why AC for aluminum?

The DC Electrode Negative arc has no cleaning ability since the energy is traveling from the tungsten to the metal.

This can be done, and is used for welding heavy aluminum plate, but you need a helium shielding gas and pristine aluminum. The penetration is amazing. You have to shove filler metal in as fast as possible. The weld looks like crap, but if you grind off the top, it will be sound underneath. I have done it. It is not a delicate proceedure.

It is nearly impossible to remove all the aluminum oxide from the surface of aluminum. If there is oxygen present with the aluminum, then there is aluminum oxide on the surface.

Try it and you will understand why it is not used much. Only for really heavy plate.

My understanding is that the aluminium oxidises instantly. Maybe it could be welded with DCEN if you would grind it in an Argon atmosphere...

Apparently, it does. If you try it with a regular tig torch (say, because you forgot to switch the welder to AC...), you will see that the puddle looks like it has a "skin" that holds it together like a bag and prevents the rod to wet it. I believe that this "skin" is actually aluminium oxide.

Yes, aluminum oxidizes almost instantaneously in a normal atmosphere. It is one of the most reactive metals, i.e. it forms oxides very readily. As others have mentioned, you can actually see the oxide surface sputter off when welding with AC. But there is an important principle behind this. The aluminum oxide on the surface has a melting temperature of about 3700 degrees F while the aluminum base metal has a melting temperature of about 1200 degrees F.

Most metals will form a thin oxide surface fairly quickly but this is not as much a problem with some of them. For example for steel (iron), the base metal melts around 2800 degrees F while the oxides (there is more than one type of oxide) melts around 2900 degrees F, i.e. there is not much of a difference. Yes, you want to remove "heavy" oxides (e.g. rust, which can be combination of oxides and/or hydrated oxides) but this has more to do with potential contamination problems and arc stability problems (the oxides are less conductive and prevent smooth running of the arc) than physically making a molten weld pool. For aluminum, in addition to the oxide "contamination" of the weld pool, part of the problem in establishing the molten weld pool is the large difference in the melting temperature of the oxide versus the base metal.

Note that there are also other important physical effects which make aluminum more difficult to weld, specifically very high thermal conductivity, high specific heat and low melting temperature. This combination means that it is hard to heat up the base metal (because of the high thermal conductivity and high specific heat) but once it does heat up, you are quickly at the molten stage (due to the low melting temperature). For us unpracticed welders, that is why we end up going from no molten pool to suddenly finding a big hole. :-) The surface oxide problem on aluminum makes it just that much more difficult because if the oxide is not removed (by the sputtering action of AC) then the surface does not melt easily and greatly inhibits "wetting" of the surface my the filler metal.

I apologize if this turned out to be more than you wanted to know.

Unless aluminum is cleaned and maintained in an inert gas environment, it will have an oxide layer. Aluminum is extremely reactive with oxygen, even at room temperature, and will almost instantly form an aluminum oxide surface when exposed to air. There is no such thing as a *clean* aluminum surface in air.

It can't, but it will already be oxidized from its exposure to air before you start welding.

No. The agitation of the welding puddle by the arc will ensure that the oxide is mixed into the joint. Since aluminum oxide has a *much* higher melting point than pure aluminum, these solid inclusions will weaken the joint by preventing it from properly wetting out. If you are doing deep penetration welding of thick aluminum, this won't be too much of a problem, but for normal thicknesses, it is.

Gary

Thanks to those who have posted responses -- so my next question is, bearing in mind all the issues already raised, why is it possible then to oxy-weld aluminum?

Is it the agressive flux that disperses the aluminum oxide as it forms?

Could you use the same flux with DC TIG to the same end?

-- you can contact me via

Without the flux it is impossible.

I suppose, but the flux makes the aluminum extremely fluid. I feel the TIG arc heat would make it so liquid it would be unmanageable