Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

How should the powerwall charge during an outage?

This site may earn commission on affiliate links.
We had an outage on Tuesday:

Shygar 7.500kW | Live Output

When looking at it, it seems like even though the sun was out, there wasn't enough solar coming in to charge the battery. But the battery was still at a pretty high level.

Does anyone know how this is supposed to behave? Does it not charge until a certain battery percentage or something like that? Just wondering what expected behavior of charging while using only battery power is supposed to work.
 
We had an outage on Tuesday:

Shygar 7.500kW | Live Output

When looking at it, it seems like even though the sun was out, there wasn't enough solar coming in to charge the battery. But the battery was still at a pretty high level.

Does anyone know how this is supposed to behave? Does it not charge until a certain battery percentage or something like that? Just wondering what expected behavior of charging while using only battery power is supposed to work.
It looks like your panels/inverter shut-down as it somehow wasn't seeing a valid AC wave-form on the line.

The generation side of the data should be unrelated to any consumption whether it be loads or PW2 charging. Your generation data went to basically 0.
 
We are in the same neighborhood of the "planned" PG&E power outage. The Powerwalls were able to turn on and off our Enphase microinverters just fine.

PG&E Working on the line:
20180417_115523_resized.jpg

Tesla app during the outage:
Screenshot_2018-04-17-10-31-54.png

Tesla app power flow history:
Screenshot_2018-04-17-17-02-10.png

The Powerwalls seem to drain to 95% with the solar off, turn on the solar till 100%, then repeat.
 
Thanks @GenSao I was curious what happened to yours also. So if that's true that it drains to 95%, then I don't think mine ever actually got to 95% in the 5 hours (no one was home using power). I have a single Delta inverter. Now that I have my car, I can simulate an outage and a big load to see if mine behaves the same way as yours.

When you say "planned", did you get a notification about the work? They are supposed to upgrade the transformer from my request to switch to the EV-A rate (although it was already scheduled from my solar install last year).
 
When you say "planned", did you get a notification about the work? They are supposed to upgrade the transformer from my request to switch to the EV-A rate (although it was already scheduled from my solar install last year).

PG&E's outage website that morning stated it was a planned outage in the area. I never got a notice. =(

If I did, I would have permitted the battery to drain down the day earlier so solar production would not be wasted. I currently am on backup only mode.

I updated my outage notifications with PG&E online for next time.
 
Thanks @GenSao I was curious what happened to yours also. So if that's true that it drains to 95%, then I don't think mine ever actually got to 95% in the 5 hours (no one was home using power). I have a single Delta inverter. Now that I have my car, I can simulate an outage and a big load to see if mine behaves the same way as yours.
This is the confusing part. Your PVoutput said you dropped to 94.239% (from 100%) at 10:45 and gently slipped to 91.110% at 14:00.

Then, your generation came back at 14:05, and battery started "charging" and hit 100% at 14:25.

So unless reserve-level, TBC, or something else, was preventing the battery from topping off in those 3+ hours ...
 
This is the confusing part. Your PVoutput said you dropped to 94.239% (from 100%) at 10:45 and gently slipped to 91.110% at 14:00.

Then, your generation came back at 14:05, and battery started "charging" and hit 100% at 14:25.

So unless reserve-level, TBC, or something else, was preventing the battery from topping off in those 3+ hours ...
My reserve is set to 20% and set to self powered mode. I wonder if it's what the API reports vs what is reported in the app. There is about a 3% difference or so between the two numbers, which I'm guessing is reserved for something. So my 91% may really be almost 95%.

I'll try to turn off my grid breaker this weekend and charge my car to see what happens.
 
  • Like
Reactions: NuShrike
@Shygar Maybe they don't have the correct inverter and grid control scheme set when it was commissioned?

I also have an enphase system and can confirm that mine come back online somewhere around 90 to 100%
I think one time when I was testing I saw 95% and then the second time it was around the 90% mark, so I'm wondering if perhaps it depends on the house load.
 
  • Like
Reactions: Shygar
@Shygar Maybe they don't have the correct inverter and grid control scheme set when it was commissioned?

I also have an enphase system and can confirm that mine come back online somewhere around 90 to 100%
I think one time when I was testing I saw 95% and then the second time it was around the 90% mark, so I'm wondering if perhaps it depends on the house load.
Yea good question. I'm hoping they do as the solar and powerwall was all done by Tesla/SolarCity on the same day.
 
We had an outage on Tuesday:

Shygar 7.500kW | Live Output
Side-thread, why does your production fall off so drastically after 3pm?

I'm looking at your production graph (on a good day such as Shygar 7.500kW | Live Output) vs optimal insolation -- for your reported mount azimuth and tilt.

At 16:15, you should be getting 3933W of power, but you're only generating 1872W. The delta is even wider at 17:00.

Shade probably?
 
Playing around with it today:

Shygar 7.500kW | Live Output

I turned off the breaker for most of the house, but I had to drain it with the car first. I don't have my whole house backed up and I don't have a main breaker I can throw to test. Seems like it will charge periodically during an outage.
That looks normal now. PW panel curtailment stopped at 10:45 @ after 10 minutes at 85%? That's when you were done draining.

Then, PW kept topping itself from there off to hold 97% until you turned everything back on at 13:15. The battery then charged to 100%.

Why didn't it do this the other day?
 
My guess is that depending on PV system size, that maybe it would have to go to a lower SOC before it starts up the PV inverters again. My rational is that with a high SOC, perhaps charging tapers off, and without anywhere to dump the power with a 7.5kW system, then where might the excess power go if the powerwall has to limit charging current perhaps when SOC is high?

Also, depending on how quickly the battery drains based on load, when it does hit a certain SOC (lets say 90% for arguments sake), by the time the frequency control shifts back to 60hz again, that it might be 5 min before the inverters sync and output again, and with a 5 min delay, the battery % may have dropped a varying amount. The amount below 90% could be a little or a lot depending on the battery load I guess.
 
Last edited:
PG&E's outage website that morning stated it was a planned outage in the area. I never got a notice. =(

If I did, I would have permitted the battery to drain down the day earlier so solar production would not be wasted. I currently am on backup only mode.

I updated my outage notifications with PG&E online for next time.
Lovely; that means you make way more solar than you use energy. I'm in the opposite boat on solar, having both insufficent solar and insufficient battery. (5.8kWp solar, 2 PowerWall 2's, one room electric heat in cool coastal climate, electric water heating, and electric clothes dryer; the gas oven and gas home heater both have electric elements that also run while in use.)

PowerWall Home PowerFlow

I wanted to get thrice the solar and thrice the batteries, and convert all energy use to electric, but I ran out of money, space, and permission.
 
Last edited:
My reserve is set to 20% and set to self powered mode. I wonder if it's what the API reports vs what is reported in the app. There is about a 3% difference or so between the two numbers, which I'm guessing is reserved for something. So my 91% may really be almost 95%.

I'll try to turn off my grid breaker this weekend and charge my car to see what happens.
I came here to respond to that specifically: in firmware version 1.15.1 of Tesla PowerWall 2, there is a 5% difference between what I see in my graphs at PowerWall Home PowerFlow and what the iPhone app version 3.3.5 (a80e894b4) says according to this one data point: just now, the app said the batteries were 17% full, and my graphs said 22%. When I started collecting data for my graphs, the spread was 0%. They steadily increased it up to 5%. That must be to save battery longevity.

(Yesterday was cloudy, today looks cloudy, my batteries are low, and I'm about to wash and dry clothes, so I'm about to use a bunch of electricity, and I'm hoping I can drain it below my 17% (really 22%) setting to 8% (really 13%?), and hoping with luck tomorrow when the whole household is at work, the sun comes out and recharges them.)