I bought a solar PV system a couple of months ago. It has 11 410-watt panels for a total of 4.5 kW.

The first month it was cloudy and never generated over 3.2 kW of power. Now that it's been bright and sunny, I am still only getting 3.2 kW.

After doing some checking, it looks like the micro inverters are only rated for 300 watts each (0.2998 kW). I got the 400-watt panels to get more power but it looks like I would be generating the same power with the 330-watt panels they had.

Does this sound correct? Is a 25% overload on the inverter normal?

I think the inverters should be closer to 400 watts and the system should be generating close to the 4.51 kW.

There doesn't seem to be a reason for the inverters to be undersized so much. I live in California where we have plenty of sunny days. My system has been peaking out at 3.2 kW at 9:30 in the morning.

My house is fairly new with a 220A main. My system is from a local franchise (?) of a large national company. The system does not have batteries.

  • 7
    This seems like a question for the business you purchased the system from. Did they suggest the increased panel size or did you?
    – Ecnerwal
    Commented May 31, 2023 at 17:27
  • do you have back up batteries
    – Traveler
    Commented May 31, 2023 at 18:32
  • 4
    At least on large installs, panels are cheap enough that they use 10% or more panels than are strictly needed so the install delivers rated power on a sunny day even as things age or are knocked out of alignment.
    – Jon Custer
    Commented May 31, 2023 at 19:26
  • upgrade one of the inverters and see if it helps. If so, do the math to see how worth it replacing them all would be. From what i've seen, unless you have dismal conditions (shadows, bad angles, etc), microinverters don't help much; I think companies like to push them so they can sell you more high-markup equipment (cells themselves are lower margin).
    – dandavis
    Commented Jun 1, 2023 at 4:50
  • I suggest you try to find a US version of this calculator, which works out how much power you will generate, broken down by month and time of day, given a location in New Zealand, and the orientation of your panel. Of course this will only give you a theoretical maximum, as others have noted. Commented Jun 1, 2023 at 21:18

3 Answers 3


Solar panel rating is based on absolutely ideal conditions.

Absolutely ideal. In order to get those conditions you need to have a perfectly clear day, freshly cleaned panels, your panels mounted on a heliostat, or 1-axis tracker with elevation tuned for today, so the panels are dead square on to the sun's rays. It helps to be near the equator so you get less angled travel through the air.

If your panels are oblique to the sun's rays, which will be true all the time on all fixed installations except possibly a 5-minute period 3 days a year... then you will get less than rated power. And you can do the trigonometry to figure out to what degree the panel is degraded by not being square-on.

So you're going "well hold on! Those ideal conditions will NEVER happen for me!" Yes, and the solar panel installer should have set that expectation with you i.e. educated you as to this reality. Unfortunately when you're in a 30 minute consultation, thousands of words get said, most highly technical, and laymen just can't remember all of it.

Given that max is never gonna happen...

how do we size the inverter(s)? Is there any disadvantage to just going ahead and using a 410 watt inverter so that your best day, at least, really is your best day?

Yes, there is a disadvantage, actually. Electrical panels are limited on how much solar they can support. For instance a 100A service with 100A-bussed panel can support 3840 watts of solar inverter. Using the max possible inverter capacity limits your array size - if you had used 410-425W inverters, you'd be limited to nine of them. By dropping the inverter to 300W, you can have twelve. How much to downsize is really a system-design decision.

Panels are cheap

I know you just got done paying a 5-digit sum for all this, but the cost of it wasn't in the solar panels. Solar panels are 50 cents a watt. The lion's share of the cost was in the

  • entitlements (permission to build)
  • system engineering
  • roof mounting hardware
  • Labor (perilous work)
  • wiring
  • insurance
  • financing
  • customer acquisition / sales funnel

The low cost of the panels themselves, lends itself to solving difficult problems by throwing cheap panel at them.

So what you actually bought was a 3300 watt system, in which they enhanced your system's performance by oversizing the solar panels, because doing so was very cheap to do.

  • 1
    I live in a sunny area in California in a recently built home with 220A mains. The panels were priced at about $1400 each - I had the salesman run numbers for different panel counts. The difference between the 11 330-watt and 11 410 watt was about $4000 - though the make was different (Pan v SP). I usually hit peak power before 10 AM. It seems like I would be getting more power if the inverters were closer to the panel rating. Commented May 31, 2023 at 22:55
  • 2
    @Hannover $1400 is outrageous, they're more like $200-250 really. Google 400W solar panel. That must have meant the "All in, total project cost" covering inverters and every expense I mentioned. Peak power being 10-2 makes sense, it'll be at absolute peak at high noon depending on roof angle. But hey, try it - slap the bigger inverters on there and see what you get. Commented Jun 1, 2023 at 6:05
  • 1
    Why does a 100amp panel limit (100 amp * 240 volt=24kW) have a solar limit of 3.8kW? It seems like such a small fraction of its total. What is that limit for something like a 200 amp panel (double)? I'd assume people buying 4 kW solar systems are more predisposed to having bigger panels. Commented Jun 1, 2023 at 15:14
  • @Dean To avoid overloading the service panels. The rules are somewhat arcane. Yes, 200A breaker on 200A bus panel allows 40A solar. However the vast majority of 200A-breaker panels are actually 225A bus, which (calculation here) allows 70A solar. There are also other ways around it, such as a "solar ready" meter-pan, or tricky breaker arrangement. With those, the service is the limit. Commented Jun 1, 2023 at 19:10
  • 1
    > It helps to be near the equator. Ideally, on a ship cruising the tropics so the sun is always overhead at noon each day.
    – Rich
    Commented Jun 1, 2023 at 21:29

