[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: X-10 WARNING



Original poster: "K. C. Herrick" <kchdlh-at-juno-dot-com> 

Quite a while back I posted a warning about zapping household X-10 products 
with Tesla discharges.  It's very likely I did that but I now, belatedly, 
post a caveat.

About the time I had that problem, I had installed a new computer (but not 
for the reason that it, too, had been zapped), and ever since that time I 
have been baffled by the fact that one of my X-10 locations has continually 
been plagued by incomplete operability, even though I have not, of late, 
been making Tesla sparks.

So I finally have made the connection:  Home-automation products marketed 
by companies such as X-10 depend on power-line-transmitted rf signals 
occurring near mains-voltage zero-crossings.  Electronic apparatus 
incorporating switching power supplies can put EMI on the power line that 
will interfere with those low-level signals.  I found that that's what my 
computer was doing.

X-10 markets a filter (their model XPPF) that is supposed to take care of 
that kind of problem.  Finding that it did not in my case, I made up a 
simple trap that works very nicely and which might do the same for other 
coilers.

The X-10 products send and receive their signals within 200 us of the 
zero-crossings.  With the nominal 160 V peak amplitude of (U.S.) mains 
voltage, about 4 V of amplitude is reached at 200 us.  What the trap does 
is to keep the mains-circuit to the load essentially open for that 200 us 
by the simple expedient of connecting a string of back-to-back diodes in 
series with each side of the mains circuit.  I connected 3 diodes in 
series, in each side, paralleled by another 3 reversed.  That provides the 
"dead band" of about +/- 4 V.  I added 0.15 uF capacitors to ground at the 
load-side of the diodes to further reduce the EMI.

The trap is, of course, too good by half: it not only blocks the EMI during 
the interval but also the mains voltage itself.  But not to worry: with 
mains voltage at nominal, most electronic equipment should not be adversely 
affected by the loss of 2 1/2% of the peak.

Ken Herrick