professionally installed system energy production curve flat in mid-afternoon

[gpsimms]

Hey y'all,

First of all, sorry to say that I am super lame and I didn't install my own system. I am just looking for a forum to post my question so I can see what the possible explanations are before I complain to the installer about it. If this is too off-topic, or I should ask somewhere else, then mods/everyone accept my apology and please direct me to somewhere you think this question would be appropriately posed/answered.

Anyway: I've had the system for a couple of months, and on days where there is sufficient sun, the system production rate appears to "top-out" at about 3.2 kWh. Here is a graph from the app that came with our system install:



The system is 4.48 kW, I live in Virginia. The total production for this day in September was about 23.0 kWh. Is the flat part of the graph some sort of weird artifact of the app software? Or am I really having my production limited for some reason? Does that total output sound about right (maybe that's not possible to tell, because there are too many individual factors like roof orientation, trees, etc)?

I've seen this effect on basically every sunny day until now, because now the sun exposure has already decreased so I'm not hitting this "saturation" point. Anyway, have any of you seen anything like this? Is this normal? Is it possible my production is limited by an undersized inverter? I have no idea, just looking for any possible explanations before I contact the installer.

Thanks!

 

[snoobler]

4.48kW for what?

Panels or inverter?

Common to put more panels on an inverter than it can handle as panels often can't or won't deliver their peak output, so the inverters are "over-paneled".

 

[gnubie]

It could be that is all the inverter is rated for, but if this is a grid tied system a contributor may be the inverter limiting due to voltage rise on your grid supply. If you post the model inverter that would help.

 

[gpsimms]

Thanks for the quick replies!

4.48 kW is the rating of the system. I have fourteen 320W panels, so the panels make a total of 4.48 kW. The inverter is rated to 4.48 kW, according to my contract, but I don't know without looking at it at home what type it is. I will reply again when I do.

Edit: and yes, this system is tied to the grid.

 

[svetz]

Looks like clipping.

 

[gpsimms]

Looks like clipping.

Thank you. I agree that's what it looks like. So, my question is now: if I am getting clipping in sunny September days does it mean that the inverter is *too* undersized? Or is what I'm seeing sort of typical, and I'm not missing out on that much energy? Also, FYI, I forgot to visually check my inverter size when I went home last night. There may have been something else to stress and worry about, but I can't remember what it could have been. : )

 

[svetz]

...my question is now: if I am getting clipping in sunny September days does it mean that the inverter is *too* undersized? ...

Some people will tell you that even throwing 1 watt away is sacrilegious so they oversize the inverter. Personally I'm okay with some clipping. That is it's a cost trade-off, and to a large extent just a personal preference you work into the system economics.

You posted you have a 4.48 kW array, but not your inverter size.... so can't say if you're around the guideline values.

Clipping is just the panels producing more energy than the inverter can convert. So, if the day that graph was recorded was overcast and rainy then for sure you've a problem. But if it was sunny, cool, the panels perpendicular to the solar angle, and there was pleasant breeze... it might have been just right.

Most people tilt the panels for most power year-round, so about perpendicular during spring/fall equinox and that's when they'd see the most clipping (that is, some clipping around solar noon for a few weeks every year, the rest of the year no clipping).

For example, I have 340W panels, each of which has a 290W Enphase microinverter giving me 117% over-paneled. Mine are flat on a flat roof, so I expected to see some big clipping come summer was the sun nearly ninety degrees with long days. Didn't happen, to many clouds and storms and if not the panels were so hot they produced less. Turned out my peak was 4/22 and my microinverters over-produced (post).

So, hopefully you see that the guidelines are just guidelines....the actual ratio depends on the tilt and weather at your location. A few hours of clipping a few days a year is probably right-sized. Don't forget too that panels degrade a few percent the first year and then something like a 1/2% each additional year, so as your power goes down the clipping goes down.

The good news is, you can tell what's right-sized for your area. Download SAM, plug in all the data, and take a look at the results. SAM is one complicated beast and uses your historical weather data, so watch the video tutorials and ask questions. From there pull up the clipping report and see what it predicts and if you want to live with that or not.

 

[gpsimms]

Thanks for the explanation. Yeah, I clicked over at SAM and it looked like it was too much work to figure out right now, but I will check into it later for sure. Just realized I could probably look up my inverter situation on the enphase app, and I can.

Looks like I have 14 IQ7 microinverters. From the datasheet I found, it looks like they are usually paired up with 235 - 350 W panels, and my panels are 320W.

It's good to hear that clipping would be the worst in fall/spring. I was imagining it would last all summer. For the record, it looks like I had approx...

~3 hours of clipping on 5 of 20 days in September (the system was turned on Sept 10).

