# What's a reasonable expectation for efficiency in modern solar panels?

I recently had a solar array installed on my roof with four different sections connected to three inverters - the total array includes 56 panels for a total of 13.44 kW.

However, I'm not seeing anywhere near that much production, even during the peak of the day. I obviously wouldn't expect to product 13 kW, as the panels aren't 100% efficient, there are clouds, the inverters lose 10% or so, etc. However, what's a reasonable expectation for the system production or an array that size?

What I'm seeing is that production will rise pretty quickly in mid-morning to 6.68 kW, and then sit there (at that exact number) until late morning. Then it will jump pretty quickly to 7.38 kW, hang for a while, and then up to 7.7 kW once all banks of panels are in the sunlight, and then it will sit there until early afternoon - however, that's only 57% efficiency for the system, and that seems a bit low.

Here are some questions I have:

• Maybe I mis-understand how the panels work. I'd thought that they'd produce more power when the sun is directly over them, as opposed to at an angle, but the production numbers don't seem to support that. Is that correct?
• Obviously the panels have a maximum production level, but could my inverters be limiting total production? If my panels are making 10 kW but the inverters can only process 8 kW of it, is that something I can see happening?

I realize there a lot of variables to get my exact number of reasonable production, but I'm just wondering what's reasonable and if it's something I should dig into a bit more to ensure that they were set up correctly.

-

If you're in the northern hemisphere, the lower angle of the sun in the winter is probably contributing to your low numbers. (When the sun is low in the sky it passes through more atmosphere to reach the ground.) Presumably you'll do better in the summer.

Also, the tilt angle of the panels should be matched to your latitude. If your panels are mounted on your roof and aren't at the optimum angle (or aren't pointed directly south), that would decrease your efficiency. In any event, unless you have tracking panels or constantly change the tilt, your panels will only be pointed exactly at the sun twice a year, and at all other times will be slightly misaligned.

This website has a detailed description of tilt angles, but the net result is that if you have a fixed array at the optimum angle, you'll only be getting 71% of the exposure than if you had a tracking array constantly pointed at the sun. If your angle is not optimum you will do even worse, obviously.

EDIT: I should also point out that your definition of "optimum angle" for a fixed array is dependent on what you're trying to optimize. There are a couple of ways to look at it:

1. Generate as much power as possible over the entire year. The 71% refers to this option.
2. Generate as much power as possible in the summer, when electricity usage may be higher due to AC and/or power may cost more to buy from the grid. This will be a lower angle [array pointed more upwards] than 1), since the sun is higher in the sky in the summer. Winter usage will be lower than 1).
3. Try to generate power evenly throughout the year. This will be a larger angle [array pointed more towards the horizon]. This approach tries to increase power generation in the winter relative to 1), at the expense of some summer power.
-
I'm in Arizona, so pretty solid sun. So, just to confirm that I understand it properly - let's say my panels are set up pretty optimally (they are), but they're stationary. If it's a 13.5 kW system, and I have 71% production because they're stationary, that's only 9.6 kW. If I lose another 10% in the inverter, that drops me to just over 8.6 kW, which isn't much more than I'm currently seeing. Based on that, it seems like my production is pretty much in line with what I should be expecting. – SqlRyan Feb 14 '13 at 22:34
You've basically got it. The 71% is an annual average. At any given moment the actual array output may be higher or lower based on how closely you're aligned to the sun direction. Twice a year your array should be pointed directly at the sun, and you should get around 100% at that moment (minus any other losses, etc.). – Henry Jackson Feb 15 '13 at 5:10
Interestingly, here in Scotland, we would have "Generate as much power as possible in the Winter" as most important, as that is when our electricity usage is highest. – Rory Alsop Feb 15 '13 at 8:52

You have fifty-six 240-watt modules. They each produce 240 watts under standard test conditions (STC). STC is defined as 1000 W/m^2 of irradiance, 25ºC temperature and AM1.5G spectrum. These conditions are rarely reached outdoors. If the temperature is higher or the irradiance is lower than STC, you'll get less power.

You can usually only achieve 25ºC module temperature in full sun if the air temperature is close to freezing, so in Phoenix you will probably never get this. The temperature coefficient of power for your modules should be listed on the data sheet. For silicon it is about -0.45%/ºC. So when it's 120ºF outside and the modules are running at ~25ºC above that, the temperature alone will cause the array to operate at ~22% below the nameplate rating, even if everything else is perfect.

The irradiance you care about is plane-of-array (POA) irradiance, which is the amount of sun that falls on an area that is tilted to the same angle as the array. By residential standards your array is absolutely enormous. If it is all at a single tilt angle (close to latitude-tilt) and if it's all at a single azimuth angle (pointing south), you might expect to reach nearly 1000 W/m^2 POA irradiance on most sunny days during all times of year. If different parts of the array face different directions, one will usually perform better than the others. If this is not considered during the design by using multiple inverters or a multiple-string inverter, you could be losing power from this mismatch. And, as others have mentioned, if the azimuth angle is not due south and the tilt angle is not close to latitude-tilt, you'll lose power because of the sun's low angle of incidence on the array.

You also pointed out correctly that even just connecting the modules to each other and to an inverter can further reduce power. If any of the modules are partially shaded, you lose more power than just the fraction that is shaded. If your inverter is undersized or is operating at very high temperature, it will reduce the amount it produces by operating the modules further away from their maximum power point. You can check the capacity of your inverter on its data sheet. If it's below the total power rating of the array, send it back.

Use this site to see the temperature and irradiance conditions in Phoenix. Note that you'll have to calculate POA irradiance from the direct and horizontal irradiance values given here.

Use this page to calculate the position of the sun. If you put the surface azimuth rotation and surface slope angles in, you can get the angle of incidence between the sun and your array.

-
Thanks for all the additional information. The array is pretty much my entire roof, and it's broken into three inverters to match the segments up by angle, so it seems that's done correctly. Also, I was told my array would have lower efficiency during the summer (when tempts get into the 110's), so it's good to know the effect this is expected to have. Thanks! – SqlRyan Feb 17 '13 at 4:14
The three sub-arrays explains why you never achieve the array's rated peak power: all three arrays are never pointed right at the sun at the same time. But, as discussed in other answers, you'll get a lot more energy than if you just had a 7.7kW array. – ArgentoSapiens Feb 18 '13 at 1:10

That does seem a little bit low but not crazy. I have a 3.3 kW DC array on my roof and during the peak times of the day I can get 2.7 kW AC after the inverter, or about 80% efficiency. I live in Northern CA though, where you can have cool and sunny days in the summer.

Since you are in the US, one thing you should check out is the PVWatts calculator. This little tool will take a lot of data from you such as panel size, rated efficiency, installed angle, inverter efficiency, etc and then will cross reference that with historical solar data for your specific location (latitude/longitude) to give you an estimate of how much you should be generating on a month-by-month basis. It's probably the most accurate tool available. If you aren't in that ballpark, it may be time to talk to your installers.

-