GE Z-Wave® Dimmer Switch (3-Way Kit)

Other Devices & Integrations Lighting & Load Control
101

@oTi@:

I have traced and traced and traced; but certainly could have missed something.

In the diagram I posted,

I believe the far “left” switch (front door) to be where the HOT comes in (in fact, that part I do have verified 100%); it then has a set of wires going to the “middle” switch (about 15’ feet away, ‘garage door’)). This is a red/green/ground/neutral set, AFAIK. This was a 3-way switch.

The middle switch is where the old 4-way switch was located, the only thing in that box was the prior 4 way switch, and two sets of cable going in/out - each containing 1 green/1 red/1 white/1 bare/ground. The two whites are wire-nutted together in the back of the box.

The “right” switch (per the diagram) is “upstairs” in a box with 2 other switches. The prior 3 way switch there had a set of cable coming in, presumed to be from “garage door”/middle switch. The wires are red/green/ground/neutral. There is then a single black wire which I believe goes to LOAD.

Again, I have tested all the switches in multiple locations and have had almost the exact same result. 80%. I have also isolated it right now to where I have basically jumped everything (with exception of RED/traveller in my case) from “front door”, to “garage door”, to “upstairs”. Upstairs is where I have the master (NO auxes connected!). As SOON as I connect to traveller (red) the lights only come up to 80% again, even if traveller is capped off at the next known/assumed junction point (“garage door”).

I am 99% this is not a 5 way configuration as i have not seen any other wires that would leave me to believe so, and I don’t think there are any other switches for this circuit that I have discovered.

PS - In my “final” configuration, where should my master “truly” be. At LINE or closer to LOAD? Where LINE comes in (“front door”), as mentioned, “LOAD” only goes to the “garage” (old 4 way) switch box. “Upstairs” has LOAD but no true “LINE” of its own, only if I tie the wires together at “garage”. The only thing about “UPSTAIRS” is it does share a box with two other circuits, which both DO have “LINE”. The difference being, they both are on a different breaker than the “front door” switch so I would assume I do NOT want to take “LINE” for the master switch from a different circuit? (I hope this makes sense in the way I am typing it)

Your response was/is greatly appreciated. I am really hoping to get this sorted out as it’s driving me nuts (and my switch plates are all torn apart!)

102

The colors in your description don’t entirely match up. You say red/green/neutral/ground is leaving the left switch, yet you got red/green/black/ground coming in on the 4-way?

The primary switch must be connected to its own LINE and the corresponding NEUTRAL, and yes you can’t just grab a LINE from somewhere (as also mentioned earlier in this thread).

Anyway, just going with the diagram that you referred to, whether it is your layout or not, a lot can be done. Your primary switch can go wherever you want. Do you have a preference? Perhaps closest to Vera / optimal Z-Wave reach, or wherever you think the blue LED is most useful.

In essence, what is black in the diagram is going to be used as LINE and then where the primary switch is, it becomes LOAD. Red is ‘SIGNAL’ and connects to yellow on the GE’s. White stays NEUTRAL. (And of course everything is connected to ground.)

103

[quote=“oTi@, post:102, topic:167413”]The colors in your description don’t entirely match up. You say red/green/neutral/ground is leaving the left switch, yet you got red/green/black/ground coming in on the 4-way?

The primary switch must be connected to its own LINE and the corresponding NEUTRAL, and yes you can’t just grab a LINE from somewhere (as also mentioned earlier in this thread).

Anyway, just going with the diagram that you referred to, whether it is your layout or not, a lot can be done. Your primary switch can go wherever you want. Do you have a preference? Perhaps closest to Vera / optimal Z-Wave reach, or wherever you think the blue LED is most useful.

In essence, what is black in the diagram is going to be used as LINE and then where the primary switch is, it becomes LOAD. Red is ‘SIGNAL’ and connects to yellow on the GE’s. White stays NEUTRAL. (And of course everything is connected to ground.)[/quote]

My apologies, the sentence regarding the middle box was meant to read ‘red/green/white/ground’; I updated my original post as not to confuse anyone else who reads it first!

That being said, I don’t care where the master switch resides, as long as it works! :slight_smile: As mentioned, I have tried it in all three locations and gotten the exact same functionality. :frowning: (80%).

