How does this sensor work?

Maybe it's a miniature rotary optical encoder. Think OptiSpark trigger wheel. Two pins would be power to the LED and one would be output. There could even be multiple encoder tracks inside with a chip to process/MUX and provide a single digital output. Something like that would be hard to bench test unless you knew what's inside apriori. Potentiometer would have been my first guess as well, but they don't often turn 360°. As a gear position sensor, I have to believe it's configured to report absolute rotation vs. relative. 

You could try checking ohms between pin pairs with leads "forward" and "reversed"--if there is an LED inside, that would provide a clue.

Maybe you can get hold of a junk one and dissect it. 

Nockenwelle said:

Maybe it's a miniature rotary optical encoder. Think OptiSpark trigger wheel. Two pins would be power to the LED and one would be output. There could even be multiple encoder tracks inside with a chip to process/MUX and provide a single digital output. Something like that would be hard to bench test unless you knew what's inside apriori. Potentiometer would have been my first guess as well, but they don't often turn 360°. As a gear position sensor, I have to believe it's configured to report absolute rotation vs. relative. 

You could try checking ohms between pin pairs with leads "forward" and "reversed"--if there is an LED inside, that would provide a clue.

Maybe you can get hold of a junk one and dissect it. 

I hope it's not that complex because if it is it's out of my wheelhouse!

It occurs to me I could just use that transmission shaft and adapt it to a 6 position switch and just have a different colored light per gear.

I'd bet on Hall Effect.

On a 5V Hall sensor, the three pins are: ground, +5V power, and variable ~0.5-4V signal which depends on magnetic field strength.  Minimum signal is >0V, max is <5V, and you get built-in fault detection unlike a potentiometer.  With power on (and hooked up the correct way) it looks sorta like a pot but slightly different voltage range.  Without power, it just sits there and a resistance check shows open circuit.

The shift position sensor attaches to the end of the shift drum.  If you read the voltage off the sensor, you can figure out the rotation angle of the shift drum, and you know what gear your in.

Looks like earlier MT09's used a separate mechanical contact for each gear position, "simpler" but less reliable.

Free service manual for a 2021 MT-09: https://www.manualslib.com/manual/2804501/Yamaha-Mt-09-2021.html

That manual may or may not be worth anything, but some gear position sensors on ebay said they worked for 21-23 or 21-24

On the wiring diagrams, the shift position sensor shows three wires all coming from the ECU, and does not indicate what each wire does.  Wonderful.  I didn't see a test for the shift posn sensor, but I could well have missed it.

pg 419 of the manual: the test for the throttlebody servo uses 3V (2x 'C' batteries), so the ECU seems to run other stuff on less-than-12V?

Very surprisingly, the crank position sensor is 2-wire variable resistance?  I have no idea how that works.

Here's a 'simple' DIY Arduino project that includes a gear indicator.  The basic idea seems to work, the difficulty is filtering out 'noise' to make it more robust/reliable (which is the difficult part of many 'simple' Arduino/etc projects)

https://forum.arduino.cc/t/motorcycle-gear-position-indicator/238303/6

The previous generation FZ engine cars did have a gear indicator. I suppose that contact style sensor was easier to adapt. I suppose it would be an idea too use the old gear indicator and adapt that older sensor to the newer motor.

I have seen those gear indicators on Amazon and eBay but I think they are calculating the gear from engine RPM and ground speed. And most likely that information is already available in the bike ECU but not the car ECU.