Mig vs Tig for aluminium

[Carl Wilson]

Someone asked me a question today and I felt I didn't give a satisfactory answer. It was can you mig aluminium, and if you can, why tig? My answer was that tig is cleaner, more precise, produces a better weld with better fusion and the Ac gets through the oxide layer. That sort of thing.

So I realised I've no real idea about mig for aluminium. How does it work? Is it Ac too to clean off the oxide? How, given aluminiums specific heat and conductivity can you achieve speed but still get good penetration?

I'd be grateful if someone with in depth knowledge of the process could fill the gaps in my education.

Carl.

 

[carbon]

No in depth knowledge from me unfortunately, but occasionally mig weld aluminium. Usually in the 5 - 6mm range with a dc inverter, spray transfer approx 280 amps and electrode positive

 

[Hood]

As far as Mig is concerned with Alu then a standard Mig is for 4mm upwards, a pulse Mig can go way down, I have done 1mm.

Tig is, or should be, neater.
Regarding penetration/strength then if anything I would say possibly Mig has the edge as long as it is done right.


Mig welding Alu is, or should be, done in spray transfer mode hence the problems with thinner gauge material. There is a lot of heat involved but the start of Mig weld can be cold. Some welders have a Hot start function which allows you to use a higher current at the start to get the initial penetration/heat and then drops down to a more suitable current for the heated Alu for the rest of the weld.


With Tig it is the positive part of the AC that cleans the Alu.
With Mig it is DC positive so the cleaning is done no problem.

With Tig if you use too much positive or for that matter DC positive then there is a tendency for the tungsten to melt.
The reason you can use Mig DC positive is that the electrode is the filler so you want it to melt.

 

[Carl Wilson]

Thanks for that good explanation. I've done lots of aluminium tig but no mig.

 

[Hood]

I do both depending on the job. For heavier stuff it is Mig, for thinner stuff most likely Tig. Having said that if it is a big job on thin material that doesn't have to look the best it can then I will Mig now that I have the EWM Phoenix Pulse Mig as it is just so much faster and actually often not too ugly either. I sometimes do the opposite, I tig heavier stuff when I want it to look pretty but really 6mm is the max although I do on occasion weld 8mm.

 

[KemppiFrog]

Hood
Elsewhere, there was a thread on the construction of an ally 90ft fireboat. All the welding was mig, as far as I could see, quite impressive stuff, but thick plate, no 1mm there!

 

[Hood]

Yes, nearly all boats are done with Mig, would take way too long to do it with Tig. There will be some bits that are tig welded though. On the boats I fitted out at the beginning of the year it was a mixture of Mig and Tig but mainly the former.

 

[Richard.]

Both processes have there place and as hood rightly said Mig is best suited to thicker sections unless you have the facilitie to control heat input via a pulse.
Now I wouldn’t say Mig is a better penetration process as it really depends on lots of stuff. Amp per amp you’ll get better penetration and fusion out of a tig
Certainly for speed and production work a Mig runs rings round tig. Now people say for neatness it’s tig all the way and I would of agreed with this a good few years ago however these migs are so sophisticated and finely tuned nowadays you can achieve results very much as clean and neat as a tig weld but at speed.
Yes it’s positive polarity that dose the oxide layer removal so on a Mig there is no problem there as the electrode is the consumable it’s all positive.
In tig the electrode is not consumable and there for you need balance between pos for cleaning and neg for penetration so you don’t melt the tungsten.

 

[Carl Wilson]

Hmm. If mig is Dc then that will remove the oxide but you must need high current settings indeed to get the aluminium to melt and form a puddle. How can you get decent penetration at speed? I can't see how you could get enough heat in and quickly enough with DC without using high current settings.

For example. I've welded 6mm thick aluminium plate using tig at 150 amps and about 70% balance or thereabouts. It still takes time to melt and form a pool (though I often tack on higher settings to get some heat in). I can't imagine how you get any penetration with mig.

 

[dannyp]

if theres lots to do, mig. if its possible to mig it i'l try mig it unless it needs to be really pritty

 

[Hood]

Hmm. If mig is Dc then that will remove the oxide but you must need high current settings indeed to get the aluminium to melt and form a puddle. How can you get decent penetration at speed? I can't see how you could get enough heat in and quickly enough with DC without using high current settings.

For example. I've welded 6mm thick aluminium plate using tig at 150 amps and about 70% balance or thereabouts. It still takes time to melt and form a pool (though I often tack on higher settings to get some heat in). I can't imagine how you get any penetration with mig.

Yes current is fairly high, probably around the 190-200amp on 6mm although I can't be 100% sure as I have never really looked.