The only thing I forgot about was ,there is some kind of weird CFL fixture in the chain of lights as well. I have the bulb removed but the transformer is still in the fixture. Is there ANYTHING the CFL could be doing that “upsets” the 3/4 way configuration? (As mentioned, if I wire everything up to be just one master and no AUX’es, the lights do come up to 100%)

Thanks again for your time. Please let me know if there’s any other details I can provide.

104

Went through tonight and traced all switch boxes; the cabling is indeed run as I mentioned (“front door”, → “garage door”, “stairs”)…

I removed the CFL fixture which was hardwired in; this had no effect except that the breaker no longer buzzes in the garage (which is good!)

Two things I did take notice of:

If I test the voltage on the RED wire (from the red/green/white/ground pair) at any switch box, with the LOAD off it (RED) reads 0 volts. As soon as I tie together the the greens to the LOAD or HOT (depending on the wiring setup I am testing! Not together!); the RED shows a voltage of > 0 v (it appears to be approximately .5 volts). If I tie together the wires in the “front door” box, put the “master” switch in the “garage” box, I see a steady 0.5V on the RED traveller coming from “master”, even though I have it CAPPED at “master”. If I test the RED going FROM “garage” box and the switch is OFF, it reads 0 volts. if the switch is ON it reads ~ 0.5 volts (it lowers as the switch is dimmed). Does this make sense as I am explaining it? If so, is it possible that somehow the 120 on the GREEN is “leaking” a slight amount of voltage over to the RED (traveler) which is confusing the GE master switch?

105

OK. So you can have your primary switch in any of the three boxes.

Can you try putting it in the right box. Test it. (Sounds like you have done this.) And then see if you can hook up an aux switch in that box (candycane to neutral; yellow directly to the yellow on the primary; groud)? Test it. Then hook up the second aux switch the same way. Test it.

This would at least tell you that all your switches are good, and the problem really is with the red wire.

Have you connected your aux switches to neutral, or to line? It appears that folks have better results connecting it to neutral.

106

It might be the dimmable CFL that is causing 80% issues (I don’t know for sure, but most of these cheaper dimmers are listed to work with incandescent fixtures, even though some people have had success with CFLs and LEDs with some quirks e.g. LEDs don’t dim all the way down). It would be worth trying it with an incandescent bulb in place, just to remove that as a possible problem.

I did a sketch showing the two ways you could wire this either with the “candycane” wire going to neutral or to hot for the slaves. Let me know if you see anything wrong in the drawings.

107

What I gathered is that the CFL is already out of the picture and the issue remains. And the issue is triggered by connecting just the red wire (no aux switch attached to it) to the primary switch.

did a sketch showing the two ways you could wire this either with the "candycane" wire going to neutral or to hot for the slaves.
Candycane to neutral seems preferred. It also allows the primary switch to be put anywhere. (With candycane to line, the primary switch can only be on the right, as per your drawing.)
108

[quote=“oTi@, post:107, topic:167413”][quote=“shady, post:106, topic:167413”]

did a sketch showing the two ways you could wire this either with the “candycane” wire going to neutral or to hot for the slaves.

Candycane to neutral seems preferred. It also allows the primary switch to be put anywhere. (With candycane to line, the primary switch can only be on the right, as per your drawing.)[/quote][/quote]

I think he should try it both ways to see what works best for him. I have the “candycane” wire from the slave attached to hot as per the instructions and mine seems to be functioning well (although not used that often). I guess the slave just needs enough voltage to generate a 0-5V signal and either the neutral or hot will supply that.

If everything is take apart at each switch location, I’d take the time to verify each wire using a continuity tester. Turn off power. Find a pair of wires in a romex that you think is going to another switchbox, check for continuity (should be none), find the same wires you suspect match on the other side then wirenut them together, now go back to first box and check for continutiy between the pair (wirenut should give you continuity between them). Now you know they connect between boxes (may be connected to other things, but with travellers this is a good way to find what box connects to what box).

FYI, this may not apply to you, but I thought I should mention it for people with phantom problems or mismatched wire colors. During new construction, nails occasionally go through wires and electricians don’t opt to re-pull new wires, but rather “make it work” by swapping wires around e.g. since my hot(black) is shorted to ground with a nail and my white is undamaged, I’ll make the black my neutral (since it is connected to the ground at the panel anyway, so the short doesn’t matter) and make my undamaged white my hot(black). I’m just saying that this stuff DOES happen (right or wrong) and using a continutiy tester you can check to see if any wires short to any others and use it to identify wires too.

