Plate Bend Test Failures GMAW

Date 07-13-2007 16:13 Edited 07-13-2007 16:16

"the scale or the glass like residue was not removed between passes"

Bingo! Fail every time, it promotes incomplete fusion between the passes/layers.

edit: about running multiple passes without stopping to cool.....I have found the heat usually will help the bend test, too cool will make it brittle. Some may disagree on mild steel, but this has been my experience.

By 803056

Date 07-13-2007 17:37

I agree with your assessment John. Failure to remove the "glass" oxide islands between passes is a invitation to fail the bend tests.

My experience with the temperature of the test coupons while bending has been they must be near ambient temperature. I have done experiments with bending A36 carbon steel samples at various temperatures. I found that if I could not hold them comfortably in my bare hand, they were too hot and would fracture when bent using the standard 4T bend diameter.

I have found that I have to let the samples cool after grinding off the excess weld face and backing bars or they have strong tendency to fracture with the "grainy" appearance mentioned in the post.

High interpass temperature while welding carbon steel hasn't been a problem based on my past experience. The high interpass temperature usually makes it uncomfortable for the welder, but unless notch toughness is concern, interpass temperature hasn't seemed to adversely affect the results of the bend test.

Best regards - Al

By jerrykroll (**)

Date 07-13-2007 16:19

If you are doing overhead you must be using short arc (GMAW-S). It is notorious for lack of fusion/penetration, particularly on sections over 1/4 inch or so.

In my experience, to get good fusion, you must stay ahead of the puddle and put in smaller weld passes.

If you are weld testing to code you most likely need to qualify the procedure. It is not considered prequalified by D1.1.

A standard WPS (SWPS) does exist - B2.1....004, but it covers only up to 10 gage (0.135")

Consider doing a weld test in the flat position, just for assurance your parameters work.

good luck and good welding

By js55

Date 07-13-2007 18:35

I think John and Jerry both are on to something. No cleaning and short circuit transfer are problematic. Also, the overhead focator may play. My guess is that though overheads failed flats may pass. They're easier.
Carbon steel takes heat very well. It precipitates carbides and promotes pearlite at the expense of polygonal or especially acicular ferrite which 'softens' the microstructure. It also however enlarges grain size. So while the heat may hurt impact 'strength' it can help ductility for bends, and becomes somewhat unpredictable for lateral expansion for impacts due to the ductility factor in LE measurements.

By commonarc (**)

Date 07-16-2007 01:37

What is the technical term for the glass like substance left on a GMAW bead after welding with solid wire? My reference books fail to mention what this actually is. Thanks.

By aevald

Date 07-16-2007 04:08

Hello commonarc, I believe the term for this glass like residue is "silica", it is also an additive in wires to assist in carbon distribution as I recall. I wouldn't quite agree with js55's assessment that flat welds are easier to pass than overhead welds though, as gravity will assist in bringing out certain types of inclusions that flat welds would otherwise trap due to the involvement of gravity, this is just a personal thought of mine and based on being a reasonably experienced welder myself and having done a fair amount of destructive testing to verify this. Just my $.02. Regards, aevald

By js55

Date 07-16-2007 15:00

Allen,
I'm not sure gravity has as big an effect on inlcusion location as differential surface tensions, differential melting temperatures, and incompatible soldification matrices (slags and oxides if large enough) being pushed by solidification fronts.
I believe this is why oxides and slags have not only little trouble finding their way to the top in the great majority of cases, but is actually critical to the the welding processes. Not to mention that I believe the primary metallic matrix (Fe, Mn, Cr, Mo, etc) is more dense than slags and oxides and would therefore argue that the motion of these inlcusions would actually oppose gravity in a metallic solution.
Having said this it would certianly be interesting if somebody has a study demonstrating greater density of inclusions in 'lower' layers (gravitationally speaking) of weld passes that successfuly isolates other variables that might argue for gravity's influence.

By aevald

Date 07-16-2007 17:01

Hello js55, thanks for the very specific and scientifically explained comments in regards to my post. I am certainly not a metallurgist or engineer in the sense of being educated as such. I many times make my observations from what I see as a practical result(don't have the scientific jargon or knowledge to back it up), it is very likely that my use of gravity to describe the propensity of welding failures more so related to flat test plates over the overhead plates could be due to other phenomenon. Having said that, when you look at most slag inclusions concerned with overhead, flat, or horizontal plates, the inclusions will generally be greater for flat and horizontal plates and be physically located at either the edges of respective beads on the flat plate or on the lower edges of respective beads on the horizontal plates. That is why I included the statement regarding gravity. I also realize that machine settings, operator techniques, and a considerable amount of other variables can greatly influence these results. Cleaning of slags, silicon deposits, glass islands, and any other deposits between bead deposition is still paramount to having success on clean and clear multi-pass joints, as well as using proper machine settings and operator technique. Thanks again js55, you and the others always keep me thinking and learning. Regards, Allan

By js55

Date 07-16-2007 17:42

Allen,
Having said all that stuff, I thought it important to preface with "I'm not sure". I think my argument is sound, but I had not the intention of disputing your experience which I am just as convinced is valid, though your results of slag predominance haven't been my experience. I find that easier manipulation, lesser oscillation extremes, less variation in arc length or arc destabilization, and higher welding currents and voltages associated with flat welding assist in minimizing slag inclusions more so than with overhead where manipulation struggles and parameters are cooler. Perhaps our agreement lies in the middle ground of horizontals, where your thinking and my thinking appear to come closer.

