Blue 2-1 switch. Firmware 2.15, driver date 2023-04-11 This error shows up in my logs a random couple times a day. I can’t correlate it to any activity related to the switch. Parameter 33 does not appear to be listed in the driver.
Very interesting.
Parameter 33 is (will) be used for Overheat Protection in the newest production units with the MG24 chip. But I didn’t think the the 2-in-1 switches with MG24 were shipping yet. I’m currently working on the update drivers for the MG24 chip based switches.
Good news is a value of 0 means its not in Overheat mode, but I can’t explain why your switch is randomly reporting that. When I finish updating the driver at least it will tell you its not overheating instead of “undefined” 
I am seeing the same from my switch shipped earlier this month, 2.15 firmware (as shipped) and the driver dated 2023-07-01. My error’s line# is different, but I think the driver version would cause that.
Are we sure this batch of switches has the MG24 chip? I thought @Eric_Inovelli said they were confirmed to have the MG21.
Also, -while I’m here… - I am seeing an ‘Alexa Heartbeat’ log entry. What would cause this? I don’t have anything connected to an Alexa. As far back as my logs will scroll, it’s paired up with the ‘MULTISTATE.OUTPUT.CLUSTER’ seen right above it.
The above switch is an early replacement for the first batch of Blues with questionable id codes. I just installed it after several months.
Could it be possible that the MG21 have built in temp sensor as well? Or the firmware was ported over and left in by mistake?
I don’t think there’s any MG24s in anyone’s 2N1 switch at this point. I don’t think the manufacturer gave free upgrades.
1 Like
I believe that may be the case. I never saw any information on the temp sensor parameters (P32-P33) until the MG24 Fan firmware was released. I am putting the final touches on the Fan driver and then will begin updating the 2-in1 driver to support these parameters as well
2 Likes
To be honest, I’m not sure how the Alexa integration works. I believe the heartbeat is needed as part of the discovery process and therefore the switch periodically blurts out “I’m here” so it can be discovered by Alexa. That is just my guess. In this case the driver is simply reporting it and no further action is taken
That’s the problem, I think. I don’t have the Alexa integration.
Is this the only device that is getting the Alexa heartbeat?
I thought I remembered (in the beta discussions) something to do with a heartbeat packet (cluster 0x0006) mechanism and happens ever four hours, also for Alexa.
Yes, only device. I’m getting the heartbeat log entry every 12 hours.
Hi Mark!
I got a wierd one for ya…
I have a newer-model (blue case-back) Blue 2-1 out in my detached (and currently VERY cold) garage.
I recently added some new LED notifcations to it (but no parameter changes), and it then started lapsing ito a blinking-red led cycle.
After checking everything I could think of, I noticed that it keeps bouncing back & forth between “Over Heat = True” and “Over Heat = False” statuses. But even when False, Parameter 33 always seems to stay = 1.
My garage is well below freezing right now, so there is no way the switch is actually overheating. Occaisionally, the indicated temp shows an absurd value in the 490s, but it typically stays at 32 (I’m guessing that’s as low as it can show).
First question – is that “Over Heat” condition causing the led bar to blink red?
Second question – any thoughts on why it may have started this behavior suddenly?
I’m on Hubitat with latest versions of Hubitat f/w (2.4.0.146), Blue 2-1 f/w (2.18) and your latest Blue 2-1 driver (5-30). All 8 of my other Blues are rockin’ along great.
It seems odd that just adding some new LED notifications could’ve somehow started this Over Heat thing, but the timing otherwise seems way too coincidental. But who knows - stranger things, I guess…
I’m going to try swapping in a spare Blue that I have this afternoon.
Thanks for any thoughts you may have!
The plot thickens…
Previously, I tried pulling the air-gap for a couple minutes, and also tried turning off that circuit breaker for a few minutes – neither seemed to help. I did verify the circuit itself was OK – power coming that box tested nice & steady right at 120.
But when I disconnected that Blue and put in on my plug-in testing rig, it’s been fine for the past hour – no overheats. So I just put in back in the switch box and now I’ll monitor it.
When I disconnected it, the wiring all seemed tight & secure, but perhaps one of the connections was somehow just flaky enough to cause a problem (perhaps exacerbated by this dang recent polar-votex cold) – I can’t think of another possible explanation.
ETA – when I added the new notification LEDs, I was doing a bunch of various paddle taps & holds as I was testing the notifications, so perhaps that extra action either jarred a connection loose or somehow induced the Blue into thinking it was overheating 
I’ll keep an eye on it over the next day or so… If the problem comes back, I’ll just proceed with swapping in my spare Blue.
Good times in the ol’ smart home – it’s always something’!
Don’t forget that metal will shink when it gets cold. So if you installed a “warm” switch, there could be the possibility of a less than optimal connection when the metals got super cold.
2 Likes
Ha, yes, one way or another, I suspect some shrinkage somewhere is behind this.
1 Like