3 PROJECT OR PLANT (S): Diablo Canyon Unit 1 & 2 ORGANIZATION AFFECTED: Pipe Support Design Engineering INDIVIDUAL RESPONSIBLE FOR RESOLUTION: G.V.Cranston Project Engineer SUBJECT (ITEM / ACTIVITY): Unbraced Length of Angles A * " 'C " * * **
REFERENCES:
See attached underlined copy of AISC. DISCREPANCY: Hany angle members used to build pipe supports exceed the maximum length for which the allowable bending stress may be taken as .6(Fy) PROPOSED, ACTION: ese members should be mod m ed by boxhg or brachg at critical compression points to comply with AISC or the allowable bending stress reduced and Interaction Equations reverified. l l l SCHEDULED COMPLETION: , Initiated by: _/ /> I Date: /d f 8 i Approved by: Date: l ACTIONS TAKEN: l CLOSED Approved by: Date: Concur
- Date:
Chief. Engineering Quality Control
, ,, . . . . . . .. .- ~ " ~ ~ ~ ~ ' ~
1 . * ' , t
+s, ,
Saructural Sue' fo- Bur.daags . 5 17 use.i in the gas metal.are process shall con. 1.5.1.0.2 On the gross section of arkUyloaded compress'en members
' 7d b:s.' E:cesrodes ict Ces Mescl A rr We'isne, v. hen L*l r exceeds C,:
die prosons of Sect.1.17.3; ECOT or E70T , r . .. . . L, , 12r$EU-
- i.arr crocess shall conform to the Specisco. .
i:r Flus Corrd Arc Weldmg, AWS As.20, F. . (1.5-2) et Gect.1.17.3. 1.5.1.3.3 2 Sl')'. On the gross section of axiaDy loaded bracing and secondary n shaJ cons:itute . . . -suf5cient
- evidence of con . ~~~** ". ,'
members, when l/r exceeds 120*.9.. ',= >rt.P.a*P' ' 7_ , F. (by Formula (1.5-1) or (1.5 2)) LE STRESSES
- i ggg 1.6 - ' - -
- 4. 1.6. 1.7, 1.10, 1.11 and in Part 2. all com. 200r .
o' e so proportioned that the stress, in kips . 1.5.1.3.4 On the gross area of plate girder stiEeners: . 'nd ths following values, except as they are F. = 0.60F, .
.n -
2.5.1.3.5 On the web of rolled shs' pes at the toe of the Ellet n: rippling, see Sect.1.10.10): et pin holes: F. = 0.75F, Es = 0.60F, d 1.3.1.4 Bending ., T gre - 1.5.1.4.1 b "- o sninirnum tensile strength of the steel. Tensic.n and compression on extreme fibers of compact hot-alain syebers,. pin-connected plates or built- rolled or built-up members lexcept hybrid girders and members of A514 steeli symmetrical about, and loaded in, the plane of their minor axis and meeting the requirements of this section: ets see Table 1.5.2.1. F. - 0.66F, In order to qualify under this section a member must meet the following require:aents: T. = 0.40F, .
- s. The flanges shall be continuously connected to the web or webs.
d cad fabricated shapes may be taken as b. The width-thickness ratio of.unstiEened projecting elements of the 5 cnd the thickness of the web. See Sect. c mpression flange, as defined in Sect. _1.9.1.1, shall not exceed in webs. For discussion of hiirh shear stress 52.2 /VF,. , n:ctions of members whose webs lie in a c. The width thickness ratio of stirened elementsof the compression Sect. 2.5.1.2.i flange, as denned in Sect.1.S.2.1, shall not exceed 190 'VE,. The depth-thickness ratio of the web or webs shall not exceed the ~D value tion cf axis!iy loaded cornpression members cnderness ratio of any ur.cracc:i segment as C,: S M' d.tN*- - 412 1 - 2.33 VF, (1.5-4)
) _ 'IlOctri'
- except that it need not be less than 257 'VE,.
e
" p' , , . ~.#- , , ". 1.T-1 Tne compression fiange shall be supported 1sterauy at intervals 20.000 U. 5 .:t to exceed ~C.CL T F, nor(d A.a.r,_
2-:I -- - m nyer:.: gir: ers ar.d members of A514 stee:. beams and ;irders
.\ E,
- a. . ..
-mbera cesier.ed on the basis of composite action which meet th* requnements of suo. paragraphs a, b, c, d and e above and are con-uncal values fer vanaut trades af stui enm. *
- hr tia sse E is taken as unity.
- . u. -. :,
p j gyms,u-=, . . .- 7T- -m= _ stan
' 3_yg ,=..- - -
M*
"' 1 Hh M Bb;azrm&4 16 4
, g ---
g .- n .ni n - x-1 w -- "
- eh , W
- if g
. 5
! ., ' ' {. ,.
.L + '" h*
=
w .. W W . Mi >,
~
s. G.M w S-C---
.m 7
g,**- '.
..:..Ji.:a. = M ik M F M =
r: . ' , - .6;- Qr:r.S:rwswulSand /sr Basiass . 5 19 f, '
. - . .~. . ?.
x ?.4.y 1rrp.: - <;:.9,. , .e..: I
.zs -
f propertioned for tio of the negative When ifr !5 .2. - *: . Uf- ^ - .- Yg whleh are maximum at po.tnta of r 3 % .10 .
. F, x: _,: 10'C. W-..c /a3i..+
.:. *r-.-m 1. -
... . s. ; ng- .-33..m ;.w ; t- . ."- ^ -
hers, the maximum positive moment re n:gative moments. This reduction -- - -= ;- - I*._."_170 x 10'C.s ^ ~ 1 D. * ' sy 12cdmg en cantilevers. If the negs.
.. " ' ' ' ~~M~ .
s
. . L.pp q . W.~' .
ii . . - e
' . a. .: 3, i ...
(1.5-6b) gidly framid to the beam or girder, the 4. ;,
- Or, when the compression flange'Is'soHd aid approxidiately re'etangular in rtioning the column for tlie combined cross-section and its~area is not ^less ~ 'than
" ' that " ~ of the tension aanse that the strass f., due to any con- - - -- W- -
a not sxceed 0.15F.. F, . 12 X 10 T. . . . . _ . 1 (1.5-7)
', ., . n.Id/A . .g :, -:. ,. . , . . . - ';
~ -
rid gird:rs and members of A514 steel) , ,;, .-. 7
.1.5.1.4.1, except that 6 /2t., f exceeds In the foregoing. - - '- . msy be designed cn the basis of an ~
f - distance between cross. sections braced against twist or lateral
, displacement of the compression flange ).0014 VF' (1.5-51 '
rr - radius of gyration of a section comprising the compression flange
\2t/ f , plus one. third of the compression web area, taken about an axis in the plane of the web salon en extreme fibers of doubly- A, - area of the compression flange -
s meeting the requirements of Sect. C. - 1.75 + 1.05 (3fi/3I,1 + 0.3.f3fi/3f )', but not more than 2.3*, i 3 bent cbout their minor axes fexcept where Af, is the smaller and Af the larger bending moment at the md equars bars; and solid rectangular ends of the unbraced length, taken about the strong axis of the member, and where Afi/3fi, the ratio of end moments,is positive 0.75T, when AI and AI have the same sign ereverse curvature bending) and negative when they are of opposite signs (single curvature sai:n en extreme fibers of box. type bending). When the bending moment at any point within an Bang 2 er web width. thickness ratio unbraced length is larger than that at both ends of this length, the
.cte 1.5.1.4.1 but does conform to the value of C shall be taken as unity. C. shall also be taken as compnssi:n flange is braced laterally unity in c mputing the value of F., and F., to be used in Formula nIs the transverse c,: stance out.to-out (1.6-la). ' See Sect.1.10 for further Imutation an plate girder flange stress.
100F, _ For hybrid plate girders, F, for Formulas Q.,E:Q,a) and U-$$ is the 3ield stress of the compression flange. Formula (1.5-7) shall not apply to .
.ers of flexural members not covered in hybrid girders.
5.1.4.4: 1.5.1.4.6 b Compression on extreme fibers of flexural members in-cluded under Sect.1.5.1.4.5. but not included in Sect; 1.5.1.4.6a: O rim 2 fibers of flexural members in-S axis of symmetry in, and loaded in. $ F. - 0.60F, m on extreme Abers of channels
- bent ce ecmpmed by Formulas il.~i-Gai or provided that sections bent about their major axis are braced laterally in
- css a hicher value can he nati5ed the region , 4 of compression o stress at int,ervals not exceeding 76.05..VF,. -
- hm not more tn:n 0.60F,.' / k..,.,t 1 4.
Bearme2S' D . b3 b , 3 3 ,1 ion contact areal 14 L g 5o.47" ImC. 1.5.1.5.1 .TIllied surfaces including bearin;; stifieners and ;: ins in
;-, turne.i. c.riaea. or oored holes: -tr': .
(1.5-6 E, - 0.90F," '
;0 x 10*C.,lgF,
. C. can be conservatively taken as unity. For smaller values see Appendix A.
Fig. A1. p. .%104.
- :7,eh. " When pam in contact have diferent yield stresses. E, shall be the smaller value.
1.51.4.6. lan twn parngraphs. ~
.. .?--:~ ; :-
. s ;,..
.- . . .-ty,. .,
dM ^
~ . a w,. .
/.
M
; f;- _ - g -
. _ r- -
g $. ,. - fgrmuh
- s- ~
?^ Wal k] ,1 ,
n r
- ~~ Ewa $,.
, _ ~
g . K;JM,n. f *
- v. ~
&.jf44f : .
FM,, .- . p. . -m _ _ - __-u _ 7.i j 6-
. 'a% .
W rN 4 1 .
.,7 '; ;.._. ' ' - - ~ - - -
- - -e N
My T^ x?
.c 1
o I ___ -
#---I
___n___ j _ _ _._ n - ---.m__ __"_ _. hap _t-
.a m,___
A
,_xx,, ,
s - 1 K W4N 4h
.c . _
,;$n % a Og q. :-
- - - . . - . - - - M.-b -.h , , ye .- 'e5-,...
s .118.4 . AISC Specu%uon : ' ' . .L W. ' :.-.;-'* . i:-25= ~' * = + h " ' s t-- - - , - - - -
- 1 1.5.1.4 Bending l- SECTION1.11 COMPOSITE CON 1 1.5.1.4.1 Delete subparagra;5h d m its ennrety and substitute the IOU 1.11.2 Design Ass.amptionas ;. -
**E % ~
! 1.11.2.2 At the beginning of the
" d. The depth. thickness ratio of the web or webs shall not exceed the i "For construction without tempor3ry value given by Formulas (1.5 4a) or :1.5 4b: as applicable. ; modulus of the transformed composit4
-d - - 412 _ 1 - 2.33 f,
']-; (referred to the bottom dange of the ste stitute the fonowingt when /' -- '
t vy, F, F, ( 0.16 (1.5-4a)
,l . .x . .
"For construction without tempor.
i stress may be computed fro:n the total d 257 e I
-- - formed section modulus S.,, axcept tha when y' > 0.16 (1.5-4 b)"
Y I' - shah not exceed that of Formula (1.11.: appropriate value ofSect.1.5.1." 1.5.1.4.2 Immediately foUowing the words "of Sect.1.5.1.4.1" add a comma. and immediately foDowmg the words "except that bf /2tf " delete tDe Comma SECTION 1.15 CONNECTIONS Change formula number ":1.5 5)" to "il.5 5a)". 1.15.5 Restrained Members
- In the fi.r-t line of the second par strained".
1.5.1.4.3 Add a second paragraph as foUows:
"Doub'i.sym=etrical I. and H-shat = rnernbers bent about their minor axis eercept hybrid girders and c: embers of A514 steel meeting the require-rr.ents cf Sect. 1.5.1.4.1. subparagraph a. except where b f/Offexceeds SEC'TIONI.23 FABRICATION 52.2 vF, but is less than 95.0 'VF,. may be designed on the basis of an 1.23.1 Straightening Material al~owab:e bending stress Delete this subhead and the entire t.
the words "RoUed materia!". and substit 0.933 - 0.0'53
~
F, - F, y *i- '1.5 5b'" reading as fobows:
,, \ 2: 7 (
"1.23.1 Cambering. Curving. and S.
1.5.1. 4.6a Immediately foUowing the words "under Sect.1.5.1.4.5." The local appUcation of heat or me add: "and meeting the reau:rements of Se:t.1.9.1.2.". troduce or correct camber, curvature, a of heated areas. as measured by approve for A514 steel nor 1200'F for other steels. 1.5.1.1.6 b Immediately foDowing the words "under Sect.1.5.1.4.5." add: "and meeting the req uirements of Sect.1.9.1.2.". 1.23.6 Welded Construction
""? -
In Table 1.23.6. for thickness "To {.
; the beading "We! ding Process", change "
^"
- SECT 10N I.10 PLATE GIRDERS AND ROLLED BEAMS 2.10.5 Sticeners l
l 1.10.5.3 In the third pararraph. :mmediately foDowing the werds I
- holes sc.3 t e such that", delete: "the :=a..er panel di=ension a or h.
i shall net exceed 345t V/," and substtu:e tne words "f, does net exceed the va'ue given by Fcrmula (1.10-1;" c-
.f '
7 ....L.. .J.I, ..i. ,:. ~ .
. . .....n~-=---
......._ '- a ;...
~...,g,.,-von , . , . =
. e. . .
. .. . . a T. - 2ih - 1.6/, ( 20.0 1.8.2 Sidesway Prevented - ' J ". ~~ ~ ' ~ ~
- T. - 53.0 - 1.6/, ( 27.0 ' .. .. . .- .
