Hi everyone!

I'm working on a RPM-Logger (tacho) project. It works fine on my desk (using
a computer with small programm to generate the ignition signal) but in my
car it runs lousy.

First some words about the project:

A PIC 16F84 @ 8MHz.
The Timer0 is running to messure the time between two ignition signals
(TMRDiv 1:1).
The Ignition Signal provided by the ignition-system of the car is 12V with
falling edge to 0V when it fires. (2 times per revolution) is connected via
a resistor an a zener-diode to RB7 of the pic. I use the RBIE (RB Port
Change) Interrupt to start and stop (read out) the timer and the addtional
counters.
On RA3 a PC will be connected to read out the processed and calculated
signal from the pic. ("just a wire" - serial communication)
On a other RA a circut is connected to switch a bigger relay.

Everything works fine on the desk at home. Good readings (RBIF) and the
transmission to the computer (with Data-Logger SW) works like i want.
The relay will be switched at the right RPM.

In my car (it's a honda) it seems to work fine until the RPM goes up to 3000
RPM.
Will the RPM raise about 3000 RPM the result jumps 200 RPM up and down.
The relay will be switch around 200 RPM below the desired position. (see a
line befor ...)
But the serial transmission works still fine!

I've added a routine which will toggle a port when the RBIF occure and use
my scope to see what happens. It seems like the interrupt will be triggered
at the right time. But the difference must be only some 100 uS. I've no
possibility to check this out with my old scope.

The supply looks fine. (100 uF, 7805, 10 uF, 100 nF) I can't see any drops
or noise.
The input signal looks fine. (i've also added a 100 nF to ground. nothing
happens)

Now i don't really know what to do?

I've also tried a other method to read out the signal (thank you Tony Nixon)
but the result is the same and this method is to slow for my car. (It can
rev up to 10.000 RPM)
I think it can be a problem of the noisy enviroment in the car.

Has anyone an idea to solve this problem?

My next step is to try the LM 2940 Voltage regulator ...

TIA

Christian Dorner
aka, Doc ...

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
"[PIC]:" PIC only "[EE]:" engineering "[OT]:" off topic "[AD]:" ad's