Pulsed TIG vs MIG for aluminum welding

Date 03-22-2011 04:11

Aluminum mig welds are made every day that are more than good enough to pass any weld inspection. New technology is great but it is by no means the only way to get the job done. There are many benefits to TIG or MIG just depends on what someone wants to qualify or accomplish.
Typically dcen al welding is done on thick al sections. Al mig can get the job done much faster and just as good if done properly. Depends on many things but if given a choice I use mig or pulsed mig on anything over 10gauge and there are exceptions to even that. I have welded 16gauge sheet with pulsed mig to match the original welds and I have ac tig welded 1/2" al because that is what the customer speced out.
Take a look at an al semi tanker and you will see that they are fabricated almost entirely with mig.
Mig is faster and can not only look great aestheticlly but be just as sound as any other process. For me it makes me money faster and cheaper than tig in most cases.

By Lawrence (*****)

Date 03-22-2011 04:16 Edited 03-22-2011 04:18

GMAWP can be done with aluminum and pass PQR's for sure.. Excellent for fillet joints even down to .040 in thickness. Much higher travel speeds than GTAW and a much shorter learning curve for the welders. Miller 350P has tremendous bang for the buck with a push-pull setup... Lincoln Pulse on Pulse technology can produce excellent ripple patterns if cosmetics are an issue.. Both Miller and Lincoln have pretty nice Hot-start and crater fill programmable options as well as factory presets for both... and Fronius can run circles around both of the others for another couple thousand shekles. But if your simply doing thick aluminum fillets you don't really need GMAWP... Any good wire delivery system will run excellent CV GMAW aluminum welds that can pass PQR's.

As far as pulsing DCEN GTAW for aluminum (I would guess at high Hz) I can't say... But my opinion of the "stirring action" with GTAWP with high Hz on steels, stainless and super alloys is that it's just salesmenship. If you are married to GTAW I would think about an Asymmetric AC power supply and DCEN current for over 95% of the balence with a high blast of DCEP for 1-5% of the balence dwell time... It will run alot like DCEN but with a puff of cleaning action that will still allow for a sharp electrode prep... Helium/Argon Mix will work. May be some time spent on R&D with this method. It would really depend on alot of thickness/joint configuration issues.

By Zeek

Date 03-22-2011 14:44

Thanks for the feedback. I am certainly not limiting myself to TIG, but high quality is the first priority with the parts we are welding. These welds have to be acceptable per AWS D17.1 Class B, which limits us to 0.50T max subsurface porosity. Most of these weld joints are going to be a double sided fillet on t-joint (0.125" base material thickness). But, some of these joints were designed with dissimilar thicknesses such as 0.125" to 0.312", 0.5" to 0.312", 1.0" to 0.5". The dissimilar thickness design doesn't concern me too much, however the alloy we are using is 2219 which becomes more difficult to weld than 6061. The aesthetic appearance of the weld is not important to us as long as it meets the acceptance criteria. The subsurface quality is what I am most concerned with, as it's something we continually monitor and struggle to resolve. I've talked to Miller and Lincoln and both are not convinced that high quality can be achieved with aluminum MIG, but I was interested to see if anyone on here as had experience with it.

But I was also curious about the pulsing function on TIG, as we don't currently use it. We weld most of our thick joints with straight DCEN. We meet our tensile reqts, but the process is slow and subsurface quality is not always 100% perfect. We've thought about using a variable polarity process, but some of our joints are pretty thick. One groove weld in particular has a 1/8" root thickness, but with fill it ends up being over 7/8". With AC or VP, we assume that the heat input would be pretty great, especially on the last few passes and I'm concerned about the resulting strength. We've tried some testing in the past with a 75EN/25EP balance but it was difficult to get the focused penetration that we needed. We ended up having to increase our amperage almost 40% to get the same penetration. At that point, I become skeptical about maintaining proper levels of heat input. Another concern is arc rectification. We chemically etch all of our parts because cleanliness is extremely important. Even at a 90 or 95% balance, if there is not layer of oxide there, the arc becomes unstable.

