I think HOMER is calculating the PV array output correctly. In the DMap you sent, black does not necessarily mean zero, it just means less than 5.4 kW. If you double click on that DMap and choose the ‘fine rainbow’ color scheme (below), you’ll see that in mid-summer the PV array does produce power late into the night. But the sun is in the northern part of the sky at such times, and because the panel faces south at a high angle, no beam radiation strikes the panel. With only diffuse radiation striking the panel, it produces little power.
The clearness index is the fraction of the extraterrestrial horizontal radiation that strikes the earth’s surface. That is to say, it’s the cloudiness index. The extraterrestrial horizontal radiation is the amount of solar radiation striking a horizontal surface at the top of the atmosphere. Here it is for Ambler:
That DMap, which I created by clicking Plot in the Solar Resource Inputs window, shows that the extraterrestrial horizontal radiation changes with season. (Of course it drives the seasons.) You don’t have to tell HOMER how it changes with the seasons; it figures it out from the latitude. But you do need to tell HOMER the state of the atmosphere – how much of that radiation striking the top of the atmosphere makes it through the atmosphere to strike the earth. That is what the clearness index specifies. The extraterrestrial horizontal radiation times the clearness index is equal to the global horizontal radiation on the surface of the earth.
So the latitude determines the quantity of radiation striking the top of the atmosphere. Weather patterns determine the clearness index. What strikes the ground is the product of the extraterrestrial radiation and the clearness index.