GTAW of Aluminum

[sn0border88]

This is the byproduct of me having way to much time on my hands, and noticing that about every week the same questions pop up. Hopefully this can shed some light about the mystery and general misconceptions about TIG welding of aluminum.

Polarity: AC current consists of a positive and negative portion, and alternates very rapidly between the two. This produces the "buzzing" sound that you hear, and what contributes to the nature of the AC arc. The AC balance control determines how much time is spend on either portion, a value higher on the negative side will giver better penetration and arc stability with decreased cleaning action, and vice-versa.

The positive portion of the wave (electrode positive) puts about 70% of the heat generated into the tungsten, this causes the tungsten to melt and beings to form a ball at the end. This portion of the wave is also what produces the cleaning action, in which ions bombard the surface of the material, removing some surface contaminates. DCEP can be used for welding of very thin aluminum but the tungsten will rapidly overheat.

The negative portion of the wave (electrode negative) puts about 70% of the heat generated into the workpiece, yielding much better penetration. This also enables the tungsten to better retain a point which makes the arc more stable and easier to control. This part does not offer any cleaning action. DCEN welding can be done on aluminum but it requires removal of all surface contaminates and generally done with helium shielding gas. (I dont know why, if any one can shed some light on this than im all ears)

Shielding Gas: 100% Argon is the most common, while mixtures of Ar and He can be used to achieve greater heat input. Argon is preferred because it is cheaper than He, very dense (better cleaning action) and produces a stable arc. Any mixture containing O2 or CO2 cannot be used because of the rapid formation of aluminum oxides, along with the rapid erosion of the tungsten.

Tungsten: The electrodes used on AC welding vary greatly with personal preference, but I do not know of any commonly found electrode that cannot be used. Some of the more popular ones include 2% ceriated, 2% thoriated, 1.5% lanthinated and pure tungsten. Pure tungsten is not required for use on any machine, in fact most inverter machines will not proper functionally with pure tungsten.

Forming a ball on the end of the electrode is not necessary, often it is preferred to weld with a slightly blunted point. If the ball grows too large and the arc seems unstable, stop welding and regrind the tungsten. If it continues, either decrease amperage or go to a larger tungsten diameter.

AC Controls: Older machines will likely have a simple balance control which was explained earlier. Newer inverter machines typically have an additional frequency control. This allows the user to adjust the amount of times (per second) that the arc switches from EP-EN. The standard is 60Hz, but increasing this ,typically >120Hz will stabilize the arc, making it "stiffer" and increasing penetration while reducing the width of the weld pool.

This is by no means the definitive guide, obviously I hardly even touched the surface. If you have something to add, do me a favor and just make it easy to read but feel free to add it in here. The goal is to make this a one-stop-shop for aluminum. If you disagree with something, PM me and we'll work it out.

That's all for now, more to come tomorrow.

[sn0border88]

Cleaning: Welding of aluminum is made difficult because of how rapidly it can form an oxide layer on its surface. While this is what makes it so resistant to corrosion, this layer must be removed prior to welding. This can occur in a number of ways, either by mechanical cleaning like a wirebrush or abrasive pad; chemical cleaning using a variety of industrial solvents or electrically by the bombardment of ions.

Stainless steel wirebrushes are a good choice, as they are both very effective and cheap. Care should be taken to prevent contamination of the brush (either from oil, dirt or rust), which could impart these contaminates onto the material and into the weld. Powered wire wheels should be avoided as these have a tendency to smear the material and further trap contaminates instead of removing them. Steel brushes should be avoided if possible because they can leave behind iron oxides (rust) on the material.

Wiping with a solvent such as acetone, denatured alcohol, or others is a good idea, but make sure that all solvent has been evaporated or removed from the weld joint before an arc is struck to avoid the possibility of dangerous fumes or fire. However these solvents will not remove the oxide layer, only excess oil or dirt.

Abrasive mediums like 3M pads, sanding or grinding wheels or stripper wheels work well for removing heavy contamination like is often found on castings or parts that have been in service. These will remove surface oxides as well, but can also smear the material if to much pressure is applied.

How clean is good enough? I would be lying if I said Ive never successfully welded material that had heavy oxides on the surface, or had some leftover dirt that got pulled into the weld. Ultimately the decision is yours to make, but keep in mind that the cleaner the material, the easier it will be to weld, lower chance that cracks will form and overall leads to a better quality weld.

[sn0border88]

Common problems with welding:

I cannot get a puddle to form: Possibly not enough amperage, aluminum requires more amps than steel does for the same thickness material. You might also be on DCEN instead of AC, this will generally create a "skin" over the puddle making it hard to distinguish from the base metal. Your balance might also need to be adjusted, as a value high on the EP side will reduce the heat into the weld pool and a value high on EN might not remove enough oxides to allow the puddle to be seen.

