Hi All;
I am new here so please forgive me if this not the correct place to post this.

I have modified my mini-mill by adding stepper motors and inductive limit switches to the mill.  Works very well with Mach 3 at the helm.  I have milled many parts but sometimes have problems with very fine engraving jobs where objects are out of alignment.

I am using BobCad to design my parts and generate the g-code output.  BobCad can generate g-code for many makes of machines.  Through trial and error I managed to find a Fanuc post processor that is somewhat compatible to Mach 3.  I managed to edit the Fanuc post processor to make it more compatible but still occasionally come across some code problems.

My question here is what flavor of g-code does Mach 3 except, Fanuc, Samsung, Hitachi...?  What co-processor would work best?

Thanks.

Hi,
all CNC solutions like Mach3, Fanuc and Mitsubishi differ slightly. There is no such thing as an absolutely standard Gcode.
The closest approach to a standard is ANSI standard code, and that was modelled after what was then, widespread industry
practice, which means in real terms, Fanuc. At that time Fanuc represented about 80% of the market.

Mach3's Gcode closely follows Fanuc and in most cases a Fanuc post will produce compliant code with a few exceptions.

Mach4's Gcode is very strictly compliant with Fanuc 21B which is in turn compliant with ANSI. I have used Mach4 for five years
and found Mach4 Gcode to be compliant to the letter.....which can be problematic, we used to take shortcuts in Mach3
but they fail in Mach4. Its not that Mach4 is wrong but rather we got into bad habits using Mach3.

Craig

There may be a mechanical reason why you are having problems with very fine text, backlash in the drive screws, belts, bearings, etc.  It can also be you do not have the resolution to produce very fine detail.

Thanks for the feedback.

The Mach 03 and Fanuc relationship was useful as I found that Mach 03 responded very well but not perfect to all the g-code that BobCad generated.  BobCad is also a very old program so I suspect it can be finicky as well.  As to an ANSI standard g-code thanks for the insight, I can relate.  As I an Engineering Electronics Technologist I have also seen the ANSI standards as in RS 232 communications not being very standard at all.

I managed to find a users manual for Mach 03 that details the g-code subprograms very well as well as Mach 03,s response to a g-code function. So I should have enough information to finish customizing the post processor.

As for the machining problem again thanks for the feedback.  I am aware of the backlash sues as I am using an open loop servo system (without positional feedback).  Fortunately Mach 03 can be setup to compensate for backlash and lead screw nonliterary.  Nonliterary has not been a problem and I have corrected for the backlash.  Unfortunately the backlash correction is not quite symmetrical so I have a 0.5 though backlash in one direction and 0.2 though backlash in the other on all three axis.  All in all, the mill is quite accurate considering the setup.  I purposely did not over pre-load the bearings too prevent premature wear.  Maybe I can install a strong wave washer behind the bearing to help with the backlash vs pre-load.

I have reviewed my part problem and have found that I had used the "scribe" function and not the "engrave" function in the past.  That may solve my problem.

Anyway, Thanks again for all the feed back.  I love Mack 03 have not tried Mack 04.  I think that there is allot of power packaged in the software.  Keep on making!!!

Case closed.