TIG question on Al

[HWooldridge]

I didn't want to hijack the other thread so will pose this question here to the local TIG gurus. I have used my TIG on stainless, titanium and mild steel with great success but never welded aluminum. Last week at work, we found an aluminum casting on one of the production lathes that was broken so they asked me to fix it. This is a hollow housing with approximately 3/16 walls and there were three pieces knocked out of one side but only two pieces were found. I burned the oil out of the housing and the pieces with an O/A torch, then sand blasted everything to remove the residue. Parts were very clean when I went to welding.

Since I had no idea of settings, I consulted the Miller book and based on that, I used a 3/32 electrode, on AC, continuous hi-frequency, at about 125 amps and 15 cfh of argon with a #6 gas cup (I have a 250 amp Synchrowave AC/DC machine). Unfortunately, all I have is 2% Thorium electrodes and the book said pure Tungsten or 2% Zirconium is better. However, I still got a nice ball on the end and the arc seemed normal. The problem is that the sides of the weld would not close up even when stirring with the filler rod and there were lots of little pin holes and black speckles in the weld. In addition, the material flowed to the inside of the casting so I got ugly blobs under the weld. I suppose a backup would have been better but almost impossible to put inside a rough casting and make contact over the whole surface. I also believe that I should have been able to weld 3/16 without support as that is pretty substantial material.

I finally slathered enough weld metal on top of the breaks that a little mound was built up and it did not break when I ground it down so maybe the weld took. Unfortunately, I still have to add a blank to replace the piece that was never located and I am dreading trying to weld that.

So here are my questions:
1. Do I really need pure Tungsten or 2% Z - or - can I use the 2% Th with AC?
2. Is there anything else obvious that I could try to get better welds? It took way longer than it should and made a mess plus I know TIG does a better job than this.

[yesteryearforge]

I have heard that the naked eye has never seen clean alum.
used castings are noturious for not being able to clean good enough and the sand blasting probally was not a good idea as it more than likely impregnated the alum with sand.
But I must confess ( I am not a welder )

Mike Tanner

[arftist]

None of your problems sound unusual for cast Al. Aluminum is difficult to weld at best, and casting alloys contain many impurities. Start with a new stainless steel wire brush. Brush the work severely, imediately before welding, and between passes. Only use pure tungsten, and check the alloy of your filler rod. Be very careful grinding Al, unless you have a disc made for Al. The main problem with welding Al is aluminum oxide, which has a much higher melting point than Al. I think your amp setting is rather low, but I wasn't there either. Some cast weld well, some do not. Remember that Al is a very good conductor of heat.

[Hillbillysmith]

You're doing most of everything right although I can make a few suggestions. You can use 2% thorium tungsten on aluminum with no problems at all. I know, I've done this many times on actual work for people and it welds just fine but I would use an 1/8th inch electrode and sharpen it to a very fine point. The heat from welding will ball the electrode for you. The cup size is ok but I would turn up the gas to 20-25 CFH.

I have a couple of questions about the amperage settings though; are you using some type of rheostat? (ie; foot pedal, slide control/roller on the torch?) If you are, and I personally prefer a foot pedal, turn the amperage up to about 175-200. If you have an adjustment for the AC sign wave (max clean/max penetration) turn it to about 25% of max penetration (ex; if the knob on the machine is numbered 1-10, 1 being max clean and 10 being max penetration, set it about 2-3 and make sure you get it hot enough. One trick that I've learned and tell people all the time is let the area that you're welding get to a pool and wait until it gets to a "liquid mirror" look and then just shove the filler in. You don't just dip the filler like you do stainless. Just blob it in there (don't get the weld TOO big though). Cast aluminum is some nasty $*it to work with. Use 4043 aluminum for the filler material. While welding aluminum, the puddle will not just "snap" together like steel/stainless steel will. You just need to get the first ripple and then the rest should just move right along once you get the hang of it.

Btw, SOME aluminum CAN NOT be welded. If you know or can find out the grade of aluminum the casting is, I can see if it's weldable or not. I have a bunch of books on aluminum (roughly a dozen of my own) that I can reference for you.

And remember: practice, Practice, PRACTICE makes perfect. :D

Edited July 7, 2008 by Hillbillysmith

[jeremy k]

I weld lots of castings and I only use 2% thoriated tungsten (yes shame on me for not using pure - only been doing this 20+ yrs). Castings can be a problem - use the lowest settings to still do the job to help with porosity - still it can be a big problem to get a sound weld on something that acts like your trying to weld a sponge. all I can say is you don't learn this overnight - and it takes awhile to get comfortable with castings of aluminum. - motor cases are bad because of the oil impregnation. Take your time and don't rush it by putting to much heat into it - like i said use a lower amperage to slowly do your weld - those pin hole areas are small gas pockets from casting (correct me if I'm wrong) and a high amperage just makes it harder to fill them. - Don't know what else to say but practice and more practice. Most general castings of aluminum are best welded with 4043 from what I've found - as it gives a good liquid flow of the weld - JK

Edited July 7, 2008 by jeremy k

[Hillbillysmith]

Jeremy k; you are right all the way, but even with your years of experience I am going to have to disagree about the high amperage harder to get rid of porosity. I have found that as long as you keep the puddle liquid until it gets a smooth silver finish.

