Analog sensor on EOAT- cable down the side of robot

    I have a fanuc robot that I want to put a laser sensor on. The sensor would be mounted to the EOAT. We'd get analog data on 2 conditions: I'd be checking to see if a nest is orientated correctly and if a part is oriented correctly in a lathe before starting the machine. Misloads create wrecks in the lathe. The condition hasn't shown up yet, but it's a matter of time. The keyence IL-300 is capable of using a window with upper, lower, and good condition and changing it to a discreet output. However, I have one output left in the robot's EE connector and need to detect 2 distance/measurement conditions. This could be controlled by teaching the robot in different positions to make the output work for all cases, but I think this will be more complicated.

    So, I want to run the amp cord down the outside of the robot and back to the main control panel. Has anyone ever done this? Can I make a few standoffs and holders to keep the cable safe and flexing at a minimal? I really prefer not to do something crazy like get a leoni package for 1 measely cable.

    Any help and suggestions are appreciated.

    Sent from my Pixel 2 XL using Tapatalk

    Has anyone tried something like this? I searched this site and found old discussions about analog sensors on EOAT for measuring gripper positions and things like that but the cord routing wasn't discussed. What do you guys think? Any suggestions or help? Any stories good or bad from your own experience?

    I'm starting to think maybe I should just use the 1 input on the robot that's available. I will only be able to have one set distance to detect both conditions I want, but maybe it's not a big deal. I can just teach my robot points where the measurements are taken in the proper window where they need to be. Maybe it's not a big deal at all?

    Sent from my Pixel 2 XL using Tapatalk

    Edited once, last by Schroeder (February 10, 2019 at 9:58 PM).

    The general term is "dressing" or "robot dress." And the answer is... it depends.

    In general, you can run cables down the outside of a robot arm, but they'll need:
    1. To be hi-flex or some other cable rated for taking LOTS of flex over its lifetime
    2. Strapped down to the robot arm at various points along its length to keep it from dangling like spaghetti. Most robot arms provide some threaded holes for attaching tie-downs of some type.
    3. Enough, but not too much, "extra" at the joints -- as the arm extends and the wrist contorts, the cable will alternately be drawn up or relaxed loose. You have to find a happy medium here that will cover your programs' range of motions, without stretching, snagging, or binding.

    Also, you'll usually want to use simple connectors en route. Think about the most likely places for the cable to get snagged, damaged, wear out from flexure over time, etc, and bracket those areas with some kind of reliable connector. Then have your cables made to those lengths, with those connectors, so you can quickly swap damaged sections without having to re-wire everything from scratch.

    EMI could be an issue for low-amplitude signals, especially if your cable runs near the servo motors, so you may need to look into using shielded cable, or using signals with better noise immunity (RS-485, for example, uses dual opposing voltages for noise immunity, and 4-20mA analog signals have better noise immunity than 0-10V analog signals).

Advertising from our partners