Strange behavior by my SmartPie [SOLVED]

[Lollandster]

Hi, I have had some issues with my SmartPie lately that I could use some help getting fixed. Since I'm not very good at describing faults I made a short video showing my problem http://youtu.be/1ru0V48RGII As you can see the motor wont start by it self and I have to spin it and when it gets going it doesn't run smooth and it looks like it's trying to spin in the opposite direction. The problem disappears again after a few minutes. I'm guessing I have an issue with one or more hall sensors, but I don't know where to start. Any help would be appreciated. Thank you.

[Bikemad]

If the battery voltage is satisfactory (above 37V) and doesn't drop dramatically while the fault is evident, it seems likely it will be a hall sensor type problem.

If the fault only occurs from cold, and disappears when the motor is warm, it may be due to a build up of moisture on one of the hall sensors, which is drying out as the stator warms up during use.

I think you may need to remove the speed controller and take a look inside to check the hall sensor connections, and also to see if there is any sign of condensation in there. But as you bought the motor from GM in China, I suggest you email David (wyh@goldennmotor.com) first, and ask his opinion, before you dismantle anything.

I suggest you also take a look at this post and this video in the meantime, for more details on testing the operation of the hall sensors, just in case the problem is not moisture related.

I suggest you also read this post before attempting to remove the controller.

Alan
 

[Lollandster]

Thanks, it might be a moisture problem since I have mostly been using the bike in wet conditions (raining season here now until it starts snowing instead). A moisture problem explains why it isn't consistent. I'll let the bike dry out for a couple of days before I start checking my warranty options and/or open up the motor. I'm not sure where the moisture may escape, but if I keep it warm enough it'll find a way.

[Bikemad]

If moisture has found a way in, it should be able to get back out again under the right conditions, but it may require some assistance (better ventilation in a warm environment) to allow it to evaporate.
If water has got in already, it can get in again, unless the point of entry can be located and properly sealed.

I would still email David in the meantime and ask permission to remove the controller if required, in case the fault does not sort itself, as this could save time waiting for an email response if you find you need to take further action.

I also suggest you read Mikko's post regarding a similar problem, which was probably caused by ~10ml of "moisture" which managed to find its way into his controller housing.

Alan
 

[Lollandster]

The problem got worse and I contacted David at GoldenMotor and got permission to open up the controller. When I got inside the controller I noticed that the plastic connection that holds the sensors was a little loose. I pushed the plastic against the circuit board and tested the motor and sure enough it worked perfectly. So I added some hot glue to the connector to make sure this doesn't happen again and assembled the bike. That was all, a very easy fix.

Notes:
There was no water at all or any traces of it inside the controller.
The controller circuit was coated with a plastic layer probably to protect it from water.
You need a PH2 Phillips head screwdriver to loosen the small screws. They are made of very weak metal and you will destroy the head with the wrong size screwdriver.
Gary's video removes all the guesswork involved in removing the controller.

I have attached some ebike porn for those who like that (softcore)

[Bikemad]

I'm very pleased to hear that your problem is now sorted, I just wish that all the other problems we encounter on the forum could be fixed as easily.

Now you can carry on riding with your Smart Pie again.

Alan

[Morgen 3Eman]

Hi Lollandster

Did you happen to notice what the voltage rating was on the electrolytic capacitors on your SmartPie control board?

TTFN,
Dennis

 

[Lollandster]

Yes its 63v, you can see it on the picture I took. thomab wrote earlier that it uses IRFB3077PbF mosfets witch are 75v max if that is useful info to you