Measure Fuel usage using DVM

[cwa]

I have connected one of the Injector Signal (& GND) to DVM (Digital Volt Meter) and here are the results.

Park = .47 V - .51V
Neutral (idel) = .48 V - .52V
D (With Break) = .49V - .54V

Neutral (2500rpm) = .95V - 1.00 V

I have not tested while moving as DVM cables are short and it was dificult to have a proper connection.

Any similar experience or meter readings?

Also when I put the DVM to check the ACV, I got zero reading.

CWA

[skewbe]

you should try something like a dwell meter

[bearleener]

(I have a 1996 VW Golf with throttle-body fuel injection, i.e. only one injector.)
I think I had similar values when I set my DVM to DC volts. In my car the injector line has on-board voltage (i.e. about 13.5 V) most of the time but gets pulled down to ground periodically to trigger the injector's solenoid. These pulses come with a frequency of twice the engine RPM (4 cylinders, each firing every other revolution) and a duration related to demand.

When you measure with a DC voltmeter this "dirty rectangular wave, pulse-width moduluated" signal gets averaged, so it's a decent measure of what proportion of the time (duty cycle) the injector is on. Keep in mind, though, that when the on-board voltage changes (e.g. when you turn the headlights on) this reading will change. You could probably eliminate this problem by measuring the voltage across a diode (connected in series with, say, a 50 kOhm resistor to limit the diode current).

Many newer or more expensive DVMs can measure the frequency (Hz) and duty cycle (%) of a signal. I bought a relatively cheap one of these (can't afford an oscilloscope) and can now directly measure the injector's duty cycle (and therewith the fuel rate).

I can't remember the exact duty cycle figures but it was something like this:
12% at idle a few seconds after a "cold" start (at 20 C or 68 F)
4.7% at idle when engine warm
10+% maintaining constant speed (60 km/h)
45% during full-throttle acceleration.
One thing became clear: the higher the gear during cruising, the shorter the duty cycle.

So you can indirectly measure momentary fuel consumption (in terms of a rate, so many milliliters per hour). To convert this to mpg or l / 100 km you'd need the vehicle speed signal and (with a microprocessor, usually) divide the fuel rate by speed. That's what ScanGauge and the others do.

It would be nice to have an average fuel consumption, too. Since the ScanGauge doesn't work in my car I intend to do this with my own microcontroller, which is somewhat complicated.
I don't know much about electronics, but you may be able to do an analog average fuel rate simply by smoothing the DC voltage measured at the injector with a big capacitor.

[landspeed]

Bearleener - I came to this site while trying to figure out how to get instant MPG too!

Take a look at the 'SuperMID' (type in SuperMID on the search function). It is made by Yoshi on here, and is a ready-made injector/speed sensor reading device that works on almost any injection car (including my 87/89 Nissan Bluebirds, and 97 Micra)! It would probably save a lot of time from making your own from scratch!

[cwa]

Thank you very much. I have already place an order with Yoshi and awating for Supermid.

Reg.

CWA

Originally Posted by landspeed

Bearleener - I came to this site while trying to figure out how to get instant MPG too!

Take a look at the 'SuperMID' (type in SuperMID on the search function). It is made by Yoshi on here, and is a ready-made injector/speed sensor reading device that works on almost any injection car (including my 87/89 Nissan Bluebirds, and 97 Micra)! It would probably save a lot of time from making your own from scratch!