Welding thin Aluminium U channel

admin

HiMy first post and with it a great big thank you for all the knowledge beingshared. It has helped me tremendously in my lurking phase.I am a self taught welder from South Africa. I have a Lincoln Invertec205 AC/DC. Fantastic welder, not a single problem in 5 years. I boughtthe machine for fun, but now it enables me to make a living.I started off using it as a stickwelder doing light mild steel fabricationand repairs using mainly 6013 rod.Then I got into SS and aluminium welding. Better money and more interesting work. I have welded SS balustrading, piping etc etc.I now have the opportunity to weld thin aluminium frames out of U channelAluminium which is about 1mm thick. In inches I think that would be about1/24th of an inch or thereabouts. The sides of the profile are about 1/2of an inch long.The grade of the aluminium is unknown, but it welds well with 4043 rod.I set my machine at 120 amps, use pedal to control, gas lens, 6 cup, 6 litres argon (About 5 CFH I think - Sorry we are metric here).I set balance to 75, no prep, clean material. Tungsten is 2.4 mm thoriated.Point is maintained and no balls form. No pulse, 110 frequency.Welds are nice and shiny. My problem is that I cannot get a full penetration weld over the short distance of the weld. Even using a backing bar of steel does not really help.Is there a way to get full penetration welds on short welds on thinAluminium when trying to butt weld?As mentioned I am self taught, so any input would be most welcome.
Reply:Tricky stuff that thin Aluminium !I'm self taught - so that's my disclaimer.I do a fair of welding with thin Al tube and angle, often 1.2mm thick stuff, 25mm sections. Butt welds at this size are really susceptible to cracking - so I can see why you're concerned about penetration. I like to make sure I have a decent volume of bead lapping the join to help with the cracking issue.The amp setting seems high to me, I typically would use half that, and still use the foot controller to throttle down when needed. Pulsing is good too. With high amps its a very fine line between no penetration and a big hole!A photo would be good, but in the absence of that I suspect that you are running a bead very quickly over the surface of the join. If you turn the amps down, make sure your cut edges and weld area a really clean then you should be able concentrate on getting a small weld pool established before any filler is added - the base metal needs to heat up a bit. Only add filler and continue with the weld once you get a decent involvement of the parent metal. What size filler rod are you using? I use 1.6mm.Personally I wouldn't use steel backing . Copper is less likely to contaminate the weld.you mentioned that you don't prep the material. I usually don't either with new stock - but you should trial it. Flapper disk to take off the skin (or anodising!), acetone wipe. Do this on both sides of the weld.Weld from the outside edges of the channel in towards the thicker metal so the heat of the weld works with you.Regards- Mick
Reply:Thanks for the reply Mick. That is the reason I love this forum. You get a different perspective and more often than not I am in the woods and Ido not see the trees.I have 2000 of these frames to do and perhaps going too fast hasactually slowed me down.I will try your suggestion of lowering amps. I will also try small beveland different backing.Thank you
Reply:Honestly, if you've got 2000 to do, then I wouldn't worry about bevelling. I think the lack of penetration is more that the molten weld pool doesn't want to flow through nicely to the other side probably due to surface oxide on the back side (the back side doesn't get the benefit of the cathodic etching in the AC cycle). If you have a tight fitting copper back plate, then another thing you could do is make up a jig so that you can consistently have a small root gap (say 0.5mm) to allow easier penetration (etching and argon gets right through). If you wanted to get really tricky you could keep the copper bar heated with a small burner, which would preheat the aluminium to avoid the cold start/hot finish issue.A few other things to consider: if the weld gets too hot, then you risk having a brittle weld, and also the heat affected zone will cause the aluminium to lose its temper on either side of the weld. Check that the aluminium doesn't get all soft and bendy around your welds. Not sure if this is important to you but thought I'd mention it.Are you using cutting fluid to cut the pieces? contamination?Once you get in the groove you should be able to weld pretty quick. Start the weld slow, get the parent metal going, then run the bead with filler rod constantly in the puddle. It'll probably take 10 secs a side.Cheers- Mick
Reply:Have you considered putting the pieces in a rotating double vice and have a fixed bracket to holdthe torch so you could sort of semi-automate it ?   You could rotate the vice outboard to center 2 timesfor each joint while feeding rod with the other hand.  You could leave a very small open root and have afixed block on the back (or even some small copper capilary tube ) to shoot a small fountain of argon on the back of the open root.  T it off the torch line so its not running all the time.This  "jig" might alow  for a very fine control of the current and heat input and good repeatability.You would have to build it but for 2000 pieces it might be worth it..........I am thinking of a manual version of those welding heads they use to weld around small tubingwhere it clamps around the tube and rotates.  Just a thought !Tim
Reply:Can you explain why you are using u channel.Is their something inside the channel that needs to slide, cause full penetration will cause you some trouble their and have to carefully focus your heat.Better off to spend some time practicing on this thin stuff till you get it down, Also don't quench it after welding.
