Question about total_pack_energy from the Tesla Customer API

[barely]

I have a newly commissioned 8.19 kWh PV + 2 Powerwall system. The hardware itself was installed 7/16. It took a while for the building inspector to sign off on the install (he found chipped roof tiles that Tesla had to replace). PG&E sent the PTO about 8 days after Tesla submitted the interconnect agreement. That happened a couple of days ago.

Before the PTO, I spent time playing with the system. I set up a Grafana dashboard-based monitoring system to watch the operation of the system (thanks to Jason @ rhodesman/teslaPowerDash for the great template to work from).

The first thing I noticed is that my dashboard is showing 25.4kW for the pack's total energy. Looking at the JSON returned from Tesla's API:

curl --request GET --header 'Authorization: Bearer XXX' https://owner-api.teslamotors.com/api/1/products | jq .response[2].total_pack_energy

Note: I have 2 cars appearing first in the returned array, hence the index 2 for the energy site​

I see "25376". The spec for each powerwall is 13.5kW usable, so the pair should be ~27kW. I would expect the returned value, especially on a brand new install, to be a lot closer to 27000. What values are others seeing on their systems?

 

[Rhodesman]

First of all, you are welcome! Glad my guide on Github helped get you setup with the Power Dashboard.

Second, I too have noticed that the powerwalls total pack energy is far less than what I would expect them to have available, especially from new. Mine is reading: "total_pack_energy": 23543 right now and I believe it's been dropping since I got them (but I lost my original data from earlier this year).

I called Tesla about this discrepancy a couple months back and they claimed they "fixed" it as in it is just an improper reading of the total pack power. But I suspect mine has been slowly decreasing since I got them installed and I am 100% certain I've never seen them anywhere close to tesla's stated pack energy.

 

[Rhodesman]

I do not have a lot of data on this unfortunately due to a bug in my original power data API server that I didn't bother to fix until just now, but going back through the data that I do have, I found something interesting. If you take a look at the graph, you can see for the week of July 3rd, I put my powerwalls on backup only and then before and after I had them set to "self-powered" with a buffer of 25%. You'll notice that when the Powerwalls are in self powered mode, the total_pack_power starts to drop off. However, when I put the powerwalls in 100% backup only mode, they got a boost half way through the week in total pack power (23.92kWh to 24.27kWh).

So maybe, in some ways, we should be switching between self-powered and 100% backup? I dunno, for the winter months I had them set to 75% backup, 25% self-powered but considering how little sun we're getting, I might move that back to 100% backup and allow the house to either be on grid or solar when available.

Curious what others might have logged as well as yourself @barely .