This is a response on a query regarding use of the driver with the PMP
modem.	The I/O assignments in the driver are compatible with the modem
used with the Baycom package which interfaces through the modem control
lines of a PC serial port.  The PMP modem interfaces through the printer
port when used with the PMP S/W but can be rewired to the serial port
for use with this driver.  The BAYCOM doc is a good source for more info.

							     73 KE6HA

Subj:	AX25 driver and PMP modem
Return-Path: <JALOCHA@vsk01.ifj.edu.pl>
Date: Wed, 22 Apr 1992 11:58 GMT+1
From: JALOCHA@vsk01.ifj.edu.pl
Subject: AX25 driver and PMP modem

I just uploaded a bit "improved version" of the ax25 driver to ucsd.edu
in /hamradio/packet/tcpip/incoming directory. File name is ax25drv.lzh.

PMP modem should cooperate properly. In fact _any_ modem would do the job
if it meets some minimal requirements. Basic rules are:
1. The modem should provide decoded received data on CTS pin.
2. The driver outputs data to be transmitted on DTR pin.
   Modem should modulate and pass this signal to the radio.
3. Pin RTS controls the PTT of the radio. When it becomes positive
   the radio (together with the modem) should go into transmition mode.
4. Optionally the modem may provide carrier detect signal
   on DCD pin.

The problem with most modems is level conversion... RS232 ports
of a PC use +/- 12 V while most modems need TTl levels (0..+5V).
The simple (but not very elegant) way of doing conversion is to use
CMOS inverters (I use 4049 with my TCM3105). Like on the picture below:


                _____    |\
RS232 output --|_____|---| \o_____ TTL modem input
             50 or 100 K | /
                         |/

                _____     /|
RS232 input ---|_____|--o/ |_______ TTL modem output
               2.2 K     \ |
                          \|

Inverters are supplied from 0 and +5 Volts - same levels as the modem.

The first scheme _does_ work because CMOS inputs accept voltages outside
power supply range thanks to input protective diodes.
The resistor limits the current flowing through the diodes.
You must _not_ avoid it !

The second scheme does work as well because the threshold between
logical states in PC's RS232 is a bit _above_ 0 Volts.

	Hope it helps.

			Pawel