Re: [CAD_CAM_EDM_DRO] emc g92

HI Tom,

No plans right now for further tests, I'll just check the next release
from Sherline. Sounds like restarting was an issue, and I like to do
that all the time. That's where I do my G92's.

Alan KM6VV

Tom Hubin wrote:

> Alan Marconett wrote:
>
>>Hi Tom,
>>
>>Are you running The Sherline version of EMC? I simply MUST have G92 for
>>touching off in X and Y, and also the tool length in Z. I'm currently
>>using my STEP4 controller, not EMC. I want to run Sherline's EMC, as
>>soon as I get it's machine running again!
>>
>>I use G92 as Jon suggests to set X,Y and Z, but not IN the program, I do
>>it manually with MDI. Does EMI tolerate this? I don't currently do
>>tool changes in the program, I just make separate programs. I take it
>>you're using G92 to set Z 0.005 or whatever after the M06? Humm, maybe
>>I can incorporate something like that. Do you have a code snip?
>>
>>I was thinking of putting in a dialog box that would automatically set
>>the 0.1" offsets for LL, UL, UR, LR for me.
>>
>>Alan KM6VV
>
>
> Hello Alan,
>
> I tried Sherline's EMC version and the regular version 2.18, both from
> the Sherline BDI CDs.
>
> I don't think it matters if G92 is MDI or within program. EMC just does
> screwy things whenever you try to use G92. All of the offered solutions
> form the EMC lists, none of which are acceptable to me, avoid the use of
> G92 altogether.
>
> I could not get anybody on the EMC developers list to fix this. They say
> that it works as defined in the spec cuz the spec does not define it.
> How's that for catch 22. Nevertheless, nobody on the EMC list has come
> forward to say that they actually use G92 to do anything. So, if you ask
> me, EMC's interpretation of G92 is somewhere between useless and
> dangerous.
>
> I did all of this in June so some of the details are muddy. Bottom line
> is I cannot use EMC with my Sherline mill until G92, or some other
> Gcode, sets the axis values with no screwy side effects.
>
> Let me know if you have better luck with G92 and I will try again with
> whatever version you are using.
> Tom Hubin
> thubin@...
>