View Single Post
Old 02-01-2004, 10:39 PM   #9
vrg3
Scooby Specialist
 
Member#: 25796
Join Date: Oct 2002
Location: Cleveland, OH (sometimes DC)
Vehicle:
1993 Legacy Turbo
Burgundy

Default

I bet you could do it without microcontrollers.

When I have more free time I can come up with something more concrete, but here's what pops into my head:

If you tap the ECU's "tachometer output" wire, you get a 0-12v square wave whose frequency depends on RPM.

If you tap the ECU's "vehicle speed sensor" wire, you get a 0-5v square wave whose frequency depends on engine speed.

If we're lucky, one of these is always at a higher frequency than the other. I would guess the tachometer signal is higher frequency than the VSS. Let's work with this assumption for the moment.

Set up a counter to count clock pulses from the tach signal (through a frequency divider if necessary). On the rising edge of the VSS signal, reset the counter. On the falling edge of the signal, store the counter's output into a latch. Connect the output of the latch to a decoder.

If done right, each of the decoder's outputs will be active only in a particular gear (modulo clutch slippage or being in neutral, although the latter can be detected via the neutral switch). These can drive a little LED single-digit display with a bunch of diode logic OR gates.

Hm, in words it probably sounds much more complicated than it is. And I'm sure someone well versed in microcontrollers could do it in like 2 seconds with almost any PIC or AVR. Oh well.

It would be useful to know the actual VSS and tach signal frequency specifications.
* Registered users of the site do not see these ads.
vrg3 is offline   Reply With Quote