Thanks for the replies please find my answers below.

On a philosophical point, how do you know that the controller is wrong,and not the receiver ?

I was wondering that, the receiver works OK on everything else, so I would "assume" that it is OK.




Are you aware that you need to delay the turnaround (enabling the 176 for xmit) by 1/2 bit time after the interrupt?. This is beacuse the interrupt happens in the middle of the stop bit. Your problem 'smells' turnaround problem

Yes I was aware of the requirement for a delay (thanks to you sometime ago) at the end of the message before switching the '176 back to Rx Mode, I have some nominal delay in there of about 2 odd mSec which seems to work fine. I also thought that it was a turn around problem but it is occuring at the start of the message, rather than the end, indicating to me that the '176 (485 chip) is switching to TX too slowly, however even with a delay in between the TX control and actually sending the message has no effect (actually for some other reason I get a whole heap of nulls sent - maybe this is a clue?).





//before sending first byte of the string

DELAY(for time to send 1 byte);
TI=1;

Yep I have tried this, with delays inserted all over the place but all to no avail, the framing errors on the first byte still occur.


I was wondering if it could be a noise thing? anyway I will continue to test and let you know the outcome.


Regards
Marshall