Robot welding with interchangeable fixtures

    Hello all,


    This is not a problem thread but rather a question for your experience.


    Have any of you integrated a robot welding application where the welding fixtures need to be switched out because of part variety?
    We're trying to build a case for such an application but we don't, at the moment, have a working example to use as a basis. We're worried about splatter build up from welding and how it would impact the repetability for the fixtures when we take one out and replace it with another. We've got about 5 varieties of parts to weld and each one needs a differet fixture from the rest. It would be in our best interest if we could simply switch out the fixtures when we change to a different part to weld.

    Build 5 square tubing frames that will hold the 5 fixtures, next build A fixture that locates the tube frames. the tube frames all need to be the same size. Then write your 5 programs, one for each part.
    If you do this, any future fixtures that can be mounted to A tube frame will go right in. Then you just need to change the program being run. We do this with good results, and quick change of parts made.
    Care must be taken so that any fixture mounted to A tube frame can only go in one way. Best to make sure the square tube frames only go into the locating fixture in one orientation. That way fixtures are put
    in the same every time.

    Our setup is similar, with one robot servicing two different stations, and each station capable of accepting a large variety of tooling. We use RFID on the tooling along with an orientation pokeyoke to ensure that the program matches the tooling, and we use a User Frame programmed to each station. Using this, the tooling allows the PLC to choose the correct user frame and program on the robot. We do our program selection with a main program loop launched by PNS, and then digital inputs which decide what is run from that Main loop program.

    I have full confidence that we will get this up and running like a dream in terms of programming, but it's the mechanical aspect that may prove the most difficult for this application.


    It's good to know that it's been done. Now I can properly twist my boss' arm around until he agrees to the concepts we're giving. :biggrins:

    I built a single robot cell before to utilize multiple fixtures. I had 2 pins, one on each side and 4 pads for the fixture to sit on, the stand was a U shape, with the pins and pads on each side, once the fixture was on it was held in place with 4 bolts. Quick to remove and swap. I had the robot running the cell so I had dogs on the side of the fixtures, 4 sensors on the cell work table and configured the sensors to a group input which gave me 15 possible fixtures, though we only used 4.


    It worked well and the customer is still using it.

Advertising from our partners