ML20080A611
| ML20080A611 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 01/28/1984 |
| From: | Doyle J Citizens Association for Sound Energy |
| To: | |
| Shared Package | |
| ML20080A604 | List: |
| References | |
| NUDOCS 8402060158 | |
| Download: ML20080A611 (15) | |
Text
9 -T l
UNITED STATES OF AMERICA - I NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of APPLICATION OF TEXAS UTILITIES I I
Docket Nos. 50-445 GENERATING COMPANY, ET AL. FOR and 50-446 AN OPERATING LICENSE FOR I COMANCHE PEAK STEAM ELECTRIC I STATION UNITS #1 AND #2 I (CPSES)
AFFIDAVIT OF JACK D0YLE 1 Q: Mr. Doyle, do you have any comments regarding Applicants' 1/17/84 2 Motion for Reconsideration of Memorandum and Order (Quality Assurance for 3 Design)?
4 A: Yes, I do.
5 Specifically, on page 12, last paragraph, continued on page 13:
6 Not only is this paragraph inaccurate, but a Senior Vice President of Gibbs 7 & Hill (the Architect / Engineer at Comanche Peak), Mr. Scheppele, went to 8 great lenths to prove that what went on outside my little box was none of 9 my concern. See my Deposition / Testimony, CASE Exhibit 669, pages 257-260, 10 for example.
11 On page 20, regarding Mr. Kerlin: I had indicated that Mr. Kerlin 12 held a responsible position, not' necessarily a supervisory position. One i 1; point is certain: I was not of equal rank with Mr. Kerlin; it is my under-14 standing that I was in fact two grades lower in rank, as was Mr. Walsh (even r
15 though Mr. Walsh was the group leader).
16 On page 20, re: "Had Mr. Doyle reported this alleged deficiency l
8402060158 840201 PDR ADOCK 05000445 O PDR t
1 to his supervisor . . ." etc.: This is not true. For example, I complained 2 of cinching up of U-bolts and lack of inclusion of the structure of the support 3 in seismic calculations. (See CASE Exhibit 669, Doyle Deposition / Testimony, 4 pages 29 and 30; see also pagn 95, lines 15-24 and page 105, lines 1-17.)
5 Every time such problems were brought to the attention of Mr. Krishnan, our 6 supervisor at Comanche Peak, he either ignored the problems or told us to 7' forget it. In reference to instability, it was a useless attempt in any event, 8 since by his own testimony, Mr. Krishnan admitted that he had no idea what 9 instability is. (Tr. 3940/1 through 3941/25, 3942/7 through 3943/23, 4908/15 10 through 4909/25,4913/13-25,4939/22-25,4945/4-12,4954/1-22,4967/14-25, 11 4972/8-25 through 49/5/13.),
12 On page 21, 2nd paragraph: Applicants' argument is without merit.
13 During testimony by Applicants' witnesses, the instability problem was ad-14 mitted to exist and again by Applicants' own admission corrective action was 15 not instituted until sometime much later. (See. CASE Proposed Findings, 16 Section. III . ) Beyond this, the problem at Comanche Peak was more severe
- 17 due to rotational problems of the clamp about the pipe. The problem (which l
l 18 was corrected) at the Fast Flux Test Facility (FFTF) involved clamps and 19 struts (double pin ended), not box frames and U-bolts with a clearance to 20 the pipe.
l l 21 Page 21, last sentence, regarding "potentially unstable supports":
l l 22 I disagree. The testimony was in reference to unstable supports and in this
( 23 particular case, like in pregnancy, the degree is irrelevant.
24 Page 21, footnote 22: The fact that an. unstable support (rotational 25 capability) exists in effect means that there is no support at this point.
l
r ,.
- F , t. s s T
. J Q/^- ' '
O ('N F m N'. , ;
, s.
1
-And this, fact means that a significant deficiency exists, as is attested s N N2 , e to. by. the contents of NRC IE Bulletin 79-14, which was written to (among 3 o'ther reos'ons) address supports which were missing. In short, the instability 4 addres e'dsin tnese hearings is in reference to supports which could carry 5 no load in one principal direction. See CASE Exhibit 669, Doyle Deposition /
6 Testimony [p' age;104, lines 1-8.
. 's C 7s .
Pdge 23, last paragraph, continued on page 24:
This paragraph N 8 is inaccurate. See for example, CASE Exhibit 669, Doyle Deposition / Testimony, i, .
N 9 T_ , pages 257,-260 (in reference to the bcx in which Applicants like to place s ,J . ~.