In frames where lateral stability is provided by adequate attachment to
.c :n ess arca T, - 29.0 - 1.6f, ( 20.0 diaron.nl bracing, shear walls, an adjacent suucture having adequate lateral .,,~..:.,., ~ ~
stabmty, or to floor slabs or roof decks secured horizontally by walls or T, - 50.0 - 1.6/, ( 40.0 bracing systems parallel to the plane of the frame, andin trusses, the effective
. type Jo:r.u r , = 4 0.0 - 1.6/, < M.0 . lengthgfactor, K, for the compression 3 by the same forces, shall not eaceed the gg mem. bers shad be,,taken .
a.s un.i.t.y...
- - ~
< p rg. -{ . ;-Q y.-1.;. , ,f;- - -
- icir.u. ti.e shear stress allowed m. Sect. 1.8.3 Sidesway Not Prevented '-- -- F M' .
In frames where lat'eral stability is dependent upon the bending stiffaw. . T ( 15.0:1 - fA./T.) of rigidly connected bnma and coluinns, the efreetive length Kl of compres. . T, ' 20.01 - /A./T.) sion members, shall be determined by a rational method and shall not be less than the actual unbraced length. a due to a direct io.id applied to all of the - - s;niaed pretervion load of the bolt. ,1.8.4 Maximum Ratios The slenderness ratio, Kl.'r, of compression members shall not exceed . 'D CONNECTIONS SUBJECT TO 200. ll1ATION OF STRESS (FATIGUE) preferablyThe slenderness shouM not exceed: ratio, Kl;r, of tension members, other than rods, *f cation. 2s cefined as the damage that may For main members . .
. ......... . .. 240 number of fluctuations of stress. Stress For braem.g and other secondary members . . . . 300 . of these fluctuations. In the case of a be computed as the numerical sum of
- pressive stresses or the sum of manmum SECTION 1.9 WIDTH. THICKNESS RATIOS retion at a given point, resulting from 1.9.1 Unstiffened Elements Under Compression 1.9.1.1 Unstiffened (pwjectings compression elements are those hav.
in conventional buildings need to be de-c:angrs in such structures occur only a ng one free edge parallel to the direction of compression stress. The width
.s oniy minor stress fiuctuations._The of unstiffened plates shall be taken from the free edge to the first row of earthquaxe loads is too infrequent to fasteners or welds; the width oflegs of angles, channel and zee f!anges. and .
si:n. However. erane runways and sup. stems of tees shall be taken as the full nominal dimension: the width of ni equipment are cften subject to fatigue flanges of I. shape members and tees shall be taken as one-half the full nominal width. The thic.tness of a sloping flange shall be measured half. way between a free edge and the corresponding face of the web. g 1.9.1.2 Unstiffened elements subject to axial compression or g pression due to bendirm shall be comMeH m My eFmire when the ratio
.s. subje : to is:i ue loading as defined on Wtn to tmexness as not greater than the following:
ed to satisf" the streu range limitations a b Single. angle struts: y double. angle struts with separators. 76.0 N , D SI.INDERNES.: RATIO 5 ts e m risine doulp angles in comam anpu er niatex pro ectmg num ctraers. cosumns or uther c":nnression memtiers: co:npression fian:es of beanu: sti.% .cr on
- nd :: tne 2:ru::u a s a .. ;.. and for Pi ste Firders . .
95.0 s F. Stems of tees . .. . . . . . 127 V F, s1.5. ratio .3.3. of taw an axiali[v ienr h sizail ioaded ecmpression be taken as its When the actual width-to-thickness ratio exceeds these values. the npo..:: . . . . . ........s
, . . c f .. .rauu.. cesign stress snall be governed by the provisions of Aptwndix C.
.h .S ,
. ..- ws :.~ .
n - ._. _ ,= .. . -- --
" - - ' ~ -
(* y g _l_ -- - -[ M M C
-a.
'Ef
'8Bmed,2mg&_ ,a
~
y ! q y
- . . .p. 'MY N h t-j
'M r s c h- '
. Nf
-w F.
$..5A.WT-3$$'$Q M+1-m%:y. (;:.4h
~
D..'
~
4 * ~ fr- .
,b .;
~:
F- * .-mW't'M mi.MI&. .' y~? l y.A..e . 7. w .. . _ . _ . - - - D h., .
- r. t
}.{ _qh _
,e
1 i h t#b.G- , s.di w-$' 3 ,n 5 4-r m.r /* .* yww w WD n& 5
%y=2 A Ej.
m U.*. s g
.a e ;. %-
Y =^
,e w e m M'.
Mtw @,, ee E
@TEQhr f,s , "W. %r M.f > .- 1 g . g $l' %
** ,,h .?1 ps Q'y y. % ~ "
- vth:* ,p 4 Weh C. .
h.m. .
.~ ..
(%. A - q c
,p -. &%p 2:
w l ,
, 7.y.$ fe. n - <
M *WM, . g wwme. r per - %m
,g e
_, 4m. _ .__ A_ _a
/
5e t me.:n u: A sc 5pecin:. :wn . . ---- . . .
