Calling all you crack experts. No, not that kind of crack.

By Kix

Date 04-16-2009 14:09

Material is a 10 gauge, A572 GR50 TPPO (tempered pass pickled in oil). Process used is GMAW-S, .035 ER80S-D2 wire and 90/10 gas. Problem is, we have had about 6 cracks now on 6 different hydraulic tanks. Some are brand new, never been in service. Cracks are found at the end of the weld. There is a baffle inside the tank that gets 2 fillet welds. You can see where these two fillet welds are on the outside of the tank from where they dimple the steel. The cracks are found in the HAZ just at the end of the dimple.
The crack is so thin and small, that it can barely be seen with the naked eye when no hydraulic fluid is leaking out. The weld on the inside is smaller then an 1/8" fillet. The crack runs almost perpendicular to the weld, just a few degrees off of perpendicular. The crack is about 3/4" long.
So thoughts and opinions are welcome. I'm wondering if the wire might be causing this, but I still think the steel would be durable enough to take the shrinkage. The welder is supposed to be using an ER70S-3 wire, but has apparently been using the ladder shop wire instead of the right wire. I need to go over how to read the wire classifications with him now. ;-)

By ravi theCobra (**)

Date 04-16-2009 16:38

You might be having a problem with the cold rolling. Look at 572 Gr 50 and it is only a hot rolled product -

By hogan

Date 04-16-2009 16:46

I'm no expert but the first thing I thought of was using an 80K wire, short circuiting, on A572 that thin, with that small of a weld may be cooling to quickly. Ant chance of doing a macro on a test piece (and possibly WFMT)?

By Kix

Date 04-16-2009 17:41

I'm working on getting some scrap pieces of this material and doing some tests of my own to get to the bottom of this. The company just wants to test the tank better for leaks and repair them as they come up. I want to know why this is happening so none make it out into the field and crack there. Apparently some did make it to the field and cracked there.

By defaced (**)

Date 04-17-2009 13:09

Does the crack follow the rolling direction of the plate? If so, represent that, along with the level of restraint when you do your tests.

By Kix

Date 04-17-2009 13:46

I just don't see 12 and14 gauge A572 gr50 cracking for any reason. You can fold it like a piece of paper back and forth on top of itself and you woun't see any sign of failure.

By ravi theCobra (**)

Date 04-16-2009 23:47

was that 572 Gr 50 you actually got ? What did the test report say ?

There is no referenct to A 572 Gr 50 TPPO within the ASTM spec

By Kix

Date 04-17-2009 12:19

Yeah, it was A572 gr50 TPPO. I've been welding up sample pieces and destroying them to try and replicate what might be happening, but to no avail. I'm stumped.

By 803056

Date 04-17-2009 16:49 Edited 04-17-2009 17:01

Are you really using short circuiting transfer?

Try undermatching the filler metal. Something has to give as the weld solidifies and cools down to ambient temperature. The higher strength filler metal forces the base metal to act as a weak link in your system, it fails before the weld deposit yields due to the residual stresses developed by the welding operation.

Try a E70S-3 filler metal that is called for. The weld should yield while the weld cools rather than cracking the base metal.

A small scale experiment doesn't replicate the rigidity offered by the actual tank you are building.

Crack result from tensile forces that exceed the tensile strength of the weld or base metal. The residual stress from welding is always in tension, so you have to allow one of the materials, i.e., the weld or the base metal to acommodate the stresses. In your case, the transverse cracks would indicate the weld is stronger than the base metal, thus the base metal, lacking sufficient ductility to accommodate the contraction and stress, simply cracks.

Just a thought after reading your other post, how are you removing the oil from the material before welding? If the oil isn't removed, you could possibly be increasing the carbon content of the opposite side of the plate and the weld by the presence of the residual oil. After all, the oil is a hydrocarbon which can contribute hydrogen and carbon to the weld and possibly diffuse some carbon into the base metal if it is heated to a very high temperature on the opposite side of the plate. The increased carbon increases the hardness and sensitivity to cracking as does any diffusible hydrogen.

Some photos would help us.

Stronger is not always better!

Best regards - Al

By Kix

Date 04-17-2009 18:35 Edited 04-17-2009 18:39

I agree that stronger is not always better. The welder is supposed to be using the 70S-3 wire. I did the samples with the 80S-d2 wire just to keep everything the same and I couldn't see anything wanting to start happening after some severe punishment. About the rigidity of the tank, it's made from 14 gauge a572 gr 50, 5' tall by 3' wide with a baffle runing down the center, so it's not real rigid, but I know what you're getting at. The oil is wiped off, but there is still residue on the sheet metal. So ya think I may have some surface hardening going on here. Very well could be, but I just think that it would take more then that to make such a malleable material hard enoug to crack like this, don't ya think. After all this though, it's really the only thing I have to go on. Thanks for the input Al! I'll look into having the guys watch out for the oil residue left on the sheets.

By GRoberts (***)

Date 04-18-2009 13:53

I think you may have answered your own question right there- "but there is still residue on the sheet metal"

If you material is cold rolled, the tensile strength could be significantly above the minimum.

You are welding short-circuit, and a very small weld so it cools off quickly

You have a source of hydrogen

You have small cracks in the HAZ

You can't replicate it on less restrained (and possibly better cleaned) samples.

You are using a higher strength than required filler metal

Sounds like classic symptoms for hydrogen cracking to me. Don't let the sheet metal throw you off, I have heard of similar situations in the auto industry.