Hacker News new | past | comments | ask | show | jobs | submit login

So you are comparing two energy sources, one that provides constant power 99.999% of the time and one that provides volatile power ~35% (if best) of the time?

These price graphs hide the obvious fact. Normalize solar and power by adding the price to make the source constant(shitload of ESS maybe?) and removing government incentives.




Yeah, I, too, read slashdot when those arguments were still current. Now, they just aren’t. The stats above are already levelized. The change in output is also smaller than what people believe: with solar and gas as two independent systems, some improvements to interconnects to allow averaging over larger regions, smart devices on the consumer side picking the right times to recharge (and maybe even de-charge when it’s useful), and the improvements in battery tech, the path is all but inevitable.

It also wasn’t solar or wind power that lead to blackouts in France last summer, but nuclear plants that ran out of cooling opportunities because the rivers they use were overheating. And in Texas, it was natural gas that failed.


Not a single one of the power source/price studies I've seen has normalized the price in regards to making the source constant or government incentives. Not one.

Energy sources have their caveats. Nuclear plants using rivers decrease output when rivers overheat. Solar panel efficiency falls 0.38% per degrees celcius under and over 25. We think failing nuclear and gas is a big deal because it's reliable most of the time. We do not think failing solar panels in night (because they can't produce energy) or hot and cold weather is not a big deal because it's designed to be that way. If that's the case, how is it viable to compare the prices?


Comparing the price of unreliable energy with the price of reliable energy is like comparing the yield of a bond without taking default risk into account. It's just terrible.

Similarly thinking you can combine a bunch of unreliable energy sources and "tranche" them in order to get a smaller stream of reliable energy is very much like the gaussian copula problem that got us into the financial crisis. Yes, it's possible to do it in theory, but very hard to do so in practice due to the financial incentives involved and lack of knowledge about risk.

The level of discourse in our energy markets really needs to be improved, and quickly, otherwise we are going to make a lot of foolish choices.


Nothing runs at 99.99; 85% is a realistic nuclear reactor uptime. And remember to have multiple gensets so you can take them offline for maintenance.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: