Total kW output in relation to string sizing

2 posts / 0 new
Last post
Total kW output in relation to string sizing

I have an unresolved discussion with our electrical engineer over a simple issue of string sizing. Everyone understands the benefit of running the inverters "hot" with longer string sizes but as the financial guy I am only interested in total output.

We have a simple project with 90 Yingli Green Energy 260 watt panels connected to two Fronius 11.4 Uni inverters (Lat 44.92). I suggested we use nine strings of ten instead of eleven strings of eight(roof limitations prohibit any other configuration)because SAM output comparisons show the 90 panel configuration produces 600+ more kWhs annually. He says that is incorrect. He claims that the 88 panel configuration will produce more because it has higher efficiency. Its not a big deal except that 6000kWhs is worth ~$4000 to the customer over ten years.

His claim is that the SAM program does not account for the gained efficiency of running higher voltages per string.

Can I rely on the SAM modeling to provide me with the highest output or do I have to rely on the engineers to make the call.

Thanks for your input.

Paul Gilman

As a general rule, I think I would consider the recommendation of an experienced and qualified engineer for decisions like PV array layouts to be more reliable than results from a model like SAM.

You didn't mention the array's orientation or a specific location, but for the system you describe in Portland, OR (latitude 45.6 degrees, similar to the 44.92 degrees you mention), the 9-strings-of-10 system generates 27,845 kWh of AC electricity annually, while the 11-strings-of-8 system generates 27,190 kWh. That's a difference of 655 kWh per year, which is similar to the difference you mention, and is roughly 2% of the total annual output.

A 10-strings-of-8 system generates about 1% more electricity per year than a 8-strings-of-10 system, the only difference between those two systems is the array voltage -- I used the same number of modules per string as in each of your two systems, but kept the total number of modules the same to investigate the impact of array voltage on system output.

How confident are you in your assumptions?

I used the TMY2 weather file for Portland for that comparison. Using the TMY3 weather file instead changes the 9-strings-of-10 output to 27,991 kWh, an increase of 146 kWh per year.

The default soiling factor on the Array page is 95%. Changing it to 98% results in an annual output of 28,672 kWh, about a 1% increase from the original 27,845 kWh.

My point is that the level of certainty of your input assumptions may mean that the 2% difference between the two layouts is within the error margin of the results. A possible $400/year savings over 10 years for a roughly $80,000 system may not be strong enough basis for a decision about numbers of modules per string.

I hope that's helpful.

Best regards,

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer