here is how my software UARTS work, assuming 10-bit frame:

I trigger on the falling edge of the INT pin, so I get an interrupt.
This is serviced immediately  upon entry into the interrupt routine
so that the software delays before the service routine are known.
At that point, I set the bit count for 10 bits, start TIMER1, preset
the value of TIMER1 to a spot that will cause the first trigger to occur
right in the middle of the start BIT, then I turn OFF the INT pin (I only
use it to locate the beginning of the serial frame).

Every time Timer1 interrupts, I immediately capture the data  bit state,
which occurs approx in the dead-center of the  bit. I count down the
bit counter, and when all 10 bits have been captured, I make sure the
FIRST bit is LOW and the LAST bit is high, then extract the 8 bits in
between. I immediately turn off  TIMER1, turn on INT pin.

Because there is a gap from the middle of the stop bit until the next
start bit begins, a small amount of timing error can be easily handled if
a small EXTRA delay is issued. If it is NOT added, I might miss the start
of the next serial frame.

--Bob

William Chops Westfield wrote:

> On Feb 3, 2005, at 9:25 AM, Bob Axtell wrote:
>
>> from [start bit] on through the  frame until the end of the STOP bit,
>> timing is precise. The receiver UART locks on at the beginning of the
>> START bit and subjects the remaining bits to a rigid timing test until
>> the end of the  STOP bit, then the UART is released from its rigid 
>> timing.
>
>
> Presumably the "problem" is with software uarts that check the full stop
> bit, and don't have time to "unsynchronize" before the next start-bit 
> starts.
>
> BillW



-- 
Note: To protect our network,
attachments must be sent to
attach@engineer.cotse.net .
1-866-263-5745 USA/Canada
http://beam.to/azengineer

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