Need Help!- Can a gear be cut with a tap?

[Alex Funke]

Hello! I'm new here, and don't know all the ropes; but CNCZone is an amazing source of information! So please bear with me as I learn to use the forum.....

Hope somebody knows about this: I would like to cut relatively fine gear teeth (say 28 teeth per inch) in the edge of a disc, say 6 inches in diameter and about 1/4 inch thick.

I vaguely remember seeing something in HSM about mounting a plug tap between centres and letting the tap drive the disc round and round as it gradually cut the teeth.

The 'gear' (if we can call it that) is to engage a threaded rod, not another gear--so the fact that the teeth would be concave in the edge of the blank is ok. It's sort of a worm drive for a camera mounting device.

Does anybody have an idea if this is moonshine, or a real technique?

Cheers, and keep the faith! Alex Funke

[Geof]

Not at all moonshine. Have a look at Post #6 in this thread: http://www.cnczone.com/forums/showthread.php?t=20781

In this example the gear and tap were driven but it is likely you could let the tap actually drive the gear blank as you suggest. A probable difficulty could be getting the correct starting diameter though. You need your blank to rotate at the correct speed so that when the cut section gets back to the tap it is 'in phase' and this means the pitch has to divide evenly into the circumference of the disc.

Go ahead and try, if you are successful post some pictures.

[juergenwt]

NO! You would be making a thing with some type of teeth around it but not a gear.

[george4657]

If you want to make a worm gear, this site has a good explanation of how to make one with the method you are thinking of.

http://www.majosoft.com/metalworking...ar_reduce.html

George

[drwc]

The method you propose will form a worm gear, but not a spur gear. So if you are making a worm gear drive (as you propose)your in business!, but if a spur gear as part of a gear train is the plan this method won't yield good results.

Wayne

[Geof]

Originally Posted by juergenwt NO! You would be making a thing with some type of teeth around it but not a gear.

Not entirely correct. When an acme, trapezoidal in Europe, tap is used it does generate a proper worm gear tooth form. When a standard tap is used it generates a profile that will rune successfully against a bolt, or threaded rod, of the same diameter and pitch. In the latter case with a standard tap the load carrying capacity is not great but with a bit of care it is possible to make a worm reducer that has very little backlash.

This is what Alex is seeking to make; a lightweight fine worm drive for moving a camera. Unless he is into filming Imax movies I suspect the camera is not very heavy.

[Alex Funke]

Great information! I knew that somewhere out there would be the facts.

Indeed, it's not a 'gear' as such, but it will be acting like a worm and wheel. It's not moving anything very heavy--in fact, I'll probably make it out of PTFE.
It's just to move a digital stills camera.

And the wheels will run very slowly--they are not the tracking part of the system, but the part that will reposition the camera (in pan and tilt, so to speak) to acquire a series of overlapping views of the heavens. The whole thing sits on a 'barn door tracker' to counter the motion of the earth.

Many thanks to everybody! This is fun! alex

[Geof]

Use Polycarbonate, Delrin or Nylon, PTFE (Teflon) is not very strong.

[HuFlungDung]

How to explain this: if you try hobbing a gear starting from a blank with a tap, and feeding the work into the tap radially, you will experience a creeping advance of the teeth along the circumference. For a standard wormwheel (with a bit of a rounded throat on top of the tooth), the circumference at the OD of the blank has enough length for 3 more teeth than can fit at the pitch diameter of the gear. The pitch circle is where the worm technically contacts the teeth.

So the tap is to some degree, going to mash its way around the blank, and give you +/- 2 or 3 teeth from what you actually want

To try to avoid that, you should have a positive drive to the blank that will force it around at the proper ratio. That is not likely in the realm of possibility for a "one-off".

You could try a tangential feed method, where the tap and blank are positioned at the correct radial center distance, but the tap is displaced axially so that it only cuts on the very end to begin with. If there is sufficient contact to drive the blank around, you might luck out and get the correct number of teeth.

You might need to provide some soft 'dummy stock' on each side of the gear blank to create a constant engagement with the tap so that the blank never can come clear by springing into the flute of the tap.

To generate a wormwheel on the lathe can also be done with a threading method, and this involves some sort of indexing plate, with enough holes to match the tooth number being sought. Basically, you build a boring bar that runs between the chuck and tailstock, with a single tooth flycutter sticking out in the middle. Then you simply use the carriage half nuts and engage to thread the blank past the revolving bar. This cuts only one tooth to full depth, and a couple of partial cuts on either side of that one. Then stop and index the blank one tooth, and do it again. How you index the blank is usually done with a plate drilled full of holes in a regular pattern. You might do this on a cnc, or you could also use a degree wheel on a small rotary table to hold and index the blank.

You might want to opt for a coarser tooth pitch than 28 tpi for this method.

[Geof]

Originally Posted by HuFlungDung .....To generate a wormwheel on the lathe can also be done with a threading method, and this involves some sort of indexing plate, with enough holes to match the tooth number being sought. Basically, you build a boring bar that runs between the chuck and tailstock, with a single tooth flycutter sticking out in the middle. Then you simply use the carriage half nuts and engage to thread the blank past the revolving bar. This cuts only one tooth to full depth, and a couple of partial cuts on either side of that one. Then stop and index the blank one tooth, and do it again. How you index the blank is usually done with a plate drilled full of holes in a regular pattern. You might do this on a cnc, or you could also use a degree wheel on a small rotary table to hold and index the blank. You might want to opt for a coarser tooth pitch than 28 tpi for this method.

You are a glutton for punishment if you did this!! Before building the rig described in the thread I linked I considered this approach; tedium multiplied by 60 teeth quickly discouraged me.

[HuFlungDung]

Yes, there is likely no need to have so many teeth on a wormwheel: just turn the worm slower

I have used rigid tapping on the Haas to generate wormwheels in a similar method. I simply grind an 'acme-style' tool bit and single point the gear. Because the tool is not supported on the outer end, one cannot be real aggressive in the cutting, but the machine basically runs itself (with the 4th axis), so you can go and internet for a couple of hours while it works

[Alex Funke]

I think I'll call it a toothed wheel instead of a gear!

If I were to cut a thread of the desired pitch on a rod of the proper diameter, thus making the actual drive worm, and then took a short piece of it, slotted it at three positions lengthwise, polished the cutting faces, and hardened it (in effect making a 'tap'), wouldn't this produce an exactly mating set of teeth on the disc, which would match the master worm?

Hadn't thought about the softness of PTFE. I have some Delrin also and will use that Thanks for the suggestion.

Many thanks for the useful technical answers that have come in on this question. Cheers! alex