Testing a PAS sensor

[passpato]

Can somebody help me please. I am repairing an ebike for a friend. Unfortunately its been messed about with before. Im pretty sure the issue is with the pas sensor or its curcuit inside the speed controller. The owner had already started fitting a new pas sensor and unfortunately I no longer have the old one that was removed. I can make the motor work by fitting a twist grip throttle and using that but I cant make the motor spin from turning the crank passed the pas. Does the wheel need to be turning at the same time as the pedal crank? How does this basic 5 magnet sensor work? I would expect to get a reading across the blue and red pas wires with an ohms meter but I get nothing. I thought I would be testing a coil. How can I test this sensor to see if it is working properly? Can I fool the speed controller with a pulse from something else to test that? Help appreciated.

 

[E-HP]

Can somebody help me please. I am repairing an ebike for a friend. Unfortunately its been messed about with before. Im pretty sure the issue is with the pas sensor or its curcuit inside the speed controller. The owner had already started fitting a new pas sensor and unfortunately I no longer have the old one that was removed. I can make the motor work by fitting a twist grip throttle and using that but I cant make the motor spin from turning the crank passed the pas. Does the wheel need to be turning at the same time as the pedal crank? How does this basic 5 magnet sensor work? I would expect to get a reading across the blue and red pas wires with an ohms meter but I get nothing. I thought I would be testing a coil. How can I test this sensor to see if it is working properly? Can I fool the speed controller with a pulse from something else to test that? Help appreciated.

PAS sensors are expecting a certain rotation of the magnets and magnet poles to function. Obviously, you don't want it to kick in when pedaling backwards, for instance. Because of that, most are made specifically to be mounted on the drive side (chain ring side) or non-drive side. Also, they are expecting the magnetic ring to be in a certain orientation (many will have directional arrows, to signify which direction the disk should rotate. Lastly, the sensor and disc magnets need to be very close (e.g. 1mm) to function.

First look at the spacing and alignment and adjust if necessary. Turn the crank in the forward direction and see what happens. If nothing, try turning the cranks backwards. If that works, you may need to move the disc and sensor to the other side. If it doesn't provide a response, you could try flipping the disc over, and testing again (forward and backwards).

The other issue is that the wiring isn't standard, so the three wires in the connector could be in a different order than what the controller uses. In that case, you may need to move the pins in the connector so that the +5V and Gnd wires align, and the third wire, the signal wires also lines up with what the controller's order is.

 

[passpato]

Thanks for the reply but I knew all of that except that most pas sensors clain they require a 3mm gap. I will try reducing he gap. I really want to know how the sensor works so that I can test it. Im guessing the possitive and negative wires go to a coil that provides a magnetic field. When the magnet passes the field breaks down and sends a pulse down the signal wire but in that case I should be able to measure a resistance with an ohms meter across the red and black wires.

 

[amberwolf]

There's a few threads that discuss the specific signals and hardware in various PAS sensors (LewTwo has a recent post you can find from his profile that shows the electronics inside one), but all of them that I have seen use a hall sensor (or multiple hall sensors) and some logic to create a digital signal with about 0v to 5v. Ebikes.ca has a page about PAS sensors that gives some examples of waveforms you can look for with an oscilloscope. Different sensors have different waveforms depending on pedal direction and possibly other characteristics; that page covers some of that, too.

 

[LewTwo]

... (LewTwo has a recent post you can find from his profile that shows the electronics inside one)...

That is at the end of this thread: https://endless-sphere.com/forums/viewtopic.php?f=49&t=116822&p=1730802#p1730802

Ebikes.ca has a page about PAS sensors that gives some examples of waveforms you can look for with an oscilloscope.

https://ebikes.ca/learn/pedal-assist.html :thumb:

 

[docw009]

Always a PITA to get an unknown sensor and figure out which of the two wires is power, and which is output. I bought a $15 ebike tester years ago and a couple of trial/error tries makes that easy,

After that, you still have to figure out if the magnets are spinning into the sensor with their N/S poles aligned right and know whether they are coming in with the right direction. I'm talking about whether the disk is mounted in the right orientation and whether it spins into the sensor in the right direction. I've got one install where it only works when the disk spins opposite to the arrow molded on it.

I sympathize. It gets frustrating when you don't even know if the controller can interpret the signal.

Anyway, there is no coil. It puts out a train of pulses that are assymetric so the controller can tell whether it spins forward/backward. Some controllers expect these pulses to be within certain frequency bands, so they need settings for disks that have more magnets. If your controller/LCD has that option, you may have to set it properly. I recently upgraded a controller and it didn't work with the existing PAS sensor because of a settings issue.