Aluminum tig welding cracks.

admin

some of our aluminum tig welds tend to crack when we weld the thicker aluminum eg. 32mm thick welded to 6mm aluminum. Would it be better to weld it with pulse? (Then we will have to buy a new machine)Yes I definitely pre-heat the thick material. I am using 5356 filler rod the 32mm is 6062 aluminum and the 6 mm is 6061. Cant use an oven either because masts and x-beams are somewhere between 5 and a 100 meters long.I can see the crack when I stop welding - so therefore post-heating wont stop the problem either. It cracks all down the side of the weld on the 6mm material. I'm using straight argon.yes i do fully button off when stopping to get a new rod.joint design problem - yes that i suppose is possible but how would one design it?
Reply:This is the weld that cracksLast edited by Anton Kleyn; 02-06-2014 at 04:11 PM.
Reply:Could you try 4043 filler? Looks like you may need a few more amps, and push some more filler into the joint.Airco Ac/Dc 300 HeliwelderMillerMatic 200 (stolen)Miller Maxstar 150STLMiller AEAD200LE (welding and generating power) Hobart MIG
Reply:i'd try pre-heating that thick heat sink.miller thunderbolt 250vlincoln square wave tig 175 prolincoln idealarc mig sp250everlast tig 210EXTeverlast power plasma 50chicago electric (hf) 130 tig/90 arcchicago electric 90 amp flux wire3 sets oxy/acet
Reply:Originally Posted by boatbuoyi'd try pre-heating that thick heat sink.
Reply:Are the tacks only cracking! I don't see any crack in the weld itself. If it is just the tacks I would say it's from having a on off switch type torch with no downslope to it. Two dissimilar thicknesses cooling at a different rate and the fat guy always wins the tug o war. A foot pedal would help you here, or a machine with slope functions if that's the case.Last edited by weldermike; 02-06-2014 at 07:54 PM.I hate being bi-polar it's awsomeMy Heroes Have Always Been Cowboys
Reply:+ 1 to what mike said, is it cracking while in use? May be a design flaw or load bearing problem in that case.
Reply:You all need to re-read the post, he said he preheated this.
Reply:What is 6062 aluminum?5356 filler has a higher tendency to crack than 4043.Welding/Fab Pics: www.UtahWeld.com
Reply:He also posted over there: http://forum.weldingtipsandtricks.co...php?f=5&t=4300More info on that thread.It's cracking all down the toe on the thinner side, if I read it right.Dave J.Beware of false knowledge; it is more dangerous than ignorance. ~George Bernard Shaw~ Syncro 350Invertec v250-sThermal Arc 161 and 300MM210DialarcTried being normal once, didn't take....I think it was a Tuesday.
Reply:Pulse won't help you I have a pulse machine at work I always turn it off all it does is throttle the current up and down.4043 might help but it is a weaker fill mat.  Are you preheating both pcs or just the heavy plate ?I used to tig weld wake board towers I would preheat both parts to be welded with a tiger tourch I heated till the water droplets evaporated and weld right away. The other thing that could cause it is lack of filler adding more fill could help
Reply:Yeah pulse does nothing for you. Try more amps, bigger filler, preheat, and blend some helium into your argon. Don't know what kind of power source you have, but transformer tigs are notorious for cracking on thicker alum. Like said above by Abec, you could slow down and add more filler to every dip to thicken the throat. When stopping taper off slowly and terminate with a little bump of thickness. Start again and back into that bump and continue. Again adequate reinforcement is crucial to prevent the crack. What is your power source?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:Do you use a particular code to weld aluminum?If not, Eurocode 9 contains some useful information, especially the calculation part. That might be helpful in your choice of filler material.
Reply:deletedCon Fuse!Miller Dynasty 350Millermatic 350P-Spoolmatic 30AMiller Multimatic 200Hypertherm PowerMax 1000G3Miller Maxstar 200DX
Reply:What temps are you preheating to and how long of a soak?   Can you post pics of where the crack occurs after a weld attempt?  Without knowing what the service life (temp exposure, chems and stress)  I agree that a 4043 filler would be more ductile in this application.   Does this part have to go thru any kind of post weld heat treat?   If you got a WPS or Code your welding to please provide some info as well. off the cuff advice without knowing jack about whats going onPreheat both materials to about 350 make sure the thick piece is preheated thoroughly but you still must warm the thin side as well.  You should be running enough amperage that the thick piece wets easily....very easily...this means your thin part will be super dang hot if not careful with the torch.  Apply THICK FILLER  to your heavy piece and kinda wash/wet into thin piece .... do circles or what have you to distribute the heat as you can.....Backstepping your welds is probably a good idea...your main piece is so thick you may have to have someone applying heat as you weld from the opposite side.  Consider a change in filler if that is an option.Another person spoke of throat pen....if you are doing this in a single pass that could be an issue by itself.....really thick part that may require a RT/BEAD pass first  small and hot as you can and then a fill/cap at a diff heat.   ALSO you may need to increase your tacks significantly  not bigger....smaller just more of em with a LOT more heat ....backstepping the welds would help with reducing stresses from heat input as well.   More info needed......   Trying to help thats allLast edited by tommyjoking; 02-07-2014 at 10:35 PM.
