Path: utzoo!utgpu!water!watmath!clyde!att!pacbell!lll-tis!lll-winken!uunet!mcvax!ukc!dcl-cs!nott-cs!pyr1.cs.ucl.ac.uk!william
From: william@pyr1.cs.ucl.ac.uk
Newsgroups: sci.electronics
Subject: Re: Powering circuits from serial p
Message-ID: <44000014@pyr1.cs.ucl.ac.uk>
Date: 22 Jul 88 15:54:00 GMT
References: <1712@iscuva.ISCS.COM>
Lines: 23
Nf-ID: #R:iscuva.ISCS.COM:-171200:pyr1.cs.ucl.ac.uk:44000014:000:1264
Nf-From: pyr1.cs.ucl.ac.uk!william    Jul 22 15:54:00 1988


Yep, you can do this.  In fact I sucked power from everything coming in
including Rx.  I had the RS232 inputs all pumping into a bridge
rectifier feeding a pretty big capacitor, and this could drive
a little CMOS processor and UART.  The advantage was that the system
could enter a very low power mode to maintain its memory and charge
the cap when there was no Rx data, but the processor could be
woken up by a change in Rx and could process signals.  It also had some
control work which required the power in the cap.  I imagine that a 
cap could be used to store a +/-12 volt supply for short transmissions, and all
you would need to control it would be a couple of transistors or something.

I don't recall the exact spec, but I seem to recall that RS232 levels were
defined as valid from (+/-) 3-12V.  I think.  I remember that I had to
decide whether the above circuit should have to be designed around the fact 
that levels may be as little as 3V.

			... Bill

************************************************************************
Bill Witts, CS Dept.     *
UCL, London, Errrp       *       Don't believe everything you hear,
william@uk.ac.ucl.cs(UK) *       or anything you say.
william@cs.ucl.ac.uk(US) ***********************************************