Wierd Pedal Assist Problem

[turbulent]

Hello everyone,

So something weird is going on with my pedal sensor. A few things actually. when starting from stationary it works and as long as I keep pedelling it keeps power. As soon as I stop pedelling and restart pedelling it doesn't kick back in until i stop the bike and start again. Also, if I start off pedelling too hard it doesn't kick in at all. I've cleaned the sensor and magnets. Tried moving the sensor different distances from the magnets, nothing works.

Have any of you ever had this specific issue before and if yes, what was the fix, a new sensor or something else?

It's a bafang hub 250W 36V 11.5AH - on a Lombardo Valderice.

Thanks for any help you can offer.

 

[turbulent]

Here's a vid to show the problem

 

[Nealh]

You mentioned 3mm gap, it is to large push the sensor to wards the disc so the gap is 1mm or nearly touching.
At 3mm rotating fast the sensor simply can't pic up the magnet signal, rotating slowly is has chance and can do so.

 

[turbulent]

Hi, yeah that's not it unfortunately. Like I said, if I keep pedelling, it's fine. If I stop pedelling for just 1 second and continue, no more power assistance until I come to a stop. Then it kicks in again, as long as I don't stop pedelling I get power.
I tried moving it closer anyway, just to rule it out, but nothing.

Just praying it's not a controller issue as it's all built into the battery system, semi intergrated

 

[turbulent]

forgot to mention. The sensor has an LED which lights everytime a magnet passes. It's lighting up fine, so the sensor must be close enough.

 

[Sturmey]

I had a similar type (kingmeter) sensor with red led that started to behave erratically and finally totally failed. I did not know initially what was wrong but the bike worked on walk assist worked ok so I suspected the pedal sensor. Anyhow, I removed the sensor (by unscrewing the small screw) and found that it was very wet and the rubber sealant had broken down. I removed all the old sealant and dried the unit over the fire and had to resolder the cable. I put the whole lot together and used clear silicon bathroom sealant to seal as a temporary fix. Its still working. I reckon these units have a hard life and are exposed to a lot of water splashing.

 

[Nealh]

Likely as Sturmey says the final option is water ingress affecting the 5v line.

 

[guerney]

It might or might not be that the magnet isn't close enough - strange things can happen even if the sensor is lighting up as the magnet passes... the weird problems can occur at higher speeds. Even though it seemed that my magnet was close enough to the sensor, my display wouldn't register any speed over 28.7mph: just kept showing 28.7mph, then 0mph, then 28.7mph over and over again till I slowed down. I was going downhill and my bike is limited to 15mph on PAS, but I wonder how the controller's software would have applied power with a strange alternating hi/no speed calculation like that - in the event of a speed sensor error, or impossible/out of bounds speed calculation, a system shutdown of PAS until the bike sufficiently slowed down, stopped, or resumed normal speed calculations again, might be a sensible bit of built-in PAS system programming for safety reasons. After I moved the magnet extremely close to the sensor, the problem disappeared. I am about to apply some self-amalgamating rubber tape around the speed sensor connector, because I certainly don't want 5v line water ingress.

 

[turbulent]

Thanks for your help guys.
I'll be taking it apart tomorrow to inspect it for water damage. I ordered another one just in case it's beyond repair. Always good to have a spare anyway as they do seem exposed down there.
Hopefully it's that, sounds right to me.

I'll let you guys know.

 

[turbulent]

update:

It's not the sensor at all. It's the bafang hub.

I switched back wheels with my wifes bike and it gave the same issue. Wish I'd tried that first.

So, this just got a lot more complicated.

 

[Nealh]

Not much within a hub to go wrong.
Speed sensing magnet may have moved or dislodged so maybe is affecting the running.
Motor Hall sensors could be faulty, these are easy to test using a meter.