A capacitor is a low impedance to high frequencies and that added capacitance along with the inductance of the transformer usually increases the transfer of X-10 signal from one leg to the other. It also changes the phase relationship of the signal but that is usually of little consequence. Commercially available passive coupling devices (like the ACT p/n CP000 or the evil Leviton p/n 6299) do far more than simply use a capacitor. Both utilize twin parallel tuned circuits for 1:1 signal transfer and for safe separation of the high voltages to which they are often connected.
Most DIY’ers find that passive coupling is sufficient for the majority of X-10 installations. Many of the them will suggest using a 0.1mfd capacitor across the legs to allow for signal travel between the two phases. But remember that the original transmitter only sends out about 1/3 watt of signal power. As long as most of it is on one leg, then signal strength remains high on that leg. As soon as a low impedance pathway in introduced, some of the signal will be reduced on the source leg. Even though it doesn’t make my company a penny, I have to admit that most of the time it works just fine.
In more complex installations, a capacitor across the two legs will both help and hurt. DIY’ers have long used this method of providing a low impedance pathway for X-10 signal to travel from one leg to the other and it is especially beneficial in situations where only phase-to-neutral receivers are used, and there is plenty of original signal to go around, it just needs a way to get from the source leg to the target leg.
It does, however, have 3 drawbacks. First, a capacitor is non-selective. It will also pass noise. (Admittedly, a minor consideration in 99% of installations.) Secondly, installing a capacitor into your breaker panel is contrary to National Electric Code. (Also, a minor consideration in 99% of installations. We know that if properly rated and properly installed, its safe. NEC inspectors are often overly rigid.)
Thirdly (and most important), because it passes the exact same voltage reference, phase-to-phase differential signal potential is occasionally canceled out completely. Imagine a battery connected to point “A” and point “N”. Then a wire is connected between point “A” and point “B”. Then a voltmeter is used to measure the potential between point “A” and point “B” and it (of course) measures zero volts.
Since many people have asked about a “plug-in version, we might as well discuss that, too. As far as why we (ACT) or Leviton or X-10 do not make a coupler that plugs into a 240v outlet, it is simply a question of economics. Every house that has X-10, also has a breaker panel. However, they may ‘not’ have a 240v outlet. There is also the question of ‘which’ style to use. Depending on the current draw, there are several different styles of 240v outlets. Then there is the cost of the receptacle/plug assembly (with the coupler built inside it). That would more than triple the cost of the phase coupler and not save you anything. Finally, there is a technical issue. The coupler works best if it is installed at the ‘high current’ point in your system. That means the breaker panel.
Imagine an X-10 signal that originates on one leg (or ‘phase’, if you like that term better) and arrives at the main hub of all the other circuits. If the coupler were installed at the end of another circuit, that signal would have to travel downstream to that point, then jump over to the other leg, then travel back to the panel before it would be available to the other circuits. This is very inefficient and detrimentally effects the signal strength.
While many of us tell home owners to “turn on your stove” to see if the installation of a passive coupler will help, that test is not absolutely conclusive. Sometimes the coupling afforded by your stove or clothes dryer is marginal, but it illustrates that a passive coupler, ‘properly installed’, will have an even better effect. Sometimes the stove trick does not appear to help, but if a passive coupler were to be installed anyway, the homeowner would see some significant improvement (but not as much as he would with a coupler/repeater). It is not a matter of “yes/no”, it is a matter of degrees.