"ANyone have any better ideas, bit-banged serial, etc?"

I have used opto-isolators in the past on the
receive end to differentially 'receive' the
transmitted signal - while simply driving the
twisted-pair line directly with a TTL-level
serial signal.

The receiver end used an opto-iolator that
was "pseudo-matched" to the line impedance
(approriate series/parallel resistors were
used to match the twisted pair's Z to the
Z presented opto-isolator).

I used one of the fast 'active' opto-iolators
that had some speed capability to it - made
by HP (Hewlett Packard I believe) at the time.

I had practically no limit on baud rate that way.

The only thing I observed was - the HP active
optos had a common-mode slew rate (input to
the LED) that had to be observed otherwise
they WOULD output a pulse for the duration of
that common mode pulse (that portion which
exceeded the slew rate would generate a pulse).

Jim



----- Original Message -----
From: "Amaury Jacquot" <sxpert@ESITCOM.ORG>
To: <PICLIST@MITVMA.MIT.EDU>
Sent: Wednesday, April 03, 2002 9:30 AM
Subject: Re: [PIC]: Ling distance i2c between a pair of '872s?


> Quoting Tim Thompson <irule@IRULE.NET>:
>
> > Hello all,
> >
> > I'm working on a project where I will need a pair of 16F872's
communicating
> > at a distance of about 30ft wire length.
> > I was thinking of using the i2c on them both, since i'm familiar with
the
> > code required for this, but was wondering how reliable i2c would be over
> > that distance? ANyone have any better ideas, bit-banged serial, etc?
>
> one solution would be to use a basic communication protocol over a pair of
> plastic fibers.
> I have a system that can go up to 1Mbps with that fiber
>
> Amaury
>
> --

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics