CMM not taking hit at theoretical coordinate

[JABrown]

Why are my measured values of a vector point so far off from my theoretical values? I'm measuring points on a blade airfoil and for example my vector point theo is (X 1, Y 1, Z 1) w/ (I 0, J 1, Z 0) but when my program runs in takes the hit at (.994, .999, 1.001) (0,1,0) The .006 off in the X is a big difference in my world. I don't understand why the probe would hit that far off when it is approaching the part parallel to the Y-axis at the given X & Z coordinates.

[Matthew D. Hoedeman]

Sloppy machine, slack in the drive system. You might see better results with a larger prehit/retract value, give it more time to settle in.

[DAN_M]

Please post your alignment code and then code for the section of your program that is giving you trouble.

We'd like to help...but this will be very hard without more info.

[JABrown]

Here is my code:

ALN_PART_THEO_CENTER=ALIGNMENT/START,RECALL:TG37728_TBALL,LIST=YES
ALIGNMENT/TRANS_OFFSET,XAXIS,2.8661
ALIGNMENT/TRANS_OFFSET,YAXIS,-0.2178
ALIGNMENT/TRANS_OFFSET,ZAXIS,4.1724
ALIGNMENT/ROTATE_OFFSET,0.3,ABOUT,ZPLUS
ALIGNMENT/END
MOVE/POINT,NORMAL,<-3,0,10>
TIP/T1A90B0, SHANKIJK=0.007, 1, 0, ANGLE=-179.9977
ALIGN_START=LABEL/
MOVE/POINT,NORMAL,<0,2,5>
WORKPLANE/YPLUS
MOVE/POINT,NORMAL,<-.445,2,-0.4>
DAT_X1 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<-0.445,0.241,-0.4>,<0,1,0>
ACTL/<-0.4533,0.2531,-0.4035>,<0.0005148,0.9999997,0.000604>
TARG/<-0.445,0.241,-0.4>,<0,1,0>
SNAP=NO
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=BOTH,DISTANCE=0.25
SHOW HITS=NO
DAT_X3 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<0.36,0.202,-0.4>,<0,1,0>
ACTL/<0.3516,0.2137,-0.4041>,<0.0005148,0.9999997,0.000604>
TARG/<0.36,0.202,-0.4>,<0,1,0>
SNAP=NO
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=BOTH,DISTANCE=0.25
SHOW HITS=NO
MOVE/POINT,NORMAL,<0.36,2,-0.4>
MOVE/POINT,NORMAL,<0.36,2,0.3>
DAT_Z1 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<0.5984,0,-0.0731>,<0.1000067,-0.0048222,0.9949751>
ACTL/<0.5903,0.0113,-0.0773>,<0.1008071,-0.005475,0.9948909>
TARG/<0.5984,0,-0.073>,<0.1000067,-0.0048222,0.9949751>
SNAP=NO
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=BOTH,DISTANCE=0.25
SHOW HITS=NO
MOVE/POINT,NORMAL,<0.36,2,0.3>
DAT_X2 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<-0.155,0.5644,1.8605>,<0.2086474,0.9755214,-0.0694565>
ACTL/<-0.1613,0.575,1.8573>,<0.2090935,0.9754558,-0.0690357>
TARG/<-0.155,0.5644,1.86>,<0.2086474,0.9755214,-0.0694565>
SNAP=NO
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=BOTH,DISTANCE=0.25
SHOW HITS=NO
MOVE/POINT,NORMAL,<-1,2,1.8605>
DAT_Y2 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<-0.5266,0.1719,1.8605>,<-0.9774794,0.2102724,-0.0178762>
ACTL/<-0.5331,0.1827,1.8573>,<-0.9773851,0.2107859,-0.01696>
TARG/<-0.5266,0.172,1.86>,<-0.9774794,0.2102724,-0.0178762>
SNAP=NO
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=BOTH,DISTANCE=0.25
SHOW HITS=NO
MOVE/POINT,NORMAL,<-1.5,0.1,1.8605>
MOVE/POINT,NORMAL,<-1.5,0.1,-0.34>
DAT_Y1 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<-0.6621,-0.1,-0.34>,<-0.9961947,-0.000926,0.0871508>
ACTL/<-0.6705,-0.0878,-0.3432>,<-0.9961244,-0.0004663,0.0879545>
TARG/<-0.6621,-0.1,-0.34>,<-0.9961947,-0.000926,0.0871508>
SNAP=NO
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=BOTH,DISTANCE=1
SHOW HITS=NO
MOVE/POINT,NORMAL,<-1.5,0.1,-0.34>
MOVE/POINT,NORMAL,<-1.5,0.1,5>
MOVE/POINT,NORMAL,<0,2,5>
DCC_PART_ALIGN=ALIGNMENT/START,RECALL:ALN_PART_THEO_CENTER,LIST=YES
ALIGNMENT/ITERATE
PNT TARGET RAD=0.002,START LABEL=,FIXTURE TOL=0,ERROR LABEL=ALIGN_FAILED
MEAS ALL FEAT=ALWAYS,MAX ITERATIONS=4,LEVEL AXIS=YAXIS,ROTATE AXIS=XAXIS,ORIGIN AXIS=ZAXIS
LEVEL=DAT_X1,DAT_X3,DAT_X2,,
ROTATE=DAT_Y1,DAT_Y2,,
ORIGIN=DAT_Z1,,
ALIGNMENT/END
GOTO/ALIGN_PASSED

[JABrown]

The X is a known value at all of the points except DAT_Y1 & DAT_Y2 so I need the machine to hit there. The Z is known for all points except Dat_Z1 and the Y for all except DAT_Y1 and Dat_Y2

[ctw00]

JABrown,

The ACTL numbers shown are based their position relative to your previous alignment (ALN_PART_THEO_CENTER). From all of the blades and vanes I've inspected, an iterative alignment of the airfoil target points to the stacking axis is the correct way to go. The iterative alignment will repeat until it places all of the points within the target radius of .002. If you put a location dimension below your alignment your points should be very close to nominal. For aero and IGT blades/vanes I have always had good luck with a target radius of .010 and never saw any benefit going tighter than that.

[Matthew D. Hoedeman]

if you are talking about the results from an iterative alignment, yeah, you won't see exact values for the non-locating axis.

[RandomJerk]

Under Parameter Settings (F10), Probe options, look for positioning accuracy. This is how close to the target point the machine must come to call the move complete and move on. A lower number may cause a bit of cycle time increase, a really low number WILL cause cycle time increase.