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

Difference Between App From Grid kWh and Utility Energy Use kWh

This site may earn commission on affiliate links.
That’s interesting. Do you know if you were on 1.50 or 1.50.1? I guess if it was fixed in 1.50.1 they would have put you on that.

I’m on 1.50.1 and I just compared the data in my Tesla app to my utility over the past few days and they were all within 2 or 3%, which I think is totally acceptable. One thing I did note that I thought was kind of interesting is that some days my utility was slightly higher than the Tesla app and some days the utility was slightly lower than the Tesla app.

Was there any obvious correlation to your net usage? I mentioned elsewhere that my usage has consistently been higher according to my utility than the Tesla data, with the daily difference being roughly between 0.1 and 0.7 kWh. But the higher my net usage, the higher the difference.
 
Was there any obvious correlation to your net usage? I mentioned elsewhere that my usage has consistently been higher according to my utility than the Tesla data, with the daily difference being roughly between 0.1 and 0.7 kWh. But the higher my net usage, the higher the difference.

There didn’t seem to be a correlation at all, although the numbers for the past few days have been pretty similar. Also, it’s worth noting that I’m not comparing the net usage, but rather the amount of power I pulled from the grid. For some reason my utility only allows me to see usage data for the power that I pull from the grid in their online system. I can’t see my net usage or the power I feed into the grid at all.

But for example, here are my numbers for the past 3 days. The Tesla app is listed first:

23.1, 23.22
28.3, 28.46
22.8, 22.75
 
Hi, I'm also experiencing a similar challenge.

I recently completed inspection/granted PTO for an 8kw Solar Roof and am seeing my utility reporting a much higher draw than what the Tesla App is showing (often a 20-25% discrepancy). Below is a daily breakdown of what my utility is showing vs what tesla is reporting. Everything is on the backup circuit fyi.

Any insight would be appreciated.
Ba8EV7eTT8BWwhJ.png

IMG_6066.PNG
 
I have a solar roof (9.9 kW) installed in December and also noticed my utility reporting much larger grid values than the Tesla app (via powerblaster for me since I do not have powerwalls). I'm slowly debugging things, but have yet to determine what is happening. My hunch is that the utility/meter values are correct and that the problem is with either the powerblaster's connectivity or its CT setup.

I've connected directly to my inverter at various times and its solar production numbers match those in the Tesla app. Earlier this week, I pulled data from Tesla's Energy API and learned ...

The calendar_history API shows that home usage is derived from stored grid_power + solar_power (home usage is not stored independently).
The zero home usage values I was seeing in the app and in the app csv export were not really zero, but were actually small values rounded to zero because the app only shows hundreds of watts (tenths of kW).
The API power value unit is watts, allowing you to see the small daytime usage that gets dropped by the app with rounding.

For the day I checked, the app "lost" 700 watts over 5 daytime hours.

So, there are some rounding errors in the displayed Tesla app data. However in my case, the rounding errors are too small to account for the large difference I'm seeing between the utility and tesla app grid numbers.

I don't have any easy way to check daily utility values since my utility only sends me a monthly total on each bill. I've been running outside to my meter to record numbers from the meter display. The couple of days I've checked, the Tesla grid numbers are ~65% of the delta I see on the meter - big difference! The difference is so big that I'm worried I'm miscalculating something somewhere.

I'm still investigating, but am eager to see what others learn (or if anyone has suggestions as to what might be wrong).
 
I have a solar roof (9.9 kW) installed in December and also noticed my utility reporting much larger grid values than the Tesla app (via powerblaster for me since I do not have powerwalls). I'm slowly debugging things, but have yet to determine what is happening. My hunch is that the utility/meter values are correct and that the problem is with either the powerblaster's connectivity or its CT setup.
I feel like this is the most likely answer as there really should not be any big differences, and it seems more likely something would be missed on the Tesla side.

I mentioned above that I see differences, and they appear to be directly correlated to my net usage - the higher the net usage, the higher the difference, but the largest difference I've seen is about 0.75 kWh for a day. Attached is a quick chart I threw together showing my utility (PEPCO) data versus the Tesla grid data. As you can see, PEPCO always is above the Tesla number, even when net usage for a day is negative, but there is a clear correlation between net usage and the discrepancy.

Tesla vs PEPCO usage.png
 
  • Informative
Reactions: trautmane2
I feel like this is the most likely answer as there really should not be any big differences, and it seems more likely something would be missed on the Tesla side.

I mentioned above that I see differences, and they appear to be directly correlated to my net usage - the higher the net usage, the higher the difference, but the largest difference I've seen is about 0.75 kWh for a day. Attached is a quick chart I threw together showing my utility (PEPCO) data versus the Tesla grid data. As you can see, PEPCO always is above the Tesla number, even when net usage for a day is negative, but there is a clear correlation between net usage and the discrepancy.

View attachment 634301

This all sounds super similar to a problem our friends across the pond were having with their smart meters: Beware SMETS2 meters | Tesla Motors Club

My understanding is that the meters were incorrectly configured to measure total power (reactive + real) instead of just real power. Since residential usage is almost entirely real typically billing is done with real but if the meter is configured for total you'll see higher bills compared to what your CTs measure.
 
This all sounds super similar to a problem our friends across the pond were having with their smart meters: Beware SMETS2 meters | Tesla Motors Club

My understanding is that the meters were incorrectly configured to measure total power (reactive + real) instead of just real power. Since residential usage is almost entirely real typically billing is done with real but if the meter is configured for total you'll see higher bills compared to what your CTs measure.
I had thought about that, but the magnitude here seems to be much smaller than what they found, so if it is something similar, at least the effect is going to be much smaller (on the order of $10/year) even if the Tesla numbers are actually perfectly accurate. And, at least some part of it is likely loss somewhere between the gateway and the meter, since even with negative usage, PEPCO exceeds Tesla numbers. (Quick linear regression in excel puts it at 0.270 + 0.011 * net usage.) This suggests (perhaps) that roughly a quarter of a kW is lost daily somewhere between what Tesla is measuring and the meter, and then there is a smaller adjustment for net usage. I do not know how this would compare to others who seemingly have good numbers, as opposed to those above who clearly have larger issues with the data they are getting.