<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<HTML>
<HEAD>
<TITLE>LIRC - Enrique Vidal's improved transmitter circuit</TITLE>
<LINK REL=stylesheet TYPE="text/css" HREF="lirc.css">
<LINK REL="shortcut icon" HREF="favicon.ico">
</HEAD>
<BODY BACKGROUND="images/marb18.jpg"
BGCOLOR="#FFFFFF" TEXT="#000000" ALINK="#9090FF">
<TABLE WIDTH="100%" BORDER="0" CELLSPACING="0" CELLPADDING="0">
<TR>
<TD CLASS="menu" WIDTH="100%" HEIGHT="150">
<IMG SRC="images/diode.gif" ALT="diode" WIDTH="300"
HEIGHT="150" BORDER="0" HSPACE="20" VSPACE="0" ALIGN="LEFT">
<IMG SRC="images/lirc.gif" ALT="lirc-logo" WIDTH="300"
HEIGHT="150" BORDER="0" HSPACE="20" VSPACE="0" ALIGN="RIGHT">
</TD>
</TR>
<TR>
<TD WIDTH="100%"> <BR>
<!-- Text -->
<HR WIDTH="70%">
<H1 ALIGN="CENTER">Simple signal-powered, middle-range
infra-red transmitter</H1>
<HR WIDTH="50%"><BR>
<TABLE WIDTH="100%">
<TR>
<TD WIDTH="15%">
 
</TD>
<TD WIDTH="70%">
<center>
<IMG SRC="images/improved_transmitter.gif" ALT="improved transmitter">
</center>
<h3>How it works:</h3>
<P>
While the computer serial device is open and idle,
DTR is negative w.r.t. GND and the large 4000
µF capacitor gets full charge through D2 and
the 2k2 resistor in a few seconds. Meanwhile, the
input to the NPN transistor remains blocked thanks
to D1.
</p>
<p>
As IR commands are sent to the serial device,
positive pulses arrive to DTR. These pulses have no
effect onto the capacitor, since they are now
blocked by D2. While the capacitor has enough
charge, it powers the couple of NPN/PNP transistors,
which behave as a simple conventional pulse current
amplifier. They yield enough (pulsed) current to
the pair of infra-red diodes to allow mid-range
operation; typically beyond 5 meters. The
(optional) red LED and 0.5k resistor are intended
for visual monitoring of the outgoing IR bursts.
</p>
<p>
Obviously, the current pulses tend to quickly
discharge the capacitor, but 4000 µF can hold
enough charge for regular typical operation with
most IR commands, even in those cases where bursts
are repeatedly sent while
button-pressed. Fortunately enough, even in these
repeated-burst situations, the serial device is in
fact idle most of the time, thereby allowing the
capacitor to recover its charge and be prepared for
powering new pulse bursts.
</p>
<h3>Notes:</h3>
<p>
Any type of low/medium-power diodes and transistors
are suitable. The only critical components (apart
from the IR diodes themselves) are the two voltage
offset diodes. These diodes are marked as
"silicon", meaning that they must be "normal"
silicon diodes, which have a significant offset
voltage (+0.7 volt, approx.). Other kinds of diodes
(e.g., germanium diodes) may have much lower offset
and are inadequate for the intended purpose.
Tolerances are not critical for any component of
this circuit (±10% or even ±20% should
be ok).
</p>
<p>
Of course, the range depends on how much power is
applied to drive the IR diodes and of the type of
diodes themselves. I have tested several kinds of
IR diodes (intended for standard remotes) with no
much difference. As to the power, it depends on two
main factors: a) the value of the (22 Ohm) resistor
from GND to the emitter of the PNP transistor and b)
the large (4000 µF) capacitor. The range can
always be increased by lowering the value of the
resistor (and/or adding more IR diodes) but than the
value of capacitor needs to be increased
correspondingly.
</p>
</TD>
<TD WIDTH="15%">
 
</TD>
</TR>
</TABLE>
<!-- Text -->
<BR> <BR>
<CENTER>[<A HREF="index.html">LIRC Homepage</A>]</CENTER>
<BR> <BR>
</TD>
</TR>
<TR>
<TD CLASS="menu" WIDTH="100%">
 <BR>
</TD>
</TR>
</TABLE>
</BODY>
</HTML>
