Critique this Weld

Date 07-30-2009 15:04

Bob, Will the UT guy have enough material to perform his scans?......I ran into something like this recently

Attached is a ACAD drawing of a 2.5" thick piece of material that would be scanned with UT(45°,60°,70° shear waves)

Attachment: SketchesforUTreportsModel1.pdf (12k)

By aevald

Date 07-30-2009 15:31

Hello Bob, that is a rather interesting call out. I say that for a couple of reasons. The first one that I noticed was his "requirement" to layer the beads on the horizontal plane as the groove is being filled out. Most all of my experiences with this sort of weld-out would stack the beads on the vertical plane and apply the layers in that manner until the groove was filled. Additionally, the detail doesn't specify an alternating progression so possibly a greenhorn might take this to mean that he/she should weld one side and then the other(although most would certainly not approach this in that manner). I am also making an assumption here that this isn't simply A-36, the peening requirement would possibly indicate that. Another item that I would feel is lacking is a specification for back-grinding/gouging for the root interface of the joint, as detailed as this description is, I would expect this information as well. These are some of the questions that I might have with regard to what you have presented here. Lifting Lug points are certainly something that needs to have special attention paid to, as life and liability hang in the balance(pardon the pun). Interesting post, thanks for sharing, looking forward to others sharing their interpretations/comments. Best regards, Allan

By Bob Garner (***)

Date 07-30-2009 16:24

The material callout is not clear. There is a callout for A36 and a duplicate callout for A572 Grade50. In addition, there is a callout for a maximum carbon equivalent of 0.40 (This drawing is preliminary and the material specs are not clear yet, as you can see.)

One confusion I have is the requirement to make a Charpy test near the center of the weld. Isn't a Charpy a destructive test? Is it intended that a sample weld be Charpied instead of the actual production weld?

I have a constructive opinion on this weld but I want to learn from you all before I say somethin' stupid.

Bob

By aevald

Date 07-30-2009 16:51

Hello Bob, anytime I have seen Charpy's called out they are definitely not a part of any production welds. They are a stand-alone seperate test that is generally used to determine a workable filler/base metal regimen prior to production welding. That's my interpretation anyhow. Best regards, Allan

By ssbn727 (*****)

Date 07-31-2009 01:40 Edited 07-31-2009 02:55

Hi Bob!

Is this USN, or ONRL?

I ask this because it sounds like either a mild steel (Although, I agree with Allan that A-36 doesn't come to mind here) to High Tensile Steel CJP joint, or a HTS
(Reason for recommending E-7018 (Even though there is a caveat as far as the engineer's preference on the Electrode Filler Metal in the notes) with plenty of control on pre-heat & interpass temps) to a Q&T steel call out, and guess who does tons of that sort of stuff?

It looks like standard stuff for naval lifting pad eyes... The peening used to eliminate most of the trapped residual stresses (even though VSR is also recommended also), and what looks like a tempering bead technique used to avoid a higher PWHT temps is something very standard when working close to stuff related to USN surface, or even SUBSHIPS and their associated ancillary components... I mean, you know how the Navy is... ;) :) Btw, That Charpy test is to be performed on a sample, and not on the pad eye itself unless someone doesn't know their "azz' from their elbow. :)

It's either their way, or you end up back on the highways - "Land-lubber!!!" ;) ;) ;) (Not you Bob! ;) )

Respectfully,
Henry

By aevald

Date 07-31-2009 08:19

Hello Henry, another astute observation on your part, very well could be that this individuals background has connections to military work. Earlier in my career we would quip a bit when we had jobs that were Navy, U.S. Army Corp of Engineers, and others. Government jobs were often described as: "an elephant is a mouse built to military specification". I mean no disrespect when I say that. Best regards, Henry, and thanks for also pointing out those items, they do bring back some of the memories of work that I performed in that arena, challenging and very thorough. Allan

By MMyers

Date 07-30-2009 16:34

I agree, this doesn't sound like a common every day steel. Hydrogen control, preheat on cutting, preheat and post weld bake on welding, and peening tell me someone is concerned about hydrogen induced cracking. This may be the reason for the horizontal stacking sequence, or it could be for stress or distortion control, I don't know.

By js55

Date 07-30-2009 17:18

Charpies at 70deg?
How will you verify 33% dilution?

By DaveBoyer (*****)

Date 07-31-2009 02:49 Edited 07-31-2009 02:53

I don't know sh_t from shinola, but is it really practicle to do a magnaflux between layers when You have pre & post weld heat treating requirements? Can it be done at 300f?

