The majority of power usage in homes is heating, cooling, cooking, refrigeration and appliances. Lighting is already falling down the list of energy usage because of more efficient lighting solutions that are already widespread. It is a total red herring to discuss the power usage of computing devices, because they already constitute a tiny minority of power consumption.
Adding battery packs significantly increase the cost of a system by an order of magnitude. These are all true but they do not factor cost into the equation. With any of these additions, solar will still be far more expensive than baseload grid power, and for most will still not negate the need for a grid connection.
All these technologies will be fantastic for remote areas to obtain a higher standard of living, but essentially have zero chance of displacing baseload grid power as the majority of power usage in urban areas.
The only way they will win out in the long run is with punitive taxation of large-scale power generation, which is essentially a political dead end for all those who try it.
It's great to be enthusiastic about solar power, I personally like solar technology and will buy when it makes sense to do so on it's own merit. But there are limitations that are very difficult to overcome.
There is no requirement to use batteries to back up a grid connected solar house, any of the current generation options work fine.
The question raised was could solar PV supply a majority of residential electric load if it was basically unavailable at night.
In ten years, when the OP expects solar PV to be cheaper than all other forms of generation, how much power will really be needed at night?
I can easily imagine a house with very little night time consumption. Natural gas furnace, gas stove, gas water heater and dryer; nothing exotic. Refrigeration electric load can be shifted to the daytime by simply setting the controller to over-chill the freezer during the day.
That leaves computing devices, lighting, and entertainment for the night time electric load.
Don't forget hair dryers, vacuum cleaners, and microwaves, all devices that (in northern latitudes) are used before sunrise or after dark during the colder 5 months of the year.
Maybe we can totally flip culture on its head by going to work when it's dark and enjoying the few hours of sunlight all to ourselves.
> Adding battery packs significantly increase the cost of a system by an order of magnitude.
At current prices. Grid-scale batteries are expensive now because each is a unique specialty product. If we built sodium-sulfur batteries in the same volume as we build gasoline storage tanks, the price would decrease dramatically.
Efficiency gains also matter, because net system efficiency is a geometric process. Consider a 20% efficient solar cell and a 20% efficient battery: the system efficiency is a ghastly 4%. And that's not counting power conversion and transmission.
If the solar cell and battery were boosted to 40% efficiencies (physically possible), then the system efficiency would rise to 16%, a factor of 4 increase. That's halfway through the order of magnitude price problem you give.
Then there's heating and cooling efficiency. American buildings and refrigerators tend to have horrible insulation. In many localities, merely using proper insulation closes the rest of the price gap.
Lead acid batteries are 85% to 95% efficient. A good AGM is 98% efficient.
As a data point, I sized the batteries at my cabin to provide two days of power under total cloud (more than enough for getting to 24 hour solar power) and the batteries cost about half what the panels did.
(That discharges the batteries to 60% capacity. You also need to consider that the batteries will last about 6 years then get recycled. I mentioned AGM because they can charge much faster. If you are only interested in 24 power shifting, it might be cheaper since you'd need to buy more capacity in lead acid batteries to tolerate the charge rate. )
Does the actual efficiency of the solar panel matter? Of course it factors into the area needed to generate a given amount of power and given equal production costs more efficient is better, but people with lots of roof space are generally going to install the cheapest panels that meet their needs, not the most efficient panels they can find.
It matters to the extent that you need enough roof space. The last time I calculated I figured that the available south facing roof of a typical midwestern United States house was about right for the non-heating power needs of the house with current solar technology.
Adding battery packs significantly increase the cost of a system by an order of magnitude. These are all true but they do not factor cost into the equation. With any of these additions, solar will still be far more expensive than baseload grid power, and for most will still not negate the need for a grid connection.
All these technologies will be fantastic for remote areas to obtain a higher standard of living, but essentially have zero chance of displacing baseload grid power as the majority of power usage in urban areas.
The only way they will win out in the long run is with punitive taxation of large-scale power generation, which is essentially a political dead end for all those who try it.
It's great to be enthusiastic about solar power, I personally like solar technology and will buy when it makes sense to do so on it's own merit. But there are limitations that are very difficult to overcome.