Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic American Welding Society Services / Certifications / Root Bend Test Evaluation
- - By flaguy (*) Date 09-27-2012 15:27
The attached picture is from a welder qualification root bend test.  A36 steel, E7018 electrode, 3G up with backing.  There is a series of tiny pin holes along the root fusion lines, they are not vey deep, just barely below the surface.  What would cause this?  Would this coupon Pass or Fail?  The face bend test was fine.  Thanks
Parent - - By OBEWAN (***) Date 09-27-2012 15:32
Did it have a crack or tear?  Usually that is what fails a bend test.

If there is a porosity requirement then that is usually examined with a macro etch test.
Parent - By flaguy (*) Date 09-27-2012 15:45
I finally got the picture to go up.  Anyway, it did not crack or tear, just two lines of pin holes for lack of a better descrption.
Parent - By waccobird (****) Date 09-27-2012 16:00
flaguy

Not enough information to say whether it passed or failed.

What Code?

But as to the cause of these discontinuities (I am going on a limb as I wasn't there witnessing the fit-up)

It appears there was space between the base metal and the backing bar that allowed slag to flow in thus preventing fusion.

A lot of people prepping the test plates grind too much off the backing bar side of the base metal thus creating a void along the joint.

Just my ΒΆΒΆ's

Marshall
Parent - By aevald (*****) Date 09-27-2012 16:17
Hello flaguy, I might offer a slightly different explanation for what I see in your pic. You said that this was a 3G with backing. This would lead me to believe that the plates were bevelled to a knife-edge and fitted to the backing. If the preparation of the knife-edge portion of the bevels wasn't ideal there might have been a bit of space between the edge and the backing that could have trapped some irregular pools of slag at the fusion line. Additionally if the contact surfaces where the knife-edge and the face of the backing weren't ground to bright metal there might have been some mill scale or other contaminants/barriers to prevent thorough, even fusion where the backing and knife-edge plate contacted and overlapped one another. I always try to fit my plates to a point where they are as flat and parallel to one another as possible and also try to ensure that they have all of the mill scale or other possible contaminants removed prior to fit-up. When I apply the weld metal on the root I attempt to use a speed and technique that allows both edges to be fused and tied to the backing. Just a couple of items for your consideration. Best regards, Allan
Parent - - By Lawrence (*****) Date 09-27-2012 16:51 Edited 09-27-2012 16:58
My  "opinion"

I see this regulary with SMAW and my welding students.

For me it is a good opportunity to introduce liquid penetrant inspection.

A 10x glass is also helpful or an optical comparitor

I would wager to guess that 75% of the time those pin holes (indications) are linked (with cracks too tight to view when you are past age 50)  But a simple red dye penetrant check will reveal linkages between obvious discontinuities.  I don't give a darn about sniveling that "the code doesn't call for magnification or dye"  an indication is an indication, a crack is a crack, no matter how tight, its there!

For guided bends we can assume they have already passed visual inspection via (table 6.1) and we walk away from that table and move forward to 4.8.3.3

In 4.8.3.3 we jettison terms like "porosity" and simply look for "discontinuies" and we measure them and compare them to the acceptance criteria. 

If the pin holes (indications) are linked than the measurement is taken of that length,,, Longer than 1/8"?... Fail... If shorter, contunue...If there are several than the cumulative length must be added ..  Along with the length of every single other indication (there is no minumum length of indication as there may be for porosity and undercut in 6.1)  Even those tiny indications must be assigned a measurement value and added to the caluclated total.

Only then can you determine if the thing passes or fails.

Again  assuming D1.1

I think Allan's discriptions of cause are excellent.
Parent - - By 803056 (*****) Date 09-27-2012 17:33
I believe Allen and Lawrence and I are in the same corner in this discussion.

As for the evaluation; any single discontinuity longer than 1/8 inch is usually cause for failing the sample. AWS D1.1 requires the length of the multiple open discontinuities to be summed. If the sum is more than 3/8 inch, the sample fails. If the applicable code is ASME Section, the open discontinuities are not summed and the sample passes.

As for the cause; the welder was on the low side of the amperage range and did not hesitate when he reached the corners where the bevels met the backing bar. From the photograph, I cannot see whether there is slag in the open discontinuities or if they are "clean". I have seen similar results with GMAW-S as well as SMAW and FCAW. If the discontinuities show no evidence of slag, it was simply a problem with securing complete fusion at the corners of the bevel and the backing. Technique problems.

