The hypoid gear such as applied in automotive differentials appears to be a very complex machining operation if one were to make it in a home shop environment in a hobby scale. Has anyone in this forum witnessed the machining process involved in the creation of this type of gearing? What is involved in the machining aspect? Does it require a specialized machinery? Can it be created using the usual home shop machines like a vertical mill and lathe? It is certain that CNC would be helpful.
As for the geometrical forming of the teeth themselves, does anyone know of a book or reference that provides a most basic knowledge for understanding for this type of gearing and the particular aspects of the relationship between ring and pinion? I am not an engineer. Thanks in advance for any input and direction.
As I understand it, the process for hypoids is a generating process, and as such is pretty much relegated to a machine specially made for it. The subject has come up here in the past in relation to getting hypoid gears cut for replacement parts for old autos, and the short answer was, for at home, "No frikken way!".
The cutters are special to the process and expensive, as is machine time, so it was essentially a non starter, even as a professional endeavor.
I don't know if the more up to date CNC (4 axis, mabe 5) would be capable of carving one out (I would think it should) but I can imagine that the run time would be huge.
The generation of involute gears in the home-shop is well-covered in Ivan Law's book, "Gears and Gear Cutting". The book describes home-shop gear making in detail. It explains how to make 2-button, tool-post mounted, gear cutters from drill rod. Two cutter buttons, properly located, can generate a reasonably facsimilie of the involute curve. Ostensibly, one can make a set of cutters for all diametral pitches.
But, hypoid gears are twisted in several planes at once. I don't think they can be cut with a single cutter, as they twist and taper at the same time. So, the diametral pitch changes along the radial direction of the gears. I think you have to make a hob for it, and run it in a gear hobbing machine. One could make all this, of course, but it is much more complicated that cutting a spur or helical gear, or even a bevel gear.
Jon sez: "> But, hypoid gears are twisted in several planes at once. I don't
I think Ivan Law addressed that issue in the paragraph introducing involute teeth where he said: "It can be seen that cycloidal teeth can vary considerably in shape and yet still fulfil the needs of smooth action and constant velocity between the mating gears. So long as the generating circles are the same diameter for any two gears running together, then whatever the resulting shape of the teeth may be, the two gears will run together satisfactorily.
Previous to this the curves shown were labeled epicycloidal and hypocycloidal. Is the term "hypoid" a mere contraction of hypocycloidal ? I'm not sure.
Quoting further under the intro par. Involute Teeth, he says: "Gears based upon the cycloidal curve do maintain constant velocity between the two P.C.Ds but this is only true so long as the two P.C.Ds are touching each other. If, for any reason, the gear centres are opened out and the P.C.Ds lose contact then the constant velocity condition will not be met. This means that if the wheel centres are not accurately set in the first place or, as is more likely, the wheel bearing wears thus allowing the gears to spread, the constant velocity condition will be lost. The gear ratio between the gears cannot alter as this is determined by the number of teeth on the respective gears but the gears will not revolve evenly; speeding up and slowing down will take place as the teeth come into and out of the engagement, thus introducing vibration and noise in the mechanism. If the gear teeth are shaped on the involute curve then the constant velocity condition will not be affected by a small amount of spreading of the gear wheel centres, which means that any normal wear in the gear wheel bearings will not adversely affect the proper action of the gear teeth . . ."
All of the above is only talking about plain old flat type gears. Hypoid is not the same as hypocycloidal. Epicycloidal and hypocycloidal curves are two ways of generating the curves that define the tooth shape of a plain gear tooth.
A hypoid gear is the type found in the rear end of a vehicle drivetrain.
Sorta a combination of a skewed spiral bevel worm gear.
Now if the terminology of a gear tooth that only curves along in relation to one plane causes you some confusion, think how far out of hand the math gets for a gear that combines all the above.
They are a major gear cutting machine producer in Rochester, New York. I worked at the place for over 30 years. I have run a majority of their gear cutting machines. Believe me when I say that trying to cut a set of hypoid gears is quite complicated. It is something that can not be accomplished on a table top machine. IIRC William Gleason was the inventor of the hypoid gear. This allowed the hump in the middle of the car to be lowered. Hypoid pinions can be offset up from the center of the gear or offset below the center of the gear. The amount of math involved is quite over whelming. I believe Trevor mentioned CNC. IIRC the last I knew I believe a Gleason CNC hypoid cutting machine has something like 8 axis of motion. This includes both the cutting motor and the work holding head. It's been 7 years now since I retired from the place and memory is getting a bit fussy on all that stuff.
I've seen a web page describing the 2 button method you describe, but I forget who the author was. While I haven't gotten around to making my own gears, the 2 button cutting tool method for making a hob seems like a very practical approach.
Thanks everyone for the input. I had a feeling it was even more complicated than I imagined. 8 axis cnc machine? Temendous geometry and math involved? Gotta get my head out of the clouds! Thanks again.
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.