SMA 3000TL with 4kW Solar Panels

[pookey]

Hi guys,

We are configuring up a 4kW solar pv array and the Sunny Design software (latest DB) is recommending an SMA 3000TL with our LG 250w - 4kW system.

The annual yield production shows a difference of 6 kWh compared to the 4000tl !

Thinking this was a bug I called SMA and they confirmed that the difference between the inverters for a 4kW array is hardly anything. They said that although you wouldn't catch the peaks on the 3000TL, it would pick up quicker and would work more efficiently.

Can anyone lend their thoughts to this please?

Many thanks.

 

[SolarCity]

It is an interesting question.

I agree with SMA's answer that the 3000TL will work more efficiently. However, at Segen's technical conference earlier in the year, SMA's Brad Morgan suggested that undersizing inverters would actually result in loss of yield on the peaks yet give no extra advantage in efficiency. I caught up with him afterwards to make sure I understood him correctly and he confirmed it.

It left me scratching my head as it goes against what I have understood in the past and certainly goes against what analysis software suggests.

 

[JulianC]

Aren't there two separate, albeit related elements at play in undersizing inverters?

1. The array peaks are infrequent in the scheme of things so yield loss is minor or insignificant in relation to increased inverter efficiency the rest of the time. Or apparently not according to Brad Morgan.

2. The latitude, inclination and azimuth of the array is such that its kWp rating would never be attained in reality. Your 4kWp array is therefore effectively downgraded to a 3.6kWp array for example, thus being suited to a smaller inverter.

 

[SolarCity]

To be fair, I have seen 4kWp arrays produce far more than 4kWp before but I understand your point.

 

[pookey]

Thanks for your comments guys, I'm leaning on the view that in the real world a 10% loss would be realistic for a 4kW system so at 3.6 kW the 3000TL would be fine over the longer term. It's just interesting to know that we have another option as far as inverter choice is concerned.

The technical chap I was passed to was based at SMA in Germany, so it's interesting he took a slightly different view to Brad.

 

[moggy1968]

does location affect this as well? On the south coast my arrays produce in excess of 3.6kW quite often from a 3.8-4.0kW system. I would presume in Northern Germany or UK this may not be the case and so downsizing may be more appropriate in the North than the south or on East or West facing arrays.
The main advantage I see from going to a 3.6 is the G83 threshold. If the threshold were higher I would probably look at 3.8 or 4