A question of welding to tool steel; what metal, how to heat treat...

I have a project on the table, of making a new axle for an older skid-steer tractor (similar to a Bobcat.)

The axle is a short stub, about 7" long, and splined. The original snapped at a machined step, where the shaft was reduced for an oil seal. Replacement shafts are apparently unavailable or fabulously expensive if you can find one, so we're going to make a new one.

The machining is simple and straightforward. The only trick is there's a wide flat plate welded to the outer end- the axle looks like a giant thumbtack.

That plate is what the wheel bolts to- it appears to be mild steel, about 3/8" thick, and is MIG welded to the axle shaft. We are presuming to either bore out the old shaft and reuse the drive plate, or just fab up a new plate out of plain hot-rolled.

I have two questions: First, I had planned on using 4140/4130 to remake the shaft, because it's easily available locally. The max OD of the shaft is about 1.5", and about 1.200" where it broke. Would there be a better choice for material, without getting into something really exotic?

Second, we're presuming the shaft will need to be heat-treated after machining, to make it as strong as possible. I know the heat-treating won't make a difference on the mild steel, but how about the welded area? Will this be a problem? Should I use a special filler rod/wire? (Meaning something other than the usual mild steel wires/rods?)

Besides the cost of machining, it's time-consuming to install, so naturally neither I nor my customer want a part that's just going to break or fail- neither one of us can really afford to "keep trying 'til we get it right".

Doc.

There may be others with different opinions but what I would do is use 4140 and make a new plate that has a hole in it big enough to pass the shaft through. I would mount the flange on a faceplate and turn a decent bevel on the outboard side. I would also bevel the end of the shaft slightly. Pass the shaft through the flange and make sure it is good and square. Tack it in several places and check again. I personally would either wrap the shaft in wet rags or set it in a bucket of water right up to the flange. If you get a coffee can or something just the right size then you could support the whole thing with water right up to the flange which the heat is not going to affect. The bevels will help you to get good penetration and keep the weld below the surface so it doesn't get in the way of the wheel bolting on.

OK, it's a pain, but doable.

You need to use low hydrogen electrodes and there is a strong chance of embrittlement so you need to do a PWHT. If PWHT is done, a preheat and interpass temperature of 400°F minimum is suggested (that's the low end, I'd go 500°F), followed by transfer into furnace immediately after welding for PWHT at 1100-1250°F (593-677°C) followed by slow cooling.

Hardening is accomplished by heating to 1550-1600°F, oil quenching and tempering at a temperature which depends on hardness and other properties required. A 1000°F temper results in a hardness of about 35 HRC whereas a 1200°F temper results in a hardness of about 25 HRC with increased ductility.

Not easy! Immediately means right now, not as soon as you can get to it.
JR

I woukd start by testing the hardness of the broken shaft and the material. If you can't do that you are guessing at how strong it was to begin with. You might find it was just some cheap stuff and soft to boot. If thats the case you kight be doing really well just making it out of some 4130/4140 prehard. If that were the case you might be just fine welding it up and slapping it in with no further HT.

I've built a few narrowed full floating rear axles for rock buggies on the cheap. Dana 70/80 and Ford sterling axles.

I turn the shaft round where it will be cut off, cut off the flange, cut the shaft to length and center drill the cut off end, drill/bore/bevel the flange, preheat a few hundred degrees and weld the flange back on with 7018 or 70 series mig wire. Put the shaft back in the lathe and skim the mounting surface on the flange.

My question is how big is the plate that mounts the wheel?
Reason i am asking is that i just bought some forged axle blanks with integral flange made form 300M for a pretty low cost (about $130.00 i believe)
Can't buy the material for that.
Cheers Ross

Only way to go IMHO.

I suspect if you took out a broken welded axle, it was some of Bubba's "repair" work

AlfaGTA said:

My question is how big is the plate that mounts the wheel?
Reason i am asking is that i just bought some forged axle blanks with integral flange made form 300M for a pretty low cost (about $130.00 i believe)
Can't buy the material for that.
Cheers Ross

Click to expand...

what machine? i have an axel or two from a JD.

DocsMachine said:

I have a project on the table, of making a new axle for an older skid-steer tractor (similar to a Bobcat.)

The axle is a short stub, about 7" long, and splined. The original snapped at a machined step, where the shaft was reduced for an oil seal. Replacement shafts are apparently unavailable or fabulously expensive if you can find one, so we're going to make a new one.

The machining is simple and straightforward. The only trick is there's a wide flat plate welded to the outer end- the axle looks like a giant thumbtack.

That plate is what the wheel bolts to- it appears to be mild steel, about 3/8" thick, and is MIG welded to the axle shaft. We are presuming to either bore out the old shaft and reuse the drive plate, or just fab up a new plate out of plain hot-rolled.

I have two questions: First, I had planned on using 4140/4130 to remake the shaft, because it's easily available locally. The max OD of the shaft is about 1.5", and about 1.200" where it broke. Would there be a better choice for material, without getting into something really exotic?

Second, we're presuming the shaft will need to be heat-treated after machining, to make it as strong as possible. I know the heat-treating won't make a difference on the mild steel, but how about the welded area? Will this be a problem? Should I use a special filler rod/wire? (Meaning something other than the usual mild steel wires/rods?)

Besides the cost of machining, it's time-consuming to install, so naturally neither I nor my customer want a part that's just going to break or fail- neither one of us can really afford to "keep trying 'til we get it right".

Doc.

Click to expand...

Pretty routine.

High reliability welding of alloy steels like 4140 to mild steel is a common problem that requires a little research. The things to consider are: weld prep, fit up, metallurgucal condition of the materials (annealed etc), filler metal, sequence, pre-heat/interpass/post-heat, and subsequent heat-treat.

I'm experienced at alloy steel welding but I'm not a welding engineer. My little card index of methods is based on engineered procedures. I won't add my procedure to the list because I'd be adding to the confusion. You don't know me from Adam so how can you tell from my well written, highly specific, plausible sounding response that I'm competent?

Ask a bunch of people how to... and you get nearly as many answes as there are people. High confidence welds require proven methods and good technique. It aint rocket science but it aint a consensus decision either. We have a great bunch of people here on PM but few of them are welding engineers.

A welding procedure is like a cake recipe. You don't add a can of pork and beans even if you like pork a and beans. You don't leave out the eggs because you don't like chickens. You don't use part of one recipe and part of another. Follow a single recipe - the full recipe - and you'll make a good cake. Same with a good welding prodecure right down to the last post heat step.

So: use welding procedures and filler metals reccommended by competent authority.

Here's my advice: go to your local welding supply. They have handbooks and weld engineers on call to help you with specific welding problems based on the reesources you have available. They offer this free service just to sell you a couple pounds of filler metal. They'll tell you filler metal, Volts, Amps, heat control, sequence, root pass treatment, distortion control, weld prep etc, everything. Use them. I did for all the years I was welding; and I did some fussy work; a few of which did not require a full rod to complete.

Sounds like a rather puny axle considering the flex stress that occurs on the wheels of a skid steer.

Do what you can to re-engineer the axle to incorporate generous sized fillets at every shoulder.

Ordinary weld deposits do not heat treat with much beneficial effect because they are low carbon steel. However, the heat affected zone (HAZ) will heat treat, and probably what you want to do is get the HAZ to temper down from extreme hardness that can occur from welding on an alloy steel. But prevention of a super hard HAZ is better, that's why you want the high preheat temperature, because you could get a crack started even before you're done the welding, and you'll never know it, although you should listen for any 'ting' sounds that indicate a cracking event.

I'd usually heat shrink the fit to the new plate, mostly to ensure that the shaft stays square for welding, but it will likely still pull a little. However, welding on one side of the plate will bow it, and open up the bore a little bit on the other side. That indicates the high stress on the joint just from welding. Relax those stresses with the post weld anneal.

The part failed once, so it will fail again. You can't fix what wasn't engineered to last forever. But from my experience, large fillets are the key to making a stressed part live as long as it can, if you have done what you can to prevent the super hard HAZ.

Edit: make a large knob on the end of your 4140 axle to incorporate the proper filleting and such, and make your new plate with a large diameter hole in it and weld somewhere on a 2.5 to 3" diameter circle, instead of around a 1.5" root.

Welding hardenable alloy steels isn't trivial. I think it would be a lot easier just to hog out the hole thing from a solid chunk of prehardened 4140HT material. You say the plate is 3/8" thick - what are the other dimensions?

Personally, I would leave your flange fat when you cut it out. From some bitter experience, the warpage can be suprising when mating tool steel to plain jane plate steel. After your done with the welding, recheck it in the lathe and turn it all true.

You need to use low hydrogen electrodes and there is a strong chance of embrittlement so you need to do a PWHT. If PWHT is done, a preheat and interpass temperature of 400°F minimum is suggested (that's the low end, I'd go 500°F), followed by transfer into furnace immediately after welding for PWHT at 1100-1250°F (593-677°C) followed by slow cooling.

Hardening is accomplished by heating to 1550-1600°F, oil quenching and tempering at a temperature which depends on hardness and other properties required. A 1000°F temper results in a hardness of about 35 HRC whereas a 1200°F temper results in a hardness of about 25 HRC with increased ductility.

Click to expand...

-Thank you! That's exactly the kind of information I was hoping for.

Unfortunately, as I have no heat treating furnace, I may not be able to do this project. The preliminary plan was to turn the shaft, weld the flange, then mail it off to a professional heat-treater. I could do things like pack it in insulation to let it cool slow, but I have no way to accurately raise the temperature (apart from a weed burner or the like) nor to hold it there for a controlled cool.

Which, I guess was one of my questions: Can I, somehow, weld it (I have access to MIG, TIG and stick) and manage to let to cool properly in order that I can ship it off to a professional heat-treating shop? If the "transfer straight to a furnace" part is a hard requirement, I may have to pass on the job.

I've built a few narrowed full floating rear axles for rock buggies on the cheap. [snip]

Click to expand...

-Your suggestion still involves welding, but you do give me a good idea. I might have to check and see if maybe we can cut down an old car or truck axle- turn it as one piece, rather than dealing with the weld.

My question is how big is the plate that mounts the wheel?

Click to expand...

-Again a good idea. The problem being the drive flange is (it's not in front of me at the moment) about 8-10" in diameter. The "blank axle" might be an option, but as above, I might be able to hit a wrecking yard and look for a truck axle with enough meat we can shave it down.

I suspect if you took out a broken welded axle, it was some of Bubba's "repair" work

Click to expand...

-We have the axle out. The fracture was at a 'step' in the unmodified factory shaft, well away from the weld. As others have noted, the point of failure had, in my opinion, a much too sharp shoulder, and we will, of course, be giving the replacement a more generous radius there. The weld is clearly also factory- it's a single circular and very smooth MIG pass, almost certainly done robotically.

what machine? i have an axel or two from a JD.

Click to expand...

-Not sure, off the top of my head. I can find out, of course, but all I can say right this second is it's a "skid steer" tractor, similar to, but not, a Bobcat.

Here's my advice: go to your local welding supply.

Click to expand...

-I did. Most of the welding industry around here is oilfield based, which means it's 95% pipe welding. The guys at the local suppliers are extremely knowledgeable... in that area. You want to stick mild steel together, overhead, in the wind, and have it pass X-ray? No sweat. You want to TIG stainless? Well, we have some stainless rod around here somewhere, go give that a try, lemme know what happens.

If it were CM-to-CM and heat-treat and/or ultimate strength wasn't an issue? That's easy. If it were all mild steel or all stainless steel, no problem. CM-to-mild, then heat-treated and needing it to be stronger than the original part? That's getting tricky, as JRI's instructions well show.

Sounds like a rather puny axle considering the flex stress that occurs on the wheels of a skid steer

Click to expand...

-I agree. But the thing is apparently some 30-35 years old, and only recently broke, so it couldn't be too weak. Like a regular car axle, though, there's a stout bearing just inboard of the wheel flange- I actually suspect the axle broke due to HP, not flex.

I'd usually heat shrink the fit to the new plate [...]

Click to expand...

-We were looking into other options, like a keyed shaft, press fit, smaller flange to which the larger flange bolts, etc. Keys and press fits, however, are nonstarters- the flange is only 3/8" plate. I suspect I can extend both shaft and plate outward, similar to the old 2-piece flanged axles, but there's considerably more work and time, and I'm not sure what the customer's budget is like. He may not want to put a $1,000 axle into this thing.

I think it would be a lot easier just to hog out the hole thing from a solid chunk of prehardened 4140HT material. You say the plate is 3/8" thick - what are the other dimensions?

Click to expand...

-Considered that too. The drive flange is 8"-10" in diameter, and the axle shaft is about 8" long. It wouldn't be godawful if I had the big Springfield up and running, but trying to take off that much meat with a 10" Sheldon would take weeks.

Without the flange, the shaft is 1.5" bar stock, gets two relatively shallow steps, some splines, and a threaded stub on the end for a retaining nut. Easy even with wimp machines.

Personally, I would leave your flange fat when you cut it out. From some bitter experience, the warpage can be suprising when mating tool steel to plain jane plate steel.

Click to expand...

