Harold Hallikainen hallikainen.org> writes: > It may also come from mechanical teleprinters which idle in the mark > condition (from Morse's original recording telegraph where loop current > caused the tape to be marked). Teleprinters had a loop current of 60mA > (series pulling magnets) or 20mA (series holding magnets) between > characters. This was probably carried over to UARTs where loop current was > represented by a high. The RS232 drivers and receivers PROBABLY invert > because it takes fewer transistors to make an inverter compared to a > non-inverting buffer. I know for a fact that having - voltage on the lines was a corrosion-preventing feature. Most POTS lines still carry the -52V CO voltage when idle for the same reason, whether buried or air lines. It turns out that the wire set to negative potential wrt. ground will only supply electrons, and not ions, so it stays in one piece when wet, instead of disappearing into ionic solutions of rainwater and dirt. The same trick was used in telegraphy times, when it was discovered, and it appears to have been grandfathered down to current RS232 systems, be them POTS, telegraph, telex or computer data. The relatively high line current is a requirement that also dates back to air lines which were typically assembled by twisting (Westinghouse tie etc). The copper/copper connections quickly became copper oxide/copper oxide connections, with lousy transmission and bad noise in phone lines. The high (>50mA) current and the high enough voltage (50V) would break down the oxide and provide wetting current to the twisted connections, when the line started passing current. To this day, rural air lines are 'fixed' by putting a capacitor discharge current through them if noisy. If that does not help, then it's time to break out the tall ladders and the service truck. It also seems to help with noisy phone lines in older buildings, and I have used it at least twice to make faxes and modems work on lines that wouldn't. That saved a lot of time in spelunking in ancient uncharted connection boxes. The 'why is it inverted' question probably goes back to the first line repeaters, which used one valve per circuit each and were inverting when the input and output audio transformers were bypassed. I assume that the people who made the first RS232 interface systems faced the problem of interfacing their low voltage logic to the lines, and used the simplest available thing, a standard line repeater. Peter PS: I am not that old but I often work with old equipment and I had to pick up the skills. I am not sure but I think that the bog overland DC HV lines are also set to a negative DC potential wrt ground for corrosion and corona corrosion (erosion) reasons. The Mark and Space states seem to go back to the Morse tape inkers, where Mark meant pen down and Space the other state. I do not know how this relates to the inversion issue. It is possible that Mark (which is -Vcc on the line), was selected as above, i.e. Morse key down, with -Battery on line, and Space was key up, with nothing on line. Later the key up state became +Battery on line. Later RS232 came around and inherited these states, while logic circuits usually used a '1' for something 'On', or 'High'. Matching the two, the line and the logic state, would require an inverting interface. I saw a schematic of an old American railroad telegraph once on the net, and I think that it also had -Battery on the line, but it was the series connected key system (party line). Engineers must have discovered early that wires connected to +Battery out in the weather would disappear or corrode away. -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist