Tesla vs pvwatts vs real life actuals

[bevo]

Tesla estimated my annual production to be 12,500 kWh for the year. After entering in azimuth, array tilt, and size in pvwatts, the estimate there is 12,300 kWh for the year. I left system losses and inverter efficiency at default so that's why the number may be lower than Tesla's.

I have less than 1 month of data but it's looking like I should be able to generate >13,000 kWh for the year (hopefully).

What is your actual production vs tesla estimates vs pvwatts estimates? Is it true that Tesla's estimates are conservative?

 

[wjgjr]

A while back, when a similar discussion came up, it seemed like most people got higher numbers from PVWatts than Tesla, which led to the idea that Tesla was being conservative. So, it is interesting that your PVWatts number was lower. In any case, I think that Tesla tends to be at least a bit conservative because they don't really want to deal with people calling, asking why their annual number didn't meet the estimate. Whether or not Tesla has changed that policy as they have gotten more real-world data, I don't know.

For our install, Tesla's number was definitely lower than PVWatts, though I concede I may not have gotten some of the parameters perfect for a solar roof. Right now, we are trending between the Tesla and PVWatts numbers, and it is pretty amazing how much weather has a huge impact on output - our Jan/Feb numbers seemed to be really low (below Tesla numbers), while March so far has been great (exceeding PVWatts,) likely due to the weather patterns. I also wonder if Tesla might do a better job of accounting for shading from other structures than PVWatts, as we do see some of that in the winter near sunrise/sunset.

While weather variability and shading may not be an issue for everybody, I would still say to give it a bit longer before being to confident on the numbers. We are at about 8 months, but I am waiting for a full before drawing too many conclusions about performance.

 

[Redhill_qik]

The PVwatts numbers for my PV array configuration also came in slightly under Tesla's estimates at 10,965 kWh vs 11,548 kWh (5.3% higher). I attributed the difference to the Tesla components have less loss than what PVwatts was defaulting to. Here are my results so far:

Month	Actual​	Tesla/PVwatts​	Pctg​
Nov	529.6​	486.7​	9.3%​
Dec	463.3​	383.7​	34.4%​
Jan	473.5​	433.0​	8.8%​
Feb	666.3​	495.6​	20.7%​

 

[bevo]

The PVwatts numbers for my PV array configuration also came in slightly under Tesla's estimates at 10,965 kWh vs 11,548 kWh (5.3% higher). I attributed the difference to the Tesla components have less loss than what PVwatts was defaulting to. Here are my results so far:

Month	Actual​	Tesla/PVwatts​	Pctg​
Nov	529.6​	486.7​	9.3%​
Dec	463.3​	383.7​	34.4%​
Jan	473.5​	433.0​	8.8%​
Feb	666.3​	495.6​	20.7%​

Looks like you are on pace to do substantially better

 

[Redhill_qik]

Looks like you are on pace to do substantially better

Looking at the PVwatts daily data shows multiple days with really low numbers 1.6-6.2 kWh during this period and the weather has just been better than that. I don't have any tree shading and only a small chimney that creates a shadow on 2 panels during late afternoon that PVwatts couldn't factor in to its estimates.

 

[wjgjr]

Looking at the PVwatts daily data shows multiple days with really low numbers 1.6-6.2 kWh during this period and the weather has just been better than that. I don't have any tree shading and only a small chimney that creates a shadow on 2 panels during late afternoon that PVwatts couldn't factor in to its estimates.

As a general rule, PVWatts daily data should not be used for comparison to actual results. It appears PVWatts generates numbers by using historical weather data, and the result is that if a given day happened to have been cloudier, it will show low numbers. Over the course of a month, and particularly a year, it is probably pretty accurate, but the daily numbers can be all over the place.

 

[charlesj]

Weather has a great deal to do with production especially with what the planet is gong through, warming that upsets historical patterns.
I am about to finish my 9th year of operation on the 29th. The 1st full year was the best then it was all over each year.

I have no idea how PVWatts handles weather historically and the previous year.

 

[Kodemonkey]

I was able to get PVWatts to come within a hundred or so of the Tesla numbers. So I have been comparing monthly output vs PVWatts and when it’s not raining I see almost 20% better actual production over the PVWatts. This only for a short sample of time so far, but I echo what others have said;”let the customer be pleasantly surprised”. Sunrun and Tesla had similar numbers when they quoted, sunrun being a little higher. But so far I am pleasantly surprised

 

[winfield100]

I would suggest you go to the PV watts site, and download "System Analysis Monitor" (link should be in upper right corner) that can give a more granular result
You can enter all your parameters, if you have a split system like mine have 2 arrays, enter latitude & longitude, enter specific panel types, (Mine are 37 Hanwha Q.Peak Duo 315) inverter(s) (Mine are 37 Enphase IQ 7) enter degrees tilt, enter degrees N, S etc (mine point 187 degrees S and 97 degrees easterly) on 2 flat planes with 1 a tiny bit shaded, enter your elec company and it's various rates, charges (my minimum is $21.69/month mandatory connection fee) etc
you can get very granular tho it is a bit of a learning curve, it's _free_ (you paid for it with your taxes if you live in the US)