I think that the asynchronous transmission bit format is distinct from the signalling specification (i.e., voltage levels). I looked up the two ICs which Barry mentioned and both of them appear to output TTL levels on their serial out lines. These ICs still require a level converter IC to be used with RS232. So, for whatever reason, RS232 was developed with a low voltage indicating the active (mark) state and a high voltage indicating an inactive (space) state. Also, TTL was developed with >2.4V indicating a high and <0.8V indicating a low (if I recall correctly), and then the logic could either be active high or active low. Many TTL UART ICs were developed using active high logic since active high logic is more common. These were sometimes used (with a level converter) to do RS232 comms, but that is not their only possible application. Sometimes you might have asynchronous serial communications over a radio link, through optocouplers, within the same device, etc, and in those cases TTL levels may have been used directly to feed the destination device. When it came to level conversion for RS232, the level converter converted the active TTL state (usually a TTL "high") to the RS232 active state (a "low" voltage). This isn't really inversion since the logical sense is not being inverted (active is mapped to active and inactive to inactive). It just so happens that those two standards developed with opposite relationships between high/low and active/inactive. Sean On Sat, Nov 14, 2009 at 3:22 PM, Barry Gershenfeld wro= te: > Well, the first UART I was associated with would be the mechanical stuff = in > the teleprinters. =A0Later, after they "invented computers", I build myse= lf a > "serial to parallel converter" (from TTL) to make my TV Typewriter work w= ith > a modem. =A0 Subsequently, I was at a hamfest (swapmeet) and a guy had so= me > 40-pin parts which he described as apparently the same thing. =A0My eyes = grew > wide and I purchased one of them, took it home, replaced my board with > that...It was a TMS6011 if I recall correctly, later discovering the more > "standard" AY-5-1013 from General Instruments. > So, these seem to have been the "first" UARTs. =A0And it never seemed str= ange > to me that 0V was considered the "active" state. =A0Not just because "mar= k" > was "current" and all that, but I had gotten used to CS# being low active, > MCLR# being low active, and so on. =A0It just seemed normal at the time. > Keep in mind that the TTL interface was never intended to relate > electrically to the +/- 12V one. =A0That was the converter's job, period. > =A0This apparent "UART over TTL" idea really was just the result of some = folks > who saw a conversion shortcut they could make using a diode and a resisto= r. > Barry > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > -- = http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist