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

Why did my Powerwall not satisfy all house demand

This site may earn commission on affiliate links.
This morning I noticed that my Octopus Smart meter showed energy usage from the grid when I expected it not to. The Powerwall had capacity and had already started feeding the house during the grid peak time. But yet at one point there appears to be draw from the Grid. The power draw is well within the capacity of the Powerwall, so why did the Gateway 2 not just let the Powerwall deal with the Load demand, but chose to pull power from the Grid.
You can see in the next graph the battery does kick in later to satisfy demand but it's almost like it's not reacting fast enough to the house's demand.
 

Attachments

  • Screenshot_20231114_081511_Octopus.jpg
    Screenshot_20231114_081511_Octopus.jpg
    91.1 KB · Views: 44
  • Screenshot_20231114_082027_Home Assistant.jpg
    Screenshot_20231114_082027_Home Assistant.jpg
    142.2 KB · Views: 26
The two screenshots tell the story. The first shows you used 1.00 Wh of energy from the grid (which is very little) at about 7:47 while the second shows a peak power usage of about 1200 W during the same time. I suspect there was a brief transition where your system was switching in the battery to support a large load. During that transition, the system took some time to enable battery power causing a spike in demand for grid power, just as you suspected. The 1200 W of grid power demand was very brief since it averaged over an hour to be only 1.00 Wh of energy, about a factor of 1200 less. This implies the spike lasted about 1/1200 of an hour or about 3 seconds. After that brief period, the battery kicked in to supply the power your home needed. I’ve seen this behavior with my system as well and I think it’s normal. I wouldn't be concerned since the cost of 1 Wh of energy is minuscule.

Btw, I think the second graph is deceiving since it implies the 1200 W lasted for 1/6 of a minute (10 seconds), not the 3 seconds I calculated. But 10 seconds of power at 1200 W would result in an energy consumption of 3.3 Wh, not 1.0 Wh as shown on the first graph. The 10-second time interval is probably just the sample rate for the power measurement, not the true time that the 1200 W actually was being drawn from the grid.
 
Last edited:
You can see in the next graph the battery does kick in later to satisfy demand but it's almost like it's not reacting fast enough to the house's demand.
I wouldn’t be surprised if there is a delay responding to a sudden load. Don’t worry, it responds faster with the grid disconnected.

When grid connected, the Powerwall has to “push” against the flood of grid power to keep net flow at zero, but not so hard as to be exporting power. This delicate act is like balancing a broomstick on the palm of your hand. Then someone smacks the broomstick with a sudden load and the Powerwall takes a moment to adjust.

Without the grid, a sudden power load will literally pull what it needs out of the Powerwall. This is akin to what a cranking (non-EV) car does to a car’s 12V battery when it starts. No delicate balancing act, just demand and supply.