ah, that is a shame. So LR is a whole little universe by itself?
More or less.
The hub can talk zwave mesh AND zwave LR at the same time (so you can have both paried to the same hub), but zwave LR end devices only talk straight to the hub and nothing else.
1 Like
You mean LR devices doesn't mesh even with other LR devices? They are a star topology from the hub?
1 Like
Correct
3 Likes
Seems like a lot more than a couple in the last 9 months.
Last 9 months:
7.15.2.0 Pre-Certified GA released January 27, 2021
7.15.1.0 Pre-Certified GA released December 9, 2020
7.14.3.0 GA released October 14, 2020
7.14.2.0 GA released September 9, 2020
7.14.1.0 GA released July 29, 2020
7.13.9.0 released March 3, 2021
7.13.8.0 released October 28, 2020
7.13.7.0 released August 12, 2020
Prior to 9 months
7.13.6.0 released May 27, 2020
7.13.5.0 released April 29, 2020
7.13.4.0 released April 15, 2020
7.13.3.1 released March 27, 2020
7.13.3.0 released March 20, 2020
7.13.2.0 released February 21, 2020
7.13.1.0 released January 24, 2020
7.13.0.0 released December 13, 2019
1 Like
Your right. I was only looking at zwave 700, which is 7.14.x and above in the production releases, if I remember correctly.
7.15.x is where zwave lr was added.
Most .x.y releases have very few fixes. And no hub I know of actually implements all of them. Most do maybe update 1-2x/year unless there is a severe bug.
Those are not ZWare fixes for controllers. There are for the SDK for End Devices.
ZWave SDK for End Devices started with 7.0.0 for the 700 Series.
So 7.13.0-7.15.2 have been all End Device updates. That's the issue. The Hubitat can and should include LR IMO.
To be fair, the reason I want a hub is for LR support later. That's why the Hubitat over the SmartThings. Samsung is so far behind the curve.
They will - that really isn't in question (they have already said so).
Well, unless there is some unseen technical issue that makes it undoable on current hardware - but I think that is unlikely. They likely won't guarantee multi km range or anything with the current antenna implementation, but even as-is should be usable in normal home scenarios.
2 Likes
OK, so I have a question though: why would you not want your zwave devices to act as a mesh? What are the disadvantages of a mesh compared to every single device directly connecting with the hub?
My thought is complexity for everything involved.
Managing a mesh is a lot of work for all the devices involved.
Example: if the Hub goes down for a restart, traffic from the devices to the hub starts freaking out and trying all their neighbors. Quickly leading to a mesh storm.
Plus, all the devices have to be coded up to handle all the other traffic.
And, everything has to deal with devices on the path disappearing.
About the only benefit of a mesh is coverage. So, if a higher power radio can handle that, then why not?
3 Likes
In addition to all that @rob9 mentioned, here is my major consideration. Less points of failure. The most reliable integration I've had with Hubitat (and HA briefly before Hubitat) is/was the Lutron Caseta integration, which has a radial topology - no real mesh to speak off, yet remains extremely reliable.
3 Likes
Multi-pathing has its place in a resilient network. Especially if the interference is variable. That is why it is used very extensively in industrial settings - someone parks a big truck or a crane in your signal path, you need it to re-route automatically. Or in other industrial cases there is no direct signal path to the hub due to piping and other thick metal objects in between the sensor and detector. You HAVE TO use a repeating topology in those cases.
I think mesh is a little less useful in home environments where the interference is a lot more static. That said, there are potentially places where zwave lr is not going to work without careful placement due to the direct hub connection. Where there is a lot of interference between the device and the hub (sensor is next to an oven, in a freezer, etc), mesh network may (likely will) operate better.
No free lunch / situation that is better 100% of the time either way. But where a direct signal pth can work, less is more in terms of complexity.
Less repeating also (typically) reduces latency of the signal, thus speeding things up. Also, less repeating=less radio airtime used, freeing up that airtime for other messaging.
6 Likes
I didn't know Caseta was ZWave?
Lutron is not ZWave, but it is amazingly well integrated to Hubitat via the Lutron Integration to the Lutron SmartBridge PRO (via telnet.)
The 8-10 Pico's I have scattered around my home uses the Lutron Integration and it is fast and reliable, It even works across multiple hubs. I do not have any of Lutron's switches, dimmers or fan controls, but I believe what @aaiyar said was exactly the same.. it's an amazing integration and takes Lutron's high reliability and gives that to Hubitat.
If you can afford the price of their products and like the look, it's an amazing solution. If you have even more money, then kick up to the RadioRA series. 
3 Likes
It's worth noting the fact that Lutron invented this entire space of radio controlled dimmers and switches with the original RadioRA system (now called Classic, and discontinued for years). The technology they came up with was protected by a patent portfolio. I don't know the specifics, but some of this drove the design of Z-Wave and Zigbee which came years later. One example is that Lutron had a specific patent claim about a switch instantly reporting a change of state to the controller. Control 4 infringed this claim, Lutron sued and won. Control 4 had to pay a very high royalty on every switch and dimmer for years after that. The Z-Wave spec was fixed to allow for this by those vendors who obtained a license from Lutron, including Leviton (the Hail command). You could not sell an instant reporting device without that license. This is why many older Z-Wave devices have to be polled. or are very slow to report.
Lutron uses a star topology they refer to as fixed repeater. The main repeater, or SmartBridge, during setup sends a distributed database to each device, so each device knows what commands it should respond to. When a command is given, it is broadcast by the main repeater, and the devices respond accordingly, allowing for very compact messages. A Lutron system can include a fixed auxiliary repeater device that simply re-broadcasts any transmission in the system.
Further to all of this design, Lutron obtained licensed spectrum in the 433 MHz band. The FCC rules for this band include low power requirement, and low channel occupancy -- exactly fitting the way a Lutron system works. As a result there is little interference in this band, and small messages have pretty high probability of getting through. The propagation characteristics at this frequency are pretty good for penetrating walls.
So, along comes Z-Wave LR basically mimicking this original design. The major difference is that Z-Wave operates at higher unlicensed frequencies, and has more interference issues to overcome. So, it needs higher power. Mesh is a poor design for this application. It's interesting to note that the battery life of a Lutron battery powered device, e.g. Pico, exceeds the battery life of pretty much every Z-Wave battery powered device -- one of the arguments to support the concept of mesh being that it allows for lower power consumption in the transmitters. Ha!
16 Likes
Awesome post!
Thanks Hal!
I am reading some stuff on here about how a ZWave device wouldn't need to upgrade the HW to use LR but wouldn't they have to do so to get the longer range?
700 series would probably need a firmware update but shouldn't need a hardware upgrade according to what I've read to date.
That may depend on what version of the chipset they have.
- ZGM130S037HGN2 is the latest, the one on the SiLabs LR Dev Kit, and verified it supports 14dBm and up to 1mi.
- ZGM130S037HGN1 is the previous one, the one most manufacturers used, and says up to 13dBm at 100m.
Not sure a FW update will be all that is required. Plus, most vendors qualified the device with the FCC and were required to keep the RF lower so the hardware may not support that extended range.
The only difference for the first few 700 series products would be a star topology in LR instead of the Mesh.
1 Like
I know this may be a bit much for the layman but everyone on here seems to be Engineers or Doctors.
2 Likes
