Oversizing inverters

[bcroe]

Considering a decade aging, converter losses, and DC wiring
losses (loops up to 400 feet), the effective DC:AC is around 2.1:1.
To my surprise, measurements in 2016 indicated I hardly needed
any south facing panels for best performance. That upgrade has
not been done, but there is still room for getting more energy from
the 15KW inverter plant.

The peak power applied is probably more like a 1.2:1 at any time,
more measurements could check. Good luck getting the inverter
mfr to approve this. Working here for 10 years. Bruce Roe

[Ampster]

This. Output can be reduced as much as 70% depending on the types of clouds for the time of day. My advice for people who live under high exposure to clouds should maximise DC/AC ratio as much as possible within recommendations of the inverter manufacturer to maximise output.

If I am not mistaken, that installation has a DC to AC ratio of nearly 2 to 1 with multiple orientations. Bruce can clarify.

[davidcheok]

In reality other considerations enter the picture. Starting with the above
idealistic approach, it was found that the frequent clouds might reduce
output to 1/8 or 1/4 of design. It was clear that regardless of any changes,
the $ per KWh per year was going to be higher than a cloudless situation.
I had the fun of watching all that equipment frequently doing next to nothing.


NScurve.jpg

This. Output can be reduced as much as 70% depending on the types of clouds for the time of day. My advice for people who live under high exposure to clouds should maximise DC/AC ratio as much as possible within recommendations of the inverter manufacturer to maximise output.

[Ampster]

It is not clipping a lot........

I know that, I was using the flat top to illustrate my point that too many people focus on the flat top and do not see long production and utilization of the inverter capacity for a greater portion of the day.

[bcroe]

Som people would look at that flat top as a lot of clipping. I see it as maximizing inverter output for nine hours from 8AM until 5PM a few days before the Equinox. It will only get longer until June.

It is not clipping a lot, I have not built the meter to see how much.
Guess a plot with one string disconnected would show, still pretty
flat top.

Unfortunately the sun moving north shades the panels at the day
start and end, so I am not adding hours. If I rebuild, that will be
taken into account. Bruce Roe

[Ampster]

Som people would look at that flat top as a lot of clipping. I see it as maximizing inverter output for nine hours from 8AM until 5PM a few days before the Equinox. It will only get longer until June.

[bcroe]

All the above are fine based on some assumptions, for a fixed array.
1. All devices are unrestricted in capacity, such as array, inverter(s),
wiring capacity, and Power Co contract;
2. cost tradeoffs are capacity of inverter vs array panels;
wiring, panel mounts, etc ignored;
3. loss of irradiation from clouds, etc is negligable;
4. converters and wiring are 100% efficient;
5. ROI over rides all other considerations.

In reality other considerations enter the picture. Starting with the above
idealistic approach, it was found that the frequent clouds might reduce
output to 1/8 or 1/4 of design. It was clear that regardless of any changes,
the $ per KWh per year was going to be higher than a cloudless situation.
I had the fun of watching all that equipment frequently doing next to nothing.

What to do? Building an additional identical system could boost total KWh
to the level desired. But I would then have the displeasure of seeing
TWICE as much equipment doing next to nothing. Anyway that was not
possible, because of all the wiring capacity and the PoCo contract.

Collecting more energy would take more panels, which have become
cheap. But could they still be used with existing plant? Orienting panels
in groups to favor East or West would allow getting to peak power very
quickly, and staying there till much later in the sun day. Because only
half the groups could reach capacity at any time, the power available to
the inverters would be essentailly the same as before, just for hours longer.
Understand, the inverters only see the available power, and do not know
the actually the panel ratings. Similarly, all the plant wiring capacity is
retained, even panels of different orientation may share individual wires.

What about the panel mounts? The sturdy ground mounts used here,
with good foundation and wires trenched in, cost more than the panels
mounted on them. So making them double sided allows double the
energy capture, for little increase in mount cost. Wind loading is about
the same

And clouds? Clouds disperse the light, so any panel orientation is
working. With double the panels, cloudy output is more like 1/4 to 1/2
of capacity, with light clouds allowing 100% output.

Conclusions.
1. The more clouds, the more the panel multi orientation advantage.
2. With the considerable savings of 2 sided mounts, and all the rest of
the plant, the cost per KWh delivered is competetive, if not cheaper
than a basic setup.
3. Inverters are not additionally stressed, providing the array is carefully
designed to limit peak power. Experiments show this is working.
4. Experience at this 42 deg latitude, is the seasonal sun rise/set varies
about equally north or south of a straight E-W line. So the double
sided panel mount is not badly aligned on the E-W line.
5. The elevation of E-W panels is set to LEVEL hourly power, not to
accomodate the season, so NO seasonable adjustment is desireable.
This angle turns out to be good for snow rejection as well.
6. Energy collected here on a sunny day is barely into clipping for
8 hours a day, 15KW of inverters produced 137KWh yesterday,
record is 158KWh. This fixed array is matching any tracker of same
AC rating.
7. Despite all the clouds, the multi orientation array here delivers more
energy per year than the basic system of same AC rating in the sunny
SW desert. It totally out performs a tracker under clouds.
8. The energy needs of this application are being met, with only a few
pecent to spare, fossil fuels are eliminated, ROI might be competetive
with other solar in the area but is not high on the list.
9. Effective DC to AC ratio should be adjusted down for inverter and DC
wiring loses.
10. This sort of system is easier to set up with string inverters and ground
mounted panels.

