jlapsley%D.CC%berkeley@sri-unix.UUCP (12/16/83)
From: The tty of Phil Lapsley <jlapsley%D.CC@berkeley> Mike, et al. Yes, just this afternoon I got a call from a friend, and we talked for a while, and then I hung up to call him back later. When I hung up, the U.S.R. S-100 jumped into the circuit and sent an answer tone (it then decided that the dial tone was another modem and tried to talk to it for a while, without much success). I have found that the same thing happens if I switch-hook or click on my home made hold button. I haven't noticed it happening when I *call* somebody, but I haven't had it for very long. I'd be willing to bet that it will, however. If I recall, when one calls a long distance party, there is a brief reversal of polarity (this is the same reason why the touch-tone pads on some pay phones don't work when one places a long distance call). This may make the modem think that there is an incoming call. It seems to me that there are about three solutions. First, you could set up your BIOS to tell the modem not to auto-answer. I don't think too much of this idea (my terminal program, though, does do this, and when I switch-hook or whatever, I get a RING result code back). Second, you could put in a switch between the phone line and the modem, so that you could switch the modem out of the circuit when you don't want it there (which is what I am getting close to doing). Third, if you are a hardware hacker, you could do as I have done and write to U.S. Robotics and ask for a schematic. It must be that they are using really brain- damaged ring-detection circuitry, and some modifications might be in order here. Whatever option you choose, it strikes me as if none of them should be necessary. I would suggest writing to U.S. Robotics (1123 West Washington Blvd., Chicago, IL, 60607) and complaining about this. Perhaps if enough owners do, they will consider doing a bit of re-work on some of their products. Good luck, and let me know what happens. Phil (jlapsley%D.CC@Berkeley)