Bad Welds?
- Thread starter Scm
- Start date
Since the repairs have also failed, I wonder if your welder ground and or drilled out the cracks before welding or simply welded over the cracks?
You made the comment that you "don't care about looks", but there's often a correlation between how a weld looks and how it functions.
Those “fix” welds have some porosity and likely weren’t done well. As noted, the prep is important and the amount of heat/material added and HOW it is added may need some addition thought. Did he preheat anything?
I just cut rectangular holes in rectangular tubes for smaller tubes to pass through and will say in my case of passing smaller 5/16" wall tubes through larger 5/16" tubes there's more resulting strength than butt-welding, and I would think that in the case of the front-end loader if a smaller round tube was passing through the larger loader arm rectangular tube the strength and rigidity would be increased. Mind you I'm not a tractor or loader engineer, so it's only a guess.
From looking at the initial pictures it looks like poor heat penetration on the loader arm side of the weld (a "cold weld") resulted in a stress fracture. Welding is supposed to be using an arc to heat both parts (the loader arm and cross-tube) to the point they melt and steel from each flows together into a pool that when cool forms one continuous part. The filler wire is to make up for the loss in material that vaporizes due to the electrical arc--it's not glue and doesn't add strength when used as glue; in fact too much weld reduces strength (it's a bit more than worth detailing here, all that is in the AWS book).
I say "looks like a 'cold weld'" because from the pictures it looks like the filler sits on top the cross-tube and loader arm instead of the cross-tube and loader arm being slightly melted away and into the filler; any time I see a "caterpillar" type weld that's usually the case, the filler is forming a convex (bulging out) round ridge mostly on top the donar parts with little burn-in. A weld with good burn-in generally shows material contribution from the parts being welded and is concave (bowed in) or flat across with smooth (not sharp) transitions between the parts and filler. (As a note, sharp transitions usually result in high concentrations of stress at the sharp transition. In this case smooth transitions are a function of sufficient heat to make a good weld and also benefit the weld by not creating a localized high-stress point.)
Also note we MIG weld low-carbon steel plate (and tube, angle, bar, etc) daily and there's no problem with burning through the scale and creating a strong weld, I've posted quite a few photos of heavy welds we've done and burning through mill scale isn't an issue.
Looking at the repair I see a multi-pass weld that looks "pretty good." I don't see porosity, there are a small amount of slight voids but it's not a shielding gass issue. To me it looks like the welder pulled the weld (instead of pushing it) and the welder likely kept a good puddle and constant heat, so all good there. BUT, look at the weld distribution. I'd guess it was welded out of position and there was not enough heat--lots of motion on the part of the welder caused a good looking weld though not a strong weld--it looks like there wasn't enough penetration for good penetration as evidenced by the lack of burn-in. (See how on the left there's a small smooth radius between the filler and the tube, and in some areas the filler sits on top the tube? And on the loader arm there's a sharp transition between the filler and the loader arm? To me, from one picture, I would *guess* there's not much heat penetration. Looking at the steel and looking at the paint, I'm not seeing evidence of melting the cross-tube and loader arm together, rather it looks like the opposite happened with both being unscathed.)
I say this hoping to be helpful, not criticize anyone's work. When I saw the repair I cringed and likely left an :O or
face out of concern for the result, which it seems showed up much sooner than I'd expected.
---
If you look up the word "weld" with the gear icon and enter WI_Hedgehog (or leave me out, whichever) there are other threads on welding that might help. I do have a few pictures in the D-ring thread, though that also doesn't make me an expert--in fact I should be the first to tell you there's a lot more involved than what has been talked about here as certified welders will tell you. And as has been mentioned by McMXi, a picture can only provide some of the information, which is why I say "guess" and "looks like" a bunch, plus I don't have the experience an instructor or certified inspector has, so it's a guess based on experience, but still a guess.
From looking at the initial pictures it looks like poor heat penetration on the loader arm side of the weld (a "cold weld") resulted in a stress fracture. Welding is supposed to be using an arc to heat both parts (the loader arm and cross-tube) to the point they melt and steel from each flows together into a pool that when cool forms one continuous part. The filler wire is to make up for the loss in material that vaporizes due to the electrical arc--it's not glue and doesn't add strength when used as glue; in fact too much weld reduces strength (it's a bit more than worth detailing here, all that is in the AWS book).
I say "looks like a 'cold weld'" because from the pictures it looks like the filler sits on top the cross-tube and loader arm instead of the cross-tube and loader arm being slightly melted away and into the filler; any time I see a "caterpillar" type weld that's usually the case, the filler is forming a convex (bulging out) round ridge mostly on top the donar parts with little burn-in. A weld with good burn-in generally shows material contribution from the parts being welded and is concave (bowed in) or flat across with smooth (not sharp) transitions between the parts and filler. (As a note, sharp transitions usually result in high concentrations of stress at the sharp transition. In this case smooth transitions are a function of sufficient heat to make a good weld and also benefit the weld by not creating a localized high-stress point.)
Also note we MIG weld low-carbon steel plate (and tube, angle, bar, etc) daily and there's no problem with burning through the scale and creating a strong weld, I've posted quite a few photos of heavy welds we've done and burning through mill scale isn't an issue.
Looking at the repair I see a multi-pass weld that looks "pretty good." I don't see porosity, there are a small amount of slight voids but it's not a shielding gass issue. To me it looks like the welder pulled the weld (instead of pushing it) and the welder likely kept a good puddle and constant heat, so all good there. BUT, look at the weld distribution. I'd guess it was welded out of position and there was not enough heat--lots of motion on the part of the welder caused a good looking weld though not a strong weld--it looks like there wasn't enough penetration for good penetration as evidenced by the lack of burn-in. (See how on the left there's a small smooth radius between the filler and the tube, and in some areas the filler sits on top the tube? And on the loader arm there's a sharp transition between the filler and the loader arm? To me, from one picture, I would *guess* there's not much heat penetration. Looking at the steel and looking at the paint, I'm not seeing evidence of melting the cross-tube and loader arm together, rather it looks like the opposite happened with both being unscathed.)
I say this hoping to be helpful, not criticize anyone's work. When I saw the repair I cringed and likely left an :O or
face out of concern for the result, which it seems showed up much sooner than I'd expected.---
If you look up the word "weld" with the gear icon and enter WI_Hedgehog (or leave me out, whichever) there are other threads on welding that might help. I do have a few pictures in the D-ring thread, though that also doesn't make me an expert--in fact I should be the first to tell you there's a lot more involved than what has been talked about here as certified welders will tell you. And as has been mentioned by McMXi, a picture can only provide some of the information, which is why I say "guess" and "looks like" a bunch, plus I don't have the experience an instructor or certified inspector has, so it's a guess based on experience, but still a guess.
Old Machinist
Well-known member
Equipment
Kubota LX3310 cab, JD 4310, NH 575E cab backhoe, JD F725, Swisher 60", etc.
A butt welded torque tube was a shit design to start with. Kubota should eat all of these and replace them with a proper through frame torque tube design.
In the picture your right side of the weld has a crack down the middle of it, an incomplete bridging of the hot metal, Vertical welding is very difficult, if your able to find an experienced welder I'm sure you would be able to make it work. God bless.
The cross-tube is pretty big, a butt-weld should be fine if done properly. I just inspected a machine frame we're putting together and it's all butt-welded tubes with gusseted corners, it'll hold.
Before weld:
Two of the same after weld:
Notice the small yet visible material sacrifice outside the filler (hard to tell from a picture, much easier in-person), fairly flat welds (not like caterpillars sitting on top), and the visible color changes in the tubes from heat--they got hot, but not "too hot" (also hard to tell from a picture).
Also note we've dealt with Chinese steel, and the industry literally prefaces the material with "Chinese" like "Chinese A36" or "Chinese 4140" because China has their own, lower-quality specifications. I am not knocking it nor making some sort of political statement, it's lower-grade material at a lower price, and by "lower-grade" I don't mean "worse," I mean not as strong (like ASTM A36 should have a 36KSI yield strength and HRB ~76, you're not going to get that nor the chemical composition out of Chinese A36) which means it'll twist up more when heat is applied (like machining/grinding) and it won't be as strong, but Kubota is likely checking the materials when they come in and making sure they meet requirements....or at least get close to requirements....
We don't use Chinese steel, we don't like it. I've used Chinese 304 (stainless steel) in personal projects because it's dirt cheap compared to U.S. stainless (and available), but 1 out of 10 parts might have issues that make the part scrap...usually I'm still money ahead, even accounting for the wasted time in replacing the part (not so much if it wrecks a mating part and I have to replace that too), but I've over-designed to accommodate using lower-spec Chinese parts. However, I will not use Asian bolts in tractor projects, that's likely not going to end well...
BUT, you can weld Chinese mild steel (potentially used in the loader) just fine. (Almost always.) Steel is almost 100% iron, welding Chinese steel shouldn't be an issue, strength after welding can be an issue and I personally would worry about it as Kubota is pretty light duty compared to say CAT, but that's not the problem in the pictures from what I can tell, it looks like the weld is at fault in both instances.
Last edited:
Russell King
Well-known member
Lifetime Member
Equipment
L185F, Modern Ag Competitor 4’ shredder, Rhino tiller, rear dirt scoop
@WI_Hedgehog
I am not 100% sure about that chemical list for the A36 material. When I knew more about the actual specifications there were really zero chemical properties in the specification, only the 36ksi yield strength requirement. So the steel may have more carbon in it than that limit you show. Where I worked the heavy fabrication shop we used complained that the carbon content was actually getting above 0.25% and getting up into the mid carbon steel range (0.45%?) and making the welding process more challenging than low carbon steel.
You might discuss that with your steel supplier to be sure what the ASTM A36 specification requirements are today. We were “lazy” and just changed our material specifications to say something less specific like minimum yield of 36ksi and carbon content less than 0.25% (or some number that was the top of low carbon steel allowance. There were alternative specifications that did the same thing we were just tired of changing our material specifications to suit what the steel industry was trying to make, so controlled only what was needed.
I am not 100% sure about that chemical list for the A36 material. When I knew more about the actual specifications there were really zero chemical properties in the specification, only the 36ksi yield strength requirement. So the steel may have more carbon in it than that limit you show. Where I worked the heavy fabrication shop we used complained that the carbon content was actually getting above 0.25% and getting up into the mid carbon steel range (0.45%?) and making the welding process more challenging than low carbon steel.
You might discuss that with your steel supplier to be sure what the ASTM A36 specification requirements are today. We were “lazy” and just changed our material specifications to say something less specific like minimum yield of 36ksi and carbon content less than 0.25% (or some number that was the top of low carbon steel allowance. There were alternative specifications that did the same thing we were just tired of changing our material specifications to suit what the steel industry was trying to make, so controlled only what was needed.
@Russell King
Correct, it depends on the thickness of the steel. Here's a current listing, which is pretty much on-par with the previous chart:
www.theworldmaterial.com
We buy domestic prime whenever possible, every item has a heat # and corresponding mill cert. That eliminates the high-carbon issues, though there was a porosity and lamination issue related to the reduction ratio when 1018/1020 (mined) switched over to A36 (remelt) due to govt. regs., and a serious hard-spot issue due to pushing whole cars into the vat and not mixing it enough so there were areas with chrome that would break cutters, and with the EPA removing lead....
I'd like to see Kubota use Hardox 450, but I'd also like tractor prices to not double, so there's that...
Correct, it depends on the thickness of the steel. Here's a current listing, which is pretty much on-par with the previous chart:
ASTM A36 Steel Properties, Modulus of Elasticity, Yield Strength, Material Density, Hardness & Equivalent
ASTM A36 steel is a widely used mild carbon steel, young's modulus of elasticity, properties, material density, yield strength, hardness, equivalent, pdf, tensile strength
www.theworldmaterial.com
We buy domestic prime whenever possible, every item has a heat # and corresponding mill cert. That eliminates the high-carbon issues, though there was a porosity and lamination issue related to the reduction ratio when 1018/1020 (mined) switched over to A36 (remelt) due to govt. regs., and a serious hard-spot issue due to pushing whole cars into the vat and not mixing it enough so there were areas with chrome that would break cutters, and with the EPA removing lead....
I'd like to see Kubota use Hardox 450, but I'd also like tractor prices to not double, so there's that...