On Thu, Oct 18, 2001 at 07:35:04PM +0100, Tony Duell wrote: No, the square brackets always look like square brackets. It's the underscore that may be a left-pointing arrow. The circumflex may be an upward-pointing arrow. In a _really_ old version of ASCII, escape (a.k.a ALT MODE) was in a different place too (125 or 126 instead of 27). I think that version was defined without lowercase, but Stanford University used lowercase and never bothered to move escape, so they had to shuffle the right brace and vertical bar. For good measure they even defined printed characters "on top of" the control characters. Aren't character sets fun? At least ASCII doesn't generally use shift characters, like the Murray code or the FIODEC code I'm currently playing with. Thanks for the explanation of current loops. Could you call RS-232 a current loop too, or is there something else that remains constant, or not? Now that I think about it, I believe the US and UK phone systems are incompatible for a similar reason. -- Derek

Derek Peschel wrote: Ok, practical, how-to info here... Assuming your using a 12531-series interface in your HP... No current loop to RS-232 adapter is needed, the 12531 is able to generate and accept RS-232 signals with minor cable modifications as follows: Jumper pins 16 and T to pin V. This brings the current loop transmit data into the RS-232 converter. RS-232 transmit data is not available at pin W. Jumper pin Y to pins 4 and D. This brings receive data from the RS-232 converter into the current loop data input. RS-232 data is now applied to pin X. (These modifications are made inside the hood that attaches to the TTY board) Now then, data format issues...this gets just a little screwy here.... Older HP software will want to get seven data bits with 'space' parity when a program is running, but when reading a paper tape the data format switches over to eight data bits and no parity bit. This is due to HP software, and not the interface hardware. Contrary to some advice on the list, Hyperterminal will work just fine, and it does support ASCII file transfers. The problem is that HP software treats reading a paper tape differently than when reading keyboard input. When reading a paper tape, the 12531 interface sends a READ signal by grounding pins 13 and P through a NPN transistor (Q5). There is no 'easy' way to turn this signal into an RS-232 compatible signal for automatic format switching without adding extra hardware. Lastly, the 12531 A/B interfaces can run up to 2400 baud if given an external clock signal as a TTL signal on pin L on the interface connector. Jumper W1A must be moved to the W1B position for this to work. The 12531 C interface can run up to 9600 baud using this trick. I'm glad this question came up, I need to ship a tested RS-232 cable and 12531 board to a friend on the list...