Magic Pie III - Full throttle for no reason.

[fx-82au]

Hi everyone, firstly, what a great forum. It's so nice to find such a great support community for GM products out here.

I have a problem. I have a brand new Magic Pie III (Internal controller, 48 V, 10 Ah battery 26" Rear wheel.

Backstory
I'm putting this long version of the problem here for reference in case others have the same problem.
While riding home tonight, I came to a stop at some traffic lights, then felt my bike shudder forward. I thought I'd knocked my thumb throttle somehow, so I pulled on my mechanical brakes (I do not have regen brakes fitted yet, so they couldn't disconnect the motor) but the motor kept pushing. My next thought was that must have knocked the thumb throttle, then hit the cruise control button, so I released the brakes and allowed the bike to accelerate while I hit the cruise control again to disable it and slow down. This did not work. The motor continued to accelerate at full throttle and I got no response from my controls. I eventually reached around to my battery pack and switched the motor off at the key.
When I dismounted and turned the key to ON again, the throttle lights turned on again, indicating full charge, but the motor would not respond to any throttle input. I turned it off and on a few times and eventually the throttle worked again, but again while stationary the motor started to accelerate with no input from my throttle! I spent the rest of the ride home alternately pedaling with the battery off, and using the throttle up hills, ready to turn around and switch it off as soon as it went to this unwanted full throttle mode. This occurred several times again on the ride home.

Short Version
New GM MP3. For some reason motor is ramping to full throttle with no input from my thumb throttle. It is happening automatically and intermittently. Once it starts the only way to stop the bike is to switch it off at the key. I have mechanical brakes only, no regen system installed so far.

I have done a fair bit of reading on the forums and suspect it may be a Hall Effect Sensor failure, though not all the symptoms fit. When I turn the key on I hear NO beeps from the motor. When I push the bike very slowly, I feel only a very slight juddering, but not the shaking that everyone else describes. I suspect I am only noticing it because I am expecting to feel some juddering. When the bike moves a bit faster (>3km/h) I cannot feel this judder at all.

Additional Info
The weather was very cool at around 15 C
There was some rain today. The throttle got a little wet, but it was only trickling. This occurred 6 hours before the problem started so it should have dried right out. I rode it straight after the rain and did not experience this problem then, so I think it had nothing to do with the rain.

Does anybody have any suggestions? I'd love to feel safe on this thing again and not expect it to just take off when I'm not ready!

EDIT: I understand that

A broken throttle ground wire or a poor connection in the ground circuit could cause full throttle. -Alan

This feels to me like it may the the source of my problem. On one occasion, I unplugged the throttle from the main connector, thus creating an open circuit, and I zoomed off at full throttle straight away. So If I have a faulty connection in my ground wire, and it occasionally causes an open circuit, this would send me zooming off into the sunset unintentionally, right? GM China sent me two thumb throttles, either by mistake, or because this is a problem they anticipated, so I will swap tomorrow and report back.


Julian

[Bikemad]

Hi Julian andto the forum.

The problem you have described is definitely not Hall Sensor related, and as it also exists with the throttle unit unplugged, it is not being caused by either the throttle control or a broken ground connection.

I suspect the problem may be due to water ingress on the main eight pin round connector where the front harness plugs into the motor harness, and I suggest you have a good close look to see if there is any sign of moisture in there. If water has managed to get inside the connector housing, it can conduct enough voltage from the battery gauge supply to operate the throttle.

The reason I know this can happen, is because I have just performed a very simple test to confirm my suspicions:

I plugged 2 wires from a disassembled battery balance lead extension cable into the to Battery + and the Throttle Signal connection on the motor harness:



and then gradually dipped the connector end into a small amount of tap water, and the motor instantly started to run the moment the contacts touched the surface of the water!

Inside the eight pin connector, the  Battery + and Throttle Signal pins are adjacent to each other and are only separated by a gap of around 2mm, which is the same as the spacing on the balance lead connector terminal that I used for the test.
If the water conducts enough volts with my 29.6V battery, the problem would occur even more easily with a 48V battery.

However, the motor did not run when I tried the same test with the connector lead attached to the +5V and Throttle Signal connections.

If you have routed the wiring down the rear of the seat-tube, the connector will probably be directly in line with any water being thrown off the tyre (unless you have mudguards fitted) and if the plug and socket joint is not 100% waterproof , the water could easily make its way into the connector and cause this problem.

I usually place some insulation tape around the joint of the plug and socket to keep any water out, and I have also positioned the connector on the front of the seat-tube, further away from the direct line of fire.

I suggest you dry the two halves of the connector thoroughly with a hair dryer, and then apply some petroleum jelly (Vaseline) around the plug before pushing it in fully, and this should hopefully keep the water out.

I really don't think that "watermatic" throttle action is likely to be of any use on an ebike.

Alan
 

[fx-82au]

Thanks for the quick reply Alan.

That sounds like it may have been the issue. Before your reply I had done some testing to confirm that neither the controller not the throttle were faulty by shorting the pins on the #3 connector and the 8 pin connector. Both of these sent the motor flying. This confused me somewhat, as while riding last night I conducted a smal test by pulling the throttle connection (#3) from the motor wiring connector, expecting this to stop me, but it sent the motor into uncontrolled acceleration. I thought this meant that the throttle operated as High resistance=high speed, low resistance=low speed (a silly and unsafe system), but what I now believe actually happened is that in pulling that plug, I allowed some more water to fall into that connection and short the wires.

I have been testing today and have not experienced the problem again. My wires are drying in the sun and needless to say they will be waterproofed as soon as they are reconnected.

Solution
As Alan suggested, the problem seemed to be caused by water shorting out some connections in both the 8 pin connector and the throttle connection.
The problem was resolved after allowing the equipment to dry and retesting.
I will now be waterproofing all connections and placing a plastic sleeve over my handlebars when the bike is likely to be left out to get wet.

Thanks for the help Alan. I appreciate the time you took to test that solution for me.

Julian