I've been using CFL friendly X10 appliance module along with the 3-way
switch units and a motion sensor for a couple of years. I've never
had a flickering or a relighting problem and the dimming feature of
the switches works just fine.
However, my problem has always been the fact that the lights do not
consistantly turn off. I can't tell you how many countless number of
times I leave for work in the morning and have to reach back into the
door and tap the switches to turn off both lights that are controlled
by the appliance module. It's maddenly inconsistent.
The weird part is that SWMBO and I can take the dogs out at night and
the lights will come on when we leave the house. By the time we get
back they are usually off, but they come back on as we walk up the
driveway. That's what I would expect to happen. However, an hour
later, I'll look out and they'll still be on. Makes no sense. Why did
they time out while we were walking the dogs, but not time out once we
were inside the house?
On the other hand, I have an X10 appliance module controlling a tube-
FL shoplight and had to add a incandescent bulb (a Christmas candle)
to the circuit to keep it from coming back on after it timed out. This
setup has no switches, just an appliance module and motion sensor and
it times out and turns off every single time.
I just mentioned to SWMBO the other night that once it warms up I'm
getting rid of the X10 for the 2 front lights and going back to
regular motion sensors for each fixture. I'll lose the dimming feature
but at least the lights won't (sometimes) stay on all night and
(sometimes) all day.
John Maddenly? (-: This maddening problem is fairly typical. In my last
post you'll notice I use a smart controller (the now discontinued HomeVision
unit) that "knows" that when it's light outside, outside lights should be
There can be a lot of reasons for the setup you've described to not work as
intended. Any time the X10 signal is marginal, as it can be with long
electrical runs typical of porch and outside light wiring, a signal can get
"lost" depending on what else is turned on at the time. Diagnosing weak
signals is an art with X10 and I was never very successful at it until I got
an X10 meter to read the signal strength of the control signals on the
It only takes one badly behaving lamp that's "electrically near" the circuit
panel to mess things up but good. I had a space heater that simply "ate"
X10 signals on that circuit and several others. When it was cycled on and
running, no X10. When it kicked off via thermostat, everything was fine.
What fun *that* was to figure out. CFL's and fluorescent shop lights might
activate with no problem but once lit they can produce enough electrical
noise to inhibit their remote shutdown. Again, an X10 meter can read noise
on the line in the 120KHz range that X10 uses to propagate its signal.
It may be that once you got back into the house, you turned on a lamp or
device that emits enough noise to inhibit the powerline "OFF" signal from
the motion detectors (EagleEye? HawkEye? DM10?) from reaching the light to
shut it off. Again, without a meter it's very hard to determine if the
signal level is marginal or good or if there's sufficient noise present to
block signals. That covers most of the problems with X10, but there could
be something else happening.
It used to be that you could sort of diagnose these issues by standing at
the circuit panel and turning off all the circuits except the ones in use by
X10 and restoring the circuits one by one until the X10 signal failed. That
method had a very low "spouse approval" factor when I inadvertently caused
the VCR to reset and miss recording the season finale of one of her favorite
This is a case where with a CFL friendly appliance module, you could scrap
It all depends on circuit paths and as Trader pointed out, phases. If
you're controlling items that are all wired to one side of the panel, things
are usually A-OK. It's when a control device on one phase is expected to
control a device on a different phase that you run into problems. In such
cases, without a coupler or repeater you're typically allowing the phases to
couple at the transformer and that *never* results in a strong signal. You
can pull the cover off your circuit panel (another low spousal approval
thing, I've found) and by inspection determine what outlets and fixtures are
on which phase.
A simple coupler can be made of a capacitor, but I recommend something
designed for the problem that not only couples the phases, but amplifies the
signal, too. I use Jeff Volp's XTB-IIR but there are other, cheaper (but
inferior) couplers on the market.
I have a Leviton HCA unit and an X10 coupler as spares I could lend you. I
believe I even have an older model of the XTB-II and even older,
non-coupling amplifiers (XTB and XTB-R) that you simply plug in the RR501
transceiver (or MaxiController or other X10 transmitter) and it boosts the
output from around 5 volts to nearly 25 volts. That's enough to couple at
the pole transformer and still have enough X10 signal left to reach the
When X10 was designed, 5 volts was enough to reach nearly everywhere. Now
with so many chargers, CFLs, UPSs and more plugged into the home wiring, 5
volts is very, very little voltage compared to what's actually needed to
pump a strong X10 signal throughout the average house.
*>regular motion sensors for each fixture. I'll lose the dimming feature
My side porch light is an independent motion sensor unit that's controlled
by an X10 wall switch. That way, I don't have to worry about the X10 motion
sensor reaching the RR501 transceiver and then the transceiver sending an
X10 signal to the lamps. I have the whole shebang supervised by HomeVision
that turns the wall switch off when it's light out.
HomeVision has a dawn/dusk calculator that's based on the user's
lat/longitude which is double-checked against a "sundowner" unit that has a
photocell that "knows" when it's dark out and sends X10 commands when
there's a transition.
(discontinued, unfortunately, but you can do the same thing with a Hawkeye
and the plus one dawn/dusk detection feature)
Like you, I often left the house only to maddeningly discover that the porch
lights were on. Not anymore.
My best guess about your situation is that the porch lights are on a
different phase from the outlet that the RR501 transceiver is plugged into
(although I dimly remember that may not be the case in your house).