10 ' employees) . See also the reference in the Circuit Breaker (given out onsite 11 at Comanche Peak) to " disgruntled and misguided former employees"; and the
~ ~
12 reference of counsql for the Applicants to me as a non-expert to whom the 13 Board should no' t ' listen in preference to Applicants' and NRC Staff's experts T
14 (in the May 1983 hearings regarding the upper lateral restraint). The intent 15 of, Applicanth' counsel and experts has already been demonstrated in the l !' \
16 record and thes Board has correctly interpretted that intent.
. , - - s s
-s -se .
7 17 Page 26, last paragraph, continued on page 27, regarding the argu '
r '
w s N 18 ~m ent in reference to 1084 supports and the 30 or 40 field tours. This argu-
! \
i h9 . ment, without listing what numbers of supports were done by each member of f ' 2h' A the group, is deceptive. (Also, Applicants have supplied no documentation l k 21 . (relating t.a what'pe{ cent of the total supports at Comanche Peak this repre-h Y22s. sents.) 'The production'of Mark Walsh and me was always at the top of the
.i
, i
' [ '23 group (this is by raw numbers only). In fact, I made a special effort to do 24 the more complet (and time-consuming) problems. Mark Wa'.sh and I obviously
\f 25 - were doing alconsiderable amount af, work and therefore had less time to wander l V l
,s _ .
- n. : :
- . . - . -~ - - - . - . . - . . .
G
.3 3 f
1 - about than is insinuated in Applicants' pleading. In fact, the 30 or 40 2 field tours referenced usually involved far less than one hour each. If P
- .3 more engineers were" encouraged to go out into the field and look at what 4 is actually in place, more problens would probably be caught.
5 It should also be noted that Mark Walsh and I saw many other problems 6 during those field tours which the Board should probably also consider; 7 I did not bring them up previously because I didn't have specifics regard-8 ing them. If the Board is interested, we could relate further problems de-I 9 tected in the course of our employment.
10 Page 28, first paragraph, in reference to U-bolt material: Appl i-I
- 11 cants appear to be implying that we are dealing with two different animals 12 here, but we are not. SA-307 is merely the designation for A36 type steel 13 used as bolting material. See ASTM Standard Specification for Structural 14 Steel, A36-77, Table 1 and item 3, Apurtenant Materials; or Section 2, Part 15 - A of ASME, which does not include Item '3 or Table 1 but only recognizes ,
16 LSA307 as low carbon bolting material.
! 17 Beyond this, the mechanical and chemical properties of SA307 bolts and 18 SA36 rods which are threaded are identical and Applicants acknowledge that 19 -" SA-36 and SA-307 materials have equivalent material properties" (page 28, 20 first paragraph of Applicants' pleading).
21 Page 28, second paragraph: Friction occurs whenever cinching is L22 utilized to resist shear to prevent rotation of the U-bolt. Applicants 23 are depending on cinching and thus friction between the pipe and the U-bolt 24 to prevent rotation. Applicants are therefore not in compliance with ASME 25- XVII-2462. See also ASME NF-3324.6 a.4,1983 Edition.
J 1
- . , + - . , , . , , , , . . _ , , - + . , , , , . , . , ,m__-..~,,%, , _ - , _ , , , -,.,,,,ce.. , _ . _ , , .....,_..-w,, . - - - - - - . ,,-..,,.e__,,-...w,,r..,---
r ~
. a
. r 1 Page.29: But this cinching up is not the practice intended by the 2 manufacturer of the U-bolts, and this can be confirmed by reference to the 3 -gap allowed for pipe expansion by the manufacturer of the U-bolt. The inside 4 dimensions of the U-bolt are larger than the pipe outside diameter to allow 5 for expansion.
5 Page 29, last paragraph: The word is not insufficient; the key 7 word is that the force is unpredictable and therefore could be greater than 8 . or. less than anticipated.
9 Page 31, in reference to (stiff) ' quoted portion: But these are 10 not conventional pipe clamps; these are cinched-up U-bolts which introduce 11 a problem of point loading and pre-torquing versus the common area loading 12 only that exists with a pipe clamp. (See CASE Exhibit 6698, Attachment 13 to Doyle Deposition / Testimony, items llYY,11ZZ,11AAA, and 11BBB.)
14 Page 31, middle of page, regarding pre-tensioning: Applicants 15 have made a factual error: we are not concerned only with the piping, but 16 with the U-bolt and the piping combination, as has been indicated in our 17 testimony and in CASE's Proposed Findings. Beyond this, the pretorquing 18 stresses induced by whatever type of clamp (sitff or otherwise) is equatable 19 with the stresses in the pipe,. as was pointed out in Board Notification 20 82-105A.
21 Page 32, first paragraph: Applicants are incorrect in stating 22 that calculations do not exist for thermal expansion between U-bolts and 23 pipes. See CASE Exhibit 763, Surrebuttal Testimony of Jack Doyle, page 13.
24 Page 32, last paragraph: Applicants may now consider local messes 25 for integral attachments, but they do not consider them for non-integral i
l . .
1 attachments', as was indicated by inference in Applicants' pleading, page 2 31 (stiff clamps vs. U-bolts).
3 Page 34, indented portion: This is in reference to procedure, but 4 does not consider the limitations on the particular procedure; for example, 5 the minimum weld which is required to insure that cracking does not exist 6 or Beta angles which insure that the weld assumed actually exists, etc.
7 These prohibitions go beyond the basic procedure. For example, in the case 8 of minimtzn weld requirements, there are no tests which will preclude this 9 requirement, because it relates to cracking of the weld relative to the stiff-10 ness of the joint and may or may not be present in any given series of tests.
11 To avoid!this requirement, a new procedure would be required controlling the 12 welding itself; for example, preheating and control of the cooling rate of 13 the total affected structure to eliminate the thermally induced strains which 14 result in the craking of the weld. These pmcedures do not exist at Comanche 15- Pesk. The fillet weld applications are approached as if they were, in fact, 16 prequalified.
17 Page 34, continued on page 35, footnote 36: The Board should note 18 that if the drag angle requirement and work angle requirement is unknwon to 19 Applicants, then it is obvious that this consideration has been violated.
20 See Welding Handbook, 7th Edition, Volume 2, Copyrighted 1978, pages 69-72, 21 Electrode Orientation (Attachment A hereto), which states, in part:
22 " Electrode orientation, with respect to the work and the weld groove, is important to the quality. of a weld. Improper orientation can result 23 in slag entrapment, porosity, and undercutting."
24 Further, on page 71, Figure 2.14 of that document, the orientation for drag 25 angle, push angle, and work angle ate diagramatically depicted. On page
1 72, Table 2.3, the angular limitations for work angle, etc., are given.
2 If the designer is not made aware of welding limitations, then his design 3 will reflect .this shortcoming and result in potentially deficient welds.
4 Page 34, last paragraph, and page 35, first full paragraph: If 5 the Beta angle was incorporated May 11,1982 (before intended plant comple-6 tion anticipated during May 1982, while Mark Walsh and I were still at Comanche 7 Peak), what procedure was used to perform the welds required during the period 8 1974-1982? (Applicants have again provided no documents.) And if, as the 9 Applicants state, the program established for qualifying welds precluded 10 consideration of each of the AWS criteria, why did this Beta criteria have 11 to be added in mid-1982? Also, Applicants have presented no documentation l
l 12 that ITT Grinnell or NPSI follows tne same Beta guidelines.
13 It appears that Applicants have proved our point.
f 14 Page 36, last paragraph, continued on page 37: The fact that I
15 something is done all the time is irrelevant, and is in fact a cause for 16 concern, because it could just mean that they have been doing it wrong all 17 the time. The purpose of minimum welding is to prevent weld cracking inter '
18 nally during cooling. Covering potential cracks with more weld material lg will not eliminate the internal cracks generated from improper welding pre-20 viously. It may look good on the surface, but you can't tell what's 'under-21 neath. To the best of my knowledge, in no code, including AWS, is the pro-22 cedure for violation of minimum welds correction spelled out, since the 23 Codes do not anticipate violation of standards as being a generic practice.
24 See AWS Commentary, on Structural Welding Code -- Steel, 3rd Edition, Section 25 2, Design of Welded Connections, 2.7.1. Minimum Fillet Weld Sizes for Prequalified
1 Joints. NOTE: The minimum weld size is required to prevent cracking, not 2 to act as a safety factor for loads which were not included in the analysis.
3 Page 37, last paragraph: The generic stiffness study which accom-4 panied the NRC Staff affidavits on open items showed clearly that the loads 5 on support points went up significantly when generic stiffness v31ues were 6 replaced with actual stiffness valuc:. And when all the stiffnesses in 7 the system (which is not the case in this study) are included, the actual 8 stiffness value will be far less than those assumed in the generic stiffness 9 study submitted by' the NRC Staff.
