[Owain1602]

Hi, does anyone have any experience in gear cutting on a milling machine or lathe?
Me and another guy from work are starting a machine shop night class in a few weeks, he's got a steam locomotive to be doing and i need things to make really.
I know gear cutting is unbelievably complex so it would take ages to learn this but would it be possible to make my own gearbox for my rally car? Choose the exact ratios i want, make the gears straight cut.
I have started reading about gear cutting and its a very large field, Cyclodic curves, Hypocyclodic, PCDs.
I could sort of have this project in the back of my mind after years of practice i might begin to make it.
The lecturer said he can make absolutley anything on the machines, i think he was just boasting. So if things are to complicated for me to make, he'll make them.
What do you guys recon? The machine shop is amazing, every machine you could ever think of, plus they do heat treatment and tempering and things there.
So, I know it is possible cause anything is possible, but just how difficult and complex is it? and whats the difference between doing it in a lathe or a milling machine?
Im in the process of looking for a milling machine and maybe lathe for my workshop at home so I will be able to practice and play about a LOT. Im not going to just go to the course for 1 term, im going there for years and years to get as good as I can. Everything I do, i want to be the best.
The gearbox project wont get under way for many years probably, because there's no point in starting if my skills aren't good enough

Thanks for your time and patience.

[lakeside]

We have this thread going http://www.cnczone.com/forums/showthread.php?t=23867 and yes depending on form you can cut a gear on a mill

[Owain1602]

is that gear cutting with cnc i take it.
what aspect of gear cutting is so difficult then? i have been told that most engineers avoid it because of its difficulty.
from my very small knowledge about the subject, it seems to me that most of the work is in the designing of the gears and obtaining the right cutter. Such as deciding your cycloid and hypocycloid curves.
In that sortware, i see blocks for entering stress angles or whatever, how are these found then?
Im 21 and a head mechanic in a fairly large workshop and regularly service and re-build car gearboxes, but this time i want to make my own gearbox using the original box casing and bearing housings but changing the design of the gear type from helical to straight cut for less transmission losses and changing the ratios to suit different evens and my engine torque curve.
Any help will be appreciated.
P.S I have a book on order "Manual Gearbox desgin" which might help me a little.
Thanks a lot

[fkaCarel]

Making gears in the traditional way using a dividing table is a piece of cake, using gear cutters. These gears are limited usable, because they have evolving errors. So for better gears you go for hobbing, where the gear is made, like it is contact with another gear. But car transmissions use different gears. All the different set of gears must have the same pitch (centre to centre distance). They must also have prime numbers of tooth, to avoid the same tooth constantly meeting the same tooth of his partner. Then they use techniques of tooth bottom strength heightening, where they enlarge the pitch circle. This is advise and expertise land. After this comes case hardening. Same land.

Conclusion:
For industrial use: Make your own, only in emergency, if not for sale and preferrably not above 1000rpm.(noise).
For automotive: outsource.

You live between open gearboxes, start measuring gears and see how it's done.

[diarmaid]

Hi Owain. Who publishes the book 'Maual Gearbox Design' ? Also if you could look at the thread in post no.2 above and give me any advice on attaching the drive shaft to the gear from your experience it would be much appreciated. There won't be any housing around the gear. Thanks.

[Owain1602]

Why are the teeth made using a dividing head and a cutter so different to other methods then? Is there a way of cutting the gears like this then modifying them to have some properties of the other processes?
They are gonna need to cope with around 150nm and about 8000rpm!
I take it making them straight cut gears will make then A BIT easier to cut than helical plus they wont have to cope with side thrust.
I have been advised by a guy to use N39 to cut the gears from by a historic car builder, any comment on this material?

[fkaCarel]

With the dividing method you use a set of 8 cutters for the range of tooth between approx 15 tooth and straight, somewhere around 135 tooth. They are approximations. The hobbing method uses the a mill with the "straight profile" and is geared to the gear to make. The profile of the gear is in this way automatically corrected. You would see, feel and hear immediately the difference of this two methods.
For "sport" appliances straight is the normal choice, more noise, no side pressure. Keep also in mind, that you will have to broach the internal splines and make the intricate connecting lips on the gears. The lips have an angle to keep the gear "in gear" and radiuses on the bottom that with high power, scratches and unoptimal hardening can come of.

