Simulation RPM with Function Generator

[ad]

Hi, I am making a device to count the revolutions per minute of the engine. To perform this function I had thought as follows:

- With the CCP1 module in Capture mode PIC18F4550 generate a TTL signal with the function generator, then generate an interrupt processing time and eventually accumulate RPM. When the frequency is varied to increase the speed.

-Another way I had thought, is to make a frequency meter (square wave signal give me the function generator), and through the relationship W = 2 * pi * f, I knew revolutions.
I wonder what you think, if I'm on track.

[Ttelmah]

It depends totally on the nature of your RPM signal.....

'Old' systems commonly used the ignition pulse from one cylinder to measure the engine RPM.

Modern ones in cars electronic fuel injection, typically have sources giving multiple pulses round the revolution. Often down to only a few degree steps.

In all cases you have two choices:
1) Measure interval between pulses. Then RPM is proportional to 1/interval (the proportion depends on how many pulses per rev).
2) Count pulses over interval. Then RPM is directly proportional to count (again proportion depends on the number of pulses).

Downside to the latter is slow update. If you have ever driven a car with a Jaeger chronometric speedometer, you will have an idea of what this is like.
Downside of the former is it requires careful damping and equal timings round the rev, if results are not going to be erratic.

Yes, you can 'simulate' with a frequency source, but there is little point in this. It won't prove the system is working on the real engine....
Real signals will have noise, fluctuations in the pulse width, etc. etc..