@JohnRob I saw the annotation and at first thought breaker as well. But after staring at the diagram for a while, I decided that didn’t make sense.
I’m still confused as to where the power comes in, and I’d like to know the connections at the fixture. Until we know the source of the constant hot and what is going on at the fixture, it’s going to be difficult (at least for me) to get this diagrammed correctly.
Let’s start with the basics. In a properly wired configuration, one of the two travelers at a switch has to be hot, so I don’t understand how you can have power to the hall switch but neither of the travelers are hot. On a dumb switch, the black screw is the common and the two brass screws are the travelers.
So let’s follow your diagram labeled Original, which looks ok to me. Working with dumb switches: I’m assuming that power comes into the circuit from the top left 2-wire labeled breaker. The white neutral goes to the light and the black hot is tied to the black going to the fixture and to the white going to the hall switch. So now we have a neutral at the switch, a black constant hot and a red.
At the hall switch there is a constant hot on the white, which may or may not have a piece of black tape on it. This white wire should connect connect to the black common terminal. So that hall switch sends power over either the black or the red. One of those two has to be hot at any given time, so if that’s not the case, I’m guessing the constant hot isn’t on the common terminal.
So now we have switched power coming into the front switch over either the red or the black. Depending on the switch position, one of those is connected to the common terminal, which has the red wire connected to it. So that red wire on the front switch will toggle hot or not as you toggle the hall switch. If that’s not working as described, you are still miswired.
So now we have switched power coming from the front switch on the red wire. That red wire goes to the light. So at the light, presumably, the red is connected to the bulb supplying the switched power and the white neutral is connected to the other lead. Presumably, the black is not connected and is just powering downstream devices.
Hope this helps. If you can get the dumb switches working properly and diagram accordingly we might be able to help with the Inovelli.
I started to redraw what I though the Inovelli diagram should be. But when I read the below I wasn’t clear about the test you made.
I tested the wires in the original config by photographing, marking, and disconnecting them after shutting off the breaker.
I turned on the breaker and measured AC voltages compared to ground.
At the “front” gang panel switch, all red and black wires were dead. At the remote hall panel the red and black wires were dead and the white wire was hot.
When you said “disconnecting them after shutting off the breaker” did you include the black coming in? If so I don’t have a good idea of what is going on hence the black wire to the com terminal.
I suggest you:
If this works the "dumb’ switch is wired in a way we can use it.
Note (1) you don’t want to feed line voltage through the white any more than you have to.
Lets test the light fixture. It seems the original drawing suggests the fixture has connects to it Neutral (white), Power (black) and switched power (Red). All three of these wires are not needed for the light. Perhaps the original electrician is feeding something else through that fixture. Logic would suggest the lamp is connected to the Red and Neutral (White).
With this info we should be able to wire your LZW dimmer.
John
OK got some time to do some probing. First the dumb switch.
So the only surprise here is the hot black for the entire panel was coming from the “fixture” romex bundle and not the “breaker” romex bundle. But all other assumptions were correct. So I am baffled about why this isn’t working properly. I THINK my circuit is the same one proposed by Inovelli…
Based on the above, I rewired the circuit to reflect a black-wire common at the hall dumb switch – tied at the other end of the black-wire in the front gang to the “fixture” load red-wire. And then connected the red and white wires at the hall to the dumb switch traveler connections – connected at the other end in the front-gang to the load and traveler on the inovelli. Then connected a black/hot wire from the front gang “hot bundle” to the inovelli line. Connected a neutral from the white/neutral bundle to the inovelli.
Switch behaves as before. Functions properly when the dumb switch is up. Shuts down when the dumb switch is down.
When the dumb switch is up and the inovelli is functioning but has the fixture turned off, I see 120v on the line, load, and traveler lines. When I use the inovelli to turn the lights on the line is 120v but both line and traveler lines are 108v. Which suggests they’re heavily loaded. But the two light fixtures only have bulbs consuming approximately 30 watts of power.
When the dumb switch is down and the inovelli stops functioning, the line voltage is 120v, the load voltage is 115v, and the traveler voltage is zero.
…and same voltages at the dumb switch…
So when the dumb switch connects the inovelli load to the the light-fixture load, the switch works.
But when the dumb switch connects the inovelli traveler to the light-fixture load, the switch shuts down.
