Measuring a two part fixture with one program

[Sean Steele]

I have written a couple programs that will measure both halves of a fixture that holes a right and left part. The programs work but are really complicated. I am hoping that all of you gurus know of an easier way. Here is what I have to work with. 1 automotive stamping fixture, used to verify the stampings using feelers and locator pins. The fixture has tooling balls with locations correlating to each part. I setup the first side as normal using an iterative alignment on the tooling balls. Then it gets tricky, I have to recall the machine alignment, the one that is there at the beginning of a new program. Then I have to manually measure the tooling balls again. I basically end up with two programs in one piece of code. What I would like to do is have it just recall the first DCC measurement of the tooling balls and align to them using the coordinates for the second side of the fixture.
Any suggestions?

TIA
Sean

P.S. If it helpes, the sides are mirror images, but, I would like to know hoe to do this with non mirror fixtures as well.

[NZ CMM Girl]

Equate Alignment?


Have you looked at the equate alignment feature?

It may or may not help in your case - but it's quite interesting. Yesterday I discovered that you can use it when you accidentally bump your part. It will remeasure your alignment features and then equate your before bump features with your post bump features so you don't have to measure them all again!!

[Matthew D. Hoedeman]

If you have 2 sets of numbers for the t-balls, one set for each side, I would simply measure them all once and make 2 regular alignments, one for each part. Save each alignment to an external file, then recall those alignments in checking programs. Construct a place and a line for the alignments. Level to the plane, rotate to the line, origin on one t-ball, offset for the axis based on one set of numbers. Save the alignment. Then, level to the plane, rotate to the line, and origin to the same t-ball, but offset based on the other set of numbers. Save that alignment.

[Paul Barriage]

I agree with Matt. Most of our alignments aresaved & recalled externally. Not only does it allow this combination, it keeps you away from the 2 alignments messing with your feature measurements.

[Sean Steele]

So there isn't a way to do it within one program using iterative alignments? I can't believe that with everything this software can do I actually have to do this the same way I did in MeasureMax! The last time I programmed one of these dual fixtures, I used standard 3-2-1 alignments but since the fixture set on the table with the x axis of the cad running in the z axis of the machine I had trouble getting the CAD to match between alignments.
Here is the code I used to get from one side to the other.
A2 =ALIGNMENT/START,RECALL:RH-TB-ALIGN, LIST= YES
ALIGNMENT/ROTATE_OFFSET,180,ABOUT,YPLUS
ALIGNMENT/TRANS,XAXIS,RH-T-BALL-1
ALIGNMENT/TRANS,YAXIS,RH-T-BALL-1
ALIGNMENT/TRANS,ZAXIS,RH-T-BALL-1
ALIGNMENT/TRANS_OFFSET,XAXIS,-2430.4
ALIGNMENT/TRANS_OFFSET,YAXIS,100.006
ALIGNMENT/TRANS_OFFSET,ZAXIS,-585.247
ALIGNMENT/END
LT-T-BALL-2 =AUTO/SPHERE, SHOWALLPARAMS=YES, SHOWHITS=NO
THEO/2430.4,-100.006,585.247,1,0,0,12.7
ACTL/2430.408,-100.011,585.259,1,0,0,12.709
TARG/2430.4,-100.006,585.247,1,0,0
THEO_THICKNESS = 0, RECT, OUT ,$
AUTO MOVE = YES, DISTANCE = 200 ,$
NUMHITS = 5, NUMROWS = 2, INIT = 0, PERM = 0, SPACER = 0 ,$
START ANGLE 1 = -360, END ANGLE 1 = 0 ,$
START ANGLE 2 = 0, END ANGLE 2 = 0 ,$
ANGLE VEC = 0,1,0
LT-T-BALL-1 =AUTO/SPHERE, SHOWALLPARAMS=YES, SHOWHITS=NO
THEO/2430.404,-100.086,-174.74,1,0,0,12.7
ACTL/2430.417,-100.102,-174.73,1,0,0,12.702
TARG/2430.404,-100.086,-174.74,1,0,0
THEO_THICKNESS = 0, RECT, OUT ,$
AUTO MOVE = YES, DISTANCE = 200 ,$
NUMHITS = 5, NUMROWS = 2, INIT = 0, PERM = 0, SPACER = 0 ,$
START ANGLE 1 = 0, END ANGLE 1 = 0 ,$
START ANGLE 2 = 0, END ANGLE 2 = 0 ,$
ANGLE VEC = 0,1,0
LT-T-BALL-3 =AUTO/SPHERE, SHOWALLPARAMS=YES, SHOWHITS=NO
THEO/2430.388,-606.885,-174.8,1,0,0,12.7
ACTL/2430.414,-606.933,-174.741,1,0,0,12.705
TARG/2430.388,-606.885,-174.8,1,0,0
THEO_THICKNESS = 0, RECT, OUT ,$
AUTO MOVE = YES, DISTANCE = 200 ,$
NUMHITS = 5, NUMROWS = 2, INIT = 0, PERM = 0, SPACER = 0 ,$
START ANGLE 1 = 0, END ANGLE 1 = 0 ,$
START ANGLE 2 = 0, END ANGLE 2 = 0 ,$
ANGLE VEC = 0,1,0
LT-TB-ALIGN=ALIGNMENT/START,RECALL:A2, LIST= YES
ALIGNMENT/ITERATE
PNT TARGET RAD = 5, START LABEL = , FIXTURE TOL = 0, ERROR LABEL =
MEAS ALL FEAT = NO ,LEVEL AXIS=XAXIS ,ROTATE AXIS=YAXIS ,ORIGIN AXIS=ZAXIS
LEVEL = LT-T-BALL-2,LT-T-BALL-1,LT-T-BALL-3,,
ROTATE = LT-T-BALL-2,LT-T-BALL-1,,
ORIGIN = LT-T-BALL-1,,
ALIGNMENT/END
I ended up measuring the same thing twice and I hate that. I don't understand why a simple translation of my alignment in the X won't work.

[Sean Steele]

I got this to work without any remeasuring. Here is my code.
DCC-RH-ALN =ALIGNMENT/START,RECALL:M-ALN, LIST= YES
ALIGNMENT/TRANS,XAXIS,LH-ALIGN
ALIGNMENT/TRANS,YAXIS,LH-ALIGN
ALIGNMENT/TRANS,ZAXIS,LH-ALIGN
ALIGNMENT/TRANS_OFFSET,XAXIS,-79.98
ALIGNMENT/TRANS_OFFSET,YAXIS,-474.99
ALIGNMENT/TRANS_OFFSET,ZAXIS,-586.99
ALIGNMENT/END
A0 =ALIGNMENT/START,RECALLCC-RH-ALN, LIST= YES
ALIGNMENT/ITERATE
PNT TARGET RAD = 5, START LABEL = , FIXTURE TOL = 0, ERROR LABEL =
MEAS ALL FEAT = NO ,LEVEL AXIS=ZAXIS ,ROTATE AXIS=XAXIS ,ORIGIN AXIS=YAXIS
LEVEL = LH-ALIGN,LH-HEIGHT,LH-ORIGIN,,
ROTATE = LH-ORIGIN,LH-ALIGN,,
ORIGIN = LH-ALIGN,,
ALIGNMENT/END

Now I just have to figure out why the CAD didn't move to match the new alignment. Other than that, everything is working. What do you guys think?