Bad Welds?

[Henro]

What I wonder about is that the OP's photos do not seem to show fresh shiny metal...looks like the surfaces are rusty brown. I could be missing something here for sure. What is the message there?

Something was broken for a while and just noticed?

My inexperienced eye is unable to determine if the weld pulled off the base metal, or if the base metal pulled apart.

Not voicing an opinion, because I don't have one. But I am interested...

 

[McMXi]

Looking at the OP 's picture I have to wonder about the design of welding that tube to the fish plate.

Someone here has a signature that states "science doesn't care what you believe". I like that aspect of math and engineering.

I've ordered a gooseneck trailer that touts "cross members every 16" that pierce the web of the I-beam" and this is a common approach with more expensive trailers. The other two methods are butt welding the cross members to the web of the I-beam, or stacking the cross members on top of the flange of the I-beam. Intuitively, the flex in the frame would be the same for the pierced or butt welded configuration if the cross members are equivalent, but piercing the web means that it isn't just a butt weld that is resisting shearing forces, since the channel is keyed into the web in a pierced design. This is not the case for a tube cross member as is found on tractor loaders. Piercing in the case of a loader simply allows for twice the weld length which means more resistance to shearing forces i.e. torque.

I found a website for a company run by an engineer who designs, manufacturers and sells trailers for hauling shipping containers. He is adamantly against the pierced beam design and much prefers the stacked cross member design. He claims that piercing the I-beam weakens the design. Is he right? Does he have other reasons for using a stacked cross member rather than a pierced cross member design?

Sometimes there are other aspects to consider when designing or building a trailer (or a loader) such as deck height, that makes one process more desirable than another. Perhaps a 15% reduction in strength is a reasonable trade-off for a significant improvement in functionality, or perceived improvement. It's a complicated world for sure.

 

[McMXi]

What I wonder about is that the OP's photos do not seem to show fresh shiny metal...looks like the surfaces are rusty brown. I could be missing something here for sure. What is the message there?

I think you're absolutely right. The weld cracked and water was able to get into the interface between the weld and loader arm resulting in corrosion. A common term for this type of corrosion is called crevice corrosion. Eventually there was insufficient strength to resist the loads imposed, and the tube separated from the loader arm.

 

[GreensvilleJay]

re: He claims that piercing the I-beam weakens the design. Is he right?

I'd say no.
Look at TJI floor joists and see how many holes can be put into them ! An I-beam is a solid steel version of a truss, TJI is the wooden version. Maybe simplistic to say, but gee they are the same looking structure.

As for the loader frame failure, my expert said it was a 'bad weld - no penetration'. Since he owns and runs run a megamillion steelfab company with papers to weld nearly everything, I take his word.

 

[McMXi]

As for the loader frame failure, my expert said it was a 'bad weld - no penetration'. Since he owns and runs run a megamillion steelfab company with papers to weld nearly everything, I take his word.

Easy to say when nothing is riding on it. Any welding expert/engineer who's paid to analyze weld failures wouldn't stake their professional license or reputation based on what they see in a photo. There are many possible failure modes, and as much fun as all this speculation is, none of us, and that includes "my expert" know with any certainty why there was a failure.

A proper analysis would at bare minimum involve cutting out some sections through the tube, weld and arm, and then acid etching them to look at the weld metal and parent metal interface, heat affected zone, etc. SEM/XRD analysis of the tube and arm steel would reveal the material properties of the steels used and so on. There's a reason why experts are bought in for this sort of thing.

 

[Chanceywd]

"science doesn't care what you believe".

I have that kind of conversation with my better half. I use my line of "it doesn't work like that" in explaining something she doesn't understand but in a polite way. My other line is" you can't beat math and science".

Getting back to the loader arms I don't have your design skills but have to wonder that loader if say you were using one side of the bucket hard. To try and cut or clear a ditch it would be stressing there at that point for sure.

Bill

 

[GreensvilleJay]

Easy to say when nothing is riding on it. Any welding expert/engineer who's paid to analyze weld failures wouldn't stake their professional license or reputation based on what they see in a photo. There are many possible failure modes, and as much fun as all this speculation is, none of us, and that includes "my expert" know with any certainty why there was a failure.

A proper analysis would at bare minimum involve cutting out some sections through the tube, weld and arm, and then acid etching them to look at the weld metal and parent metal interface, heat affected zone, etc. SEM/XRD analysis of the tube and arm steel would reveal the material properties of the steels used and so on. There's a reason why experts are bought in for this sort of thing.

my expert didn't look at the photo, he saw my loader in person, then he repaired it. If I'd still been working at STELCO, I'd have hauled it in and analyzed it in the MET &QC lab I worked in.

 

[McMXi]

my expert didn't look at the photo, he saw my loader in person, then he repaired it. If I'd still been working at STELCO, I'd have hauled it in and analyzed it in the MET &QC lab I worked in.

Oh, your loader. I thought you were referring to the one shown by the OP. Carry on.

My comments re weld analysis still stands though. When I was in welding school we cut up some pipe that we'd welded up (ASME 5G). We ground, sanded and polished the cross section, applied an acid and could clearly see the various transitions between parent metal and weld passes along with the heat affected zone (HAZ).

I worked as a welder in Oregon for a year and spent a few months doing the awful task of magnetic particle inspection of forklift parts. Crack had to be drilled out, ground out and welded. Not much fun at all.

 

[GreensvilleJay]

I finally found my pictures....2 before, one after.

 

[Henro]

I finally found my pictures....2 before, one after.

interesting. Your before pictures also do not show any bare fresh metal...Final fix seems to be good though...

 

[GreensvilleJay]

The repair was done before I got to his shop at 8AM, with the SSQA plate, they start at 6:30, so I never got a chance to see HOW he 'worked his magic' to fix it.

 

[Sidekick]

The repair was done before I got to his shop at 8AM, with the SSQA plate, they start at 6:30, so I never got a chance to see HOW he 'worked his magic' to fix it.

Using the word Magic in the same sentence as the word Weld scares me after Joe's thread 🫣. It appears Kubota didn't put enough layers of paint on to reinforce the weld .

 

[McMXi]

The repair was done before I got to his shop at 8AM, with the SSQA plate, they start at 6:30, so I never got a chance to see HOW he 'worked his magic' to fix it.

It looks to me like the fracture was in the heat affected zone in the steel and not the weld itself. Pre and post heating allows hydrogen to escape from the weld and HAZ, and 7018 rods can further reduce hydrogen ingress. Excess hydrogen in the weld can lead to brittleness, delayed cracking, or hydrogen-induced cracking. All possible explanations for the failure shown.

Also, there's no magic here. If I were doing this I'd grind out the existing weld, get the steel back to clean condition, move the tube to where it needs to be, tack it in place, pre-heat the tube and loader arm and make a couple of passes using 7018 rods, and apply some post heat. I'd stand the loader up on one of the arms so that the welding is done in the flat position. Then flip it over to do the other side. Not hard at all.

 

[Scm]

When I have it fixed (hopefully in the next week) I'll try to dig into it more. I dont have the unit here right now to look. But I'm wondering:

Did the frame steel sheer off without the weld breaking off.
(Less likely since the metal is not shiny?)
or
Did the weld break off the frame steel (no penetration)
or
Did it break between two beads

 

[McMXi]

When I have it fixed (hopefully in the next week) I'll try to dig into it more. I dont have the unit here right now to look. But I'm wondering:

Did the frame steel sheer off without the weld breaking off.
(Less likely since the metal is not shiny?)
or
Did the weld break off the frame steel (no penetration)
or
Did it break between two beads

Your failure looks very similar to that shown by @GreensvilleJay. Both look to me like the crack was in the loader arm steel very close to the weld, but it's hard to be sure based on the images. Once the crack develops, water gets in and crevice corrosion ensues until the part fails so it's not surprising that the fracture surface is corroded rather than bright. Fast fractures tend to show as bright regions since they happen fast enough that corrosion doesn't have a chance to take place.

When you get it repaired, I suggest that you discuss pre and post heating with whoever is doing the work. I think that's more important than the type of welding wire or rod, although 7018 would be about as good as it gets.

 

[lynnmor]

When I have it fixed (hopefully in the next week) I'll try to dig into it more. I dont have the unit here right now to look. But I'm wondering:

Did the frame steel sheer off without the weld breaking off.
(Less likely since the metal is not shiny?)
or
Did the weld break off the frame steel (no penetration)
or
Did it break between two beads

My opinion is that nearly all of the welding was done on the pipe and the arm was just standing by watching it happen. I said it before, that shop in Georgia needs to be moved to China.

 

[GreensvilleJay]

robots are only as good as the humans program them......