I’ve been digging into Z-Wave 500, 700, and 800 technologies and trying to understand how they interact in a mixed network. Here’s what I’ve gathered so far — please let me know if I’m missing something or if any of this is incorrect:
Summary of What I Understand
All Z-Wave generations (500, 700, 800) support mesh networking, and they are backward compatible.
Z-Wave 800 introduces Long Range mode, which is point-to-point (not mesh) and allows direct communication with the controller over much longer distances.
In a mixed network, devices must use the lowest common protocol to maintain compatibility — meaning Z-Wave 800 devices may not fully benefit from their advanced features (like battery life or range) if they’re operating in mesh mode with older 500 devices.
Battery life improvements in Z-Wave 800 are mostly realized when used in a pure 800 network or in Long Range mode. In a mixed mesh, they may be limited by older protocol behavior.
Long Range devices do not participate in mesh, so they can’t relay other devices and won’t benefit from mesh redundancy.
Given all this, it seems that adding Z-Wave 700/800 devices to a 500-based mesh doesn’t bring major improvements, unless you’re planning to migrate fully or use Long Range for specific cases.
My Questions
Is it true that Z-Wave 800 devices in a mixed mesh are “held back” by 500 devices? (we need full 800 device mesh to see improvement)
Is there any real-world benefit to adding Z-Wave 700/800 devices to a mostly 500 network?
Has anyone successfully used Z-Wave 800 devices for radiator thermostats or similar use cases?
This is not correct. Assuming you have a LR-capable controller, a 800 series device can be included into the network using either mesh or Long Range protocol. If you include it using mesh, it will act as any “standard” mesh device. If included using LR, it will have the benefits of longer range and improved battery life (but won’t participate in the mesh network.)
My network is mostly older mesh devices, but I’ve added a few new LR devices and both old and new devices are working great.
The limitation comes from how Z-Wave handles backward compatibility. If your mesh network is built primarily with 500-series devices, the 800-series devices have to fall back to the older protocol to maintain interoperability. This means:
They use the same routing and command classes as the 500-series.
They cannot leverage the advanced features of the 800-series, such as Long Range (LR) or improved power management.
The network performance is effectively capped at the capabilities of the older generation.
In other words, as long as the mesh relies on 500-series nodes, the 800-series devices lose most of their advantages and behave like 500-series devices for mesh communication. To benefit from the 800-series improvements, you need a network where the majority of nodes support the newer protocol or use LR in a point-to-point configuration.
hi @PecosKidd
You’re absolutely right that if a device is included in the mesh, it has to speak the same protocol as the other nodes. So, if you have 500-series devices in the mesh, any 700 or 800-series devices included in mesh mode will need to follow the same constraints to remain compatible. That’s what I meant: the presence of older nodes forces compromises and prevents you from fully leveraging the optimizations of newer generations.
Of course, there’s still the option to include an 800-series device in Long Range (LR) mode, which takes it completely out of the mesh and gives it its advantages (range and battery life). But in that case, it no longer participates in the mesh, which can be a strategic choice if you want to bypass the limitations of older nodes.
In summary:
Mesh with 500-series → the entire network is limited by 500-series capabilities.
LR → no mesh, but improved range and battery life.
Your system works well because you’ve mixed both approaches, but for someone who wants to keep a fully meshed network, the presence of 500-series devices remains a limiting factor.
Then buy an 800 series controller if you’re worried about it. I have a mixed network and everything is fine. If I decide to go long range with an 800 device or at 10 years with my 500 stick I’ll migrate it.
Im not certain about it (in fact I don’t believe it does downgrade the whole convo) Ask Fresh they will know. (really, they will ZWave sme there, owns the whole migration to jsiu guide… Pretty sure if Fresh says If it won’t it doesn’t. So thier post was a direct question that should have been read as: you sure bro? ) downgrading the entire comms chain. I think it downgrades to its next hop as I know for a fact that security doesn’t downgrade just because you added an s0.
In short it’ll work fine. I’d worry more about bad sticks and bad 700 series firmwares.
500 series devices do not limit anything in a z-wave mesh compared to 800 series devices. It is the mesh itself. There is no difference between a 500 series mesh and a pure 800 series mesh network.
Data from 500 or 800 devices is routed identically by 500 and 800 devices if they are not directly connected to the controller; only in this case would it be relevant at all.
The only difference is that 800 series devices has generally a little better range and less power consumption.
What makes a difference makes, is when is operate you 700/800 devices in LR mode. However, there is no influence from 500 devices here. It runs in a different modulation on a different frequency.
I think this may be the disconnect. If you truly want a mesh-only network, there is minimal reason to get any 800-series devices. But why would you want this? There is no benefit I can think of to keeping a “fully meshed network”. Why not use mesh for your older devices, and the new LR protocol for your newer devices? They won’t add to the mesh, but they won’t hurt it either.
Consider adding or upgrading to LR devices that are:
Further way from your controller (ie, that currently use multiple hops in the mesh)
Battery powered (to take advantage of the better battery life, and since they aren’t contributing to the mesh anyway)
That said, replacing an existing device that acts as a critical hop in the mesh with a new LR device (and therefore taking away that mesh point) could be an issue you’d want to avoid.