Blue 'halo' when introducing filler on Aluminium TIG weld

admin

Well - Aluminium is challenging me!Once I get going I get a decent finish, but when starting off, or when tacking, I get a blue halo around the filler rod when I introduce it and it will not stick nicely to the base aluminium. It kinda blobs up into a nasty ball, and I have to shake it off the end of the filler.  Meanwhile, I have burnt through the very thin ali sheet......I am thinking contamination of course, but I have:- wiped filler down with acetone- run filler through a green scourer pad- even tried chopping off the end from the previous weld to get to fresh filler- tried a different pack of filler (2.4mm instead of 1.6mm)- wiped/rubbed the TIG glove fingers with acetone as they were a bit blackBut nope - still get the blue halo, and blobbing effect. Not sure what else I need to consider.My setup is:- Kemppi MLS 2300 ACX MasterTig AC/DC- AC set to 40% EN and 60% EP (approx - Kemppi have a weird way of expressing this)- 1.6mm (1/16") Tungsten (rare earth) - but also tried Ceriated 2% - sharpened to a point (not balled)- Gas saver lens (inbuilt mesh etc for smoother gas flow)- 1.6mm (1/16") Filler 5356- 1mm Aluminium being welded (a bit less than 3/64" - that'd be 1.2mm I reckon)- 15 to 35 A, controlled by foot pedal- Pulsed at 1.2 second frequency- Various Hz frequencies from 60 up to 250 - Square wave- 6L/Min of Argon (that'd be 13 cfh for you guys)Any particular thoughts on where I am stuffing up?Thanks a lotDavidLast edited by SpudBoy; 07-31-2012 at 08:14 AM.
Reply:Use a larger cup with a gas lens.............. Use lots of amps to make the start and then let up on your foot pedal and start to travel.........Lincoln SA200's... at least 15 - 20. They come and go. Growing partial to the "Short Hoods" in my old age. Last count on Short Hoods was 13 in possession.
Reply:I'm using a gas lens, with a #5 cup.  Is that an OK size?With the 1.6mm tungstens I am using this is the cup size that comes with that gas lens.
Reply:Your EN and EP looks strange to me. Your EP should be equal or less than EN. I think you are not getting any penetration in the base metal.I would run your Hertz to midway at 120 or so. Why are you using pulse? Is it a cadence application? I find that pulse on alum creates more problems than it solves. Then check your gas flow. Most problems start and finish with gas flow, delivery to torch, or contamination of gas. If you have a new gas lense try it. Before changing it crack the backcap just a tad, remove the gas lense and look into the threads to see if the backcap has cut off the gas flow. Also check to see if the collet has be crushed or bulged, which will also cut off gas flow. Check your backcap o-ring, and if a hole is in your backcap. Then install new gas lense, and try again.How thick is your alum?Weld like a "WELDOR", not a wel-"DERR" MillerDynasty700DX,Dynasty350DX4ea,Dynasty200DX,Li  ncolnSW200-2ea.,MillerMatic350P,MillerMatic200w/spoolgun,MKCobraMig260,Lincoln SP-170T,PlasmaCam/Hypertherm1250,HFProTig2ea,MigMax1ea.
Reply:Originally Posted by SpudBoyI'm using a gas lens, with a #5 cup.  Is that an OK size?With the 1.6mm tungstens I am using this is the cup size that comes with that gas lens.
Reply:It sounds like you're producing more oxidized aluminum than the cleaning action can remove so, while the blue haze is thick, so is the remaining film which prevents a good bead from forming. Besides the above-posted recommendations, you might  try moving the torch away, then hitting the foot pedal, releasing, and moving the torch close again while gas flow continues; finally hit the pedal again while the gas flow is on. This assumes you don't already have preflow, of course.
Reply:Kemppi don't express balance settings the conventional way... the readout is -50 to 0 and the default setting is -25%. The numbers aren't absolute, if they were then -50 would = 100%EN (DCEN). Rumour has it that the default setting of -25 is actually about 67% EN, a setting of -50 about 90 odd %ENFrom the description it sounds to me like technique... amperage is a little on the low side for 1.2 ally (especially tacking) and the slow pulse will be furthering lowering the heat input. The way forward for tacking ally is short arc length, hot, quick. Build the heat slowly, especially on thin stuff and you heat the whole area. If your arc length is on the long side and the filler wire is kept closeish (ready to dab) then that gets hot too, can get oxidised and then when you try to add it it shrivels back/balls up and you get a manky mess. Another common newbie error is trying to add wire too soon, the arc melts it instead of the weldpool and again result is typically a manky mess. Larger diameter wire will make things worse as you need a larger/hotter weldpool to melt itThe one setting you haven't mentioned which could make a difference is preflow. Make sure you've got at least 0.5sec or, as said, activate the post flow before striking up.I'd stay with 1.6 for wire and tungsten, turn the pulse OFF, turn the amperage up to at least 40A (50 would be better for tacking), skip most of the OTT cleaning (yes, you can't get ally too clean but it's WAY overdone on many welding forums) and for the initial tacks... stamp on the pedal, be ready to add a dab of wire and back off the pedal. AC frequency... depends on the joint, higher frequencies are more useful on fillets than butt welds BUT it's not a make or break setting- everyone got by just fine before inverters were invented.
Reply:Could also try a bigger filler. I used 1/16 for years until I tried a 3/32" (I believe). The thicker filler won't ball up and run away from the puddle as easy as the thin stuff. Even on thin sheet it works, just put less filler in the weld when moving.Last edited by VPT; 07-31-2012 at 12:24 PM.
Reply:Hey!  I woke up this morning to see allthese replies - so thanks a lot for all your assistance.Firstly: Originally Posted by shovelonYour EN and EP looks strange to me. Your EP should be equal or less than EN. I think you are not getting any penetration in the base metal.I would run your Hertz to midway at 120 or so. Why are you using pulse? Is it a cadence application? I find that pulse on alum creates more problems than it solves. Then check your gas flow. Most problems start and finish with gas flow, delivery to torch, or contamination of gas. If you have a new gas lense try it. Before changing it crack the backcap just a tad, remove the gas lense and look into the threads to see if the backcap has cut off the gas flow. Also check to see if the collet has be crushed or bulged, which will also cut off gas flow. Check your backcap o-ring, and if a hole is in your backcap. Then install new gas lense, and try again.How thick is your alum?
Reply:Originally Posted by Oldiron2It sounds like you're producing more oxidized aluminum than the cleaning action can remove so, while the blue haze is thick, so is the remaining film which prevents a good bead from forming. Besides the above-posted recommendations, you might  try moving the torch away, then hitting the foot pedal, releasing, and moving the torch close again while gas flow continues; finally hit the pedal again while the gas flow is on. This assumes you don't already have preflow, of course.
Reply:Originally Posted by hotrodderKemppi don't express balance settings the conventional way... the readout is -50 to 0 and the default setting is -25%. The numbers aren't absolute, if they were then -50 would = 100%EN (DCEN). Rumour has it that the default setting of -25 is actually about 67% EN, a setting of -50 about 90 odd %ENFrom the description it sounds to me like technique... amperage is a little on the low side for 1.2 ally (especially tacking) and the slow pulse will be furthering lowering the heat input. The way forward for tacking ally is short arc length, hot, quick. Build the heat slowly, especially on thin stuff and you heat the whole area. If your arc length is on the long side and the filler wire is kept closeish (ready to dab) then that gets hot too, can get oxidised and then when you try to add it it shrivels back/balls up and you get a manky mess. Another common newbie error is trying to add wire too soon, the arc melts it instead of the weldpool and again result is typically a manky mess. Larger diameter wire will make things worse as you need a larger/hotter weldpool to melt itThe one setting you haven't mentioned which could make a difference is preflow. Make sure you've got at least 0.5sec or, as said, activate the post flow before striking up.I'd stay with 1.6 for wire and tungsten, turn the pulse OFF, turn the amperage up to at least 40A (50 would be better for tacking), skip most of the OTT cleaning (yes, you can't get ally too clean but it's WAY overdone on many welding forums) and for the initial tacks... stamp on the pedal, be ready to add a dab of wire and back off the pedal. AC frequency... depends on the joint, higher frequencies are more useful on fillets than butt welds BUT it's not a make or break setting- everyone got by just fine before inverters were invented.
Reply:Originally Posted by VPTCould also try a bigger filler. I used 1/16 for years until I tried a 3/32" (I believe). The thicker filler won't ball up and run away from the puddle as easy as the thin stuff. Even on thin sheet it works, just put less filler in the weld when moving.
Reply:Originally Posted by hobohilton----------------------Run the largest cup you can get into the space you have to work with. The more blanket the less chance of O2 getting to your puddle.Thanks,Hobo
Reply:Just to add some testing I've done: - if I weld a test bead in the middle of the 1.0mm flat sheet (i.e. - not joining anything - just a stack of dimes test weld) I don't get the blue halo, and the weld starts nicely. Does that indicate that maybe the gas is flowing away when I am doing the outside corner welds?The pool on the test welds is bigger and probably more molten when I introduce the filler.  If I get the pool that big on a corner joint I end up with a hole creeping away from the vertical plate because it's right on the edge of the sheet.These test welds end up with a nice 'shiny'; finish, whereas my corner welds end up a dull grey finish. The test welds are run much faster than the corner welds, so maybe that is why they are shinier.Last edited by SpudBoy; 07-31-2012 at 07:23 PM.
