>stephen mcalonan wrote:
>>
>> Hello all,
>>
>> I am new to this list and am having a problem with a ceramic
resonator
>> and a pic16c74a.  I am using a 4Mhz panasonic resonator w/integral
>> caps(part# efo-s4004e5).  The resonator has an error of 0.5% and is
>> directly across osc1 and osc2.  Now on to the problem,  I noticed
that
>> my rs232 communications at 4800 baud (0.16% error) were not reliable
and
>> traced it back to the resonator.  I put a scope accross osc2 and
ground.
>> My frequency was about 7Mhz!?  Should I place a series resistor
before
>> osc2?  The pic data book mentions that one may be needed, but doesn't
>
>
>Scoping the pins isn't too useful, as the scope loads the circuit.
>Obviously there's no way you were running at 7, and only off .16% on
>routines calculated for 4.  .16% also isn't enough to worry about on
>serial.
>
>To check your osc, one way is to tune it in on a shortwave radio. This
>is nice, because it dosen't affect the circuit at all. Another would be
>to program some output pin to make an output that's directly dependent
>on the clock freq, like a timer output, or a large nop-loop and I/O
>net/clear instructions.
>
>AFAIK, you should be able to run serial at errors approaching 1%
without
>any trouble.  If your bitrate is accurate (look at AAh or 55h on a
>scope) then I'd look very hard for other problems. Is it a software
>uart? Level translators working right?
>
The stange thing is that when I use my emulator, using the external
oscillator, the RS232 works well.  I am only sending data (4800-8-N-1).
I only have occasional problems when I burn a chip and place it in my
target.  The rest of the programs works well.  I have a timer overflow
interupt in my program and an available output pin.  I will toggle the
output pin and view that on the scope.  Thanks for the suggestion.
-Steve

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com