Nest Gen 3 turns off and back on within a few seconds

[Engineer]

There are multiple recent threads regarding the Nest thermostat turning the air conditioning system off when it shouldn't and then back on within a few seconds. Have the people at Nest acknowledged this issue? I am running a C wire.

[WorkPlayDrive]

Just adding to this thread. 

Same issue as many have reported above, my equipment has worked fine for YEARS and in June I started noticing that my AC compressor would shut down for 2 seconds mid cool and kick back on. 

Why is this an issue now after years of using a nest thermostat? 

[dco34]

+1.

At about 4am every morning the AC starts and abruptly stops within 2 seconds .. and then starts up again 2 seconds later.

The only potential workaround I've found so far is to turn off all manual schedules and auto-scheduling.

Super disappointing. Nest should fix this ASAP.

[Ryan_G]

Hey everyone,

Thanks for being here in the Google Nest Community.

For WorkPlayDrive and dco34, could you tell us if you have a C wire connected to your Nest Thermostat? 

Regards,

Ryan

[dco34]

Hi Ryan, I can confirm that i do have a C wire connected.

[Tbnilsen]

I have the same issue. I had the Nest connected for years without a C wire and noticed last winter that my boiler was short cycling. It makes perfect sense to me that this will happen without the C wire installed (ie battery charging). So I added the C wire, and the Nest (gen 3) correctly told me it detected power to the device and started using the power. I even checked the voltage level under the Nest information screen (Vin is 30VAC). Note: I only use the Nest to control heat via a water boiler with infloor heat exchange. 

Short cycling still happens (sometimes the Nest demands heat for as little as 30 seconds over and over again). It appears that this occurs right as the room temperature approaches the Nest’s setpoint. My suspicion is that the Nest is trying way too hard to achieve the setpoint by quickly cycling the boiler (and thereby the hot water circulator pumps). I’ve turned off overshoot learning and there are no schedules at all programmed (I only installed the Nest so that I can remotely monitor and control the house temperature since this is a vacation house. Otherwise I set it at one temperature and forget it). 

Is it possible that the Nest still thinks it needs to charge the battery? Likely the battery has been weakened after years without a C wire, but with full Vin via Common Wire should that matter?

The Nest has the latest firmware. Is there a setting for the Nest to stop it from such extreme short cycling? Please help. 

[WorkPlayDrive]

@Ryan_G   I do not have the C Wire connected.  We never had a C wire since we installed the original nest back in 2012.  Worked on my First gen and the newer 3rd gen and just became an issue this season.  

[Stevefins12]

I do have a C wire. My AC was working perfectly fine with a different smart thermostat, and since switching over to Model: G4CVZ, every ~110 seconds the Air handler turns off, then in 90 seconds it turns back on. WHAT'S GOING ON HERE?!

[horseshoehorn]

It does not seem that Nest is going to acknowledge that this problem is an issue with the thermostat programing and/or firmware. The Community Specialists continue to circle back to the presence, or lack thereof, of a C wire. It has readily been established that users with a C wire still have this problem, some after years of reliable service. It is more than coincidental that this problem is now occurring after a recent firmware update in the spring. This is a massive failure on the part of Nest. I am done with this thermostat and will recommend users stay away from it as this issue alone will compromise their AC system and compressor. Nest should issue a refund on this product as a matter of warranty. I could not possibly be more dissatisfied with this product and lack of resolution to this issue.

[Tbnilsen]

Thanks for your reply! You’ve touched on the real problem as I have read over and over again online and that is Google’s inept technical support that equates to no more than a “did you plug it in” reply. This is infuriating and I’ve seen many a frustrated (and helpless) fellow Nest owner treated to the “C wire” response. 
I won’t recommend any Google product either and will likely just eat the Nest cost and move onto a product with real tech support. 
Caveat emptor. 

[Keatsian]

Why wont the Google reps ack this problem?

why.

[Engineer]

Short cycling an air conditioning system is very hard on the equipment. I suspect Nest does not want to take responsibility for destroying thousands of systems.

[dco34]

It's such a joke. I called customer support and they said there have been no reports of this. I asked them to read this thread. They offered a warranty replacement, but I think I'm done with Nest after this issue. Going to install a different brand and move on with my life.

[Engineer]

What phone number did you call?

[Tbnilsen]

I didn’t call them. I was sent a link to a form to fill out with my issue. It’s working it’s way thru the system now. 🤞

[Oldsalty1]

After reading through these posts, I am convinced this is a firmware issue.  I wish I would have read these before dropping $1K trying to "fix" my compressor.  My background includes debugging firmware on several pieces of equipment over the years.  Luckily I did not purchasing the Nest (gen 3 learning) because previous owners of house left a brand new in cupboard in garage.  But buying a $25 "dumb" thermostat has resolved my issues of short cycling my compressor and probably saved me $5k on not having to replace my compressor.  These type of non-answers has become the norm.  Quite sad really.  I think your guess of Nest not owning up to issue is right on target.  I wonder though why they have not come up with a revision to the firmware (this is over a year old) or at least gone back to the previous version.

[WorkPlayDrive]

I just called in as well. 

They explained the C wire is required in my specific configuration due to the heating side not providing power. they shared this article   https://support.google.com/googlenest/answer/9250906?hl=en 

Despite explaining this just became an issue around May of this year after having the same wiring configuration on my 1st gen and 3rd gen nest since 2012. 

[Tbnilsen]

A friend sent me this link: 

https://smarthomeowl.com/nest-thermostat-cycling-on-and-off/#:~:text=The%20common%20causes%20of%20Ne...

if you believe the article then even with a C wire the Nest relies on its battery for at least the display and WiFi power. With an old battery it claims it will crater it and have to charge it during the heat cycle and that THAT causes it to cycle heat demand. Which makes no sense. Unless the firmware erroneously uses the “steal power trick”… or maybe the new firmware added power hungry features that craters the battery for sure… I wish I had schematics for the Nest circuitry so I could help out. A man can dream. 

[Engineer]

I believe this article is speaking about an earlier generation of thermostat. The batteries and battery charging system are very different between what they're speaking of and the Gen 3 thermostat.

[Tbnilsen]

Well that’s good news. That article described behaviors that didn’t make much sense, and I’m sure happy it’s not describing the operation of my Nest gen 3. Maybe an expert can confirm this. 

I just did another test with the Nest. I monitored the power (volt meter) to the valve which is controlled by the Nest. I created a demand for heat on the Nest and the voltage to this valve correctly jumped up to 27VAC (ie the transformer voltage). The boiler came on and all started operating. However, after 1 minute and 11 seconds the demand for heat went away for less than a second (I don’t have an oscilloscope here and don’t have access to one, so I can’t say exactly how fast it was, but I saw the very brief drop on my volt meter). My furnace controller reacts immediately to this glitch and initiates its shut down process. Once the furnace completes its shutdown along with its 1 minute lockout period, it again sees that demand exists and fires back up again. Like I’ve said in previous posts, this will continue for many hours. 

So, something is taking place inside the Nest which causes glitching to the furnace. (BTW all the while, the Nest is correctly demanding heat and is unaware of the short cycle it induced). Is it switching in the battery charging resistor used for non-C wire operation accidentally? Or even cycling the heat demand switch, and I assume one exists? 

I sure hope gen 3 Nests use the C-Wire to charge the battery, because if it doesn’t and still uses the two wire power stealing trick I’m afraid the new firmware is a battery power drain and you always have to have a low-hour battery in the Nest to avoid (or at least drastically reduce) short cycles.

[bailem2]

I, too had this "turn off, turn on" behavior around 1-2x per day when calling for cool maybe 30x per day.  I had no C wire for years and everything was fine.  I have also had no problems keeping my battery topped off except for a handful of occasions where zones that had no action on them for months on end, which is understandable due to power scavenging off of Rc and Rh wires when the zone is active.

My problem started around 5/2022 as well.  My battery voltage is fine both before and after the anomy and I observed this happening shortly after plugging it in with a full charge via USB.  Based on this, I think the lack of a C wire is a red herring.

Based my battery level, this is clearly a software problem.  While I can understand that all bets are off when there are charging or low voltage problems, I have a healthy voltage before or after the anomaly.

[Tbnilsen]

How many zones do you have? If you run two or more zones at the same time you won’t see very many or even any short cycles. That’s because each Nest will glitch at different times and as long as one zone is demanding heat properly the boiler keeps going. Try it. (Obviously doesn’t fix the problem, but can spare your boiler until software is fixed). 

[bailem2]

I have 5 Nests, however only one of them is connected to A/C so it's the only one capable of calling for cool.  The rest are in OFF mode.

[Yehanma]

Hi Everyone,

I have noticed my AC doing similar things as all yours. It came up this year only and never had anything like this before. By browsing all your comments, I think the issue came from the firmware updates and Nest doesn’t have the capability to solve the issue at this stage. I think I am going to switch to the old school simple ones and solve the issue permanently.

[Engineer]

I've learned the hard way that nobody from Nest follows this community page. I have permanently corrected the issue by installing a different brand thermostat. It can't screw up if it's not installed!

[Yehanma]

The firmware I am having is 6.2-27. The hardware I am currently having is a replaced one. I had Wifi issue before and the unit got replaced. My feeling from dealing with the wifi issue is that the technical support from Nest is useless, as the representatives only asked to do reset and cannot even debug. That’s not the way engineers solving technical issues.

[Jeff]

Hey there,

Apologies for the late reply on this thread, and for the frustrations surrounding this issue. We'd like to gather any case ID's that you may have from contacting customer support so that we can send them to our engineering team for them to investigate further. Once we receive your case ID, they will continue investigating this and we will update this thread as soon as we have any further news.

Best regards,

Jeff

[Engineer]

Hi Jeff,

Thank you very much for the reply. Aside from the apparent defect in the software, there seems to be a second issue. Yourself, Ryan, and others at Nest fail to realize that this community page is the de facto location for failure information and communications with your customers. The frustration is with the lack of response to this community page in addition to our air conditioning compressors being damaged by the change in the code that occurred in May of 2022. It seems no one at Nest has bothered to read through the two or three threads specific to the Gen 3 thermostat short cycling compressors and furnaces. Myself and others have made comments that adding the C wire has not cured the issue. I have also added plots showing the short cycle phenomena. Many people on here have already changed their thermostat from Nest to anything else. I suspect this is why the activity on these threads have declined.

[Tbnilsen]

Thanks for your calm and lucid comments about not only the actual Nest firmware  problem, but for expressing your feeling about the lack of support. I’m still hopeful that it’ll work its way thru. In the meantime I devised a workaround. I added a 5VDC relay in between the Nest and furnace that now is controlled by the Nest. Then I moved the furnace control to the relay contacts. In addition, I added a 1000uF 16VDC electrolytic capacitor to the control line of the relay (and ground). The relay module (bought on Amazon) I have powered by a USB dongle. Now when the Nest quickly (and erroneously) cycles heat demand the 1000uF capacitor holds the demand for an extra 3 seconds thereby filtering out these short cycles. I consider it good for now, at least til Google gets this sorted out. TBC…

[Engineer]

I considered something similar, but then I convinced myself that the features of the Nest thermostat just wasn't worth the development effort. It was far easier to remove the Nest and replace it with another Wi-Fi enabled device. With a big discount from my utility company, I was able to get a $75 device for $10. The best part of this is realizing that the new device has features that the Nest does not have. On a completely different note, it'll be interesting to see if Nest picks up on the fact that, when I reformatted the Gen 3, it completely lost its mind regarding its location. I don't know where it thinks it is, but it's certainly not in my hometown. I've noticed that others have complained of this issue, as well.

[Tbnilsen]

 Case ID [5-3280000032672]

In a response, support Tech claims my boiler is not compatible with Nest and that a bug was not introduced in May fw release. BTW he/she never asked what kind my boiler is….  

I hope you can help Jeff. 🙏 

[chase314]

Which brand did you go with? I'm having the same issue and am worried about ruining my new(ish) AC.  I'm considering Ecobee. 

[Engineer]

I bought a cheap one for $10 that my utility company was offering. I think it was regularly 75 bucks but you got a huge discount if you went through the utility company. The brand name is Greenlite. It's not as nice looking as the nest and doesn't have as many features, but this also means it's less likely to screw up.

[chase314]

Got it, thank you! 

[morris_7]

Anyone find a solution besides the C wire theory? I’ll be removing my nest shortly is there isn’t one. It’s not worth the wear and tear on my equipment.

[Tbnilsen]

Google (“support”, at least) is locked in on that the problem doesn’t exist, or that the problem is the home owners heating system and not the Nest. There has been no Nest firmware release since May 2022, nor does it look promising that there will be one. On top of that, it is appears that Google isn’t going to include the new home automation protocol named Matter into “older” Nest models. 

[Yehanma]

With recent layoffs, maybe these tech people have been let go, so Nest may lose the capability to solve the issue. 
Anyone has a recommendation on a reliable replacement? Old school Honeywell?

[Tbnilsen]

Good choice.

[BarnesR]

Gen 3 learning thermostat worked fine for a couple of years. Short cycling issues started a few months ago, with or without C wire connected. Replaced flame sensor, and furnace control board $$ with no improvement. Going to give up and reinstall old school thermostat.

[BarnesR]

Our gas furnace only short cycles first cycle in the morning, when the schedule turns up the heat from 66 to 69 degrees. Removing the Y1 wire for air conditioning stops this cycling, but this makes the AC inoperative. The cycling occurs with or without the C wire connected. I read a post where someone noticed the cycling will stop if you use "heat+cool" mode (set cool to 90 to prevent AC from turning on). I can confirm this works in my case.

A lot other things can cause short cycling, but in this case it seems to be a software issue which is reproducible. Google support told me there is no way for them to report this to a software engineer.

[Engineer]

I find it humorous in a sad way that Google support cannot speak with Google engineering. The issue you are describing has been there since the last update in May of 2022. As the air conditioning compressor will also short cycle, many people have removed the thermostat as this is very damaging to the compressor.