I have a couple of etches in this thread, also vids which are not very good but some have etches in them I think. Think I was using Double Pulse mostly which is slower than single pulse and puts less heat in than single pulse I suppose.

http://www.mig-welding.co.uk/forum/threads/1mm-v-1-2mm-alu-wire-on-ewm-phoenix-330-pulse.73454/

Also maybe some stuff in this thread as this was when I first got the EWM and was finding my way with it as I had never used pulse before.
http://www.mig-welding.co.uk/forum/threads/ewm-phoenix-330-expert.72053/

 

[Carl Wilson]

Thanks for that mate, highly informative. I don't doubt that it can be done because it is done. I just couldn't visualise it, never having done it myself.

 

[Richard.]

To a point*
Travel speed doesn’t play a part in penetration.
Going slower increases reinforcement and heat input.
Voltage increases your arc width and it’s wire speed that increase your current generated. Current is what gives you your penetration so rather than going slower you’ll increase penetration to a much more elaborate effect by turning up the wire speed. Now obviously there are Process limits. You can’t just whack up the wire speed without changing voltage to suit that speed.
* now I say to a point. There are obviously exceptions if you race down at a pace too fast for things to get a hold then in this case yes you’ll lack pen and fusion but going at a speed to generate the correct bead size will not effect it.
Going too slow makes your joint weaker not stronger and even more so with aluminium. The heat effected zone will be huge and let’s not forget aluminium (unlike steel) is at its most brittle when it’s hot. Steel behaves differently. High heat input with steel creates a low toughness joint but not brittle. Either way, either material going slow to generate penetration and fusion will not work as you see it.
Then you have the other problem of going slow increases reinforcement. You might need reinforcement due to the size of material and joint configuration and in this case a multi run is the best approach. If you don’t then excessively slapping in a big weld will act as a stress riser especially around the weld toes if they don’t blend nicely. Your parent material has limits and it can only take so much heat before it’s wrecked permanently. Travel speed has the biggest effect on heat input out of the 3 variables in the formula.
J=AxV/T.

 

[Carl Wilson]

Very interesting. How does mig compare to tig, specifically for aluminium in weakening of the parent material in the haz?

 

[Brad93]

Both processes have there place and as hood rightly said Mig is best suited to thicker sections unless you have the facilitie to control heat input via a pulse.
Now I wouldn’t say Mig is a better penetration process as it really depends on lots of stuff. Amp per amp you’ll get better penetration and fusion out of a tig
Certainly for speed and production work a Mig runs rings round tig. Now people say for neatness it’s tig all the way and I would of agreed with this a good few years ago however these migs are so sophisticated and finely tuned nowadays you can achieve results very much as clean and neat as a tig weld but at speed.
Yes it’s positive polarity that dose the oxide layer removal so on a Mig there is no problem there as the electrode is the consumable it’s all positive.
In tig the electrode is not consumable and there for you need balance between pos for cleaning and neg for penetration so you don’t melt the tungsten.

You’ll get more penetration with MIG at 200A than TIG on aluminium.

 

[Richard.]

At 150 you won’t.

 

[Richard.]

In fact I think you’ll be pushing it at 200 tbf. 230 plus yes you will.

 

[Hood]

Sounds like I need to do some testing

 

[Richard.]

It’s a difficult thing to test at the 200 amp mark because I believe it’s gonna be a close one. You’ve got options on both processes to take the win. Shifting balance well into electrode neg on a tig could be enough to take it as could dropping a volt on a Mig. My personal opinion is at that current level you can make either process take that. I think you need to go higher to get Mig into a winning streak and lower to see tig take things on every time. That is my view. I can do some tomorrow if I get chance as well to see what I get.

 

[Hood]

I think it would be a hard comparison myself due to the points you mention.

I actually do not have any idea what current I weld at with Mig, couple of reasons are that it is usually the voltage and the wire feed you adjust to get started, now obviously the wire feed determines (in part) the current but as that is normally displayed in m/min rather than Amps then it is not something I am used to.
Also the set current, on a machine that does display it, is just an indication, once you start welding then the current will likely be different.

With Tig I am usually 15-20% on the balance so quite far negative, also I tend to use a slow frequency compared to others, usually around about the 60-70Hz, that may also affect things. It certainly gets the heat into the Alu quicker than with a higher frequency at a given current, at least that is my observation anyway.

I will try and see if I can do some tests and see if I can keep the currents the same between processes.

One thing however, I would say, is you have a window for current in each process. Out with that current window it is either too cold or too hot so as a comparison I will also set my Mig and Tig to the settings I would normally do a given thickness at.
That may mean the Mig is higher Amps than the Tig but the point of that test will not be an comparison about which process gives the most penetration at a given current but rather which process gives the most penetration at the settings you would normally use to get a sound weld, or maybe a better name would be real world settings.

Hopefully I will get a chance tomorrow.