Its also possible that you have inadequate shielding gas coverage, this allows atmosphere to reach to weld pool and rapidly forms a thick oxide layer which can make welding impossible. Increase gas flow, or consider using a gas lens which will provide better coverage than a standard collet nozzle.

My tungsten keeps falling off: Amperage is to high, or your on DCEP instead of AC. Find a way to reduce the amps needed, switch to a larger tungsten, or increase the balance control more towards the EN side.

My filler keeps balling up: This is because the filler is kept to close to the arc, either switch to a larger diameter filler wire or keep it farther away from the arc. Balled up filler wire should be cut off, because it tends to oxidize and if added to the weld pool will make further welding difficult.

[DiabolicZ]

How clean is good enough? I would be lying if I said Ive never successfully welded material that had heavy oxides on the surface, or had some leftover dirt that got pulled into the weld. Ultimately the decision is yours to make, but keep in mind that the cleaner the material, the easier it will be to weld, lower chance that cracks will form and overall leads to a better quality weld.

I have a question regarding AC balance control and clean aluminum.

1-10 scale. 10 being super clean stainless brushed and acetone wiped aluminum.

If you have clean aluminum (10) how far toward EN can you set your AC balance before you typically START to see the black crust form? More toward the EP setting is for cleaning action right? So what if your metal is almost perfect? I say almost because nothing is ever perfect. Can you set your balance control ALL THE WAY to the the EN side and still get a great looking weld with no black stuff?

Some machines can go as high as 95% toward DCEN correct? Is that even useful when you say that at 70% DCEN there is no longer any cleaning available from the DCEP wave?

[sn0border88]

You shouldnt see any black contaminates if the material is *near* perfectly clean, only time you would see that is if it was not clean to start with.

If im working with new material, I typically run 85-90% EN. A quick swipe of a wirebrush after everything is tacked up is all I would ever do in that case, and you wont see any specs in the weld either.

I think you misunderstood what I said about 70%. This is not a point where the EP portion becomes ineffective, as70% EN is a pretty common setting giving good penetration and tungsten life without sacrificing cleaning action. Dont make the mistake of thinking that you can run 95% EP and just blast through whatever you want; regardless of your balance the material has to be fairly free to start with.

What I was referring too was that in DCEN welding, 70% of the heat generated is put into the workpiece. Opposite for DCEP welding, which is why the tungsten overheats so rapidly on that polarity.

[buster1]

Snowboarder88,

Thanks for a valuable topic. After wading through all the advertising hyperbole and half-truths, you've cleared up a lot. Short question - I've been trying to draw the line (like I initially thought DiabolicZ was getting ready to ask about) on the polarity balance features between different models. As a practical consideration (3/8" aluminum or less) and assuming money is somewhat a factor, based on what you know looking back, would the price difference between a 70-30% polarity balance machine and an 85-15% range be something you wish you would have sprung for? I'm not looking for any kind of absolute standard, but I'm trying to make one purchase that covers most jobs, even if I have to burn extra time and energy getting all the impurities of the surface. Any thought as to what else you might be giving up with the cheaper model?

[sn0border88]

I think I know what you're getting at, and its a valid question, however I dont think I can answer it well. In my experience, knowing how to adjust the AC balance to suit the task at hand is a very important part of welding. But these adjustments need to take into consideration many factors, such as tungsten, waveform and more importantly machine type.

Inverter machines seem to benefit more from being able to run in the 80-90% DCEN range than transformers because by design they tend to ball the tungsten less. Transformers dont seem to run much different from 70-85%, usually if you push them too far with blended tungstens you will start to see splitting or cracking. 70-75% seems to be the max where you can keep a good ball formed.

Different tungstens also react differently to changes in balance, thoriated tends to be the most prone to splitting, followed by ceriated and then lanthinated. With that being said, I tend to always use ceriated as it seems to hold its shape best on my machines, but im sure others would be quick to disagree.

So is it worth it do pay up for a machine that is capable of 85% EN? If the choice was between a transformer with 70% and an inverter with 85%, no question go with the 85% machine. If both machines were transformers, then I would be hard pressed to convince myself to pay up for that little extra.

[waynep10]

Excellent info here. Just perfect for a welding forum. Thank you for sharing your knowledge with the rest of us. This is virgin territory for most and exactly why we are here usually. This is one of the best posts I have EVER witnessed so please accept my thanks this is VERY unselfish on your part.

[Rizz01]

Boarder:
My helmets off to you.
Nice work providing that editorial for those less knowledgeable on this subject.
You rock, Dude!

[backhoe]

I have spool gun for my miller 212 for alu.no problems there and have tig gun powered by my ac/dc stick machine for mild steel, no problems there, was wondering if possible to do alu. on the a/c side of my stick machine? prolly a dumb ? but Idon't know and appreciate any input good and bad thank's alot and keep your rod's dry.

