Please bear with this rather long post. I would like to present a method for wiring a 3-way X10 circuit which I have not seen before, and would appreciate any comments on it. I will be referring to Phil Kingery's series of articles titled "Which One Should I Use", Part VI, which can be found at . I will try to present some diagrams of my circuits, with apologies to Phil for the ASCII art :-).

First, I have three 3-way circuits in my home which I would like to convert to X10. I have purchased all the switches, and have opened all the switch boxes and inspected the wiring. The first of my three 3-way circuits is a rather straight-forward 3-way circuit with the load in the middle of the run. It should present no problems.

The other two 3-way circuits are more interesting. To begin with, the switch-box wiring for these two circuits is as shown here:



_Each_ of the switch-boxes has _only_ three wires, and the load is physically between the two switches. So I conclude that they are what Phil's article calls a "Standard 3-way with the load at the beginning of the run and the switches split." They would then look like this:



So far, so good, or so bad, as the case may be, because the article says "Unlike most of the previous examples, converting this type of 3-way _will_ require significant wiring changes in the load j-box." It suggests re-arranging the wiring thusly:



So, I climb up and take out the light bulbs, and remove the fixture, and I find that each of the lamp j-boxes has _only two wires_, black and white!



Hmmm... what's going on here?

Since each switch box has only three wires (black, white, red), the circuits _must_ be as stated above. But the travelers _do not_ pass through the load j-box! These circuits are in my basement, and I can only guess that there must be some hidden j-box where the cables from the load and each of the switches meet. But as my basement is entirely finished, there is literally no way I can find out without tearing open the walls :-(!

The next step is to figure out which of the two switch boxes receives the live feed, and which feeds the load. But even if I figure that out, how am I going to re-arrange the wiring in the hidden j-boxes?

So I go and do the first 3-way circuit, the one with the load in the middle of the run. All is working, and as it's getting rather late, I knock off for the day and hop in the shower. While showering, I'm thinking of those two troublesome 3-ways, and Phil's article. And suddenly the idea comes to me: "Why can't I just wire it like this?":



The idea is that the hot wire from the panel will feed the slave switch, and from there it will _also_ feed the master switch via one of the travelers. The second traveler will be the slave's control signal to the master. So the master receives live and control over the travelers, and then feeds the load. Thus, no re-wiring required in the load j-box!

Now, I've been in electronics for many years, but I am _not_ an electrician. So if I've overlooked some incredibly obvious reason that this won't work, or that it isn't code, or whatever, someone please let me know! But if anyone can confirm that this ought to work, please let me know that, too :-)! Finally, how can I identify which of the two switch boxes receives the live feed, and which feeds the load?

This message was edited by PChek on 08/18/03 05:11.

PC, you "discovered" the simplicity of X-10, that it is designed to replace standard switching.

If you're talking about typical inexpensive X-10 switches, where the slave has two wires of the same color (blue?) and one of the other (red?), and the master has a red, a blue, and a black, then it doesn't matter which is the hot and which is the load switch.

The two blue wires are internally tied together inside the slave switch, by the way, and it works by simply connecting the red to the blues, meaning that it parallels the contacts inside the master. It's doen that way to make it easier for people to replace switches.

However, if you do need to know which switch is which, the hot one is the one whose common (dark) terminal tests hot to ground when the load is off. The other common will test hot all the time. (If you have no ground at the switches, use a 3-wire extension cord.)

So, your solution is the regular X-10 method of switch replacement. The main concern is if you're using X-10 switches that require a neutral. As for the hidden junction box, it should legally never be hidden. Are there no blank plates anywhere?

Larry
www.fineelectricco.com

Larry, thanks for the response.


The switches I have are WS12A dimmer masters, XPS3 appliance masters, and WS14A slaves. The WS12A has black, blue, and red wires, but the other two have screw terminals rather than wires. The slave does have two labeled 'Live', which parallels what you said about the blue wires.

But, are you saying that I can put the master in either switch box, and the slave in either switch box?


Ok, but I'm a bit confused here. Or maybe it's my terminology that's wrong. I would have thought that the one that's hot, would test hot all the time? And the other common test hot only when the load is on, since it feeds the load, which returns to neutral?


Ok, the only reason I thought it was different, is because of Phil's article mentioned above, which does say that a "standard 3-way with the load at the beginning of the run and the switches split", "_will_ require significant wiring changes in the load j-box."


No, not that I've found :-(. I suppose that it was originally in the open, and then someone finished the basement and covered it up?

Yes. The 'live' terminals will pass the current either way, whether it's the always-hot wire or the switched-hot wire.


My bad. It was early (!). The always-hot will test hot all the time, and the switched-hot will test hot only when the load is on.


Not with switches that don't require neutrals. The re-wiring he shows can be done, if you need the extra wire in Fig. 13 for something else, but it's not required. Standard X-10 wiring will suffice. The only difference is that current will flow to the (again, using Fig. 13) slave switch and back again, which is unnecessary, but would allow easier replacing of the X-10s with standard switches in the future by someone else.


It happens all the time.

Larry
www.fineelectricco.com


This message was edited by Larry Fine on 08/19/03 11:34.

Thanks, Larry. I'm all sorted now :-). It took a bit for me to wrap my mind around being able to put either switch in either box :-). I see now how that can work, assuming that the slave switch can trigger the master by momentarily switching the control lead to either hot _or_ to neutral!

Anyway, I wired them in today as per the last diagram in my original post above, using your advice to measure for the always-hot and switched-hot, and all's working well! As you said, doing it this way will make it easier to switch back to regular switches, either for someone else, or for me, if I move and want to take my system with me :-). And, I didn't have to tear open my walls to find the hidden j-boxes :-) Yeah!

No, not to the neutral, only to the line wire. Remember, the entire switching arrangement is in only the hot wire, never the neutral. This is true for all line-voltage switching.

The switches don't know or care which direction the feed is or which direction the load is, as long as there is voltage across it and at least a trickle current through the load.

Very light loads, such as night-lights and other below-25-watt lamps often don't supply the X-10 dimmer with enough current to charge up its internal power supply.

Larry
www.fineelectricco.com

Ahhh... [Reaches down to retrieve fallen penny :-)]. It finally dawns--when the load is off, it is the _trickle_ current which supplies the slave with the signal to toggle the master [when the switches are in the opposite j-boxes from the way I placed them in the last diagram above] :-)!