10 The generic stiffness study does not indicate that with a 200% increase 11 in load a problem does not exist throughout the plant. It only indicates 12 that on that particular system, which had relatively light loading, no problem 13 existed, probably because the stresses in the supports due to the light loads 14 were only a few thousand psi and doubling or tripling would have no signifi-15 cant effect on the stress ratio. However, on other systems, as little as 16 a 20% or 30% increase in loading could have a significant effect on stress 17 ratios.
18 Page 37, last paragraph: As an example, using 7th Edition AISC 19 steel tubes, the following will clarify the point. Assuming two conditions 20 with a horizontal run of pipe, 20" off a wall, and supported by cantilever-21 type supports, two supports would exhibit the following: For one case, 22 assume 800 lb. vertical load downward on a 3x3xl/4" tube with a moment of 23 inertia of 3.16"4 and a section modulus of 2.10"3 The stress would be 7,620 24 psi. Assuming an allowable of 24,000 psi, the stress ratio would be .3175.
25 The deflection would be .024 inches. This would be within all of the allowables
1 'for this type of support at Comanche Peak.
2 On the other hand, for a 20" cantilever 4x4x3/8" tube with a moment
.3 of inertia of 10.2"4. and a section modulus of 5.1"3 and a 4,000~1b. vertical 4- load, the stress level would be 15,686 psi. With the same 24,000 psi allow-5 able, the stress ratio would be .6536. The deflection for these conditions 6 would be .037 inches. Again, this would be within the allowables at Comanche 7 Peak.
8 If the loads were to be increased by 60%, it can be seen from the above 9 that the support with the 800 '1b. load would still be within the allowables.
10 But if we assume that since this is the case for the first instance that it 11 also must be true for all cases, it is obvious that we would be wrong, since 12 in the case of the 4,000 lb. load, the stress ratio would exceed 1 and the 13 deflectic. level would be marginal relative to a 1/16" allowable. Beyond 14 this, as the generic stiffness study showed, the variation can be greater than 15 60% increase and in fact can be well over 100%(200% of the original load and 16 in some cases, far higher per cent, as was pointed out in the generic stiff-17 ness study in which one case indicated loads increased over 600%). Also, 18 there are many supports with stress ratios of more than .654. I know because 19 I worked on a number of them myself.
20 Page 39, last paragraph, continued on page 40: The problem goes 21 beyond these tests (which were made with single bolts), since each bolt or 22 two bolts of a group must be capable of carrying the full load or a substantial 23 portion of the load assigned the group, and not the group load divided by 24 the number of bolts. The interaction of the loads vs. a bolt system is depen-25 dent on the location of each of the bolts in oversize holes vs. the bearing
e 1 area which is available to transmit the loads.
2 3 Q: Mr. Doyle, do you have any connent regarding Applicants' proposal 4 that additional hearings'be held on the closed Walsh/Doyle items?
5 A: Yes, I do. It seems incongruous that after 18 months, we now find 6 ourselves facing an Applicant whe was all too willing to decide the issue 7 on the evidence available in September of 1982 now wishing to go on to a 8 new presentation under _new ground rules. And this would appear to be an 9 unprecedented procedure in any legal forum. Without resorting to profanity, 10 I can think of no words to express my feeling on the Applicants' new desire 11 to pursue this matter ad infinitum until such point as Applicants can con- -
12 vince themselves or the Board that Applicants are correct.
13 14 15 16 17 18 19 20 21 22 23 24 25 l
Attachment A The Five Volumes ofthe We.C Ing HanC 300 <
WeldingHandbook, SeventhEdition '
Seventh Edition, Volume 2 1 FundamentalsofWelding .
2 Welding Processes- WeldingProcesses-arc acas wans ac uinz.
Brazing, andSoldering Arc and Gas Welding and Cutting, 3 Welding Processes-Resistance andSolidState Welding und OtherJoining Processes
. 4 Engineering Applications-Materials t
5 Engineering Applications-Design i
! WH. Kearns, Editor AMERICAN WELDING SOCIETY 2501 Northwest 7th Street Miami, Florida 33125
. =
s
..A,.
