OK - that’s a big difference! When you posted your original post, I guess I wasn’t paying close enough attention to the pic you used - I assumed you had 14/3 running between the gangboxes! So…now that I know that, let’s take another stab at this!
Another epiphany - after cruising the internet and looking more closely at the diagrams provided by Aeotec, I’ve determined that the 3-way switch mode depends on a “return to state”. If you look at their diagram, you see they have both poles (ie travelers) bridged together and entering their switch terminal. So apparently, they are relying on detecting the momentary interruption in voltage for operation. So, yeah, your older switches might be affecting this, although I think this is a stupid design decision personally. What you did was exactly right (at least I think!) by bridging the traveler terminals.
Now, all that being said, I think the diagram below would be an “approved” Aeotec configuration. But maybe we need to let it figure out what it is dealing with. So, unless you’re just tired of dealing with this, try the wiring below but this time, don’t set the switch manually - try the automatic mode. Maybe if you do that it will realize how the switch operates and make some internal adjustments or something. From the manual: “Quick press the Action Button 4 times, which will activate the automatic identification mode for external switch S1, you will see the blue LED fast blinks.” I’m assuming you would then need to flip the switch(es) at least once. Worth a shot…
That’s how I have it wired, except the output of the switch in Gangbox 1 is going straight into S1 and the switch in Gangbox 2 is wired to Line and the output is going to S2.
So the last time I tried to put it in auto-sense mode, I don’t think it sensed the switches properly. It set parameters 120 and 121 to 0, and I had to use ZWT to set them back to 2. It probably just doesn’t like the switches.
Probably not going to take it back out unless I can’t figure out this last issue and need to re-pair it, or we determine that my current wiring isn’t close enough to your ideal posted above. Alexa will send the On/Off command to the switch, and the lights follow correctly, but the Z-Wave status is never updated. So the phone app always shows it as being On. What the heck is going on with this thing >.<
This is quite interesting. I think the old switches are not the problem but most likely the floating voltage during transient. Instead of having the 2 switches going into S1 and s2. Try @doncaruana second diagram. Maybe that will lower the floating voltage to a workable level.
If I get bored tomorrow, I might give it a try. In the meantime, the switch started reporting status after resetting it at the breaker, and I just spent a few hours making WebCore pistons. So I’m pretty tired and satisfied at the moment.
I get it! That’s a lot of work for something that should have been so simple! FWIW, I opened a ticket to Aeotec about this to validate some of the above at the beginning and never heard back from them.
I’m convinced that my last diagram is the best shot at making this work, I just don’t know if it will ever work, even if you try autosense again. The only thing is maybe trying it in this config, with everything going to one switch will help. But clearly no guarantees here.
Like I said, I think the premise of detecting a voltage drop in the time it takes a switch to go from one pole to the other is completely stupid. I don’t know what Aeotec was thinking.
So, if you ever feel like trying it again @kev2, please report back and let us know. Otherwise, I hope I at least helped a little!
Also, I think the way it’s currently hooked up is exactly the same as what’s shown in the manual for 3-way switches, including the optional switches. So it should be “Aeotec approved” too. But let me know if they get back to you with anything different!
Jesus, I just left the switch configured in 3-way switch mode. It’s annoying to have to really flick the switches fast, but when done correctly you get a normal 3-way toggle. If you leave it in 2-way switch mode and turn the switch off virtually like via the SmartThings app, the lights get out of sync with the physical switches. So for instance, in 2-way switch mode, if I turn the lights off with the switch, then on via SmartThings, the next time I flip the physical switch the lights will try to turn on again instead of turning off (as you would expect with a 3-way switch configuration).
Like I said, I think the premise of detecting a voltage drop in the time it takes a switch to go from one pole to the other is completely stupid. I don’t know what Aeotec was thinking.
This was one of my primary issues with the these switches as well. Great news, however! At the end of May, Aeotec made v2.01 firmware images available for these switches (ZW116 / ZW139). The linked official documentation is still in need of some improvement - but I believe these to be issues only with the write-up, not with the implementation. After having had put my implementation efforts effectively on-hold over the past half-year waiting for something to improve here - dare I say, things are now working perfectly, and I’m excited and looking forward to getting some more Nano switches.
