Reading automotive signals

Dave Paton

Specifically, I'm interested in reading some of the engine signals directly for the purpose of some advanced datalogging. Injector duty, RPM (tach sense) and possibly distributor signals (ignition advance, dwell), along with more pedestrian things like pressures and temperatures.

I'm under the impression that I can safely piggyback from the injector and distributor signals with a Schmitt-triggered optoisolator. I'm not sure, however, how to read the normal sensors (analog inputs). The option of using secondary sensors seperate from the main ECU sensors just isn't an option, given space and mounting limitations.

Thoughts?

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
This is not a sig. This is a duck. Quack.

parsko

Dave,

It is my understanding that one should be able to piggyback on other signals as long as the piggyback is of high resistance, so it (effectively) does not interfere with the original signal.

For the analog signal that change resistance (RTD's), it should be a matter of developing a divider circuit placed on the input of an op-amp, which then drives an ADC.

For analog signals that (don't) change resistance (aka provide a voltage), you could pump it into an op-amp in the "follower" format, then into an ADC, or simply pump it into an ADC input, but an inductor may be needed (see below).



The biggest concern is noise. Opto-isolate/schmitt trigger anything digital, if possible. Inductor between an analog input and the actual input (or in series if using an RTD) should work. Also, filter caps on any inputs, as close to the pin as possible.

Keep in mind, this is "IMHO", as I am no EE, and have not yet developed a working circuit in an automotive environment, but it is my primary goal with microcontrollers. I'm just researching and experimenting right now. Experts need to verify!!!

-Parsko