Hacker News Reader

Comment by pfdietz | Hacker News Reader

That conclusion assumes no long term storage. Including hydrogen drastically reduces the overbuilding required. Put another way: if you overbuilt that much, you can make hydrogen for basically the cost of the electrolysers, since the input power is often free.

View on HN

You are correct that you can't create a stable, resilient grid just with intermittent renewables, you need storage.

But I don't think anyone has demonstrated long-term storage at grid scale, capable of storing power across seasons, which is needed in many areas. To be feasible the storage system has to be cheaper than maintaining some fossil fuel plants.

I haven't seen much about promising hydrogen storage systems.

> But I don't think anyone has demonstrated

Arguments like this really annoy me, if it's clear there's not a serious obstacle. It's just a "nothing can be done for the first time" argument.

I'm not saying it can't be done. I'm saying it hasn't been done yet. How about we try it before pretending that we can shift to 100% renewables or before legislating transitions that we don't know how to accomplish.

Hawaii is trying hard, but it doesn't seem to be going well with the grid becoming less reliable. BTW, this video is a solid overview of the power situation in Hawaii: https://www.youtube.com/watch?v=bbECmVdyWlQ&themeRefresh=1

I believe parts of the Azores have tried, but have not been successful.

I would be very interested in pointers to successful projects in this area.

I don't like how we talk about 100% renewables as if failing that means failure.

I'm happy with 90% renewables, a much easier goal, while we figure out the diminishing marginal returns later as technology breakthroughs and cost curves do the work for us.

Our objective is to minimize the area under the curve of future emissions. It's not "net-zero" in the abstract. Net-zero is just a proxy for the true objective. Getting to 90% renewables soon means less emissions (while being cheaper) than waiting an extra 10-15 years for 100% nuclear.

I'm not entirely sure what you mean by 90%. 90% of what?

You can't just have 90% of power generated by renewables and 10% by fossil fuels because there are times when you get 0% from renewables and so your fossil fuel plants need to provide 100% of the power. So you spent all that money on renewable infrastructure and didn't even get the benefit of shuttering your fossil fuel plants.

90% of power generation on average.

What you've written here isn't persuasive. Are you trying to say that keeping gas peaker plants as a backup will mean renewables are more expensive than nuclear? Where is your analysis that fleshes that out?

Especially when a simply cycle turbine plant has a capital cost per kW maybe 5% that of a nuclear plant.

It's harder to do it on islands with only a single weather system than on continents which have diverse weather systems.

You need a significantly higher capacity (read: expensive) grid to get continent-scale weather diversity.

> I'm not saying it can't be done. I'm saying it hasn't been done yet.

So, what point are you making then? That because it hasn't been done it shouldn't be done?

In any case, there's a perfectly fine explanation about why it hasn't been done: people largely can burn cheap fossil fuels and not have to pay for the negative externalities that imposes on the world.

You are trying hard to misunderstand me. Don't communicate as if it is a solved problem if your solution has yet to be actually been built and successfully demonstrates the feasibility of the solution.

I'm not sure how to interpret your last sentence. A demonstration project, especially subsidized by the government, doesn't need to adhere to any sort of market pressure regarding the price of fossil fuels.

It's a solved problem in the sense that it involves components that are all understood to work, integrated. This is the surest kind of innovation.

Now, we don't know how cheap it will ultimately be once these things are integrated and run down their experience curves. But pretending there's any serious doubt that they would work is more dishonest than arguing they would.

Your definition of "solved" is quite a bit different than my definition of "solved". Especially if you are going to just pretend that the economics aren't to be included in the solution.

Yes, your definition of "solved" is useful for nothing but obfuscation and denial. Mine is useful for planning what to do.

Come on.

https://news.ycombinator.com/newsguidelines.html

> it's clear there's not a serious obstacle

However, building grid-level storage is a serious obstacle

> It's just a "nothing can be done for the first time" argument.

No. The "it's easy, just build a lot of storage" argument is magical thinking.

How is it magical thinking? The plan is very explicit and all the parts are available: We build gas peakers that are H2 compatible; we build electrolizers; we upgrade the existing methane storage to be H2 compatible; we store hydrogen and burn it when needed. There is no "magic" there. Hydrogen is needed anyway to decarbonize industrial processes, so the cost for the infrastructure has to be paid.

You're omitting one tiny detail: we just have to build that at scales never attempted before.

Also, omitting the tiny detail of how complex it is to store hydrogen which you assume is the way to store energy.

We literally have the infrastructure to store TWh of methane, so we have already built at the necessary scale in the past. It is also not terribly complex to store hydrogen if you're okay with losing a few percent per year. You might be thinking of cryogenic hydrogen storage, which is definitely not the plan.

Everything is simple if you ignore everything. Because in the mind of renewable proponents all that grid-level storage is either already available or can be easily and cheaply built, no biggie.

"We have existing infra" ignores that this infra took decades to build.

"We know how to build" ignores that the storage costs can vary between 7 EUR/kg and 1040 EUR/kg [1]

and so on and so forth.

The capability for wishful thinking in renewable space is staggering.

Note: no one is saying it is impossible BTW.

[1] Great article: https://www.sciencedirect.com/science/article/pii/S036031992...

I don't think anybody ignores the costs. It's just that every other option is more expensive.

Yes, an energy system to supply the world is not a small thing, regardless of where the energy is coming from. The world is going to spend something like a quadrillion dollars on energy over the next century. There's a very large budget to do this.

I will note that the same hand wringing would apply to nuclear, only in nuclear's case it means we can't keep using today's burner reactors, since cheap uranium runs out. In contrast, we already have electrolysers (the pacing technology) at less than $3/W in China.

The thing is, that "hand wringing" is what renewable proponents always apply to nuclear. However, the moment you ask them about storage, "oh it's a solved problem"

No, what renewable proponents do is point out nuclear's egregious demonstated economic flaws. This is not a case of hypothetical problems, it's a case of actual repeated failure.

> is point out nuclear's egregious demonstated economic flaws

No, they don't. They point at an industry which was for decades was vilified, ignored and obstructed, and say "see, this is hard, costly, and nearly impossible"

The industry was for decades vilified and obstructed, but not ignored, because of the nature of the failure modes.

Mean cost vs. worst-case cost.

The mean cost of cleaning up nuclear reactors is tiny.

Eyeballing this graph: https://ourworldindata.org/nuclear-energy I think we've had around 100 PWh total generated by nuclear reactors over history?

There was only one Chernobyl ($340 billion inflation-adjusted), only one Three Mile Island ($2 billion adjusted), and only one Fukushima (probably around $200 billion adjusted).

If so, those accidents cost $542bn/100 PWh = 0.5¢/kWh, which is a trivial excess on even cheap (5¢/kWh) electricity: https://www.wolframalpha.com/input?i=%24542bn%2F100+PWh+in+u...

Unfortunately, two of those three demonstrate how expensive these can be when they go wrong, and that number is sufficient to just bankrupt quite a lot of countries — for example, Ukraine, which is where the Chernobyl reactor is, has a nominal GDP less than the cost of the Chernobyl cleanup: https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(nomi...

> basically the cost of the electrolysers

Yes, that's the expensive bit.

The reason you don't get much of this already is exactly that the capital cost of electrolysers is very high because of the requirement for platinum electrodes. And big capital investments are not free.

Platinum isn't required for alkaline electrolysis.

The reason we don't have a lot of hydrogen right know is because we don't have a lot of surplus energy right now. It makes no sense to buy expensive equipment and run in only a few hundred hours a year.

We have huge amounts of hydrogen right now. At standard temperature and pressure, the amount of hydrogen made globally each year would occupy 700 cubic kilometers of volume. It's just that it's made overwhelmingly from fossil fuels, mostly natural gas (6% of global gas production goes to making hydrogen).

That this is being done instead of making green hydrogen is because CO2 from the fossil fuel process is not being taxed or otherwise forced to account for the negative externalities.

This also illustrates that producing hydrogen without CO2 release is not an optionally solvable problem. Without the ammonia made from hydrogen (which is where most of it is going) billions will starve.

Reply on news.ycombinator.com