Key to this improvement is the following line from the firmware update page:
Added the configuration 0xF7 parameter to configure the external two-wire switch mode
… expanded with some details from another support ticket I had opened looking for some clarifications around this:
This is now an adjustable setting which the default sets both S1/S2 on how they would react with a 2-state switch via V2.0 and later
Significantly, the default configuration value for 0xF7 here is “0” - which in my experience is the “new” functionality. Setting this to 1 (for S1) or 3 (for S1 and S2) would appear to restore the “old” functionality, which causes the “out of sync with the physical switches” concern (@kev2).
To be clear - it now doesn’t matter if I have a single 2-way switch, or a series of 3-way/4-way switches. I will have everything configured as a 2-way switch: 0x78 / 120 = 1 for S1 (or 0x79 / 121 = 1 for S2, if using the second switch input) - and keeping the new 0xF7 / 247 as 0. I now have no more “out of sync” issues, and I have 100% responsiveness (no more trying to detect a voltage drop between switching of poles).
Ok So those are the software settings, can you diagram how you have it setup physically? I’m having trouble figuring out exactly what I need to do for a setup of 3 way>4way>3way>light. If I’m clear on what the ideal setup would be it would be the incoming hot to L and IN and S1 to what used to be the entrance of hot to the circuit on the first switch, then somehow connecting the OUT to the last leg of hot to the fixtures.
That is unless you have a better wiring pattern than aeotec…plz say you do.
So basically here’s what I’ve found. You put the nano at the end of the series of switches, the out(to load originally) side of the final circuit FROM the switch goes to S1 on the nano, the neutral to N which is no big deal, The wire that originally was connected to the out of the switch (fixture on the other end) goes to the Out on the nano, but heres the bugger, you need to get the hot from the original box, before it ever goes into a switch, all the way to the nano and connect it to the L and the IN on the nano. So unless adding a new wire between the first and last box is possible, or you have unused wires to replicate this all the way through (unlikely) a GE or Leviton in wall switch and add on switches is far more easily installed and more economical if its being professionally installed. The nano is great when the circuit is a 2 way, but if you have a 3 way (2 or more switches that control the same device) there are better options.
A related question - I’m about to dive into installing an Aeotec Nano Dimmer ZW111 on a stairway which has 3 switches (4-way configuration). The wiring for this configuration is clearly shown in the Aeotec article on 3-way/4-way/5-way switch methods.
However the nano dimmer limits max current to 1.2A. I’m wondering if the same wiring diagrams could be used to wire the Nano Switches ZW116/ZW139 to provide additional load capacity up to 15A? Is the COM terminal on the ZW111 dimmer effectively the same as the IN terminal on the ZW116/ZW139 (i.e. runs at 120/240V)?
nano dimmer is 1.2 amp, and no, once its amp load has been reduced you cant just add another anything down the line and increase the rating since power is still flowing through the first device that’s limited to 1.2 amp. Your lighting circuit should never even approach 15 Amp anyway in total draw.
That’s also a new drawing and hasn’t been available that long, it involves a lot of splicing and reworking the original circuit so that the nano is actually technically the last device in the circuit before the load and the switches just interrupt its flow.
I’ve read through this entire thread and, unfortunately cannot use the final diagram where line and load are shown in gangbox 2. The diagram in this message where you pass line to gangbox 2 via the black traveler from gangbox 1 would be the way I’d have to wire my system as I don’t have line voltage in gangbox 2. Did you ever determine if the wiring in this diagram worked properly or not?
I’m trying to install one of these at my fixture (regular switch, no 3-way). From you diagram, it’s hard to tell, are L, In and S1 are jumpered together?
I came across this post last night, as I was trying to figure out how to wire the Nano Switch into an existing 3-way set up. I just used this wiring diagram as a guide, and I can confirm that this method works perfectly. I did have to quick press the action button 4 times, and once I had done that, the Nano Switch recognized both switches correctly. Thank you so much for posting this, and sorry for the necro-post, but hopefully this confirmation can help someone else.
