Heat treating gear out of 4140 Steel
Iím looking for some recommendations for heat treating a 5.3 inch diameter gear out of 4140 steel. OkÖwithout going into too much detail, the basic problem this. I want to repair a large PTO spur gear from a farm tractor. The original gear had some broken teeth, which then lost contact ratio. The gear moves on a spline shaft and mates with a stationary pinion.
I machined off all the teeth from the original gear and saved the hub, which still has the interior spline cut out in it. My plan is to machine a new gear and press fit it over the original hub so not to have to re-machine a whole new interior spline for it. My main question is should I heat treat the gear overlay before I cut the gear teeth into it, or should I do it afterwards?
If I heat treat the blank for the gear overlay first it will be difficult to machine, since I will be using a HSS cutter (the gear cutter is only in HSS). However, I am certainly going to get a correct tooth profile, since any distortion from the heat treatment will be machined out.
If I heat treat after I machine the gear overlay Iím now restricted to using only two methods: induction hardening of the teeth, or flame hardening of the teeth. I really donít know the original composition of the gear, so I donít want to touch the hub. I cannot use my furnace, since I will have to heat treat the entire gear including the hub, which I do not want to do.
If I flame harden the teeth I have no means of measuring or controlling the actual temperature. Even if I could, I am now left with some distortion in the teeth, which I would have to grind back to the correct tooth profile, which I really donít want to deal with either.
My original thought is to rough machine most of the material, heat treat, and then finish machining.
Any thoughts or suggestions?
Make the gear. Install it on the existing hub. Use a shallow shoulder fit up so the gear doen't wobble. Allow 0.001" per inch interferance. Do not press it to seat it. Use differentail heat. Heat the gear to 400 F and drop it on the hub. Drll and tap for a half dozen setscrews equally paced right through the joint between the gear and hub. Intall in in the lathe. spin it a couple hundred RPM for the hardening process.
Here''s the tricky part: Drape the machine aganst water. Heat the teeth with an oxy-acetelene rosebud tip to red heat. Careful, it's easy to overheat the tooth tips. Heat until the gullets are red. Quench with a fine water mist from a hose nozzle that mimiks an oil quench. It should go from red to black in 3 to 5 seconds. Continue until the gear not longer sizzles. Practice on a scrap piece of like dimensions. Polish the face, re-heat, and draw to a vivid to pale blue. After all hot work is complete, remove the setcrews and install with LokTite.
Here's where most people screw-up in replaceing broken gears: they don't replace the mating gear. If teeth broke off one gear chances are very high the mating gear teeth are damaged or deformed. You hould also examne the shafting, bearings, and housing. Maybe other gears in the train.
A broken gear in a well designed train is very serious. Engineers design gear trains so all parts are just strong enough to meet predictable overloads plus a safety factor. Sometimes they'll design a "fuse factor" in a readily replacible part so it preferentially fails if the train is overloaded to the breaking point - but not always.
Once a gear fails in a train that part of the mechanism often causes recurrng problems. Final drives in heavy equipment seem to be particulary vulnerable. I once diagnosed the final drive case for a biggish dozer removing all bearing races and found two distorted bores and a bent haft that supported the same gear pair that failed regularly. The mechanic bored and sleeved etc, the shaft was replaced, and the dozer went on for years afterwards without trouble. How do I know it lasted for years? Whenever I saw the owner in the fastener store or the hose outfit he would grin, pound my back, and brag on me. For years. Good ol' Rod Bowstead.
Vacuum deposited coatings are a low temperature alternative to heat treatment. Depending on the loads involved it may well be perfectly adequate on a plain 4140 gear but usually the process is applied to more sophisticated materials to get decent life out of a gear that would be overloaded if given merely a simple heat treatment. Oerlikon-Balzers, Oerlikon Balzers Coating Services - Homepage, do this sort of stuff and seem happy to advise on the best approach. I've had good service for one and few off items from their UK arm.
Bit spendy tho' and you will have to wait whilst they assemble up a vacuum chamber full for one process. Advantage is that you will be sure that there is no distortion and you will know exactly what you have. Doesn't take very much distortion to send tooth loads through the roof.
Don't make it out of 4140. Use 8620, and get it professionally carburized and heat treated. 8620 is great stuff, quite stable in heat treat, yet has that tough chewy center that we all like I'd be afraid that water quenched 4140 is too brittle, and oil quenched won't be hard enough.
A few suggestions from my experience designing heat-treated gears....
If the application is light-duty enough, you could make the gear out of 4140 Pre-hard and not heat treat the teeth.
I second getting it professionally heat treated if you go the heat-treat route. You are looking at about $125-$150 depending on the minimum charge of the heat-treater.
Induction hardening the teeth will result in very little distortion. I doubt this application is critical enough that the teeth will need to be ground or shaved after hardening. Depending on the diametral pitch of the gear, the teeth will probably end up hardened all the way through. This is my recommended hardening method.
The teeth can also be nitrided. This results in a shallower hardness depth. However, the gear will be hardened all over. We usually only nitride larger diameter gears that are difficult to induction harden.
8620 is a great material to make the gear out of, but you will probably end up with a gear that is a lot stronger than the rest of the gears. Also, you cant induction harden 8620 due to the .20% carbon content so the teeth are more likely to distort during carburizing.
Where I work the standard spec for a high performance gear is as follows:
Material: 4140/4150 Steel preheat-treated 28-32 HRC
Induction harded teeth 48-53 HRc .040 min depth.
We only use 8620 for extreme-duty planetary gears.
now that we have some other "nope my way" s under our belt
i say screw the 4140
make it out of D2 and use as machined
how ' bout that?
Thanks for all your suggestions. Yes...the original pinion looks good to me. Acutually, that's how I determined the module for the original gear.
Unfortunately, I already have 4140 stock available. So from an ecomonic point of view I prefer to use that material, since it's already available.
I am a little uncertain about using visual color as an indicator for temperature. Everyone's visual imagry of color is slightly different. A rosebud color, as viewed by two different peiople, can be as much as 170 dec C off (enough to be outside of the critical temperature).
I do like the idea of practicing on a scrap piece before making the actual part. Even 4140 has slighly different properties from different manufacturers.
Is the tractor truly so rare that used/aftermarket/oem gears are no longer available?
If I was choosing to make a gear to practice gear cutting I would not make a PTO gear. The fact your original is damaged suggests that the factory units weren't strong enough for the engine power.
The PTO gear train of a tractor will see full developed engine horsepower and if somebody tweaked the injector screw the gears could get more power than the design level. There are a lot of tractors that can stand a turned up pump on the pulling gears because of the inability to put that power to the ground. The PTO gears don't have the luxury of the cushion of the tires and die like yours.
Yes...that was the first thing I did was look for a replacement part as oppose to making a new one. I'm not an expert on tractors, but it's my understanding from the owner that this tractor was Japanese built and the brand name is 'White'. I believe this tractor was manufactured around the 80's.
I like Forrest's idea of making one gear slightly weaker ('fuse factor' he calls it), thereby limiting the damage if there is an overload problem.
Normally, this is done by adjusting the tooth thickness of the pinion and gear by doing an addendum modification. However, in this case I'm limited by the current geometry of the original design and also the limited equipment I have. I do know that the original gear is extremely hard ~Rc 59. My though is to make this gear ~Rc 45-50; less hard than the pinion.
Thanks again for all your suggestions.
Well, there's "hard" and then there's "strong". Also, RC59 on the outside doesn't tell you anything about the RC's on the inside. For this application (low-speed, shock loads), I should think a decent RC thru-n-thru would be better than high RC skin over soft core. To that end, and also to save some $$'s, you might consider a 4340HT blank with no post-machining HT. You get RC-35 or so thru-n-thru, and a lot of toughness. You can machine this mat'l with HSS involute gear cutters no problem.
Originally Posted by wepllc
Just a thought ...