I am going solar

They're not wrong, though. I have 8 panels at 260W each, for a total of 2.08 kW. I don't have microinverters, but rather a standard inverter that can handle 3 kW, so my system never clips. But I can tell you that it rarely gets that high. The all-time maximum (measured as an average over some number of seconds - I'm not sure exactly how many, but probably 15-60 or so) is 2.2 kW - ironically, on a partly-cloudy day. (When the sun shone through the holes in the clouds, it got full direct sunlight - plus all the light that was diffusing through the clouds all around the sun, hence more light than you'd normally expect.) The most I've ever gotten in a 1-hour period is about 1.7 kW. So that means that my peak sustained power is about 82% of the rated 2.08 kW - and 82% of 310 W is near enough to 250W to make no difference. So it's not unreasonable to assume that if I had 310W panels and the 250W microinverters, I'd rarely or never see clipping.

That said, YMMV somewhat - my panels face east (although that's only maybe a 5-10% penalty - they're slightly more efficient in the cooler morning temperatures, slightly less efficient due to the common morning fog, and the fluctuation between winter and summer is higher than for a south-facing array). But I am in California, and we get lots of sunlight - certainly more than you would in many of your locales (like the Netherlands, PA, or NJ).

And even if there is clipping, it's not losing a huge amount of power. I get maybe 3 hours of near-peak power production in the middle of summer. Let's say for some reason, your sun is a lot brighter than mine, so your 310W panel actually produces 310W during those 3 hours, but the inverter can't handle it all. So you get 750 Watt-hours instead of 930 Watt-hours - a loss of 180 Wh. But at least for me, I only produce about 40% of my daily power during those 3 peak hours - the other 60% of my daily power is produced during the rest of the day. So in this case, that'd mean that a perfect inverter would get you 2,325 Wh, or the 250W inverter would get you 180 Wh less, or 2,145 Wh. You're still making better than 92% of the power you would have if you had a more powerful inverter! And that's on the brightest sunniest day of the year! In the spring, fall, and winter, when the sun is lower, you'll almost certainly never have clipping, so you'd probably lose at most 1% of your power. I can guarantee you that upgrading the inverters to get that last 1% of the power will increase the total system cost by well more than 1%. If you had two systems for the same price - one with 310W panels and 300W inverters, the other with 310W panels and 250W inverters - but one extra panel - the system with the extra panel would almost certainly produce more power.

So yeah, if your system clips, you might feel a little sad that you're losing out on some power - but trust me (I'm an engineer!), it's not very much.

That said, if/when they do offer a higher-power microinverter, I wouldn't be surprised to see it marketed as Jan suggests - but not because you really need the extra capacity, just because it'll make you feel better, and they can get you to pay more for the better inverter.

 

Not only were the panels briefly getting extra sunlight, they were also COLD from all that previous time covered by clouds. Cold panels produce more, until they start warming up in the sun and lose a bit of efficiency.

 

In our areas the following picture is often used to estimate the annual loss percentage when your panels have no perfect orientation.

​

The thin concentric circles indicate the angle, from 0° (horizontal) in the middle to 90° (vertical) at the border. The radial lines indicate the facing direction. The thicker white curves indicate the situations with equal estimated annual loss percentages.

In our case the ideal orientation indicated by 100% is directed to South with an angle of 35°. Facing East with the same angle would give a loss of just a bit more than 20%. But for facing East it is somewhat better to have a lower angle. The best would even be 0° (horizontal); then the estimated loss percentage is about 15%. But horizontal panels easily get dirty as the rain has less effect on them.

Do you also have such pictures for your areas?

 

During these peak intervals the determining factor is the angle between the sunlight and the panel, in particular the cosine of the angle between the sun rays and the line perpendicular on the panel.

I did some calculations for this cosine (right column) for your case of facing East and a sun at 60° from South for different angles of the panels (left column):

00 ---- 0.89
10 ---- 0.88
20 ---- 0.84
30 ---- 0.77
40 ---- 0.69
50 ---- 0.57
60 ---- 0.45

If I did not make any error, this shows that, for example for a panel angle of 20°, you will not be able to have more than 84% of your maximal panel power (Wattpeak). As you indicate 82%, could it be the case that the angle of your panels is around 20° or 25°?

So, therefore in cases like yours indeed you can use inverters with about 15% less Watt than your maximal panel power (Wattpeak) without having any clipping over longer time intervals.

 

The best resource I've found in the US is the PVWatts Calculator from the NREL.
You can plug in just some basic data to get a decent estimate; in my area with my roof pitch (18.5°), an east-facing array produces 14% less than a south-facing array. On my particular house, the difference is less due to more shade on the south-facing side.

One of the "advanced parameters" you can set is the DC to AC size ratio - 1.0 means that your inverters have exactly the same capacity as your panels. 1.35 is about the highest I've seen recommended, which means that for a 250W inverter, you could use a 337.5W panel. With a ratio of 1.35, they say a south-facing 2.08 kW array would produce 3,237 kWh - while a 1.0 ratio would produce 3,245 kWh. So by their estimate, you'd lose 0.2% of your total production over the course of the year by not having equally-sized inverters and panels. On an east-facing array, you could probably oversize the panels even further.

My roof is a 4/12 pitch, or 18.5 degrees.