Also, having reviewed our posts I think we are talking about two different, though related things (my misunderstanding). Non metallic inclusions not so influenced by technique, and slag (FCAW/SMAW) or silica (GMAW) inclusions, which are greatly influenced by technique and weld parameters.

By aevald

Date 07-16-2007 18:09

js55, thanks for the additional comment. likely you are correct in where we are coming from. Regards, Allan

By Lawrence (*****)

Date 07-16-2007 04:13

Not a small question.

Most of what those islands are would be silicon/silica deposits added to the filler wire as deoxidizing and scavenging agents. They float the contaminants out of the molten pool and they must be removed if multiple passes are required.

An example (a simple one) would be ER70S-6 Vs ER70S-3 The -6 wire has greater silicon and deoxidizing agents to allow welding on steels with moderate mill scale or even rimmed steel, with the -3 electrode having less of those agents and more suited to a very clean prep and killed or semi-killed steels.

Here is a good article by the Fabricator
http://www.thefabricator.com/Consumables/Consumables_Article.cfm?ID=653

By Lawrence (*****)

Date 07-16-2007 04:33 Edited 07-16-2007 04:41

An additional thought unmentioned in this thread.

Rolling/Grain direction.

Omer Blodgett says...... (nobody knows better than this guy)
"Steel is not fully isotropic, and ductility is sensitive to rolling direction. When steel is strained parallel to the direction of rolling, the highest values of elongation are obtained. When it is stretched perpendicular to the rolling direction, ductility is reduced. The poorest ductility usually is measured in the through-thickness direction."

The full article and drawings are here
http://www.weldingmag.com/323/Issue/Article/False/13719/Issue

I have this printed in color, laminated and posted

Just this summer I experienced the ugly side of this phenomina whilst doing some contract training at a local fabricator... The shop forman decided during the afternoon shift that he wanted to do bend tests the next morning and had somebody "prep" the coupons for me....... thanks alot!

Every weld broke, face, root all of em... I was beside myself. This was simple solid wire spray transfer on plain carbon and I monitored every pass. We did things over with properly prepped material and everybody qualified, but I'll tell you it is a terrible feeling watching one strap after another not even get to the bottom before it snapped!

This did nothing good to my already notable reputation as an autocratic control freak on the shop floor.

By ssbn727 (*****)

Date 07-16-2007 04:56 Edited 07-16-2007 05:00

Hey Larry!!!

So I can add your name to the not so perfect person's list??? :) :) :) just kidding friend!!! ;)
Btw, I'm on the list also so, "weldcome" to the club :)
By admitting your shortcomings whether or not they were a result of your doing or not, Let me also "Weldcome" you to the: "It takes a very strong and humble person to admit catastrophies/mistakes/ shortcomings in public." list... It's a pretty exclusive club, and I for one am a very proud member of this club :)

Btw, We're looking for many more members to join so, if anybody needs to know the prerequisites for joining just refer to the above stated requirements :) Discretion is an option :) Honesty is a REQUIREMENT!!! Larry, YOU IMMEDIATELY QUALIFY :)

Respectfully,
Henry

By Lawrence (*****)

Date 07-16-2007 06:03

Henry.
It was a seemingly small detail........I thought.

I originally contracted to just provide process training... But late in the week it was decided that some guided bends would be a nice cherry on top..... I'll never forget that sick feeling and the look on the faces of those welders as every strap busted...

Gotta oversee every detail.

By ssbn727 (*****)

Date 07-16-2007 11:30

I know what you mean friend :) It's happened to me in different situations more than once also - darn it!!! :)
You know the phrase - manure happens :)

Respectfully,
Henry

By 803056

Date 07-16-2007 13:31

Hello Lawrence, et al:

Maybe I missed somthing, but in this case I thought it was the welds that failed.

You are on the mark about the direction of rolling, but my experience has been that the majority of failures occur in the carbon steel base metal if the direction of roll is incorrect.

Over the years, I've had many clients insist on providing their own test coupons to "save money" when testing their welders. It isn't uncommon to have a high failure rate due to the "direction of roll" issue. Even after warning them, they fail to understand the importance of D or R until the welders start failing. I've even had clients call me in for cracking issues when bending hot rolled sheet. D of R is often the problem and solution.

I cut my test pieces from HR bar to eliminate the D of R issues. There is no question of what is the direction of rolling with bar stock.

Best regards - Al

By ssbn727 (*****)

Date 07-16-2007 21:47

Please take off those x-ray glasses Al ;)
You're reading way too into the "thickness" of what I was alluding to, and noticing some shadows that are'nt there :) In other words, I was making some humor out of what Larry experienced. :) I apologize if it went over your head :) WWWOOOOOSSSSHHHH!!!!! :)

However, I do concurr with the D of R issue which I'm sure all of us at one time or another have become victims of this condition when it comes to bend testing. :)

Respectfully,
Henry