- - ~.- - +-----,:~.-. . . . - ,
I . ~, . L t L ha !vnt embm arr torsionJy s em i.m:.
.d.' Tne critica] , Formula T1.5M is a convenient ap
- ,a'
.rm n* a mer.o-torswna; oucilm;. for tn . ccmprewee dange ence of octb iatera! bendmg renstance
' :- i-un. wued m tr.e t/.ane of its n am a as to t.end about Due to the difference between nange n _
Jr. m am ur. i. < ht.un -c us.ng Fermula I: 1 mtn an equiva:ent girder, it la desirable to base the lateral s t rro ra to um e.. re&ics torsion of the flange. Hence. use of For: members Its ag eement with more [l\ r , ,,. . , , 41%
} d[,
_ strength ofintermittently braced Eezw neous sections having substantial resista u bre ! nt in the case of doubly-symmetncal secti s dinance between pomts of lateral support and S,. I, and J For some sections having a comp: m re m n e tn ma or aus w:tien modulus mmor axi, mo:nent of than the tension Bange area. Formula 0
- re ua an: ine .< r> :nal connant of the beam cruw-se:ucn. It can be its use is limited to sections whose cc snt .n tnat u ren d 105 and I b < 2.500 F,. the showable compressicn sarge st- great as the tension Bange. In plate inxiic.ned ny the abov e e ;uanen will appretimate 0.60F,. higher d. Af ratio than rolled W shwes.
bewn'i t:a 1;m:t dedectwn rather than stress is likely to be the design the conservative side. For such mem For=ula fl.5-6a) and. at times by Forr 1.i.1.1.', cu._ I .5.1. 3. 6 Tne aihb!" ban i:n: streu for all other mate of buck!mg strength. While toes G ar.d n.en.oe, r. rn en as O M,. pree um u mpessicn Bange is strength somewhat because they ignere er <' - at cutiu.v ese intenals :! 3. the prodle, this ngidity for such sestion Mem: r. nent aoout tae:r mapr au and nas mg~6.0%F). an ax:s of symnetry f vercon.wrv tism. therefore, is likew-m.., .
- n. :va _...., ma s be ad eq uately tiracu., :a t era h.) at grtater inter-It should be noted that Formu;a (1 expresaons it replaces, is written for th
. . t .r
..a.... . .. b(.n u:.s, s t ren. is reauted su:nc.cr tly to prevent pre-m ure oui dr; of tne compressmn fiange ~Ma6matirai exnre cinns tign is n t provided for t, n.is :crmula i_
* ; -:.n-r i i m o i ~ nue of the i- ' a r e -. r:t n < <: .
~
= , - w nen actual conditions of loac, app'.: cati
_ u-- .o s u . _;r,m .-
- t. are considered, any unconservante err r . .: i .w: , t r + m- r. .r n : x: w .
t e emen f:me . Singly-symmetrica!. bu.lt-up. ; 1-sr.
- r
. m :.n -
w ar mo~ -
- n -,m . - g:rders, often have an incrcased como:
~e r ur urn re. .nc. .utt- e. e . ocnt u mn r,encmg due to lateral loadir.g action in 7*
.recara.n: t r at .n - c f ,ta: eral su ppe r Such members usually can be troportii
~. -
.:r: ' nm < r r. .- ..an erg 2n - r:rg
*" msms:ym p 7' loac." :ng. " "M. nere . ,,
trae Ia:.ure moae c: a
._I- ** ** ...q 7 havin; a la eer comaren:cr inan tener
'nic '. n h . .
- re - nrenar nr tne re missa.e ver.cin; stres- car N s a h- o' ' 1 .5- 7, on c: cwn. Fermulas 1 - -
.. Through the introduction of the r ar
-e! on t he asum: .:cn inat c r.!y tne r. en mr4 l
i
- o n ta n;" w ;., pr ven: the !.ceral a.isplnerent of stress is parmissiole when there is mome
- -' - . . er , 7 >,'e+t - eracing u ..
ine reu _ r ormu.u 1.a..ca . and except where, in the~ case of combine.a
. -
- e t w e iroer Form ui.4 4 m t w o u ays: adjustment is provided by the facter C n nea u :.e earner prm iens requ:re: no wress reducu.nn wnen . Formulas (1.5-6a; and 1.5 6b n
. u- -
t r.a n 4 0 r e c a rd w.s3 of y:i. i s:ress value and then a 5,enant and warping ters;on by sum. w rm m tne s aiue cMa:ned :. om tne rara!n,c expression, the equivalent raaus a of gyration. r....,. can priate expresstien giving twe critical e.s. rw 'em :ax by increasmg F, at 1 - O irem 0 60F, to ?F, 3, . s. ro. dange of a beam
- with that of an axiall s a centmucus s:re ss relanornh:p v i:h ine unbraced length e
-n ! < re.iuced from the maumum pcrm:sChle value of 0.60F,. ..