I'm getting a little off topic here, but I guess one of my questions would be ...at what thickness does using AC create too much heat input? I would assume that industry experience would show that at some thickness level, you are better off using DC. Or is this not the case due to the ability to adjust the balance?

By Lawrence (*****)

Date 03-22-2011 16:30 Edited 03-22-2011 16:36

The ability to adjust balence *must* be a factor in calculating heat input when you are controlling it with the percentages involved in our conversation... I'm sure somebody has done the calcualtions.. Not me :) heh. As far as Too much heat input? Your PQR tensiles may be the only *real* way to judge if the heat input is reducing your 2219 below requirements. Edit: One last thought on this part.... With manual GTA welding of aluminum, in my opinion; operator travel speed has a greater bearing on total heat input than Polarity or Pulsation. The OSU grads may disagree; but I think a puddle is a puddle, if it's too big or too slow, your going to add heat into the work that will draw Hy into the soluable puddle and subsurface porosity is the result.

If the arc is becomming unstable an increase in AC frequency to around 200 Hz is pretty effective in keeping the arc on point... Any increase over 220 Hz is just going to trigger migranes, with no additional arc control in my opinion.

If Miller and Lincoln don't have confidence in there products than I'm at a loss... Are your welds very short? I really don't see why poroisity would be an issue that would cause them to hesitate in recommending GMAW for fillet thicknesses and inspection criteria you are describing.

Are you joining base metals that have been previously annodized? The last 15 years have seen the advent of new annodized surfaces that are quite different than in the past; That can open up some pretty serious subsurface porosity issues, even with surface prep techniques that were previously thought sufficient.

By Zeek

Date 03-22-2011 17:09

Thanks for the response Lawrence. Your replies are always filled with good info.

The weld assembly that we are currently discussing is pretty big and has lots of fillet welds. We are manually welding it with DCEN right now because the welders are most comfortable with that. I'd rather be using AC, but the welders need more practice to be able to use this process and produce acceptable welds. The subsurface quality always seems to vary when they manually weld with AC. We are trying to look using a robotic welding system for this assembly and have been looking at what our options are. The joints arent too complicated and the shortest weld is probably no less than 20". In terms of joint access, using the TIG process might present some challenges due to one weld joint we have. I would prefer to use MIG if we can get consistent high quality.

I've talked to both Miller and Lincoln and it's interesting that they both feel that MIG, most likely, will not produce the results we want. However, these parts will be machined and chemically etched prior to fitup to help maintain cleanliness. There is talk of possible alodining these parts so that if they sit around, they won't start to pit. But, regardless, they will still be chemically etched prior to fitup. That is one issue that I've heard mixed reviews about. Some say that when you alodine aluminum, the coating actually migrates into the material quite a bit further than people think and that the only true way to remove it is to machine the surface off. Have you heard this in your experience? I have assumed that a chemical etch that removes 0.0004"-0.0006" material would be enough, but haven't really done any testing to verify that the surrounding grain structure is free of the chemicals in alodine. Even though we chemically etch 1" on either side of the weld joint, it still would be good to know how deep this coating really goes. Is it possible that the heat from the weld would draw in some of the components from the surrounding alodine, even 1" away? If porosity is just coming from hydrogen, then the source has to be coming from the way we are cleaning the part.

Anyways, we are looking to to some weld testing with lincoln to see what sort of quality we can achieve, but I'm am somewhat curious as to why they feel that process is not capable of achieveing the higher quality results.

In regards to pulsing, my interest was to see if that had any effect of porosity formation. Our automated procedures right now just use straight DC, but I read that powerful short pulses have a better effect of boiling out contamination. I would assume that if you were to pulse, the amount of time on the low side of the pulse would have to be pretty short so that you aren't freezing the puddle too quickly. Does that sound reasonable, or is that too big of a generalization? If we can achieve better results with pusling, I'd imagine we could move faster as well.