I'm trying to troubleshoot a problem with welding up a long rectangular-section tube. The cross-section is about 7 x 15", about 12-feet long, and is made up of 1/8"-thick aluminum sheet (sorry, I don't know the type of aluminum). It is made up from two pieces of sheet bent into an L-shape, then the sheets are put together with their edges at two of the section corners. The weld is a continuous bead along the outside corners where the edges of the sheets meet.
There are two problems, which I believe are closely tied together; I think that if I can solve one, both will be fixed. One is that after the section is (TIG) welded together, the cross section shape is no longer rectangular - the angles inside the section where the corners were welded are less than 90-degrees. It has become a parallelogram. The second problem is that as you sight down the length of the tube, the edges which have been welded are bowed inward toward the center of the tube. These parts need to be pretty straight and have 90-degree corners.
I think I can understand why the long edges are bowed inward - the cooling weld pulls along the length, only on the outside of the tube, which sucks the corner in toward the center. I'm not really sure why this causes the corner angles to be less than 90-degrees, but I think that is the case.
I think a joint redesign might help, but the pieces are made and ready to weld, so that isn't a very attractive option. Additionally, I understand that these parts have been made years ago with an identical joint design, and this problem didn't come up. (They had different problems back then!)
I'd like to hear from anyone who might have ideas on how to change our procedure to get around this problem. The parts are tacked in place along the length of the tube, then a continuous weld is run over the top. I believe that the continuous weld is not done all at once starting at one end, but rather is welded in 12" or so sections, jumping around until everything is welded.
Any ideas or suggestions?
I weld a lot of aluminum. I am not an expert however. One thing that helps is to have a backup (Chill Bar) that also restrains the movement of the material as it shrinks during cooling. In your case a piece of heavy angle 3/8" to 1/2" thick and 4" to 6" on the legs. Heavy tubing would be ideal. You must have another piece of steel clamped on to the outside of each piece to be welded, like a sandwich. The welding should be by the back step method. Back stepping is where you start to the left and weld to the right. Move over to the left and weld to the right again. It can be done right to left as well. Do this in three to six inch welds. Do not remove the restraints until cool. I am in California and would be willing to set up a time to talk if you email me. Just click my name at the top of this response to see my email.
Dennis
Thanks for the comments. I'm not sure from a practical point of view how I can clamp pieces of heavy chill bar to the inside and outside of the joint when the part is 12 feet long, and is a closed tube. However, I'll give it a try and see if I can come up with something that at least improves the situation.
Do you have GMAW equipment available? You are in the thickness range for GMAW (1/8 inch) and 12' GTAW welds put an awful lot of heat into the work. GMAW would be considerably faster and overall would put less heat (a major distortion factor) into your work.
Combine that with well thought back step sequencing, maybe some angular pre-loading and what ever heat sink is practical will take you a long way to removing your distortion.
It's a big process change, but if you have alot of these pieces to build a switch to GMAW may be a reasonable consideration.
I believe we have a machine we could set up for aluminum. I'll give that a try and let you know how it works out.
As mentioned, chill blocks, heat sinks, or backing bars may be used to minimize distortion. Distortion is directly related to the total heat input or the size and quantity or frequency of welds. Because of the rapid dissipation of heat when welding aluminum, the greatest contributor to distortion is weld metal shrinkage. Molten aluminum shrinks approximately three times than that of a similar volume of steel. As Lawrence mentioned, the GTAW process has a higher total heat input than the GMAW process for equivalent welds, therefore where distortion is critical, use the GMAW process if available.
If it's feasible, you may try a backstepping technique also.
If you have a lot of these to weld, all the same size, you could build a jig.
In otherwords, Make a tube out of steel that fits the exact ID of your aluminum tube. This would act as both a heat sink and structural strongback in order to eliminate warpage. Just make sure you let the aluimnum tube cool before removing it from the jig.
I also second the motion about switching to MIG. That would probably cut you labor and consumables expense in half.
Let us know how you wind up working it out.
Tim
It would indeed be nice if I could build a jig that fit the ID of the tube snugly. But, it would need to be collapsible, or I could never get the part off of the jig after welding!
I discussed the MIG suggestion with the welder in the shop, who had nothing but negativity about the idea. I may have to wield my seniority and get him to at least try some of these ideas.
I will try to let everyone know how it all works out. Thanks for everyone's input. This is a great system!
switching to GMAW is a must here. Better yet, if you have PULSED MIG capabilities or some of the new advanced waveforms that are available,,,some of them work very well
In addition to all the good advice above, if you bend the parts so that what becomes the long diagonal is a little shorter the distortion will move things toward rectangular. Some experimentation will be necessary.
Attacking the distortion with a porta power may work.
Peining the weld without supporting the other side will tend to expand the side peined and can help remove distortion.
Bill
A method that will work to hold your 10 gage aluminum is the heavy tube I mentioned in my post above with an outside clamping member made from a "WT" section clamped on each side of the weld joint. The way it works is easy. First cut a wide flanged beam down the middle with a torch. It must be longer than twelve feet. The wide fanged beam should be a W12X26 min. The two pieces of beam are now WT's and they look very curved. Put the heavy tube section on pipe stands so that any corner is up. Put your tacked aluminum over the tube so that one corner is up. Take one of the WT sections and attach the flat side (flange) to one end of the tube, with a strong "C" clamp making a sandwich of the WT ,aluminum, and the tube. Now clamp the other end. As you tighten the second "C" clamp down the curved WT becomes a clamp that applies even pressure down the full length of the weldment. Put the other WT on the other side of the joint and you have exo-fixture that requires little brake down. Weld as directed above with back step and MIG.
Forum