Reply:Hi GuysThank you for all the input. The frames will be used as outer strengthenerson shipping boxes that will be used repeatedly.The customer wanted a smooth finish. I explained about strength anddrop through etc etc. Customer knows best, so I made samples to specand they broke. I am learning and did not tell them told you so. I steered them so thatthey came up with the suggestion to leave the bead as is. I am not the best welder, some of the welds I see on the forum areway better. But by now my aluminium welds are pretty and strong, mostof the time, and I am proud of them. So I did not want to sand them downin the first place.One of my biggest learnings this week is that you can be technically thebest welder, but if you are a one man show you need customer skills.I am finding, to my great shock, that even companies that I regardsas technically more sophisticated than me have the stick it together, what's the problem approach to welding. You have to be good at welding,but the need and importance  to educate and handle the customer cannotbe underestimated. The more I weld the more I begin to appreciate how much I still have tolearn. Every time I weld I learn, and I want to keep it that way. Your Answers help me do that and I appreciate the time you take to providethem.Made new samples now and took one and bent the frame into a pretzel.No break. I went down one size in tungsten, 75 amp to establish puddlequickly, back off, 110 frequency and all is good. I make the puddle alittle bit wider than I normally wood. Produces a flatter bead likecustomer wants and still gets enough material down to give strength asneeded. HAZ does not seem to be a problem.This is my first order and I will do it manually. I do think that a semiautomatic setup will make sense for repeat orders and I will comeback for advice.The point about inside distortion is very valid and I did use it in mydiscussions with the customer.Thanks again for all the answers.Chris
Reply:HiI tried to attach photo of weld on one corner of pretzel.Let's see if this works. Thanks again for all the input.With this weld I wanted to reproduce production conditions. No cleaning,no wiping with acetone etc. Cut, put on jig and weld.Chris Attached Images
Reply:1 mm = 0.040 inch  Thin stuff.  You shouldn't need to bevel aluminum that is that thin.Aluminum needs to be clean to be welded.  No grease or oil or wax or paint or ink or whatnot.  A quick dunk-n-wipe in some acetone usually gets rid of most 'goop'.  There are other cleaners that can work too.Freshly cut clean ends of the channel and a quick swipe with a dedicated-for-aluminum-only stainless bristle 'toothbrush' should help in making sure the welding zone is clean and oxide-free.Cut, wipe/dip clean, bristle brush, wipe/dip clean again, then weld.  IMHO.Your original tungsten size was probably too big for the material.  Go down to an 0.040 (1 mm) or a 1/16 inch ( 1.6 mm) tungsten, they both have enough amperage/heat capacity for working on 1mm thick aluminum using AC.Gas flow rate for 'small' stuff is usually 10-20 cfh, which is 5-10 lpm (liter per minute).  Your stated 6 lpm should be fine.2000 assemblies to make?  Jig it up to make the overall job go better and faster.  The best laid schemes ... Gang oft agley ...