Welding Questions 1

[MrGearhead]

I am surprised there is no welding subthread here in the mechanical area... At least not one that I saw..

Issue: .090" thick 6061-T6 Welded Enclosure popped a butt weld seam during either a manufacturing vibe checkout or a low temperature ice test. Assumption is that it was the low temp during the ice test that popped the weld seam. Investigation showed that the weld was not really centered on the seam joint and therefore only a small part of the burn-though made it through the joint. It appears that based on the bow of the .090 plate that popped, there was theremal distress from the original weld op waiting to be "set free" by a significant low temp excursion... My questions are very general in nature:

1) How will a rework of the same weld seam work (or not)? Assumption is that a 6061-T6 was locally heat treated at the weld seam from the initial weld op. How will the 6061-T6 base material strength be affected?

2) For corrective action, an additional fillet weld on the inside of the enclosure and some gussets are assumed. Any comments?

3) Somewhat related... How watertight is a continuous weld expected to be? Aside form demanding a leak test on every box that goes through manufacturing, is there a weld callout that would basically force the box to be "watertight" at all the weld seams?

 

[kingnero]

That's a whole can of worms you've just opened.

1) about welding 6061-T6, you'll find plenty of threads in the metal/metallurgy ; materials ; and Welding/Bonding/Fastening subfora.
Basically, you'll fall back in the zero temper condition.

2) I think i need a sketch of the setup before answering that one.
Much depends on how much of the previous weld you're going to remove.

3) noe, there's no specific symbol, as all welds are supposed to be watertight.

 

[fegenbush]

1) How will a rework of the same weld seam work (or not)? Assumption is that a 6061-T6 was locally heat treated at the weld seam from the initial weld op. How will the 6061-T6 base material strength be affected?

The as welded condition for the weld should be assumed to be fully annealed. The strength values for 6061 (no temper) are published. Additionally, it is possible to check the weld procedure qualification report (PQR) to review any results of mechanical testing performed.

2) For corrective action, an additional fillet weld on the inside of the enclosure and some gussets are assumed. Any comments?

The design of the gusset would need to be analyzed. The presence of a gusset alone is insufficient and may have adverse effects (e.g. change in failure mode from rupture to burst)

3) Somewhat related... How watertight is a continuous weld expected to be? Aside form demanding a leak test on every box that goes through manufacturing, is there a weld callout that would basically force the box to be "watertight" at all the weld seams?

The weld callout would not force watertightness, but I believe you are looking for a seal weld. The only way water tightness could be assured is testing. Vacuum box, hydrostatic test, or dye penetrant testing are some options to consider.

 

[stanweld]

Weld soundness may also be an issue. How was the weld made - with what process - with what welding filler metal?

 

[tbuelna]

MrGearhead-

There seems to be a couple things that do not make sense regarding your description of the problem.

You came to the conclusion that the butt weld failed as a result of exposure to low temperatures during a portion of a qual test, but not as a result of dynamic loads experienced in the random vibe test phase immediately preceding it. How did you establish this fact without performing some kind of NDI on the weld area between test phases?

It would also be wise, before conducting any further qual testing, to perform a thorough stress analysis on your welded enclosure that takes into account factors such as weld defects, reduced mechanical properties in the weld HAZ, and residual strain from the weld process. Normally, the purpose of testing is to validate design/analysis work.

I would also recommend implementing tighter controls in your manufacturing processes, and additional QA steps both during manufacture and prior to delivery or testing. As you have seen, it is a waste of time and money to perform testing on defective hardware.

It would be very helpful if you would provide the relevant parts of the engineering drawing used to manufacture the enclosure. This would allow a more complete evaluation of the possible causes of the problem and potential solutions. Adding gussets and additional fillet welds to your enclosure does not address the apparent root cause of the problem, which is a deficient butt weld. Based on your OP, this enclosure appears to be intended for use as a test article in a qualification test. If this is the case, then it is important that your test article is manufactured exactly like your production hardware will be. This means producing an acceptable butt weld joint, and not relying on added gussets and fillet welds.

Good luck to you.
Terry

 

[racookpe1978]

0.090 is just barely past the sheetmetal stage, and - contrary to what you'd instinctively think - thin metal is much, much harder to weld than thick.

Even .125 is significantly easier than 0.090, and when I am making my own assemblies and staircases and decorative railings, always pay for the thicker steel to make assembly, cutting and welding easier.

Have you tried the simple assumption that the welder goofed up? Might be because the assembly is too hard, or he had a bad day, or he goofed. Or, what you are trying to do can't be done economically within his time and budget and training and jigs and fixtures.

What pressure was the test run at? .

 

[IFRs]

You identified the problem as weld that was not centered on the butt joint, poor penetration and insufficient fusion. Is the only question how to fix it? I would fix it with TIG by hand and then review the fixture, weld procedure and process.