63 / smnDED METAL ARC %ELDING 3- Attachment A weldingProcedures / 69 h sdjusted so that the are slightly leads the mn! ten creases the size of the heat-affected zona and re-pkper control of the molten weld pool. For ver- terity. Although the are length does change to tidal and oserhead wc! ding, the optimum amper- some extent with changing conditions, certain weld pool. Up to a point, increasing the travel duces the cooling rate of the weld. Forward trwel ages uould be likely to be on the low end of the fundamental principles can be given as a guide ta speed will narrow the weld bend and increase speed is recessarily reduced with a weave bend a!!owable range. the proper are length for a given set of conditions. penetration. Beyond this point, higher travel as opposed to the higher travel speed that can ba ,
For downhand wc! ding, particularly with I
Amperage beyond the recommended range speeds can decrease penetration; cause the sur- attained with a stnnger bead. Higher travel speed face of the' bead to deteriorate and produce reduces the size of the heated-affected zone and {
should not be used. It can overheat the electrode heavy electrode coverings,the tip of the electrode and cause excessive spatter, are blow, undercut, can be dragged lightly 'ilong the joint. 'Ihe are undercutting at the edges of the weld; make increases the cooling rate of the weld. The in-and w eld metal cracking. Figurcs 2.13(B) and(C) length, in this case, is automatically determined slag removal difficult; and entrap gas (perosity) crease in the, cooling rate can increase the strength I
sh'ow the effect of amperage on bead shape. by the coating thickness and the melting rate of la the weld metal. The effect of high travel and hardness of a weld in a hardenable steel, un-the c!cctrode. Moreover, the are length is uni- speed en bend appearance is shown in Fig, less preheat of a level sufficient to prevent hard-form. For vertical or overhead wc! ding, the are 2.13(G). With low travel speed, the weld bead eningis used.
ARC LENGTH l length is gaged by the welder. The proper are will be wide and convex with shallow pene-The are length is the distance from the mol- length, in such cases, is the one that permits the tration, as illustrated in Fig. 2.13(F). The shal- ELECTRODE ORIENTATION ten tip of the ciectrode core wire to the surface welder to control the size and motion of the low penetration is caused by the are dwelling on l of the molten wcld pool. Proper are length is im* molten weld pool. the molten weld pool instead of Icading it and Electrode orientation, with respect to the i portant in obtaining a sound welded joint. h!ctal For fillet welds, the are is crowded into the concentrating on the base metal. This, in tum, work and the weld groove, is important to the transfer from the tip of the electrode to the weld joint for highest deposition rate and best pene- affects dilution. When dilution must be kept low quality of a weld. Improper orientation can result pool is not a smooth, uniform action. Instantan- tration. The same is true of the root passes in (as in cladding), the travel speed, too, must be in slag entrapment, porosity, and undercutting.
cous are voltage varies as droplets of molten gmove welds in pipe. Lept low. Proper orientation' depends on the type and size (
metal are transferred across the arc, even with When are blow is encountered, the arc length Travel speed also influences heat input, and of electrode, the position of wc! ding, and the constant arc length. Ilowever, any variation sn . should be shortened as much as.possible. The this affects the metallurgical structures of the geometry of the joint. A skilled wc! der automat-voltage will be minimal when welding is done various classifications of electrodes have widely weld metal and tne heat-affected zone. Low trav el ically takes these into account when he deter-with the proper amperage and are length. The different operating characteristics, including are speed increases heat input and this, in tum, in- mines the orientation to be used for a specific latter requires constant and consistent electrode length. It is important, therefore, for the welder A B C D E F G feed. to be familiar with the operating characteristics p ;D .
i ' d' -
The correct are length varies according to of the types of electrodes he uses in order to ree- ... ,,
the electrode classification, diameter, and cover- ognize the proper arc length and to know the f [, I ,
, .h .
g,y;;,;j ing composition; it also varies with amperage effect of different arc lengths. The effect of a i $ m,a y A 1 and w c! ding position. Arc length increases with long and a short are on bend appearance with a t ,
'l increasing electrode diameter and amperage. As mild steel electrode is illustrated in Figs. 2.13(D) [ $
a general rule, the are length should not exceed and (E). .
i
(' ]
the diameter of the core wire of the electrode.
The are usually is shorter than this for electrodes TRAVEL SPEED .
p d with thick coverings, such as iron powder or
" drag" electrodes. Travel speed is the rate at which the elec-A Too short an are will be erratic and may trode moves along the joint. The proper travel short circuit during metal transfer.Too long an speed is the one which produces a weld bead of i ^
I are will lack direction and intensity, which will proper contour and appearance,as shown in Fig. ] .
tend to scatter the molten metal as it moves 2.13(A). Travel speed is influenced by several : k q from the c!cctrode to the weld. The spatter may factors. Some of these are ,
g (1) Type of welding current, amperage, , i be heavy and the deposition efficiency low, , ,
Also, the gas and flux generated by the covering and polarity t Y.: .q g are not as effective in shielding the are and the (2) Position ef welding 1 ,p , .
weld metal from air. The poor shic! ding can (3) Melting rate of the electrode ; Y ,c jf dm cause porosity and contamination of the weld (4) Thickness of material .
