I don't have a C4, so I can't comment to that, but I've seen the same behavior with my Nortek stick; Z-Wave devices pair the first time and pretty much every time. A couple of the GE/Jasco dimmers had issues, but I think that's just because the older ones suck. In contrast, the same devices with the C5 consistently fail to pair multiple times and then just works out of the blue.
I'm attempting this at the moment. I'm starting with following your Hubitat > NodeRed > MySQL > Grafana (LONG READ)
but it may be a bit over my capabilities.
2 Likes
After "knowing" you here for a few months, I know you'll get it. If you do get stuck, PM me and I'll give you a hand.
As for HA, do you have an install already setup?
2 Likes
Not yet, I was seeing if I could accomplish this part first. Then I was going to move to that next step. Do you have HA running on your server or a Rpi?
I have it running on my server (my horsepower, though HA doesn't really need it). However, if you have a rPi around, I'd throw it on there along with a Nortek stick. That way, you can easily move it around should you need to for pairing.
1 Like
@waynespringer79:
If I may ask, are you using Lock Code manager on your zwave locks?
I found that my zwave locks (Kwikset), have dropped from my zwave mesh after using that program.
Using from RM, from the device page, and from dashboards - no problem.
P.S. I have found that it's critical to have a "beaming" device near the lock (I have a Leviton Zwave Plus Switch nearby).
Yes I'm using Lock Code Manager and I have 5 beaming Aeotec Range Extender 6's
1 next to the hub and 2 EACH by each problem lock.
I have yet to prove it, (I'm currently away from my home for some time), but I believe that the lock code manager is causing locks to drop from the zwave mesh. I know that it happened to me.
Again, I stress that I haven't yet been able to prove it (because if I could prove it, and show that it is repeatable, I know that HE Staff would try their best to fix it).
If you don't use LCM, I would wager that your locks will not come off the mesh!
Interesting theory. Next time they drop I will delete LCM and test it.
My theory is the hub's power supply is under powered for "secure" zwave communication with "secure" zwave devices, but like yourself I'm unable to prove this. Makes me wonder why they set up the zwave radio to "choose" between "Locks and Garage Doors" and All Secure Zwave, as well as having this ability to "limit" the secure communication of zwave is a first I've seen on the 3 different platforms I've used.
I ran this test on one of my many excluding everything and start over attempts. The All Secure Zwave causes far more "communication" issues than just the "Locks and Garage doors" given the appearance of the "secure" communication of the radio as being the problem.
Another data point for my theory of this is here lately after they have dropped off, if I simply do a hub reboot they all come back working again, until a few hours later.
I don't think that the problem. I think the following is the problem:
(from an excerpt of a solution, posted in 2014):
"Z-Wave devices form a mesh network and a beam command from your controller to the lock may have to pass through other devices in the network to get to the lock. These intermediate devices need to recognize the beam command as a valid command and then retransmit it to the next device in the network. If the intermediate device doesn't support the beam command, it may simply drop it on the floor and it will never get to the lock. If your controller is able to be in direct communication with the lock, without any intermediate devices, then it wouldn't matter whether the intermediate devices support beaming or not. But you won't really know whether that's possible until you install the devices and allow the mesh to be configured."
I believe, (again no proof), that the Hubitat zwave radio is not as powerful as the ST radio, and therefore it relies on other intermediate devices, re transmitting messages. However, as the previous excerpt said, all of those intermediate devices must support (in one way or another), the beaming class command. Therefore, you should be looking at your mesh.
As many others have said, "it's all in the mesh!"
I'm not buying to much of this explanation as the problem didn't exist on two previous platforms (yes I also believe their radios are stronger) using far less repeaters and probably none had "beaming" capability. And the locks didn't drop off.
Right now I have 23 repeating devices with at least 5 have beaming capability for only 14 battery devices (3 of which are locks) So I have almost 2 repeating devices for every 1 battery powered non repeating device. As well as 2 beaming devices for every lock that is further than 10 feet from the hub. And STILL have this issue.
1 Like
Wayne, that's precisely my point.
If other systems had hubs with stronger radios, then possibly they could attach to the lock directly. They wouldn't need intermediate, repeating devices, and it would be irrelevant if all devices in the mesh could "beam".
However, some other hubs, require intermediate repeating devices, and require that all intermediate devices support beaming.
Do all your intermediate devices support beaming?
(To a certain extent this is "shooting in the dark", because, in general, we don't know what our z wave mesh really looks like.)
1 Like
Under my latest test currently going on now for 3 days YES. As I have a completely separate hub by itself with the following devices ONLY:
5 Aeotec Range Extender 6's (beaming supported)
3 Schlage BE469 Locks
And the locks have still dropped off.
This just isn't possible. Lcm communicates with the lock driver not the device directly.
The range extenders stay connected now? Or do they drop too?
Last night I tested them again and the "test packets" received were dropping down again unlike all receiving all 30 back instantly after the firmware updates. I'm rebooting that hub again at the moment as all 3 locks dropped off again after reboot I'll test the range extenders again.
If I remember correctly, you have one range extender like right next to the hub, correct? Do you see the same dropping of frames on there too?
Yes although not as big of a drop it went from 30 to like 26
I see a drop from 30 to 25 every once in a while when I haven’t run the test for a longer time. But I get 30 if I run a second round right away. Do you see the same?
I tried that it was hit or miss.
Some (especially ones that scored low at the first test) I tested again. Sometimes the test improved and sometimes after the improvement I tested again and it went down.
