Electric Bicycle Controller Board

[ChrisOfBristol]

My electric bicycle has stopped working.

I have checked* the voltage regulator (7812), the Dual Schottky diode (MBR20100CT) and the NMOSFET (RFP70N06). I haven't checked the low power chips which I think are unlikely to break. These are a pulse width oscillator (TL494CN) a dual op-amp (LM358P) and quad nand gate (CD4001BCN). I have checked the "peripherals" like the throttle and they all work except the speed sensor which I am suspicious of.

Since it is a low power bike (150W - the limit is 250W), it is unlikely to exceed the speed limit, so I could just disable the speed sensor. I believe that it produces a pulse with frequency (or width) related to speed. I am also guessing that the PWM produces a pulse with the frequency (or width) modulated by the throttle. The difference probably produces a pulse for the Schottky diode. Is this, or something like it, likely to be correct? If so, could I short circuit the comparison somehow?

* Thanks to help on this site on a different thread about a different electric bicycle.

 

[Diver300]

I guess that it's a DC motor, and the N-channel MOSFET supplies battery power to the motor. The power will be controlled by PWM. The Schottky diode is a freewheel diode to prevent the inductance of the motor windings from blowing the MOSFET when the current turns off. The inductance of the motor windings will mean that when the MOSFET turns off, the current in the motor can't stop instantly. When the MOSFET turns off, the current will run through the Schottky diode instead.

The speed sensor is likely to be shorting the sense wire to -ve each time a magnet passes the sensor. The +ve is likely to be the supply to the sensor only.

 

[ChrisOfBristol]

I guess that it's a DC motor, and the N-channel MOSFET supplies battery power to the motor. The power will be controlled by PWM. The Schottky diode is a freewheel diode to prevent the inductance of the motor windings from blowing the MOSFET when the current turns off. The inductance of the motor windings will mean that when the MOSFET turns off, the current in the motor can't stop instantly. When the MOSFET turns off, the current will run through the Schottky diode instead.

That all sounds right to me.

]The speed sensor is likely to be shorting the sense wire to -ve each time a magnet passes the sensor. The +ve is likely to be the supply to the sensor only.

I originally assumed that it was a Hall effect transistor, but the supply is 12V and they are only 5V.

 

[Diver300]

I originally assumed that it was a Hall effect transistor, but the supply is 12V and they are only 5V.

Hall sensors may look like transistors but they are ICs.

All the ones that I have seen will work to 12 V, or even 24 V. They will also work down to about 3 V.

 

[ChrisOfBristol]

I have a spare 49E but this says max 10V.

 

[Diver300]

I don't know if the sensor on your bike is a 49E. If it is one, it shouldn't be run above 10 V.

I've seen the Allegro sensors used in lots of applications, and they go to higher voltages:-

https://www.allegromicro.com/en/pro...-latches/hall-effect-latches-bipolar-switches

 

[ChrisOfBristol]

I've seen the Allegro sensors used in lots of applications, and they go to higher voltages:-

That is more likely then. Someone gave me the impression that there was some fancy IC in the sensor with an oscillator that produced a pulse with frequency (or width) related to speed, not just a pulse every time a magnet went past. I took that to mean that it wasn't a simple Hall sensor, although I couldn't see the point in doing it in such a complicated way. Also I would have expected it to motor to run without it.

However I think you may be right in thinking that it is just a higher voltage Hall sensor chip. I could do some tests with a resistor perhaps, it may need some default voltage output from the sensor to work.

 

[Diver300]

It could be that the bike will only run for a really short time if there is no pulse, or it might even let the rider get the bike going before allowing the motor to help.

A pulse each time the magnet goes past does produce a frequency proportional to speed. That is how rotational speeds are often measured. Commercial vehicle tachographs and car ABS systems do that.

 

[ChrisOfBristol]

A pulse each time the magnet goes past does produce a frequency proportional to speed.

Exactly, that's what I meant by "I couldn't see the point in doing it in such a complicated way". Why have a circuit that senses the magnets passing, then uses that to control the frequency of an oscillator? You've already got that. I think the guy who told me this didn't really know what he was talking about, or alternatively explained it very badly.

 

[Nigel Goodwin]

Exactly, that's what I meant by "I couldn't see the point in doing it in such a complicated way". Why have a circuit that senses the magnets passing, then uses that to control the frequency of an oscillator? You've already got that. I think the guy who told me this didn't really know what he was talking about, or alternatively explained it very badly.

It could be either - a pulse per revolution IS a frequency change with speed, as you say, there's no point feeding an extra oscillator.