Pretty much everything in a PIC has a requirement for Fosc timing. That
is, it usually requires 1/4 of your instruction time. For a 4 MHz xtal,
you need a pulse width of 250nS.
Andy
At 09:36 PM 7/23/98 +1200, you wrote:
{Quote hidden}>I am using PORT B "interrupt on change" on a 16F84.
>With the signals I am using it works well, but I can find no data sheet
>information on several parameters which could be critical.
>
>My query - is there a minimum hold time for the "difference" signal which
>triggers the interupt.
>The data sheet shows the comparison circuitry but does not indicate whether
>RBIF is set asynchronously and essentially instantaneously or whether it
>has some synchronous timing constraints (perhaps clock frequency or cycle
>frequency). Is there a minimum time constraint and is it based on clock
>timing?
>
>While I could carry out measurements (which may or may not be conclusive)
>there are no doubt PIClisters who have experience here, So -
>
> - How long a "pulse" do you need to trigger this interupt?
> - Is the timing asynchronous relative to the clock.
> - Is there a known formal specification?
>
>
==================================================================
Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA
==================================================================