Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!zardoz.cpd.com!dhw68k!felix!dennisg@felix.UUCP
From: dennisg@felix.UUCP (Dennis Griesser)
Newsgroups: sci.electronics
Subject: Re: IR Reception - Modulation and Quality
Message-ID: <160142@felix.UUCP>
Date: 1 Apr 91 23:20:47 GMT
References: <2766.27E40773@ofa123.fidonet.org> <159639@felix.UUCP>
Sender: daemon@felix.UUCP
Reply-To: dennisg@felix.UUCP (Dennis Griesser)
Organization: FileNet Corp., Costa Mesa, CA
Lines: 60

In article <159639@felix.UUCP> I complain about X-10 signals interfering
through the power supply with the Sharp IR receiver modules sold by Radio
Shack.

John Whitmore was kind enough to reply via E-mail, but my mailer can't respond
to him.  Since the discussion is perhaps interesting to other experimenters,
I'll take the liberty of replying here...

> 	This is very odd indeed; the full-wave rectifier in your
> power supply should only turn 'on' the diodes near the AC peaks, 
> and the X-10 burst is near the zero crossing (not the peak),
> so should never get past the diodes (let alone the filters).

Yes and no.  Each X-10 burst is sent three times, to coincide with the zero
crossings in other phases of the AC line.  So we might well get an X-10 burst
at (almost) any time.

Also, different X-10 transmitters produce signals of different amplitudes.  At
home, I have observed the following progression:
	maxi_controller < CP-290 << TW-523
If the bursts are strong enough (say 100v p-p), it doesn't matter when it
hits, it will still get through the bridge.

Perhaps I should try a MOV.

> 	The LM309, however, is a terrible excuse for a regulator
> at high frequencies; the gain of its internal error amp is low at high
> frequency (and you get a millivolt of AC out for each volt of AC in
> at circa 1 MHz).  The plot shows 30 dB rejection at 100 kHz (which I
> understand is the X10 frequency).  You can double that with an
> RC filter ahead of the regulator (it would be something like
> your 10 uF capacitor with a 16 Ohm resistor), which might be a help.

The X-10 frequency is actually spec'ed at 121 kHz, but most writers just
round it to 120 kHz.  This is probably close enough to 100 kHz to make your
figure from the plot at least ballpark.

Good idea for the R/C.  I'll try it in a couple of days.  Can you give me a
formula for attenuation as a function of R/C/f?  [Give me a break, I'm a
software type.]  Otherwise, I'll try that 16-ohm/10-mFd combination.
 
> 	Maybe it would pay to use smaller current-carrying capacity
> in your full-wave rectifier; the big ones have a lot more stray
> capacitance than the smaller ones.  

I have actually tried two power supplies, both exhibiting the same problem.
The one that I use more frequently has a bridge rated at only 1.5 A.  Doesn't
seem all that high.

In the meantime, I have experimented with outlandishly large values of
electrolytic filter capacitor in front of the 3-terminal regulator.  A couple
thousand mFd extra eliminates the interference from the CP-290, but lets
the TW-523 through.  The latter starts to get better around 47000 mFd and
is eliminated around 80000 mFd.

In addition to cost and general lack of style, the huge-capacitor approach
is tough on the bridge rectifier when you turn it on.  There's got to be a
better fix out there.

[Thanks again for your response, John.  Anybody else care to step up to bat?]