Reply:Originally Posted by SpudBoyJust to add some testing I've done: - if I weld a test bead in the middle of the 1.0mm flat sheet (i.e. - not joining anything - just a stack of dimes test weld) I don't get the blue halo, and the weld starts nicely. Does that indicate that maybe the gas is flowing away when I am doing the outside corner welds?The pool on the test welds is bigger and probably more molten when I introduce the filler.  If I get the pool that big on a corner joint I end up with a hole creeping away from the vertical plate because it's right on the edge of the sheet.These test welds end up with a nice 'shiny'; finish, whereas my corner welds end up a dull grey finish. The test welds are run much faster than the corner welds, so maybe that is why they are shinier.
Reply:You only want as much electrode positive as you need. 60% is too much and will cause problems. Try 30% EP (70% EN). You should see a mirror-shiney puddle form very quickly, and it will accept filler.Welding/Fab Pics: www.UtahWeld.com
Reply:Originally Posted by SpudBoyHaha - I am SURE it is technique!!  Manky mess is a good description I have tried the balance at 0%, 25% (the default for AC on a Kemppi) and 50%. For tacking I am taking it easy and putting the heat in slowly. I am mostly trying to join 1.0mm sheet at a 90 degree angle, and welding from the 'outside' edge (i.e. I am not doing the 'inside' 90 degree joint fillet joint, but rather the 'outside' 270 degree joint).  Once I get the puddle going on the flat sheet, it starts to keyhole and I get a gap pulling away from the vertical sheet.  Maybe I need to hit it harder to start with.To make a tack, I am taking 15 or 20 seconds from the time I start to heat things up to the time I finish - sounds like this is too long.I have tried tacking without filler (at higher amps) but the two weld pools will never 'join up'. The molten aluminium on each side of the join just seems to repel each other, rather like reverse mercury!  I'll give it a try without Pulse and with higher Amps and let you know how I get on.
Reply:Originally Posted by MikeGyverYou only want as much electrode positive as you need. 60% is too much and will cause problems. Try 30% EP (70% EN). You should see a mirror-shiney puddle form very quickly, and it will accept filler.
Reply:Originally Posted by shovelonHow does your tungsten look? Any discoloring?
Reply:Originally Posted by hotrodderThe bit i highlighted in bold... yep building the heat that slowly ain't gonna help. This example is a fair bit easier as we're talking a butt joint and 1.5mm wall 1.5" tube but the technique is pretty much the same...With an outside corner and thinner material there's no margin for error and very little material to make an autogenous tack with i.e. better off using wire but you need to have a VERY short arc length (less than the diameter of the tungsten), hit it hot and be quick. Without wishing to sound rude unless/until you can 'produce the goods' on 1.5mm or thicker ally while talking to someone/singing along with the radio/lying on you back welding overhead etc then 1mm ally will be a nightmareEP is what provides the 'cleaning'. EN puts less heat into the tungsten and more into the work.There's a good chance it's as simple as the bit in bold but gas coverage could also cause this... the test welds were done on flat sheet, the geometry of an outside corner is different making good gas coverage harder to achieve as you said- many people often turn the gas flow up when they have/suspect coverage problems but with an outside corner this often makes things worse. 6L/min is in the ballpark, knocking it back to 5 or just under might help... it depends and without pics we're all guessing as there's dull grey from overheating things a bit (too slow) and then there's dull grey...The test welds didn't invovle an actual joint so you sidestepped the issue you've been having with tacking up/starting an outside corner...
Reply:kempi is retarded in their settings. That is so hard to easily relate to. Why not use a direct percentage of a full sine wave? In what world does -50% equal -75% (according to diagram)??Lincoln Power Mig 210MP MIGLincoln Power Mig 350MP - MIG and Push-PullLincoln TIG 300-300Lincoln Hobby-Weld 110v  Thanks JLAMESCK TIG TORCH, gas diffuser, pyrex cupThermal Dynamics Cutmaster 101My brain
Reply:If you are from Finland it might make sense - who knows.  Apart from that, it is a great little machine. All other controls make sense, but that one always causes me to scratch my head.
Reply:The only time I get any green or blue glow is when the material wasn't cleaned good enough and your seeing the cleaner whatever you used burning.....That would be my guess  Miller Dynasty 350Twenty Six HammersThree Crow BarsBig Rock
Reply:My thoughts are to drop the pulse for a bit, and try a few things: Freq around 70-90. Tungsten looks ok, so no change. As was stated, EN balance around 70%. Set amps to 40. Use the pedal, and try to start the puddle by starting hot, and back it off. See what it takes to get the start up going well. The welder can tell you peak amps, and last amps used, and maybe ave. amps for that last function. Try to match those parameters with your pulse setup.If the welder can not give you the amps info, then by starting with a round number as 40, or 100, you can pretty easily approximate what you did that worked- ie 'Started wide open, then dropped to half withing 2 seconds. Well, there is your setup! Peak amps at 40, base amps is 20, and 30 pulses per minute, meaning .5 PPS.Set the ramp up and ramp down times very low, .25 second. No need to soft start and then no need to downslope a bunch at the end.And then, after so much work...... you have it in your hand, and you look over to your side...... and the runner has run off. Leaving you holding the prize, wondering when the runner will return.
Reply:Originally Posted by SpudBoyThis is a (not very good) photo of my tungsten. Sharpened, but with a slight ball at the very tip.Is that discoloured?Originally Posted by DSWIt's hard to tell from that pict, but the black on the tungsten and that dull "frosty" section looks a lot like what I often see with alum contaminated tungstens. I'd expect to see the nice shiny area above the black all the way down to the very tip. Alum will wick it's way up a fair ways some times and will keep contaminating the tungsten if it isn't all removed.
Reply:Originally Posted by B_CThe only time I get any green or blue glow is when the material wasn't cleaned good enough and your seeing the cleaner whatever you used burning.....That would be my guess
Reply:Originally Posted by RojodiabloMy thoughts are to drop the pulse for a bit, and try a few things: Freq around 70-90. Tungsten looks ok, so no change. As was stated, EN balance around 70%. Set amps to 40. Use the pedal, and try to start the puddle by starting hot, and back it off. See what it takes to get the start up going well. The welder can tell you peak amps, and last amps used, and maybe ave. amps for that last function. Try to match those parameters with your pulse setup.If the welder can not give you the amps info, then by starting with a round number as 40, or 100, you can pretty easily approximate what you did that worked- ie 'Started wide open, then dropped to half withing 2 seconds. Well, there is your setup! Peak amps at 40, base amps is 20, and 30 pulses per minute, meaning .5 PPS.Set the ramp up and ramp down times very low, .25 second. No need to soft start and then no need to downslope a bunch at the end.
Reply:So, I did the following:- Cut off my tungsten at the black mark & re-ground to make sure there was no aluminium on the tungsten- turned the Argon down to 5L/min- Turned off the Pulse- Increased the Max amps on the foot pedal from 40 to 50- held my tungsten closer to the work- hit it much hotter and much quickerI am getting less of the blue halo when I introduce the filler, and the tacks are not eating away my 1mm sheet like was happening when I was taking 20 or 30 seconds per tack. These were done is 3 or 4 seconds I reckon.The nasty crackling and spitting sounds I was getting have gone away.HUGE improvement, so thank you all very much.  Much appreciated. Who'd have thought that more heat would give less melt through.....Got to work on removing the crater, and then doing a full edge, but the tacking side of things is hugely improved.This is what I am getting now (tungsten touched on the 2nd to top tack): Attached ImagesLast edited by SpudBoy; 08-01-2012 at 04:32 AM.
Reply:Aluminum welding 101: You essentially floor it to develop a puddle as quickly as possible, then back off as necissary and weld.Welding/Fab Pics: www.UtahWeld.com
Reply:All in all, looks like you are on your way.Weld like a "WELDOR", not a wel-"DERR" MillerDynasty700DX,Dynasty350DX4ea,Dynasty200DX,Li  ncolnSW200-2ea.,MillerMatic350P,MillerMatic200w/spoolgun,MKCobraMig260,Lincoln SP-170T,PlasmaCam/Hypertherm1250,HFProTig2ea,MigMax1ea.
Reply:Don't get too nervous about touching the tungsten when tacking, I sometimes actually bump the tungsten into the weld joint to get a tack especially if there is a gap, it's no big deal......I call it GAGING  the tungsten, you'll come up with all kinds of tricks before your done......  Miller Dynasty 350Twenty Six HammersThree Crow BarsBig Rock
Reply:Originally Posted by B_CDon't get too nervous about touching the tungsten when tacking, I sometimes actually bump the tungsten into the weld joint to get a tack especially if there is a gap, it's no big deal......I call it GAGING  the tungsten, you'll come up with all kinds of tricks before your done......