I've been doing some research on how to read RPM from the spark plug wires; and for Coil on plug cars, from the coil pack wires.
I've come to realize that this product is a current sensor that works off induction. Which is exactly what I want. The thing though is that the datasheet is showing maximum voltages much lower than what a coil wire would carry (30KV vs 2.5KV in the Datasheet ).
Is there anything else i can use? Am i reading the datasheet wrong? I'm making a device to read engine RPM.
There is a number on there 2.5KV/1A/1min but look to the left of it and tell me what that number represents. Is it anything about the maximum voltage in the wire?
For coil on plug it will be a "low voltage" input to the cap. You could tap into the coil on plug trigger input directly as long as you clamp its input to the Arduino to 5v or less.
However they can be expensive, not because of how they are made but because they have the word "automotive" associated with it.
You can make one with a toriodal ring that will fit over the lead and using some 240V rated wire wrap 4 or 5 turns around the toroid.
This forms an inductive pick up with the same sensitivity as the clamp.
The insulation on the HV lead and the 240V you use will be enough, if you keep the coil away from the ends of the lead.
The clamp shown above only uses enamelled magnet wire and U and I ferrite cores.
Then use a circuit similar to here;
The lead from the pickup to the rest of the circuit needs to be twin core with a shield.
Thanks for that circuit. I could make this as I have a timing light, i could use the part.
About what voltage is coming in at L1? would this be a multiple of the spark wire 30,000V?
Or does the clamp itself reduces the voltage before the circuit above?
The only thing Is I plan to sell this equipment (it's going into an existing software product) so I was looking for something less expensive.
Are you trying to read the voltage from the spark plug wires? If all you want is the led strip to react to your engine rpm, it might work. But the voltage is not necessarily something that truly tells the exact rpm, independent of say the low voltage on the ignition coil. You might get different readings for same say 3000 rpm, depending on how much load there is on the alternator (head lights, outdoor temperature, battery charge).
Or is it so that the circuit in post #6 manages to turn the frequency of peaks at L1 into a current at the base at Q1, independent of the average voltage at L1?
Johan_Ha:
Or is it so that the circuit in post #6 manages to turn the frequency of peaks at L1 into a current at the base at Q1, independent of the average voltage at L1?
Yes, to measure RPM accurately you need to measure the ignition pulses or a signal off a shaft sensor or an encoder.
The output of the transistor will be a pulse produced by a spark current in the ignition coil.
The are many Arduino Tacho sketches, just google.
There used to be a tacho system for diesel engines that used the AC voltage off one of the alternator 3phase windings.
This was fine for low speed engines like diesels, but these days with electronic direct injection, its not used.
Tom...
You really might want to consider getting the rpm off the crankshaft sensor. Much less hardware and wiring involved, all you really need is one optocoupler, one resistor (probably in the 1K-5K range depending on your car) and a couple of wires.
rpm can then be calculated by counting the number of pulses from the sensor using the Arduino's timers.
TomGeorge:
The output of the transistor will be a pulse produced by a spark current in the ignition coil.
What I was after (I'm no good at figuring out circuits), was a circuit that turns the "analog" signal in the spark plug cable into a "logic level" pulse. The first stage would be a pulse with a constant length, shorter than the fastest pulse rate from the spark plug cable. So no matter what the actual voltage is in the spark plug cable, the pulse has always the same voltage and length, probably created with some opamp circuit. In the second stage some capacitor-resistor-inductor? circuit would hold the short time average voltage of the pulse signal in the first stage. This would in theory measure the actual RPM from the spark plug cable, not only the voltage. Maybe it's overkill for this purpose, but I believe the voltage varies a lot depending on the conditions I mentioned earlier. Add to that the condition of the spark plug.
You're really making your life a hundred times easier if you just tap into the crankshaft sensor. This will eliminate the need for almost all of the hardware that has been discussed in this thread.
And the crankshaft sensor already offers you a ready to use logic level pulse. All you have to do is find out how many teeth there are on your car's crankshaft flywheel, which induces the logic level pulses in the crankshaft sensor. Then divide your measured pulses per time interval by that number (plus one, for a complete revolution) and extrapolate to one minute.
Beware! some electronic ignition systems use multiple sparks. My old International Scout befuddled the emissions testers trying to read the RPM by inductive coupling. Idle showed thousands of RPM.
I once built a strobe unit for timing light on miniature engines I build here.
It was from a Gompy site and used a coil pickup.
You may be able to modify to suit your needs as it fed a 555 timer.
See if i can get the files to upload.
Carguy : And the crankshaft sensor already offers you a ready to use logic level pulse
Not all of them - some are simply a coil wound round a magnet and generate successive + and - going pulses of a volt or so, and so still need a bit of signal conditioning.
carguy:
You're really making your life a hundred times easier if you just tap into the crankshaft sensor. This will eliminate the need for almost all of the hardware that has been discussed in this thread.
And the crankshaft sensor already offers you a ready to use logic level pulse. All you have to do is find out how many teeth there are on your car's crankshaft flywheel, which induces the logic level pulses in the crankshaft sensor. Then divide your measured pulses per time interval by that number (plus one, for a complete revolution) and extrapolate to one minute.
...and Robert will become a close relative.
No, my friends, I'm not measuring voltage. I'm measuring pulses over time.
I cannot tap into the CPS. I already even have a circuit that taps into the distributor 12 signal wire that they use for tachometers, or tap into the coilpack signal wire. But since the industry is used to clamping a wire over the HT spark plug wires in 5 seconds and go, I'd love to use that method instead of tapping anything.
Also the CPS is the most sensitive sensor any little interference can throw the ECM off.
TomGeorge, I ran your circuit through a simulator. It seems to be frying the parts. But a circuit like that is something i need.