It is possible, even likely, that you may get more times of 3.2KW output (and thus, greater total output for a month) on suboptimal days with 410W panels than you would with 330W panels, but it's difficult to know to what extent that would apply without having the two systems set up side by side and comparing them. As panel output gradually decreases with age, full power output will also be maintained longer.

There may also be other factors in play such as the capacity of your electrical system for solar power, on both technical and regulatory levels that might preclude higher output. Technical being "it would not be safe due to the capacity and loading of various components" and regulatory being "you are not allowed to have more than X output as a homeowner."

  • 2
    All of which your installer should be able to answer. If you don't like their answers, it may be worth getting a second opinion. Whether you can make them do anything about it without charging you is going to be a matter of what they committed to and what you signed off on. But most solar installations do run below 'ideal"/"peak" output, not being perfectly aligned to the sun.
    – keshlam
    Commented May 31, 2023 at 18:28
  • I think saying it is "possible, even likely" is a pretty drastic understatement. Commented Jun 1, 2023 at 15:58

NREL has a solar calculator on their website here

Using an arbitrary point in California I ran some simulations (using the default value for everything but the DC size and DC/AC ratio)

DC rating KW DC/AC ratio yearly AC kWh yearly DC kWh
4.5 1.366 7610 8015
3.63 1.1 6168 6466
4.5 1.1 7647 8015

The difference in effective annual AC production is 1442 kWhs for the panel upgrade. Had you also upgraded the inverters, you would have only realized an additional 37 kWhs.

That's just in the first year. Over time the panels will degrade such that your effective DC/AC ratio will get lower. Your current setup has cushion in the DC/AC ratio so effective degradation will be less pronounced than with a smaller DC/AC ratio (bigger inverters).

Additionally, inverter efficiency is maximized at high output. If you download the hourly csv file you can compare the dc and ac production. You'll see that the ratio of produced dc/ac is lowest when output is the highest:

enter image description here

The output_ratio is the DC power over the AC power for each hour of the aforementioned simulations color coded by the row of the above table. You can see how the red and blue dots follow the same curve because they have the same inverter size. The bigger inverter's curve is pushed out to the right so you can't get as much AC power when DC power is low.

Here's a histogram

enter image description here

You can see that with a high DC/AC ratio that row 1 (your actual setup) has a huge count at its inverter capacity. This is what you're seeing day to day. Intuition is telling you that if you had big inverters that you'd simply push that blue bar to the right and it'd be equal height. However, that's not the case. If you increase the inverter capacity, what this shows is that you don't move that tall bar to the right. You just kind of smoosh it down into marginally higher production that peters off pretty quickly to the right of where you are. This just reinforces the idea that having bigger inverters would only have gained about 37 kWhs/year.

As an aside, in wholesale solar farm sizing, having a big DC/AC ratio on the order of 1.5, sometimes called the ILR (inverter load ratio) is normal. That's despite the fact that in a commercial solar farm the panels are on trackers so that they don't even have the same angular losses that you experience with rooftop solar.

  • "won't notice panel degradation for several years" is not entirely accurate. It will be hidden only when the inverters are saturated. The other 80% of the day, degradation will directly impact power output.
    – Ben Voigt
    Commented Jun 1, 2023 at 15:25
  • @BenVoigt "It will be hidden only when the inverters are saturated." This is definitely true. In my head I deliberately used the phrasing "won't notice" rather than to say "you won't have any realized degradation" with the context that the salient observable is max power. If I create an ACproxy which is just DC*.9579 and then capped at the rated DC/ILR and sum that up, then create a degraded year2 which just takes the existing DC*0.996 and then take the same ACproxy logic as before; the difference is 25.5Kwh for row 1 and 30.7kWh for row 3. Probably doesn't notice the degradation either way. Commented Jun 1, 2023 at 15:57
  • But your analysis that the degradation is too small to notice doesn't support the claim that OP won't notice panel degradation because "you have cushion in the AC/DC ratio".
    – Ben Voigt
    Commented Jun 1, 2023 at 16:01
  • 2
    The analysis is that getting bigger inverters bears little fruit. The tangent about degradation is just meant to convey that when degradation is factored in, the situation gets worse with bigger inverters. Commented Jun 1, 2023 at 16:13
  • @BenVoigt I edited my claims about degradation to be less sensational. Commented Jun 1, 2023 at 16:18

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.