PAS problem

[Robert M]

Hello guys

I hope one of you can help

My PAS sensor has recently developed a fault. The cable had initially an intermittent fault very close to the sensor head and is now dead !

The sensor is labeled as a power assist sensor V series

It was supplied with +5V (red), 0V black and and the function on peddling was to pull down the a signal on the output wire (green) from 5V to something just below 4.2 V.
That was enough to tell the controller that it should power the motor. Energy to the motor is set on the display unit of the controller.

I have bought a PAS sensor on ebay, but that seems to have a different function. The signal on the output wire is 0V and on pedalling the signal voltage increases. I haven't connected that PAS sensor to the controller in case something might go boom in the controller box.

Can anybody please advise me on what PAS sensor ( and where ) I should get.

https://www.dropbox.com/s/ur2axhf2guyjfg2/IMG_1718.JPG?dl=0

 

[Deleted member 4366]

The controllrr counts pulses. Nearly all sensors pulse on and off, so they should work as long as you match rhe three wires: 5v, 0v and signal.

 

[Robert M]

Dave,
Thanks for your info and help.
I've grabbed myself an oscilloscope from work and looked at the signalss and yes, you are perfectly correct. The sensor is generating pulses. For the electronic wizards amongst you, I 've found that the sensor acts as a sink of a 5V signal which comes from the controller.
The sensor is supplied by +5 and 0V on red and black.
If you just supply the sensor with the supply voltage, you will see that the signal line is open circuit if there is no magnet in the vicinity. )n a scope you will typically just see some 50Hz hum or noise at the 10-30mV level. When you bring in a magnet, the sensor switches on and connect the signal line to ground, hence the hum will disappear. The controller has a 5V open collector output on the signal line and when the signal line is connected to the sensor, the signal line will be pulled to zero when a magnet is moved across the sensor.

One interesting thing I noticed is the way the controller recognizes which direction you pedal. I.e if you pedal forward you want you power to the motor - however if you pedal backwards, the motor will not be energized.

I looked at the timing of the high low signals on the scope.
When you pedal forward, the duration of the low signal is shorter than the high signal, however if you pedal backwards, the duration of the low signal is now longer than the high signal. It appears that the controller looks at this timing ratio and only switches on the motor if the low signal is shorter than the high signal.

Does anybody know how this is accomplished? I've taken apart the faulty sensor and it appears that there is only one component inside the housing. From what understand, this is just a single simple hall effect sensor. The only other component I found in there is a small capacitor soldered directly between 5V and gnd.


Even though the magnets are moving across the sensor regardless if they move forward or backwards,

Regards
Robert

 

[Robert M]

Just to one thing to add.
The controller switches the motor on after recognizing 5 pulses within 3 seconds.

Also, when I say "open collector", I am not absolutely sure if that is the correct term for such a connection pin, i.e. a current limited connection on an IC or transistor which is floating at a high potential (5V) when unconnected and is switched by pulling to ground.

 

[Deleted member 4366]

For the signal wires, there's often a pull-up resistor in the controller that holds the input high. The external device (switch, jumper, etc) pulls it to ground.

I made a throttle enabler that only allowed throttle when pedalling. It used a normal pedal sensor and a Picaxe type microprocessor. It didn't work until I put a pull-up resistor on the pedal sensor input.