~2 hours of clipping on 7 of 31 days in October

~2 hours of clipping on 1/3 days so far in Nov

Again, I'll click around the NREL tool a bit to figure out if that's a significant loss. Thanks for the help everyone!

 

[svetz]

Sounds more than average, but you may have just had better weather than average.

...It's good to hear that clipping would be the worst in fall/spring....

Depends on panel tilt as to when it's theoretically possible.

 

[Hedges]

Did your paperwork from vendor list what performance is supposed to be? (Our locality now requires many disclosures.)
You should see if peak production higher than 3.2kW was claimed (which would suggest the wrong inverter model was installed), or if 3.2 kW was indicated (in which case they over-paneled, reasonable to boost your output closer to 3.2kW on hot days, under overcast conditions, other seasons.)

IQ7 is 240 W continuous:

https://enphase.com/sites/default/files/downloads/support/IQ7-IQ7plus-DS-EN-US.pdf

240 x 14 = 3360.
Maybe 3200 is a slight mis-reading by you, or few percent measurement error of equipment?

It shouldn't be called a 4.48 kW system. Better called "overpaneled with 4.48 kW of PV panels to make up for days with less sunlight"
It should be called a 3.36 kW system.

See what wattage panels you were given.
Make sure with 14 units delivered you weren't short-changed.

 

[gpsimms]

You are right. The bar graph has bars going about to approx. ~0.8 kW with 15 min bar widths, but it looks like the total over the whole hour was 3.4 kW, exactly as you say.

Agree that it seems like a misnomer to have been billed as a 4.48kW system. The guarantee that Sigora makes is with regard to annual production, so they may make me wait until the final yearly number comes in before making any changes to the system (assuming it comes in low).

The total annual output is supposed to be 5344.7 kWh, and I'm currently at 822.5. This is in 56/365 days.

Assuming the 56 day block I've been operating in is approximately average (seems like it might be a reasonable assumption for mid Sept - early Nov), then I am actually almost exactly on pace to hit their projected number, though I do think the weather has been unusually nice this fall. Guess I'll just cheer for an unusually cloudy year so I get a system upgrade on them : ).

 

[Hedges]

The total annual output is supposed to be 5344.7 kWh, and I'm currently at 822.5. This is in 56/365 days.

5344.7/365 = 14.64 kWh/day
14.64/5.5 effective average hours (for my San Jose location) = 2.66 kW system.

Your 3.2 or 3.4 kW is well sized to deliver that.
Because it is clipping at that, over-paneled, it will do even better.
So it appears they are delivering more than promised.

(look up your location average hours for fixed array orientation.)

 

[gpsimms]

5344.7/365 = 14.64 kWh/day
14.64/5.5 effective average hours (for my San Jose location) = 2.66 kW system.

Your 3.2 or 3.4 kW is well sized to deliver that.
Because it is clipping at that, over-paneled, it will do even better.
So it appears they are delivering more than promised.

(look up your location average hours for fixed array orientation.)

Yup. I came up with about 4.9 effective average hours for the year, which comes to 2.99 kW system. I also found that the sunlight average hours for Sept-Oct basically comes up to 4.9 as well. So, in other words, this ~50 day stretch should be almost exactly representative of what I should expect to see all year. Given that, I am 15% of the way through the year and have generated almost exactly 15% of their guarantee. Looks like I am exactly on pace.

Thanks for all the input. Hopefully this thread will be useful for others who are confused about clipping. It was for me! Thanks again to everyone for the input.

 

[snoobler]

Interesting for me to learn that the microinverters have a peak output even if they can be used on a higher wattage panel.

 

[Hedges]

Interesting for me to learn that the microinverters have a peak output even if they can be used on a higher wattage panel.

Small variety of Enphase inverters, large variety of PV panels.

Enphase Energy - Wikipedia

en.wikipedia.org

With only a few wattage inverters to choose from, it will always be under-paneled or over-paneled.
Me with string inverters, or you with DC charge controllers, can put many panels on one unit, fitting close to its wattage capacity or deliberately over-paneling for times of less sun.

We can also parallel panels of different orientations, to get more hours usage out of the electronics.
One might be able to do that with smaller panels on an EnPhase micro-inverter, but the IQ7X delivers 315 VA and new panels these days are up to 500W.

These days it looks like they market their inverters based on how much gas they can burn rather than how fast they can go.
Had to click through and read data sheet to find power output, not just maximum recommended PV panel.

But they easily deal with a variety of orientations and shading, and for OP the vendor installed something that should deliver what was promised or more.

I am curious how well these interact with battery inverters, since people often want to add that later.

 

[svetz]

IQ7 is 240 W continuous:

Depends on the IQ7 sub-model, for example the IQ7PLUS-72-2-US is 290W and peaks to 295.