120v MIG weld on a 3/8" thick (bevel 60deg)

[Sodo]

This is excerpted from a thread that got outta control, and closed.
Plate beveled at 60deg because everybody knows a 120v MIG can't melt far into a butt weld on 3/8" steel.


======= mostly a re-post =====================
This is Weld#2, the middle weld.

Root weld. (1/4" bolt is to show scale)

Third pass

Bent 90deg


Weld#2, it was very strong, had to reef on that 20T press to bend it to the 90-ish deg as shown.
But even though weld #2 was not 100%, it was firkin' strong (would not let trailer parts loose on the highway :laughing::laughing::laughing: )

Today I cut 'coupons' (sort of) so it would be easier to bend them with the 20Ton press. And busted them too. Will post those pics later as time allows.

This was kinda fun, even if rushed for time.

 

[Sodo]

OK heres welds # 1 & 3, split into appx 1" coupons, then bent. I expected that #3 would be the strongest weld, but it broke. Not easily though.






I figured weld 1A would break about the same as 1B, so I bent it the other way instead. It didn't break.

 

[MinnesotaDaveChalmers]

Although the welded area predictably (in this experiment) ended up not as strong as the parent material, it's still an interesting and instructive test.

Good post for discussion in my opinion.

 

[Mark @ Everlast]

Technically, the best weld you had would be a fail according to standards. Not even close. Can't figure the nail out though unless you used it to hang it.
What you don't figure in is repeated loading stress that you'd find on a trailer bouncing down the road and the cracks that would begin to form where you have a pull or tear or some sort of incomplete fusion over time that would eventually work it's way through the metal for a sudden failure. These extreme bend tests are designed to reveal these spots. A bend test isn't designed to illustrate a one time "load failure", but rather is severe enough to indicate potential spots for long term failure. The other thing that is lacking is a face bend. A failure or weakness of one parts even such as this(if n can lead to subsequent "cascade" failures of other parts as well! You only a root bend here. To properly determine strength, you should have bent the face as well, as loads, especially in the case of a trailer, are in both directions.

 

[Mark @ Everlast]

Technically, the best weld you had would be a fail according to standards. Not even close. Can't figure the nail out though unless you used it to hang it.
What you don't figure in is repeated loading stress that you'd find on a trailer bouncing down the road and the cracks that would begin to form where you have a pull or tear or some sort of incomplete fusion over time that would eventually work it's way through the metal for a sudden failure. These extreme bend tests are designed to reveal these spots. A bend test isn't designed to illustrate a one time "load failure", but rather is severe enough to indicate potential spots for long term failure. A failure or weakness of one parts even such as this(if n can lead to subsequent "cascade" failures of other parts as well! You only a root bend here. To properly determine strength, you should have bent the face as on all welds as well, as loads, especially in the case of a trailer, are in both directions.

 

[muddstopper]

Technically, the best weld you had would be a fail according to standards. Not even close..

Please explain exactly why the 120v mig welds on the 3/8 steel would all fail. Is it the Weldor error, or the choice of equipment, or a combination of the two.

 

[muddstopper]

Although the welded area predictably (in this experiment) ended up not as strong as the parent material, it's still an interesting and instructive test.

Good post for discussion in my opinion.

why do you say predictably, its as if you knew the welds would fail, even before the bending took place

 

[Furu]

why do you say predictably, its as if you knew the welds would fail, even before the bending took place

From the pictures posted it could be seen that the welds did not have full penetration and there appeared at least to me to have flaws in the tie in to the base metal on the face as well. I am sure that someone will be able to describe it better than I have.
I too would have liked to see a face bend as I wanted to see what would happen to the weld. It is my understanding that most bend tests go to nearly if not a full 180. It appears that these made about 90 degrees +\- when they broke.

 

[roadhunter]

Please explain exactly why the 120v mig welds on the 3/8 steel would all fail. Is it the Weldor error, or the choice of equipment, or a combination of the two.

I think he is referring to a bend test. He is correct in that sense although the vast majority welds you see are not capable of passing a bend test or dont have 100% penetration. Go look at something like a cattle panel, brush guard, lawnmower deck, etc... and you will find welds that are not perfect in every way. Just sitting her at my desk I found some imperfect welds.

 

[Mark @ Everlast]

Furu,
There was one face bend (on a root bend fail), but I couldn't completely make it out but it did look like some pulls along the toes.


Mudstopper. Technique was at play for sure. I see too thick or passes, too wide of passes, under reinforcement, under penetration, porosity, lack of fusion at the toes etc. Most could be attributed to technique but then again, they could be attributed to the machine. (Every weldOr has blamed the machine at one point or the other in his career!) But as unstructured as the test was, not adhering to traditional bend test standards of grinding the welds flat on both sides,and rounding the edges and the odd nails welded that added some sort of strange element, it is hard to determine the cause but what is undeniable, is they failed even in a softball bend test like this.