The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
Subject: Re: RS485 help #2
X-Newsreader: Forte Free Agent 1.21/32.243
X-Original-Trace: 5 Nov 2002 20:18:38 +1000, 188.8.131.52
Date: Tue, 05 Nov 2002 09:18:42 GMT
NNTP-Posting-Date: Tue, 05 Nov 2002 20:18:42 EST
Organization: Comindico Australia - reports relating to abuse should be sent to firstname.lastname@example.org
On Thu, 24 Oct 2002 09:09:25 -0500, "EBG" wrote:
>I designated this #2 as I wanted to start fresh with a new post.....thanks
>for all the response on my previous "RS485 termination" post a few days ago.
>Let me ask this.....
>The fact is, I don't care whether my interface really conforms to a specific
>"standard"...RS232 422 or 485.
>My goal is to have a "star network" (central PIC transmitter with branches).
>I neeed to transfer data at a mere 1200 baud (fact is, it can even go lower)
>to up to 10 PIC receivers, which simply do some switching tasks.
>I need however, to eliminate the EMI emmisions coming from the cable by
>using a balanced connection.
>How about this.....
>Has anyone built their own balanced network using a dual optoisolator (say a
>PS2501-2)at the xmtr and each of the receivers? Couldn't this be done? Just
>split the RS232 coming out of the pic, invert one side and drive the
As Frank Bremelman said in his post - the PIC doesn't do RS232 - but if
it has an on-chip UART, then the i/p & o/p pin signals are the 0-5V levels.
Generally a separate RS232 chip is used to connect the UART pins to the
RS232 comms lines to provide the +12V / -12V RS232 signal levels.
As for using opto's - as another poster has pointed out, you will need
to have a separate floating power supply to drive all the opto LEDs -
I'm presently designing a system that uses an optocoupler on the transmit
side to drive an output transistor (rated to 2 amps) which pulls the Comms
signal line (which can be connected to up to 32 opto-isolated receivers - mine
is in a single loop configuration - not a STAR set-up) low - which activates
all the receiver LEDs (the anodes are connected to the separate isolated
+12V Comms supply) at the same time (thus the 2 amp rating on the send
transistor). I'm using digital on/off signalling as I don't need to worry
about EMI emissions as there is only the one two wire signal cable used - so
turning the Tx & Rx LEDS on/off at 150Kbit speeds is very useful.
However for your needs, I suggest you consider the possibility of using FSK
signalling to do your lower speed 1200 baud comms - say 38.4Khz for a high /
19.2Khz for a low - and transmit & receive the FSK signal 'linearly' across
the Tx / Rx LEDs (more as a linear sine wave than as a digital signal) - this
will dynamically reduce any EMI that may be induced. Use a NE567 to
recover the FSK signal back to digital at each receiver point.
In my system, my comms signal is half-duplex, so I have both transmit &
receive circuits on the same 2 wire comms line - for your set-up, one Tx
opto / output transistor combination should be able to drive up to 32 (or
even more) receive opto LEDs - either in a single net config (serial) or in
a STAR config (parallel). An output drive transistor rated at around 2 amps
or more is recommended to be able to drive ALL of the receive opto LEDs
similtaneously without degrading the signal.
Your suggestion of converting to 'balanced' RS232 (or RS422 / 485) lines
may not be necessary if you use FSK signalling at your 1200 baud speed.
Anyway, I hope my comments are useful.
(Retired - sometimes electronics tinkerer from OZ).
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup