Re: Bypass GFCI using Transceiver?

["Robert Green" <ROBERT_GREEN1963@xxxxxxxxx>]

"Chris Shearer Cooper" <chrisnews@xxxxxxx> wrote in message
news:11tj92atmbp20a6@xxxxxxxxxxxxxxxxxxxxx
> I've got some circuits that are protected by a GFCI outlet - which, after
> doing some testing, seems to also mean that those circuits are protected
> from receiving X10 signals.

What kind of testing?  Did you get hold of an X-10 signal strength meter?
Did you replace the GFCI with a standard $.79 non-GFCI outlet and things
then worked?  Those would be the only tests that an X-10 "pro" would
consider as irrefutable proof the GFCI was the culprit.  If you pick up a
different brand, you wouldn't even have to wire it into an outlet box.  You
could bench test to determine if X-10 propagated correctly downstream of the
new GFCI before going through the hassle of pulling the old one out of the
box.

> (yes, I've tried the various other suggestions about things on the
circuits
> that can cause noise that can drown out X10 signals - at this point I've
> pretty well convinced myself it's the GFCI that's the problem).

Well . . . you might have convinced yourself, but signal sucking is pretty
complex business. I don't mean to sound snide, but many of us have been on
the same path you're on and many of us have been very surprised by the
things that end up being signal suckers from space heaters to UPS's to
laptop power supplies.  Unless you've got meter results or have swapped out
the unit in question, you're still just guessing.

You can end up chasing your tail for a very, very long time with X-10 unless
you have a meter or some sure way to determine a unit is troublesome.  In
your case you just can't unplug the GFCI as easily as you can unplug a
suspect appliance - but in reality you *have* to remove it from the circuit
so that you're sure it's the cause.  If you're going to pull it, then you
might as well replace it with a unit that you've bench tested beforehand to
insure that it's not affecting the X-10 signal.  What does a new GFCI cost?
Ten bucks?  Even if you value your time at minimum wage, you blew ten bucks
of your time a long, long time ago on this particular problem.

> I want to control things on this circuit using my computer interface
> (currently a CM11A but I'll probably buy a CM15A soon).

You'll need one upstream of the signal sucker and you'll need one downstream
of the signal sucker.  If the GFCI is the source and it's not a perfect
filter, then you've set yourself for the wild and wacky world of X-10
collisions where devices receive partial signals from both transceivers.

If the *only* lights you'll *ever* want to control are downstream from the
GFCI it might be possible to control those units using a CM15A via radio
control BUT the signals won't go beyond the GFCI so those are the only
lights you'll ever be *able* to control.  That's not the typical pattern of
X-10 use, BTW.  You'll eventually end up wanting to control more items.
Virtually everyone does.  Then you'll be stuck because your transceiver is
behind an inadvertent firewall.

> Can I put a transceiver (like the RR501) on the circuit, on the far side
of
> the GFCI, send radio signals to the transceiver from my CM11A

You can't send RF to a CM11A - it's a powerline device only . . .

> CM15A, and
> have the transceiver send those signals via X10 to other X10 devices on
that
> circuit (simple lamp controllers)?  Is that the whole point of the name
> "transceiver", or am I just having an extraordinarily dense night?

If you're right about the GFCI being the signal sucker, there's really only
one GOOD solution and that's to replace the offending device.  Attempts to
work around the problem with multiple transceivers will probably cause new
and different problems even if they do manage to solve the immediate
problem.  The proper solution is to put devices that suck signals behind
X-10 filters.  Since that's not possible with a GFCI, the only real solution
to the problem is to pull it and replace it with another GFCI that's X-10
friendly although it doesn't *seem* that way to you at the moment.

If you live near Washington DC I'll even come by with both my X-10 meters
(yes, they are *that* critical to successful X-10 operation that I have
two - a straight up LED bar meter and a much more complex Power Line Signal
Analyzer known as "the Monterey") and we can determine with great precision
what is interfering with the X-10 signal.  If you've got a noise generator
that's causing the problem, and not a signal sucker, not even your proposal
to put a transceiver downstream of the GFCI will help.  You can't determine
whether your lines are noisy without an o'scope or a powerline meter or
analyzer.