Heartbreaking Bug Revisited

Got time to look at the problem in much more detail. Last night I was just too upset to start work so close to bed time. Today I discovered that the baud rate was indeed correct. What i was measuring yesterday was two bytes back to back without the start and stop bits. And hence the baud rate that was about 50% of my  9600.But why does my Linux box see two bytes back to back that are the same when I have logic to prevent that? Well I switched over to reading the data from my laptop and the problem goes away. On the Linux box I get the problem when the first byte is alternating bits (0x5 or 0xa). Now I don’t have root cause, but I can only figure that I have a noise issue. Tomorrow I plan to remove the 30 foot kludged cable and co locate the detector and computer and see if the problem goes away. In the mean time, I need to look at RS-422 or wireless solutions.