Graph below shows sunny 15KW inverter power output. Bruce Roe

NScurve.jpg

[Ampster]

Here is a GSES Technical Paper on the matter........

https://www.gses.com.au/wp-content/u...Oversizing.pdf


I followed their recommendation and used 1.5 to 1 DC to AC ratio. Has worked great for me....just make sure your inverter specs can handle it. I have two 7.5 Fronius transformer based string inverters and they have been quite happy with the increased power for the past 11 years. (Knock on wood)

That is a good discussion. Most people see clipping as a loss without adding in the additional production gained from faster ramp up of production in morning and slower ramp down in afternoons. It is total production for year that matters. Temperature matters also since panel output is reduced as temperature increases. Clipping on cool days in March does not always mean you will get more clipping on hot days in June. That was my experiene with a system with a 1.5 to 1 ratio in the wine country where the clipping in the summer was less that on the cool days of March.
The cost equation can be influenced by limitations of the main service panel busbar to handle a larger inverter. Main panel upgrades can cost several thousand dollars.

[davidcheok]

Here is a GSES Technical Paper on the matter........

https://www.gses.com.au/wp-content/u...Oversizing.pdf


I followed their recommendation and used 1.5 to 1 DC to AC ratio. Has worked great for me....just make sure your inverter specs can handle it. I have two 7.5 Fronius transformer based string inverters and they have been quite happy with the increased power for the past 11 years. (Knock on wood)

This is another good one.

https://global.apsystems.com/wp-content/uploads/2021/02/Appropriate-PV-module-over-ratio-can-increase-in-power-generation.pdf

[DanS26]

Here is a GSES Technical Paper on the matter........

https://www.gses.com.au/wp-content/u...Oversizing.pdf


I followed their recommendation and used 1.5 to 1 DC to AC ratio. Has worked great for me....just make sure your inverter specs can handle it. I have two 7.5 Fronius transformer based string inverters and they have been quite happy with the increased power for the past 11 years. (Knock on wood)

[davidcheok]

err..

The main reason why modules are oversized is because most real world environments do not align with STC conditions. Real world conditions will often produce less so system design standard is 1.25x the output of the inverter. Certain areas especially hotter climates require a much higher DC to AC ratio in order to provide sufficient energy to the inverters to convert to AC. Inverters only draws what it can use from the array. STC conditions that matter the most are the solar radiation of which 1000w/m2 is used and ambient temperature of 25C at sea level. Wind is also a factor.
There are two ways/schools of looking at system design. One is based on the output of the panel itself in which designers focus on converting every joule of energy from the panel and another is based on a fixed conversion in which the panel output is calculated so the inverter is producing maximum output over the longest time.

A lot of large scale solar farms use the latter where they aim to expand the curve outwards to either side and take the hit off clipping when solar radiation is at its peak. This produces a better yield over an extended period as the larger surface area will be able to produce more at all light levels.

'Clipping' does not mean the inverter shuts down. All it means is it is not using up as much solar energy collected. Throttling is when the inverter temperature has reached maximum safe operating range and the energy conversion is reduced in order to allow the inverter to cool before ramping up again.

Having a larger array size will always be better because it means you have more surface area to collect energy to feed the inverters.

An example of the system I have is the array is designed to provide my micro inverters enough energy (DC/AC 1.38) to maximise production when the panels are at its hottest i.e. 75-85C (our ambient is often 30-33C and wind is maybe 5-8m/s. Therefore for my system, at the hottest time of the day i.e. 2pm, my system output is still at maximum inverter output whereas a standard DC/AC ratio of 1.25 would have reduced output due to the hot panels at this time. You can calculate this roughly using the panel's temp coef which is roughly around -0.24% to -0.4% (mine are -0.34%) per C above 25C.

Screenshot 2023-04-27 at 6.47.07 PM.png

[foggysail]

NOMB, a respectful suggestion: You might want to get what looks like personal info off that shot.
Just sayin'.

Yes-- the only way I knew to edit that picture was to delete the post. I will try a different method later showing only the performance without my personal stuff. Again---THANKS

[J.P.M.]

Ipad screen shot just taken showing performance of my 28 400W solar panels

NOMB, a respectful suggestion: You might want to get what looks like personal info off that shot.
Just sayin'.

[Annlodge12]

Hello, I know this subject comes up a lot.

and I understand that a 400 watt module never puts out 400 watts, so some oversizing is logical.

but don't you think if you are oversizing at 150% that you are going to get some significant clipping?

one question I have - what exactly is "clipping?" when an optimizer/inverter is "clipping" does it shut down completely or does it continue to put out the max output?

If it shuts down at the max inverter output, i would think this is very undersireable because it's right at the time when you would be getting the max output. wouldn't it just make sense once you get over 125% or so, to just get the next size inverter?

gary in vermont

The oversizing of a solar panel is indeed important to ensure you get the most out of your module. It's recommended that you should aim for around 150% more than the module wattage, but this can vary depending on the application and components used. And clipping is when the inverter will shut down past a certain point, usually around 125-150% of the module wattage. This is done to protect the system from further damage or overloading, and it is desirable as it can help prevent costly repairs in the long run. It's also important to note that when an inverter clips at its maximum output, it can reduce efficiency and increase losses.

[J.P.M.]

29 degrees

Thank you.