- W at rt tne earber smg:e Formu!2 4 at:::ed esen in the range (f+ '~
t udhng streu .cn the a< mm;:tmn that Fo mula (51 , . _ v 1. rn t he rer'.xemen t of T rm.;;a (4 ' is liberah:ed in -
- ne aautwn of an Eug - 3 & egrenion, since i,n:s .
- Cduren Research CourS Gu de to '
i - i . u y 5 c .-- M err.bm Mc-i Ed.ne- E; f 5
' h.L. E: . :3
. i,_-
u
. r .- - ...c . Comp rewn
. h.i.Ep 4W r.
1" v .
.= e -w .
w
. . , . , . - . ,-.r-.. -...-~-- .c < *r<--- 4 . . .
- e. y ~~~ ~ ~:'~~~'. ~
< _-q StrucsuralSuntforBube
- E 127 * .
. . a ". 2 .: '
,..:.. . v.r . ni - ' Tne crideal Fr.rmula (1.5-7) is a convenient approximation which assumes the pres-
- .:. ter in , . ..orenian fiange enee of both lateral bending res: stance and St. Venant torsional resistance.
., ;nct r.x., o o to be:ni about Due to the difference between aange and web yield strength of a hybrid .nt. 1.51 w:tn an equivalent girder,it is desirable to base the lateral bucklmg resistance solely on warping _- . torsion of the aange' Hence, use of Formula (1.5-7) is not permitted for such. ,
- members.~ Its agreement with more exact expressions foe the buckling
.U5, strength of intermittently braced flexural members
- is closest for homoge .' "
~g~* .r- . neous sections having substantialresistana to St.Venant torsion, identifiable e nir-r - in the case of doubly-symmetrical sections by a relatively low d/Af ratio.
~
inter i supt > ort and S,. I, and J For some sections having a compression aange area distinctly mmaller a m.'es. r.:n axis moment of than the tension Bange area, Formula (1.5-7) may be unconservative; hence, oc.m crou.section. It can be its use is limited to sections whose compression fiange area is at least as 9 F.. tne siiowable compression great as the tension fange. In plate girders, which usually have a much * *
- atiesn will a;micr.: mate 0.60F,. higher d/A fratio than rolled W shapes Formula (L5-7) may err grossly on
,tre,s is lii.ely is be the cesign the conservative side. For such members the larger stress permitted by Formula (1.5-6a) and, at times by Formula (1.5-6b), affords the better esti-
- t. e cending nren for sll other-mate f buckling strength. While these latter formulas underestimate this cided the compression Sange is , strength somewhat because they ignore the St. Venant torsional rigidity of the profile, this rigidity for such sections is relatively = mall and the margin
- s I b. . u. 0 y r:- .
of overconservatism, therefore, is likewise small. and havine an axis of symmetry It should be noted that Formula (1.5 7),like the more precise, complex uraced laterally nt greater inter-expressions it replace I, is written for the case of e.lastic buckling. A trans!. uced sufhciently to prevent pre-tion is not provided for this formula in the inelastic stress range because, ni:e. .%1athemnLical expressions when actual conditions ofload application and variation in bending moment stren;th of stien members which are consideren, any unconservative error without it must be small. tcout their ion;:sturiinal axis iSt. Singly-symmetrical, built-up, I. shape members, such as some crane
- nen o! Incar compression fange girders, often have an increased compression Sange area in order to resist
- torsion . are too complex for bending due to lateral loading action in conjunction with the verticalloads.
neir accuracy is dependent upon Such members usually can be proportioned for the full permissible bending
-stnt at points of !steral eupport stress when that stress is produced by the combined vertical and horizontal an be na more than engineering loading. Where the failure mode of a singly-symmetrical I shape member
.. having a larger compression than tension flange would be by lateral buckling,
. r 1.5.o... . nn d ,. .o. .-. provices the permissible bending stress can be obtained by using Formula (1'5-6a) .
