Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!olivea!apple!amdahl!key!jbr
From: jbr@key.COM (Jeff Rubin)
Newsgroups: comp.dcom.modems
Subject: T2500 and PP9600SA dropping V32 connection
Message-ID: <2583@key.COM>
Date: 3 May 91 19:10:45 GMT
Reply-To: jbr@key.com ()
Organization: Key Computer Laboratories, Fremont
Lines: 42

I have a Practical Peripherals 9600SA modem (Rom version 1.29) which I
use to connect to a Telebit T2500 (Rom version 6.1).  I connect using
V.32/V.42/V.42bis.  The T2500 is the originating modem (I used callback
security).  The connection works okay for a period of time and then
the connection gets broken.  The 9600SA reports various causes (S86 = 4:
physical carrier loss, or S86 = 12: normal disconnect initiated by remote
modem, or S86 = 13: remote did not respond to repeated message; after 10
attempts local modem terminated call).  Upon a suggestion from Telebit
technical support, I tried three things with the following results:
1.  Turn the speaker on and listen for retraining.

    Indeed, the connection hangs and then drops at a time when there
    are (what I guess are) retrain tones on the line.  In fact, when
    things are okay, there are no tones; when transmission stops there
    are a few bleeps, and then a long continuous tone, and then the line
    is hung up.

2.  Try MNP.

    Again, when okay there are no tones, then when the trouble occurs,
    there are a couple of beeps, then about 1/2 minute with no beeps,
    then another couple of beeps, and then hangup.

3.  Try no V42 no MNP

    This time when the trouble occurs, there is simultaneously a burst of
    garbage characters and beep tones.  This is followed by recovery in
    which there is normal transmission.

4.  With a person at each end, find out who is hanging up first

    I haven't been able to do this yet.

I should mention that the problem occurs even if I don't use V.42bis.
On a previous call to Telebit tech support, they suggested I try
ATJ6S140=1 to cause the T2500 to not attempt to retrain.  This didn't
solve the problem.  I also tried simultaneously doing AT&B0 on the 9600SA
which is supposed to disable retraining.  Still no luck.

Does anyone know what's happening and how to work around it?

-- Jeff