109

I cannot stress this enough but I genuinely appreciate the help from both of you!

The diagram is (two) of the scenarios I tried. Still had the 80% issue. For the sake of clarification, CFL is totally removed.

I did a continuity test on the wires last night, all came back as expected, even polarity.

Regarding the potential nail intrusion, I coiuld see that being a possibility except the fact that the red gets the 0.5v on two different bundles/boxes, so I don’t think both wires being damaged is likely?

I have not tried wiring the two switches ‘together’ yet but will try tonight or tomorrow night.

Fyi, I have tested the switches (aux) with common and ground but got the same thing, partial dim. The only difference is when wired to HOT pushing dim OR bright on the secondary causes full brightness momentarily (while held down), whereas wiring to neutral only causes lights to go bright when pushing the top paddle on secondary… :frowning:

110

I have the aux hooked to line as well. BTW, I would guess maybe the signal wire is at half line voltage, so the aux switch sees ~60VAC regardless of whether it is connected to line or neutral.

[...] "make it work" by swapping wires around e.g. since my hot(black) is shorted to ground with a nail and my white is undamaged, I'll make the black my neutral (since it is connected to the ground at the panel anyway, so the short doesn't matter) and make my undamaged white my hot(black).[...]
:o So where do you presume your return current is going?
111

[quote=“oztiks, post:109, topic:167413”][…]
I have not tried wiring the two switches ‘together’ yet but will try tonight or tomorrow night.
[…][/quote]
OK; definitely try wiring the switches all together in one spot (i.e. no existing red wire in between) just to confirm that all your switches are working properly / nothing got fried in the process.

112

Yup. Tried it. They work tied together at any point as long as no (external) traveller is involved :frowning:

This is immensely frustrating. I have three 2-way configurations (one of which is a MASTER from a 4-way kit) working fine in my kitchen area; one 3 way configuration working between my 2nd and 3rd floors, but this 4 way switch doesn’t seem to want to play nice. I think it’s the only 4 way switch in the house, too, so I don’t have any other circuits to try it against.

One thing I did forget to mention but noticed tonight - on the “front door” (HOT) box, there are the 4 way circuit switch, and another switch which is the light at the front door. The front door light is a standard decora/rocker switch, but if I turn the light on or off, it makes the z-wave switch MOMENTARILY shut off (immediate, no dim) and then immediately resume to its previous dim point. This seems a tad strange…

113

Well, at least we know you’re switches are still good!

This is immensely frustrating.
No kidding.
[...]This seems a tad strange....
Just a tad. So your front light is powered by that same 2-wire romex coming in on the left? What kind of light is it? Have your tried your set-up with the front light fully disconnected?
114

I’ve had a few discussions with Jasco over the last few days regarding these switches with the 80% problem. I received some interesting information and testing results.

The biggest surprise to me is that these switches are not reliable using travelers greater than 20 feet. Yes twenty feet. I have a real hard time understanding this for the following reason:
If 20 feet is the max these can only be used in a 10x10 room because with 8’ ceilings that’s 4’ up, 4’ down and then 12’ to cross around the room. Seriously 20 feet? Forget about trying to use this in any setup where stairs are involved! I cut a 20’ piece of cat 5 to use as a traveler and guess what… it worked. WTF? My traveler in this one room is no more than 24’, crappy design.

Who the hell needs dual switch setup in a 10x10 room? I can touch two opposite walls of a 10x10 room with ONE step…

I have tried both version 2 and version 3 switches with the exact same results. Lights will only go to full power only when the switch is pressed. Which leads to the next problem… Obviously the signal can be picked up, but why is this system failing to maintain the full voltage to the load?

There is good news… the next version of these switches is not limited to 20’! And they’ve got rid of the lead-lines, and are switching over to terminal switches. Yeah on both counts!

115

@ Jeff… did Jasco mention any trade in program for the “older” switches??

116

@JeffD

I found the same to be true running just about anything as an independent traveller works. I tested mine running through 50’ of romex and . . . it works. It just will not work with the traveller being in the same core with the other lines. Best advice I found is to use the main 3-way dimmer switch as a single and throw away the Aux unless you have easy open access to run an independent traveller.