[sn0border88]

Possible yes, but not as is. You would need a HF module that produces a high-freq current inline with your power source. As the AC wave switches between EN and EP there is a brief transition point that involves very low voltages, making it very hard for the arc to keep bridging the gap between the tungsten and the work. What the HF box does it provides a low amp - high volt current that allows the arc to stay lit through this transition period.

Some machines have refined the wave to the point that they no longer require continuous HF current, only for arc initiation. But your stick welder is not one of them

I would advise against it, and save your money and headaches for a dedicated TIG machine.

[Mu-T]

I have seen the 2% thoriated tungsten to "spit" and show up in x-rays. 2% zirconium tungsten did not have this issue and it is what I use on all aluminum projects I do.

[woi2ld]

[ame]http://www.youtube.com/watch?v=vWNZioJ_FOc[/ame]

[curb guy]

Thank you ,sn0border, this will be a much needed and usefull thread!

[SuperArc]

When your screwed up like me, MIG IT!!!!! Spool gun ready!!!!!

[shovelon]

Do I assume you are spamming a sticky?

[WeldorWes]

Do I assume you are spamming a sticky?

sure looks that way
to bad its still quoted

[zapster]

Do I assume you are spamming a sticky?

Not any more.

...zap!

[soutthpaw]

Just thought I would add that there really is really little use for a machine that has a balance range. <35% or so as it will just put so much heat into the tungsten that it just melts away. Balance is not a substitute for proper cleaning.

[Alan Murfee]

I have always found aluminium wielding to be the trickiest due to the emergence of the electrode balls at the end. Next time I would definitely try with increased tungsten diameter. In addition to it, the presence and early appearance of oxides leads to a rather tedious task of cleaning which makes the work more time consuming.

[Hickory]

I have an old Marquette, pretty simple, but what are the gap points in the top for.

[nmaineron]

Great info here,I am trying to connect a Hi Freq box to my Lincoln 225/125 AC/DC welding Machine and I am having trouble with what to do with the voltage sensing connection.I don't know if I have a transformer or inverter type machine.The manual says that if it is a transformer machine thaen I should connect the sensing wire to a 115v fused connection within the welder and the fan connection being the desired spot,however the wiring scheme simply shows the sensing wire being connected to the work side or negative side of the welding machine.Can you clear this up for me? Thanks, Ron

[nmaineron]

Well,I have determined that I don't have a transformer in my machine so I connected the sensing wire to my work piece and Viola, she works I couldn't do any real welding because I thought that I had some Alum Tig wire and I don't so I will have to wait a couple of days.

something I did try and was quite surprised with is that I used an Alum Flux dc rod ,4043, on DC reverse polarity with my tig torch and actually got a pretty decent bead.I was surprised that the flux didn't contaminate my puddle and the flux chipped off easily.

I will have to do some experimenting.Ron

[Donald Branscom]

Common problems with welding:

I cannot get a puddle to form: Possibly not enough amperage, aluminum requires more amps than steel does for the same thickness material. You might also be on DCEN instead of AC, this will generally create a "skin" over the puddle making it hard to distinguish from the base metal. Your balance might also need to be adjusted, as a value high on the EP side will reduce the heat into the weld pool and a value high on EN might not remove enough oxides to allow the puddle to be seen.

Its also possible that you have inadequate shielding gas coverage, this allows atmosphere to reach to weld pool and rapidly forms a thick oxide layer which can make welding impossible. Increase gas flow, or consider using a gas lens which will provide better coverage than a standard collet nozzle.

My tungsten keeps falling off: Amperage is to high, or your on DCEP instead of AC. Find a way to reduce the amps needed, switch to a larger tungsten, or increase the balance control more towards the EN side.

My filler keeps balling up: This is because the filler is kept to close to the arc, either switch to a larger diameter filler wire or keep it farther away from the arc. Balled up filler wire should be cut off, because it tends to oxidize and if added to the weld pool will make further welding difficult.

What you are leaving out is the SIZE of the weldment.

The greater the surface area ,the more heat is needed.

[nmaineron]

I attempted to weld my transmission case with the HI FREQ box connected to my AC/DC machine.I had no problem as far as function however I had big problems with cracking after welding.I preheated the case with a propane torch a few times for cleaning and to expose any cracks.Cracks were identified and welded successfully.I had two spots where the case `was worn through.I managed to weld these spots up by building up the edges and had good solid cover.The cracks started after the the holes were filled.Initially they ran parallel with the weld but outside of the fillet.I tried to gouge the crack and fill it but after I did that, new cracks appeared and ran perpendicular and were very pronounced,so much so that I feel that there is too much to repair. i would appreciate opinions.Thanks,Ron