Help with an undersized inverter used in my system!

[Ian S]

Thank you for pointing out the image attachment feature. This is a day from March clearly showing the clipping at 5,110w AC generation for several hours. I highlighted the missing energy and will use basic calculus to determine the area under each curve in an attempt to identify/quantify the amount of power that was not realized.

[ATTACH=CONFIG]1966[/ATTACH]

OK, when I see that, it's looking maybe more like 10% of the kWh clipped. Roughly, you've got two triangles: the overall unclipped estimate and the amount representing what got clipped. I think you have reason to be ticked off. I'm not an installer but it would seem to me that there was no excuse not to give you a properly sized inverter. Maybe naptown can chime in with some rationale for doing so but if not, then their design is faulty. It would bug me every time I checked my daily production and saw the clipping! File a complaint with the BBB and the appropriate state licensing agency.

[sun2energy]

Some Thing 2 Consider

Inverters have a performance curve. Often what appears to be a undersized inverter even if there is clipping will produce more energy on a year around basis than the next size larger inverter which may not operate at pike performance as often as the smaller inverter.

To confirm your assumption that a larger inverter would generate more kWh/yr you will need to graft a years production and compare how the two different converter would have performed. Often I find the smaller inverter makes more kWh because it operates closer to pike performance more of the time.

Trust the manufacture and installer they have the experience with their products. In your case the installer did not save any money installing the small inverter because like he said a 5100 and 6000 are the same price wholesale.

[Naptown]

Well before we go trotting off the MHIC and the BBB
What is the estimated peak output you estimated during the times the inverter clipped.
Although I agree that the PV watts derate is too low perhaps we are maybe over estimating production.
Example
PV Watts standard uses a power tolerance at .95 meaning worst case scenario regarding the + - watts tolerance of the modules.
This alone if the Sunpower modules are -0 +5% could account for a lot of the difference. If the modules are all putting out +5% this could be a cause. Very evenly matched modules reducing the module mismatch category and short runs of wire reducing the DC wiring losses will account for more harvest.
Essentially what did they tell you the system will produce annually?
Is the system producing more or less (not counting the inverter clipping)

[billvon]

To give you an idea of how much clipping occured by the inverter! For the past 297 days we had 117 days of clipping (i.e. days where the inverter maxed out in its DC to AC conversion @ 5,110w AC). That's 40%!

Right - but on those 40% of days, you are only clipping 50% of the time, and then you are losing, at most, 15% of your max power. That's very approximately (.4 x .5 x .15) = 3% of the potential power harvested. Is that worth fussing over? Perhaps; that's up to you. But it doesn't seem like it's fraud or anything.

[Ian S]

Inverters have a performance curve. Often what appears to be a undersized inverter even if there is clipping will produce more energy on a year around basis than the next size larger inverter which may not operate at pike performance as often as the smaller inverter.

Well, in that case, you've got me worried because my Sunpower installer has my 6.9 kW system hooked up to a 7 kW inverter which I thought made sense. Now you're telling me that I should have the 6 kW inverter?

What I'd like to know from our gurus here is what is best practice when matching inverter to system size. What parameters go into that decision?

[Naptown]

Inverter manufacturers generally produce string sizing charts. The general rule is 20% +/- the STC wattage of the array for Peak year round performance.

[PseudoCode]

Well before we go trotting off the MHIC and the BBB
What is the estimated peak output you estimated during the times the inverter clipped.
Although I agree that the PV watts derate is too low perhaps we are maybe over estimating production.
Example
PV Watts standard uses a power tolerance at .95 meaning worst case scenario regarding the + - watts tolerance of the modules.
This alone if the Sunpower modules are -0 +5% could account for a lot of the difference. If the modules are all putting out +5% this could be a cause. Very evenly matched modules reducing the module mismatch category and short runs of wire reducing the DC wiring losses will account for more harvest.
Essentially what did they tell you the system will produce annually?
Is the system producing more or less (not counting the inverter clipping)

Hi Rich,
We do have a very short run (1 story home) from roof to inverter, clear view to the south. I could somewhat understand that say a “large” inverter that would produce 10Kw AC matched with a 6Kw DC system may have some inefficiencies. Per your suggestion I have provided the PV Watts calculations and actuals for Jan -> April below.

I never felt that the installation company attempted fraud; just an architectural mistake or simple goof of inverter selection (I’m trying to think positive here!). I may be repeating myself but Sunpower did admit that with this 5000m inverter, we are not fully realizing all the DC energy being produced by a factor of less than 1%. I’ve asked them how they concluded this percentage using PV Watts data but have no reply. I also inquired as to how they match/size an inverter to a set of panels. Still waiting for a reply which I’ll post back when I hear something. It’s unfortunate that I know not a whole lot, but just enough about PV systems to get into trouble…

To some it may be appear to be splitting hairs over 1%, 10% or 20% whatever the actual unrealized loss may be , but to the owner who coughed up a good chunk of change for the system knowing that its full potential may not be reached is tough to swallow. Let’s just say I’m the type of owner who does clean the panels from pollen, snow, bird droppings, etc. to ensure peak operation. It’s also compounded not only with long term loss of energy production to lower our bills but missed opportunity in SRECs too. Thank you again to everybody for their helpful comments and feedback!

"PVWATTS v.2: AC Energy and Cost Savings"

"Station Identification"
"Cell ID:","0263375"
"State:","Maryland"
"Lat (deg N):", 39.18
"Long (deg W):", -76.67
"PV System Specifications"
"DC Rating:"," 6.2 kW"
"DC to AC Derate Factor:"," 0.850"
"AC Rating:"," 5.3 kW"
"Array Type: Fixed Tilt"
"Array Tilt:"," 39.1"
"Array Azimuth:","195.0"

"Results"
"Month", "Solar Radiation (kWh/m^2/day)", "AC Energy (kWh)", "Energy Value ($)"
1, 3.47, 580, 104.42 <- Kwh for the month was 560
2, 3.95, 589, 106.04 <- Kwh for the month was 657
3, 5.23, 837, 150.69 <- Kwh for the month was 803
4, 5.34, 805, 144.92 <- Kwh for the month was 851
5, 5.61, 844, 151.95
6, 5.85, 822, 147.98
7, 5.68, 820, 147.62
8, 5.31, 770, 138.62
9, 5.26, 748, 134.66
10, 4.78, 737, 132.68
11, 3.53, 547, 98.48
12, 3.17, 513, 92.36
"Year", 4.77, 8611, 1550.24

[PseudoCode]

Well, in that case, you've got me worried because my Sunpower installer has my 6.9 kW system hooked up to a 7 kW inverter which I thought made sense. Now you're telling me that I should have the 6 kW inverter?

What I'd like to know from our gurus here is what is best practice when matching inverter to system size. What parameters go into that decision?

Ian I

[Ian S]

Inverter manufacturers generally produce string sizing charts. The general rule is 20% +/- the STC wattage of the array for Peak year round performance.

Well, then Pseudocode's inverter is on the hairy edge of what's acceptable. IMHO, one of two things happened: a mistake was made and the wrong inverter was used or the 5 kW inverter was available and the 6 kW was not. Perhaps they figured no big deal, only a few percent loss due to clipping. And without the monitoring capability, who would have noticed? I guess the silver lining in this would be that as the panels age, the percentage dropoff in production each year will be less!

[sun2energy]

Sunpower makes a great panel at a premium price. I am not sure what panels you have but Sunpower panel do over produce but remember these panels will degrade over time. The area on top the curve you are missing really looks nothing like the attached jpeg chart. The curve at pike sun flattens as the heat goes up mid day. By year two or three the pike production will be well under the inverter cutoff. If you had a 6000w inverter you would then be losing production because you would outside the sweet spot on the inverter curve.
All installations have variables and compromises. I am no fan of Sunpower but your system looks to be optimized for long term production.

This may not be what you want to hear but remember the system is designed to produce the maxium kWh over 20 years not just the first year or two. Sunpower has lots of experience in long term production, as the panels production decrease with time I think you will find your inverter is a long term winner.

[Ian S]

Sunpower makes a great panel at a premium price. I am not sure what panels you have but Sunpower panel do over produce but remember these panels will degrade over time. The area on top the curve you are missing really looks nothing like the attached jpeg chart. The curve at pike sun flattens as the heat goes up mid day. By year two or three the pike production will be well under the inverter cutoff. If you had a 6000w inverter you would then be losing production because you would outside the sweet spot on the inverter curve.
All installations have variables and compromises. I am no fan of Sunpower but your system looks to be optimized for long term production.

This may not be what you want to hear but remember the system is designed to produce the maxium kWh over 20 years not just the first year or two. Sunpower has lots of experience in long term production, as the panels production decrease with time I think you will find your inverter is a long term winner.

So then you are confirming that MY system (6.9 kW Sunpower panel array mated to 7 kW Sunpower inverter) is poorly matched? Because, I don't see how BOTH can be optimum.

[billvon]

To some it may be appear to be splitting hairs over 1%, 10% or 20% whatever the actual unrealized loss may be , but to the owner who coughed up a good chunk of change for the system knowing that its full potential may not be reached is tough to swallow.

Keep in mind that if he goes to a larger inverter his total production may be LESS because tare losses go up with larger inverters. Ideally you want an inverter that limits just a bit under ideal conditions, because tare losses will be reduced - and you pay for tare losses every second that inverter is on, not just on ideal days.

[sun2energy]

PseudoCode,

If you have a Facebook page engage Sunpower with a complaint and then the rest of the world will see how they respond. I find that companies that use social media like Facebook want to have happy customers and will engage you to resolve your concerns in a more public way.

Making complaints with forums there are not enough eyes to force a company like Sunpower to respond and bloggers like myself give companies like Sunpower cover. Complaining the Yelp or Google etc. does not give acompanies a the chance to respond in a meaningful way. Facebook and twitter allow customers to demand a response and the others that are interested in doing business with that companies see how they react to problems.

[sun2energy]

Ian S,

Once again each system has its own set of operating conditions. As a general rule I would say yes it is over sized but there many be good reasons for over sizing it too.

An example is when we have a inverter that can not be located in a cool area and has to sit in the sun all day I might over size the inverter to lower the heat stress on the inverter. Inverter hate heat and will fail prematurely if over heated thus increase the maintenance cost for the system.

Every system is different, we have general rules but they are only a starting point. The design of a system has some assumptions that can only be verified after installation and even then there are trade offs because solar PV systems degrade over time.

[kimberlymitchell87]

No, not a problem. The inverter will only output as much power as is "safe" (design and temperature limits).

Also, it is not unusual to install an inverter that is ~86% of the solar panel rating. Unless it gets really cold and you have snow fields reflecting light onto the array--Your system will rarely (few minutes to a few tens of hours a year type thing generally) more than 86% of the solar panels rated (marketing) output.

Yours is:

6,000 / 6,110 = 0.98 of rated array power

Not a problem/issue/or real loss of revenue at all (going to the next size inverter may cost you more than the little bit of lost revenue will ever gain back).