Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/5/84; site avalon.UUCP Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!mtuxo!pegasus!phoenix!avalon!merlyn From: merlyn@avalon.UUCP (Steve Humphrey) Newsgroups: net.micro.att Subject: Re: Hayes modem and the 3b2 Message-ID: <103@avalon.UUCP> Date: Thu, 30-May-85 11:44:20 EDT Article-I.D.: avalon.103 Posted: Thu May 30 11:44:20 1985 Date-Received: Fri, 31-May-85 04:15:18 EDT References: <101@gwsd.UUCP> Organization: AT&T-Information Systems, Lincroft, NJ Lines: 52 > There is a problem using the Hayes modems as an ACU. > > To originate calls (outgoing) you must disable the carrier detect (switch 6). > To receive calls (incoming) you must enable carrier detect. > > Has anyone figured out a way to allow you to use a Hayes modem on a uucp > bi-directional line? > > Bob No, it is not possible with the standard software to use the Hayes modem bidirectional. Reason: The cu/uucp software accesses a modem in a way that requires DCD to be up. The 3B2 is capable of allowing access with DCD down, and I've modified the cu/uucp software to do that. Even running that modified software, though, I still had minor problems with the modem, thus I don't use it any more. (I never identified the problems, so I don't know who to blame.) I sent my software modification to DAN (as in Honey-DanBer UUCP) so it may appear in a later release. If you find it expensive to (1) toss out a pile of Hayes modems and buy other modems (e.g. Ventel, Penril, AT&T 2224C all work fine), (2) use two ports on the 3B2 (one incoming, one outgoing), or (3) buy the source so you can do the software fix, then consider a hardware work-around: The Hayes keeps the CTS (Clear To Send) signal up while connected to the 3B2. The work-around is to build a small circuit that uses CTS and DCD from the modem to build a pseudo-DCD signal for the 3B2. The trick is to keep the pseudo-DCD signal up while the CTS signal is up, regardless of the state of the DCD signal, EXCEPT when the DCD goes from high to low (i.e. when a connection to a remote machine is dropped). At that transition, the circuit drops the pseudo-DCD for a short interval (about one second should be enough) then raises it again. This tells the cu/uucp software that the connection dropped, so the software quits; raising the pseudo-DCD back up again lets another outgoing call be placed. What does the circuit look like? I wish I knew, sorry. We were starting to design the circuit, since we have a closet full of Hayes modems that we'd like to use, but other work took precedence. Interested parties can drop me a note and I'll give you what info I have, but be warned: I'm not a hardware jock so I don't know much more than I've described above. Steve Humphrey ...!avalon!merlyn AT&T -- Steve Humphrey hogpc!merlyn or lz3f310!merlyn