Automatic control of home lights, appliances, window shades, environmental
variables, etc., especially with computer controlled X-10 (type) systems. 

22 responses total.

(I was talking to a person the other day who wants to add handshaking to
the X-10 products.  To confirm that a given module did indeed do what it
was told to do.  As it stands today, X-10 only gives commands.)

I meant to enter more information on X-10 after I started this item, but
got called away. But, its fine for it to grow on its own. For my first
contribution, here is a newsgroup on home automation. Even this short list
gives an idea of the breadth of interests and applications: 

======   9 unread articles in comp.home.automation -- read now? [+ynq]
22727 Re: MOUSEHOUSE MAC software for CP290?
22728 Re: LonWorks versus CEBus
22729 Re: Halloween Automation
22730 Line Sense Relays for Home Control FS
22731 Re: 70V line audio
22732 Re: LonWorks versus CEBus
22733 Re: Light switch that is hard to turn off
22734 Re: CK11A Two-Way Kit by X-10 ONLY $79.45 Wow, what a deal!!!!
22735 Re: Halloween Automation

My X-10 system is controlled by a CP-290 programmed from an XT. I control
20 lights and 1 heater. All house outside lights are controlled with on
and off times both adjusted for the season and with somewhat randomized
switch times.  I run all outside lights with 60 watt bulbs at 80% power so
they last a long time. Inside lights also follow the seasons and our use
patterns There are many special uses possible: for some time the kitchen
light also flashed on and off at the time our daughter was supposed to
leave for school. Because dimming modules stay *off* if a power failure
occurs, I have several on times programmed during the whole intended on
duration. 

We discovered that, even though there are more lights on than before the
control system was set up, our power use (and electrical bill) dropped.
This was because of both running lights at reduced power and, probably
more important, *turning off lights* when they should go off. My wife was
rather a sceptic (X-10 is another one of those "men" things, for some
reasons - technology, I guess), until she heard a noise one night I was
away, and pushed the All Lights On button. The house really lit up. She is
now "sold" on the system. 

I use my X-10 system for my wife who is deaf.  I built a controller interface
that flashes the lights in coordination with a ring signal on the phone line
and it will also turn on the same lights for about 10 sec. when someone rings
the doorbell.  I used a standard $10 controller and wired relay contacts
across the ALL ON and ALL OFF pushbutton contacts.  It's been in operation
for baout 8 years and has worked *very* well.  I've lost a few lamp modules
for whatever reason overh the years though.  Don't know why.

The X10 FAQ from comp.home.automation is in /u/rcurl/X10.FAQ - all ca.
141K of it. It is hard to use it is so big. You can grep for things, but
then its hard to find what you find. It would be more useful on the web,
with a hot-linked index.

The FAQ contains some interesting mods - for example, to disable the local
sense, so fluorescent lights won't turn back on, as they sometimes tend to
do, on Appliance modules, or adding local dimming to wall switch modules.

[I'm ticked, though, that it omits my design for a resonant signal
bridge.]

Resonant signal bridge?  What's that good for Rane?

Your house is (probably) wired from both sides of of a 240 volt two-phase
service. The X10 transmitters inject their signal into only one side. It
get across to modules on the other side probably mostly through loads, but
the signal is weaker. A signal bridge jumpers the two phases with a
narrow-pass filter for the X10 frequency. An *active* signal bridge is a
repeater, which keeps the synchronization of the signal and the power
correct.  A passive bridge is just a filter, so synchronization is not
assured, but it usually helps. Some people jumper the two phases with a
0.1 ufd cap. I tuned out the reactance of that with an inductor, so the
reactance is very small at 120 KHz.

Ahh.  So you have a series resonant circuit bridging both sides of the
line, resonant at about 120 KHz?  .1uf in series with 17.59 uH?
Perhaps they didn't include your suggestion because of safety issues
with people fooling around inside their fuse boxes, etc.  It might
also be a good idea to include a small fuse in your circuit should
the cap short, etc.  I get all sorts of neat sparks inside my fuse
box during close, violent, thunder storms  :-)  Don't know what that
might do to your average 500 vac cap.

I included a fuse. I built the circuit into a 240 V plug that I put in an
unused outlet for an electric clothes dryer (I guess...), so I didn't do
any "wiring". You have to use a low resistance 18 uH choke (off the shelf.
I also used a cap rated at 630 VDC (which is like a 500 VAC cap). 
Hmm..maybe I better check if it's blown the fuse yet.... The fuse is
actually the largest resistance in my circuit. The total resistance is 2.8
ohms, and it is resonant at 120.8 KHz (measured), and the Q = 4. The
half-power bandwidth is ca. +/- 20 KHz, so there really isn't anything
critical about tuning it.

Here is the only mention of the 0.1 ufd bridging cap in the FAQ: 

"one side will not transmit reliably to a receiver on the other side. Try
your X10 system with and without your electric stove turned on; turning
the stove on may bridge both sides of the power line, but is not the
recommended permanent solution. A better way would be to install a signal
bridge which is available as a commercial product. See section 2 below for
details. An alternative solution is to install a 0.1 microfarad capacitor
(240 VAC or 600 VDC) across the 220 volt line "hot-to-hot". A qualified
electrician can do this across any 220 volt double pole breaker. This will
bridge the signal from one side to the other." 

Actually, I got fifteen seconds of my alottment of fame when my usenet
posting was republished by "Uncle Phil" in the Advanced Control
Technologies Inc newsletter for their installers. It turned out that ACT
also sells such a tuned passive coupler - which of course they recommend
over building your own. 

