Gear Position Sensor- Again

Ducati Forum

Help Support Ducati Forum:

Joined
Sep 29, 2019
Messages
2,045
Location
London
Hey guys,
so I was at the dealership, getting a new sensor under warranty. I have no real world data on whether the hirschmann or novotec one is better, but I suspect the hirschmann is
I did splice the hirschmann into my 2022 loom, to test it, but it gave back completely nonsense in terms of which gear.
However, I was talking with a tech, and we ended up talking about the gear learning procedure. The bike must be plugged into their diagnostic tool, and then ran up and down the box.
And it struck me, could the ecu 'learn' the hirschmann GPS?
The tech didn't know, and I'm willing to give it a try, if my sensors fail again....
 
The V4 and the 99s share common shift mechanism parts and functionality. On the 99s their is also talk of this gear learning wizardry. Never seen it. You can take a shift assembly cover (spare, off the bike) attache the bikes gps to it, grab a magnet, drop it in the drum hole on the cover, turn it with your finger and watch the gear display on the dash move.

That as simple as it is. The flow of information is GPS, to ECU to dash display. How in the hell does that get electrically “lost” when it is driven by a non adjustable, (unless you physically rotate the GPS or drum assembly holding the magnet) mechanical drive?

Whenever I have assembled, the gearbox cover or swapped the GPS, I screw the GPS to the cover, put the gearbox in 1st, power the bike on, stick the star drum in the cover and rotate the drum until the display says 1. Stick the star drum in its posts, install the cover, done. Just don’t understand how swapping a GPS screws that up.
 
I would imagine, Endo, that the GPS outputs a different voltage for each position it's aligned with the magnet. And whether it be manufacturing tolerances, or pixie dust, there's a relearning procedure programmed in to the ECU, to allow for variance. Maybe.
 
I’ll measure that for you. I have a bench setup with a functional electrical system. I’m betting there is no voltage change. It’s a simple magnetic trigger. So I’ll hook a meter inline with the GPS and look at voltage signals as I manually rotate the drum in the cover assembly and take a look. My point was is that it’s a fixed physical interface. If you unscrew your GPS right now, turn the bike on, hold the GPS in place on the bike and start rotating it, tell me what’s going to happen. Correct, your going to be emulating the magnetic shift drum rotating as it does when you change gears. How is a fixed mechanical assembly now being misinterpreted by the ECU?
 
Last edited:
This is as simple as it is, v2 and v4 are the same. You have a magnet in the drum with polarity. You have a magnetic pickup sensor that reads where that polarity is in relevance to its rotational position. In this depiction, I have 6 possible position options within the magnetic drum and the star selector and 4 possible position options between the cover and the sensor.

Once this is set the physical mechanical setup in the shift assembly is set in stone. It is mechanically correct in relation to the actual gear and the sensor. Once this assembly is mechanically manipulated to correct reference on the dash, the only way to screw that up is if the ECU looses its sync. Why is that happening? Again I can swap any ECU in this bike and it’s never out of sync.
 

Attachments

  • 59B73F34-0A79-4E33-9F23-A27FC548EEEC.jpeg
    59B73F34-0A79-4E33-9F23-A27FC548EEEC.jpeg
    43.6 KB
  • C7F63432-53A2-464B-B88A-EF1DC80A5F4C.jpeg
    C7F63432-53A2-464B-B88A-EF1DC80A5F4C.jpeg
    45.5 KB
Hey BP, I do have the race belly pan fitted, but I have the Ducati ducting for the gear position sensor. Also, my belly pan has cutouts for the Akrapovic exhaust, so plenty of airflow in the belly pan. On the V4R, with the air duct, GPS hasn't failed, same belly pan.
 
This is as simple as it is, v2 and v4 are the same. You have a magnet in the drum with polarity. You have a magnetic pickup sensor that reads where that polarity is in relevance to its rotational position. In this depiction, I have 6 possible position options within the magnetic drum and the star selector and 4 possible position options between the cover and the sensor.

Once this is set the physical mechanical setup in the shift assembly is set in stone. It is mechanically correct in relation to the actual gear and the sensor. Once this assembly is mechanically manipulated to correct reference on the dash, the only way to screw that up is if the ECU looses its sync. Why is that happening? Again I can swap any ECU in this bike and it’s never out of sync.

I don't know how it works Endo, but it's not that simple. If you fit the sensor upside down, it will read, but gears in the wrong order. But does increment +1 each turn. With the Hirschmann sensor, it reads really randomly. 3,5,N,1,N,4,6 for example. And I tried the Hirschman upside down too, no joy.
 
Rick, is the GPS that you referenced in the video you just posted now not operational? Junk, going in the trash can?
 
The 2023 models have a new air conveyor (+ new part number for lower LH fairing), looks like this is aiming air towards the GPS?

V4 MY23 air conveyer_1.jpg
 

Attachments

  • V4 MY23 air conveyer.pdf
    270.8 KB
That's interesting. One of the tech bulletins mentioned the magnet placement was different. I don't get what the learning thing really does if it wouldn't adapt. I wonder if its just a check for the ECU which just verifies that all 6 gears are read correctly since so many functions are dependent on each other.
 
That's interesting. One of the tech bulletins mentioned the magnet placement was different. I don't get what the learning thing really does if it wouldn't adapt. I wonder if its just a check for the ECU which just verifies that all 6 gears are read correctly since so many functions are dependent on each other.

Yeah, me neither. You're probably right, just verifies what it's expecting.
 

Register CTA

Register on Ducati Forum! This sidebar will go away, and you will see fewer ads.

Recent Discussions

Back
Top