NNTP-Posting-Date: Tue, 22 Oct 2002 22:59:50 -0500
Subject: Re: RS-485 termination
From: Bill Paul
Date: Tue, 22 Oct 2002 22:59:51 -0500
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly
"EBG" wrote in
> Up to now I've been using a PIC'S software SEROUT RS-232 to drive an
> optocoupler at only 1200 baud to feed a group of receivers in sort of an
> optoisolated current loop situation and it's worked flawlessly.
> I need though to eliminate the RFI being generated by the unbalanced
> connection and want to change everything to
> RS-485, (something I've never "tutored myself" on).
> One thing that puzzles me though, if I were to have these
> systems where the end-user can install anywhere from 1 receiver to a
> maximum of 10 receivers on the line, what kind of termination resistance
> could I put on each receiver so that there is a "window" that will be
> acceptable? ie..
> maximum resistance with one.....minimum with 10?
> I'm staying with a mere 1200 baud.
In practice, you won't really need any termination until you get into
faster data rates, or longer runs. I built a system that was running at
115.2K baud, total length was 150' or so, and 40 receiver stubs of about
24" each. On the scope I could see the difference with and wthout
termination, but without termination the signal was marginally acceptable
and 0 bit errors.
So the short answer is, for your 1200 buad and 10 receviers, you shouldn't
need any termination.
Also, you're not supposed to terminate each recevier, only at the end of
your RS-485 bus.