Reply:What have you guy's found TIG pulse useful for?  I'm getting off track, because my tig with pulse is DC, so not for aluminum.  It seems like it just makes a much smaller bead, but I haven't found a use for it.  My AC Tig doesn't have it.
Reply:Autogenous welds on thin metals, mainly automated stuff.
Reply:Anton, I'd say the cracking was the result of #1 joint prep and #2 the joint prep's effect on that cross section of weld's cooling contraction.I don't use 40series filler for other than cast aluminum so I can't say much about that alloy but I've done this weld with 5086 for 40 years without incident and believe that alloy will give the strongest weld.The lack of a groove or bevel in the thinner material allows the weld to 'lay' on the thin material and not have a root face area in proportion to the final weld.  By grooving/beveling/chamfering the thinner material so the single pass weld wets the entire (cross section) face of the parent metal at an angle to the section; the root face will not be contracting away from the planar surface of the thinner parent piece.This means the slot,. where the thicker material is placed, would be beveled to allow the weld to penetrate the entire thickness of the thinner (horizontal) material in the images shown.Next is the pulse question, yes that would drastically help this weld since the problem is one of cooling contraction. Pulsed (or  automatically timed, amperage surge) weld modes allow a higher peak amperage to be coupled with a low background amperage to cool the puddle more thoroughly as you progress.  This results in two benefits that are often not given their due; #1 a net BTU/wattage/heat/power reduction to get the parent metal fused.  & #2; A relative cooling of the weld as you go along reducing the amount of heat to be radiated, convected and conducted from the final weld- therefore reducing the net amount of overall shrinkage the weld experiences.The cracking you mention is over the weld face; toe to top of weld where the 'tearing away' is happening at the toe of the weld nearest the thinner material.  By grooving this material the weld will not be contracting on the same center as the 'surface mounted' weld you show.  Also by using pulsed (pedal or timer in the power supply) welds this weld will have less NET heat and will be cooled more thoroughly as it progresses and I use both techniques building commercial fishing equipment  (davits, booms, padeyes on masts, cable clevis mounts in cabins, decks and topsides where 3/4" to 1/4" is common)  and have had good performance in both instances.I note the material does not looked brushed and the argon cleaning of the beads shown is seems low? Last, I'd add more filler per puddle so the face is not concave but is flatter or even slightly convex.My reasoning about the bead profile comes from the idea  of looking at a cooling casting or a TIG puddle that is 'stopped' abruptly.  If you're welding along and just stop- we all know the cooling and contraction leave a dimple or recess in the top of the previously full looking puddle!  What's up?  Molten versus frozen metal.  Expansion and contraction of the overall weld is no different from the stop in TIG.IF you pulse the weld as you go, the next puddle's liquid can and is used to fill the 'last' puddle's contraction and if you fill just a little 'over full', when using pulsed or amperage stepped auto-timed weld modes, the result is less internal stress due to puddle contraction and less cracking.Just my experience doing these joints in both 6016 and 5086 materials.[FULL DISCLOSURE: I use a cold wire feed TIG gun to do all TIG welds but I believe the underlying principles to be the same for two handed dip rod methods.]Cheers,Kevin MorinKenai, AK
Reply:That is what i did to try and distribute the heat a bit more evenly - it is still crackingWhen I'm finished with the first run it has a crack all along the toe of the weld against the thin material. So I decided to give it a second weld all along the toe to cover up the crack - it still cracks but much less - still not exceptable.I cant use 4043 filler cause it must get anodized. When pre-heating I concentrate on heating the thick material only - because the 2 are close together the thin one automatically gets pre-heated in the process.I pre-heat for about 5 to 10 minutes (don't know what temp but it is very hot) with a blow flame.Does this part have to go thru any kind of post weld heat treat? No it doesn't.Does one need to wire brush ally when tig welding? As far as I know the machine does that for you (the reversed polarity or high frequency or something like that).Last edited by Anton Kleyn; 02-10-2014 at 04:36 PM.
Reply:Anton, I'm not all that convinced the larger padeye insert needs to be preheated. I think the TIG heat of fusion would be adequate since the block is fixed in size and not plate/sheet sized where the heat sink is 'infinite' compared to the other metal piece.I don't preheat if the 300A power supply I use is adequate; once in a while it is not, and if I do preheat I always use a temp crayon ( http://www.caltechindia.com/TempSticks.htm .... not this brand but this type of indicator) so the temp of my preheat is below 275-300deg F as the added weld heat will not only keep the thicker material warmed, once the weld is begun, but will elevate it somewhat from that preheat level.I brush all welds immediately before I weld them in aluminum, MIG or TIG and without exception.  Yes, the AC will tend to lift an oxide film but a scarified film will lift in pieces that have edges which are more effectively blown off the puddle surface by the hot argon- resulting in a cleaner finish AND less impurities in the final weld, but the mill scale on extrusions may be contributing to the toe cracks- so I'd say to eliminate that possibility it would be normal for me to brush first with a power brush when doing the overall prep after beveling and washing with acetone, and THEN to repeat with a hand (tooth) brush just before I put the weld in- to scarify the oxide film.I don't know the extent of the re-inclusion of the oxides into the puddle as it freezes if the film is not fully cleaned, but I do know that if the edges of both sides of material is not cleaned the bead will include the contaminates on the 'off side' of the weld.  But because I brush every weld, and clean both sides, I try to avoid these conditions whenever possible.I do know that mill scale (generic term for mill produced surface of sheets/plates/extrusions and other 50 & 60 series aluminum) is not aluminum, and melts at 2x to 3x the temp of aluminum, holds water molecules, is not just aluminum oxide, and is cleaned off (mechanically) in 100% of the weld spec.s I've seen for these alloys. So I brush as part of the weld.You don't show or mention the beveled weld prep and don't show the weld bedded into the entire thickness of the thinner material, (so I'm)  just checking in- that you did in fact bevel for this weld and the bevel was 45 deg of the entire thickness of the thinner material?I'd suggest, if the power supply will provide the heat of fusion with a limited preheat or NO preheat (?) to do the weld in that condition and see of there is any reduction in cracking?Cheers,Kevin MorinKenai, AK
Reply:To answer some of your questions:The 300A power supply is not adequate, it will take about 20min before it starts to melt.When I lift my welding shield the crack is there.4043 is not an option.No this 32mm to 6mm we have not done this before.4943 can it be anodized and have no color change?Yes I did preheat the blocks - I made everything the same tempOk what I have done:After I welded it right around - I then turned my amps down by about 70 and welded one run right around again over the crack. This is not ideally what I want to do but it worked.On the next one I welded, it obviously cracked again so like the previous time I welded over the crack again BUT this time it cracked again so I welded it a 3rd time over the crack before the crack was gone - **** - this is not how I want to do this?What I will try next time: (some of the things you suggested).Normally I would weld the bottom part first (the underneath side) which is much smaller than the top side and it doesn't petrude very much (more like a but weld). Next time I will weld the top side first (this side requires a lot more heat than the bottom side). This should allow for more movement in the thick material when welded (less restrained). I will preheat and wire brush before I do any welding. I will add more filler rod per dip. The 2 flat plates that cover the end of the tube, I will leave them off until all the welding is done and then only weld them on. I will bevel the tube (I did not do this in the previous welds), make a full penetration root pass, then weld over it all again. It sounds like the last run should be a thick run? (someone mentioned heat, penetration and the size (big) of the weld).Lets hope for the best! Any other suggestions let me know. I will let you know what happened after I tried this. Hell I hope this works!!!
Reply:Originally Posted by Anton KleynTo answer some of your questions:The 300A power supply is not adequate, it will take about 20min before it starts to melt.When I lift my welding shield the crack is there.4043 is not an option.No this 32mm to 6mm we have not done this before.4943 can it be anodized and have no color change?Yes I did preheat the blocks - I made everything the same tempOk what I have done:After I welded it right around - I then turned my amps down by about 70 and welded one run right around again over the crack. This is not ideally what I want to do but it worked.On the next one I welded, it obviously cracked again so like the previous time I welded over the crack again BUT this time it cracked again so I welded it a 3rd time over the crack before the crack was gone - **** - this is not how I want to do this?What I will try next time: (some of the things you suggested).Normally I would weld the bottom part first (the underneath side) which is much smaller than the top side and it doesn't petrude very much (more like a but weld). Next time I will weld the top side first (this side requires a lot more heat than the bottom side). This should allow for more movement in the thick material when welded (less restrained). I will preheat and wire brush before I do any welding. I will add more filler rod per dip. The 2 flat plates that cover the end of the tube, I will leave them off until all the welding is done and then only weld them on. I will bevel the tube (I did not do this in the previous welds), make a full penetration root pass, then weld over it all again. It sounds like the last run should be a thick run? (someone mentioned heat, penetration and the size (big) of the weld).Lets hope for the best! Any other suggestions let me know. I will let you know what happened after I tried this. Hell I hope this works!!!
Reply:Have you confirmed that the thinner material is in fact the alloy you think it is?
Reply:I would push more filler on the root pass so that weld is not concave.. the 2nd pass pulled the crack up to the surface again. likely when the root was made you're pretty much all in on the heavy side, letting the puddle find it's own way over to the thin stuff and it just needed a bit more filler. keep gravity in mind too and position the part so the puddle does not want to run away from the toe line on the thinner material. that dye test on aluminum is pretty high scrutiny.. I had to make some adjustments to welding technique first time i was tested like this. It's definitely more of a test of welder attention to detail than structural soundness. Kind of depends on the cwi/testing company if they are by the book or look at the part for what it is sometimes this test is not necessary.