By 803056

Date 07-31-2009 04:28

I could write a book on this problem. You've already answered some of my initial questions. The questions I would toss out for the engineer are:

1) what is the base metal (you have already stated that hasn't been determined).
2) how was the preheat determined without knowledge of the base metal or the carbon equivalency?
3) which leg of the fillet weld measures 1/2 inch and which one measures 3/4 inch?
4) a few light blows of the slag gun fitted with a round nose chisel is hardly considered to meet the definition of peening. Peening has to cause plastic deformation of the weld deposit. A few pecker marks of the surface of the weld face don't do it.
5) without knowing what the base metal chemistry is, a slow cool down from 300 degrees isn't necessarily going to do much good. A hydrogen bake out is performed at higher temperatures with a hold time of several hours.
6) some base metals are cooled to room temperature to ensure the transformation of retained austenite to martensite which is subsequently tempered at some temperature below the low transformation temperature and then cooled rapidly (Q&T HY steels in mind).
7) what welding code or standard is applicable for the purposes of qualifying the WPS and performing the NDT requested, i.e., what acceptance criteria is to be applied.
8) if notch toughness is a concern, interpass temperature is something that needs to be monitored and controlled.
9) notch toughness is lowest in the HAZ, therefore the test samples should be oriented such that the notch is within the HAZ. Notch toughness must be performed when the welding procedure is qualified. The test temperature is often required to be the lowest anticipated service temperature (depending on the welding standard referenced). It appears this aparatus will be used on sunny summer days only.
10) Oxy-acetylene cutting is abbreviated as OAC.
11) preheating before OAC cutting is required, but what is the preheat temperature and what is it based on if it hasn't been determined what base metal (specification) is?
12) as mentioned by Allen, the sequence of welding should alternate from side to side if possible. If all the welding is completed on one side before welding the second side, the depths of the grooves should be deeper on the second side to be welded.
13) vibratory stress relief hasn't been all that successful on higher strength base metals, i.e., the higher the strength, the less effective the process. I don't believe it has been accepted by any welding code in lieu of thermal stress relief.
14) it is preferable to position the weld joint so the weld can be layered perpendicular to the orientation shown in the sketch.
15) mag particle testing is a good idea if the base metal is susceptible to hot cracking or the base metal is allowed to cool to ambient during the course of welding over several work days or shifts.
16) UT should possibly be delayed a day or two if hydrogen cracking is a concern. That is, don't perform UT immediately after the weld has cooled to ambient. Allow time for the hydrogen to do it's dirty little deed. UT serves little purpose if the hydrogen cracks develop after the UT is performed and go undetected.
17) depending on the type of base metal selected, there may be heat input considerations. That is, a maximum heat input may have to be defined by the engineer.

However, it appears the engineer has put some thought into the process and I'm taking pot shots at it without knowing all the details. Dirty pool on my part, but my comments are intended to give the engineer some additional things to consider.

Thanks for the challenge!

Best regards - Al

By aevald

Date 07-31-2009 05:03

Hello Al, your post had many items that I would not have thought about. Whether applicable in this instance or not, you certainly gave a host of items for consideration that should be observed. Thanks for the very concise narrative.
Even though the fitter/welder isn't necessarily responsible for seeing that these questions are answered, I do believe that when they are aware that these considerations exist they might be more apt to become a part of the QC process and avoid some of the costly mistakes that seem to be a part of doing business in many shops. Best regards, Allan

By js55

Date 07-31-2009 12:53

Al,
"9) notch toughness is lowest in the HAZ"

Not to digress but, this is not necessarily the case. Especially with a material like A36. If the base metal is a large grained structure (as is A36), the grain refinement of the HAZ can improve the toughness. And most likely will. This was the basis of the old nuclear 'HAZ must equal or exceed the UBM" requirement (that still exists although it has recently been revised to make it possible with finer grained modern steels) back in the days when base metal wasn't so good. Now if your talkin SA-516 then there's little chance the HAZ is going to be equal.

In fact, the first time I qualified nuclear WPS's under this requirement I was stupid. I used SA-516 because I wanted a good material for nuke quals. But then I couldn't get the HAZ to beat the BM. Duh. Then a wise old friend said to use A-36 and told me why. Problem solved.

By 803056

Date 07-31-2009 13:46

No disagreement from this quarter, I was trying to toss out those items that deserve consideration by the designer. I should have been more careful with my choice of words. You bring up some important considerations.

I was thinking along the lines of high strength low alloy steels, but failed to limit the discussion to those alloys. Open mouth, insert foot!

Thanks.

Best regards - Al

By js55

Date 07-31-2009 14:33

Al,
With HSLA's (TMCP's - Q&T's, etc) the HAZ will definately be problematic. The welding regime destroys essentially the processing utilized to achieve the exceptional mechanical properties.

