Re: [SI-LIST] : Input switching threshold & CPCI

About this list Date view Thread view Subject view Author view

From: Adrian Shiner (adrian.shiner@virgin.net)
Date: Mon Jan 03 2000 - 13:06:46 PST


Nice one D.C.

Many decades ago when I was training, I was one half of a two man team which
designed a passenger entertainment and services ordering system for aircraft
(this was a 6 month training project). Signals were 0 to 10 volts nominal
but referenced to the supply. Lower half of signal was multiplexed pulse
width modulated audio (10 channels as I remember). Upper half of signal was
multiplexed digital coded menu items (food, drinks etc). Each seat had a
unique address. Signalling from seat to receiver was by loading the
transmission line at the appropriate address frame. Only one transmitter in
the whole system. The "receiver" at the transmitter end sensed the drop in
line voltage due to line loading (no reflections either!) A prototype worked
with transmission line length with seat decoders suitable for passenger
aircraft in use at the time. The system was covered by a UK patent which has
long since run out.

This is just one example of thinking slightly differently about a
communication problem in a digital (and analogue in this case) system.

This is possibly another case to take a clean sheet of paper and climb out
of the convention rut! Fixed switching thresholds are bad news. A simple
demo is to make up a simple logic circuit with (gasp) TTL logic gates and
expand one or two inputs with diode OR function. Check the operation as the
circuit temperature is varied.

Adrian

----- Original Message -----
From: D. C. Sessions <dc.sessions@vlsi.com>
To: <si-list@silab.eng.sun.com>
Sent: 03 January 2000 16:39
Subject: Re: [SI-LIST] : Input switching threshold & CPCI

> Jon Keeble wrote:
> >
> > I hear that the cpi/cpci guys are looking at a new termination technique
for
> > backplanes that supports more devices on the bus with the existing
> > controllers.
> >
> > I guess this involves parallel termination at both ends in Zo to Vt with
Vt
> > close to the threshold region (?).
> >
> > How much variation in Vt exists across the range of controllers, and
across
> > the range of fpgas and cplds that people use for pci?
>
> PCI does the old (idiotic) 0.8-2.0 input thresholds that were first
> documented on cave walls. In contrast, anyone trying to do serious
> signaling at nontrivial speeds uses very tight thresholds, usually
> scaled to the driver supply and usually with differential receivers.
>
> The implications of this difference are left to the reader to infer.
>
> --
> D. C. Sessions
> dc.sessions@vlsi.com
>
> **** To unsubscribe from si-list: send e-mail to
majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE
si-list, for more help, put HELP.
> si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list
> ****
>
>

**** To unsubscribe from si-list: send e-mail to majordomo@silab.eng.sun.com. In the BODY of message put: UNSUBSCRIBE si-list, for more help, put HELP.
si-list archives are accessible at http://www.qsl.net/wb6tpu/si-list
****


About this list Date view Thread view Subject view Author view

This archive was generated by hypermail 2b29 : Thu Apr 20 2000 - 11:34:23 PDT