writings       links       about me       music

In the interests of keeping this site somewhat useful to people who live lives outside the internet, non-musicians, non-guitar builders, and others, I present you the following:

One of the character traits that makes me good at my job is that I'm remarkably, occasionally insufferably, lazy.  This is a desirable trait in a network engineer, as it leads those of us who pursue this noble profession to automate our jobs as much as possible.

I carry this wonderful behavior into my personal life as well.  My lifelong quest to eventually move into a powered wheelchair, tethered only to a bag of glucose, was given an assist by the noble folks at x10.com (warning, clicking on the link will result in rape of your eyesockets - their site is the perfect meld of the internet and the infomercial.  Also, turn your sound off before you click.  YOU HAVE BEEN WARNED.)

Kidding about the wheelchair - but not about their site.  Good lord, it is awful.  It pains me to give folks such as these any money at all, but they are the people who manufacture the world's most ubiquitous home automation hardware and software, and moreover, they're the cheapest.

I can now control all the lighting in my house from my recliner, and soon I'll be able to do it from anywhere in the world that has an internet connection.  X10 has really done a great job of implementing home automation in a simple and straightforward manner.  If you can set an alarm clock, and change a lightbulb, you can set up X10 automation in your home, if you desire.

However, X10 has a bit of a problem.

Compact Fluorescent Lamps (CFL) are an issue for X10 - specifically, if you are controlling CFLs with an X10 control module, they have a really annoying habit of turning on at random times - and of turning back on immediately after you turn them off.

There is a reason for this behavior, and it has nothing to do with gremlins or a ghost in the machine.  Nor, as it has been posited on some internet posts, is it due to interference with the X10 control signal being interfered with by EMI (electro-magnetic interference) from the CFL bulbs themselves.  

Let's dispose of the interference hypothesis first.

X10 works by means of a signal broadcast over the AC wiring in your home.  Now you may be thinking "gosh, those wires are full of electricity, how is that even possible?"

Well, it's actually pretty easy.  I used to use my home's AC circuit as an antenna for my radios when I was a kid.  Now, granted, I was a precocious tot, and knew how to do this without burning the house down.  Let's have a look at how electricity works.

You need two wires:

Wire A is what we will call "positive" (that's what the + sign is for).  Wire B is what we will call "negative" (that's what the - sign is for).  Electricity can be thought of as a stream of water.  Water always flows downhill, right?  Well, with electricity, electrons always flow from positive to negative:

I'll use the water analogy again to help explain some electrical terminology; "volts" would be the amount of water pressure.  "Amps" would be the amount of water.  

Now, what I've shown you is the simplest form of electricity, what is termed "direct current", also known as "DC".  It's what batteries produce, and is good for low-voltage, short-distance circuits.

What we use in our homes is "alternating current", also known as "AC".  Invented by Nicola Tesla, who was a freaking genius, AC forms the base of everything we use that works on electricity, from computers to light bulbs to electric motors.  It's hard to overstate the genius of Tesla, by the way - I've been reading his collected works recently and it's frankly creepy.  The electrical infrastructure of modern society blossomed from this man's brain fully formed, tested and ready to go.  I'm not one to believe in this kind of thing but the only explanation I can come up with for how Tesla got his knowledge is that perhaps he was a time traveler from the future.

At any rate, the way AC works is that, unlike the circuit shown above, where the positive and negative are fixed, AC circuits alternate their positive and negative orientation.  They do this, at least here in the US, 60 times per second.

Here's a picture to show what DC (direct) current would look like if you were capable of seeing it:

The red line would represent a signal of 120 volts DC.  It never changes - just a straight line for the entire duration of the graph.

120 volts AC behaves quite a bit differently:

You can see here that the current is rolling between positive and negative.

Here's where X10 comes in.  In order for the X10 signal to carry through the AC line, it needs to broadcast at a time when there is no voltage on the line or else the signal would get buried in the noise.  So it broadcasts when the line voltage hits zero, which happens sixty times a second.  The zero-voltage point is marked with green here:

The idea behind the interference hypothesis (go back up to the top and reread the part about X10) is that interference generated by the CFLs would be seen as an X10 "on" or "off" signal, causing the device to trigger on its own.  This is not what happens with the "random turn-on" and "refusal to turn off" problem.

How do I know?

Two ways:

1.  The CFLs on the circuit are already off!  Therefore they can't interfere with the X10 unit.

2.  There is not a problem with the X10 devices turning off randomly.  If there were, this would be proof that the devices were receiving interference and perceiving it as a command signal.  No, the problem only happens when the CFLs are off.

The problem is a little bit stranger in origin than mere interference, and is a problem inherent in the way that X10 modules and CFLs interact.

Whoever designed the X10 circuit was not stupid.  They realized that people wouldn't use them if there was no way to manually override the device.  If you turn on a lamp attached to an X10 controller, all you need to do is to turn it on twice - the second time the light will turn on normally.

In order to know whether the light is "on" or "off", the X10 controller draws a small amount of power from the circuit it's attached to - about 2 watts - and feeds a small amount of voltage to the light that it's controlling.  With normal incandescent lamps, this works fine - the light stays dark.  But with CFLs this becomes a problem.

Fluorescent lights work off a ballast - a device inside the lamp that, for want of a correct explanation, stores up electricity until it has enough and then fires off a jolt to start the lamp.  Because the X10 controller is feeding the lamp low (not none) voltage, the ballast will sit there for a while - sometimes hours - but it will eventually store up enough voltage and try to fire the lamp off.

How do I know this to be the case?

One strange, hot night last week, I woke.  I went to my living room, where I was treated to a magical sight.

Every one of my CFLs was flashing, very dimly.  They resembled nothing more than giant fireflies - every few seconds, one would flash.  I was puzzled about this and unscrewed one of the bulbs; much to my shock every one on the circuit popped on, nearly resulting in a dropped bulb.  I had removed one of the drains on that circuit, increasing the available voltage to the rest, and that was enough to spark them all on and to trip the X10 controller into "thinking" that someone had manually turned them on.

How to fix this?

As it turns out, there is a really simply way to fix this.  It is inelegant but it works very nicely.

Grab an unused "wall-wart" transformer and plug it into the same circuit as the CFL being controlled by the X-10 unit:

The "wall wart" pulls just enough power to prevent electricity from building up in the CFL's ballast, preventing it from firing off and turning on everything connected to the X-10 controller.

Problem solved.

Site design and administration by Apocalypse Web Designs, NLC, a wholly-owned subsidiary of kRAPcO Industries, NLC
Love it? Like it? Approve? Tell me! 
all content © 2011