But I think your dead on. There is info missing and info that seems either non applicable for the intended service or just plain overkill.
But it is a suspect indictment since, as you say we don't have the full picture.

By aevald

Date 07-31-2009 04:47

Hello again Bob, forgive me for sidetracking your thread here, but likely you might appreciate some of the responses of the questions that I am proposing here. Have heard of Charpys being discussed and referred to many times and not really paid all that close attention to them. Since the mention of them here and also the response from Gerald about temperature and dilution I would like to ask a few questions. When specifications call out for a particular filler metal to meet certain strength/temperature requirements does this mean that you weld out a weld sample of sufficient volume as to allow a specimen of the appropriate size(that is comprised solely of weld metal) to then be chilled/heated to that temperature and the Charpy test applied to it? Also, is Charpy testing only applied to either weld metal or parent metal and not a test of any sort of weld/parent metal interface? I did some googling, reading, and such and believe I have a little bit better idea of the testing but not necessarily it's real applications.
Bob, if you are completely aware of all of this please forgive me, I am not and I would guess there are others as well who would benefit from a rundown on this. Gerald, others, I would greatly appreciate some enlightenment here. Thanks and best regards, Allan

By Bob Garner (***)

Date 08-03-2009 15:38

First, thanks to everyone for their comments. When I first saw this weld spec, I realized I was totally unsure what the Engineer was trying to accomplish. Or more accurately, I knew what he wanted to accomplish but I didn't know how this spec was going to get there. I had a conversation with the engineer this morning and this spec was a compilation of Navy and AWS specs. The weld sequencing requirements were from studies of the failed beam-to-column connections after the Northridge earthquake. The materials are A36 throughout and this fabrication will not be subject to temperatures below 32 degrees. There were a number of questions that were posted to me that he could not answer, and this concerns me.

My training has been to understand welding but more for strength and serviceability requirements. The specifics of the weld are developed through the WPS and PQR. We do design work and as the engineer, we typically rely on outside expertise for things beyond our level of expertise such as geotechnical recommendations and welding. I was taught (and your replies have confirmed it) that welding is a specialty and that the engineer must know what he wants the weld to do, but the engineer should leave the weld creation to specialists. To do any more is to take on more responsibility than what is considered the standard of care today (our legal system, love it or not). However, I acknowledge that if an engineer knows welding intimately, there should be no problem with specifying more information to attain the desired goals.

I mean absolutely no disrespect to the engineer specifying this weld. This is a learning experience for me.

The good news is that this detail is going out to a fabricator that we have worked with for years so it will be done right in the end. And I've learned something here. Learning that you don't know is learning nonetheless.

Again, thank you for each of your replies.

Bob Garner

By HgTX

Date 08-03-2009 16:49

It's probably a learning experience for the engineer as well. Design engineers don't always know a lot about fabrication, construction, or materials, and sometimes they'll just throw everything they can find at a problem. I had one project where someone had specified all the supplemental requirements available for the particular ASTM steel specification used. Not because they'd thought about it and decided they were necessary, but because they were there and more is better.

Hg

By ssbn727 (*****)

Date 08-03-2009 17:49

It's no wonder this one was such a Cluster***K!!!
Combining Navy & AWS specs is very much like attempting to mix oil and water. ;) :) :)
Good luck with this one Bob!!! :)

Respectfully,
Henry

By js55

Date 08-03-2009 17:09

Allen,
When CVN (Charpy V Notch - there used to be others such a a keyhole notch, sometimes still used but not popular)testing filler metals it is usual to follow the AWS specification method. Weld a groove, smack 5, drop the high and low.
When testing for weldments it is usual to follow the joint design most commonly used in fabrication and follow ASME IX or AWS D1.1, with construction code specifics included.
As for application to WM, HAZ, or BM, it depends upon code and temperature. ASME III requires BM comparisons. Most other codes do not. WM is a given if testing is required, and then sometimes HAZ's as well, generally as it gets colder.
Some codes require tilting of the specimens for HAZ's to allow as much HAZ in the specimen as possible.
B31.1 does not require them AT ALL.
CVN's are used generally for colder temp applications but are increasingly used for higher alloy ferritics, bainitics, martensitics, as a sort of verification at room temp. Such is common with Grade 91.
Research guys prefer CTOD's becasue they can tied into design calcs and actual matrial properties. But they are slow and expensive.
And then there's notched tensiles for Ni base alloys etc. when it really gets cold.

By aevald

Date 08-03-2009 18:23

Hello Gerald, thanks for the reply. When I get a little bit of spare time I'm going to look into all of this a bit more indepth. Best regards, Allan