Detecting Injector pulse without cutting wire?

Hey guys!

I'm attmepting to make a MPGuino like project of my own but design it a little differently.

In order to calculate the MPG of the car I need to measure the injector pulses. I want to do this without having to cut or damage the wire, and I was wondering if it is possible to detect the signal travelling through the wire without phsycially touching the copper together, but using something like an EMF detector.

I'm not a complete novice with electronics... but I am fairly limited with my knowledge. Programming on the other hand is fine.

Any ideas please? Thanks in advance!

You can try using a hall-effect (magnetic flux) sensor. Current through a wire generates a magnetic field. By measuring the magnetic field you should be able to determine if current is flowing.

Do injectors always squirt the same amount of fuel?

I assume they do or this approach won't work, therefore you can also get the RPM from any other source and derive the fuel consumption from that. Or is there only a single injector into a manifold?


I'll have to look into the Hall-Effect option, thanks :).

And yes, injectors run at a specific pressure which the fuel pump maintains, and injects fuel at a steady rate. The amount of fuel the engine receives depends upon how long the injector is pulsed for (the pulse width) basically.

I can't use the RPM method unfortunately as the fuel flow is determined by throttle position, RPM is directly linked to the road speed and the current gear (e.g. 3000RPM in one gear will be different fuel usage to 3000RPM in another gear, just as 3000RPM whilst accelerating will be more than 3000RPM while cruising).

As far as I can tell, on my engine (1.8l petrol Rover K-Series) there are four injectors, each one operating at 3.5bar pressure. The ecu sends a pulsed signal to them in order to allow fuel through. The pulse width and speed changes dependant on how much fuel is required.

What I want to do, is measure how many microseconds the injector is open in a given period of time. Multiply this by how much fuel the injector should pump per microsecond, and I know how much fuel has gone through! Very simple in theory :)

All I want to know is whether I can do this without actually cutting and joining the injector signal wire with the one going to the arduino. I want to be able to simply take it to the car, clip the signal wire on to the injector, clip the earth wire onto body and away I go.

I hope this explains what I'm doing a little better ;) thanks for the input guys.

If this is a “regular wire”, not a ribbon cable, or coax cable, etc., I’ll suggest a “compromise”. [u]This type[/u] of connector will slice though the insulation and make contact without cutting the actual wire. They are available in two different sizes to fit most common wire-sizes. You can buy then online (obviously) and you might also find them at an auto-parts store.

Yep that's clear now, do you know the typical pulse length?

I've only used HE sensors for simple stuff like a speedo sensor on a tail shaft, that's pretty slow and of course I don't care about the pulse length. I don't know if they would be good for this but I assume they would be. You would have to look at response times (they may not be fast enough) and also if the rising and falling edges are symmetrical.

For example the ATSD682 is good for 8kHz which should be fast enough but it may not be able to detect a 3uS pulse.