Depending on what is powered up at the time, the signal either makes it to
the lamps or not. Usually it's a noise issue. Something between the RR501
and the modules you are controlling is a device that interrupts the X10
signal either by interfering with it (noise). But it could also be that the
signal is getting absorbed by an attenuator, aka a "signal sucker." UPSs,
PC power supplies and surge protecting powerstrips are all candidates for
signal sucking because they often have integral noise filters and to them,
the X10 signal is simply RFI.
From what you've described, I suspect that when you return home from your
walkabout, you turn something on in the house that's inhibiting the
powerline signal from reaching the device. An electrically noisy CFL, even
one not controlled by X10 but on the same circuit, would be enough to screw
things up. With Hawk/Eagle-eye motion detectors, standing between the
sensor and the RR501 receiver is often enough to prevent the RF signal from
getting through. They are basically what ham radio guys call "flea powered"
devices and plaster/lathe construction often blocks the signal entirely. Is
there more than one RF motion sensor in the house? They love to "fight"
with each other under certain conditions.
I no longer use the stock X10 transceivers for RF reception but have
switched to a WGL all-housecode transceiver that feeds an XTB booster. The
WGL has coaxial aerial connections that allowed me to run multiple antennas
on all four sides of the house feeding into the WGL transceiver via a
four-way splitter and RF amplifier. I can turn on the porchlights with a
keychain controller from over 100' from the house. The bathroom light (one
of the few controlled by a Hawkeye directly) activates 20 times a day
without failure, so it can be done.
I consider the XTB and WGL additions mandatory for maintaining the peace
with SWMBO. Her tolerance for lights coming on (and especially going out)
at random dropped lower with every failure. My situation is a little
different though. I've been using X10 since the '80's and everything in the
house is remotely controlled so it was either "get it to work" or "get off
your lazy ass and turn the lights out by hand like every other human being!"
So I got it to work. (-:
I know that you'd prefer (like most people) not to spend any more money on
something that already seems dubious but meters and repeater/couplers really
are getting to be the price of entry into the world of X10. One of the
vendors used to rent out X10 meters for very little money because end-users
usually only need the very rarely to find the big problems but I wasn't able
to find anyone that currently rents them.
I wish meters and repeater/couplers had not become "entry level" items but
on the other hand, I've gotten some tremendous bargains from people who have
given up on X10 and put all their equipment for sale on Ebay. Their loss
was my gain. I recently saw such a lot go for over $400!!!!! So there's
still a lot of interest in X10 out there to command prices like that. Every
cloud has a silver lining and every tornado has some loose change spinning
around in it. (-:
The stock X10 transmitter modules should emit a more powerful signal but
from what Mr. Volp tells me it takes some pretty big capacitors to spit out
a 25V X10 signal. There's no practical way to make the standard X10 gear
more powerful without making it much larger. The XTB boosters and repeaters
are quite a bit larger than stock X10 modules because they need much larger
and more powerful components to produce higher than stock voltages.
Are you comfortable working in your circuit panel? Do you have a 240VAC
outlet in the house that you can plug a coupler/repeater into for a test?
Coupler/repeaters require access to both hots and the neutral wire and that
can often be found at a dryer or range outlet.
My XTB-IIR repeater/coupler/amplifier is connected to the main panel via a
tandem 20A breaker connected to both phases. I put a Hubbell twist-lock
plug on the end of the repeater and wired a 240VAC twist-lock receptacle to
the XTB-IIR so I could swap it out easily if it ever failed or if I wanted
to compare repeaters. That's preferable to the dryer outlet technique
because the closer to the circuit panel, the better the signal propagation.
My 240VAC outlet is connected with about 10' of Romex, so it's very close.
X10 forced me to learn enough about home wiring to be able to install a
coupler/repeater and that's a good thing.
I'm with you on X-10 being unreliable and not suited to
any serious home automation project. It can be fine
if you want to spend $25 on Ebay to control a light that
isn't essential. I remember recommending it a few
years ago to you for your control of your outside light.
You had problems from the beginning, but I think you
eventually got it to work, at least to being usable. At that point in
time, my limited experience with X10 was that it worked.
Even then, it wasn't 100% reliable, but my experience
had been that it was good enough for that kind of
motion sensing, turn on the light kind of application.
And I'd still use it for limited apps like that. But you
quickly ran into problems even using it for that.
Since then, I've seen more than enough to convince
me that the whole system is crap and you're probably
going to be disappointed and frustrated with it at some
point. Either from motion sensors that don't last or
from the inability to get it to work where you need it to
work, etc. The whole thing has one big flaw from the
start. It's a one-way protocol. The sender simply sends
a command to turn on or off and assumes that it got
I helped a friend install a system that uses a motion
sensor to turn on outside lights by the garage, turn
on porch lights at dusk, and turn on/off a few lights
in the house randomly at night. We got it working
and it was OK, with some exceptions. The motion
sensor was far more sensitive to stray movement of
a tree branch etc than other motion detectors I've
used. We had problems getting the right modules to
work with CFL, as per this thread. At it's best, it
was probably 95% reliable. There were still times
when you'd find a light didn't go on or off as it
should. But most frustrating,
after working for a few years, one of the light modules
simply would no longer respond to the controller
one room over. We tried different modules, different
outlets, removing any nearby cable box, etc. It just would not
work. And there was no new appliance, no moving of
an appliance, nothing that could account for a change
in the house.
I said in another post that I don't know the specifics,
but I think the company that was the main driving force
behind those modules that are out there has exited the
business. Clearly you can see from what's available
that not much, if anything new is being developed.
Could still be OK for a limited investment and application,
but not what I'd choose for any home automation
project of significance.
HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.