I replaced a compact-fluorescent lamp bulb in a floor lamp in which I use
two such lamp bulbs, controlled by an X-10 appliance module. While the
first pair of bulbs worked fine with the module, the new bulb caused the
lamp to refuse to be turned off: it would come back on any time from
immediately to ten minutes after being turned off. This is a "classic" 
problem with fluorescents on appliance modules. The X-10 FAQ suggests
disabling local sensing by cutting a wire in the module. I sought a less
drastic solution.

Since local sensing works with incandescent bulbs, I tried a low (4) 
wattage bulb in parallel with the lamp. This stopped the lamp from turning
back on by itself, but it also disabled intentional local control. A
single fluorescent bulb exhibits a DC resistance of 13 megohms. Apparently
switching this in parallel with a 4 watt incandescent bulb, which has a
resistance of only a few hundred ohms (when off), causes too little
current change to be sensed by the module. I therefore tried a larger
parallel resistance. My choice was based on using the lowest resistance
that dissipated ca. 1/4 watt (as I had only 1/2 watt resistors), which is
47K. This is still a lot less than 13 megohms - but it solved the problem: 
the lamp does not turn itself back on, and local control works. I do not
know, however, what is the optimal parallel resistance to use, nor how it
depends upon the fluorescent unit. 

I put the 47K 1/2 watt resistor in a replacement cord plug, and plugged it
and the lamp plug into a multiple socket adapter on the X-10 module.

Interesting.  I suspect the problem with the compact-fluorescent bulb was
that the first one you had in the used an "iron" ballast while the more
recent one uses an electronic ballast.  I suspect that the electronic
ballast would show a much higher impedance.  They are also suppose to be
more efficient.

The one that failed was the same make and model - though it was four years
old, and a slightly different shape. So, I tend to agree that the
difference was a change in the electronics. The "mate" is also the current
design.

There is a circuit diagram for the X-10 Appliance module (AM486) at
http://www.nysaes.cornell.edu/cc/staff/pool/homeauto/am486.html

There are circuit diagrams for five X-10 modules - appliance, lamp, chime,
wall switch and computer interface - at
http://www.io.com/~hattersj/index.html
Follow the "How to modify....." link. These drawings are all in .PCL format,
so you have to print them to a HP laser printer, and they are also in A4
format, which apparently prints OK on legal size paper. 

I visited the site yesterday.  I couldn't establish a link to the server
that the .PLC drawings were on either, as you had mentioned.  I'll have
to see if AutoCad will read .PLC files...  I also now have a HP 11" X 17"
plotter acctached to my PC at home.  Perhaps it'll plot them if .PLC is
anything like HPGL.

The site came up this morning, and I downloaded the files - still can't do
anything with them, though.

Rane, you will be happy to know that I downloaded the .PLC files and was
able to print them to my HP Desk Jet 500C *after* setting the dip switches
to tell it that we were using A4 paper (DIP switch A-6 up.)  I printed up
a set for you.  Petty cool :-)

THanks! I look forward to them! What did you print from, a PC? I have
a DeskJet 500, but didn't think I could deal with the .PCL files from
an XT. Could I have? Now that you've told me how, I may just try it.

Your XT and the Desk Jet 500 should work just fine.  Just be sure to
filp DIP switch A-6 up and cycling the power before printing.  I used
legal sized paper.  I also plan on trying to print them to my HP
plotter... Perhaps tonight.

My XT-class machine (Zenith 150) and DeskJet did work just fine (with
switch A-6 up). I had to make A6 paper, however. The figures are not quite
as crisp as I guess an HP laser printer would do, very adequate. The
*hardest* part was remembering DOS commands, finding PKUNZIP again, and
other arcane trades that have faded from short-term memory.

I've attempted a repair on a wall switch module (WS467), which went dead.
I found R2=22 ohms burned and open. I replaced it, but it still doesn't
work. It looks like that resistor is in the emitter (?) circuit of the
SCR. Could a failed SCR have led to R2 being sizzled? Or, if not that,
what might be the next culprit to check?

An SCR has three terminals.  Anode, Kathode and a Gate.  The turn-ON
signal is applied between the Gate and the Kathode.  A good SCR will
have about .7 volts between the Gate and the Kathode.  I suspect,
however, the device you are talking about is a Triac. acs are
sort of like back-to-back SCR's and will pass both sides of of an AC
wave.  SCRS are more like diodes and will only pass one side.  A triac
has three terminals too, MT1, MT2 and a Gate.  The controll signal is
usually applied between the Gate and MT1 (MT = Main Terminal.)  Pull
the Thyristor (SCR or Triac) and measure the resistance between all the
terminal combinations.  If any one of them is a dead short, the device
is history.  Replace it.  It could well bethat a bad Triac caused R2
 to fry...  The other question is if the current got into the IC and 
fried that too?

I had the same problem with local sensing and flourescents that Rane mentioned
way back there.  In a nutshell, the inductive load of the flourescent tended
to make the module switch right back on after it had been switched off. 
Damned annoying, and I'd even had an appliance module or two die.  My
solution had been to plug a little surge protector into the load side, which
absorbed the power spike from the shutoff.  However, a couple of new compact
flourescents I just got still caused the module to be fooled.  So I finally
spent some time with the not-very-clear FAQ directions and a module and made
sense of the mod to disable local sensing.  It's pretty easy *and* reversible;
all you have to do is unsolder one end of a jumper.

By the way, compact flourescents have gotten much nicer since I'd last given
up on them 2-3 years ago.  I bought some Feit Electric units at Meijer (these
are the ones with the interestingly twisted tubes), and I'm very happy.  Only
$10 and totally silent.  My biggest (perhaps only) problem before had been
the noise flourescents make; these are dead quiet even to my sensitive
hearing.

Response not possible - You must register and login before posting.

- Backtalk version 1.3.30 - Copyright 1996-2006, Jan Wolter and Steve Weiss