Importance of Inverter Sizing - Good Read

[Poweri]

I did previously ask if it was ok to post links to external blogs that contained industry related and on topic content. I'm assuming it is ok as no one has said otherwise.

Inverter sizing article from our blog

Solar Inverter Sizing By Chris Roberts | Poweri Services Ltd

Feel free to comment and discuss

Poweri

 

[SibertSolar]

Useful post, thanks. I've often had this discussion with clients in the UK. We tend to limit undersizing to 20% below nominal kWp @ STC.

One key thing/rule of thumb I've learnt from inverter spec's is that the peak efficiency value is not as important as the average EU efficiency value and the delta between the two values is a good indicator of the inverter's quality and ability to harvest power at the lower end of the irradiance/power spectrum. Poorly designed inverters tend to have reasonable peak efficiency values but their efficiency tails off considerably at the lower end of the power/irradiance scale - making them not particularly ideal for the UK climate.

 

[Gavin A]

bit odd to pick an SB3000 inverter as the example IMO - 3-4 years ago maybe, but now I'd expect that the vast majority are TL inverters which have different curves and are more efficient at higher voltages and top end of the power range vs the SB3000 which is the other way around.

TBH I agree entirely with his point about undersizing the inverter / oversizing the array for the latest TL and HF inverters, but wouldn't so much for the SB3000, so it's a very odd example to use to explain the point.

Also, peak output would usually be in March / April when the sun is high enough, but the weather is cooler IME, rather than the peak of summer.

somewhere I've got an MSc project that demonstrates that systems with oversized inverters perform significantly worse than those with correctly sized inverters based on analysis of UK installations.

 

[BruceB]

And as long as you trust their modelling, then Sunny Design gives you an indication of annual ouput for different sizes/types of inverter on the same array(s).

 

[Earthstore]

This is where system design comes into play, an experienced PV company will design a system that will work at it's optimum for the customer, it is not just a case of bolt 4KW of panels to a roof then wire it up to a 4KW inverter and away you go.
Of course it will work, but it could work better if designed well for any given situation.

Inverter choice and size is probably the most important factor in getting the best returns from any PV system, (there are of course many other factors too), if the inverter is too large then it will be later starting and shutting down earlier every day, and on cloudy days may not start up at all, where as if it is too small you may miss all the "higher peaks" of production and potentially stressing the inverter shortening it's lifespan.

Even though there are several design tools for inverters available for the professionals, each and every system is different, so experience plays a large part in good system design, there also is the UK weather to try and predict too, even though it is good for PV production it is very varied.

I do agree with the other comments from the experts, that it is normally in the customers interest to undersize an inverter, but as already written by how much depends on many factors.

 

[Poweri]

This article was written some time ago, which will explain the SMA SunnyBoy inverter . The topic is still important today though as i'm sure you'll agree.

A previous company I worked for installed systems based on PV-GIS information alone! This made me quit the industry, however I'm back now working as a web admin for Poweri.

 

[Ian-LS]

Whats wrong with only using PV-GIS? If the correct info is used its a very very useful estimation tool. The database has been used for the new (highly flawed) MCS approved design and estimate method from May 1st 2013.

 

[Poweri]

Whats wrong with only using PV-GIS? If the correct info is used its a very very useful estimation tool. The database has been used for the new (highly flawed) MCS approved design and estimate method from May 1st 2013.

When I say PV-GIS alone, I mean no survey, no idea of shading, supporting structure of the roof, condition of electrics etc.