Gear hobbing on a CNC actually the right way to go

[CarbideBob]

The min and max effictice range with this tooth angle sort of scares me here.
How do you plan on checking this? How will your customer check it?
Everything else can be in and this be off.
On axle shafts hobbed or rolled I have fought this darn gauge for hours and hours, entire shifts, three shifts.
Start over, change hobs, racks, centers and try again.
Now you are a bottleneck to production and there is plant somewhere that will not be able to assemble cars
Call in the engineers in the middle of the night. They are not sure what to do and upset to be called in the middle of the night. Maybe just "cheat" the gauge a bit.........
....Seems that it should be easy if all the other dims are in.

I am sure for others this goes easy and clean.. For me it seems if it can go bad it does go bad.

 

[triumph406]

I made 8620 starter shafts for small UAV motors (120cc) years ago

The length and diameter of the shaft portion not disimilar to yours.

They were rough turned with centers at each end, teeth were hobbed, then the shaft was carburized and quenched, the shaft was then ground. Evidently the OP could skip the carburizing.

Maybe for the quenching ensure the shafts are vertical going into the oil to minimize distortion.

Seem to work ok. UAV life expectancy was short, much shorter then the shaft

 

[cnctoolcat]

No way any kind of hobbing or milling machine is cutting a 16Ra finish on splines...that's grinding territory. (Sounds like you're gonna have to educate the customer about splines...)

All hobbing and milling processes will leave teeth marks on the sides and bottom of your splines (dependent on feedrate, rigidity of machine and setup, tightness of machine, cutter runout, etc.).

Especially trying to hob on a CNC lathe...where you're not going to get the finish a dedicated gear hobber can make.

Older examples of gear hobbing machines like the venerable Barber-Colman 6-10 will make great splines, but not a 16 finish. 125 maybe, and that's with a tight machine, along with a perfect cutter, part, and machine setup.

 

[EmGo]

We regularly machine post HT in 15-5 and 4140 so anything <50Rc doesn't phase me. Hob life is really the only unknown to me. I'd assume you could at least get several thousand parts per hob. The teeth are *only* .025 deep or so.

Think of it like using a hss end mill in hardened steel - with a sharp corner that has to stay sharp. The hobs won't break or anything, they just get dull on the edges and there goes your shape and size. HSS hobs hold up good in softer stuff but Rc 45, you'll want carbide.

And then you get to sharpen them, not fun Plus *must* be done correctly or your tooth shape will go to hell.

I was looking all over for this finish requirement, it's not on the print. Who the hell wants 16Ra on the teeth and why ? That's silly. These aren't a roller bearing. Serves no purpose. No one does that.

If you have to, because the customer is stupid, then it's Reishauer time. A reishauer would eat these things up for breakfast, then ask for a cookie. But more $$ and floorspace and hassles. The thing is tho, the turning center attachment certainly isn't going to get that finish either. You'd have to go grinder for that. Maybe the buyer needs some edumacation ?

The min and max effictice range with this tooth angle sort of scares me here.
How do you plan on checking this? How will your customer check it?

Over wires. Even gives you a number. It's only a thou on tooth thickness, at 45* what's that over wires, +/- .001 ? Pretty easy. I thought you Big Boys had the Good Stuff, Bob ? I bet rolling is harder to control size, +/- .001 is not that bad even with antique junk like I always owned.

Should probably get a go gage made to drop over every one, then do wires every ten or twenty or something, depending on how they hold size. If he has to go grinding, can just skip that pointless measuring step

 

[Mr.Chipeater]

Think of it like using a hss end mill in hardened steel - with a sharp corner that has to stay sharp. The hobs won't break or anything, they just get dull on the edges and there goes your shape and size. HSS hobs hold up good in softer stuff but Rc 45, you'll want carbide.

And then you get to sharpen them, not fun Plus *must* be done correctly or your tooth shape will go to hell.

I was looking all over for this finish requirement, it's not on the print. Who the hell wants 16Ra on the teeth and why ? That's silly. These aren't a roller bearing. Serves no purpose. No one does that.

If you have to, because the customer is stupid, then it's Reishauer time. A reishauer would eat these things up for breakfast, then ask for a cookie. But more $$ and floorspace and hassles. The thing is tho, the turning center attachment certainly isn't going to get that finish either. You'd have to go grinder for that. Maybe the buyer needs some edumacation ?



Over wires. Even gives you a number. It's only a thou on tooth thickness, at 45* what's that over wires, +/- .001 ? Pretty easy. I thought you Big Boys had the Good Stuff, Bob ? I bet rolling is harder to control size, +/- .001 is not that bad even with antique junk like I always owned.

Should probably get a go gage made to drop over every one, then do wires every ten or twenty or something, depending on how they hold size. If he has to go grinding, can just skip that pointless measuring step

The customer is supplying Go/No-go gages for checking the spline.

I should mention that the part is from the 60's and the customer had been making them to print in house for decades. They have said they used a carbide hob. I'd imagine getting a profilometer to fit on such a tiny tooth would be a challenge so the 16Ra might be more of a visual thing.

 

[triumph406]

The customer is supplying Go/No-go gages for checking the spline.

I should mention that the part is from the 60's and the customer had been making them to print in house for decades. They have said they used a carbide hob. I'd imagine getting a profilometer to fit on such a tiny tooth would be a challenge so the 16Ra might be more of a visual thing.

Not a gear guy, but maybe it's safe to assume shops who have been making the splines since the 60's ignored the 16 Ra requirement. Or did they get a deviation everytime they delivered?

Does the customer have an inhouse inspection dept? Last thing you need is some young inspector who doesn't know anything who might try and hold you to the 16Ra requirement.

 

[Mr.Chipeater]

Not a gear guy, but maybe it's safe to assume shops who have been making the splines since the 60's ignored the 16 Ra requirement. Or did they get a deviation everytime they delivered?

Does the customer have an inhouse inspection dept? Last thing you need is some young inspector who doesn't know anything who might try and hold you to the 16Ra requirement.

Well they themselves are the end user so I'd imagine it would be easier to give themselves a deviation. They did provide some "known good" parts to us as a sample and it doesn't look like a 16Ra to me.

 

[EmGo]

Does the customer have an inhouse inspection dept? Last thing you need is some young inspector who doesn't know anything who might try and hold you to the 16Ra requirement.

It's not on the print, so ....