GearHeads Corner

Thanks Art for your diagnosis on my tool crib issue(s).  For now, I have changed the properties to open always as admin (although I expect I will move it all to C:\gearotic soon).

With that, I've now been able to layout my first project.  I started with a few pinions around a wheel, all boxed.  I sent the gears and the box to the workspace, deleted the front of the box (retaining just the rear).  Highlighted them all, had it create and post the g-code.  I used a 1/4" cutter so I can understand the lack of detail in the gears, but what I cannot explain is:

(1) Spoke holes aren't round.  See photo, both on the rear box panel, and on the gears.
(2) The left boundary of the rear of the box didn't cut properly.  See photo. Was above the surface (after cutting in the air for a sequence, see below for more on that), then ramped into the material.

Also, what is not evident in the photos but what I believe I observed:

(3) The code when run on Mach3 ran the cutting routines almost as if it thought it was 1/2" material (i.e., path was cut "in the air above the material) instead of the 1/4" plywood I had used.  But since I zeroed out at the surface I would have expected it to cut 1/4" into the waster board.  So, I must have some setting not the way I thought.

(4) I use a SuperPID to control the router on/off and speed.  Not sure I describe this using correct terms, but the router turned on/sped up, the CNC would cut some aspect of the job, then the router would turn off; then it would turn on/speed up after it moved to a new part of the job and begin cutting (as described in 3, above about 1/4" more or less - I wasnt going to put a tape measure near!); then it would move into the wood.

Although very much a newbie, I've cut some gears when I exported the g2 design into a DXF, then running the DXF thru Cut2D; and then in any instance cut using Mach3.  I also successfully (in a version of G2 from about a month ago), exported g code from a G2 design and cut a couple of gears.

Also, the tabs are 1/4" thick - but if I now understand better the "final pass" setting that might be because I had that set at 1/4".

Thanks.  And having fun, really.

Brian.

Brian:

  That looks very much like Ive introduced a bug in the code.. Can you confrm the version that your using, Ill run
some tests and get a fix out in the morning...

  Youre tabs will get thinner as you said if you make a finish pass thats less thick.. Your center holes may be affected because they were
half tabs.. I think Ill add a fix for that as well. Ill report in the morning after Ive solved the issue..

Thx
Art

Hi Brian:

I ran a few tests here.. it seems to be working well. I DID notice that if the final pass is equal to the initial pass, that it does a
pass at zero... Ill have to add a fix for that, ideally you shouldnt set the final pass to the same depth as the final depth and initial pass...

  I also noticed that perhaps tabs shouldnt be in the shaft hole.. unless pretty small. I do remember fixing a problem a version or two ago
that could create that bad ramp you have on the plating.. coudl you make sure your up to data e and try that job again with a final pass of 1/8" or so,
... I recommend a somewhat smaller tool as well...but thats opional.:)


Art

Art:

Version 2.25800

I've attached the g-code in case it would be helpful.

Brian.

Hi Brian:

  Yes, a large part of the problem is that the depth was a single pass with a final depth equal to that pass. I need to fix this this
though so it doesnt happen again. Hopefully by later today..

Your file start with no Z move because it thinks it started at zero and needed to cut at zero.. Im not sure why its going to .25 .. But Ill see whats
causing the issue and kill it. Ill post there when Im done..

Thx
Art

Hi Brian:

  Can you tell me your  ZDn/Pass , Depth and final pass settings on that job? Im setting up safeties to make
it so that kiond of error cant occur anymore.

Thx
Art

Art:

First, let me caution you on the settings I will relay in response to your request: I cannot be certain I didn't (and I believe I might have) overwrite the gth file from the version I used to create the g-code I used to cut the project.  That said (or written, I guess), I've attached a snippet of the status screen from G2 to provide you with the information you requested

Hoped the g-code might have been of more help.

Regards,

Brian.

Hi Brian:

  Wont download for some reason.. reports 0 bytes.

I assume fomr the code that the finalpass was set to .5" while the depth was set to .25.

This ,  I think, caused the algorithm to see the first pass woudl be below final pass start, so it started
the first pass at full depth - final pass ( +.25") which made it do its first pass above the material. (Ive made thgis impossible
as of next release.).

  Im still unsure why the ramped edge I see in your phot though.. Ill see about tightening up some more logical tests . Id try again
with a small finalpass so your tabs are liek 1/8" or so..  Ill try for a release tomorrow.

Thanks
Art

Art:

I've attempted to attached the tap file again.  Hope it was properly attached/downloaded.

Thanks.

Brian.

Hi Brian:

It was the photo of settings I couldnt open...tap file was fine. :)

Art

Snippet of the settings was less reliable anyway given the likelihood of changes to the settings since I created the g-code.

Brian:

Thats what I figured. Ill add a couepl more safeties, then Ill issue a new download. I think it was just having a final pass too large that did you in..
Let me know if you do another and anything looks strange.

Art