-I'd planned on giving the flange at least a light skim cut after it was all said and done, but really, minor warpage here isn't a huge issue- it's a skid steer, top speed is probably 10-12 mph.

Doc.

Post weld heat treat can be done with a weed burner,its not text book or good practice, but it is very possible, key things being able to measure the parts temp and making sure you get it hot enough all the way through.

Then the other key thing is a slow cool. So long as the cool down is even your normally ok on all but the most specialist of alloys. The key is that the whole piece loses heat some what evenly so there’s minimal induced cooling stresses.

I love 8620 for low-speed off-road equipment axles--tough on the inside, hard on the outside wear surface, after the easy-peasy hardening routine. It's often used for this. Cheap too--Speedy metals has it, for one supplier.

The old axle made of unknownium snapped at a stress riser, right? Whatever you make the new one from, try a little radius at the diameter steps, if the seal design allows.

The plate could probably be made of anything, it didn't fail, the axle did. 8620 and mild steel are an easy weld.

JR told you how to do it the right way to get a high confidence weld. There is a significant chance the cost of the part will exceed the value of the skid steer after reassembly.

You haven't told us some details that are significant. Describe the broken surface. If the break was fatigue failure and started from a stress riser there will be half moons centered around the initial crack, and then the area that broke catastrophically will have a rougher and fresh break appearance. What is the ratio of the area that broke catastrophically compared to the area that cracked progressively? What is the grain size of the broken area? Compare it to sandpaper roughness. Fine grained on the outside .02" to .06" and coarse grained in the center the part was case hardened. What does a file test suggest for hardness?

Post some pictures of the broken surfaces if possible. It may be possible to make an axle without sending it out for heat treatment.

Sea Farmer, please give us some details of how you easy-peasy case harden 8620.

gbent said:

Sea Farmer, please give us some details of how you easy-peasy case harden 8620.

Click to expand...

Depending on the size of the part, I either put it in a small cast-iron box, or wrap it in SS foil surrounded by Kasenit. Kasenit is apparently no longer available, Cherry Red is the substitute. I bet crushed charcoal briquets would work too, but still have a couple of cans of Kasenit left.

Put it in either the heat-treat oven or forge (I have both) get to around 1700 degrees F, soak at temp around 20 mins, quench in oil. When you're done a file will skip over the surface.

I always temper afterwards but don't know if that's strictly necessary.

One of my old cans of Kasenit came with a chart showing how long to soak to obtain deeper hardness, but I no longer have the chart. 20 mins seems to be enough.

I'm sure a small part could be heated with a torch if you don't have either a HT oven or a forge. 1700 is about when orange color starts to turn yellow. Then roll the part in a pan of carburizing medium of your choice, let it cake on, then quench. That would probably result in a much thinner and perhaps uneven case-harden.

I got my cast-iron box, slightly larger than a brick, at the hardware store in the BBQ section, not expensive at all. It was made to hold wood chips for smoking, so the box is solid but the lid is perforated to let the smoke out. I just put a piece of SS foil under the lid to block the holes.

Looking over what I've just written I suppose its only easy-peasy if you have a forge or HT oven--but if you don't have a source of heat, you can't HT anything anyway. Its the technique thats easy I guess.

johnoder said:

Only way to go IMHO.

I suspect if you took out a broken welded axle, it was some of Bubba's "repair" work

Click to expand...

I have (3) different skid loaders, and (2) of them, the older ones, both have welded axles,
These are factory made this way (Bobcat and NewHolland)

There appears to be a plug weld holding the flange to the main axle, I assume it's pressed
on, and the weld in down in the bore of the flange, I have not had a good look at the fillet
xistion to the flange, I did have the NewHolland one out for a seal change, BTW it was
made in about 1982-86 ish.

DocsMachine said:

Unfortunately, as I have no heat treating furnace,

Click to expand...

My stuff is taken from our weld procedures. That used to be my department.

If you absolutely can not do a PWHT on the part (this is a work around, not a recommended procedure), immediately after welding, raise the temp to 600°F and hold for an hour per inch of cross section. The part can then be slow cooled and then shipped to a heat treater. The part should be kept rotating during this time so that the flange cools evenly.
JR

BTW, I don't have the $ that DD does, I only have 1 skid steer. I do know that on the early Bobcats, they had a taper with key on the end of the axle then axle flange was fitted and then welded.

JRIowa said:

BTW, I don't have the $ that DD does, I only have 1 skid steer.

Click to expand...

I think its the other way round as to the dollars. 3 old skid steers has got to rank up there with a boat as to their ability to suck up money.