[Owain1602]

You seem to know your stuff about gear cutting.
So do you recon the hobbing method will be doable in the machine shop they have at the engineering college im going to? Its a well equipped shop.
So say i want a 5 speed box, i will need 10 gears plus an idler for reverse. Will i require 11 different cutters for this then?
I know what broaching is but how difficult is this to do? I guess all the gears can have all the same splines on them so id only broach them all the same then?
What dissadvantages would i see from making the gears from a dividing head and cutter over the other? I dont really care about noise from the box anyway, the 4 throttle bodies i have sucking away make quite a bit of noise already.
Thanks a lot

[fkaCarel]

Well, you need a hobbing machine. They are certainly more rare than lathes and mills.
The thing I worry about is the elevated pitch circle. Once I had to outsource primary transmissions for racing motorcycles. I had everything sorted out, according to industrial standards. The manufacturer, who was experienced in this (and also in hardening) told me, that standard sizing, for the small gear, would make the teeth to vulnerable. The correction for low tooth gear weakens the base. An extreme example is the ancient roll dryer, used for drying clothes. The design (roll dryer) allowed only for four extreme corrected and vulnerable teeth. When you see one, look at this detail.

So he designed within the dimensions his own pitch circles and hobbed the special gears with a standard hobber. These gears never broke down and a championship proved he was right. Knowing this, and putting your caliper here and yonder, you see this is widespread. Transmission gears are high-stressed components.

[Owain1602]

Ah i see what a hobber is now, looks like a torture instrument.
Can you use the hobbing bit in a lathe or mill and use a dividing head or is that being stupid? Im sorry in rubbish at this right now as i have no experience.
I dont see that being a consideration really with gear sizes, they all seem to be of a fair size, cant remember diameter of smallest gear right now. So you're saying smaller gears tend to be weaker because of the smaller number of teeth or smaller diameter? Because the teeth can still be same pitch cant they?
Sorry, what do you mean by elevated pitch circle? Do you mean Pitch circle diameter? sorry again

[fkaCarel]

I'd say google a little on "Module" or "Modula", read up a catalogue of industrial gears, find out how a hobbing machine works, reread what I've written and come back next week. It doesn't sink in just on one saturday night. :cheers:

[Owain1602]

module or modula in conjunction with what then? Modula the programming language?
Sorry, thanks.lol

[diarmaid]

One good turn deserves another so heres a site that you will probably find useful Owain. Unfortunately most of the stuff is for sale not free, but theres a couple of relevant links on the site such as the second link below.

http://www.drgears.com/

http://www.geartechnology.com/

[itsme]

Hi there,

I have done a fair amount of gear cutting at home using a dividing head and involute cutters. Most of the gears I have made are lightly stressed, so high-tech alloys and heat treatment haven't been necessary.

As fkaCarel mentioned, I would think that hobbing (or some kind of 'true' generating method) would be needed for what you want. I have no experience with car gearboxes, but I would think that designing/generating the teeth would only be half the struggle over. I would imagine that the material used, along with the heat treatment would be quite critical.

Regards
Warren

[fkaCarel]

Here is a fine example of base-width versus number of tooth. Both gears are hobbed at the same factory according to the same industrial specs and the same modula. Yet you see that the tooth of the smaller gear are more corrected, leading to a smaller/weaker base.

[chich2]

This is going off the subject a little bit but I made this toothed pulley for a supercharger last week end and discovered the following: The manufacuter's information for the toothed belt stated the pitch as being 12.7mm. I calculated the diameter of the pulley to the closest number of teeth to fit the material I had. I milled out the teeth using a dividing head and discovered the pulley diameter I calculated was too big. I believe the reason for this is because the belt has a very strong braid inside it that forms the strength of the belt. The rest of the rubber surrounding the braid is "there for the ride" you may say. If the belt is running in a straight line it's eask but if you bend the belt around a curved surface, you have to take into consideration the compression or expansion of the rubber around the center line of the braid (which remains constant). To cut a long story short, the correct pulley radius is from the center of the pulley to the center of the braid minus the distance from the center of the braid to the carrying part of the toothed belt. Clear as Mud!!! Hope this didn't confuse anyone. Here are some Pix to have a look at.......

[chich2]

Yea here are the pix......

[diarmaid]

Chich, which type of machine is shown cutting the gear in the first photo? Thanks.

[Owain1602]

If the heat treatment required is going to be very specialised then i will send the gears away and get them treated by someone who can do it.
I dont mind forking out for expensive metals as long as i dont waste it by starting to make the gears when im not ready and stupid things like that.

Sorry but i cant see the differences in the gears you refer to
My eye isnt trained like yours

Do you see any harm in having a go at this project when i think ill be ready and sort of make the gearbox, try it out, then strip down, re-asses things and maybe change the plans a bit then try again. This is how a lot of things are developed isnt it?

Maybe give someone like Quaife, TranX or hewland a call and ask them if they would be willing to give me a bit of advice. Probably wont share their secrets with me.

[fkaCarel]

Chich, you could have looked it up in the catalog. Biggest H(12.7): 40tooth, Pitch circle diameter: 161.70mm. Difference Pitch circle(neutral line)-outer diameter: 1.37mm
If I counted correctly, you made a 47 tooth pulley.
((47 * 161.7) / 40) - 1.37 = 188.6275 mm. Right?