A pf.' ,f ,p 8
, . ;,p, -i.t met *il by ongen or nitrogen, or both.The qual- (5) Surface condition of the base metal _
ity of the wcld will be poor. (6) Type ofjoint
, fig. 2 n-W. cifect of welding amperage, aic length, and travelspeed on covered electicele weld Control of arc length is large!y a matter of (7) Joint fit-up beadu fAlproperam;erage, arclength, andtravelspeed; f8) amperage tco law; fC) ampengc ao wetkr shill, s.nvolving the welder's knowledge, (8) Electrode manipu'at. ton ! l.:gh: fo) m length too sha:t; i:) crc Isn5th too long; 41) travel speed too s!ca; fG) travel reced experience, visual perception, and manual dex. When v c! ding, the travel speed shoulj be t o /ist
_ Attachment A operairng ruccaures't 71 70 / simmo umt.ute wnntso *
. Push ang'e for
- jcint. ~! ravel angle and work angle are used to The purpose of the tapping and scratching motion forehand wefding specify electrode orientation. is to prevent this. When the electrode does stick, Drag angte Travel angle is tl.e angle which the electrode it needs to be quickly broken free. Otherwise, it '
for backhand
.will overheat, and attempts to remove it from the **IdI"8 makes with a reference line that both lies in a plane through the axis of the weld and is perpen- workpiece will only bend the hot electrode. Free-dicular to that axis. Work angte is the angle which ing it then will require a hammer and chiscl.
e
['
The technique of restriking the arc once it /
the clectrode makes with a surface of the base metal in a planc perpendicular to the axis of the has been broken varies somewhat with the type I weld. When the electrode is pointed in the direc- of electrode. Generally, the covering at the tip of tion of welding, theforehand technique is being the electrode becomes conductive when it is g g(( Eg/ f wa,g used. The travel angle, then, is known as the he:ited during welding.This assists in restriking __ b(,[*// / "
angte push angle. The backhand technique involves the arc if it is restruck before the electrode cools.
'/ "
A
- pointini; the electrode in the direction opposite Arc striking and restriking are much easier for N 4 that of welding.The travel angle, then,is called electrodes with large amounts of metal powders N. g N s the drag angle. These angles are shown in Fig. In their coverings Such coverings are conductive .
when cold. When using heavily covered etee-
$c\69' ogre 5'*9 2.14 x Typical c!cetrode rsrientation and welding trodes which do not I. ave conductive coatings, "
technique for groove and .illet welds, with carbon such as E6020, low hydrogen, and stain! css stect l steel cicetrodes, are listed in Tabic 2.3. These c!cctrodes,it may be necessary to break off the Axis of weld
( may be dilferent for other m.iterials. Correct projecting covering to expose the core witc at the (Al Groove weld l orientation provides good control of the molten tip for easy restriking.
! u cid pool, the desired penetration, and complete Striking the are with low hydrogen elec-fusion with the steel base. trodes requires a special technique to avoid poro-Work angle A large tavel angle may cause a,conycx, sity in the weld at the point where the are is staned.
poorly shaped bead with inadequate penetration, This technique consists of striking the are a few Push angle for whereas a small travel angle may cause slag en- electrode diameters ahead of the place where forehand welding ,F trapment. A large work angle can cause under- welding is to begin.The arc is then quickly moved '
/ e' ,
cutting, while a small work angic can result in back, and welding is begun in the normal manner. / ~~~ 7 Drag angle for lack of fusion. Welding continues over the area where the arc originally was struck, re-fusing any small glob- g sp: / [Y, ,,- backhand welding 4/4, . 'M /,$,e ules of weld metal that may have remained from
~M W WELDING TECHNIQUE strik.ing the arc. "
During ov elding, the welder maintains a g sopoo eSS#
The first step in SMAW is to assemble the Axis of weldf. $e preper equipment, materials, and tools for the normal are length by uniformly moving the cice.
