hexagon logo

In Car position/Iterative Alignment

Genius' of the PCDMIS Forum, good morning and I present to you another aggravating problem that landed on my desk.

I have a unique situation. Our forming branch needs a program made for a window trim, assumed it'd be easy enough as its just reporting some XYZ set profile points with in car position XYZ set coordinates. I have done this before however they require utilization of a slot for the alignment and the only other time I did this, was for a gage fixture that had 3 spheres that gave the offsets for the in-car coordinate.

After some research on here I was able to discern that I would likely need to use a constructed point off center of the slot (thanks Matt), and an iterative alignment. I have refreshed my memory with them via the training and lab offered in CMM 211 extra content (has a lab with Manual/DCC iterative using world coordinate), however am still failing miserably. I even copied it near verbatim with no success. I was considering downloading the CAD and IGS file with points to see if I can even replicate it to see where I am messing up.

I was originally gonna cheat and just use the f5 "lock car coordinate" toggle so the alignment never moves away from that setting, however I am unsure how to align the part in order to run clean in DCC with that being set to locked world coordinate system (keeps telling me I am out of machine capability). That's what I assumed the iterative alignment was going to fix but once I hit the features manually even with a massive point target radius, the circle and slot they use to rotate and origin end up larger/not where they are on the CAD (almost like when hitting manually in the wrong workplane). It then DCC's over to a random location on the table after telling me to place the probe near my initially hit point. I used read vector and flipped them when it was facing the inside, and have the part fixtured on 2 posts with spring clamps with the slot and hole facing my table X- with the probe in A90B90 so it can hit all required alignment targets.

Attached is a picture of the alignment scheme to get an idea of it as well as a snapshot of what happening after I manually align. The 3 points they use for the planar alignment are on a long curved feature, which is skewing the alignment I think, as the vector's were not given to me.

Program Code below (not functioning)

__________________________________________________ ___
PART NAME : WINDOW RAIL REAR (LH)
REV NUMBER : 02
SER NUMBER : 78593265 RH
STATS COUNT : 1

STARTUP =ALIGNMENT/START,RECALL:USE_PART_SETUP,LIST=YES
ALIGNMENT/END
MODE/MANUAL
PREHIT/0.8
RETRACT/0.8
MOVESPEED/ 500
MANRETRACT/1.5
FORMAT/TEXT,OPTIONS, ,HEADINGS,SYMBOLS, ;NOM,TOL,MEAS,DEV,OUTTOL, ,
LOADPROBE/TC3_2MMX40MM
TIP/T1A90B90, SHANKIJK=-1, 0, 0, ANGLE=90
PNT1 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<3111.95,1045.66,1793.8>,<0.1039575,-0.9905949,-0.0889636>
ACTL/<388.454,1059.505,-245.271>,<-0.9979487,0.0048044,-0.0638382>
TARG/<3111.95,1045.66,1793.8>,<0.1039575,-0.9905949,-0.0889636>
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=NO,DISTANCE=10
SHOW HITS=NO
PNT2 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<3108.75,1029.42,1988.21>,<0.1044568,-0.9905058,-0.0893706>
ACTL/<419.196,303.223,-766.675>,<-0.999997,0.00245,0>
TARG/<3108.75,1029.42,1988.21>,<0.1044568,-0.9905058,-0.0893706>
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=NO,DISTANCE=10
SHOW HITS=NO
PNT3 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<3124.77,1001.93,2267.42>,<0.1036899,-0.9880009,-0.1144669>
ACTL/<415.904,577.075,-759.519>,<-0.9999993,0.0008247,-0.0008247>
TARG/<3124.77,1001.93,2267.42>,<0.1036899,-0.9880009,-0.1144669>
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=NO,DISTANCE=10
SHOW HITS=NO
CIR1 =FEAT/CONTACT/CIRCLE/DEFAULT,CARTESIAN,IN,LEAST_SQR
THEO/<3119,1004.65,2238.9>,<0.103662,-0.9873932,-0.1196197>,8.1
ACTL/<335.907,1068.95,-703.679>,<0.103662,-0.9873932,-0.1196197>,55.148
TARG/<3119,1004.65,2238.9>,<0.103662,-0.9873932,-0.1196197>
START ANG=0,END ANG=-360
ANGLE VEC=<-0.7557154,0,-0.6549002>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
NUMHITS=4,DEPTH=0.6,PITCH=0
SAMPLE METHOD=SAMPLE_HITS
SAMPLE HITS=0,SPACER=0
AVOIDANCE MOVE=NO,DISTANCE=10
FIND HOLE=DISABLED,ONERROR=NO,READ POS=NO
SHOW HITS=NO
PNT4 =FEAT/POINT,CARTESIAN,NO
THEO/<3119,1004.65,2238.9>,<0.103662,-0.9873932,-0.1196197>
ACTL/<335.907,1068.95,-703.679>,<0.103662,-0.9873932,-0.1196197>
CONSTR/POINT,CAST,CIR1
SLTR1 =FEAT/CONTACT/ROUND SLOT/DEFAULT,CARTESIAN,IN
THEO/<3110.95,1030.66,1977.08>,<0.103662,-0.9873932,-0.1196197>,<0.0320666,-0.1168873,0.9926274>,8.1,12
ACTL/<339.228,1060.674,-672.636>,<0.103662,-0.9873932,-0.1196197>,<0.0267596,-0.1174554,0.9927176>,2.993,9.214
TARG/<3110.95,1030.66,1977.08>,<0.103662,-0.9873932,-0.1196197>,<0.0320666,-0.1168873,0.9926274>
MEAS ANGLE=90
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
NUMHITS=6,DEPTH=0.7
SAMPLE METHOD=SAMPLE_HITS
SAMPLE HITS=0,SPACER=0
AVOIDANCE MOVE=NO,DISTANCE=10
FIND HOLE=DISABLED,ONERROR=NO,READ POS=NO
SHOW HITS=NO
PNT5 =FEAT/POINT,CARTESIAN,NO
THEO/<3110.95,1030.66,1977.08>,<0.103662,-0.9873932,-0.1196197>
ACTL/<339.228,1060.674,-672.636>,<0.103662,-0.9873932,-0.1196197>
CONSTR/POINT,CAST,SLTR1
A1 =ALIGNMENT/START,RECALL:STARTUP,LIST=YES
ALIGNMENT/ITERATE
PNT TARGET RAD=5,START LABEL=,FIXTURE TOL=0,ERROR LABEL=
MEAS ALL FEAT=NO,LEVEL AXIS=YAXIS,ROTATE AXIS=XAXIS,ORIGIN AXIS=ZAXIS
LEVEL=PNT1,PNT2,PNT3,,
ROTATE=PNT5,PNT4,,
ORIGIN=PNT4,,
ALIGNMENT/END

Attached Files
Parents
  • you MUST use 3 surface sample hits for the hole/slot features, even for the manual iterative alignment, otherwise the 'through' axis will be in machine co-ordinates.


    Hey it's the legend himself! Love reading your past answers on the same subject (I even felt it was necessary to mention you for such knowledge)

    Gonna try that and will report back. If that's all it is, I am going to be elated.
Reply
  • you MUST use 3 surface sample hits for the hole/slot features, even for the manual iterative alignment, otherwise the 'through' axis will be in machine co-ordinates.


    Hey it's the legend himself! Love reading your past answers on the same subject (I even felt it was necessary to mention you for such knowledge)

    Gonna try that and will report back. If that's all it is, I am going to be elated.
Children
No Data