Running default PVWatts residential results in 0 generation

I've been running the attached csv file through SAM with the default PVWatts residential profile (not changing anything) and I seem to be getting 0 output for the year which I don't understand.

Also, every now and again when I run similate, it will refuse to run and will give me the following error:

missing diffuse irradiance nan W/m2 at time [y:2018 m:1 d:1 h:0], exiting

Thanks, Jack

Hi Jack,

PVWatts requires a weather file with DNI and DHI data. Your file has DNI and GHI -- it will work with the Detailed PV model, but not with PVWatts.

To use the weather file with the Detailed PV model, you'll need to change the irradiance option on the Location and Resource page as described here:

sam.nrel.gov/forum/forum-general/3184

Best regards,
Paul.

Thank-you so much for your help Paul, I followed your instructions and it runs with the detailed model set to only have DNI and GHI. Unfortunately when I run it with the default settings, for some reason I only end up with a capacity factor of 2.0% and some really high losses - (-7.8% reflection, -22.6% model deviation from STC, -11% inverter power clipping). I apologise for spamming you with all these basic questions, but I can't understand why the model is predicting such incredibly low levels of generation.

Thank-you so much again for all your help,
Jack

Hi Jack,

It looks like the time stamps in the weather file might not be in local time, or the data is shifted. If you plot the DNI and GHI data from the weather file (click View Data on the Location and Resource page), you'll see that the solar irradiance data appears to be offset by about 12 hours. For example, in January, the sun rises around 6 pm and sets around 6 am.

SAM determines whether the sun is up or down based on the time of day, and does not calculate array output unless the sun is up. In this case, the solar irradiance data is zero or very small during the daytime hours, so you get a very low capacity factor.

I tried shifting the irradiance and meteorological data down 12 rows, and the moved the last twelve rows of data to the beginning of the year and got a more reasonable 20% capacity factor.

Also, be sure to change the array azimuth angle on the System Design page from 180 to 0 degrees for a north facing array.

Best regards,
Paul.

Paul, I genuinely cannot thank you enough for taking the time to identify this issue and for your advice on how to resolve it. I am so deeply grateful to you for all your help.

Many, many thanks,
Jack