"";*"'. rm. -
or (1.5-7). . a o:. tne dpec:ncation. Formulas Through the introduction of the modiSer" C., some liberalization in
-u ptmn tnat .7nil Ihe hending stress is permissible when there is moment gradicat over the unbraced length
.vm the later.it u:wl.1 cement of , except where,in the case of combined bending and axial compression, this ne new corrnus::.4 1.a_-oa i and adjustment is provided by the factor C., in Formula (1.6-la).
i" t* " *3 FS! ~ Formulas (1.5-6a) and (1.5-6b) may be refined 6 include both St. quircd no stren retiuctign when , Venant and warping torsion by substituting a derived value forr r . This
- e; . stren value anc tnen a
, equivalent radius of gyration r,,,,,, can be obtained by equating the appro-
.:r. the Inrn:->.tc ewrenton. the priate expression giving the critical elastic bending stress for the compression 1
- lange of a beamt with that of an axially loaded column.;
*fr.* sam
- 6w!:n :rsm tne 0.60F u .rw. w 2F* k..
tenet num .v minib ' .. u- i 9.&.'T . a: hr en " in-. ran::e
~
Q :. ' . r *". w. f- *
^
* . . .~~ *- ~
- C:.iu=n Research Council Guide to Design Cntena for Sletal Compress:en
.- .. . .. . = :..cr u:- .
m yy gq_ 43 ,
" Ibid.. Eq. 4.13.
- n Cntena for 11e:al Compression t Ibid.. Eqs.14.9es, (4.30), 44.31) or (4.32).
.. Ibid., Eq. 2.2).
t . r ~~ _ ;7
-Y E w = rum m h-G
- v" _-- -=
=-+ A m-
= C f M N_ g &_ d 4 qum b, ase
_ -n m
~
J W em f ' 4
~
r , _ - . _.- n ---
. %. ,ic. pap.-'
"igh "InfBol9Fa -s@m%EP;DM Y-M . E 1 t [mz syc_ __ _
l
, - 5 EW ...:___ ~
M. . ~ 1 e __ _., -- _._= - wn- '
- . - N .c
- - - c --
MMED
._ IL -_-
.'" q #ike Mre-m w-:rr
- g r,i 7"w.c f'TYAFW p q ;L't;iyW
, u: 6he . h M M - d%1#
s '
~'
_pnn gw -- 4
- , 3;f -
-
- 5 r V Yti *
%rg: w A
< m ee w>. 7,
- v. . .: ~
n , a w s c . .. -t-4 #4 .s - 6
- y. g
..a cQty y - -
G, _[ er r-
,6 -
yg' M - wm 1 3 s . D.
.. .m. .*s a,
",* , 3,a 1,c
.s
. +
.n p..-
-t .+q3 .
M. -
," e . ," c ~
# . 2 =4 At -
i a . s < - na i
.u .
La4 t
- 4vs ;.
w [ g- N.W ~ ke
- .sapes w e**h*- "h v.v - w t**FV';f* - ., M [ . .,3%ygg;_ __M,
}y g g s 1:' Cm:m on Atsc sparkn:.m ~ . . ?... l __
~'
? m :ne case of a cc,uMy-s cn.n c:nci Jona;e beam, 1.5.2.1 Shear r ,-~r
..n.. .
-.n _ o-' ^ " Connections w hich transtnit load -
- h. 1, nre categorized as "inction. type" or
- vr -
. , a t r.- nu n a m me tre.' r.,2nen:a c: t. > r . n:
. ., .. ti:un sd.::er th nig. can ang fcrce tc
.3 a a mawt
.h C U~n: n.J:.u.and rr.e i .tter e nt n _4 upon cen:a:t o.f eie bO:e. to tran.3.e: the load :.rcin one cor - .
~
<.t ,
s.. . The amount of clamn.nr force des J=~.i -
- 4" '
3 3
. vent cc=p!ete 5:ippare at the Nnnist
, connecions and co.mactions made vit 1.3.1.5 hearing '
beann g.t y pe. Tr.e *J;h clampin; force-l 1.5.1.5.1 c ength bd:s is cu5cient to prevent 5 As usei througnout ti.e S;-ri:icatien Ine terms mihed sur- ' equal number of these bolts are substit.
- .ne. ' t.!Le d" o r ' - hng" are 2ntended to inc.ade ruriares which have would be required to transmit a given ;
r~n accurate:y owed or f.nnned to a true p;ane t9 any smable meara nvets and AWD idts for A502 Grade '. l ~r.onmn H r-anrc r re% on rm.s u nu in- un:e .ts fr- nvea. Tne The cicien:v cf threaaed fanent ow er uue. run -ter.:Ls cif he . .. s: rew cf ' ne ra ant.urang tre pin l connecuens is relaced w r.en the inrt
> . o ras ::c.s a saferuard a;a:nc mnab as of tr e r .a: tm c nd the nole,'
l.v: ween the ccnnec:ed pr rts. In the c;
. . c. coraitraM .arrer JJ n a m et rm.e. 2:e d4a 3Jm '.h are pren ore shear ;2ane and er.e where it u not 1.5.2 Encts. Holts, and Threaded Parts b h** E' tr.e m d .WJ M .
l tend in:o the shtar ; lane and the aho l 1.5.2.1 Tension Fross area, ts reduced accord:rgi.s. As m earher ed.uora. t-nns.ib:e crew fr nvcu ar pren in terrns
.Na t o t oe r.o:nma. cro u mt wna: area of tne - - wicre cirn in,_ < 1.5.2.2 Bea rin g 1 or : mater cons en:cm* m the ; r+ - !wr.; cf n:;r. t en;- h bc ted con. Ikaring values are provi-ied. nc.: a :
r- m t r- . teste m fcr ne , s are ps t r.m tern ap: :: cab:e ta the:r , m no e mm, mu3
! " . T. : (" 1s area. I t i art .t Of ': e ur.' r U:s s -, }{pw eg cr [or .q,. ;;
,, g d , 3?n ?C2nL9 % ..
