HELP

    Hello everyone. I cannot solve the problem.
    after scanning (Scansonic), the robot makes an involuntary impact on the product.
    If you cause an error after scanning, then no impact occurs.
    I tried to solve it with KAREL, but I couldn't find a command that could reset a custom error in automatic mode.

    During the transition from point 7 to point 8 without user error, a hit occurs.
    If you call a custom error after scanning, and then reset and continue, then no impact occurs.

    Maybe some of you have already encountered a similar problem.

    So you are saying that the robot will collide without an alarm, but if you insert a user alarm it doesn't? Have you tried jogging the robot from point 7 to point 8 to see if there is an obstruction between points? Also it appears that you have remarked out the weave circle but left the weave end in place. Have you tried remarking both out to see if that helps resolve the issue?

    Quote from white_raven

    So you are saying that the robot will collide without an alarm, but if you insert a user alarm it doesn't? Have you tried jogging the robot from point 7 to point 8 to see if there is an obstruction between points? Also it appears that you have remarked out the weave circle but left the weave end in place. Have you tried remarking both out to see if that helps resolve the issue?

    yes, I tried to come to each point separately and sequentially in step without scanning, there are no problems, there are no obstacles between the robot's points.

    the problem appears exactly at the end of the scanning track, while the length of the scanning path is not important, the problem is always when moving from the end of the scanning point (during scanning) to the indentation point (the robot moves to a safe distance from the product)

    OK, have you calibrated the scanning head? It might be that the calibration is off. I have only set up 1 laser guided welding application and I know that on that system, if the calibration was off then it would cause the robot to weld in the wrong location. What might be happening is that the sensor was bumped, or has moved slightly since the last time it was calibrated. This would cause the robot to scan properly, but when the application sets the coordinates, it is off enough to cause a collision. Sorry, im not very familiar with the scansonic application, but I do have experience with irvision and other laser welding applications, so im not sure exactly how it sets its coordinates.

    Quote from white_raven

    OK, have you calibrated the scanning head? It might be that the calibration is off. I have only set up 1 laser guided welding application and I know that on that system, if the calibration was off then it would cause the robot to weld in the wrong location. What might be happening is that the sensor was bumped, or has moved slightly since the last time it was calibrated. This would cause the robot to scan properly, but when the application sets the coordinates, it is off enough to cause a collision. Sorry, im not very familiar with the scansonic application, but I do have experience with irvision and other laser welding applications, so im not sure exactly how it sets its coordinates.

    all instrument and laser adjustments were made correctly, the software shows that the robot is correctly plotting the coordinates.
    but it is precisely after the weld end that it jerks randomly quickly when moving to another point. Sometimes it hits not the product, but vice versa from the product. those. there is some kind of collision area after scanning.
    however, neither fanuc nor the representatives of the laser sensor could help solve this problem

    Quote from Kawazaki02

    all instrument and laser adjustments were made correctly, the software shows that the robot is correctly plotting the coordinates.
    but it is precisely after the weld end that it jerks randomly quickly when moving to another point. Sometimes it hits not the product, but vice versa from the product. those. there is some kind of collision area after scanning.
    however, neither fanuc nor the representatives of the laser sensor could help solve this problem

    That's very strange. So basically the robot scans and welds fine, but when leaving the last weld point and moving to a safe point away from the product the robot decides to go a different direction? It appears as though the next motion is a linear motion and should go directly to that point. Its hard to know what is going on without getting my hands on it, but I would still assume an issue with the scanning application because the problem doesn't arise unless it is implemented, correct? I would double and triple check the scansonic application to ensure it is not sending the robot faulty positional data at the end of its scan cycle. Also, is it possible to bypass the scansonic, and create old fashioned welding instructions to do the job, or is there too much variance from cycle to cycle? Also, if that is not an option, I would check the coordinate data for point 7 and point 8 after a crash before the robot is manually moved off of the product to see if the last coordinates are where they should be.

    Quote from white_raven

    That's very strange. So basically the robot scans and welds fine, but when leaving the last weld point and moving to a safe point away from the product the robot decides to go a different direction? It appears as though the next motion is a linear motion and should go directly to that point. Its hard to know what is going on without getting my hands on it, but I would still assume an issue with the scanning application because the problem doesn't arise unless it is implemented, correct? I would double and triple check the scansonic application to ensure it is not sending the robot faulty positional data at the end of its scan cycle. Also, is it possible to bypass the scansonic, and create old fashioned welding instructions to do the job, or is there too much variance from cycle to cycle? Also, if that is not an option, I would check the coordinate data for point 7 and point 8 after a crash before the robot is manually moved off of the product to see if the last coordinates are where they should be.

    checked everything. this problem has been in our enterprise for a very long time.

    but still the question remains, why, after scanning, if you cause an error, the robot moves on as if nothing had happened.

    If you have checked all positional data, and the coordinates set by the application process are not to blame, then I would guess it is a software compatibility issue. Inserting a user alarm probably interrupts the software process allowing the robot to continue normally. Was this scansonic equipment recommended by FANUC or was it something your company decided to use. FANUC usually recommends the equipment that is most compatible with their software, and can provide tech assistance with diagnosing these kinds of issues. Have you tried recording P[8] as a position register and using that in place of the point? It might not help, but there is a possibility that using a position register could interrupt the end of the application process enough to prevent a crash because those applications do not typically modify register values. Jog the robot to P[8], go into your position registers and find one that is not used, record the position to that register, then in the program curser over to the 8 and use the function key for choice, select PR, then type the number of the PR that you used. Make sure it is still set to a linear motion, and that the speed of the motion is still the same, then try to run the program without the user alarm. I hope this helps. If nothing else, you could try to rewrite the program, and use that in its place if for some reason your data is being corrupted. I wish I could be more help, but without being able to put my hands on the pendant to check the application data, it's difficult to diagnose.

    Quote from white_raven

    If you have checked all positional data, and the coordinates set by the application process are not to blame, then I would guess it is a software compatibility issue. Inserting a user alarm probably interrupts the software process allowing the robot to continue normally. Was this scansonic equipment recommended by FANUC or was it something your company decided to use. FANUC usually recommends the equipment that is most compatible with their software, and can provide tech assistance with diagnosing these kinds of issues. Have you tried recording P[8] as a position register and using that in place of the point? It might not help, but there is a possibility that using a position register could interrupt the end of the application process enough to prevent a crash because those applications do not typically modify register values. Jog the robot to P[8], go into your position registers and find one that is not used, record the position to that register, then in the program curser over to the 8 and use the function key for choice, select PR, then type the number of the PR that you used. Make sure it is still set to a linear motion, and that the speed of the motion is still the same, then try to run the program without the user alarm. I hope this helps. If nothing else, you could try to rewrite the program, and use that in its place if for some reason your data is being corrupted. I wish I could be more help, but without being able to put my hands on the pendant to check the application data, it's difficult to diagnose.

    Okay, I'll try at work tomorrow.
    I will write about success

    To track sens a weld you will need to do a weave sign not a weave circle. It needs the weave to properly track as it is sensing the feedback voltage from the seam. I have seen improper weld tracks cause collisions.

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