Dwayne Reid wrote:
> Using the prescaler with TMR0 makes a 16 bit counter which is essentially
> independent of the PIC core.
>
> Using the PIC core to generate the gate signal can be very deterministic.
> If you make a loop that consumes 100,000 clock cycles, you will create a 100
> mS gate (assuming 4.00 MHz crystal) with accuracy determined by the crystal
> used.
>
> In other words, this can function exactly like my old Fluke 1900A does.

The application assumes a calibration for each equipment, generating a
conversion
factor from the reading pulses to what it should be in real, so if the
PIC clock
oscillator frequency doesn't change so much with temperature (ppm), it
will be
ok. The main idea is to gate "x" time using the prescaler, read it as
the high
8 bits, and then gate "x/256" without prescaler, so it would be the
lower 8 bits.
For an external frequency of 20MHz, "x/256" gate needs to be a max of
12.8us,
while "x" would be 3.27ms.  Assuming a 20MHz PIC clock osc, 12.8us
represents
256 instruction cycles, plenty to control the timming, I guess, Is that
right?
--------------------------------------------------------
Wagner Lipnharski - UST Research Inc. - Orlando, Florida
Forum and microcontroller web site:   http:/www.ustr.net
Microcontrollers Survey:  http://www.ustr.net/tellme.htm