hexagon logo
Register Sign in
  • Replies 6 replies
  • Subscribers 221 subscribers
  • Views 109 views
  • Users 0 members are here
Options
Related

Measuring Point Profile from THEO and ACTL Values

UgurEmanetoglu1
Hello! First of all, I would like to mark that I am not experienced in operating CMM.

I wanted to check the coordinates of our measurement and thus required a detailed log of the process from the operator. I am assuming that THEO and ACTL values are theoretical and actual coordinates of the points. Also checked if my assumption is true on Nexus Hexagon docs.

Anyways, I started to calculate the distances between the THEO and ACTL coordinates. Then compared to the ones given as profiles on the report.

PNT11 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<-4.524,21.406,-5.05>,<0.2517976,0.1223556,-0.9600141>
ACTL/<-4.529,21.41,-5.036>,<0.2517976,0.1223556,-0.9600141>
TARG/<-4.524,21.406,-5.05>,<0.2517976,0.1223556,-0.9600141>
SNAP=NO
SHOW FEATURE PARAMETERS=YES
SURFACE=THICKNESS_NONE,0
MEASURE MODE=NOMINALS
RMEAS=NONE,NONE,NONE
AUTO WRIST=NO
GRAPHICAL ANALYSIS=NO
FEATURE LOCATOR=NO,NO,""
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=BOTH,DISTANCE=0
SHOW HITS=NO
MOVE/CLEARPLANE​




According to this calculation, i thought that my measurement should be off by 0.015 mm. However, as you can see below, the result was 0.019 mm off.

​


So, at this point i am curious how deviation is calculated by PC-DMIS. Can anyone care to explain?

Thanks in advance!
  • OK, your calculation is a 'little' off as you are ignoring the vector values of the point. IF you turn SNAP=ON, it will remove any machine drift from the ACTL values, and then your formula will (SHOULD) work correctly.


  • https://www.pcdmisforum.com/forum/pc-dmis-enterprise-metrology-software/pc-dmis-for-cmms/30290-measurement-properties-snap-auto-features

    Considering your explanation about SNAP feature, if we turned SNAP ON and Y and Z coordinates were the ones we cared about, while X is not, would it give an output as below?

    THEO/<-4.524, 21.406, 5.05>,<0.2517977,0.1223557,0.9600141>
    ACTL/<-4.524, 21.407, 5.022>,<0.2517977,0.1223557,0.9600141 >​
  • On first post I forgot to add what puzzled me.

    PNT11 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
    THEO/<-4.524, 21.406, -5.05>,<0.2517976,0.1223556,-0.9600141>
    ACTL/<-4.529, 21.41, -5.036>,<0.2517976,0.1223556,-0.9600141>
    TARG/<-4.524,21.406,-5.05>,<0.2517976,0.1223556,-0.9600141>
    SNAP=NO
    SHOW FEATURE PARAMETERS=YES
    SURFACE=THICKNESS_NONE,0
    MEASURE MODE=NOMINALS
    RMEAS=NONE,NONE,NONE
    AUTO WRIST=NO
    GRAPHICAL ANALYSIS=NO
    FEATURE LOCATOR=NO,NO,""
    SHOW CONTACT PARAMETERS=YES
    AVOIDANCE MOVE=BOTH,DISTANCE=0
    SHOW HITS=NO
    MOVE/CLEARPLANE


    PNT12 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
    THEO/<-4.524, 18.708, -11.565>,<0.2517963,0.7653502,0.5923156>
    ACTL/<-4.535, 18.679, -11.582>,<0.2517963,0.7653502,0.5923156>
    TARG/<-4.524,18.708,-11.565>,<0.2517963,0.7653502,0.5923156>
    SNAP=NO
    SHOW FEATURE PARAMETERS=YES
    SURFACE=THICKNESS_NONE,0
    MEASURE MODE=NOMINALS
    RMEAS=NONE,NONE,NONE
    AUTO WRIST=NO
    GRAPHICAL ANALYSIS=NO
    FEATURE LOCATOR=NO,NO,""
    SHOW CONTACT PARAMETERS=YES
    AVOIDANCE MOVE=BOTH,DISTANCE=0
    SHOW HITS=NO
    MOVE/CLEARPLANE


    Now, when i do the same calculation for distance between PNT11 to PNT12;

    ​

    I get the exact measurement obtained from the report.​

    ​
  • in reply to UgurEmanetoglu1
    SNAP=ON projects the measured point onto the nominal 3D vector line perpendicular to that 3D vector line.

    You will see this extremely well if you measure a point with a 'perfect' vector (0,0,1 or 0,1,0 or 1,0,0). There shouldn't be any deviation in any axis with a zero vector. BUT, no machine is perfect and there will be some drift. SNAP=ON removes the machine drift from the results. Your point has vectors in all 3 axis, so it will not make any of them show zero deviation. BUT, by using it, the sqrt(Xdev^2+Ydev^2+Zdev^2) will equal the T value and SHOULD equal the profile value for that single point. BUT, for single points, it is easier to just dimension XYZT at +/- 1/2 of the tolerance (for equally disposed)
  • in reply to UgurEmanetoglu1
    I'm not sure if you meant to do this, but your calculator data input looks wrong if you're using the actuals.

    X1 = -4.535, Y1= 18.679, Z1= -11.582

    X2= -4.524, Y2 = 21.41, Z2= -5.036

    Your X2 should be 4.529 

    PNT11 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<-4.524, 21.406, -5.05>,<0.2517976,0.1223556,-0.9600141>
ACTL/<-4.529, 21.41, -5.036>,<0.2517976,0.1223556,-0.9600141>
TARG/<-4.524,21.406,-5.05>,<0.2517976,0.1223556,-0.9600141>

PNT12 =FEAT/CONTACT/VECTOR POINT/DEFAULT,CARTESIAN
THEO/<-4.524, 18.708, -11.565>,<0.2517963,0.7653502,0.5923156>
ACTL/<-4.535, 18.679, -11.582>,<0.2517963,0.7653502,0.5923156>
TARG/<-4.524,18.708,-11.565>,<0.2517963,0.7653502,0.5923156>



    Not that it matters much, your math came out to be the same

  • So with SNAP ON, when I test for coordinates, say (1,2,5) and the probe touches at coordinates (1.1, 2.1, 4.8), they will be snapped onto a point with the closest distance and on the x=1t, y=2t, z=5t vector. If I understood correctly, I assume deviation calculation is done as if the SNAP was ON behind the scenes. And actual coordinates fall onto a coordinate 0.19 mm away from theoretical ones, which is lying on x=-4.524t, y=21.406t, z=-5.05t vector.



    ​
    Thank you for wasting your time on a CMM illiterate. Smiley Having someone walking me through this stuff is great after brain-dead people working in my company.
© 2024 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement