hexagon logo

[2022.1 SP7] Adaptive Circle Scan, Variables, & Clearplanes

[2022.1 SP7] - Just updated from 2021.2 this week. ***Title states (fixed now) SP4, which is an error. I am running SP7.***

Odd issue (which I solved). I have a program that runs an adaptive circle scan on a small arc (20°), which uses clearplanes to hop around the part. No issues with this program, until this update.

I found that the software would just "bypass" the clearplanes, which would cause the probe to run into the part flying at 425mm/s. To add, Path Lines showed the clearplanes to be working. This only happened during actual execution of the program on the physical CMM.

TIP/T1A0B0, SHANKIJK=-0.001, 0, 1, ANGLE=-0.281
CLEARP/ZPLUS,25,ZPLUS,0,ON
DIM LOC1= LOCATION OF PLANE DATUM_B UNITS=MM ,$
GRAPH=OFF TEXT=OFF MULT=10.00 OUTPUT=NONE HALF ANGLE=NO
AX NOMINAL MEAS +TOL -TOL DEV OUTTOL
X 9.271 9.443 0.381 0.000 0.172 0.000 ---#----
END OF DIMENSION LOC1
ASSIGN/V9=0.783+(LOC1.X.DEV/4)
MOVE/CLEARPLANE
CIR5 =FEAT/CONTACT/CIRCLE/ADAPTIVE_CIRCLE_SCAN,CARTESIAN,IN,LEAST_SQR
THEO/<V9,0,19.87>,<0,-1,0>,19.05
ACTL/<-0.716,0,19.1>,<0,-1,0>,16.745
TARG/<V9,0,19.87>,<0,-1,0>
START ANG=230,END ANG=250
ANGLE VEC=<1,0,0>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
MOVE/CLEARPLANE
CIR6 =FEAT/CONTACT/CIRCLE/ADAPTIVE_CIRCLE_SCAN,CARTESIAN,IN,LEAST_SQR
THEO/<V9,19.87,0>,<0,0,1>,19.05
ACTL/<-0.613,19.043,0>,<0,0,1>,16.676
TARG/<V9,19.87,0>,<0,0,1>
START ANG=230,END ANG=250
ANGLE VEC=<1,0,0>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
MOVE/CLEARPLANE
CIR8 =FEAT/CONTACT/CIRCLE/ADAPTIVE_CIRCLE_SCAN,CARTESIAN,IN,LEAST_SQR
THEO/<V9,-19.87,0>,<0,0,-1>,19.05
ACTL/<-0.419,-19.727,0>,<0,0,-1>,18.164
TARG/<V9,-19.87,0>,<0,0,-1>
START ANG=230,END ANG=250
ANGLE VEC=<1,0,0>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO​


What I found out is that the V9 variable in the X was the culprit WHEN it was an adaptive circle scan. Once I changed it to TTP, the program worked fine. But I want to keep the adaptive circle scan because I don't want to take 11 hits using TTP at 2mm/s touch speed. Slows down the program.

Solution? Plug V9 into a generic feature, align to that, set my X to 0, and it works:

TIP/T1A0B0, SHANKIJK=-0.001, 0, 1, ANGLE=-0.281
CLEARP/ZPLUS,25,ZPLUS,0,ON
DIM LOC1= LOCATION OF PLANE DATUM_B UNITS=MM ,$
GRAPH=OFF TEXT=OFF MULT=10.00 OUTPUT=NONE HALF ANGLE=NO
AX NOMINAL MEAS +TOL -TOL DEV OUTTOL
X 9.271 9.444 0.381 0.000 0.173 0.000 ---#----
END OF DIMENSION LOC1
ASSIGN/V9=0.783+(LOC1.X.DEV/4)
F1 =GENERIC/POINT,DEPENDENT,CARTESIAN,$
NOM/XYZ,<V9,0+0,0+0>,$
MEAS/XYZ,<V9,0+0,0+0>,$
NOM/IJK,<0+0,0+0,1+0>,$
MEAS/IJK,<0+0,0+0,1+0>
R9525_ALN =ALIGNMENT/START,RECALL:FIX_ALN,LIST=YES
ALIGNMENT/TRANS,XAXIS,F1
ALIGNMENT/END
MOVE/CLEARPLANE
CIR5 =FEAT/CONTACT/CIRCLE/ADAPTIVE_CIRCLE_SCAN,CARTESIAN,IN,LEAST_SQR
THEO/<0,0,19.87>,<0,-1,0>,19.05
ACTL/<-1.562,0,18.981>,<0,-1,0>,16.5
TARG/<0,0,19.87>,<0,-1,0>
START ANG=230,END ANG=250
ANGLE VEC=<1,0,0>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
MOVE/CLEARPLANE
CIR6 =FEAT/CONTACT/CIRCLE/ADAPTIVE_CIRCLE_SCAN,CARTESIAN,IN,LEAST_SQR
THEO/<0,19.87,0>,<0,0,1>,19.05
ACTL/<-1.396,19.346,0>,<0,0,1>,17.375
TARG/<0,19.87,0>,<0,0,1>
START ANG=230,END ANG=250
ANGLE VEC=<1,0,0>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
MOVE/CLEARPLANE
CIR8 =FEAT/CONTACT/CIRCLE/ADAPTIVE_CIRCLE_SCAN,CARTESIAN,IN,LEAST_SQR
THEO/<0,-19.87,0>,<0,0,-1>,19.05
ACTL/<-1.644,-18.883,0>,<0,0,-1>,16.238
TARG/<0,-19.87,0>,<0,0,-1>
START ANG=230,END ANG=250
ANGLE VEC=<1,0,0>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
RECALL/ALIGNMENT,INTERNAL,FIX_ALN​


I will submit a Hexagon ticket on this, but I figured I would let Don Ruggieri and the rest of the forum know of my finding in case they have a similar situation. ​
  • ***Title states SP4, which is an error. I am running SP7.***

    I just downloaded installed 2022.1 #455, which is SP7. What build # did you install? Can you provide a screen capture indicating SP4? Go to Help - About and send a capture of that dialog please.
  • ***Title states SP4, which is an error. I am running SP7.***

    I just downloaded installed 2022.1 #455, which is SP7. What build # did you install? Can you provide a screen capture indicating SP4? Go to Help - About and send a capture of that dialog please.


    Can you please change the title to SP7? I am running Build #455.

    Matt Beecher at Hexagon got back to me and did state in my case: "While not this exact issue, I did file a report related to use of variables in autofeatures within 2022.1, which was resolved in SP5. This makes me wonder if your issue could have been related to what I filed back then. Of course, it could also be a separate issue."

    So this could quite possibly be a bug that has crept it's way back into the system?
  • Change what title? This is what I see -




    Can you please capture what you are seeing where it says SP4?

    PS, Sorry, I get it now, the title of the post, not the software
  • On to the subject matter, we have some questions. What kind of probe assembly are you using? Your code shows T1A0B0 but your features have three very different vectors and you never change tips?


    THEO/<0,0,19.87>,<0,-1,0>,19.05

    THEO/<0,19.87,0>,<0,0,1>,19.05

    THEO/<0,-19.87,0>,<0,0,-1>,19.05​
  • On to the subject matter, we have some questions. What kind of probe assembly are you using? Your code shows T1A0B0 but your features have three very different vectors and you never change tips?


    THEO/<0,0,19.87>,<0,-1,0>,19.05

    THEO/<0,19.87,0>,<0,0,1>,19.05

    THEO/<0,-19.87,0>,<0,0,-1>,19.05​


    *Side Note: This is a very old program. I just noticed the -1 in Z. Doesn't affect the program due to the angle vector being correct, but I will adjust that to clean up from previous programmer mistakes.

    1mm cylinder probe. I'd post a picture, but I cannot. I am measuring a round conical-shaped part with many different radial contours. Part sits on it's side in X, nested in a holding fixture. I measure a small arc in A0B0 in Y+, Z+, and Y-. The clearplane does not work when I move to A0B0 into the first scan right off the bat. However, when changed to TTP or with the fix I posted, it works great.
  • T1A0B0 implies that the tip is pointing down (Z-) on the typical bridge CMM. What direction is it pointing for you? What kind of machine do you have? Since the Y and Z axes are 90 degrees apart, how can you measure both with this one tip?
  • Don Ruggieri

    See the attached for how the part gets measured and why A0B0 is measuring the way it does. Current view is in X. Probe moves in the X+ direction to measure the radius on the outside of the part along the cylindrical cone.

    Programming code aside, it still does not negate why when TTP is enabled and my fix above (using adaptive circle scans), clearplanes work. However, when Adaptive Circle Scan is enabled with a variable in X, the software wants to bypass any clearplanes for these features.

    Attached Files
  • "Programming code aside, it still does not negate why when TTP is enabled and my fix above (using adaptive circle scans), clearplanes work."

    I agree, I understand that. As of yet we have not been able to replicate your issue. We want to fully understand what you are doing in order to be able to repeat it. In your illustration it seems you are measuring a sphere, yet the three features you've shared code for are circles.

    Is there something you could write on a more generic part that you can share with us? A ring gage? a sphere? A demo block? etc.