Sharp corners on test coupons cause failure? Yea or nay?

[MarkG]

Well, last week's bend test (3/8 plate, open root, uphill MIG) I thought everything was OK and wanted to ease or radius the corners some before I bent them. The instructor seemed to either be in a hurry to get out of there or something and said 'it doesn't matter'!

Well, I know it matters and that a sharp corner can be a stress riser, but since I'm the student and he's the teacher, I let him test it before I could dress the corner as much as I would have liked, and sure enough, it cracked on a corner!! I was so ticked.

Aren't we allowed to ease the edges of bend test coupons? I feel like it could have made the difference in this case.

Thanks. We're getting down to the wire----I really want to pass this thing.

Fast forward to today:

I re-took my vert uphill MIG test and will cut the test coupons tomorrow. I will be 'easing' the edges to eliminate sharp corners as a cause for failure at least!!

I'd be interested in hearing how others have treated test coupon corners. I'm sure it must have some, perhaps a LOT, of effect!

 

[lotsoftools]

I always smooth over the edges on my test coupons. The smoother the better is what my instructors told us.

 

[BLJ]

I always smooth over the edges on my test coupons. The smoother the better is what my instructors told us.

same here.

 

[jhelrey]

Huh? Test coupons?

 

[jrlp]

Of course it matters. Radius the corners, blend with a flap. Hell, polish it mirror smooth if they let you. The smoother, the better.

 

[lotsoftools]

Of course it matters. Radius the corners, blend with a flap. Hell, polish it mirror smooth if they let you. The smoother, the better.

I like to use an 80 grit flap wheel to make them shine.

 

[lotsoftools]

Huh? Test coupons?

Welding bend tests.

 

[MarkG]

Thanks. Yep, I'll do that on tomorrow's tests. Taking some fine sandpaper with me too, to follow up the flap wheel.

 

[Ozwelder]

Radius of the corner edges is good but needs to be in context with your qualifying code.

I do welder qualification testing to (Australian) codes regularly as part of my job.

You did not mention the code you will be tested to , but it should reference the dimension or limit of dressing permitted. The dressing is best done on a belt sander/linisher so that the abrasive lines are lineal ( lengthways) not transverse
( across).

A successful weld will take a full 180 degree bend .Die and former openings are specified as are strip width and lengths.

Best of luck in your test
Ozwelder

 

[welder4956]

AWS D1.1 and ASME Section IX both permit a slight radius of the edges. The acceptance criteria also states that cracks originating on corners are to be disregarded unless there is evidence they originate from a defect.

 

[5mall5nail5]

MarkG no offense to anyone here but you want to go over on Welding Web. A LOT of guys familiar with your test there. I hear so often that things "fail" that aren't failures or that teachers "teach" based on what they "think". Go find out whats up.

 

[MoonRise]

The specific 'allowed' corner radius on the test specimen depends on the Code being used.

But generally, yes, you slightly radius the corners of the test coupon. 1/8" radius max on the corners/edges is usually 'allowed' (transverse face and root bends), but check the specific Code for any details.

re: surface finish or roughness, AWS usually says 125 micro-inch Ra or better (smoother)

"Cracks that occur on the corners of the specimens during testing generally are considered irrelevent." exact quote from an AWS insection guide. But again, see the exact Code that the test is being conducted to.

 

[welder4956]

OK, here is the wording in both AWS D1.1 and ASME Section IX on corner cracks on the bend test. I don't know what Code you are testing to, but these two are the most common.

ASME Section IX - 2010, QW-163: Open discontinuities occurring on the corners of the specimen during testing shall not be considered unless there is definite evidence that they result from lack of fusion, slag inclusions, or other internal discontinuities.

AWS D1.1 - 2010, clause 4.9.3.3:
(3) ¼ in. - the maximum corner crack, except when that corner crack results from visible slag inclusion or other fusion type discontinuity, then the 1/8 in. [3 mm] maximum shall apply.

Specimens with corner cracks exceeding ¼ in. [6mm] with no evidence of slag inclusions or other fusion type discontinuity shall be disregarded, and a replacement test specimen from the original weldment shall be tested.