Posts by Leon

    First you have to initialze the aggregate like {x 0, y 0, z 0, a 0, b 0,c 0}

    Are you sure about this. In motion commands you can just do it just as Skyfire explained. In case you are using variables then you have to initialize them. important difference i think

    That is implementation detail.

    In my case i dont think it is, because it is hardwired in the safety systems. Then maybe it is a remnant of the old volkswagen control. Software is changed but the safety hardware is still there.

    KSS even allowed jogging while in T2

    I know, since a about of month we have 1 "newer" robot that is running kss 5.6 and i kind of miss not being able to jog in T2. It is not a problem to switch back to T1 but it is something i am still getting used to.

    I know this is going to be a useless comment but i am going to mention it anyway, but i have 3 robots where the gates need to be closed to move in T2. so it used to be possible. These robots are not the newest that you can find (kss 4.7) so i know that it is not really relevant for this case. But when i was reading this post i was wondering if somewhere the safety standards where lowerd? or was kuka more strict back then?


    I personally find it a bit more comforting that the gates are closed before the robot runs at full speed.

    Oke, good to know. so you already have a PLC communicating with robot then? That should be your first point of research to find out how that works and which protocol that is using. after that you can look at the software.


    But to comment on your flowchart i would use a interrupt function as panic mode described. than you can move down with a set speed, and the interrupt will cancel motion and record position when the switch is hit.

    To be clear here, do you have an already working robot as a milling machine and you want to ad additional functionality to it? Or are you trying to figure out how to build the complete system from scratch?


    Because if you already have a working set-up you only have to add to that and that is a lot easier to explain. if so be specific with what components you have installed, and not just KRC2, but also with what KSS (software version) and what I/O options that are installed. Same on the PLC side.

    First off, do you know that the deadman switch has 3 state's: Not pressed, activated and pressed to far which deactivates it again.


    In T1 and T2 you will always need the deadman switch activated. If you want to run in automatic mode you have to make sure your safety's are correctly wired. Emergency stop and some sort of user safety. without that you wil not be able to turn the drives on in automatic and depending on your configuration maybe also in T2.

    panic mode


    Thanks for the video, that is exactly what i want to attempt. It does raise a question though. they are basically only using axis 2 and 3 and axis 6 seems to be stationary. seems to me they either missed a opportunity to get some extra speed, or they know something that i don't. that could be something fun to find out. :winking_face:


    @SkyFire


    It is not my intention to tamper with the robot in any way. in the end this is just for a bit of fun for perhaps an hour or so and then robot has to go back to work.

    I have got a 9,5m high ceiling, so i am not that worried about the height. The overall size of the cell could be problem and that is probably going to determine the length of the arm that i can mount.

    So how to begin, This may seem like a strange topic but i am going to explain this as best as can.

    Each year during the Christmas party in our company (except for the last year because of some reason:nauseated_face:) we like to something weird but fun, usually with one of our production machines. af few years ago i modified one of our milling robots in to a chocolate rotomoulding machine (the old topic can still be found here: Kuka robot as rotational mould machine).


    For this year as the title of the topic implies, i want to try to turn the robot into some sort of catapult. To be clear i dont want to make anything dangerous, but i think throwing something like a tennisbal should be fine. If can keep it simple enough i will change 2 robots and they can trow balls at each other but i am getting a head of myself.


    How i want to accomplish this is by emulating the motion of Trebuchet with a long arm (dont know how long probably need to do some math on that one) mounted perpendicular to the flange. So most of the motion will be done by A2,A3 and A6.


    As icing on the cake i am probably wiring a big red button to one of the inputs as a launch button.


    What i want the know or better said what i am worried about is the amount of torque i am putting on A6. The software will stop if i in someway overtorque the axis but i want to be as carefull as possible. Is there anyone with experience with high torques on specifically A6 or maybe suggestions on a better approach?


    At the moment i have not decided which robot i am going to use but i can choose from:

    KR150 with KRC2 kss 4.97

    KR150 with arm extension with KRC2 kss 4.97

    KR210 KRC2 Kss 4.97

    KR150 but with a krc2 running kss 5.6

    Quote

    i'll try what dogeautomate said

    why?! Because A it should not be in the SPS! and B it is also incomplete, ask yourself when is metal_surface become false? if the answer is when inductive_sensor is false then congratulations you just added a few pointless lines of code.

    I am reading between the lines here but maybe your mentor wants a separate function/program/subroutine instead of everything written in 1 .src file.


    Because using the SPS is like what hermann said it doesn't make any sense.

    That is a hard question to answer, but i will try.


    First off, using any approximation will change your TCP speed when you get to the target point. In which way depends on a lot of variables, The TCP speed you have set, acceleration (TCP accelleration and individual axis), Corner angle, which type of approximation, and at which setting you have the approximation. I you have points close together that can also be of influence.


    for example if you want to make a 90 degree corner with a TCP speed of 0.05 m/s and a C_VEL of 90, You would probably get exactly what you expect. the speed will drop to 0,045 and the robot should move in a smooth motion through the corner.

    If you want to do the same corner at a higher speed, lets say 0,2 m/s you are probably going to see the speed drop more than the 90% you asked because the acceleration and deceleration of the individual axis' cant keep up


    Another example if can give is if you have to make a corner of 180 degrees. Basically approximation does not work (obviously). The robot has to fully stop and accelerate again. But in your case i expect you could have 2 90 degree corners close together. If the 2 corners are to close you will also see the speed drop more then you programmed.


    In your case Trial an Error would be the best approach. But keep mind that when using approximation the robot doesn't have to fully stop anymore, so your deceleration distance can only become less, not more.

    Well that's a bit vague to give any reasonable answer.


    What kind of object are you trying to locate?

    How much data do you need from your picture 2D/3D

    What kind of accuracy are you expecting.


    Basically the only thing i can tell you is that, No the robot controller does not process images. But a industrial camera can send coordinates to a robot controller.


    So first figure out what you need, after that you can think about how you are connecting it to the robot controller.

    Watchdog error points to a bad connection between the KCP and the cabinet. in the worst case scenario the cable itself is damaged, but more likely in my experience is a bad connected plug. This could be a in the connector to the cabinet, but also on the connectors on the ESC board. So check for connectors that are not seated properly or pins that are bent or pushed into the connector.

Advertising from our partners