Because total load on the power grid peaks at 8 PM when solar is providing nearly 0% of the power [1].

This means that no matter how many additional solar panels you add to the system, you are not displacing a single fossil fuel plant. Sure it's good that those plants are running nominally fewer hours per day per additional solar panel, but the cost to own, operate, and maintain the panels is roughly constant.

This results in the effect of increased prices for fossil power in a way that can't be offset by solar.

[1] https://en.wikipedia.org/wiki/Duck_curve

Industrial usage makes up a large percentage of total energy usage.

Industrial energy usage can be time shifted.

Who cares about residential power, if it is a small percentage of total energy usage?

That's what's generally happening. It's a bit of a limiting factor, though. As you have more and more renewable, you need more and more gas plants which mostly sit around doing nothing, which isn't free.

There's also the opposite problem, particularly in the case of wind turbines; if you have enough, then your base-load (ie coal, nuclear, hydro) plus renewable may sometimes exceed demand. This happens from time to time in Ireland, for instance, due to a large installed base of wind. At that point you have to turn off renewable, and that's not always trivial or instant. Storage would help there, too, by absorbing the excess.

> As you have more and more renewable, you need more and more gas plants

This is only true if the load on the grid is growing. In North America, it is not. Otherwise, you already have enough dispatchable generation available when the sun isn't shining or the wind isn't blowing.

This is assuming that you're replacing old, bad power (ie coal, or old-fashioned unresponsive gas plants) with a solar/responsive gas turbine combo. The US still has a fair bit of coal that could stand to be replaced, but I doubt there's enough gas to take up the slack of a solar/gas system without building more.

Over the past decade (since the wide-scale commercialization of fracking in the US), coal plants have been retired or repowered (with gas) at a record pace. I believe the share of generation has dropped from 50% to 30% in that period.

> This is only true if the load on the grid is growing. In North America, it is not.

The article I saw the other day showed growth of 4.5% (comparing Q1 2017 with Q1 2018)

Surely that growth in production is matched by a growth in load?

https://renewablesnow.com/news/wind-solar-farms-produce-10-o...

> Surely that growth in production is matched by a growth in load?

It is not, although consumption growth may trend upwards again as transportation is electrified (and oil consumption moves to electrical consumption).

https://www.bloomberg.com/view/articles/2018-03-01/americans...

The North American grids will need to start growing again, if electric vehicles become common.

That is true, and existing infrastructure does (mostly) handle the gaps.

How much does "existing infrastructure" charge to handle the gaps? This calculation differs wildly from as low as $6 to as high as $15, but I think $12 is the right answer.

So add $12 to solar only as a rough rule of thumb. Note that this $12 still comes with transmission issues, and environmental impact.

Without a buffer (like a battery) solar energy fluctuates highly because of clouds and other things that get in the way for split seconds during the day.

That seems like a point where distributed rooftop solar has a pretty good edge over utility solar (ie huge fields of solar cells). Spread over a large enough area all those tiny peaks and fluctuations will even out largely.