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

"Sent To Grid" energy mismatch from Tesla app versus meter reading?

This site may earn commission on affiliate links.
I tried to do a search on this, but the wording is tricky, and I was coming up empty:

tl;dr:
We have a huge mismatch between the Tesla app "sent to grid" figure (91kWh) and our upstream "sent to grid" (41kWh) meter. Given our setup (see ASCII diagram below), is our house capable of consuming excess solar before any is sent back further upstream to the grid and getting registered by the upstream meter?

Background
We've had our small 4.1kW + 2 Powerwall system running on our detached garage/home office since October 2019. We intentionally eat most of what we grow, but now that daylight has returned to the northeast AND we're not driving due to self-quarantine, we're sending "a lot" (for us) more solar energy back to the grid.

The Numbers
Our "Lifetime > TO GRID" value in the Tesla app shows 91kWh (80kWh in 2020).
We have a separate, dedicated backfeed meter installed (thanks, Pennsylvania!) and the day they installed it, it read 0kWh. Today it reads 41kWh, or less than HALF of what the Tesla app claims has been sent back to the grid.

Help plz!
Does anybody have any personal data they can share that might help shed some light on whether this is "normal" or "abnormal" ?? I can understand a slight deviation of these values due to losses and rounding, but less than half seems wildly off to me.

Possible explanation?
I have one idea that this could explain what might be going on and it would be related to our somewhat unusual setup that involves our single grid feed coming in to a single meter, then splits into two 200 amp "Main House" and "Garage" breakers/feed (400A total service from grid).

Having hardly any understanding of electricity, does this theory make any sense?
I'm wondering if perhaps some portion or nearly all of our upstream solar from the garage is actually getting consumed by the main house before any additional overages are sent further upstream and hit our upstream meter? Our main house daytime instantaneous draw tends to be very low (0.2kWh - 2kWh) and consistent (about 18kWh /day) while the garage tends to be very "peaky" (0.2kWh idle - 12kWh instantaneous with car charging, on-demand electric hot water, electric heat pump, etc.)

Our setup
electric.JPG
 
Not totally understanding the "picture" above, but ignoring that for a second, can you compare the instantaneous readings for what the Tesla app says is going to the grid (at like 1pm on a clear day when the sun is at its peak and production [and house consumption] are hopefully stable), and compare it to what the meter is reading? I'm assuming it's a digital meter that gives a clear number, vs. having to count revolutions on the old-school wheel.

If what the app and meter report are way off then that's probably a different issue then if they agree on an instantaneous reading but differ long-term. I know some people have seen the solar instant reading match the inverter display, but the daily production be off by like a factor of 2, and I think those were some configuration issue on Tesla's side that they had to correct. I haven't heard of that on the grid reading, though.
 
  • Helpful
Reactions: Cirrus MS100D
That's a good idea! And something I should be able to test on an ideal (sunny) day, now that we're home 24/7.

My poor attempt at a graphic (I originally did it in ASCII in the post, then because it removed all my spacing, I took a screenshot and pasted it as a JPG) is to illustrate what I believe to be a somewhat unique arrangement whereby we only have a single utility line coming into a single utility meter that has a "backfeed" utility meter in parallel with it. From the "downstream" meter, we immediately split out utility power to (2) separate 200amp runs, one goes to the house main CB panel, the other runs underground about 300 feet to the detached garage CB panel.

Only the garage has solar and Powerwalls, but I'm wondering if it's "physically" possible for the solar overproduction to get consumed by the house before it makes it back to the main grid, which would explain our large divergence in kWh.
 
Only the garage has solar and Powerwalls, but I'm wondering if it's "physically" possible for the solar overproduction to get consumed by the house before it makes it back to the main grid, which would explain our large divergence in kWh.
That was my guess but I don't know where your differential meter was placed.
 
  • Helpful
Reactions: Cirrus MS100D
That's a good idea! And something I should be able to test on an ideal (sunny) day, now that we're home 24/7.

Heh. Glass-half-full guy, eh? :)

My poor attempt at a graphic (I originally did it in ASCII in the post, then because it removed all my spacing, I took a screenshot and pasted it as a JPG) is to illustrate what I believe to be a somewhat unique arrangement whereby we only have a single utility line coming into a single utility meter that has a "backfeed" utility meter in parallel with it. From the "downstream" meter, we immediately split out utility power to (2) separate 200amp runs, one goes to the house main CB panel, the other runs underground about 300 feet to the detached garage CB panel.

The update is much more clear, thanks.

Only the garage has solar and Powerwalls, but I'm wondering if it's "physically" possible for the solar overproduction to get consumed by the house before it makes it back to the main grid, which would explain our large divergence in kWh.

It's definitely true that the house will typically prefer locally-produced energy over the grid, since the source of the grid is so much farther away.

That seems like a really odd location of the production meter, given where the production source really is (I thought those meters were usually right at the output of the inverter, not farther upstream). Where is the Gateway in this picture?
 
  • Helpful
Reactions: Cirrus MS100D
Thanks, everybody!

Since I'm WFH today (lolz) I thought maybe a photo of the outside might help. I probably should've posted this earlier.
In this photo:

1 - Grid power coming in from the pole
2 - Our "normal" meter (only measures grid consumption)
3 - Our "backfeed" meter (only measures excess solar production going back to grid)
4 - 200A to main house
5 - 200A to detached garage

So the thing I guess that still isn't clear to me is how electricity can come out of 5 (excess solar) and flow/feed into 4 (main house) without "doing something" with the 2 "normal" meter. I guess what I'm learning from you all is that the excess solar is flowing inside that 2 "normal" meter box, but isn't registering to the "3" backfeed meter, but this is OK because even though power is flowing through 2, it also isn't registering any GRID consumption, thus we're HOPEFULLY not getting charged for producing and consuming our own solar.

So in a sense, we're not getting "credit" for sending any excess solar back into the grid from our garage because we're immediately consuming it (in our main house) before it reaches the meter, which nets out to be the equivalent.

IMG_1713.jpeg
 
Only the garage has solar and Powerwalls, but I'm wondering if it's "physically" possible for the solar overproduction to get consumed by the house before it makes it back to the main grid, which would explain our large divergence in kWh.

That is exactly what will happen. The PoCo meter sees the net usage-generation of both 200A breaker panels. If the gateway is monitoring the garage 200 panel, then it will read differently from the PoCo generation meter which records the gateway and the house.
Just like the demand meter will record both house and garage loads (assuming no solar/ Powerwall activity)
 
  • Helpful
Reactions: Cirrus MS100D
I have something similar and it confused the Tesla installers until I told them what it really was. What looks like a second meter downstream from the main utility meter is really only measuring one solar inverter and it's also used to communicate it's state to an outside service. Not quite the same as your situation but has the same effect.

20200102_150325.jpg


The far left meter on the dark gray panel is the utility co meter which shows the net for the entire premise. The meter just to the right of the gateway in the light gray panel is the meter that only measures the output from the inverter next to it.
 
  • Informative
Reactions: Cirrus MS100D