What am I doing wrong here?

[JamesTSG]

Hi all, novice CNC and Gerotic user here.

Been testing out cutting gears and I am having a problem that I can't figure out. The gears look properly shaped in GM, and AFAIK I have the settings set properly but when the mill cuts them, the come out looking more like fan blades!



Both are 60mm in diameter 45 degree helicals. Left one is the default settings just with reduced teeth to get into the diameter, the right has reduced module (3.8 ) to do the same.  In both it looks like the mill is cutting too much of the tooth, as if the wrong cutter diameter was specified, but it is set correctly I think at 1/8".

I'm using GM v4.88 on a Taig 2000 4 axis Mill running Mach3.

What am I doing wrong?

[ArtF]

Hi:

  This looks liek either the actual tool diam is not matching the setting.. ( Id measure it to be sure ) OR the center distance is off. The distance is highly critical and can make teeth wider than they shoudl be ( or narrower ) and its surprising just how little you have ot be off before it shows up that way.

Thx
Art

[JustinO]

The NC file says " (Tool Radius: 0.6250 ) ".


The .6250 would be interpreted as mm with the G.21 in effect? Giving a tool offset of .0246 inches?

Where am I and reality parting?

[ArtF]

Sounds like the wrong size bit. If your in metric mode, you should enter the bit size in mm's. Try again with a bit diamter of .125 and see if the code then shows it properly. You do have to specify the bit diameter in whatever system your in
mm or inch..

Art

[JustinO]

I'm not the OP.

The header of the NC file says .625 or 5/8. Maybe it was intended to say .0625 or 1/16. I guess the program divides by two to give the radius from the entered diameter value which was maybe entered 1.250? So, it sounds like 3.175 (mm) should be entered, and would wind up in the header as 1.5875.

Sound right? Or something else entirely!

[ArtF]

Yes, I suspect thats the trouble. Ill list changing that from radius in the file to diameter to make more sense.


Art

[JustinO]

And perhaps with the unit of measure based on the mode the program is in?

[JamesTSG]

I will check and try that when I get a chance.  But I think if I made a mistake in tool size specification,  I would think the teeth would be too big instead of too narrow.

[JustinO]

If the program thinks the tool is 0.00mm in radius, it will move the center of the tool along where the tooth is to be cut, cutting too much. If the program thinks the tool is the correct radius, it will back off the path of the tool's center by the amount of the radius, so the tool's edge is moving along where the tooth is to be cut.

[The .625 in the header file is being interpreted as mm, not inches.]

Please be sure to show us how things go -- we're interested in seeing a successful outcome!

[JamesTSG]

Hi guys, 

Yes, I had mis-input the tool diameter.  It's always the little things that get you... :-X

Now onto my next question:

Are helical bevels not implemented yet?  When I try to output the G-code for one, I get a file with nothing but the comments and no instructions.

[ArtF]

Hi:

  I think helical bevels were turned off until straight bevels were tested. As yet, they havent been with the new code. They require a
milling bit tapered at the pressure angle. The problem with bevels is they are very hard to cnc, while 3d printing them is easy, cnc'ing them
requires a special blank as well as the tapered bit. The older code used a stright flute and worked, but peopel wanted smaller bevels so we
switched to tapered bit and as yet it hasnt tested out. I plan to offer a new improved version in the fall for cnc bevels, but till then your limited to straight ..

Art

[Mooselake]

Art:

I don't have a 4th axis so this is mostly academic, but why do you have to have a special blank?  Wouldn't it be easier to make one that's somewhat close (bandsaw to the general diameter, say, from the proper thickness), then mount it in the 4th axis and mill it to the proper diameter/shape?  Could this be something that GM could do for you?

Be gentle, way out my depth here.

Kirk

[ArtF]

Kirk:

  Its due to the angles involved, getting a smooth blank of the right size in a 4th axis would take a very long time..
take multiple tool sizes to do it efficiently..basically a pain in the ass. Its why special machines make bevels, you rarely mill them
or make them any other way..

  Now 3d printing seems the way to do.. Im of the opinion even to print one to use as a mould is probably quicker and better quality than trying
to machine one...

Art

[JamesTSG]

Good for quick prototyping and yeah making a plug or mold, but problem is that most additive materials and techniques aren't up to the requirements that a bevel gear train needs.  If not peak loading, then long-term wear, not even metallic sintering.  And even then with most of them, you will still have to machine finish the gear faces...


Very cool stuff nonetheless.

Thanks,

-James

[ArtF]

James :

All true. However, using additive methods for moulds is pretty good for bevels if one needs the strength
and I suspect a nylon bevel is pretty strong for most applications for small mechanisms. ( Windo blind winders and such).

  Im seeing many posabilites. I printed a bearing this week, runs pretty well so I can see where
GT will have to offer parametic bearings for printing, hinges..and a variety of other 3d objects.. HOW one
deals with making them Ill leave up to the user. Its possible to download models of many things these
days, but GT will concentrate on proper mathmatics of models and parametric ways of creating them. Ill
focus on output options that reflect whats possible for each model type.

 
Art