How do you pair a device that is permanently installed and more than 30 feet away from the hub? I would think this is a common problem but I have not seen a solution. Perhaps I just overlooked it. Can the USB stick be placed in a laptop for pairing? What else do you do if you cannot bring the zwave device to the hub?
If it is a Z wave plus device, it will likely pair in place using NWI (networkwide inclusion).
Otherwise people will bring the hub to the device using an extra long ethernet cord or you can plug a Wi-Fi booster that has an ethernet port on the side into the room where the device is and plug the hub into that.
Here's just one example:
Thanks JD. I was hoping there was a better solution built into the design of the product. One of my devices is outside and dragging an Ethernet cable out to it, or buying additional hardware to pair it is not ideal. Certainly not impossible.
Brand and model of the outside device?
@kmheid I have had the same problem and put the Hubitat on my tablet and have had success adding Z wave products without dragging my hub around. I would try that if you have a tablet.
You mean you plug the hub’s z-wave/zigbee usb dongle into your tablet?
I have paired many z-wave and Zigbee devices without being close to the hub. As long you has a solid mesh network, the devices should pair. The one exception is z-wave devices that require Secure pairing (like locks).
Have you just tried it?
I have a Intermatic z-wave device thats connected to my pool pump. You have to be a foot away from it to include it. So I drag a network cable and a power cord all the way out there. I wish there was an easier way but I haven't found one.
This is true for initial network join of an out-of-range Z-Wave Plus device when there are in-range Z-Wave Plus repeaters joined to the controller. However an older non Z-Wave plus device still needs to be in radio range of the hub when joined, at least based on a few things I've read (for example, this: Understanding Z-Wave Networks, Nodes & Devices | Vesternet ) specifically the section 'Challenges in Typical Network Configurations' under 'Z-Wave Network with One Static Controller'. It also is consistent with my efforts to join a bunch of non-plus GE in-wall dimmers, some of which were on the hairy edge of being in range (now I know where the 'hit the button 10 times' folklore comes from) and a few outliers which required a long ethernet cable connected to my Hubitat.
Understood. I have some very recent experience with respect to this topic...
When I initially starting migrating GE Z-Wave (non plus) switches to Hubitat back in February, I had a similar experience as to what you're describing. Hub placement seemed to make a pretty big difference. All of my Z-Wave devices at that time were old Z-Wave, not Z-Wave Plus. It did seem to get easier to pair these old Z-Wave devices as more and more of them were added (about 7 in total.)
A few weeks ago, I added a few more GE Z-Wave (non Plus) switches to my house, about 60 feet away from the Hubitat hub (opposite end and on a different floor.) These were some left-over switches from an old bulk purchase that I had never deployed. These two classic z-wave switches paired almost instantly, with no grief whatsoever. I can only surmise that the existing Z-Wave mesh aided in the pairing process.
Today, I just finished installing 4 brand new Ge Z-Wave Plus switches. These are located at the opposite end of my house from the Hubitat hub ( ~40ft away, on the same floor, through about 5-6 walls.) Every one of them paired within ~3 seconds and the correct driver was auto-assigned (thanks @mike.maxwell 
) I suppose it is possible that these all direct-paired with the hub. If so, Z-Wave Plus For The Win! However, there are quite a few older Z-Wave GE switches that are much closer to the location where I installed the new ones. Perhaps these were used to aid in pairing the switches?
Just wanted to share my Z-Wave experience.
1 Like
Interesting. For sure there's a lot of anecdotal information out there re: pairing that sometimes leads to frustration since everyone's environment is different, especially with regard to the propagation of RF signals in the home. It certainly makes sense to try it in place first, and resort to relocating the hub if that doesn't work out.
My last GE switch to migrate was only 23 feet (as the crow flies, if a crow could fly in a slanted trajectory through a hardwood floor and at least a dozen wall studs) from my hub and was one of those that required digging out my 25' ethernet cable. I was sure that this one would pair in place, since it did so easily with my ST hub which was located in approximately the same place as the Hubitat that replaced it. But it wouldn't; go figure. Not all Z-Wave antennas are created equal.
I can understand all the problems that people run into when they turn off their ST hub, drop in a Hubitat in the exact same spot and try a general exclusion with an older Z-Wave device. It may fail, and of course then it will never subsequently join. And if they do get it to join (by moving the hub or using a long cable); they may forget to do a followup Z-Wave repair-- which means that though joined, the hub (when back in its permanent location) won't be able to do anything useful with the device till it can route properly on its own through closer Z-Wave routers,
Complicating that is the scarcity of documentation (especially with the GE/Jasco devices, which just refer you to your hub's documentation). Turning to the internet for advice is kind of hit or miss since some of the undocumented methods that are randomly successful may be lucky coincidences. I was able to find a post from someone who actually contacted Jasco support that confirmed that the air gap isn't used in the exclusion process, and a single tap, not ten as I've often read (of the on or off button, doesn't matter) is enough to initiate the exclusion (or join, depending on the state of your hub). Why they don't put that info on the flyer in the package escapes me.
It's also worth noting that the nominal range is usually quoted in the most benign conditions (likely open air) and the attenuation of common building materials (to say nothing of furniture or appliances that may also be in the way) makes this spec pretty much useless. FWIW, I found a fascinating chart in this document (granted, its Zigbee, not Z-Wave so not the same, but interesting nonetheless) which speaks to that issue:
https://www.control4.com/docs/product/zigbee/best-practices/english/revision/A/zigbee-best-practices-rev-a.pdf
1 Like
Local architecture can have weird effects on how signal bounces around the property, Making it difficult to draw conclusions just from observing the end result. But with Z wave, it's super simple:
In order to be included into a Z-Wave network, a device must either be in range of the controller or support Network Wide Inclusion.
The strength of the mesh is irrelevant. Any other repeaters are irrelevant.
The new device has to be able to reach either a controller or it has to support NWI (networkwide inclusion), and there has to be a pathway from the device to a controller through repeaters that also support NWI.
I understand that some people think they have observed other behavior, but I guarantee you if you mapped that network you would find that the issue was either a local signal bounce or a local signal block.
The reason I know this is because it is up to the primary controller to assign the device ID to the device at the time of joining. Repeaters that don't support NWI are not capable of passing along the new device ID. So either the new device talked directly to the controller, or it routed its join request through a pathway of repeaters that support NWI.
4 Likes
Yeah, and don't forget mirrors. Mirrors bounce signal, leading to some really weird results. 
1 Like
Check the conformance statement on the official Z wave alliance website:
https://products.z-wavealliance.org
NWI was available beginning with Z wave version 4.5
I believe all zwave plus repeaters will, but I don't know for sure if exceptions were allowed.
1 Like
Based on that site’s information, it appears my older GE 12722 switches are z-wave v4.55. So, they probably do support NWI. Which in turn supports my anecdotal experience with these older switches.
2 Likes
