Pedelec sensor

[Fordulike]

Sometimes when i stop to admire the scenery, i have a habit of pedalling backwards. I would expect the bike to move forward as the pedelec sensor detects rotation of the crank.
I just wondered how the sensor can tell that i'm not pedalling forwards.
It appears to be only a disc with little magnets in it and a block of plastic with a wire leading to the controller.

 

[Mussels]

The magnets aren't evenly spaced so produce a different pattern of signals depending which way round the pedals go.

 

[Leesome]

More wisdom, just braw.

 

[Ratzakli]

The magnets aren't evenly spaced so produce a different pattern of signals depending which way round the pedals go.

Such a simple, yet brilliant solution!

 

[onmebike]

If you turn your magnet disc over you will need to pedal backwards to get drive. Magnets have opposite pole's, one way they attract, the other they repel. The sensor must also have magnetic properties to respond in the opposite direction depending on which side of the disc faces it.
Dave, Old timer has done quite a lot of experimenting with the disc's, more magnets cause power to come in earlier or rather at a slower pedal rate, whereas less magnets have the opposite effect. Too many magnets will also confuse the sensor and drive becomes erratic.
Dave's disc's as with mine have evenly spaced magnets.
On his Alien conversion for instance Dave increased the magnets from 5 to 6 to bring the power in a little earlier, the sensor thinks the pedals are turning slightly faster. Increasing them beyond 6 caused the drive to become erratic.

 

[Fordulike]

Thanks for the information

 

[NRG]

These pedelecs sensors are Hall effect sensors with additional logic to create single direction square wave pulses that can be fed into microcontroller inputs directly. As Onmebike says they are sensitive to one pole only due to the sign of the charge carrier used IE: negative or positive so they will only be triggered by a north pole magnetic field or a south pole field. Get the pedelec disk the wrong way around and they won't work. The additional logic used to create a clean squarewave output also ensures the passing direction of the magnetic field will only trigger in one direction only, hence pedaling backwards won't run the motor.

 

[Fordulike]

Crikey and i so wanted to believe they worked by magic

 

[lemmy]

Crikey and i so wanted to believe they worked by magic

Any sufficiently advanced technology is indistinguishable from magic
Arthur C. Clarke's Third law

 

[Deleted member 4366]

I want to make my own pedelec sensor because I don't have enough space to fit the standard disc one supplied with my Sunlova kit. I have the idea to stick magnet/s to the inner chainwheel and position a hall effect sensor in an appropriate position.

A couple of questions for you technical guys:
What sort of hall effect sensor do I need? Is it a latching one?
I still don't understand from the above posts why the controller won't give power for back-pedalling. Could someone give me a definitive technical explanation. Is it because the sensor gives a different result if the magnetic flux comes in the oposite direction, or is it because a latching sensor is used to give a square wave pattern, analysed by the controller, or is it something else?
Alternatively could someone give me a product designation for the sensor and tell me what type of magnets to use and their pole arrangemets and spacing.
Thanks in advance for any help given.

 

[NRG]

I want to make my own pedelec sensor because I don't have enough space to fit the standard disc one supplied with my Sunlova kit. I have the idea to stick magnet/s to the inner chainwheel and position a hall effect sensor in an appropriate position.

Not worth the effort IMHO, just position the supplied sensor on the other side of the bottom bracket...simples

.....I still don't understand from the above posts why the controller won't give power for back-pedalling. Could someone give me a definitive technical explanation. Is it because the sensor gives a different result if the magnetic flux comes in the oposite direction, or is it because a latching sensor is used to give a square wave pattern, analysed by the controller, or is it something else?

I suggest you look up some data sheets on hall effect sensors to get a more in-depth technical answer, however, it will be along the lines of my post above above...

 

[Deleted member 4366]

Not worth the effort IMHO, just position the supplied sensor on the other side of the bottom bracket...simples

Not simples! Didn't someone say they don't work backwards. I bought a left hand sensor for exactly that purpose. Unfortunately, it doesn't fit!. There is insufficient clearance between the pedal and the BB. The kit supplied a longer replacement spindle, but it is too long on the chainwheel side, so I can only use the smallest gear (of 3) and there's a 15mm gap between the inner chainwheel and the BB
.

I suggest you look up some data sheets on hall effect sensors to get a more in-depth technical answer, however, it will be along the lines of my post above above...

I have looked up every data sheet on every type of hall effect sensor and I now have a good understanding of how they work, but there's lots of different types that do different things. If I knew the signal that the controller would respond to, I would be able to choose the appropriate one and associated magnet/s. This is what I really need to know. Thanks for taking the time to help.

 

[NRG]

I have looked up every data sheet on every type of hall effect sensor and I now have a good understanding of how they work, but there's lots of different types that do different things. If I knew the signal that the controller would respond to, I would be able to choose the appropriate one and associated magnet/s. This is what I really need to know. Thanks for taking the time to help.

There are a number on the forum who have fitted the sensor to the opposite side of the BB, myself included. Perhaps a picture of the problem would help, including the other side of the BB.

The controller should respond to a positive going pulse at 1.5Hz to 2Hz and be at maximum from 6 to 7Hz. (5 magnets at 20 to 80 crank RPM)