Rotary position sensor - how might this work?

[metrologist]

I'm curious how this might work. It seems to end up having ~100 different positions it can detect. I drew the wipers in white which work in tandem to bridge the traces. What might the traces be made of? Some kind of resistive carbon film? And, is the representative schematic correct?

[mikerj]

This is the type of sensor you might find on e.g. a throttle pedal in a car.  The schematic is not correct, the device is much simpler.  It is effectively two potentiometers with the tracks connected in parallel and with one wiper increasing and one wiper decreasing as the pot is turned, which gives a way of verifying the position and correct operation of the sensor since it is a safety critical part.

[metrologist]

This was my first crack at the schematic, but then I saw more paths between pins.

And yes, it is a throttle position sensor.

It is becoming an increasing failure. People are measuring ohms and not seeing a problem. This one opened and no signs of physical wear or damage. But a new part solves the jumpy throttle issues...

[mikerj]

The circuit is more like this (pin numbering left to right as per your photo).  The resistance between the wiper pin and the wiper itself will vary a little over the range, but provided it's connected to a high impedance circuit this won't make any significant difference.

Intermittent operation without a worn carbon track is likely down to oxidation of the wiper material, contamination on the surface of the track or perhaps insufficient wiping pressure.  Is this a genuine OEM part or a cheap aftermarket replacement?  A lot of aftermarket automotive stuff is absolute junk.

[metrologist]

These are genuine Yamaha TPS. Nobody has reported seeing a problem when using a DMM to measure resistance. Also the diagnostic mode, when it's installed, does not indicate a jump in reported numbers, but in actual operation the throttle operation is jumpy or rough, and replacing the TPS always fixes the problem.

I have not experienced it myself, and I am aware that the design changed in recent years from a single wiper to dual wiper design. Each wiper has 4 fingers on each track, and you can see marks where they rub, but they are not worn through. The unit is well sealed and nobody has reported a visual sign of what is causing the problem.

I suspect that whatever it is probably happens quick, so measuring voltage with an o-scope might be required to see the issue. I also thought of contamination on the track. I was also suspecting a connection issue to the TPS itself.

It's not like a rampant plague, but is one seeming benign component that gets replaced often. I was surprised to see an analog potentiometer used in this application. Why not a digital encoder? This is an important device that is a matter of safety. I've seen more robust twiddle knobs on test gear that merely changes a setting. This device controls the operation of machinery that could easily kill if that device goes haywire.

[mikerj]

The vast majority of cars and bikes use potentiometer style throttle position sensors, even the ones with "fly-by-wire" style throttle control.  They are cheap, (usualy) robust and do the job.  Hall effect sensors are starting to be used however.  An incremental digital encoder has the potential to be reliability disaster, missed steps would play havoc with the engine control.  An absolute encoder would need a lot of resolution and therefore be more expensive and more delicate.

The fact that it only gives a problem in actual use suggests a problem with vibration and maybe heat.  Is the hub that carries the wipers a nice snug fit in in the body at the point it rotates, no slop?  Is there a spring inside to bias the sensor toward the throttle closed position?

Which bike is it from?

[metrologist]

This is used in several Yamaha bikes. Maybe more resolution is used in operation, but the diagnostics only gives 99 steps if I recall that right, or it starts at 16 up to 100 on the dash readout.

There are brass bushings on the center shaft and a wavey washer to apply pressure to the center assembly with fingers against the traces.

I think the part is something 40 or 60 USD, but it is very difficult to get it out and back in because there is a lot of other stuff that needs to come off.