New Welding Thread ?'s for beginners...MIG

[JoshStratton]

Josh,
That's why you want to get even spaced tacks all the way around and progressively close it so that the metal pieces don't move (as much) from the solidifying puddle.

jp

Oh boy. Now that is something I would have never thought of. No wonder body shop guys make 50+ bucks an hour.

A good rule to live by is the weld should be as big (leg length) as the base metal is thick (thinnest member).

This rule would not work for sheetmetal, would it?....god I hope not. Something like 18 gauge would be pretty hard for me to do.

After seeing your definitions, I totally realize how ugly my welds are. They are too wide and have a lot of undercutting. Is this something that could possibly lead to a dangerous situation? I plan on using my cart so I can get under it and work. I had the whole car on it, minus the front sub, at one point. It has only gotten MUCH lighter since.

[Matt@RFR]

This rule would not work for sheetmetal, would it?....god I hope not. Something like 18 gauge would be pretty hard for me to do.

After seeing your definitions, I totally realize how ugly my welds are. They are too wide and have a lot of undercutting. Is this something that could possibly lead to a dangerous situation? I plan on using my cart so I can get under it and work. I had the whole car on it, minus the front sub, at one point. It has only gotten MUCH lighter since.

No, that rule of thumb pertains to any joint besides a butt joint. Butt welds don't have legs, so you're left judging the weld "size" solely by penetration.

There's two downsides to undercutting. 1) It creates a stress riser. Stress risers usually only fail when under cyclic loading/vibration. 2) At the deepest part of the undercut, if you can picture this, the metal is thinner there. So basically, if the weldment is over engineered with material selection, it will probably be fine.

I'll leave the decision up to you, as to whether it's safe or not. Too much liability involved for anyone but you to make that call.

John, that's a tough question to answer, since it depends so much on the end use of the weldment, and how it was engineered (additional gusseting, supporting elements of the weldment, etc.). But, generally speaking, my quickie drawing depicts good penetration.

You can test your own welds: Section the weld, polish to a semi-shine or so, and etch it with muratic acid, or equivelant. That will show you exactly what's in my drawing.