Aluminium tig guide getting started.

  1. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    As has been requested by a small number of members I'm happy to start a small guide as a starting point for newcomers into welding aluminium. The idea is to simply give you a starting point so your new toy can be taken from the box and dialed in without the grief of guess work or diving through countless threads with lots of compromising opinions and info.

    There are literally thousands of combinations as to what can be done to weld it I'm just going to talk about the meer basics and after that it's up to you.

    I'll do it over time and generate a few posts and get some pics in it to help illustrate.
    Alu tig is welded in AC and
    You require an AC DC tig machine to do this.
    AC is a fluctuation on polarity which can confuse because the output is always DC but the fluctuation is between DC+ and DC- . This is the noise you here when you strike it up. Now all tig welding requires the torch lead to be connected into the negative terminal so that the heat concentration is in the work piece and AC is no different except the current also passes into the positive side too.
    The reason. Alu has a thin oxide layer that will self regenerate and is never completely removed. The spike from the current passing into the positive polarity allows this oxide layer to be moved out of the way so that the bear Alu under is exposed for welding. This is the reason why it is known as "cleaning(DC+)" and "penitration(DC-)"
    Now this information is best understood in the form of a wave. It has nothing to do with pulse welding despite a wave used to illustrate that too as it isn't a variation on current but a variation on time spent in a polarity.
    What can we do to that wave?
    Well modern machines allow us to do just about anything to it including even changing its shape but from a beginning point of view there is usually 2 key things we can do to it.
    1) balance. Can be also known as cleaning on some machines.
    This allows you to set where you want the polarity biased.
    2) frequency (again no link to pulse) it's AC frequency and this allows you to set the amount of times that fluctuation happens over a period of time and is usually set in Hz/second. So 1 full hz is a cycle of up in the positive and down in the negative.
    This is all for now. Lots of info here so far which is an introduction to what it is and why we use it. I'll update further on it soon.
     
    • image.jpeg
    stuvy, indy4x, 123hotchef and 26 others like this.
  2. Very good idea. Info covering the basics seems hard to find. Why not a few YT vids to compliment your text?
     
  3. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    Ok ive had a look through the phone and ive enough photos to explain part 2. Balance.
    I prefer to call it compromise because that's how I see it. In a perfect world we would have lovely focused tungstens, gleaming clean welds with substantial penetration on every joint and material thickness but let's face it it's not a perfect world and some sacrifices need to be made. Balance is exactly that. If we biased our setting right into the cleaning (DC+) we would have extremely good cleaning going on, the oxide layer would be kicked right out of the equation and our welds would represent mirrors. Unfortunately DC+ is heat focused on electrode which combined with current will cause the electrode to heavily melt and cause an unstable arc as that melted electrode changes shape. The pics below show 3 tungstens all welded at 100 amps.
    1 is at 70% dc- on the balance
    2 is at 60% dc- on the balance
    3 is at 50% dc- on the balance.
    The first two are quite acceptable depending on the joint type which will be covered later but the last pic it's gone too far and is out of shape. I tried to weld with it and struggled due to the unstable arc it was trying to focus.
    Next step will be about the sacrifice based between cleaning and penetration. Which is also effected by the balance control. So you have more to compromise than just your tungsten prep.
     
    • image.jpg
    • image.jpg
    • image.jpg
    Richiew, baldy, indy4x and 7 others like this.
  4. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    So we have discussed briefly how your tungsten is effected by the balance control on your lovely new welder.
    What about the welding. Well you'll be pleased to know cleaning and pen is exactly that. You will get cleaner welds with lower penetration if the balance is biased deep into the dc+ half of the wave.
    You will get dirtier but deeper penetrating welds if the balance is biased deep into the DC- half of the wave.
    The trick is to find a compromise.
    You need just enough cleaning to produce a contamination free weld and no more. On clean Alu this is anywhere between 65-75% DC- or 25-35% DC+. This is what I would call a good compromise between both features.
    Below are 2 pics showing the balance heavily biased at both ends of the scale but not too far that I couldn't produce some sort of weld.
    Now the pic of the beads on plate might look nice and shiney but the penetration was lacking and you can see that by the raised profile. Also my tungsten was on the edge with those and arc was getting a little wondery. All 3 runs were done at 55% DC-
    The fillet however was done at 90% DC- and it is a proper scabby weld. Really contaminated with bits of oxide layer floating about in the pool. While my tungsten point was rock solid and my heat getting in was sufficent the weld doesn't look the part.
    I didn't do a fillet run to illustrate high DC+ as I was struggling with the tungsten stability on the flat with that huge ball.
    Note the White bands around the welds with the heavy cleaning. This is known as the etched zone. It's simply the oxide layer that's been pushed out of the way by the dc+ half wave.
    Note the lack of etched zone on my scabby fillet that was produce with very little cleaning at all.
    I hope this explains balance ok and gives you an idea of where it needs to be set.
     
    • image.jpg
    • image.jpg
    Richiew, baldy, TomCGB and 8 others like this.
  5. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    So we are in a position to cover a bit of frequency. Remember this is ac frequency and not the same as pulse. We know what it is with regards to the ac wave but the question remains how does speeding up the fluctuations of the ac wave bouncing between both polarities effect our weld?
    Arc concentration. When you turn your frequency right up to high Hz 150 plus the first thing you'll notice up on strike up is the much higher pitch noise the AC wave is generating the next thing you'll notice is how much that arc width has tightened up. It's more focused in a small surface area and is localising your pool which offers a narrower and deeper penetration also.
    Now slow it down. 60-70hz. The noise is deeper the arc becomes wider the pool more spread and the penetration is more widely spread below the surface but not as deep.
    This is what it does and the reason it's a useful feature is for setting your arc width to your joint. A fillet joint is the most classic joint for problems caused by the arc not been where you want it. The arc will always favour a side and its generally the vertical plate because that is the one with the exposed edge with no heat sink like a bench sat under it. So we can get round that issue to a degree by turning the ac freq up nice and high, concentrating that arc getting all the focus right in the joint corner.
    Opposite end of the scale is a thick section butt joint. You want your pool span quite wide to nicely cover both edges of the thick plate . A narrow arc would in this case offer the same problem as a wide arc on a fillet and try to favour one side of your butt joint. In this case you turn your freq hz right down to widen that arc and get your pool covering both sides of the joint.
    Quite old pics below you may have seen before
    Pic 1 fillet (120-150hz)
    Pic 2 butt joint (60-80hz).

    Now this isn't a rule book and material thickness plays a part too. A 1mm thick butt joint does not need a wide arc so the freq can be turned up higher in this case to produce a smaller more suitable size weld for the material.
    The last pic is an outside corner that I did ages ago. This would be an inbetween range as you want focus but a little spread to cover the joint too especially if it's thicker gage.(100-120hz) perhaps.
    These figures are very rough and the idea is more to explain what the function does and how best to use it. In a nutshell this is ac frequency.
     
    • image.jpg
    • image.jpg
    • image.jpg
    Richiew, baldy, TomCGB and 7 others like this.
  6. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    The idea of this thread is to just cover the AC. Every thing else to do with GTAW is heavily covered by Paul in the tutorials. This is a basic guide for beginners to plug in and understand what they are changing with AC welding. I have a few more points to add
    Down slope. This is more critical with Alu imo. If you come to the end of your run and drop straight off the power as the pool freezes you will every time watch it drop a crater right on the end. Not only does it look awful but it's very likely the first point that will crack if the joint is stressed.
    Setting a down slope for a couple of seconds allows you to get your final dab of reinforcement into the pool before the lights go out. Then as the current is phasing out simply and slowly start going backwards over the end of where you've just welded. This is a back step and the consistent movement of the torch along with the slow phase down of current will slow the solidification process to be much slower smoother and controlled thus avoiding that crator.
    Current. It's discussed a lot and it's too variable particularly with aluminium to give even close to accurate info.
    Cold aluminium requires a lot of heat to do its first two jobs of heat soaking the area and punching through the oxide layer. It's very roughly 40 amps per MM thickness. Once that part is well and truly heat soaked that current level becomes way too hot and can be decreased in some cases by half.
    The part size plays a big role in current setting as smaller parts get heat soaked quick as a flash and bigger ones suck it away taking lots more current to get going. Joints make a huge difference too an outside corner has two plate edges of surface area only and they get cut away if your not quick with fill and travel so a reduction in current would be useful here.
    A fillet weld is an enclosed joint with only 1 exposed plate edge. The flat plate on the bottom can cope with a lot more heat and more current is needed to get that pool going between both sides.
    Then we have the butt joint which is between the two.
    2 plate edges like an outside corner but not as exposed as they are butted together. Middle of the road with the current on the butt joint in most cases.
    Filler sizes.
    Very rarely with Alu should anything less than a 1.6mm rod be needed. It's nothing like DC in the respect that when its flowing nicely, a 2.4 filler rod can easily be swallowed into a 50amp pool if you wanted too. I keep 3 sizes. 1.6 for thin gage up to 1.5mm. 1.5-3mm I'm at 2.4 and above 3mm I'm away with 150 amps using a 3.2 filler rod. There is no need to make it more complex than that. Tungstens again I keep it very simple on AC. Up to 100 amps max 1.6 is fine and 100-200 a 2.4 is fine. If you have the power I'd suggest a 3.2 for over 200 amps.
    I think this is enough. I wanted to keep it simple and offer advice as a starting point only so that new welders can at least get going with something and more importantly know why there machine has these features and what impact they have on the welding.
    Thanks for letting me complete it before commenting. If you have any comments or questions or disagreements by all means tag em on now. And more importantly if you do want to take this info down take it quick before the thread becomes lost down the list.
    Cheers.
    Rich.
     
    Richiew, JP6SIX, harveyglen and 10 others like this.
  7. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    Taken from a previous post if you don't wish to read through a long drawn out post you can use this simple guide settings below

    Alu
    • Set machine to ac
    • Set balance to 70% dc negative
    • Current work around 40amps per 1 Mm of material thickness
    • Frequency 80hz for butts, 120hz fillets and outside corners
    • Gas argon 100% @ 6-8lpm
    • Pre flow 1sec
    • Post flow 6 seconds for 1mm and add 1 second for every mm after
    • Arc length 1-2mm
    • Torch angle 10 degrees from vertical lean away from your direction of travel.
    • Tungsten white, blue, gold, grey, don't worry about which people prefer at this time.
    These are very rough settings that I know work they don't cover all case scenarios but they will get you results and you can fine tune as your experience grows.
     
    Richiew, Dcal, harveyglen and 13 others like this.
  8. blenkie1

    blenkie1 Member

    What would be the chances of this being put in the main tig tutorial section? Is it a matter of cost? Its really useful information!
     
    harveyglen likes this.
  9. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    The cost in time of correcting all the spelling mistakes and miss use of grammar Alone would probably Kill it. It's been kindly put at the top of the thread with a sticky so it's very easy to get access too.
     
    blenkie1 likes this.
  10. ewmIreland pre-moderated

    Messages:
    62
    Location:
    Ireland
    downloadfile-2.jpeg

    Should it say EN-60% penetrate
    instead of EN-60% Weld
     
  11. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    It should! I didn't draw it I stole it.
    I think I explain the penetration polarity so the error isn't too bad. Well spotted though
     
  12. ewmIreland pre-moderated

    Messages:
    62
    Location:
    Ireland
    Im not trying to point out flaws Richard,just to help
    Youve done a great job so far
     
  13. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    I know mate that's why I say it was well spotted. I noticed it too and while I understood what they meant the terminology is wrong.
    I'm happy for any criticism on this thread because I want the info as accurate as poss so do say if you spot a flaw and I will see about getting it amended. Spelling mistakes etc don't count as its full of them lol.
     
    DAPPH, Appyasme1 and ewmIreland like this.
  14. Appyasme1

    Appyasme1 Member

    Messages:
    1,389
    Location:
    Lancashire
    Nice one Richard
     
    Richard. likes this.
  15. Well done that man, excellent, well explained information
     
    Richard. likes this.
  16. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    Appreciate that cheers mate.
     
  17. whm_fab

    whm_fab Motorsport fabrication

    Messages:
    348
    Location:
    ireland
    thats a massive amount of effort put in there, well done Richard great job, some fantastic information there for beginners and experienced alike.
    nice job :thumbup:
     
  18. Richard.

    Richard. Member

    Messages:
    18,137
    Location:
    Cambridgeshire
    Much appreciated.
     
    Matthew76 likes this.
  19. spencer 427

    spencer 427 Member

    Messages:
    6,156
    uk colchester
    Location:
    uk colchester
    Awesome post Richard...I haven't done any since I came to yours.but will get back in to it very soon..
     
    Richard. likes this.
  20. Piero

    Piero Member

    Messages:
    1,162
    Location:
    Uk
    Far to much time on yr hands lol.

    Top marks to Richard for his free time.

    A lot o effect will be going into this I am sure.

    Hope it gets appreciated
     
Advertisements