There are a couple threads in this forum where users are describing the same symptoms where the smart switch only functions with the dumb switch in one position. Makes you wonder if it’s the Inovelli switch that is the issue.
Then…
Hi again,
Its difficult to understand how you rewired this time, but the black wire in the Hall should not be the common.
However knowing the power comes from the fixture does not change anything, your first diagram should have worked (good job 
)
The issue others have complained about appears to be with certain bulbs. If you can find a old incandescent bulb, put it in fixture and see what happens.
If the incandescent bulb works the same as the LED bulbs (I made the assumption your fixture has LED lamps) then your symptoms would seem to indicate one of the outputs of the dimmer are not working. I suspect it might be the “traveler”. To test this connect the dimmer directly to the fixture by connecting the fixture RED to the dimmer “Traveler”. The light should work in one of the two paddle directions (likely off).
If there is no light then you found the issue. If this is the case I apologize for not suggesting you test it earlier.
If the light does not come “on” then connect the fixture red to the dimmer load. It should come on. The reason for this step is that Inovelli will ask you to perform this test when you go back and complain the dimmer doesn’t work. Remember your first diagram is correct, it would not have damaged the dimmer.
Below is what I believe the dimmer is internally.
John
Was prepared to do more testing on the dimmer and traveler lines as John suggested. But first replaced all the LED’s in the fixture with incandescent bulbs.
The switches now behave much differently. Here’s what I observed with incandescent bulbs in the fixtures and wiring unchanged from my most recent post (which is electrically the same as the first post – just changed wire colors between the Inovelli and dumb switch):
So overall very weird, inconsistent behavior vs when LED bulbs are installed, and still not functioning correctly. Me thinks there is an issue with the Inovelli.
Add’l info: There are 7 bulbs total in the circuit: Two light fixtures. One has 4 bulbs and one has three. The LED bulbs are 4 watts each and rated as dimmable and the same bulbs work correctly with other Inovelli switches I have wired as single pole configs. The incandescent bulbs are 40 watts each.
Sorry I can’t help, the issue seems to be with the Inovelli. You need to contact them for the next step. Good luck,
John
Thanks for all your help. I’ve opened a support request with Inovelli and will report back when I have more information.
I did not see anywhere here that you set the configuration of the switch. Have you configured it so that it “understands” that there is a dumb switch? This sounds to me like the LZW believe that it is alone in the circuit. The little “wiring insert” tells how to do this manually using the config button on the switch, but if you can connect to a hub like Hubitat, it is much easier to set it via Z-Wave. You want to be sure it is set that it DOES have a neutral (I think the parameter is called “power” in the Hubitat driver) and that it DOES have a dumb-switch attached. Good luck…
following along. I’m having the same issue - LZW only works with dumb switch in one position. I did do the initial programming including setting the neutral and 3 way toggle setting.
Same problem. LZW works only when dumb switch is a specific position. Parameters 21 and 22 have been set.
Thanks burns and I agree the config should be noted. In my case I set parameters 21 and 22 at the switch. I also re-checked my hubitat drivers and they show:
I was having an issue similar to what you described initially. Lights would only work with the dumb switch in one position,not the other. I set parameter 22 at the switch and still no go. But then I set parameters 21 and 22 from my hub to 1, and that fixed my issue.
I was able to solve my problem the same way. It appears that when I included the dimmer in SmartThings it reset those parameters to 0. Once I changed parameters 21 and 22 in the SmartThings app the dimmer now works as expected in a 3-way setup with a dumb switch.
@ragsnova - i seem to be in the same situation as you are, but must be missing something in the app. do you mind elaborating on how you changed those parameters?
I’m having this issue too. Inovelli switch works only when dumb switch is in a certaqin position.
However, I check my parameter 21 and 22 settings in the IDE and they both say 1. I was hoping this was a eureka moment but I guess not. 
I’m using the SmartThings Classic app. Also, assuming you have installed the device handler in the SmartThings IDE. In the app, select the switch/dimmer from the Things list, then click the gear in the device screen. Scroll most of the way down the list to find the AC power type and switch type setting.
thanks @ragsnova, i had the wrong device handler in the IDE.
@Eric_Inovelli/@EricM_Inovelli - this page needs an update for the LZW dimmers.
for anyone else who needs it, they have it in github here.
