I think this may have been debated before, but I am not certain. So here is my thinking: You do not use the 125% factor, when you obtain lighting load from table 220.12.
You say that lighting is continuous. I say that that is true, but irrelevant. I submit that the fact that lighting is continuous load has already been taken into account, when the numbers were inserted into that table.
Here are some arguments in favor of my point of view:
? Look at the first sentence of 220.12 (not the table, the article). It says that the values taken from the table give you the minimum lighting load. It does not say that the lighting load is 125% of the tabulated values, it says the lighting load is the tabulated value.
? The examples in Annex D do not add 25% to the lighting load.
? For dwelling unit calculations, the value taken from Table 220.12 gets added to other stuff, then we get to use a demand factor. To me, it does not make sense to ?biggie size? a number, and then turn around and ?downsize? the same number a minute later.
? Again for dwelling unit calculations, the value taken from Table 220.12 includes the load associated with a number of receptacle outlets. Those are not continuous loads. So out of the 3 VA/ft2 of lighting load, there is no way to determine how much to call the continuous load of lights, and how much to call the non-continuous load of receptacles.
By counter-example, I often perform a service load calculation late in the design process, by adding the actual watts of the actual light fixtures that I have placed on the plans. I do the math using excel spreadsheets in the format of panel schedules. That is actual lighting load, not a ?take a value from the table? lighting load. I do add 25% to that value, because lighting is continuous.