I am surprised to hear that some users have this switch working properly at all. I came to the same conclusion you stated about the 10’x10’ room which doesn’t really need a second switch. I would bet that GE/Jasco have a ton of customer returns on this switch model.

117

I can imagine that if it’s a 3-wire running just to the aux switch, it’s probably ok. But having a 3-wire with line driving the load (as in @oztiks’s case) in it is potentially problematic.

118

@Raskell, there may be something more to it. I remember when I was hooking up my first switch I wasn’t sure if both travelers could be connected in parallel, so I tried. The result was that it didn’t work similar to the way this other room works, 80% bright and only fully bright when the switch is pressed. I switched back to using one of the two and that worked fine in that room.

The other room different story, started with one, and that was working great then that switch started the flickering (with incandescents) and that setup hasn’t worked correctly since. I’m not sure if there’s something else on the circuit causing a problem,but maybe. There are other light fixtures on that circuit with other z-wave stuff and some other lights with CFLs installed.

It’s a PITA, that’s for sure, it shouldn’t be this hard. This is basic electronics and what, 120V isn’t enough power to send a signal from point A to point B if it’s over 20 feet?!?!? I’m half tempted to put a meter on the traveler to see what kinda of voltages they use for the commands. Now I’m having a tough time remembering if thicker or thinner wire has less resistance (I can see reasons why either would)

119

I have installed 3 of these so far in 3-way setups (and at least 20 others used as a single switch), in each case they worked the first time. Of course this house has new wiring from 2009, and neutrals in each box. In all cases, the “hot” wire comes in on a differnet jacketed NM cable, and I usually have both travellers availble, which I usually wire nut one off (usually black, and use red) and don’t even need all of the original wiring. I think it really depends on how your wiring was done, and how “clean” a job the electrician did on running the wiring in the walls as well (no nails through wire, shorts, etc)

In most of the cases, I have more than 20’ of wire between the “control lead” on the main switch and the AUX, in fact its more like 35-40’ in most cases.

120

[quote=“Jeff D, post:114, topic:167413”]I’ve had a few discussions with Jasco over the last few days regarding these switches with the 80% problem. I received some interesting information and testing results.

The biggest surprise to me is that these switches are not reliable using travelers greater than 20 feet. Yes twenty feet. I have a real hard time understanding this for the following reason:
If 20 feet is the max these can only be used in a 10x10 room because with 8’ ceilings that’s 4’ up, 4’ down and then 12’ to cross around the room. Seriously 20 feet? Forget about trying to use this in any setup where stairs are involved! I cut a 20’ piece of cat 5 to use as a traveler and guess what… it worked. WTF? My traveler in this one room is no more than 24’, crappy design.

Who the hell needs dual switch setup in a 10x10 room? I can touch two opposite walls of a 10x10 room with ONE step…

I have tried both version 2 and version 3 switches with the exact same results. Lights will only go to full power only when the switch is pressed. Which leads to the next problem… Obviously the signal can be picked up, but why is this system failing to maintain the full voltage to the load?

There is good news… the next version of these switches is not limited to 20’! And they’ve got rid of the lead-lines, and are switching over to terminal switches. Yeah on both counts![/quote]

Another newbie here, playing with the goodies from Radio Shack while awaiting delivery of my Vera 2.

Can you clarify what ‘versions’ of the switches means? I see on this page there are models 45606 (dimmer type) and 45609 (relay type) of the 3-way master (both working with the 45610 slave). There is a bundle pack (45609+45610) numbered 45614.

BUT, the model I bought yesterday when cleaning out the local Radio Shacks is a 45613 bundle, which included the 45607 (according to the manual, although the unit is marked model ZW3002-WCS with a date code of 0933 and a note of "Ver 2.0a) and the 45610.

So what does ‘next version’ mean in terms of model numbers or anything else we could distinguish? I’m about to order a dozen on/off switches and would like to get more reliable ones than the Radio Shack stock has proven.

I’ve just installed 3 sets of 3-ways and the wiring went fairly well with slaves&masters working. I did have to Delete one switch from the remote before it would Add.

BUT, I am having some flickering, some lights that won’t brighten fully, some lights that flash bright occasionally when the buttons are pressed, and one gang that has a LED light mixed with incandescents that has its own range of bizarre behaviors.