System Efficiency

balee123

Hello all,

I think I remember reading several threads describing overall efficiency from panels to usable energy being ~50%, is this correct?

More specifically, I have a 3.2 kW grid tied system and the most output I've seen from the inverter (in the late spring) has been around 2.1 kW, or about 65%. Is that expected/typical or does my system have some efficiency problems?

Another reason this seems low is I have an friend who has a 5 kW GT system, that uses the Tigo Energy "microinverters" and he claims that he sees an output at the inverters of 95% of the PTC rated system. Is this possible? If his system is that efficient, is it do the Tigo technology?

Cariboocoot

Re: System Efficiency

The over-all efficiency rating of 52% is for off-grid systems (nameplate rating of array to AC Watt hours available).

A GT system will be closer to 77% efficient (panel nameplate + MPPT function, similar to MPPT controller off-grid, no battery losses to worry about).

65% does seem rather low. If the array is 3.2 kW it should 'peak' around 2.4 kW output. However a lot of things can change that, most notably panel temperature, array aiming, and weather conditions. How's the kW hour production? If the array is a bit East or West it may never hit the maximum output Watts but may still produce the expected amount of Watt hours, which is what really matters for a GT system.

Most GT inverters themselves are 90+% efficient. All you have to do is back them up with enough panel Watts so that even a derated array can push them to full production.
Example: a 5kW inverter with 5kW array will probably peak at around 3850 Watts. Put a 6500 Watt array on the same inverter and you may see that full 5kW output.

balee123

Re: System Efficiency

Thanks for info.

For what its worth this is an older system. 32 Astropower AP-100 panels, two paralleled inverters [Xantrex STRX2500 (20 panels) and STRX1500 (12 panels)]. The spec sheet say the inverters are only 91% efficient.

The kWh production is about 16 kWh per day during that peak time of the year. I'm in San Diego, CA. The arrays face 200 degrees, so 20 degrees west of south.

Are micro inverters or this Tigo technology any more efficient compared to the standard GT inverters?

Photowhit

Re: System Efficiency

I would think your running low, as 'Coot said, I would expect a daily average peak of near 75% not an overall one time peak of 65%.

Lots of factors, roof/mount slant, direction, heat buildup...

...are your panels 5-6" off the roof's surface?, Does your roof or mount face south?(assuming northern hemisphere), is your mount pitch with in 10 degrees of your latitude?

Might also look at, length and size of wire run, and the 'typical' rating of your panels, some have pretty poor typical rating compared to panel rating.

Cariboocoot

Re: System Efficiency

16 kW hours sounds good to me (maybe Bill will be kind enough to run the numbers on PV Watts). If you divide that by a 3.2 kW array you get 5 hours of "equivalent good sun".

Micro-inverters are only more efficient on a case-by-case basis. If you had both types of system side-by-side with equivalent arrays and inverter capacity you'd be hard-pressed to notice any difference. If you have a situation where there are unavoidable shadow problems and/or a need to face panels in several different directions the micro-inverters would work better.

Consider the alternatives: replacing the inverters with a newer one to gain a couple of % improvement? Replacing the whole system to pick up 10%? Doesn't sound sensible, does it?

balee123

Re: System Efficiency

Photowhit wrote: »

I would think your running low, as 'Coot said, I would expect a daily average peak of near 75% not an overall one time peak of 65%.

Lots of factors, roof/mount slant, direction, heat buildup...

...are your panels 5-6" off the roof's surface?, Panels are tilted to 20 degrees (latitude is 32 degrees), either way the data mentioned is best/peak data for the year.Does your roof or mount face south?(assuming northern hemisphere)panels face 20 degrees west of true south, is your mount pitch with in 10 degrees of your latitude? within 12 degrees

Might also look at, length and size of wire run, and the 'typical' rating of your panels, some have pretty poor typical rating compared to panel rating.

what do you mean by "typical" rating?

Cariboocoot

Re: System Efficiency

balee123 wrote: »

what do you mean by "typical" rating?

"STC" vs. "NOC"
And throw in "vs. the real world" for good measure.

A panel rated as 'X' Watts STC (Standard Test Conditions) could be '0.8X' NOC (Normal Operating Conditions) or some other number. In the real world it could be worse. Much of this is due to the change in Voltage caused by panel heating (the higher the temp, the lower the Voltage - with resulting power loss).

You've actually got 4kW of inverter (newer ones would be about 95% efficient, btw) so if you had room and could find some panels with matching specs you could "panel up" a bit. This probably isn't practical either.

BB.

Re: System Efficiency

Using PV Watts for San Diego California, 3.2 kW, 0.77 derating, 20 degree tilt, and pointing at 200 degrees (20 degrees west of south):

A "typical system" in a "typical weather" year would look like:

"Station Identification"
"City:","San_Diego"
"State:","California"
"Lat (deg N):", 32.73
"Long (deg W):", 117.17
"Elev (m): ", 9
"PV System Specifications"
"DC Rating:"," 3.2 kW"
"DC to AC Derate Factor:"," 0.770"
"AC Rating:"," 2.5 kW"
"Array Type: Fixed Tilt"
"Array Tilt:"," 20.0"
"Array Azimuth:","200.0"

"Energy Specifications"
"Cost of Electricity:","12.5 cents/kWh"

"Results"
"Month", "Solar Radiation (kWh/m^2/day)", "AC Energy (kWh)", "Energy Value ($)"
1, 4.23, 304, 38.00
2, 4.88, 315, 39.38
3, 5.73, 412, 51.50
4, 6.67, 457, 57.12
5, 6.44, 460, 57.50
6, 6.46, 441, 55.12
7, 6.76, 471, 58.88
8, 6.90, 477, 59.62
9, 5.96, 399, 49.88
10, 5.50, 390, 48.75
11, 4.56, 316, 39.50
12, 4.04, 285, 35.62
"Year", 5.68, 4728, 591.00

Or around 285 to 460 kWH per month... Or ~9.5 to 15.3 kWH per day...

For non-billing rated systems, logging can be off by +/-5% or even as much as +/-10%... In some of the cases, the average seems to be averaging +5% too high (who would complain if a system output 5% more than "book value").

If you keep your panels clean--You can get a bit more power. If you have lots of marine layer in your area, you would get less.

Arrays pointing 20 degrees west of south--May generate a bit less power (also depends if you have AM or PM marine layer), and if you have Time of Use billing (summer afternoons are more expensive than morning/off peak... For me ~$0.30 per kWH Summer Peak vs ~$0.09 per kWH Summer Off Peak). So, you may generate more power in the afternoon, and get "better payments" because of this--even-though, on average, the system generates less raw kWH per month because of the afternoon power bias).

-Bill

balee123

Re: System Efficiency

Thanks for the analysis Bill. It appears overall that I'm not too far off.

Another thought/ question.....the DC input voltage range for my inverters is 52 - 75 volts. Are the higher voltage inputs on newer GT inverters inherently more efficient?

BB.

Re: System Efficiency

Yes, no, maybe...

You have I2R heating losses... So, if you cut the current in 1/2, you reduce the heating/wasted energy to 1/4.

However, you also have capacitors in the construction of the switching transistors.. with E=1/2 * CV² -- So when you double the voltage, you get 4x the losses due to capacitance (charging/discharging--again, more wasted energy).

How designer trades off cost of copper/steel, price of FETs (higher voltage FETs cost more, have more resistance, more capacitance) etc...

The whole design of a switching power supply is chuck full of these trade-offs/safety issues/requirements from safety/government/regulatory/marketing/customers...

Electronics, in general, have become "more capable" because smaller geometries means smaller capacitances (fast, less power), less distances (less time for signals to propagate), and more "switches" per square inch (cut the linear dimension by 1/2, you get 4x more switches per unit area (plus more heating, more difficult to image/etch/etc.)...

With "Power" -- It is not so simple--In the digital world, we went from ~15 volts to 1.5 volts or even ~0.2 volts for switching speeds/low power/etc... You would not agree with the idea that your 240 VAC input power and 100 amps was now 24 or 2.5 volts at 1 amp, and you can stack a 1,000 "micro driers" in your garage to dry a load of laundry--Just does not scale the same way.

Many times, it is a fundamental change that can save money/improve on efficiency... For example, there are now transformerless GT inverters. Saves the cost and losses of the transformer--At the loss of "isolation" between the AC output and the DC Solar input. In the US, isolation was a requirement for safety...

But you could argue what is the big difference between 240 VAC and 200-600 VDC of the solar bus... Not a lot really, so why isolate the DC side from the AC side? So--The isolation requirement was tossed (and probably some leakage/ground fault requirement was added).

-Bill