Welding Right Angles

[wilful]

For more than a couple years we have had issues welding right angles for the robot. We use right angle clamps and letting the metal cool before taking it out of the clamp but the pieces always seem to end up at funny angles.

Recently we were doing some welding for our shooter and the axle that the wheels are going to go on ended up not being as loose as it should have been.

Just wondering if anyone has some tips on this?

[sanddrag]

What is your welding process? You need to make sure the pieces are clamped well, but the heat will still warp it. It's to be expected. Typically we bend them back after welding.

[ttakashima]

Clamp in place on a flat surface, metal tables are best -> Tack a corner -> rework the angle -> tack a second corner -> fix again ->tack a third -> Check again -> finish the weld. This should fix any angle issues. Our drive base is always welded in one pass (~2 hours of welding and reworking angles) and we have not had much of an angle issue over the past 2 years. As for warping, we have had metal bow in on our chassis by about 1/32 inch per 36 inches, just use some heat and bend it back to where you want it.

[OSU_Springer]

One thing that may help you is to attach heat sinks near the area you are welding. Some chunks of brass or something similar may help keep the heat from warping everything too much.

[MrForbes]

See if you can set up the joint so that you weld an equal amount on the inside and outside of the angle. The more weld there is on one side, the more that side will shrink, thus the more it will be bent out of shape as it cools.

[Shaun Coon]

tac weld it in a couple of spots let it cool then connect the dots while welding them

[JamesCH95]

If you're welding a shaft onto a plate (what it sounds like) I would do 4-6 tack welds, then seam between the tacks in an alternating pattern like tightening down lug nuts. Ensure there is solid fit between the two pieces with little or no gaps. Gaps are a huge contributor to weld warping.

[Kevin Sevcik]

We've taken to riveting our frames specifically because of this. We're just too easily annoyed by everything going completely wonky after welding. So what I'm about to say is from theory and not actual practice welding robot frames.

Theoretically, the best answer to things warping during welding is more heat, not less. Allow me to explain for the uninitiated. Things warp during welding because you're dumping a lot of heat into a small area. That small area heats up and expands. Or tries to. There's all this cold, unexpanded metal around it that resists it expanding, you see. So it expands as much as it can, pushing against the metal around it, and if the force and stress exerted from this expansion exceeds the yield stress, then it yields. It yields and takes a set in its currently not quite as expanded as temperature requires state. And then it cools and shrinks down to smaller than its original size, on account of it's starting to shrink from a smaller size than it really should have been. There's also the fact that the filler metal itself is going to shrink an awful lot just because it froze at the solidification temperature, and it's going to cool down all the way to room temperature.

You can't completely prevent all this from happening, but you can reduce the effects by preheating your part. The closer the surrounding material is to the temperature of the welded area, the less warping you'll see, thanks to there being less difference in expansion between the hot metal and the cold metal. 6061 anneals at 775F, so you don't want to get that hot or your whole piece will get soft. 300F-400F seems to be the recommended range for a preheat temperature.

[AdamHeard]

It's an interesting topic really.

The more experienced the welder I'm talking to, the more they advocate to me that welding without post welding machining is NOT the best way to do things if you really, really want precision.

Precision is really the key. If you can design such that the warp and such doesn't cause an issue, you're golden. If not, be prepared to take a lot more effort during & after welding to straighten, tweak, thwack, etc. your frame into tolerance.

[MrForbes]

We haven't seen any warping problems in our welded chassis this year. Of course, we cheated by using steel....it's cheap, easily welded by a freshman with little experience and no formal training, and it keeps the bottom of our wide base robot heavy enough that it shouldn't tip over too many times in competition. We also designed it so that it doesn't really have anything critical as far as being in the right place, so even if it did warp some, it doesn't matter.

[kmccorm2000]

If you're welding aluminum it helps to preheat and use strong jigs.

[ratdude747]

for TIG, a trick I was taught once was to put the (pointed) end of the tunsten rod (make sure it is the "green" kind, not orange or red) about .25" INSIDE the porcelain gas shield...

also, for TIG, make sure you are set to AC, not DC if you are welding AL.

For any metal or welding type, be sure to use a wire brush to get the weld areas as oxide-free as possible... having an angle die grinder/drill with a wire brush makes this a lot easier.

Last, for TIG, this is kinda a basic thing, but make sure that you use the correct filler rod. for butt and inner angle joints, the thickness should match the thinner of the two edges being welded. for lap and outer angle joints, no filler rod is (generally) needed; sometimes it helps to "cheat" and use a bit of filler rod to get the puddle started.

[JamesCH95]

Originally Posted by ratdude747

for TIG, a trick I was taught once was to put the (pointed) end of the tunsten rod (make sure it is the "green" kind, not orange or red) about .25" INSIDE the porcelain gas shield...

also, for TIG, make sure you are set to AC, not DC if you are welding AL.

For any metal or welding type, be sure to use a wire brush to get the weld areas as oxide-free as possible... having an angle die grinder/drill with a wire brush makes this a lot easier.

Last, for TIG, this is kinda a basic thing, but make sure that you use the correct filler rod. for butt and inner angle joints, the thickness should match the thinner of the two edges being welded. for lap and outer angle joints, no filler rod is (generally) needed; sometimes it helps to "cheat" and use a bit of filler rod to get the puddle started.

Any benefit to pulling the tungsten into the nozzle? Maybe just to keep a new operator from dunking it in the puddle?

FWIW I prefer thoriated tungsten (Red, also radioactive! avoid unless you have vented grinding equipment!), lanthanated tungsten (orange), or zirconated tungsten (silver), over pure tungsten (green) for welding aluminum. Pure tungsten is a bit old-school and definitely not the best way to do it IMO as someone who earned an AWS welding certification in Al TIG.

Wire brushes, particularly non-stainless steel brushes, have a very nasty habit of making aluminum and other non-ferrous metals very dirty if used in weld prep. One must also ensure that stainless steel brushes are designated for one material only, i.e. aluminum, steel, or titanium, and that they are not used on a different material. I have always preferred abrasive sanding discs for aluminum weld prep because they don't contaminate the base metal, even if they're used on a different material. Though AC welding current (for AL) will break through the oxide layer as it is, and a break-arc technique will clean out even the toughest oxide deposits like anodized surfaces. Just be sure a good degreaser is used BEFORE and AFTER any brushing or abrasive prep.

In thinner sheet metal you could probably get away with just making a fuse joint for lap and outside corner welds. However, you run the risk of under-cutting the base material and causing a weak joint and some nasty stress risers. I would be very careful about specifying the right weld bead for each joint knowing that fuse joints are prone to under-cutting.