job. Next, the type of welding current and the trode toward the work as the cIcctrode me!!s. At polarity, if de, need to be determined and the this same time, the electrode is rioved uniformly * - 181 Fillet wefd power source set accordingly.The power source along the joint in the direeden of welding, to form must a!so be set to give the proper volt. ampere the bead. . Fig.2.H-Orientation of the electrode cha acteristic (open circuit voltage) for the size Any of a variety of techniques may be em-and type of electrode to be used. After this, the ployed to break the arc. One of these is to rapidly the crater and slowly brought forward to con- increases the prcbability of trapping slag and i work is positioned for weldmg and, if necessary, shorten the arc, then quickly move the electrode tinue the weld, in this manner, the crater is filled, thus, producing a defective weld. Complete and I clamped in place, sideways out of the crater.This techmque is used and porosity and entrapped slag are asoided. efficient slag removal requires that each The are is struck by tapping the end of the when replacing a spent efectrode, in which case This technique is particularly important for low properly contoured and that it blend smoo:hly electrode on the work near the point where weld- weldmg wdl continue from the crater. Another hydrogen c!cctrodes. Into the adjaccot bead or base metal.
Small beads cool mo e rapidly than large ing is to begin, then quickly withdrawing it a techmque is to stop the forward motion of the ones.This tends to make slag removal from small '
small amount to produce an are of proper length, electrode and allow the crater to fill, then grad-SLAG REMOVAL Another technique for striking the are is to use a ually wnhdraw the electrode to break the arc. The extent to which slag is removed from beads casier. Concave or flat beads that wash scratching motion similar to that used in striking When cortinuing a weld from a crater, the are each wcld 1c.'d b:fe c weld'ng :ver the bead smocthly ir.to the base metal or any adjoWng ,
a match. When the electrode touches the work, should be struck at the forward enj of the cr:ter. his .* firect tear *.r.f, t, . ,h: ceti y of a muhipi: beads ui..:.nize .n.Mec,tiag an:! anid a r..mg.1 there is a tendency for them to stick togeth:.. It sl.ould then quickly be mc.c4 to tha tid cf ,
.,.3 g,v3, ..:::r[.!; bar.cMhi.or.d ao.f. . r.4 Q. c *p't.f Me bea 1..'ar; %
72 / stuiint tuttTAt. Ake ut.t. DING Af.tachment A operatinx NcrJuin / 73 could stick. Finally, it is most important that The important factors influencing are sta- the power source, and the wclder, all contribute remain within the steel plates. For this reason, wclders be able to recognize areas where slag bility are to arc stability. the flux around the electrode. when the electrode cmrapment is likely to occur. Skilled uciders (1) The open-circuit voltage of the power is near either end of the joint, is concentrated ,
i understand that complete removal of slag is source between the electrode and the end of the plate.
(2) Transient voltage recovery character- . ARC BLOW .Ihis high concentration of flux on one side of neecssary before continuing a weld.
istics of the power source Arc blow, when it occurs, is encountered the arc, at the start or the finish of the weld,
( } Si f he mohen dmps of SHu metal principally with de wc! ding of magnetic mate- deflects the arc away from the ends of the plates.
WELDING GROUND and slag in the are rials (iron and nickcl). It may be encountered with Forward blow cxists for a short time at the Proper grounding of the workpiece is a (4) lonization of the are path from the elec- ac, under some conditions, but those cases are start of a we'd, then it diminishes.This is becat.se necessary consideration in shielded metal are trode to the work . tare and the intensity of the blow is always much the flux soon finds an easy path through the c. eld w ciding.The location of the ground is especially (5) Manipulation of the electrode less severe. Direct current, flowing through the metal. Once the magnetic flux behind the are is imponant with de welding. Improper location The first two factors are related to the design electrode and the base metal, sets up magnetic concentrated in the plate and the weld, the are is may promote are blow, making it difficult to and operating characteristics of the power source. fields around the c!cctrode which tend to deflect influenced mainly by the flux in front of it as control the arc. Moreover, the method of attach- The next two nre functions of the welding elec- the are from 8s intended path.The are may be this flux crosses the root opening. At this point, ing the ground is important. A poorly attached trode. The last one represents the skill of the deflected to the side at times, but usually it is back blow may be encountered. Back blow can ground will not provide consistent electrical con- welder. deflected either forward or backward along the occur right up to the end of the joint. As the w eld tact, and the ecnnection will heat up.This can The arc of a covered electrode is a transient joint. Back blow is encountered when welding approaches the end, the flux ahead of the are be-lead to an interruption of the circuit and a break- arc, even when the welder maintains a fairly con- toward the ground near the end of a joint or into comes more crowded, increasing the back blow l ing of the arc. A copper contact shoe secured stant are length.The wc! ding machine must be a comer. Forward blow is encountered when Back blow can become extremely s: vere right at witn a C. clamp is best. If copper pickup by this ab!c to respond rapidly when the are tends to go welding away from the ground at the start of the the very end of thejoint.
