Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: nyu notesfiles V1.1 4/1/84; site ur-univax.UUCP Path: utzoo!watmath!clyde!bonnie!akgua!mcnc!decvax!genrad!panda!talcott!harvard!seismo!rochester!ur-univax!ning From: ning@ur-univax.UUCP Newsgroups: net.micro.mac Subject: Re: Mac-tty pinouts? Message-ID: <16500012@ur-univax.UUCP> Date: Sun, 17-Mar-85 11:12:00 EST Article-I.D.: ur-univa.16500012 Posted: Sun Mar 17 11:12:00 1985 Date-Received: Wed, 20-Mar-85 03:25:38 EST References: <2362@nsc.UUCP> Organization: University of Rochester: Computing Center Lines: 65 Nf-ID: #R:nsc:-236200:ur-univax:16500012:000:2973 Nf-From: ur-univax!ning Mar 17 11:12:00 1985 In my previous article to this newsgroup, I gave details on how to build a cable between a Mac and a modem. Fortunately, that worked with my modem; unfortunately, it doesn't necessarily work with modems that expect a "Data Terminal Ready" as jsp@unccvax said in his/her reply to my article. Jsp was correct in saying that my cable might not work with modems that expect a DTR. And mine had the option set to ignore the DTR. I didn't realize that the cable wouldn't work until I tried to attach a hard-wired Unix on the other end of the cable. The problem was that the Unix's serial ports expected a DTR and a "Request to Send" from the Mac, and the Mac doesn't send any of these signals. The solution that I found was to trick the Unix by "bouncing" the signals sent by Unix back to itself. Since the serial port on the Unix was wired as "Data Communications Equipment," it sent the signals "Data Set Ready" and "Clear to Send." In RS-232 convention, these signals are analogous to the "Data Terminal Equipment" signals of "Data Terminal Ready" and "Request to Send." The point of this is that I could "trick" the Unix into thinking that my Mac was ready to accept data by sending the DSR back into the DTR, and the CTS back into the RTS. So whenever the Unix was ready to send data, it would find that the Mac was ready to accept; and whenever it was ready to accept data, it would find that the Mac was ready to send. This usually works, but sometimes this faked handshaking might pose a problem. (In my case, both the Mac and Unix were ready to accept data anyway.) But without another kludge, the Mac can't make use of these signals even if the Unix depended upon them: Some mainframes that don't use X-ON and X-OFF (DC1 & DC3 by other names) might use DSR and CTS to tell the Mac that it is not ready to accept data. The kludge is to wire one of these signals (depending on how your mainframe's serial ports toggle these signals) into the Mac's Carrier Detect, and with software the uses the Carrier Detect, you get one-way handshaking. So the forged handshaked, which is what I use for both the hard-wired serial ports and my modem looks like this: Macintosh Modem RS-232 Signal Name (9-pin DTE) (25-pin DCE) Pin # 3 <---> 7 Signal Ground (GND) 5 ----> 2 Transmit Data (TxD) 9 <---- 3 Receive Data (RxD) 7 <---- 8 Data Carrier Detect (DCD) +--- 6 Data Set Ready (DSR) +--> 20 Data Terminal Ready (DTR) +--- 5 Clear to Send (CTS) +--> 4 Request to Send (RTS) The one way handshake (which I haven't tried, but in theory should work) is like the forged handshake with this exception: 6 DSR 7 HSC <---< or 5 CTS Now that you see the big mess of RS-232, you can see why people prefer to use X-ON and X-OFF--that is, until they have to enable and disable it in order to use XMODEM with macput and macget. -- Happy happy Telecommunicating, John Lam_____________________ {allegra,seismo,inhp4}!rochester!ur-univax