By Leen Weijers
Wind and solar don’t work most of the time. You may think intermittency is acceptable because the sun shines for free and the wind blows for free. Capturing these diluted energy sources, however, is anything but free. If you require electricity to be available at the flip of a switch from renewables, their temporal lightness requires massive overbuilding, making wind and solar the most expensive sources of primary energy.
Solar and wind’s capacity factor, the actual energy output vs the maximum energy output over time, is only 14% for worldwide solar and only 26% for worldwide wind. If this strikes you as low, that is true, as this number is weighted toward installations by Europeans, for example Germans, who are ahead in installations in some areas most unsuitable for solar. But there are other reasons solar and wind work even less than you think.
The map below shows that the sun shines a lot more in the United States than in Germany, as it lies further to the south. In both places, sunshine duration is significantly higher than hours of electricity generation at maximum capacity, because the sun’s angle is seldom perfectly perpendicularly to a solar panel. While the German sun shines about 1,600 hours a year, generation at capacity is only 940 hours. Texas has 3,000 hours of sunshine a year, but generation at max capacity is only 1,600 hours a year.
Texas solar therefore works about as hard as a hardworking Frenchman, about 31 hours a week. German solar clocks an average of 18 hours a week, takes lots of vacation days, sick days and strikes occasionally. Both workers are highly unreliable, only show up when they want and count on other colleagues to pick up their slack at any given time. Would you hire them?
No. And that’s the resounding answer for the capital of world solar, where collecting solar power supposedly makes the most sense. As discussed by Liberty Energy CEO Chris Wright on CNBC Last Call, together with the Northern European capital of wind, these two places represent the low-hanging fruit for solar and wind, where they are backed by massive subsidies. Population centers like China, Vietnam, Indonesia, India, Bangladesh, etc., have much poorer wind and solar resources, making their rollout in these countries much more challenging.
As is evident from the above graph, US wind and solar take long vacations. The US solar capacity factor is 12% during its winter vacation and 21% during summer. Wind works a bit harder – a 27% capacity factor during the summer/fall vacation and a 45% capacity factor the rest of the year. The seasonal variations result in significant shortages that are backfilled by coal and natural gas. If you want to get your average electricity without fossil fuel backup, the average summer/fall electricity gap from current wind (137 GW) & solar (100 GW) capacity is about 3.7x that capacity.
However, it does not stop there. Just overbuilding wind & solar by 3.7x over what is currently available will not be sufficient. The problem with renewable replacements is that just matching daily average electricity generation does not work. We are used to reliability in electricity every second. Eliminating intermittency and blackouts is the top requirement of our modern electricity system because not having power available at the flip of a switch can simply be deadly.
If you want renewable power without fossil fuel backup you need to account for situations where wind and solar take their regular “sick day”, i.e. when they work at less than 10% of their maximum capacity for the day. Below is a Texas example in late January – early February 2023. In this example of the lack of power generated by wind & solar, TX electricity council ERCOT was about 12x short in wind & solar of what natural gas and coal backup provided for three full days. There were 14 sick days for TX wind and 62 sick days for TX solar in 2022.
But it gets worse. Weather variability occasionally causes the wind to stop blowing for many more days in a row. The same applies to subsequent days when the sun does not shine. An evaluation of the frequency and duration of low-power wind (LPW) in Germany found that almost every year there will be 5 consecutive LPW days; every ten years, however, there will be an 8-consecutive day LPW event.
If ERCOT decides this needs to be avoided for the occasional week without flexible and reliable backup from fossil fuels, there are two solutions, both highly impractical and very costly: (1) overbuild the current wind & solar power generation by at least a factor of twelve, or (2) rely on massive battery power backup.
Currently, TX wind and solar capacity totals 64 GW. Overbuilding wind & solar for the sake of reliability would require ERCOT to match current fossil fuel backup capacity of 50 GW by about 12x based on my simple graphic math. ERCOT would need 770 GW in wind & solar. Rough installation costs would be about $770 billion (or $30,000 per ERCOT rate payer).
ERCOTs current battery power of about 5 GWh is enough to power its grid for about 6 minutes. Enough battery power to cover demand for a week requires about 8 TWh, which requires ~20,000 battery farms each with 100 MW capacity/400 MWh power, at a cost of about $3 trillion (or $120,000 per ERCOT rate payer). This number is so outrageous I decided not to pursue possible supply chain and building area constraints.
Despite the physical and economic constraints of the real world, there is a desire to overbuild wind, solar and batteries in many places. A recent study in Minnesota by Isaac Orr showed that existing 1 MW coal power plants are being replaced by 1 MW natural gas, 0.62 MW solar and 1.6 MW wind. 100% backup thus requires a 3.2x power replacement rate, with a disproportionate share going to low-energy renewables. Another plan, for 100% solar, wind and battery backup in Wisconsin, shows that state would need to build 7.7x its current reliable power capacity in wind, solar and batteries.
ERCOTs power building additions show a similar trend – it yearly adds to solar & wind power and battery backup, but peak load remains at about the level of reliable fossil fuel and nuclear power. Do you feel you need hundreds of extra bikes if you have one car to drive the distance?