Comment by gregwebs | Hacker News Reader

Comment by gregwebs | Hacker News Reader

Parent Full thread

The article on the site about solar is equally bad. Both are full of lots of good information but then have some key misunderstandings. The solar article doesn't acknowledge that installing a solar requires installing a gas peaker plant on the grid as well. [1] Although solar certainly reduces carbon emissions (the gas peaker plants only runs some of the time), the result is that the cost to the grid, rather than being the lowest, could be one of the highest depending on how the math actually works out. Battery can be installed instead of peaker plants, but the prices of raw materials for batteries have increased dramatically recently and it isn't possible for the mining of battery materials to keep up with the future demand. Tesla's car prices are shooting up- even though they are building huge battery factories to drive down price- it's not enough- they are now exploring mining Lithium, having already bought mining rights for it in Nevada.

[1] https://www.nber.org/system/files/working_papers/w22454/w224... Does anyone have a more up to date citation?

The site actually has several articles about why this specific criticism is silly.

To summarize: we already have gas plants, and gas storage, leaving them turned off except as a fallback insurance is the sensible option. Even when building new plants, gas plants are the cheapest to build and simplest to run which matters a lot if they're mostly idle, which is why they get used as peakers.

https://austinvernon.site/blog/gasredemption.html

I don't agree with everything they write (they seem to think "virtue signaling“ is a real thing and that renewables had nothing to do with sensible government regulation but rather the free market which reveals a bias that runs through their conclusions) but the facts are fairly incontrovertible if you actually look at them so they come to the reasonable conclusions anyway.

That’s another mostly great article you linked to. Still though it doesn’t mention the dramatic cost increase of natural gas plants when used as a peaker instead of always-on, and it assumes we can produce as many batteries as we want. It also assumes we can continue producing massive amounts of natural gas in the US and export huge amounts to the rest of the world.

The author's article on renewables and batteries states "The cost of these batteries is plummeting. Material constraints are not an issue because storage batteries do not require top-tier energy density" [1]. This was reasonable to think just two years ago but now is obviously wrong- and this assumption underpins most of the author's writings on energy.

Energy policy is very unforgiving- missing one important fact or one wrong assumption can dramatically alter the outcome.

[1] https://austinvernon.site/blog/renewablevariation.html

I'm not aware of anything occuring in the last couple of years that put a hard limit on battery construction due to material constraints. Could you be more specific about what you mean?

The price of the raw materials for batteries is going up dramatically and the supply is not increasing accordingly. That's with today's demands- not a future where we need 10x the battery capacity.

There's been a lot of attention paid to nickel recently, but nickel may have enough availability with Tesla getting New Caledonia mining back online and Indonesia being ready to mine their nickel deposits. Still, the price of nickel is up ~2x right now. Cobalt is also up ~2x and has major ethical issues. But let's assume the industry can move at least grid storage applications to the LFP chemistry and reduce the pressure on these minerals.

The price of lithium is up ~5x. It's easy for this author or Elon Musk to say that there is plenty of lithium in the ground, but it's a lot harder to actually produce it. Tesla is experimenting with new lithium extraction techniques in Nevada- this will take a lot of time to come to fruition. The easiest scenario is existing mines increasing their production, but even in these cases it can be a long process to satisfy environmental concerns. Historically commodity super cycles last 10+ years with metals being much longer.

I would be interested in someone with more knowledge than me projecting the future here on both the demand and supply side, but instead in the article you linked we get "Lithium is not rare, but demand is growing so fast mining may take a few years to catch up". I don't think the author appreciates how long it takes to bring significant mining capacity online. And certainly the stated assumption that the "cost of these batteries is plummeting" is completely untrue at this point. Those with a deeper understanding are warning of a possible battery shortage [1].

[1] https://www.wsj.com/articles/rivian-ceo-warns-of-looming-ele...

There's nothing new there. Musk has been talking about how many "gigafactories" making batteries the world would need for a long time. And even battery manufacturers were a bit taken aback by the scale proposed.

> Three years ago, when the Gigafactory was first proposed, "I thought it was crazy," said Yamada. "At that time, production capacity of this Gigafactory would exceed total production of the industry. Not Panasonic. Not Japanese companies. All Japanese, Korean, and Chinese companies combined. I thought it was a crazy idea.

"But I was crazy. And I was wrong. After seeing extraordinary success of the announcing of Model 3, there is a strong demand for this battery. So, three years ago I thought this was a crazy idea. But I was crazy at the time

It's not the battery cell production that is the real issue. Battery cell production is a big issue, but production can be increased in a few years to keep pace with demand if the raw materials are available.

The new aspect is the price signal that Lithium is now sending. This means that lithium will limit the ability to increase battery production and that battery prices are not going to decline as they have in the past. The history of commodities means this situation is more likely to hold for a decade than a year.

Gas peaker plants are needed for nuclear just as much. Generally the article is looking at the most favourable conditions for the technology, e.g. only considering OCC and not interest cost for building and decommissioning of nuclear plants.

Nuclear is producing 100% output during peak power needs. Solar varies greatly but in the winter may be producing little output during peak power needs. So yes, a 100% all the time output requires peaking, but a variable power source that is not producing during peak times requires much more.

You're going to have to source this because there's been absolutely no need for gas plants in the two cities I've lived in that were both powdered by nuclear in Europe.

Energy networks are not city-specific but are cross-national. Or are these cities self-sufficient and not connected to the rest of the grid?

It's right there in the article, but it seems many here commenting did not even read it:

Many US nuclear plants are incapable of cycling power output up or down without changes to procedures or equipment.

Where? Searching for 'peak' returns no results and going through the results for 'gas' doesn't turn up anything about peaker plants for nuclear.

I just cited the section in my post. US nuclear plants can not follow load at all, some European ones have some ability, but it's generally not enough.

If you compare France and Germany gas portion of the overall electricity production is quite similar (it's a bit tricky to compare, Germany produced 12% and France 7% but in different years, France gas accounts to 10% of their installed capacity)

Reply on news.ycombinator.com