attachment to the base metal is detrimental, out, or it is short circuited by large droplets of jo nt, as shown in Fig. 2.15. The welding current passing through the the copper shoe should be attached to a plate that metal bridging the are gap. In that case, a surge Arc blow may become so severe at times work creates a magnetic field around it.The field is cempatible with the wmk. The plate,in turn, of current is needed to clear a short circuit. With that a satisfactory weld cannot be made. Incom. is perpendicular to the path of current between is then secured to the werk. l'or rotating work. ac, it is important that the voltage Icad the cur- p!cte fusion and execssive wcld spatter result.To the are and the ground clamp. The flut field contact should be made by shoes sliding on the rent in going through zero. If the two were in those who are welding with iron powder elec. around the arc is perpendicular to the one in the unrk or through roller bearings on the spindle phase, the are would be very unstable.This phase trodes or to those with other electrodes that work.This concentrates the magnetic flux on the on which the work is mounted. If sliding shoes shift must be designed into the welding machine. produce a large amount of slag, foward blow is ground side of the are and tends to push the are '
are used. at least two shoes should be employed. Some electrode covering ingredients tend to especially troublesome. It permits the molten away.The two flux fields mentioned above are, if loss of contact occurred with only a single stabilize the are.These are necessary ingredients slag, which normally is confined to the edge of in reality, one field. That fic!d is perpendicular shoe. the are would be extinguished. for an electrode to operate well on ac. A few of the crater, to run forward under the are. to the path of the current through the cable, the these ingredients are titanium dioxide, feldspar. The bending of the are under these condi. work, the arc, and the electrode.
an van US Potassiun compound @ctuding tions is caused by the effects of an unbalanced Unless the are blow is unusually sesere, ARC STABILITY ,
i the binder, potassium silicate). The ,nclus, ion of magnetic field. When there is a greater concen. certain corrective steps may be taken to climinate A stable are is required if high quality welds one or more of these are stabilizing compounds tration of magnetic flux on one side of the are it or, at Icast, to reduce its severity. All or only are to be produced. Such derects as inconsistent in the covering provides a large number of readily than on the other, the are always bends away from some of the following steps may be riceessary; fusion, entrapped slag, blowholes, and pcrosity ionized particles and thereby contributes to the greater concentration.The source of the mag. (1) Place ground connections as far as pos-ionization of the are stream.Thus, the electrode, netic flux is indicated by the electrical rule which sible from the joints to be welded, can be the result of an unstable arc.
states that a conductor carrying an electric (2)If back blow is the problem, place the Table 2.3-Typical shielded metal arc electrode orientation and welding technique for carbon current produces a magnetic flux in circles ground connection at the st:ut of wc! ding, and steel electrodo ,
around the conductor. These circles are in planes weld toward a heasy tack weld.
perpendicular to the conductor and are centered (3)If forward blow causes trouble, place Work angle, Travel angle, Techn,ique of Type of Position of on the conductor. ..
wc! ding deg deg welding joint Flat 90 5 10' Backhand posed on the steel and across the gap to be w elded.
Groove 80 00 Bac "
e turizontal 90 5-10 yt a Dackhand The flux in the plate does not cause difficulty, but unequal concentration of flux across the gap or g
K a ,
f Oniove Overhead , ,
ratet ilorizontal 45 5-10' Backhand around the are causes the are to bend away from / Ground )
Itlet verticat-up 35-55 5-10 Forchand i .he heavier concentration. Since the flux passes Fillet Os crhe.id 30-45 5-10 Backhand _,
j ,vough stect many times more readily than it flg. 2.15-The er/ect ofgroundk, cation en
- ~ - ~ T '
mmw.awnw - - _ _ - _ .
.7. . . .
I have read the foregoing affidavit, which was prepared under my personal direction, and it is true and correct to the best of g knowledge and belief.
I h 'l Y (5[gned),'y Date: m 78 /9,94
'j -
l STATE OF N e ~ h b COUNTY OF M %b On this, the 1 Ts kh ' day of bum - ,198h, personally appeared u
b4 g h m ale , kn n to me to be the person whose name is subscribed to the foregoing-instrtsnent, and acknowledged to me that he executed the came for the purposes therein expressed.
Subscribed and sworn before me on the 3%Ikdayof h%~ ',
~
198N.
c~vo es t es State of r ,- 3 l .
' MY c0?."/ISSIO!J EXplRES 'JAflUARY 9.1937 My Comission Expires:
l t
l
- . .-- . - - _ .