| ??? 12/27/07 19:54 Read: times |
#148809 - fact is ... we don't have the information ... Responding to: ???'s previous message |
Jim Deas said:
Yea, you can use mode zero but then the uart is no longer full duplex.
I kinda let my zest for creating a channel get ahead of the question. Given the limited criteria I assumed full duplex single UART. There are several variants out there with dual UARTS and several self-clocking serial encoding schemes as well. Where this started was how to drive a signal 4km. To which a protocol is in no way germane While there are many ways to encode the data into a stream the end result will be the same. It's not practical to send high speed square waves over a long twisted pair. AMI or otherwise.
Modems do it better and interface directly to a UART full duplex. I doubt anyone could manage full-duplex at 9600 bps over a 4 km line without quite a bit of dedicated hardware. What kind of UART/USRT is used is not the controlling factor. Most 805x's can manage a 9600 bps link in both directions in firmware, depending of course, on what else they may have to do. I'd say that either AMI or, perhaps easier, manchester encoding is what he should use. He can feed his NRZ to either one, and get a nominally balanced mix of 1's and 0's on his line. Moreover, he doesn't have to use synchronous methods. Even in Mode 0, the data transmission is, at best isochronous. Manchester encoding also requires only an XOR gate and a flipflop to work with a 2x clock. That 2x clock can be generated from a 1x clock with an xor and a T-flopflop, so it's not really difficult. I still don't see why we have to impose a protocol on a guy who's trying to figure out how to get signal across a 4 km line. He's probably better off considering RF rather than wire, but it's his choice. RE |
