For each of the 4 nodes (Battery/Solar/Load/Site) you have import and export, which indicates how much energy each node produced or consumed.
Did you mean to say “consumed or produced” here? That is:
“import” = “consumed”
“export” = “produced”
(By the way, thank you for your post. I was trying to help my 70+ dad set up his energy dashboard and we were having the hardest time trying to figure out what each of the sensors actually meant. Your post helped solve our problem! I’ll try to put up a PR to help improve the powerwall integration docs with your explanations)
Mine matches yours exactly. What doesn’t seem right in your dashboard?
Full disclosure, I don’t know if everything in my Dad’s dashboard is lining up correctly with the app and I’m no longer at their house to verify. Our main problem we were trying to resolve was why the “Solar” bubble was saying 0 kWh even when the app showed that solar was producing energy. Turns out we were using “Solar Import” instead of “Solar Export”.
But, after reading this thread, I noticed a few other sensors that were likely incorrect and changed them as well.
Hello,
I’m having problems with Energy Management configuration with Tesla Powerwall.
I’m trying to understand if data is correctly reportd, using Tesla App as a benchmark.
As per my understanding of the official documentation:
60,6 kWh consumed in total
43,7 kWh from the gird
7,2 kWh from Batteries
9,0 kWh from Solar Panels (direct usage)
But if you do the math 8,3 (gone into the battery) + 9,0 is less than 18 kWh generated.
From my Tesla App I see there was a small return to the grid (18 - 9 - 8,3 = 0,7 kWh), that is not reported here, in the Energy Distribution (Return to the grid is “empty” in configuration)
What am I missing?
Had anydody of you experienced same problem?
What integration are you using for your Powerwall data? I have repeatedly tried the included Tesla Powerwall integration but keep getting logged out after several minutes.
How is everyone creating utility meter sensors to measure daily home energy usage? I’m defining home usage as the actual energy consumed by my home every 24 hours.
For the last year, I’ve had a utility meter helper configured to use the sensor.powerwall_load_import sensor as the source for daily calculations of my home energy usage and I want to say it has been rock solid. It’s matched the readings in the Tesla app and also what I see in the Energy Dashboard under the home icon on the right.
I noticed today that this daily home energy sensor I setup is reporting 64.2 kWh of usage. This seems exceptionally high considering I generated 70.9 kWh from solar, had nearly a full battery at the start of the day, and neither of my cars charged.
After looking at the Tesla app, I see my home used as 40.1 kWh. My solar production contributed 24.1 kWh, which leads me to believe that this sensor is somehow adding the two values to report the 64.2 kWh of home usage today. However, sensor.powerwall_load_import shows a midnight reading of 25,998 and a current reading of 26,038, so a diff of 40 kWh, which is expected. And now that I look into yesterday (also high solar production), I did not consume 97.2 kWh. The Tesla app shows 68.9 kWh home usage…
I’m scratching my head on this as I have changed nothing on this utility meter sensor since I set it up last year and want to say that most readings have been correct, but today looks odd. So, I’m wondering if an HA update or something has changed without me knowing that I need to adjust now.
I really wish I could just extract the home value from the Energy Dashboard as that’s the value I want.
How are others measuring this value with a Powerwall & Solar?
Some time in early March my basic daily cycle utility meters started to over-report on load (but seemingly not on solar generation or site export/import).
The load utility meter is on sensor.powerwall_load_import.
I opened this bug and incorrectly suspected it might be day light savings related, but I have since found another workaround - creating a Reimann Sum Integral Sensor off of sensor.powerwall_load_now instead of using sensor.powerwall_load_import directly. Another person has found that toggling “net consumption” works for them on the utility meter…I just created a sensor to test this to see if it helps. It just doesn’t make sense to me given that the historical data on the sensor does not match the expected daily values.
Does anyone know why this net consumption toggle would make the difference? The history on the sensor doesn’t show appear to go down, so it doesn’t make sense to me that this would be required.
I suspect the issue is with the powerwall plugin (or a recent powewall firmware update) as I can see the powerwall_load_import sensor repeatedly increasing and decreasing.
It would also explain why “net consumption” works…
On further digging, this looks like a powerwall firmware issue - the values reported from the API occasionally go backward - this is from the api/meters/aggregates API:
Are the “import” and “export” cumulative sensors in the latest Powerwall Integration still not accurate? Also, I noticed that sensor.powerwall_load_export=unavailable. I wasted a lot of time presuming I could use the cumulative sensors directly from the powerwall integration; only to find it wasn’t adding up correctly.
I didn’t know what else to do, so I just created integration sensors to collect real-time data from the powerwall integration; followed by utility_meter “daily” sensors:
sensor:
- platform: integration
source: sensor.powerwall_solar_power
name: "Daily Solar Energy Produced"
unit_prefix: k
round: 2
- platform: integration
source: sensor.powerwall_battery_power
name: "Daily Battery Energy Discharged"
unit_prefix: k
round: 2
- platform: integration
source: sensor.powerwall_site_power
name: "Daily Grid Energy Consumed"
unit_prefix: k
round: 2
utility_meter:
daily_solar_energy_produced:
source: sensor.daily_solar_energy_produced
cycle: daily
daily_battery_energy_discharged:
source: sensor.daily_battery_energy_discharged
cycle: daily
daily_grid_energy_consumed:
source: sensor.daily_grid_energy_consumed
cycle: daily