I a tt. u ra. are a . . , mc y7 -
- er. and tnica ied
.G - e u. . . . . a. ,
...c
.s t-Te n.
.n .s rc : r <
.n-- - _ sar-m me - 7 ,. g . ,
Lst %, - . ' ' ,
~
r.t ern.e 1.n e
.n,,,,. n, ,.,7.,;
- > > a r. " . -
..t .
- w. . mu!:i; bd : 3 m n 1 9 ge. ;,.y g ., c,
, i - . 6i . n r.rea ; m : t r ._ . - r.een found to r--
heuun-daT 4 t' .es
- ne M t, s 'w ine tenue wren :h of 1.rger d.2:ne - :nreaced ; arts- of e.e part. In :ni: inveve:nen tre '
.u y m:en: - uv.! fer anchor t-dts er upw: roas accera:r: to the u=J convennon. as :
2n re e;- en of ti+ rrotecaon aca:n.=: no:ch e 5 :n t he thread.n;- eter and t'.:c k. ness cf t r.e con ect ed F - asur i by tne recu:rel uuua! t:;raenmg cf ri.ph s:rerrn no :5, the Re-w tween unt:ed. ear leann; and enclov
. C< und on Ra etei and ICted S: ue:ura: an:s nas rt-remmemied the recommended werbng nreu is tr.
i
; re. u n e. h;cher w erk.:n; st res in : ens:en for h:;r. < -< r.;th Edts. g. g.7, g ga.m .gy eq.
l Ans addinonal fastener tens en resu!t:n; i a n pnr action due to dis- stress recc=menied fur determimns re teima cf the connmion deuils shou:.! le a<ided to tr a cress calculated I i c.recdy i- n, the ep;;mi terson in ;mpc.-nonir; fas are for an apphed In. 'I m e. tain; ine w.fied u rming s: e-s Lea n dirg upon the 1.5.3 Welds l l r
.A n e n #m u cf the f.u:enm an ! t he c c - --- a:. th;s rning - - - -
an m3v !* n A s in t ne J .m., t r.e a'now a b'e w or m-t r: (.f :h.c :4;a] terJ:On lf b,.e O r 1: m a y l + 3 fu t' l
*! i 2003. -
. , , q g; ,
- ancrer-l
~
- g. e no uN m r m c f *u
-n - x.,: , - ""'^""' " * " " ' ='
n
"'" ' k i k ,.
b'.. d 4 - A .% ] l * ' J : ; .I r
{.. . .
. . . , . . . . . . - . . .. . ~ v ..,, a , , ..y _
- Procsdure 10.1 Attachment A Page 1 of 1 PACIFIC CAS AND ELECTRIC COMPANY ENGINEERING DEPARTMENT DISCREPANCY REPORT Control Number (3)(3) (3)(3)(3)-(3) 50 JECT OR PLANT (S): Diablo Canyon Unit No. 1 & 2 ORGANIZATION AFTECTED: Pipe Support Design Engineering I
INDIVIDUAL RESPONSIBLE FOR RESOLUTION: G.V.Cranston Project Engineer Center to center distance of Anchor Bolts SUBJECT (ITEM / ACTIVITY): Hilti Catalog, and Phillips Catalog H-9 & ESD 223
REFERENCES:
DISCREPANCYNanufactures soecify the minimtzn center to center distance to be equal to 10D where I> the hole diadutr. Design has used the bolt size on shells, ; not the hole reg'd for the shell for D. This was caused due to missing information as tn bnla =% vm'el fnr challe ferrn e m+-alev, PROPOSED, ACTION: Rechech anchor bolt calc. for shells, Reducing allowables per M-9 and rechecking interaction equations for these cases where 10D (shell hole size) exceeds that used on Dwg. SCHEDULED COMPLETION: , Initiated by: [ _4_> Date: /0[ (% Approved by: Date: ACTIONS TAKEN: CLOSED Approved by: Date: Concur
- Date:
Chief. Engineering Quality Control
- .__ _____ _ _ . _ - _ _ - _ . - - _ . . _ _ , _ _ . _ _ _ , _ _ _ _ . _ , _-_ _ _ -._- _ . _ _ _}}