Announcement

Collapse
No announcement yet.

Problem with MRI controller power input powerpoles?

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Problem with MRI controller power input powerpoles?

    One of our teams was using an MRI motor controller yesterday and ran into an issue with the power input. When they plugged in the power and turned it on, the LED would not light on the motor controller. We found that if we ever-so-gently wiggled or put pressure on the powerpole for the input to the module, that the LED would turn on. We found that if we put a little bit of duct tape on the MRI power input powerpole to put a slight bit of downward pressure on it, that the LED would light and stay lit and the controller would run for our limited test.

    Needless to say, duct tape isn't a great reliable long term fix for this problem. I haven't seen any issues like this on other motor or servo controllers, but then we're just starting to use them a lot on mobile robots.

    Looking at Mitch's picture of the controller board, it looks like the powerpole is connected with the two wires bent an 90 degrees and soldered in. There does not appear to be any other strain relief on the wires or connector. There is a slight visible gap between the connector and the case of the motor controller, so the case doesn't appear to be providing any strain relief. I wonder if slight wiggling of this connector over time is likely to cause one or both of the soldered connections on the wires to come loose?

    Has anyone else experienced a similar problem? Does anyone else have any suggestions for preventing this? Adding external strain relief may be possible once the design is solidified, but in our prototyping stages we need to move motor controllers and wires around a bunch. It doesn't seem possible to add strain relief while also allowing the power to be connected and disconnected.

    I suppose we could add a small red/black powerpole to red/black powerpole wire sticking out from the controller and then plug power in and out using the other end of the small wire so that the motor controller connector wouldn't have to be repeatedly plugged in and out. But doing that on every powerpole coming out of the motor and servo controllers would sure be a lot of time and trouble. And the whole point of having power poles instead of screw terminals was to make these connections more reliable, so having to add little wires hanging off the end of every powerpole would seem to be an admission that the new connectors aren't any more robust than the old ones.

    To be clear, the connections between two powerpoles themselves seem solid and reliable. The problem seems to be that slight motion of the connector when it is plugged in and out may be causing the soldered wires on the PCB to connect and disconnect.

    I'm hoping this is a one-off problem, though it looks like these could fail over time as the connectors are used and wiggled repeatedly. I was hoping these would be an improvement over the screw-terminals on the old controllers, but I never saw a screw-terminal + ferrule-ended-wire fail in those controllers.

    Thoughts? Suggestions?

  • #2
    I did a little research. Anderson offers "mounting wings" for powerpoles mounted onto PCBs to provide strain relief for the connectors. They snap onto the connector and then screw onto the PCB. See this link and this link for examples.

    Sadly, the MRI modules do not appear to have used the mounting wings on their board designs. So the connectors are only supported by their soldered wires and thus are subject to strain when they are plugged in and out.

    I would humbly suggest that MRI consider adding the mounting wings to future designs incorporating powerpoles.

    Comment


    • #3
      Which direction of force are you concerned about?

      The picture I posted obviously does not include the enclosure, which does provide some strain relief.

      Viewed looking into the end of the motor controller's power pole connectors, there should be adequate strain relief for up/down/left/right (the poles are captured on 4 sides by the enclosure), and pushing in on the powerpole should be relieved by the back wall of the top half of the enclosure.

      Pulling hard on a stuck cable would definitely be an issue. There's nothing to protect the right angle of the pole-to-PCB connection in this case.

      With the PCB out of the enclosure, it is somewhat fragile, I know that from some experience :-)

      /Mitch.
      Mitch Lichtenberg
      Technical Mentor, Saratoga High School Mechanical Science and Engineering Team (M-SET)
      FRC 649
      FTC 6165, 7641, 7390

      Comment


      • #4
        Originally posted by Cheer4FTC View Post
        Looking at Mitch's picture of the controller board, it looks like the powerpole is connected with the two wires bent an 90 degrees and soldered in. There does not appear to be any other strain relief on the wires or connector. There is a slight visible gap between the connector and the case of the motor controller, so the case doesn't appear to be providing any strain relief. I wonder if slight wiggling of this connector over time is likely to cause one or both of the soldered connections on the wires to come loose?
        Originally posted by Mitch View Post
        Which direction of force are you concerned about?

        The picture I posted obviously does not include the enclosure, which does provide some strain relief.

        Viewed looking into the end of the motor controller's power pole connectors, there should be adequate strain relief for up/down/left/right (the poles are captured on 4 sides by the enclosure), and pushing in on the powerpole should be relieved by the back wall of the top half of the enclosure.

        Pulling hard on a stuck cable would definitely be an issue. There's nothing to protect the right angle of the pole-to-PCB connection in this case.
        The controller I'm looking at has a slight gap between the powerpole and the case, so the case may prevent gross movement but it still doesn't hold the connector in snugly enough to prevent minor wiggling strain on the soldered wires.

        Quite frankly, I have no idea which direction of force has caused an issue. Obviously there will be kids pushing the connector in hard and pulling it out hard, and maybe even wiggling it from side to side if it is hard to get out or in. Ideally, the controller would support all types of motion like this without failing. Surely repeated insertions and removals with varying forces should be expected in a product like this.

        The "mounting wings" appear to have been designed specifically to address this: I wish they were on the boards.

        Comment

        Working...
        X