[ale] Monitoring the serial port

[John Mills]

James, all -
On 4 Sep 2002, James P. Kinney III wrote:

> The ingenuity of this group never ceases to amaze me!  Spiffy idea. Of
> the top of your head, do you know what the load limits _should_ be? 

Well, the 'tap' approach came from the Win-tool folks. My only
contribution was using it with 2 at Minicom windows.

I don't have the RS-232C standard anywhere handy, but my fogging neurons
tell me that c. 100 ft. was reliable with generic twisted-pair wiring and
one receiver per transmitter. I think the signal levels are MARK := -9>=
-12v and SPACE := +9 <= +12v. Most receivers seem to have a safety margin,
and here we were driving only a few feet of wire at moderate speeds (39.2
Kb/s IIRC). The drivers are supposed to deal with reactance of that 100'
length at higher rates, so we were fine.

Naturally you could use a pair of RS-232C -> TTL receivers and
transmitters and have enough bits left over (from things like modem
control) to drive your tap lines. This requires a source of + and - power
(MAXIM chip and +5v. supply?) and a bit of wiring. I'm too lazy.

> > I've used 'minicom' for this - in fact I had a pair of them running to
> > capture traffic both ways on the RS-232C line between a Win box and a
> > radio, roughly thus:
> >                                    'Breakout'
> >                                     connector
> >             TxD                     ........              RxD
> >          >>>--------------------->>------+--->>--------------->>>
> >   WinBox    RxD                     .    | .              TxD    Radio
> >          <<<---------------------<<----+-|---<<---------------<<<
> >                                     ...|.|.. 
> >                                        V V
> >                                        V V
> >   LinBox          RxD                  | |
> >    Minicom(1) <<<----------------------+ |
> >                   RxD                    |
> >    Minicom(2) <<<------------------------+
> > 
> > Salient points:
> >   1) A special cable was built (actually a M-F in-line breakout connector)
> >      which brought out the two signal lines and routed each of them to
> >      the RxD of a separate serial port on the Linux box.
> > 
> >   2) This clearly violates the usual load spec on RS-232C drivers, but
> >      with short cables it was no problem.
> > 
> >   3) The two Minicoms were each started with different "capture file"
> >      ('-C <filename>' option)
> > 
> >   4) The TxD ports of those two LinBox serial ports were not connected,
> >      so by turning on 'echo' at the Minicoms, I could enter notes into
> >      the log files by simply typing into the respective Minicom window.
> >      That also allowed me to establish their sequencing for later
> >      comparison of 'inquiry/command' and 'reply'.

 - John Mills


---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.