[solved!] Can't figure out the input circuitry for a 90's Toyota odometer JZA80

[deadlylover]

I have a digital odometer that has stopped counting, otherwise it powers on and the trip computer buttons work fine. The speedometer also works on the car.

The speed signal is a 0V>12V pulse, 4 pulses per revolution as is standard with most Toyota vehicles of this era.

I was able to buzz out some of the input circuitry, and in particular that (backwards?) diode has me stumped. If I use a function gen and inject a speed signal after the diode, the odometer starts counting again.

The most obvious explanation is that there must be a broken trace somewhere, but a quick once over with a microscope doesn't reveal anything. I may have to start desoldering parts just to get a better look at the PCB. Cold solder joints are the most common problem, but I have touched up anything I can access easily.

Another explanation would be what I think should be ground is actually a 12V supply that's shorted to ground. The speed signal is then just supposed to pull it to ground as the odometer is only looking for a "zero crossing". Actually I think there is a 13V zener on the display board as well which might be used as a reference...I better order a few replacements.

Any help would be appreciated, thanks!

I will try to reverse engineer more of the board on the weekend (and take photos, doh). I just wanted to know if I was missing something obvious and if anyone can point me in the right direction of how these things work. The service manuals for the car just say to replace the whole odometer unit, and AFAIK no one has drawn schematics yet.

The odometer is also supposed to take the 12V speed signal and buffer it down to 6V before sending it to the ECU. I will have to investigate the buffer circuit as well.

(I don't want to put the car model here yet until I fix it, don't want to waste other people's time when googling for this issue, apparently it's a very common issue and most people just replace the whole unit if touching up the solder joints doesn't fix it. It's for a 1993 Toyota __S__U__P__R__A__)

edit: all fixed, I found the pullup resistor!

1993 Toyota Supra JZA80 series 1 MKIV Supra

[Nikan]

If your schematic is correct, that diode in question is most probably a zener.
Did you inject that signal before the diode as well? If not do so.
If its not working that way, check the diode out, it might be bad.
If it's not bad, check if the trace before the diode is shorted to ground.

[floobydust]

What VSS does the car have? 2-pin or 3-pin? Does it have ABS? Turbo? Automatic?
3-pin VSS output is open-collector with a voltage-divider (2 resistors) to lower from 12V to around 0-5V swing, see MK4 pic.
2-pin VSS 89413-08010 looks like just the variable-reluctance sensor coil only, and the preamp is somewhere else as another module.

It's a puzzle because where the sensor pre-amp is, the level-translation, and distribution to ODO/ECU/ABS you have to consider the whole picture.
The ODO in the previous years might have taken the raw VR pickup input and had the preamp. Toyota might have then modded the ODO to work with logic-level VSS sensors.

[deadlylover]

If your schematic is correct, that diode in question is most probably a zener.
Did you inject that signal before the diode as well? If not do so.
If its not working that way, check the diode out, it might be bad.
If it's not bad, check if the trace before the diode is shorted to ground.

Yep I've tried injecting before the diode, it didn't work unfortunately. It wasn't shorted to ground but I'll double check when I can.

I thought it might be a zener as well, but the diode loooooks like a normal one and is a common component on the board, it's identical with things that should just be a normal diode. It measures okay but I may swap it with another on the PCB. I'll do it as a last resort because it's awfully close to the connector and if I damage it, it will be damn near impossible to find a replacement. 

Quite a few components are under the VFD display so it's tricky to probe around too. I need to buy some nice probe grabber hooks.

What VSS does the car have? 2-pin or 3-pin? Does it have ABS? Turbo? Automatic?
3-pin VSS output is open-collector with a voltage-divider (2 resistors) to lower from 12V to around 0-5V swing, see MK4 pic.

I'm 99% sure it's a 3 pin. The car is originally a N/A Auto, no ABS. (car is now turbo conversion with a piggy back ECU controlling timing/ignition/transmission, original engine ECU still in place, but transmission ECU removed)

The VSS looks to be functioning fine, I measure 0>12V while driving, and the speedo works no problem.

I know a common "mod" to fix a speed sensor code due to a dodgy odo buffer circuit is to just cut the pink wire to the ECU and splice it to the VSS. The odometer still counts after the "mod" so we might be able to conclude that the ECU side has nothing to do with it counting.

The more I think about it, I'm pretty sure the odometer should be "generating" ~12V on its own (either by pullup or a 13V zener on board), and the speed sensor is grounding it in normal operation (unless of course I'm totally wrong, I put this in the beginners section for a reason!  ). I guess lots of inputs/signals work this way to help protect against voltage spikes and stuff? The input diode was confusing the hell out of me, and I was wondering why the "fix" could be to just reverse it.

Thanks guys for chiming in, I'll have the car available this weekend and now I think I have some direction. For sure I'll trace out more of the schematic for future repairers, these cars have somehow shot up in value in the past few years and spare parts are getting impossible to find. A spare (possibly broken as well) odo module shows up every few months for ~$150 in Japan, I will have to grab one next time.

[Renate]

Maybe the diode is just a signal diode.
Maybe there is a pullup on the right side of the diode that goes to something less than 12 V, maybe 5 V?

[deadlylover]

Solved, thanks everyone!

Yep there was a pullup, it was kind of hiding between the plug and some chonky power resistors. In hindsight it should have been obvious because it was sitting between lots of mechanical stress from the plug and the heat from the resistors. 

edit: just had to add some flux and touch up the joints on the resistor, it wasn't making contact with the board