I know I have developed some simple rules for myself.
- Manual control has to be avaliable at all times.
- Make sure spouse know when i make an adjustment to automations.
- Listen to said spouse about potential downfalls. And either looks for conditions to help ensure reliability or abandon if to many conditons
- Start small and expand from there.
My first automation like this was two lamps that used motion lighting. We mutually decided on the appropriate level depending on sunset and sunrise. Then I added color temp for daytime and night time. Then I added a zen34 remote switch to tell the house when we were awake or asleep to add another layer of control. Now those zen34 switches also scroll through built in scene modes in the philips Wiz connected bulbs we use.
It works pretty flawlessly except for the doubletap on the remote switch which we can both be a little to quick for.
The key is always to start small, test to ensure it works as expected and something unexpected does not cause unexpected results, then move on to next part in a way that doesn't break what already works.
1 Like
No. Just because Zigbee is faster and can handle a lot more traffic than Z-wave.
No.
I’ll post when I get home
This is how all of mine are setup. Remember, I don't use them to control any type of load, just buttons for the Button Controller app. Also, some settings I have because they didn't always turn on properly after a power outage.
I know this was believed to be true based on some data collected with a smartapp that was distributed. This may not be the case since the new Zwave Firmware for the C7 was provided.
As I recall both are sub 500ms and I think Zigbee was generally between 100 and 200ms while zwave was around 200-450. This could also be impacted by how good your mesh is as well. Either way limiting how much either wireless network is used is the better way to go.
This to some extent, but I think it's more about zigbee having a lot more bandwidth than z-wave. It doesn't get bogged down like z-wave does.
A real life example would be, I have a z-wave device that was spamming the hub with battery reports every 6 seconds. That was enough to give my other z-wave devices a 10 second lag in response to commands. I only have around 30 z-wave (all plus devices) and none that report power.
That's interesting data -- and surprisingly high! What kind of "sniffer" tools do folks use to gather that kind of data?
I've got a tiny Z-Wave deployment as I'm testing things out: Just a couple switches and the hub all in a small room. My expectation was that local RF protocols would be super fast (10s of milliseconds, tops) and that the main bottleneck would be hub processing time.
After all, TCP/IP over WiFi can achieve sub-millisecond ping times on my local network. Surprising (to me) that ZWave/Zigbee would be hundreds of times slower!
That’s round trip communication with the hub from what I remember. ie the hub sends a command, device executes command and reports back.
Also, Zigbee operates at a maximum speed of 250 Kbps, compared to Z Wave’s maximum speed of 100 Kbps.
You have to remember wifi is designed to work with TCP/IP and computer type technologies that have a good amount of power to draw from. Even the old 802.11b tech had monsterous throughput compaired to zigbee and zwave. Zigbee and zwave are designed with power as a first prioroty that is why it is practical to use with battery devices with small batteries like 2032 coin cells. Zwave 700 was suppose to speed up responsiveness with zwave. I have over 30 zwave devices with a mix of battery and mains powered devices and in general they are great as long as the mesh doesn't get borked.
I have really had one occasion where it got bogged down and that was when a zooz double plug was installed. They dont play real nice with hubitat.
Zwave can be a little bit slower and zigbee can be impacted by wifi and other techs that run on 2.4ghz wifi or near it. I lost every zigbee device in my house once when I added a wifi camera that ran on 2.4ghz wifi.
I believe the app that was used for that timing was hub watchdog. The way the app works is pretty straight forward. It can turn a device off and capture its state chage timing. You can identify if the device is zigbee or zwave or virtual to get timings of how each type of device are doing on your hub. Just make sure you dont connect it to something you actively use.