[HWooldridge]

I am using a foot controller and the panel is set at 125. I tried a 1/8 electrode and burned it away on the first pass so went to the 3/32. I also tried higher amperage but the material started collapsing so I backed off on the current. I have the sine wave knob at 3 - I went down to 2 and up to 5 but 3 seemed the most stable. I agree that my gas flow is probably low and I may need a larger gas cup.

I did get the "silver puddle" but the thing that seemed odd to me was that the sides of the joint pushed apart (sort of like Moses parting the Red Sea). I'm used to steel and titanium wanting to cross the weld root with a little extra rod but the aluminum didn't want to cooperate. Another thing I did which probably wasn't code was to run the arc all around the weld and melt the lumps a little more. I need to get a refill on an O2 bottle anyway so will pick up some 4043 rod at the welding store tomorrow.

This housing is off a Japanese Miyano CNC lathe and the dealer wanted $4000 for a replacement - hence the request to have me fix it for the few hundred bucks I'll charge them. I'm quite sure this thing is oil soaked because that's exactly what it does - an oil catcher for the back of the spindle where the bar enters.

Thanks for the suggestions - I'll let you know how it turns out.

[arftist]

Reading your responce gave me some more insight or posibly another idea. One thing that makes me wonder, is after you burned away the 1/8 electrode, how did changing to a smaller electrode help? Aside form that, Al welds differently from other metals. Al's rate of thermal conductivity is second only to copper. This is why so much more heat is required to establish a puddle. However, as the welding progesses, less heat is required to maintain said puddle. I have to second hillbillysmiths suggestion that you use a higher heat setting at the machine. Stomp the petal to get the puddle started, then back off the heat to just enough to maintain the puddle. As the work heats up you will need less pedal. The phenomenon of the side of the weld pulling away is usualy caused by the fact that the melting temp of aluminum oxide(which continuously forms on the surface)has a MUCH higher melting temp than Al, so by the time you burn through the oxide, the Al under the oxide has melted away. Hence the need to vigorously wire brush the work with a new s.s wire brush. Also, if possible, practice some on some Al srap, just to get a feel for the heat input needed for fusion. Lastly, it is perfectly ok to weld over your previous beads and melt the lumps together. But then I've only been welding Al for 30 years so take it with a grain of salt.

[Jose Gomez]

Though you can get away with 2% Thoriated tungsten it is not normally recommended. This is due to the fact that Thoriated tungsten will crack when subjected to use with alternating current. These cracks (sometimes microscopic sometimes quite large) can cause several small balls to form on the end of the electrode instead of one large sphere. These small balls and or micro fractures on the end of the electrode can cause the arc to be unfocused, thereby causing the arc to wander and jump from edge of joint to edge of joint requiring slightly higher amperages to gain stability. It is for this reason that most welding equipment manufacturers recommend the use of pure, zirconiated, or ceriated tungsten for any AC application. Pure tungsten has a low melting temp and though approved for AC welding has a tendency to form a small molten ball on the end of the electrode that can sometimes drop in to the weld metal causing what is called a tungsten inclusion (which like a chip in a wind shield can eventually lead to cracking of the weld), and it can not be used for DC welding. Zirconiated tungsten is used to apply high quality (x-ray quality) welds on aluminum and Magnesium. This is because the addition of zirconium stabilizes the arc and raises the melting temp of the tungsten to help reduce the chances of a dreaded inclusion showing up in the X-ray of a weld that you just spent two days working on. Now on to ceriated (My general recommendation for anyone with a TIG machine). Ceriated tungsten is one of the only tungsten recommended for use on both AC and DC. It is for this reason that I suggest it. Instead of having to wonder if you have retrieved the right type of tungsten (after the paint has worn, ground, or been burned of the end) you know that if all you have is ceriated then you can do no wrong. Another point of interest is that Thoriated tungsten is slightly radioactive (look up thorium on the periodic table of elements). As far as the rest of the procedure you are following, all sounds fairly normal. One rule of thumb is to use 1 amp of power for each decimal of base material thickness. For example, if you are working on 1/8 inch that is .125 so 125 amps is a good starting point, or 1/4 inch .250 inch 250 amps. I know it sounds high but the extra power is needed in order to establish the puddle due to aluminums ability to quickly absorb heat (high thermal conductivity). One thing to remember, though, is that once the puddle has formed and the work piece has reached welding temperature you will have to reduce your amperage significantly (let off of the petal) in order to keep the puddle from falling through or becoming far to wide (sometimes called hot short). When you get it right you will find that you have the puddle kind of balanced on the tip of your toe, foot goes 1/4 inch down puddle grows, foot comes up 1/4 inch puddle freezes. When you reach this point you are driving the puddle, not the other way around. It is here that you get to take your time and put the metal where you want it, drop by drop. I would definitely recommend a much larger cup (around a 9) to ensure adequate weld puddle coverage. From everything else it sounds like you are on the right track. Clean Clean and Clean some more be patient and gently work over any spots where you can see porosity as you are welding along and you can normally get them to close up. If not Grind or otherwise remove the contaminant causing the bubbles (using aluminum approved abrasives or grinder wheels) and make another pass. It takes a lot of patience. Good luck

