2001 rancher 350ES 4x4

Picked it up from a buddy real cheap. Did not run an had a known shift issue. Full tune up on the motor got her running like brand new. As for the shifting I have done a ton of research into this specific issue. The bike will shift up-and-down through all gears including reverse as long as you keep the RPMs low. In any gear if you rev high for even a short period of time it will lock out in that gear. Resetting the key resets the issue as usual and allows you to down shift to neutral. This can be repeated every single time. The meter display does not work properly so I cannot see much on the display. I can see the top of these 0 for the speedometer. Speedometer always reads 0. During the tune up on the motor the clutch was adjusted and the shift motor was removed. Cleaned out the inside of the shift motor as well as old grease on the gears. Gears were then re greased. Angle position sensor was replaced from an extra one that a friend had. No change.

My question is related to voltage going into the speed sensor. Ground wire to the incoming battery voltage wire is exactly 12 V. I am also receiving 9.5 V into The speed sensor on the signal wire. Can anyone confirm if this is correct?

RESOLVED
Seems like no one posts the resolve. Found out that if the bike is let to idle for 20mins and gets to full op temp it would register speed and shift correctly. Ordered a speed sensor part number 37700-HN5-M51. This is actually not the one listed for my bike but was easier to find. Looked like the same one but for different years. Issue has been FIXED. shifts good and reads speed. Hope this makes it to someone needing help

 

Is it possible to put it in gear, maybe wheels off the ground, and turn the engine over a few times extremely slowly whilst watching the 9.5V - does it vary at all at any point as the engine and drivetrain all turns? I'm not at all familiar with quads to be honest but I'd expect to see the voltage change at some point if it's outputting pulses.. if possible maybe also try with the sensor wire disconnected from the ecu to be sure you're only looking at the hall sensor & it's output is not getting overridden by some kind of backfeed from the ecu

I was going to suggest trying to fool the ecu by connecting it to a variable resistor rather than the speed sensor, and then seeing how the speedometer responds to a 'fake' input - but without knowing the exact circuit parameters I'd be very scared of frying a good ecu that way. Maybe if you have sufficient electronics knowledge and info about the circuit you could try something like that

 

To answer your question yes I have tried To see if voltage changes at all. I even raked it up so I could watch the meter as the bike is driven naturally in gear. I have seen a couple people on forums say that the issue just like mine was resolved with a new speed sensor. Also saw that they had found broken wires. Physically inspected all the wires on the sensor and nothing is broken. I'm not one to throw parts at something expresly when a new sensor is a $120

The main problem that I'm having is I don't know if there should be any voltage going into the sensor through the signal wire. A Hall effect sensor is either a pull up or pulled down style. I don't know what the specifications are for this circuit. If possible I'd like for somebody to test the signal wire voltage with the speed sensor unplug and let me know if I should have voltage there or not. But I have been running in my mind to see if there is a way to modulate DC voltage to trick the ECU in to thinking that the speed sensor is working correctly. I'm afraid that the ECU is already badd and unnecessarily sending out voltage on the signal wire.

I've tested Hall effect sensors before on cars. Usually you can use A T pin in the back of the wiring harness and observe signal voltage while passing a piece of metal in front of the sensor head itself. This is done with the sensor removed and you should be able to see modulation.

 

In all this testing, have you actually tested what voltage is being fed to the battery with engine running. Most likely thing is voltage from regulator too high when engine is revved up, voltage sensor in ECU detects this and goes into fault mode. Later 350ES used to fault code if this occurred but the 01 model didn't code, just stopped working until switched off.

If it is a speed sensor fault, I haven't got the 350 wring diagram handy, but the early 450ES has three wires from sensor, a green earth, a pink that goes to both ECU and speedo, and a black that goes to speedo only, these wires change colour so check what they turn into (pink/green and black/blue on the 450) then bell out the one going to ECU to check it isn't broken. However, lots of wires go from ECU to speedo and any of them may be involved in speed sensing as well. The foot change 450 shows the sensor getting 12v direct from ignition switch via the pink wire.

 

Thats a good point I had not thought about that! I have checked the voltage input to the battery but cannot remember what it was. Do we have an acceptable range? I have confirmed exactly 12v volts though the system. Also traced the 9.5v out of the ecu, into the speed sensor and into the meter. Main question is if that 9.5v should be there

 

Voltage should be more than with engine switched off and less than 15v, usual max is 14.5v but up to 15v is OK. I would assume that if the foot shift speedo sensor is fed 12v then the ES would be too, its voltage may not matter as it is energising the coil that provides a magnetic signal to the black wire side, doubt if anything should go to the speedo via the black until the bike is moving, but don't know for certain.

 

I got good charging voltage. I kind of figured that there shouldn't be anything going into the sensor on the signal wire. Was hoping I could get someone that has a working one to test it

 

[Edited to remove URL of non original designer.] For "Whoever's behind the above" see next post ;-)

Whoever's behind the above probably knows! Don't know if you could contact them and ask.. try their listing on ebay for the rancher 350es. There's a 'contact the seller' option. Another long shot, hope it helps

Yep obviously the easiest would be if someone who's got one could meter it out for you, afraid I don't know of one round here at all otherwise I'd try :-(

 

The shift kit seller you mention does not know much about Honda electronics. In fact, he is selling a copy of my circuit. He did not design it himself. He does not know anything about Ranchers and Foremans, only the Recon that he owns and rides on paved roads. My shift kit is available at:
subzerosound.com/HondaECUBypass.html

I am an electronics expert. The Hall Effect sensor generates pulses at a very high frequency which would need a computer to count the fast stream of pulses. You can use an oscilloscope to see the pulses if you have one. So the steady state output signal voltage is not important. It drops to 0 volts when the magnetic field changes, caused by the notches on the gear shaped speed sensor wheel.

If you want to make your own, you can use the Hall effect sensor ($2) and magnet ($2.50) at adafruit.com. However, configuring them just right would take a lot of experimentation.