1
General Mach Discussion / Re: First post: Mach3 machine and workspace offsets, auto home, and auto tool offset
« on: July 29, 2017, 08:46:27 AM »
Thanks, Rich. I'll look for the Peter Smid book. I think I started out following exactly your suggested route. I find machine zero (Ref All Home in Mach3).
I have fixture plates, each with multiple stations that are fixed relative to the plate.
I used G54, G55 etc. to specify the locations of each fixture relative to the plate. That didn't work, since each local workspace has to be defined by offsets from machine coordinates. Not a big problem - I know where the plate is in machine coordinates, so I can translate. I define G54 etc. relative to machine coordinates. So far, so good.
Now, I measure Z offset for a tool. That applies ONLY to the current coordinate system - G54, for example. Works great on the first fixture, but not so much when I switch to G55 for the next fixture.
An added wrinkle is that currently, my machine zero to plate offset isn't as repeatable as I'd like. I've considered probing the plate and setting machine coordinates based on the current plate. Cheating, I know, but I think it would work in this situation and eliminate one source of errors.
Clearly I need to do a bit more reading. Thanks again for the pointers.
I have fixture plates, each with multiple stations that are fixed relative to the plate.
I used G54, G55 etc. to specify the locations of each fixture relative to the plate. That didn't work, since each local workspace has to be defined by offsets from machine coordinates. Not a big problem - I know where the plate is in machine coordinates, so I can translate. I define G54 etc. relative to machine coordinates. So far, so good.
Now, I measure Z offset for a tool. That applies ONLY to the current coordinate system - G54, for example. Works great on the first fixture, but not so much when I switch to G55 for the next fixture.
An added wrinkle is that currently, my machine zero to plate offset isn't as repeatable as I'd like. I've considered probing the plate and setting machine coordinates based on the current plate. Cheating, I know, but I think it would work in this situation and eliminate one source of errors.
Clearly I need to do a bit more reading. Thanks again for the pointers.