How should I qualify a an Autogenous Square Edge Butt Weld? 2

[iam42]

Hi guys,

I am scratching my head here and was looking for some advice on how to qualify (ASME IX) the butt weld shown in the sketch.

Any ideas would be appreciated. I would prefer to use no weld wire if possible.

Dimension are in inches.

 

The sketch presumably shows the configuration of the product, which looks like a partial penetration joint (some kind of seal weld?). That would not be feasible for ASME qualification since there is now way to extract bend & tensile test coupons. Autogenous welding is entirely feasible, but I would expect that small protrusions to melt back when welded, leaving it well short of 0.050" high.

Fortunately in ASME IX, joint design is not an essential variable, permitting you to devise a PQR configuration that delivers the variable you need while yielding testable coupons.


"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[iam42]

Thanks for the input Ironic, the 0.05" tab would be sacrificial.

I am still a little confused on what the test plate would actually look like.

 

[DVWE]

Is this a strength weld, or seal weld of some type?

How much penetration are you trying to get?

The devil is in the details; she also wears prada.

 

[iam42]

DVWE,

It is a seal weld for a vacuum

 

[DVWE]

I don't think Section IX covers a weld like that, but I could be wrong.

Personally, I would do a few mock-up tests and perform macro-etches to ensure the required .03 depth of fusion.

The devil is in the details; she also wears prada.

 

[meltedEng]

DVWE Sec IX says almost nothing about type of joint or use. That's other codes, if his client wants SEC IX and they sold that, then they should try to comply.

If this is round, maybe use QW-193, if client wants a Sec IX procedure, then as Ironic said, you must make a weld between samples of a different configuration on proper base metal thickness and weld metal thickness (make this weld between the same .5 x .5 blocks, but make a 1G weld, machine away the excess and test that portion.

I agree with you on path forward though, iam42 should simulate the joint and perform macro-etch tests to ensure it meets requirements. Run as many samples as needed to be confident in your results

If it is force bearing (holding a vacuum), then you may have it physically destructive tested by ASME VIII Div 1, Non-mandatory appendix A.

 

[DVWE]

DVWE Sec IX says almost nothing about type of joint or use.

I understand that. That's exactly why I didn't type "Joint Type" - I typed "Weld" - And by weld, I mean seal weld as per the OP's answer to my question.

Section IX defines what a seal weld is. It does not describe how to qualify. That's what I'm saying.

If it were me, I would explain that to my client, and I would also demonstrate how I achieved the weld size, or depth of fusion, by means of macro-etch and recorded set of welding parameters. How the weld size was derived is up to the engineer who determined it would be sufficient for its purpose.

The devil is in the details; she also wears prada.

 

[Tmoose]

If those are machined parts, cutting off all the material to leave the nubbins may cost more than the price of weld wire.

I'd think about plunging a radiused groove in flat parts adjacent to the former nubbin to create the nubbin with a much improved stess contour.

 

[gtaw]

You could make the autogenous weld, obtain some very limited penetration, machine all the unfused metal from the root side. You will be left with a thin piece of sheet. Test the sheet containing the weld as you normally would. Perform your visual, tensile, and bend tests.

Al

Best regards - Al

 

The construction code may have something to say about seal welds.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[iam42]

Thanks for the help guys, I am probably gonna try and qualify a procedure along the lines of what gate described.

 

[CWEng]

We recently qualified autogenous edge seal welding, similar to your configuration, except it was made from all sheet metal instead of a thicker piece of material. As mentioned, the joint configuration required to get bend and tensile tests are not similar to production. To qualify, we ended up autogenous butt welding tubes (.035" thick) together with a back-purge. Then did two full cross section tensile tests, and root/bend tests. It demonstrates the strength and ductility of the material combination, but demonstrates nothing about the joint configuration. For that, we also did some macros, in our case to make sure the weld thickness was at least as thick as the original base metal.

 

[DVWE]

iam42,

Just use caution with what CWEng has suggested. If you use material .035" thick, you will not be qualified to weld .5" thick base metal.

I would talk to your machinists before attempting what gtaw has suggested. That may or may not be feasible.

The devil is in the details; she also wears prada.

 

[CWEng]

I would call the base metal for his weld .02" thick, not 0.5" thick. The thickness is at the weld.

 

[DVWE]

The way I see it:

Base metal (T) is .5", and .55" if you want to be exact, counting the tab.

Weld thickness, or size, (t) is .03" per the welding symbol.

The OP has stated that the tab is sacrificial, so I assume that means complete consumption of the tab and .03" of penetration into the .5" thick base metal.


The devil is in the details; she also wears prada.

 

[CWEng]

Well, the weld size (penetration) is .03" on a .05" tall tab, so the penetration could be in the tab only. But if you are right, and they expect .03" penetration into the 0.5" thick base, and the tabs are completely gone, then that would fit your scenario. There needs to be more detail given on that issue.