With a detailed estimate, measuring from February to September this year, my actually production was only 4% lower than the estimate from the PVWatts Calculator - close enough that small differences in my shading estimate or in the weather would more than account for the difference.

 

For 18.5° the maximal achievable power for your panels with sun from South at 60° would be estimated by the calculation I made above at 85% of the maximal power (Wattpeak) of the panel. This is indeed not far from the 82% you observed for some time interval.

This PVWatts Calculator is an interesting tool. It makes immediately clear how much better the CA circumstances are compared to the Netherlands: in an example I came on about 50% more production than what could be expected for the same system in the Netherlands. Our location is more to the North and the outcomes are more similar to outcomes of places like Portland and Seattle.

It is indeed surprising how small the difference is estimated when you try different ratios between panels and inverters. For the east-facing option there is practically no difference (only 3 kWh) between the production for ratio 1 and for ratio 1.35, which is consistent with the calculations I made. For south-facing (roof mounted) options I see differences of 0.6% in this tool. I would have expected more differences than this 0.6% or even the 0.2% you saw in another example.

 

I still have some doubts about this tool. It turns out that it can also be used for locations in Europe, for example for Amsterdam, which is close to my location. The estimation of the annual production is quite realistic: 6900 kWh, which indeed is very close to my real annual production. However, changing the ratio between panels and inverters makes that production is even higher when inverters with lower power are used. I was willing to accept that the difference is not that much. But accepting that inverters with lower power than the panels make higher production, that is asked too much from my willingness; it is too implausible, as during the summer months I frequently see periods in which the production is close to the maximal power of the panels.

 

Yes, California is very good for solar. Not only because we're closer to the equator, but also because our weather is very favorable. There's hardly a cloudy day between April and October. And thanks to the drought, there's hardly a cloudy day during the rainy season, either. (Since moving here, I've become pretty accustomed to the idea that it's fairly rare for there to be clouds but not rain.)

What data did you enter for it that gave implausible results? It's certainly only useful for getting a rough estimate, but it seems to be pretty good. I would agree too that if all else is kept identical, it doesn't seem like reducing the size of the inverter could ever result in higher power output.

 

I took the data of my PV-system:

- location Amsterdam
- 7.765 kW
- roof mount
- tilt 20

For ratio 1.0 it gives 6933 kWh per year and for ratio 1.25 it gives 6962 kWh per year.

 

A similar pattern occurs for tilt 30 or 40.

 

The reason is you are shifting to the right when you look at the power curve of the inverter. Also look carefully at the inverter specs. The Solar Edge inverters, as an example, handle more solar panel wattage than rated.


iPad ?

 

Ah, yes, jzchen is correct; looking at their "technical reference" in the help area, it says they:

• Calculate the system's AC output from the calculated DC output and the user input System Losses, and using the nominal inverter efficiency user input (96% by default) with a part-load inverter efficiency adjustment derived from empirical measurements of inverter performance.

Since the inverter is more efficient at nearly-full power, when the ratio is higher, the system spends more time closer to that higher efficiency point.

I reproduced your data and downloaded the "hourly" version of their data; from what I can tell, with a ratio of 1.25, they say it will never clip (a ratio of 1.25 means the inverters could handle 6.212 kW, and the peak hourly production they estimate is 5.985 kWh). Also, the DC Array Output it calculates for both cases is identical - the difference only appears in the last step, when it converts the DC output to AC System Output. In that step, it looks like they pretty much universally think the more-heavily-loaded inverters will generate 9W more output from the same input.

 

OK, good to know that you get a small bonus when you are close to the maximal power of the inverter.

I also checked the hour data now. Their estimation of the peak power during the whole year is indeed below 6 kW. This is not realistic if I compare it to the data of my own system. At many days from beginning of May to September I get peak powers up to 6.5 kW. For just one example, this is the power curve for June 6 this year (vertical axis shows Watt):



These are real data for my whole system, which is rather heterogeneous due to the different possibilities I had: I have 29 panels at different tilts varying from 30° (10 panels) to 20° (11 panels) , 10° (4 panels) and even 1° (4 panels). This is the reason that the graph does not come close to the 7.765 kW which is the overall maximal peak power. For many parts of my system I also have more shade than expected as average.

If I consider only the 10 panels with tilt 30° I see them reach the maximal power of what they can perform, and that is 30% higher than what is calculated by the PVWatts tool.

 

@Jan Treur YMMV.

The PVWatts Calculator is not more than an estimating tool for people considering installing PV system on their roof. As such - I find it very extensive and elaborated and extremely useful. Do you have a better one? please post a link.

 

Indeed it just gives an estimate, just like some other tools that are used by installers, but which are not better as far as I see. I do not know whether a perfect tool for this exists. I do know from own experience in the research of my group that it is quite challenging to take into account all these factors in an adequate manner.

 

USB,
This is so inspiring !
I must start the getting serious about my home solar installation.

Have you considered getting a proper BEV or PHEV.
I know we all love the PIP here, but really, 'Baby, it's cold outside'.

The Chevy Spark EV is available now, one state over from you.
I call mine a mini Tesla. It's a fast, fun commuter.
And I love being able to preheat / cool it from a phone app this time of year!

 

Wow, that is huge variability for the timeframe!

Here's ours:



I'm afraid early December one panel stopped producing electricity, and even though I contacted the installer quite quickly, they have yet to come out to fix the problem!!!


EDIT- Panel failure date wrong...