Welding Tips and Tricks

I am having issues keeping my projects square, and I have tried just about everything I have seen/read/hear/dreamed of with little luck.

I am trying to build the chassis for an off road buggy, and need to keep everything square. I am working with .065" 1.5" steel square tubing. I'm having issues with both butt joints and mitered joints. Most of my joints are good fitting, with the occasional cut that wondered off.

I have tried tacking the 4 corners of each weld, then welding from inside to out. Both jigged tightly in place, and loosely. I have tried welding the inside, then outside of the joint, then coming back to the sides. The only thing I haven't tried is to weld and then wait to cool and weld again. And I haven't tried to weld an inch and let it cool and weld more etc...

In most cases, the issue has been more of a twist in the overall piece with reasonably square corners. (imagine a ladder where people on both ends twisted in opposite directions). I was able to get one piece to come back square (within 1/4" at least) by clamping the 4 corners down tight to my bench, then taking on O/A torch and heat each joint to cherry red, and systematically working around the entire piece. It took me 3 or 4 laps around it before it was in line, and then I left it clamped over night to cool.

I'm sure that my problems are either too much heat, or welding in the wrong order, but I am beyond frustrated...

Almost all material will pull when welded - steel is not too bad compared with stainless - but you will generally have to either accept some straightening after welding or heavily clamping the job & using minimal weld or the more usual practice of setting the joint off by the approximate amount it will pull when welded so the weld actually straightens it - ie if doing a miter joint then cutting the angles at 46 degrees means a close fit -up gives an overall joint of 92 degrees - but after welding will pull into hopefully 90 degrees - the exact amount will vary with material type / size / thickness.

If you can afford one buy a good quality cast iron corner clamp - this helps setup no end - Jody uses a nicer version of this in some of his video's.
On a miter joint try for a good fit but if there is any mis-match it's better to have the outside corner open & the inside touching - but whatever the fit - weld the outside corner first - then the sides - leave the inside to last & keep the weld as small as practical - generally do edge prep on the sides & outside corner - but not the inside - as to the twist aspect - I would expect that to be in the set up - not the weld pull - on small stuff laying it out on a known flat table is OK - on larger things most of us have to make do with shimming the frame on the workshop floor until it's about right with a good quality spirit level.

Hope some of that helps & good luck with the project.

I spent some time working/experimenting today, and I know the source of my twisting. It was coming from the order I finished the welds. I actually saw it happening.

Having seen the metal contracting as it cooled I now understand whats going on.

michialt wrote:I spent some time working/experimenting today, and I know the source of my twisting. It was coming from the order I finished the welds. I actually saw it happening.

Having seen the metal contracting as it cooled I now understand whats going on.

michialt -- perhaps others can gain from your problem solving success if you can provide more details on what you observed, how you reached conclusions, what corrective action(s) you are taking & what the results are...

Regards,
Lee

Lee;

I'm more afraid that even the beginners would be laughing at how I managed this... But I will try to explain how and what I did.

My first step was to clamp the two pieces of square tubing to a corner square jig that I built from Angle Iron. I then tacked the four corners. When I took it out of the jig and the tacks were cool, the metal was out of square in the direction of the last tack... I remembered reading somewhere that the metal at the end of a weld would shrink more as it cooled than at the beginning of a weld... Light bulb moment, I now see this first hand. Instead of forcing the metal joint back square I decided to play with it a little. I next welded the joint starting with the last tack, and welded so that I ended my weld such that the end of the weld would pull counter the metal back towards square.

As it cooled, it did correct the squareness, but it also pulled the joint out of square in another direction... so I welded to put the end of that weld pulling towards square... I kept repeating this process until I had welded all 4 sides of the tubing, and then I let it sit and cool, when it was back to ambient temp it was magically square, and I had wasted about 1.5 hours welding two 1.5" square tubes together....

I still haven't worked out a system for insuring that I am square every time, but I have a much better understanding of what the metal is actually doing as it welds. I also think the amount of distortion is relative to how fast/slow you lay the bead down.

The slower you run the bead the more time the beginning of the weld has to start cooling, and as it cools it is contracting, and as it contracts it is pulling the joined apart ahead of the bead and the end result is the end of the weld is pulling against a solid weld as it cools so it cannot contract as much as the beginning of the weld because the beginning is pulling against an open joint.

I think that the faster you can lay the bead the less distortion there will be. I also learned that it doesn't matter how solid you jig up the joint, it will still distort after you remove the jig due to the "elastic" like stretching of the metal. I also learned that it doesn't seem to matter how good your tack is, and no matter how square you have tacked things up, you can still have a distorted joint.

Before I was lured into two mental traps, the first was to believe that if I jigged my corner perfect AND solid and let it cool in the jig it would not distort. This was wrong... The jig did help to reduce the distortion though.

The second trap I fell into was to tack up the pieces, and adjusting to square as I tacked. It doesn't matter how good your tack is, the actual bead will still cause distortion...

End result of what I learned is that the speed that you lay the bead down, AND the order and direction that you lay the beads down is critical. And the slower you lay the bead the more critical the order and direction is...

Now for me to start putting it into practice...

Leejohnson1313 wrote:

michialt wrote:I spent some time working/experimenting today, and I know the source of my twisting. It was coming from the order I finished the welds. I actually saw it happening.

Having seen the metal contracting as it cooled I now understand whats going on.

michialt -- perhaps others can gain from your problem solving success if you can provide more details on what you observed, how you reached conclusions, what corrective action(s) you are taking & what the results are...

Regards,
Lee