Announcement

Collapse
No announcement yet.

Encoder/Controller Issue?

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

  • robgcarlisle
    replied
    Originally posted by Daniel_10237 View Post
    When our encoder wires are plugged into the core motor controller the board throws and error and dies. Any ideas as to why this is happening?
    Another set of problems with motor controllers... there seem to be several people seeing these.

    Leave a comment:


  • Philbot
    replied
    Originally posted by 5294-jjkd View Post
    This is the thread I was thinking of, unfortunately I don't see an obvious relationship to your problem....

    http://ftcforum.usfirst.org/showthre...t=Encoder+wear

    You may want to try connecting the motor controller to a PC and use the core device discovery program to exercise the functionality of a 'good' motor and compare to a 'bad' motor, see if there is a difference in behavior.
    I have just read this thread, and I'm intrigued by the possibility of the fault being related to the the encoder issues I discovered a while back...

    Taking what has been discovered at face value:

    1) After a parts replacement, the system seems to work fine for a while.
    2) After some usage, a fault condition occurs that seems to be related to the encoder.
    3) Cables and modules seem to be hooked up correctly.
    4) There is a time factor involved that seems to effect either the motor, or the controller... it's not clear to me which...

    I can throw out one off-the-wall scenario...

    What if a mechanical failure in the encoder electronics is causing undesirable voltages to appear on the encoder cable which is effecting the Motor controller.
    This could possibly be grounding on the +V line, or worse still, +/- 12V leaking onto any of the lines.

    Possible justification...
    We know that if the shaft of the andymark motor is pushed into the motor body, this can cause odd things to occur with the encoder.
    -- observed: Heavy loads in one direction on a NR40 causes the helical gear to push the encoder magnet out and reduce the number of encoder pulses.
    -- reported: At least one team reading incorrect encoder counts on NR3.7 motors.. suspect due to miss aligned encoder magnet.
    -- proposed: Changes were being made to the encoder housing to prevent this. We don't know what if any changes have been made.

    So, now with the addition or more gearbox combinations, perhaps a new fault is occuring that requires a certain amount of time to appear.
    eg: a re positioned component is being pushed out or alignment causing wear, or chafing, or vibration, or shorts.

    So, some new tests could be used to determine if this is possible. And it really relies on some post failure testing:

    1) Identify a controller / motor combination that has "failed". That is: you can't connect to the controller with the motor/encoder attached.

    Test 1) Disconnect the encoder cable and determine if the combination now runs.
    --- If it does, we know the failure is directly related to some part of the encoder pulse generation/transmission/reading portion of the system. But that it's not fatal to the controller
    --- If it doesn't, we know that the fault is causing a fatal failure on the controller.

    Test 2) Replace JUST the Bad motor with a new motor and reconnect the encoder. Does it run?
    --- If it does, we know the failure was originating on the motor, and it is most likely due to a fault in the encoder electronics.
    --- If it doesn't, it may still be that the fault is causing a fatal failure on the controller. But, remove the encoder cable and see if it starts running.
    --- --- If it does, this points to the final failure being on the Motor controller Encoder reading logic, caused by the original failure on the motor.

    Test 3) would be to replace the Motor controller, but this will probably lead to another expensive failure, so I resist doing this.

    I'd be really interested to hear the results of tests 1 and 2.

    Leave a comment:


  • 5294-jjkd
    replied
    This is the thread I was thinking of, unfortunately I don't see an obvious relationship to your problem....

    http://ftcforum.usfirst.org/showthre...t=Encoder+wear

    You may want to try connecting the motor controller to a PC and use the core device discovery program to exercise the functionality of a 'good' motor and compare to a 'bad' motor, see if there is a difference in behavior.

    Leave a comment:


  • 5294-jjkd
    replied
    I seem to recall an older thread that involved a motor being run in a particular way that caused wear resulting in encoder problems. I will try to find it.

    Leave a comment:


  • Daniel_10237
    replied
    Originally posted by 5294-jjkd View Post
    We have six Neverest motors this year, four with encoders, never a problem. In your picture, I see two motor controllers with encoders plugged in, but your previous messages indicated that only one was failing. Do both fail? What is the exact error message on the robot phone?
    This leads me to think that some type of physical strain is causing the encoders to error (which may explain why the seemingly bad motor miraculous came back to life and started working). Our robot is rather heavy and the motors buckle quite a bit. Have you ever heard of physical straining on motors causing this type of behavior (of course physical strain can break the motor or encoder, but in our case it was only "broken" temporarily)

    Leave a comment:


  • Daniel_10237
    replied
    Originally posted by 5294-jjkd View Post
    We have six Neverest motors this year, four with encoders, never a problem. In your picture, I see two motor controllers with encoders plugged in, but your previous messages indicated that only one was failing. Do both fail? What is the exact error message on the robot phone?
    The error message is "the core power distribution module failed to communicate with motor controller [serial number for right motor controller]"
    We have 8 never rest motors (they weren't bought from modern robotics, not exactly sure where they are from but somewhere legit). Going off of memory we have found 3 that have failed. We tested extensively and found that if we replaced those motors with new ones, we got no error (we had another one where the casing was cracked but that was a separate issue). However, the other day when it would seem that we found another motor that was broken, we replaced that motor with a motor which was previously thought to be broken and that worked.
    And by replace I mean just plug into the board not physically put on our robot

    But of the 2 times we have physically replaced the motor (because we believed we had a bad motor), the error was gone... For about 2-3 days. The error then came back with these completely fresh motors

    Leave a comment:


  • 5294-jjkd
    replied
    We have six Neverest motors this year, four with encoders, never a problem. In your picture, I see two motor controllers with encoders plugged in, but your previous messages indicated that only one was failing. Do both fail? What is the exact error message on the robot phone?

    Leave a comment:


  • FTC4160
    replied
    If you swap the controller with encoder and a controller without encoders, will the controller formerly being used without encoders fail? It sounds as if you have a bad motor controller; our team had received a motor controller which did not work straight out of the package this year (it is currently being sent back to MR for repairs).

    Leave a comment:


  • Daniel_10237
    replied
    Here are some images.

    http://imgur.com/WHcHfR3

    http://imgur.com/wv2GiUV

    Leave a comment:


  • Daniel_10237
    replied
    Actually, we replaced the board that was erroring with a brand new board. That fixed the problem for a day or two and now the error appears on the new board.

    Leave a comment:


  • Daniel_10237
    replied
    Only one controller has failed and it always is the one with a encoder plugged into it. And yes we have tried changing encoder wires.

    Leave a comment:


  • mikets
    replied
    Did only one controller fail or others too. If one controller failed, is it always the one with encoders plugged in?
    I've never heard of this issue. In fact, our robot has many Neverest motors and we have most of the encoders plugged in but we don't have this issue. So if the firmware is the latest, it can't be firmware issue. I assume you have changed the encoder wires too?

    Leave a comment:


  • Daniel_10237
    replied
    We have tried to simplify the problem. We have replaced every motor, motor controller, wire, and the core module and each time it worked consistently when the encoders were plugged in and failed consistently when the encoders were plugged in. We bought all of our stuff this August so it is very near the latest firmware.

    Leave a comment:


  • mikets
    replied
    Did all motor controllers disappear or just one particular one? Also, what are the firmware versions of the motor controllers. Like what 5294-jjkd said, if that's just one controller, swap it out with another one and see if that fixed the problem. If only one controller has problem, I would check the USB cable of that controller. If more than one controllers disappeared, then I would check all USB cables especially the one from the hub to the phone.

    Leave a comment:


  • 5294-jjkd
    replied
    The hub in the core power distribution module is fairly dumb, and doesn't actively 'find' the controllers. The phone is responsible for finding the controllers through the hub part.

    I would suggest you troubleshoot through simplification and substitution. First, set up the robot with only a phone, the core power distribution module, one motor controller and one motor (connected with encoder). Can you duplicate the problem with that configuration? If not, there may be a problem with one of the components you've disconnected. If you can duplicate the problem, swap parts one at a time. So, replace the motor with a different motor, leave everything else the same, and test. If it passes, swap back the original motor and confirm that the problem returns. If the problem remains with the different motor, swap back the original and move to the next part in the chain (the encoder cable). Continue this process until you get all the way back to the phone. Note that you can swap the USB part of the core hub with a 'normal' USB hub to bypass that for testing, and keep the powerpole wires running through the core If the problem persists no matter what you swap, there's something stranger going on, like more than one bad component, or you may have something else wrong, like a wire shorted to the chassis of the robot, or ???

    If the 'simple' configuration works fine, then try swapping in each of the removed parts one at a time. When you can duplicate the problem, you've just added the bad part. If you can swap all the removed parts into the simple config without a failure, then move on to building the original full robot, again adding one part at a time and testing.

    Leave a comment:

Working...
X