Announcement

Collapse
No announcement yet.

Diagnosing point of failure in BBSHD controller after short in display wire

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

    Diagnosing point of failure in BBSHD controller after short in display wire

    [At TommyCat's suggestion I've opened a post in this forum to continue a discussion opened in the Luna technical support forum here].

    Summary of issue:
    1. I shorted the main harness on the male green HiGo connection. The short was across the ground and (most likely owing to proximity) the green RxD signal pin. There was a spark and the ground pin melted.
    2. I replaced the harness with another and powered up the system. The display still works and shows no error code but the battery voltage jumps around 30-34V for a few seconds then rests at 34V.
    3. A voltage test directly off the battery shows voltage of 54V (as expected).
    4. I disassembled the controller and ran continuity tests across all wires in the circuit including main harness pin-outs, battery leads, phase wires and the 4-pin (PAS) and 6-pin (?) connectors in the controller. The results of those can be viewed in a Google sheet here.

    My knowledge of electronics is limited but from these continuity tests I've concluded:
    1. The mosfets are OK - I measure similar stable resistance from ground to all phase wires, and climbing resistance off positive.
    2. I didn't read any continuity between the 4-pin PAS ground wire and the battery ground but did read it between the 4-pin and the main-harness ground. This seems to indicate an issue in the circuit at that point?
    Responding to Tommycat's questions in the original post...

    In your chart, Main harness connection pin #8 is also good to PAS connection Black wire. So how could Pas ground not make it to battery ground?
    I see what you're saying. I may have mis-measured these points. Ground must be continuous in series through the circuit yes?

    Which display do you have? By empty battery do you mean an actual display reading, or display battery bar indication?
    I have the APT 750C display (from Luna) and that reading is coming from the display so is the (incorrect) measured voltage of the battery. Multimeter tests directly on the battery and the built-in LED battery indicator show a full charge.

    So, assuming that ground wire reading was incorrect (I can't recheck just now) I have these question. Is it possible a component on the controller has been damaged but not to the point of causing a display error? Specifically, a component for reading/carrying battery voltage to the display?

    If so and it's an IC I guess I've hit the limit of diagnosis without the skills/documentation/tools to pinpoint this further. The bigger question though - where (if anywhere) is there a fuse/circuit breaker in the system to prevent a $20 part taking down a $100 one? If there isn't one, is there a good technical reason why not and would it be possible to wire one in?


    #2
    Hi mp52,
    Good to see you.

    Taking things in order... And having the wiring diagram handy.





    Originally posted by mp52 View Post
    I shorted the main harness on the male green HiGo connection. The short was across the ground and (most likely owing to proximity) the green RxD signal pin. There was a spark and the ground pin melted.


    With the main motor harness disconnected from the rest of the harness, the only power at those pins would be the full battery voltage from the battery. Everything else has to wait for the display to energize the controller (power on), communication included. Which would explain the melted ground pin...


    Originally posted by mp52 View Post
    I replaced the harness with another and powered up the system.


    Did you go clear down to the board, or splice in some where outside? As in... any piece left of the original harness cabling?


    Originally posted by mp52 View Post
    A voltage test directly off the battery shows voltage of 54V (as expected).


    Is this with the battery still connected to the system, with the power on by backprobing the battery connector?


    Originally posted by mp52 View Post
    Ground must be continuous in series through the circuit yes?


    Or even connected at a common junction point, but certainly continuity thru-out.


    Originally posted by mp52 View Post
    I have the APT 750C display (from Luna) and that reading is coming from the display so is the (incorrect) measured voltage of the battery.


    Does the battery bar indicator show any bars?


    Originally posted by mp52 View Post
    Is it possible a component on the controller has been damaged but not to the point of causing a display error? Specifically, a component for reading/carrying battery voltage to the display?


    It's my understanding that the battery voltage display takes it's readings from the power line to the display. And the battery bar indicator gets it's reading from controller communication. (not 100% sure on this)
    I would like you to CAREFULLY take a voltage reading between pin #8 (Black-ground) and pin #1 (P+,Brown, Batt Positive) main motor harness disconnected, with the system powered.
    With that large voltage whack to the backside of all those ICs... damage remains to be seen. At the main motor harness connection where you took the previous reading. If you jumper pins #1 and #2 (Brown and Orange) this bypasses the panel, energizing the controller. Then you can check for 5vdc regulated power coming from pin #6 (RED).
    I think that a damaged component on the board would not be worth it... and frankly above my desire or expertise. But some have succeded.


    Originally posted by mp52 View Post
    The bigger question though - where (if anywhere) is there a fuse/circuit breaker in the system to prevent a $20 part taking down a $100 one? If there isn't one, is there a good technical reason why not and would it be possible to wire one in?


    A fuse to protect wiring and components is never a bad idea. And is certainly possible where ever you seem fit. I certainly don't like the idea of full battery voltage availible in close proximity to other low voltage wiring (and pins!)... but that's up to the manufacturer.

    I added an inline fuse on the positive battery voltage going to an add on volt/ammeter because of the high voltage and small wire/insulation.

    But please take no offence. The cause of the failure, working on a harness with the battery still connected and shorting out the harness, could easily have been avoided. Always remove battery when working on the harness or when changing/working on any harness connected part.


    Regards,
    T.C.


    See my completed Magic Pie V5 rear hub motor E-Bike build HERE.

    Comment


      #3
      Hi TC,

      Just picked this up and will have to do the testing later in the day but I'll cover what else I can now:

      With the main motor harness disconnected from the rest of the harness, the only power at those pins would be the full battery voltage from the battery.
      I've possibly confused things here. When I've said the main harness, I've meant the harness out of the motor with the loom splitting to display, throttle etc attached. This was the set-up when the short occurred (I was about to connect controls/sensors on the handlebars to complete my initial repair... :/ )


      Did you go clear down to the board, or splice in some where outside? As in... any piece left of the original harness cabling?
      I only replaced the splitter loom that plugs into the 8-pin. The controller is still potted and I haven't replaced anything on it.

      Is this with the battery still connected to the system,
      No - its a reading off the battery terminal cables coming off the cradle (no connection to the motor).

      Does the battery bar indicator show any bars?
      No - 0 bars. I've attached a screenshot I took of the display.

      I think that a damaged component on the board would not be worth it... and frankly above my desire or expertise. But some have succeeded.
      Indeed - certainly beyond mine and I lack the tools to perform micro-soldering.

      A fuse to protect wiring and components is never a bad idea. And is certainly possible where ever you seem fit.
      It's a confusing area. I've seen threads elsewhere debating it and concluding that all but very expensive and fairly cumbersome circuit breakers would react too slow to prevent damage. But then Luna in my original thread said the newer controllers are fused but I've not seen one (presumably the fuse would not be potted like the other components).

      The cause of the failure, working on a harness with the battery still connected and shorting out the harness, could easily have been avoided.
      None taken, you're absolutely right and I'm kicking myself! I was at the end of a long run of troubleshooting the motor which turned out to be moisture in the loom and display. In removing the crank to access the controller, the Bafang crank got stripped by the crank puller and eventually I had to cut it off to remove and order Shimano replacements from Germany. I'd gone to extra lengths to seal and waterproof the display and when the cranks arrived was working late to get my commuter back on the road. A case of tiredness, frustration and excitement of getting the wheels back on the road keeping them off the road longer! Another note to self: don't buy dialectric grease sold in metal squeeze tubes.

      Appreciate your help and will update on the measurements later in the day NZ time.

      Comment


        #4
        OK, I follow you on where it was shorted.(Thanks for the heads up) But with the display removed, still just the main power wire would be hot for the same reasons. So the throttle may be caught up in things, may be a good thing if that's all it is.

        I'd still test at the main motor harness connection to eliminate the top half harness and controls. Just a reminder to mirror or flip the diagrams pin-outs, as they will be reversed on the female side that comes from the motor. As the diagram shows the upper harness pin-out for the male side.

        No 5vdc...?

        Remove controller and disconnect all connectors. This is if PAS, gear sensor, or motor hall sensors had been shorted or damaged enough to drag down the 5vdc out-put.

        Lol... I have a lot of notes to self too. :-)
        See my completed Magic Pie V5 rear hub motor E-Bike build HERE.

        Comment

        Working...
        X