Recording system data from a FANUC robot with the Karel program

    Hello,

    I tried to program this Karel function to extract the data from the system of a FANUC LR mate 200ic 5L robot and save this data in a ".dt" file.

    But unfortunately, I got only an empty file and there is no result in the file, only that :

    03-MAR-22 17:33

    0;

    And here is an example of the file I got

    can you help me find out where the issue is coming from or can you give me another solution to save in real time the system data like internal joint torque, joint speed, etc...

    Thanks in advance

    Test.txt

    When you run the program are you getting uninitialized data on line 125 of your karel program?

    The Torque values stored in $MOR_GRP[1].$CUR_TORQUE[] are some cursed UINT Values as detailed by this thread: FANUC PROJECT: Data recording with background Karel program - Fanuc Robot Forum - Robotforum - Support and discussion community for industrial robots and cobots (robot-forum.com) I think trying to get/convert/write these UINT Values is causing your issue.

    Judging by the variable names you used, you've already looked over part of this thread, however, I don't think the thread landed on a Karel happy solution. If you remove the torque logging from your program the rest works just fine.

    Quote from Erik Olsen

    When you run the program are you getting uninitialized data on line 125 of your karel program?

    The Torque values stored in $MOR_GRP[1].$CUR_TORQUE[] are some cursed UINT Values as detailed by this thread: FANUC PROJECT: Data recording with background Karel program - Fanuc Robot Forum - Robotforum - Support and discussion community for industrial robots and cobots (robot-forum.com) I think trying to get/convert/write these UINT Values is causing your issue.

    Judging by the variable names you used, you've already looked over part of this thread, however, I don't think the thread landed on a Karel happy solution. If you remove the torque logging from your program the rest works just fine.

    Yes I tried to run the program without the part of torque, but it didn't work

    So, is there any other method to save the values in real time more reliable than this one ?

    Quote from AnisFrej

    Yes I tried to run the program without the part of torque, but it didn't work

    I would try again and make sure you remove/comment out where you write the torque to the log file. I was able to run your program and make a valid angle log using the following code:

    If you still want to log the $CUR_TORQUE, $TORQUE or $MAX_TORQUE Variables you need to use the correct DATA types. These system variables are stored as SHORT data types and your GET_VAR instructions were trying to write into REAL variables. SHORT data needs to be assigned to BYTE or INTEGER variables. After changing your data types and a couple of CNV instructions I was able to make a valid log containing both Angle and CUR_TORQUE values:

    Keep in mind, as my previous reply states, these TORQUE values are not very useful unless scaled and I haven't found any forum threads detailing how this would be done.

    Quote from Erik Olsen

    I would try again and make sure you remove/comment out where you write the torque to the log file. I was able to run your program and make a valid angle log using the following code:

    If you still want to log the $CUR_TORQUE, $TORQUE or $MAX_TORQUE Variables you need to use the correct DATA types. These system variables are stored as SHORT data types and your GET_VAR instructions were trying to write into REAL variables. SHORT data needs to be assigned to BYTE or INTEGER variables. After changing your data types and a couple of CNV instructions I was able to make a valid log containing both Angle and CUR_TORQUE values:

    Thank you very much for your precise answer :smiling_face: :smiling_face: :smiling_face: , but in the obtained file some values are saved as * as in the attached file from you .

    So do you have any idea how to extract the value correctly?

    And do you have any idea which variables I should use to extract or save the speed of the motors or encoders and the speed of the joints ?

    Quote from AnisFrej

    Thank you very much for your precise answer :smiling_face: :smiling_face: :smiling_face: , but in the obtained file some values are saved as * as in the attached file from you .

    So do you have any idea how to extract the value correctly?

    And do you have any idea which variables I should use to extract or save the speed of the motors or encoders and the speed of the joints ?

    There is a Karel Reference manual you can get off of the Fanuc portal/Fanuc CRC that will be a lot of help for this sort of thing, but the asterisks are caused by the "length" parameters you used when converting the variables to strings.

    According to the Karel Reference the CNV_REAL_STR built-in procedure:

    Quote
    • length specifies the minimum length of the target. The actual length of target may be greater if required to contain the contents of source and at least one leading blank
    • If the declared length of target is not large enough to contain source with one leading blank, target is returned with one leading blank and the rest of its declared length filled "*" (asterisks).

    In these lines

    Code
     CNV_REAL_STR(IntANGa3, 2, 2, IntANGa3Str)

    You set the length of IntANGa3Str (target) to 2. The actaul sorruce values used when running your program were these:

    1.358 fits because there is only one character to the right of the decimal, this character plus the "leading space" equals a length of 2.

    -94.739 obviously doesnt fit into the length of 2 after you add the leading space.

    However, -1.285 and especially -.417 (-0.417) are a bit tricky. I think they do not fit because the - signing plus the digit to the right take up 2 characters. Regardless of how this all actually works. Using 0 for the length parameter cleared up the issue for me.

Advertising from our partners