Driving the Griff rev counter
Driving the Griff rev counter
Author
Discussion

dnb

Original Poster:

3,330 posts

271 months

Tuesday 15th August 2006
quotequote all
I've been wrestling with this small niggle on the Griff for a while now.

My ECU has a TTL level open collector type (ie pull down but not up) style output for driving an electronic tacho. Max current is 350mA.

The tacho doesn't want to be driven by this, even with a pull-up resistor etc. (My spare impreza one works fine...)

Does anyone know anything about the waveform that drives the tacho as standard? I assume it can be quite a high voltage as it's derived from the ignition system and contains inductive parts. I know the frequency is 8x that of the engine.

Once I know this, I can make a circuit to emulate it that won't demolish the output of the ECU...

Thanks!

scoule

299 posts

313 months

Tuesday 15th August 2006
quotequote all

Did you pull-up to 12v or 5v? Pulses should be 4 x RPM if originally driven off the coil -ve.

What did the waveform look like with the pull-up in place and the tach connected? I wonder if the tach input is fairly low impedence and if the pull-up is quite high resistance maybe the tach is killing the signal?

The only other issue could be pulse width/polarity related, although I can't think of a logical reason why the rev-counter should care - unless for noise suppression.

Just rambling at the moment without a circuit diagram to hand ..

dnb

Original Poster:

3,330 posts

271 months

Tuesday 15th August 2006
quotequote all
Yes - I mean 4x that of the engine... Too busy thinking about other things as usual.

Pull up was to 12v, and was fairly high impedance to limit current, so it could be the case that the tacho is killing the signal.

No waveforms to share as my scope is dead ATM. However my DVM has a frequency setting and it seems to pick up the frequency correctly. It may not be sufficient to drive the tacho - hence my thoughts that an amplifier of some kind is needed.

scoule

299 posts

313 months

Tuesday 15th August 2006
quotequote all

My guess is that the original system will be pulled low for around 3 to 4 ms per spark ... does the manual of your ECU give a spec/waveform for the tacho output?

steve-v8s

2,924 posts

277 months

Tuesday 15th August 2006
quotequote all
What value pull up did you use ? Would think a 4k7 would be stiff enough to drive most things, but even if you went down to 1k you would still be well within the 350ma drive available. If you connect the tacho, with pull up and the multimeter does the meter still get a reading ?

mongoose

4,360 posts

284 months

Tuesday 15th August 2006
quotequote all
It's probably no help at all,but on my Omex,the 12v setting wouldn't drive my tacho,i had to set the unit to 'high votage coil signal' or some such,in order for the tacho to work.There were only 3 settings,5v,12v and 'high',so i dont know what the 'high' is sorry

dnb

Original Poster:

3,330 posts

271 months

Tuesday 15th August 2006
quotequote all
No manual for the ECU... It's open source.

I think I only have the choice of 5 or 12v - probably don't have the "high" setting. I used a 4k7 pullup, thinking it would be OK. Going to try a 1k sometime shortly.

pete500

151 posts

273 months

Tuesday 15th August 2006
quotequote all
Nice to meet you at the Growl Dave!
The rev counter needs some 200V signal, as this is what it normally gets fram the coil. I bought a "black box" frome some American company, it makes the neccessary spikes from the square 12V signal most modern ECU's make. Will supply make and model as soon as I find it. Pete

dnb

Original Poster:

3,330 posts

271 months

Tuesday 15th August 2006
quotequote all
I thought maybe 50 to 100v, but not 200v! Thanks for the black box info.

Hope you had a good time in Scotland

scoule

299 posts

313 months

Tuesday 15th August 2006
quotequote all

Would it be easier to open up the tacho and mod the input ciruitry? There must just be a F to V circuit in there with some input filtering ... LM2907/LM2917?

daxtojeiro

743 posts

275 months

Tuesday 15th August 2006
quotequote all
basically the TVR rev counter is driven from the massive pulse of the coil as it discharges, this is around 200V as has been said. The simplest way of replicating this is to use a relay coil, this is how I do it when fitting MS's to cars like the TVR. Connect one end of the relay coil to +12V ignition and the other end of the coil to the Tacho input of the dash and the tacho output of the ECU
Phil

steve-v8s

2,924 posts

277 months

Wednesday 16th August 2006
quotequote all
Blimey, would think the open collector transistor would need a bit of protection from that arrangement.

dnb

Original Poster:

3,330 posts

271 months

Wednesday 16th August 2006
quotequote all
Me too - hence why there will be at least some diodes sprinkled around the general area...

pete500

151 posts

273 months

Wednesday 16th August 2006
quotequote all
There's a 4k7 resistor between the coil and the tacho. pete

Griffter

4,143 posts

284 months

Wednesday 16th August 2006
quotequote all
pete500 said:
There's a 4k7 resistor between the coil and the tacho. pete


There's one between the LT and the ecu engine speed input - does that wire also feed the tacho?

Edited by Griffter on Wednesday 16th August 15:56

dnb

Original Poster:

3,330 posts

271 months

Wednesday 16th August 2006
quotequote all
If memory serves, there are 2 wires from the coil (each with a resistor with spade connectors on) One feeds the Lucas ECU and the other feeds the tacho.

daxtojeiro

743 posts

275 months

Wednesday 16th August 2006
quotequote all
the transistor I use is a 2N5555A well capable of running at that voltage, I had one on soak test for a few weeks continous running before releasing it as a working circuit. You could easily drive it from your existing output or spend a load of money on something similar,
Phil

pete500

151 posts

273 months

Wednesday 16th August 2006
quotequote all
I used the MSD Tach Adapter, part no 8920. It's one for sale at ebay now for USD46. pete

dnb

Original Poster:

3,330 posts

271 months

Thursday 17th August 2006
quotequote all
Now that I seem to have it working, I find that the needle is not damped and it waves all over the place.

Oh well, something else to fix