My solar production crashed! What happened?

[uscpsycho]

I have a solar roof and this is what my typical production looks like on a sunny day.

But this was my production today. Same type of sunny day and the curve looks identical but look at the solar generation and power production numbers! What in the world happened? Any ideas? Something is very very wrong. What number do I call to get this fixed?

I don't check my stats everyday but I'm so glad I checked tonight because this just happened. I'm glad I didn't discover the problem days after it started because I'm obviously pulling way more power from the grid now.

 

[Matt-FL]

I'd bet one of your strings is having issues. Is your system connected to a PW+ or the prior gen inverters?

 

[uscpsycho]

I have two PW+

 

[Matt-FL]

If you know the IP address of your PW+ or gateway, you can web into it, using the login credentials you'll find inside the device cover (username if your email used for tesla app). You should be able to click the 'System' link and view string voltages.

I'm on prior gen inverters, so I can't view string data without rebooting the device and viewing in the first 15 minutes. That's a great improvement with the PW+ inverters.

Here's a link to a reddit question, and a good response on how to access.

https://www.reddit.com/r/TeslaSolar/comments/qzehr9

 

[mek80010]

My solar system (Powerwall+ and Tesla panels) crashed last night also. The production data in the app had a hard cut off, I can't even see what was pulling from the grid past 2am.

My lone powerwall looked good (soon to be 3, my utility company will discount battery expansion, so only purchased one to get next Powerwalls with discount), had a steady green light,l (indicating no error) but my app said "no recent data available".

I got on Tesla chat this morning. They say they can communicate with the system fine, but can tell the inverter is intermittent. They ask me to contact the powerwall team 24/7 hotline.

I call and after another round of troubleshooting (and 40 minutes on the phone), I am told that there are some software updates taking place, and that they will email me tomorrow.

I still cannot see any data, including grid usage. It is prime time solar getting in July, and I'm just *****. System has only been in permission to operate for 6 days.

 

[Matt-FL]

Strange that the inverter would affect gateway data, especially at night when the inverter isn't producing. Then again, I guess the PW+ has the inverter and gateway in a single unit?

My system was originally designed with PW+, however my utility gave Tesla enough of a hard time about it that they were forced to redesign to the prior gen, separate inverters and gateway.

 

[jjrandorin]

What number do I call to get this fixed?

Tesla Energy supports phone number is 877-961-7652.

 

[uscpsycho]

So today my production looks like it is back to normal and I didn't do anything. Still gonna call Tesla to see if they can identify what the heck happened yesterday. If something is failing hopefully they can confirm it and order whatever I need before it goes totally kaput and I have a prolonged outage.

Is there something I can check myself to identify anything that might be failing? I know one day is a small sample size for troubleshooting but something was very wrong yesterday.

 

[uscpsycho]

Nope! Not back to normal, I didn't look carefully enough. Something is still very wrong today. In the morning it looks like I was not producing at full capacity, just like yesterday. Then I suddenly dropped to zero production for a period and then it surged back to normal. It's almost like something rebooted itself? This is not normal.

Anyone have an idea what happened?

 

[Matt-FL]

You will want to start monitoring and tracking by collecting data. To get that data, I'd suggest a micro-pc or raspberry pi and run Powerwall Dashboard which will poll your equipment and keep it historically for reporting purposes. That will be your best ammunition to prove to Tesla that something is wrong.

GitHub - jasonacox/Powerwall-Dashboard: Grafana Monitoring Dashboard for Tesla Solar and Powerwall Systems

Grafana Monitoring Dashboard for Tesla Solar and Powerwall Systems - jasonacox/Powerwall-Dashboard

github.com

This monitors a ton more data points than the Tesla app or the gateway web interface.

In that link, if you scroll down past the list of files, you will see a screenshot of the dashboard. Covers everything from production, to weather, to powerwall temperature, to string voltages (if PW+) and more. In my opinion, it's a must have for any Tesla equipment install.

 

[uscpsycho]

I called today and the person I talked to could see there was an issue but she couldn't diagnose it. She created a ticket and escalated to tier 2. She said they'll get back to me with whatever the issue is in 10-14 days. And then who knows how long it will be for any necessary repairs.

She said something like this could possibly be caused by a faulty firmware update. She said there are some things they can fix remotely so it's possible they won't have to send anyone to my house, but I suspect that's not the case and something needs fixing.

In the meantime it looks like I need to add another project to my to do list with the Powerwall Dashboard. Yay.

 

[CharlieDelta]

Hi, I'm a first time poster. I got my Tesla Solar and PW+ Installed in April and got PTO in late June.
I started having issues like you have right after I got PTO. For about a week it was happening daily and the Tesla chat support was saying it was probably new firmware updates while I contacted them 3 times that first week.

Well, the problem mostly calmed down until about two weeks ago. I documented 8 days where 1 or 2x a day, power would drop to zero and it would slow start back up over a 10-30 min period. I did a self check from the installer view and saw that upon restart one of my strings took up to 30 mins to sync and add power to the inverter.

So today, I decided to call the tech support number. I took a guess in selecting Powerwall support, since this is a combo solar and PW issue. The PW support person said they have the best view of data and knowledge for this type of problem.

What they found was a series of problems. All the power monitors and devices have been trying to update firmware at least 6 times and failed each time the power dropouts happened sometimes when the firmware tried to update. Grid quality issues were also shown as power dropping below 119V from PGE and the system getting confused into thinking it will be going into power outage mode but then not coming out of it when the voltage increases. I'm in a newish (25years) neighborhood and we've never had a power outage nor have I ever seen a voltage dip, usually 120-122V.

We did a reset of the gateway. The tech saw alarms and errors upon rebooting related to the firmware and also the wifi communications between the Tesla nodes. She stated that sometimes they have to come out and hardwire the communication links vs relying on the gateway wifi.

In summary, there seemed to be a combination of problems, and I also have to wait up to 15 days to hear from Tier 2 support and what they discover is the issue and resolution(s).

 

[CharlieDelta]

Examples of dropouts:

 

[uscpsycho]

It sounds like the issues you have don't necessarily have to do with powerwall (unless you consider the PW+ inverter part of the PW) but PW support was the most helpful. I don't think my powerwalls are the cause of my problems (unless it's the inverters) because I keep my powerwalls at 100% and that's where they stay. But I think I'm going to try what you did and talk to PW support.

 

[CharlieDelta]

Yes, I am considering the PW and inverter as one unit in terms of getting support. In some of my drop outs of the solar production to 0, the PW also changes reporting from 100% to 0% and back. So I see a zig zag of PW capacity. An odd behavior in the app is that there is some smoothing going on in terms of the graphing that happens days later. The app has wiped out all those drop outs on the PW reporting and put it back to 100% (which that was the true charge state at the time, it is just the reporting dropped to zero). I’ll see if I can find examples.

 

[CharlieDelta]

Here are the screen shots as example from my June 21st day taken at the day of the anomaly to what the reporting looks like today.

 

[uscpsycho]

Thanks for sharing that. I don't have the dropouts on my powerwall graph that you have but I have PW+ so I'll call PW support and hope some troubleshooting will identify the problem before Tier 2 gets back to me.

 

[taphil]

There was someone who recently posted that the dropouts in the graphs were fixed after they installed a wifi access point or extender near their PW/Gateway, making it seem like the dropouts were simply missing data points due to no data uploaded to Tesla from poor wifi reception.

 

[uscpsycho]

There was someone who recently posted that the dropouts in the graphs were fixed after they installed a wifi access point or extender near their PW/Gateway, making it seem like the dropouts were simply missing data points due to no data uploaded to Tesla from poor wifi reception.

I have an access point less than three feet from the gateway so that's definitely not the problem here.

But I seriously doubt that user's conclusion is correct. If so, we would all have dropouts like this every time our internet went down or we had to reboot our router or modems. Besides, doesn't the Tesla app pull data from the local devices rather than from the cloud? That's what I always assumed.

 

[offandonagain]

But I seriously doubt that user's conclusion is correct. If so, we would all have dropouts like this every time our internet went down or we had to reboot our router or modems. Besides, doesn't the Tesla app pull data from the local devices rather than from the cloud? That's what I always assumed.

The Tesla app pulls data from the the cloud (based on observations, and also it wouldn't work otherwise outside of your home network). So it does require the gateway to have internet connectivity and report its metrics to the cloud.

That said, I agree the conclusion is unlikely. We've seen reports that the gateway is capable of buffering days of metrics without internet connectivity, and can backfill those metrics once the network is back up. I suspect a more likely cause are firmware updates: production does drop momentarily while the firmware is being updated. A network issue might cause repeated failed firmware updates and thus repeated drops in production.