Got an interesting one for you guys!
Running PC-DMIS 2013 MR1 on an Optiv Performance 443 with a single infinite rotary table mounted in Y-
I perform weekly calibrations of the tips and then the rotary table
I calibrate 1 tip (Tp200 1x10 with a 10mm extension) manually (on calibration plate sphere), yes pos has changed, take manual hit, then dcc, yes changed, take dcc hit, then say no, then do a qualification check, verify the std dev isn't ridiculous, and repeat for the ring and rotary table qualification sphere.
Immediately after I calibrate the rotary table using this tip.
I measure from 0° to 180° CCW with a 45° increment, then repeat backwards (180°-0°)
I assign a variable to GETROTABDATA("CENTER")
I construct a generic circle at that variable's XYZ using the correct IJK vectors (0,-1,0)
And here is where my problems arise:
The actual alignment process works... sometimes.
I have tried two things:
I level YMINUS to the circle, rotate XPLUS to the circle about YMINUS, and set it as X,Y,Z origin
Immediately after creating the alignment, the Probe Readout window reflects the changes (e.g. if I move the probe roughly to the center of the rotab in X/Z, X/Z is roughly 0)
HOWEVER:
For whatever reason it doesn't work after I save. Not just saving the alignment to be used in an external program, but just saving the program itself.
I move the cursor under the alignment, but XYZ are way off (what should be 0,-XXX,0 is -308,-XXX,-12)
I have tried simply setting the circle to XYZ origin, but it does not work.
BASICALLY:
I'm trying to create an external alignment program that is in tandem with a calibration program. Essentially once you calibrate the rotab, the center of it is your 0, and any part program that uses the rotary table (every part that I work with does) is simply a Y point away from DCC (and if there are ever any issues I merely have to fix one program, theoretically...) I don't want to have to measure the rotary table if I don't have to, and GETROTABDATA("CENTER") does get me a good location. But it doesn't stick once I hit "Save", actually save just the alignment, or if I recall it from an existing program.
Is it the fact I'm aligning off of a constructed generic feature using variables for a feature that is notorious for having problems to begin with? Would I be better off using a point? Should I throw the whole thing in the ocean?
Any help would be appreciated
EDIT:
I FIXED IT!
Instead of creating a generic feature to align to, I aligned to the variables themselves!
TRANS,XAXIS,VROTABCENTER.X
TRANS,ZAXIS,VROTABCENTER.Z
I had to adjust the search paths to get it to work though, works great!