Best regards - Al
Parent - By jwright650 (*****) Date 09-27-2012 17:45
Yup lack of fusion, and when the coupon is bent, they flower open and look like pinholes.
Parent - By WeldorJoe (*) Date 10-03-2012 20:36
Just an addition to Al's good answer.....D1.1 4.9.3.3 states that 3/8 is the sum of the greatest dimensions of all discontinuities exceeding 1/32...but of course, less than 1/8.
It is hard to tell in the picture if some of the cracks exceed 1/32.
I personally would like to measure all cracks.....but hey....it is in the code.
JJ
Parent - - By waccobird (****) Date 09-27-2012 17:52
Lawrence

what is a discontinuies ?

Marshall
Parent - - By aevald (*****) Date 09-27-2012 18:11
Why Marshall, that's farmcode terminology........but you have to check the glossary there for an expanded explanation. JUST KIDDING I HOPE YOU KNOW and the farmcode reference is an inside joke that Lawrence and I shared earlier.
Parent - - By waccobird (****) Date 09-27-2012 18:17
Allan

I can't afford all them there Code Books Like Ya edumicators can.

:cool:

Marshall
Parent - By 803056 (*****) Date 09-27-2012 19:56
Never mind buying the codes, you would have to learn to read first! :twisted: Fortunately for many of us D1.1 has plenty of sketches. If not, I would never have passed the CWI!

You know I'm only messing with you Marshall. :grin:

Best regards - Al
Parent - - By aevald (*****) Date 09-27-2012 22:27
Why heck Marshall, that's why the internet is so full of all of those "bootleg" copies of necessary code. Ya jist gotta know where to go lookin for 'em. Regards, Allan
Parent - - By 99205 (***) Date 09-28-2012 03:38
I know a arrogant, over bearing person that's so smart, he bought a internet copy of the code and printed it out on his home computer.  He plans on using it on an upcoming CWI test.  Hmm, I just can't remember if he's allowed to use it in the test.  Oh well, he'll figure it out, he's really smart.
Parent - - By 803056 (*****) Date 09-29-2012 13:35
Provided it is a licensed copy he can use it.

The printed copy will state who purchased it.

Al
Parent - - By 99205 (***) Date 10-03-2012 23:14
It's a bootleg copy.
Parent - By welderbrent (*****) Date 10-04-2012 14:57
Al's answer is of course the politically correct one.

However, I can tell you that given the information in the above post he will not be allowed to use it.  It will not be a 'licensed' copy and he will not be able to prove his proper possession of that copy of the code. 

Have a Great Day,  Brent
Parent - By dbigkahunna (****) Date 10-03-2012 22:34
I can see the bats and chains coming out when you pull the dye and magnifying glass out!:twisted:
Parent - - By flaguy (*) Date 09-28-2012 06:57
Many thanks to all who responded.  I learned a lot and I very much appreciate the helpful information.  I did forget to mention the test code was D1.1 and the plate was prepared to Fig 4.31 (limited thickness).  Thanks again.
Parent - - By eekpod (****) Date 09-28-2012 19:48
lack of fusion in the root pass, I see it on occasion.  Make sure the WPS the welder is using calls for the correct amperage range and polarity, then make sure he has the machine set right.
Now of course you can do this after he took the test and bust him out for not following proper procedures and protocalls, but depends on how strict your program is.

When I give atest I give the welder a hard copy of the WPS,then I ask them "have you worked with a WPS? do you know what it is?" then they say no.
Then I explain to them what it is, how it works and what the info on it means and how to use it.  Then I explain to them how to set the machine, I don't do it for them...if they want to be a welder, they need to know how to adjust it, not just weld with it.

Some people are open and undertand, others don't get t, and others are too stubborn to listen.  Those guys always fail the first time.
They don't have 20 yrs experience,,,they have one day of experience doing the same thing everyday for 20 yrs.
There's a difference.
Parent - By flaguy (*) Date 09-29-2012 12:54
Thanks eekpod.  I do things just as you describe - make sure they understand the WPS, check the machine before the test etc.  We are lucky here, our welders are eager to learn - a good bunch of guys from the Philippines.
Up Topic American Welding Society Services / Certifications / Root Bend Test Evaluation

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill