Mike B wrote:
> FYI
>
> In follow-up work on my clock board using the PIC internal oscillator, I varied the supply to my on-board regulator, and found the internal oscillator varies significantly. I did not think of doing this before, but I had some PS problems and with my counters set accurately to a 5V supply I varied the source from ~3V-15V and found the timing off by 50% to ~500%.
>   

You got a PIC to work at 15V?
> Not a huge suprise, but just some insight to consider if anyone else has timing issues. The internal RC appears more sensitive to supply than temperature. And since various designs affect the supply, you will also be changing your timing significantly.
>
>   
Yes, I got similar results. It is NOT accurate enough for an RS232 
serial port application; Microchip's adverts are not correct.
Not only that, but the speed of the RC "wobbles", i.e. faster then 
slower, noticeable when fast events are being executed., but not
so noticeable at slow speeds.  For example,  I was able to get  RS232 at 
600baud  running (protocol N/8/2) but not at 9600baud.
1200 baud was usable and reliable enough with manual waits (of 2-3 bit 
times) between each character.

I finally gave up and used my favorite Manchester code, which is immune 
to error rates of 20% or less. Slow, of course.

--Bob  
>       ____________________________________________________________________________________
> Be a better pen pal. 
> Text or chat with friends inside Yahoo! Mail. See how.  http://overview.mail.yahoo.com/
>   

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist