Detecting Torque Levels To Divert Robot KRC2 KR210

  • Hello,
    KRC2 KR210 5.6 Software
    We are currently using the KR210 to perform material handling and processing of exhaust pipe. Well occasionally the pipe with have a defect such as a material split or whatever which would cause our end points of our exhaust to vary. Currently if we have an discrepancy, the robot will crash the material into the face of our sizer tooling. The crashing isn't a problem with damaging any tooling, downtime is the concern as they run in fully automated cells. I was wondering if there was a way to utilize Torque Monitoring or something else of that sort through the robot to detect when a collision is going to occur and have the robot abort continuing forward and then proceed along with its program.


    I have tried to have the Submit Interpreter read the $TORQ_DIFF[] when performing that particular task to try and sense excessive torque then call an interrupt and abort that step of the process. Torque Monitoring seems like it would be the perfect event handler for detect and pre-crash situation, but I do not what the robot to Brake and wait.


    Thanks and sorry for the rambling it has been a eventful day so far and this is a project to alleviate the strain on our technical support.

  • Place your Ad here!
  • You might be able to do this using a very slow motion, with an interrupt monitoring the motor torques. But the root issue here is that the robot can't detect a collision until after it happens, and enough force has been applied for the motors to feel the resistance.


    Just like a car, a robot has braking distance, and the faster it's going, the larger that distance. So at high speeds, by the time torque monitoring has detected a collision and starts applying the brakes, the damage has already been done from simple inertia.


    If you can spare the cycle time to perform a very slow motion to "push" the end of the pipe against a fixed obstacle, while monitoring the torque feedback, you might be able to tell the difference between a correct-length and over-long pipe. It'll depend on a lot of factors -- the robot is beefy enough that it'll need more than a gentle touch before the motor torque shows a clear difference. There's also the flexibility of your pipe, the grippers holding the pipe, the angles of the joints, etc.


    Frankly, you'd probably be better off adding a beam sensor or something similar and have the robot "touch off" the end of the pipe with it in between stations.

  • Thanks for the reply, a sensor is something I am trying to avoid. The robot is excessive for the task at hand, they basically have a solid aluminum block with a gripper bolted to the robot, and when we do have the catastrophic crashes, the pipe won't slip in the gripper, just usually bending 2.5" stainless exhaust and potentially braking the lags loose on whatever it had crashed into.
    When we go to enter the sizer, we are doing a LIN at roughly 0.2 m/s, which I could afford to slow if this would work. I just wasn't aware if the TM_BIB could be modified to use the torque monitoring and change the conditions when true if I set a variable to true to state I am approaching the sizer, that it would avoid a Brake and perform something else instead


    *Also torque values during normal operation are extremely low since we are using the KR210, the whole reason we have them is for reach, our heaviest part is probably 7lbs and maybe 3 feet long at max

    Edited once, last by drkuzzie ().

  • [size=2]clear values in torq_diff[] and drive robot with tail pipe close to sizer (without contact) and record values[/size]
    [size=2]then retract and do it again with slow speed you think is ok, then check captured values again. [/size]
    [size=2]if difference is noticeable you got input that can be used.[/size]
    [size=2]then try writing interrupt program using examples in manuals or on this forum.[/size]

    1) read pinned topic: READ FIRST...

    2) if you have an issue with robot, post question in the correct forum section... do NOT contact me directly

    3) read 1 and 2

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account
Sign up for a new account in our community. It's easy!
Register a new account
Sign in
Already have an account? Sign in here.
Sign in Now

Advertising from our partners