Machining Bevels.

[Chuck]

Art,

For small pitch we have the engraving tools.  These should work for about 20DP down to about 80DP

http://cgi.ebay.com/5-x-Carbide-PCB-Engraving-CNC-Bit-Router-40-Deg-0-2mm-/130528323178?pt=LH_DefaultDomain_0&hash=item1e6418266a

The stock sizes of 30° are close to 14.5° and the 40° are spot on for 20° pressure angles.

They are easy to make yourself and can be made for larger pitch from 1/8", 1/4", 3/8" or 1/2" drill rod.
With a Gcode generator program we can make any size we want with a 4 axis CNC!  I have made cutters in the lathe by turning them off center for clearance heat treating and grinding a cutting edge rack tooth endmill cutter for 6DP.

A basic spec for the cutter in sizes of 1/8" and 6mm:
http://www.arceurotrade.co.uk/Catalogue/Cutting-Tools/Engraving-Cutters

Chuck

[ArtF]

Hi Guys:

    Just a note on the bevels, its taking a bit longer than I would have thought, this isnt because of the difficult y in theory or anything, the theory is working well so far, but Im verifying very carefully the numbers and code Im doing, so its slow going ( but I hate bevels so Im going slowly and accuratel y so I only have to do this once more. )

   It looks like I can allow for lower taper values as well, so Im hooking that in as well as I go.
I hope to be able to release some tests soon. ( Just so you know work is progessin g..)

Art

[ArtF]

Hi Guys:

   I just completed a simulatio n of the bevel cutting. It seems Ill have to start the shaving rotations a bit earlier as the portion below the pitch is not getting fully shaped. Haveing no real root in the bevel makes it awfully close, but the root isnt being fully shaped by the current zone of rotations Im doing, so an update will come out for that problem.

Art

[DesertRunner]

Hi All whats the latest did you get the bevel gears sort so thsy can now be miilled.

Do I understan d correctly that its a issue at 90 degree?
Tony

[ArtF]

Tony:

   They are ready to be tested, but Im not setup to do so, so until Ive heard they work, I wont claim they do. Mathmatic ally, they work as best I can test them internall y. The process is to have a blank of proper size, tilt the 4th axis to the requested degree, and use a tapered bit, tapered at the pressure angle of the tooth form selected.
   You cant cut one at 90 degrees, the tangent of 90 degrees may cause math errors in the routines, though to be honest I havent tried one..may work.. Ill have to check. The tooth form on a 90 degree bevel is a rack form..so I dont think Ive allowed it..

Art
 

[Dan]

The process is to have a blank of proper size, tilt the 4th axis to the requested degree, and use a tapered bit, tapered at the pressure angle of the tooth form selected.

Is it only restricte d to this kind of a cutter? A straight cutter won't work? And a one having a smaller taper than the pressure angle?

Dan

[ArtF]

Dan:

   Lets put it this way.. No matter what pressure angle is selected when designing the bevel, the Code will produce a toothform with a pressure angle
equal to the cutters taper. The code attempts to reproduce a racks motion in cutting, but in bevel fashion, so if you use a straight flute, youll likely get a rack shaped
tooth. Give it a 14 degree taper, and the result should be a 14 degree pressure angled tooth form. So you can really use any tapered bit, closer to PA the better..

   I must admit, its been at least 8 months since I coded it, so while I was an expert on its operation as I wrote it, I'd have to study it now to refamilia rize myself.
Its the problem with bevel code, I cant test it myself properly as I like to do with other types, and its so complex to do that I rapidly forget the exact specifics of the
process. If someone attepts it and gets an error, or unexpecte d reult, Ill revisit the code as necessary to correct trouble.

Thx
Art

[Dan]

Thanks, Art. Still not ready to test these either, but as soon as I will, I will let you know.

Dan

[JustinO]

Wake up thread!

Did I hear someone say there was a video of a Gearotic bevel gear being machined?

What is the geometry of a machined Gearotic bevel gear?

I am of the opinion that there are two ways to make a bevel gear, the right way, and the possible way.

--Justin

[ArtF]

Justin:

 I didnt video my bevels as I cut them. ( I cut mine from foam ), and to my knowledge none has made any on 4th axis, though I know some were 3d printed.
You have to cut the beveled blank at the right angles to actually machine one, and your 4th axis has to tilt to the rigth angle. On my system this severly limits
me as I hit my spindle if I tilt more than a few degree's, so Ive never actually built one with latest code for bevels, though a mathmatic al analysis was done
on the posted files to ensure they "should" work fine.

Art

[JustinO]

Art,

I've read that the "octoidal" tooth form has been the modern epitome for a long time. I assume that the approxima tions you find in sources such as Ivan Law were trying to stay within error of octoidal while still being economica lly practical . But I don't know for sure. What are you're algorithm s aiming at? Were you shooting for a theoretic ally perfect form, or compatibi lity with what the industry has been approxima ting all these years?

I'm kind of liking the octoidal because the generativ e motions are simple. I don't know if octoidal gears can run with "Machinery's Handbook" bevel gears.

--Justin

[ArtF]

Justin:

   The specs of the bevel are as close to proper specs as I can make them, the only thing I am attemptin g is to allow them to be machines with a "simple" 4 axis machine.
They are incredibl y difficult to make which is why few have tried..

Art

[JustinO]

Art,
These gears are not correct. I tried to make them with only four axes -- two linear, and two rotary. They mesh nicely, but they're not theoretic al.

Machining bevel gears with only three linear and one rotary axis as you are is a really tough constrain t. The math and or tooling gets tough, it takes longer, etc.

I've got five axes now, three linear, and two rotary. This allows the math and tooling to remain simple and the run time short. But not everyone can be as obsessed and impractic al as I am.

--Justin

[ArtF]

Justin:

  Well, they may not be standard, but the only real important thing for any set of gears is that they mesh and work
for the job you intend. Look pretty good for a clock or orery or something . (Looks like the tooth profile needs to be
divided down by the secant to get they tooth profile shorter like in helical gears..)

  They look pretty good though, so if they work, theyt are by deinfitio n perfect.

Art

[JustinO]

Thanks Art,

They aren't as long-toothed as they look. The way the blanks were cut left the cross sections of the teeth on an angle that exaggerat es their length.

The really cool thing is that when you hold them together and roll them, they almost feel magnetic the way they align to each other - and they roll so smoothly!

There aren't many things as fun as making your own bevel gears.

--Justin