hexagon logo

[2021.1] GeoTol messes with my nominals, why?

Running 2021.1, first encounter with GeoTols. Opened a program created in 2019 (I think) and upon running the program, the reported nominals are not the same as the feature nominals - why?



Loaded CAD to verify nominals were correct, still no go. Recreating the FCF doesn't change anything either.

Program:


START =ALIGNMENT/START,RECALL:USE_PART_SETUP,LIST=YES
ALIGNMENT/END
MODE/MANUAL
CHECK/3,1
FORMAT/TEXT,OPTIONS,ID,HEADINGS,SYMBOLS, ;MEAS,NOM,TOL,DEV,OUTTOL, ,
LOADPROBE/3MM
TIP/TIP1, SHANKIJK=0, 0, 1, ANGLE=0
COMMENT/OPER,NO,FULL SCREEN=NO,AUTO-CONTINUE=NO,
Mät ytterdiameter på vänster rörände

OBS! Börja med tre stödpunkter!
Ø22 =FEAT/CONTACT/CIRCLE/DEFAULT,CARTESIAN,OUT,LEAST_SQR
THEO/<0,0,0>,<0,0,1>,22,0
ACTL/<627.234,644.389,262.38>,<0.2742149,0.1856828,0.94 3572>,21.968,0
TARG/<0,0,0>,<0,0,1>
START ANG=0,END ANG=360
ANGLE VEC=<1,0,0>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
PKT Ø22 =FEAT/POINT,CARTESIAN,NO
THEO/<0,0,0>,<0,0,1>
ACTL/<627.234,644.389,262.38>,<0.2742149,0.1856828,0.94 3572>
CONSTR/POINT,CAST,Ø22
DATDEF/B,FEATURES=Ø22,,
A1 =ALIGNMENT/START,RECALL:START,LIST=YES
ALIGNMENT/TRANS,XAXIS,PKT Ø22
ALIGNMENT/TRANS,YAXIS,PKT Ø22
ALIGNMENT/TRANS,ZAXIS,PKT Ø22
ALIGNMENT/END
COMMENT/OPER,NO,FULL SCREEN=NO,AUTO-CONTINUE=NO,
Mät spår i konsol

OBS! Börja med tre stödpunkter!
SPÅR =FEAT/CONTACT/ROUND SLOT/DEFAULT,CARTESIAN,IN
THEO/<295.6,-104.3,-147.2>,<-0.0605674,0.5000144,0.8638965>,<0.997564,0.0003141 ,0.069757>,9,11.8
ACTL/<-245.688,-241.238,-30.76>,<0.7142378,0.0169459,0.6996979>,<0.3198843, 0.8812814,-0.3478752>,9.169,13.982
TARG/<295.6,-104.3,-147.2>,<-0.0605674,0.5000144,0.8638965>,<0.997564,0.0003141 ,0.069757>
MEAS ANGLE=160
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
PKT Q =FEAT/POINT,CARTESIAN,YES
THEO/<295.6,-104.3,-147.2>,<-0.0605674,0.5000144,0.8638965>
ACTL/<-245.688,-241.238,-30.76>,<0.7142378,0.0169459,0.6996979>
CONSTR/POINT,CAST,SPÅR
A2 =ALIGNMENT/START,RECALL:A1,LIST=YES
ALIGNMENT/BF2D,ZPLUS,LEAST_SQR,CREATE WEIGHTS=NO,ROTANDTRANS,USE SCALING=NO,0,0,0,26.844
ITERATEANDREPIERCECAD=NO
Deviation Threshold=5,Pause Execution=NO
SHOWALLINPUTS=NO,SHOWALLPARAMS=NO
ALIGNMENT/END
WORKPLANE/YPLUS
A3 =ALIGNMENT/START,RECALL:A2,LIST=YES
ALIGNMENT/BF2D,YPLUS,LEAST_SQR,CREATE WEIGHTS=NO,ROTANDTRANS,USE SCALING=NO,0,0,0,21.172
ITERATEANDREPIERCECAD=NO
Deviation Threshold=5,Pause Execution=NO
SHOWALLINPUTS=NO,SHOWALLPARAMS=NO
ALIGNMENT/END
WORKPLANE/ZPLUS
COMMENT/OPER,NO,FULL SCREEN=NO,AUTO-CONTINUE=NO,
Mät ytterdiameter på höger rörände

OBS! Börja med tre stödpunkter!
HÖ RÖRÄNDE =FEAT/CONTACT/CIRCLE/DEFAULT,CARTESIAN,OUT,LEAST_SQR
THEO/<253,-142.5,-171.7>,<-0.2686794,-0.9630455,-0.0188342>,18,0
ACTL/<-285.276,-50.326,-175.445>,<-0.4032534,0.9128111,-0.0645188>,18.067,0
TARG/<253,-142.5,-171.7>,<-0.2686794,-0.9630455,-0.0188342>
START ANG=0,END ANG=360
ANGLE VEC=<-0.0699277,0,0.9975521>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
COMMENT/OPER,NO,FULL SCREEN=NO,AUTO-CONTINUE=NO,
Mät ytterdiameter på Ø16

OBS! Börja med tre stödpunkter!
Ø16 =FEAT/CONTACT/CIRCLE/DEFAULT,CARTESIAN,OUT,LEAST_SQR
THEO/<142.3,56.6,-226.6>,<0.4444073,0.8166771,-0.3681585>,16,0
ACTL/<-74.943,-141.655,-220.667>,<0.1179681,-0.9436953,-0.3090674>,16.024,0
TARG/<142.3,56.6,-226.6>,<0.4444073,0.8166771,-0.3681585>
START ANG=0,END ANG=360
ANGLE VEC=<-0.6379514,0,-0.7700766>
DIRECTION=CCW
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=NO
WORKPLANE/ZPLUS
A4 =ALIGNMENT/START,RECALL:A3,LIST=YES
ALIGNMENT/BF3D,LEAST_SQR,CREATE WEIGHTS=NO,ROTANDTRANS,USE SCALING=NO,-1.755,3.229,-0.217,-141.985,175.886,-1.101
ITERATEANDREPIERCECAD=NO
Deviation Threshold=5,Pause Execution=NO
SHOWALLINPUTS=NO,SHOWALLPARAMS=NO
ALIGNMENT/END
REF A =FEAT/PLANE,CARTESIAN,TRIANGLE,NO
THEO/<0,0,0>,<0,0,1>
ACTL/<0,0,0>,<0,0,1>
CONSTR/PLANE,ALIGN,ZPLUS
DATDEF/A,FEATURES=REF A,,
REF C =FEAT/PLANE,CARTESIAN,TRIANGLE,NO
THEO/<0,0,0>,<0,1,0>
ACTL/<0,0,0>,<0,1,0>
CONSTR/PLANE,ALIGN,YPLUS
DATDEF/C,FEATURES=REF C,,
DIM DIM3= LOCATION OF CIRCLE Ø22 UNITS=MM ,$
GRAPH=OFF TEXT=OFF MULT=10.00 OUTPUT=NONE HALF ANGLE=NO
AX MEAS NOMINAL +TOL -TOL DEV OUTTOL
X 0.610 0.000 0.000 0.000 0.610 0.610 ->
Y 3.631 0.000 0.000 0.000 3.631 3.631 ->
Z 0.000 0.000 0.000 0.000 0.000 0.000 ->
END OF DIMENSION DIM3
DIM DIM2= LOCATION OF POINT PKT Q UNITS=MM ,$
GRAPH=OFF TEXT=OFF MULT=10.00 OUTPUT=NONE HALF ANGLE=NO
AX MEAS NOMINAL +TOL -TOL DEV OUTTOL
X 294.187 295.600 0.000 0.000 -1.413 -1.413 <-
Y -104.300 -104.300 0.000 0.000 0.000 0.000 ->
Z -147.200 -147.200 0.000 0.000 0.000 0.000 ->
END OF DIMENSION DIM2
COMMENT/REPT,
Mått 295.6 (±2)
DIM DIM4= LOCATION OF SLOT SPÅR UNITS=MM ,$
GRAPH=OFF TEXT=OFF MULT=10.00 OUTPUT=BOTH HALF ANGLE=NO
AX MEAS NOMINAL +TOL -TOL DEV OUTTOL
X 294.187 295.600 2.000 -2.000 -1.413 0.000 #-
END OF DIMENSION DIM4
COMMENT/REPT,
Lägeriktighet inom Ø2(M) rel. A/C för Ø22
FCFDIM1 =GEOMETRIC_TOLERANCE/STANDARD=ISO 1101,SHOWEXPANDED=NO,
SIZE/NOMINAL=22,TOLERANCE SPECIFICATION MODE=NOMINAL_WITH_DEVIATIONS,
UPPER TOLERANCE=0.15,LOWER TOLERANCE=-0.15,
SEGMENT_1,POSITION,DIAMETER,2,MMC,A,C,<dat>,
ADD
FEATURES/Ø22,,
COMMENT/REPT,
Lägeriktighet inom Ø4(M) rel. A/C för Ø16
FCFDIM2 =GEOMETRIC_TOLERANCE/STANDARD=ISO 1101,SHOWEXPANDED=NO,
SIZE/NOMINAL=16,TOLERANCE SPECIFICATION MODE=NOMINAL_WITH_DEVIATIONS,
UPPER TOLERANCE=0.1,LOWER TOLERANCE=-0.1,
SEGMENT_1,POSITION,DIAMETER,4,MMC,A,C,<dat>,
ADD
FEATURES/Ø16,,
COMMENT/REPT,
Lägeriktighet inom Ø4(M) rel. A/C för Ø18
FCFDIM3 =GEOMETRIC_TOLERANCE/STANDARD=ISO 1101,SHOWEXPANDED=NO,
SIZE/NOMINAL=18,TOLERANCE SPECIFICATION MODE=NOMINAL_WITH_DEVIATIONS,
UPPER TOLERANCE=0.1,LOWER TOLERANCE=-0.1,
SEGMENT_1,POSITION,DIAMETER,4,MMC,A,C,<dat>,
ADD
FEATURES/HÖ RÖRÄNDE,,


​​​​​Ping neil.challinor


  • In short, the nominals displayed are the closest nominals to the measured point within the tolerance zone?


    That is correct
  • Trying to befriend GeoTol, but it is not going well...

    TP in one axis, expecting the nominal to be nominal, but it's not. This slot is parallel to the X-axis and perp to the Y-axis, so I was expecting the nominal to be... well, nominal:



    The slot should be nominal @ 387, so how should I view this report? What should I tell the machinists how to correct the part? I have zero deviation from nominal, so it should be green and within tolerance, but the nominal is wrong... Calculating 387 - 388.348 = 1.348 * 2 = 2.696 = within tolerance... instead the deviation is 9.51 which should be in the vicinity of a 4.25 mm deviation...

    I would rather see that PC-DMIS *never* changes the nominals - ever.
    I can take user error every day, I ain't proud like that, but this feels funky... Am I the only one in here running 2021.1 who gets these weird things happening??


    Program and report attached.

    Attached Files
  • and you have pressed in vinkelrät mot centrumlinjen??
    i had some trouble when pcmis calculate whit 2 or 3 axis when selecting 1 or 2 axis.
    have 2019R1 so cant look at your program.


  • I took a look at your attached .prg and it looks like the problem is being caused by an incorrect angle vector on slot Y2Z5. It was set to the same value as the surface vector (0,1,0). The geometric tolerance command is set to report the length-wise position of the slot but because of the nominal angle vector, it is not able to determine it correctly. You can see there is an issue by looking at the graphic in the report - it shows a different orientation for Y2Z5 than Y1Z4 whereas based on the measured values in the features they should have the same orientation.

    Changing the angle vector to 1,0,0 resolves the error (see below)

    Corrected slot command
    Y2Z5 =FEAT/CONTACT/ROUND SLOT/DEFAULT,CARTESIAN,IN
    THEO/<387,0,0>,<0,1,0>,<1,0,0>,9,15
    ACTL/<311.511,-0.373,228.484>,<-0.0025042,0.9999958,-0.0014923>,<0.8043944,0.0029009,0.5940886>,9.501,1 5.454
    TARG/<387,0,0>,<0,1,0>,<1,0,0>
    MEAS ANGLE=160
    SHOW FEATURE PARAMETERS=NO
    SHOW CONTACT PARAMETERS=NO


    Re-generated report following slot correction

  • Correct, neil.challinor ! Thank you for your eyes, my coworker later admitted he changed the vector (on-the-job-training) and I didn't fully check the code... Dang it!

    But...

    The last evaluations in the report is actually for the same slot - just in different directions:


    This slot is not perp or parallel to the axises, so I guess the "closest nominal to the measured point" applies here, but this will not fly at all
    when we file a non-conformance material report. The first our suppliers will react upon is that our nominals are wrong, "please correct and re-evaluate"
    and there is no way for us to do this unless reverting back to legacy I guess. But, I was expecting to see the same nominals in both evaluations,
    because the slot hasn't moved. Same datums, same slot, same center but different nominals? Y-difference of ~10 mm, Z-difference of ~5.5 mm...
    We simply cannot educate our suppliers in order to interpret our result reports.

    Is there a way in 2021.1 to "go back" to the old style and have the nominals exactly as entered, this must surely be important for machinists especially?

    Or is my understanding of GeoTol all wrong? I need the deviation from the nominal point defined by the drawing/CAD and the direction of the deviation (/wipes tears),
    not the distance to the closest nominal point within the tolerance zone.



    Kudos for the slot TP direction selection icon in GeoTol, makes it soo much easier!
  • neil.challinor Any way to go back to the "old way" of reporting or got any ETA's of a possible fix? As it stands now, I can't defend the 2021.1 installation...

    Or is it me that just don't understand it?
  • I was out last week, hence the late reply. There is no way to change the nominals behaviour at the moment, other than using legacy. As I mentioned in my earlier post, we are in the process of making changes to the way nominals are handled for the geometric tolerance command and are currently targeting those changes 2021.2 (subject to successful testing).
  • Thanks neil.challinor !

    Got a possible ETA of the 2021.2 release?
  • I cannot use it because of this. Nominals should never change. I will not be upgrading past 2020 R1 until this is fixed. Makes absolutely no sense.