I have a Vera Edge and a Vera Lite what would be the best approach to get all my devices together?
What would be the advantage of bridging versus making my Vera Lite a secondary controller on the Vera Edge network?
Bridging Veras is intended to increase a coverage area or extend the 232 device limit of a Z-Wave network by operating two discreet Z-Wave networks while still sharing some of their controllers’ logic via a TCP/IP connection.
A secondary controller is a participant in a single Z-Wave network, but it only shares devices and does not share Vera’s controller logic with the primary controller. Some interaction is still possible by using HTTP commands from one Vera to the next if needed. (Hopes of automatic failover controllers are not possible with Vera at this time.)
Now, the question is; which best suites your needs?
Do you have a large number of devices or an area so large that your Z-Wave mesh cannot cover the area(detached buildings)? Then bridging may be desirable, though discreet systems may still be acceptable/better.
Do you have two networks with too many devices to exclude from one and include into the other but would like to consolidate control/management of both networks as a single entity? Then bridging may be your only option.
But, bridging introduces complexity and latency. These are undesirable aspects of any network. So, if you have only a few tens of devices in not too large of an area you will probably be best served by excluding your Vera Lite devices and including them into your Vera Edge network and moving forward from there.
Alternatively, if your Vera Edge network only has a few devices, you may prefer to include it into your Vera Lite network, shift Edge to be primary and rebuild your logic on the Edge.
Thanks for explaining some of the options.
Can you tell me what logic is shared when bridging?
Let me give some more background to my situation. I have about 30 devices. The reason for the 2 controllers is that zwave transmissions do not travel well through my walls and ceilings. So I have controllers on the ground and top floor. If I depend on the mesh for a single network I found the latency to high for transmissions that need to ‘hop’ from the ground floor to the top floor. The logic for the ground floor devices is now in the downfloor Edge and the logic for the top floor device is on the top floor Lite.
But there is also logic needed that covers both top and groundfloor devices. This is now coverd by bridging.
Since the Edge is zwave plus and the Lite not I also have devices that still need transmissions to move through the house in case of zwave plus devices used at the top floor. I would expect this to work better if I have one zwave network. For these device latency is not a big issue. Will the zwave plus transmissions also use the regular zwave nodes to hop to the controller?
Does it make sense to create one zwave network and run the logic on the controller that is on the same floor as a device to keep the latency low?
[quote=“ReneH, post:3, topic:188958”]Can you tell me what logic is shared when bridging?[/quote]What ever logic you configure. The bridging process shares devices between units. Then your configured scenes and plugins create the logic that may be spread across devices on either unit. Later in the post you state that you are already bridging and sharing logic, so I’m not sure what exactly you’re asking about.
I have about 30 devices. The reason for the 2 controllers is that zwave transmissions do not travel well through my walls and ceilings. So I have controllers on the ground and top floor. If I depend on the mesh for a single network I found the latency to high for transmissions that need to 'hop' from the ground floor to the top floor.30 devices is a small network whose communication will likely improve with the addition of more devices. But, yours is not a 30 device network. Yours is two discreet networks of ~15 devices each. On the Z-Wave level, devices in one network do not communicate or relay or have anything at all to do with the devices in the second network. The networks are discreet even if they physically overlap.
It’s impossible for me to say what will work best for your specific environment. There are too many variables and far too many unknowns. Even if I was aware of specific distances and materials, there are still unknown noise levels and other signal propagation issues that make it impossible for me to know. But, by far, most Z-Wave networks improve communication with improved density. More devices usually results in a more robust mesh and better reliability.
Since the Edge is zwave plus and the Lite not I also have devices that still need transmissions to move through the house in case of zwave plus devices used at the top floor. I would expect this to work better if I have one zwave network. For these device latency is not a big issue. Will the zwave plus transmissions also use the regular zwave nodes to hop to the controller?Z-Wave plus is backward compatible, so a Z-Wave plus device will work fine on or through a legacy Z-Wave network. There are some very specific cases where newer encryption incompatibility creates some problems, but these can be resolved. Just remember that a Z-Wave device from one network, will not benefit or participate with another Z-Wave network in any way.
Does it make sense to create one zwave network and run the logic on the controller that is on the same floor as a device to keep the latency low?As I said above, it is impossible for me to say what will work best in your specific case. In most installations, such as you describe, I would attempt to use a single Z-Wave network and controller(Vera Edge). 30 devices in the same network should provide a decent mesh and adding more devices should improve weak areas.
Going back to your original question and comparing it to your present setup; bridging your Veras allows for a centralized management and logic of both your Z-Wave networks. If you do primary/secondary you will have a single Z-Wave network, but management/control of the controllers will be separate. You would need to login to one for its scenes and log in to the other for its scenes. With bridged controllers, you can create all the scenes on one controller and have it control devices connected to either.
My explanations are starting to seem circular, which is unfortunate, but I don’t know how to describe it better.
I can see the devices of the other controller after bridging. Scenes from one controller are not available on the bridged controller. I would asume when I make the Lite a secondary controller then I would see it’s device on the primary too and can use it in the scenes on the primary. Or is that not the case?
If you do primary/secondary you will have a single Z-Wave network, but management/control of the controllers will be separate. You would need to login to one for its scenes and log in to the other for its scenes.
Not sure but I think these means no on my previous question.
Thanks for your input!
[quote=“ReneH, post:5, topic:188958”]I can see the devices of the other controller after bridging. Scenes from one controller are not available on the bridged controller. I would asume when I make the Lite a secondary controller then I would see it’s device on the primary too and can use it in the scenes on the primary. Or is that not the case?[/quote]You should probably read up on some of this, as I think my descriptions may be creating more confusion. None the less…
When bridging, only the “slave” Vera’s devices are shared with the master. The scenes on the “slave” controller are not visible and will not become visible to the master. Only devices. (it is possible to do two way bridging, effectively allowing both controllers to see all devices.)
On the master controller, you can create scenes, PLEGs, and other plugin logic that controls devices on either the master or slave controller. It would seem likely that you are already doing this. The master can tell the slave to issue a command to one of its devices. The master cannot communicate directly with a slave’s device.
In the primary/secondary controller of a single Z-Wave network scenario, each controller has a complete list of all devices on the Z-Wave network. Neither controller is really aware of what the other is doing and they do not share or centralize any control or logic. They both see the entire network and either can issue a command directly to any and all devices. Either can have scenes or plugins that can act on any and all devices.
It is now clear to me how the two setups compare.
In this situation what happens with messages sent from a slave device to the controller. If a device sends a status change and it has a direct connection to the secondary controller will this status changed also be propagated to the primary controller? The slave device will not be aware of the network setup so for this to happen I think a controller would have to resend updates to all other controllers in the network.
Controllers do not resend device updates to other controllers.
Controllers will either see s status change or they won’t, depending on many factors including proximity of the device, if the device supports instant status, and other factors. Status updates will be directed to the primary controller.
The primary controller will not update the secondary. If the secondary controller does not happen to see the status update, assuming that there even was one, then the secondary controller will not have accurate state information until the secondary controller polls the device.
The primary controller will not update the secondary. If the secondary controller does not happen to see the status update, assuming that there even was one, then the secondary controller will not have accurate state information until the secondary controller polls the device.
So having any logic on the secondary controller is not usefull then because it can miss status changes of devices. The only use I can think of for the secondary controller then is that it is a backup when the primary controller fails.
In either of these configurations, bridging vs primary/secondary controller, is it feasible to put all z-wave devices on one unit and then all plugins (plegs, multiswitches, day/night etc) and logic (including scenes and plegs) on the other? It sounds like in most cases there might be problems with the logic devices receiving timely updates from the z-wave devices. Rational is that I have a veralite and a veraedge. The plugins take up too much memory on the lite but it is easier to add z-wave devices to the lite given its ability to run on batter.
Yes, certainly. I ran two bridged VeraLites like this for quite a while, devices on one, plugins on the other.
Thanks, then the next task is trying to get them bridged.
Which was the master, the one with plugins or the one with z-wave devices? I am assuming it was the one with the plugins.
Yes, it has to be the one which can see the other’s devices (assuming you’ve only bridged one way.)
Please note that currently the Vera Edge cannot be the master yet according to Vera Support.
To bridge two Vera’s there are several Forum post that describe how to change the IP address for the Vera’s to see each other this is not necessary. I had Vera support setup my bridging and the ip adresses where not changed. According to support this is what they did:
When I?ve connected to the VeraLite unit the VeraEdge did not appear when I?ve first scanned it from the interface (Devices > Add Device > Add UpnP Device); because of this I?ve manually discovered the VeraEdge thought ssh and after that scanned once again for UPnP devices from the VeraLite interface.
I do not know how to manually discover a Vera through ssh but someone here might know.