[HWooldridge]

Arftist,

I made a typo mistake in my second post. I meant to say that I had used a 1/16 electrode on the first attempt with a 3/32 on the next try.

Thanks again to all and I'll post results...Hollis

[arftist]

Jose, thankyou for your exellent explantion of tungsten selection.

[Rutterbush]

Everyone has great input about welding aluminum. I'm gonna stir the cooking pot just a little more. Can never have too many chefs.

Grinding aluminum and power brushing aluminum look as if it gets it super clean just like steel, but there's a catch. As soon as the grinding disc or wire wheel clean the surface the oxide rapidly reforms faster and slightly thicker than normal due to the air turbulence created by the rotating wheel. Only a hand held wire brush should be used. The brush should be pushed or pulled in only one direction just like using a file. The reason being not to drag particles of oxides back into the weld area. This information was passed to me by two of AlcoTecs field technicians who helped me when I started the boat propeller repair business. Called them up and they drove down to my shop and worked closely with me welding cast aluminum propellers.

Castings are full of hydrocarbons in the porous areas between the grain. Just like any gases they tend to expand when heated. Doesn't matter how clean you think you got it. It has oxide and hydrocarbon working against you. A crack that can't be hand brushed or filed is a nasty waiting to happen. I understand that what you're working with is not a crack,

One reason that your weld bead won't fuse with the metal is as stated earlier, there's oxide or hydrocarbon built up on the edge. Sorta like cleaning the weld area then painting over it to get ready to weld. Ain't gonna happen.

The idea is to clean it so that oxygen will hopefully not mix too rapidly, hand brush or file.

Use as large a cup as possible and have a low steady flow of pure argon, a small cup and high pressure will start a venturi affect and suck air into the weld area. a #8 cup and 25 to 35 cfh.

Keep the end of the tungsten a maximum of 1/8" above the puddle. Doesn't matter if it's balled or pointy (technical term).

Yes, you do need the higher starting heat, but back off asap or preheat and weld. Weld with the lowest weldable heat. Just like any other metal too much heat does strange things to the grain.

Another thought is since aluminum is red short (won't support itself at high heat) use a piece of stainless steel to support the weld area. Keeps it from falling through, keeps it flat on the back side and won't weld together. Don't use copper. Copper particles will alloy with the aluminum. Not good.

One final note. Never pull the filler wire out of the argon shielding gas. It only takes a split second to oxidize the hot end of the wire, then put it back into the puddle.

I hope you've got it finished by now and can tell me how full of it I am.

Of course, I'd mig weld that sucker and be done with it. ;0)

Edited July 7, 2008 by Rutterbush

[Tigman]

Might want to try 75 % Helium and 25 % argon for welding on thick nasty castings! 20 years welding cylinder heads and intake manifolds in winston cup shops.

[arftist]

Tigman, I've used pure helium, care to elaborate on the properties/benefits of the 75/25 mix? Thankyou.

[HWooldridge]

I bought some 4043 rod, a couple of 3/32" pure tungsten electrodes and a 1/2" gas lens then cranked the argon flow to 25 CFH. The welding got much easier but was still pockmarked by areas that simply would not weld together. I assume this trouble was because this repair was a casting and likely oil soaked. However, by stopping to sandblast periodically, I was able to finish the job. I also welded a piece of material across the ends of the broken fragments so they are all joined by a homogenous piece. After milling away the excess, the piece looked fresh from the factory. The maintenance guys at the plant were suitably impressed (especially since none of them were willing to fix it).

Thanks to all who answered, this forum is a real blessing to all of us who participate and contribute.

Edited July 14, 2008 by HWooldridge