Occasional Issue with RM and CT100 Thermostat

I have a CT100 thermostat connected to my HE via the Generic Z-Wave Thermostat DTH. I can manually control it quite well. I've also created some RM rules to control the thermostat mode and temperature setpoints based on weather conditions, modes, etc. The rules execute as expected, but sometimes the thermostat setting don't change. For instance, I have a rule setting the thermostat mode to off when a certain threshold is met:

This morning the temperature outside met this condition while the thermostat was in heat mode (which updated automatically overnight via a different rule), so the rule executed. I received a text message as expected. However, when I checked the thermostat the mode was still set to heat. I opened the rule and confirmed the necessary conditions were met. When I clicked on Done and rechecked the thermostat the mode had changed to off. I've seen similar issues occur with cooling setpoints for a rule that executes when the HE mode changes to Home. The rule executes but the cooling setpoint doesn't always change as expected.

Are there any known issues with RM controlling a CT100 thermostat?

FWIW, I have have 4 CT100 units that control the temperature in my house; and I do see similar behaviors, yes. The thermostats sometimes miss some commands, and/or notifications from the thermostats are sometimes missed by the hub as well. I noticed that same behavior with ST and Hubitat.
It may well be an issue with the z-wave mesh, but that looks fairly healthy otherwise...
I solved the problem by making my control apps more resilient - they basically send the command, and check a minute later to verify that the command was actually processed, otherwise they resend it and it recheck. I must say that I have adopted the same strategy for all z-wave devices that I never want to end up in the wrong state, and that approach works well for me.
That also implies that I'm not using RM much, but I'm mostly relying on custom apps.

1 Like

I have a CT100 as well and I have noticed that sometimes it misses setpoint changes as I leave,arrive and go into night mode. It's pretty rare so I just generally ignore it.

I imagine it's just dropping a couple of commands because it's battery powered and sleep. I also imagine the same could happen with mode changes... I hope my next house has a C-wire.

Good idea!

Mine is connected to a C-wire, so it shouldn't be due to a battery-related issue.

Good to know. Maybe it's just because it's a cheap device? It's pretty solid but it's by no means the most expensive thing out there.

Maybe Z-Wave Repair?

Good luck.

I use one and have filed a bug report on that issue. There may be something with the Z-Wave commands, but I'm not sure. I've managed to keep it at bay by creating a rule to "refresh" the thermostat every 1 minutes.

1 Like

Is yours powered or battery?

Mines all are.

Mine is wired.

Yea polling makes sense if wired.

It was a real PITA to run a new set of wires since I have a package unit on the roof. The other reason I have a RM refresh command is that I have room fans that I bump up to 60% when the AC is blowing and turn on to 30% when the heater is on. This was the only way I could make sure the fans return to their previous state when the CT100 OperatingState changes.

1 Like

I abandoned mine for all the reasons noted here. Went with an Ecobee 3 and am delighted.

I have one laying around if anyone wants it in exchange for a modest amount of beer money.

You can wire it up to use the fan wire as your C wire with the only compromise being you won’t have a fan only mode any more.

I am having the same exact issue as you guys however I am using a different z-wave thermostat...Remotec ZTS-110. I have issues with it not changing modes when the house mode changes. I have also created RM rules to double check the state and change it if it misses. I think there must be a bug with Hubitat. Both thermostats are very close to the Hub and I never had a sign issue when I was on a Vera controller before. I am also using the built in Generic Thermostat driver. Have any of you tried a different driver yet?

I am using a custom driver for my CT100s (somewhat similar to what RBoy developed on ST); I do see the same occasional problem, and I don't think it's a Hubitat issue - it seems more related to flaky z-wave communication. It does happen more frequently for thermostats which are positioned further away from the hub, but it does happen occasionally also for the one which is in the same room as the hub.
I must say that leaving the z-wave mesh alone for a few weeks did help stabilizing this behavior; but just in case it happens, the app I use takes care of fixing it.

@minollo, Thanks for the info. Like you said, it may be a flaky z-wave but I didn't have it happen one time in about two years that I had the same thermostats on a Vera controller. Maybe the antenna on the Vera was better, but my downstairs thermostat is located about 5 feet from my hub. Do you mind sharing your device driver code and app you are using to fix it if it misses?

Thanks!

Sure, for what it's worth: app and dh

2 Likes

A question about your app.
What does the switch to suspend polling device do?
Will this wear down the battery of a battery operated thermostat?

Thanks!

The switch to suspend polling is something (optional) which I use to avoid polling across the network while I'm repairing the z-wave mesh (it's just good "z-wave" practice to do that, I believe).

Polling does put some load on the battery, but I have tried to limit that; for example, battery polling, as you can see from the code, happens only every 10 hours (if I well remember); and polling for other information which is less important to me (getMoreInfo) is less frequent than the "normal" polling as well.
For reference, alkaline batteries in my CT-100s last about 3 months; rechargeable ones about 2.5 months; that's acceptable to me.

1 Like