Hacker News Reader

Comment by radu_floricica | Hacker News Reader

Economies of scale. Most of the cost for a traditional nuclear plant is regulatory&decomissioning. If new nuclear can get over that, and it's allowed to go through a couple of iterations (which happens much faster nowadays), there is really no way to estimate how low the price can go. A 10x reduction is not out of question, if you look at SpaceX for example, another industry where innovation has been frozen for decades.

View on HN

But the fact is we have now cheap existing clean technologies, with a fast decreasing cost (decreasing cost of energy, and decreasing cost of managing intermittence)...

It does not seems wise to bet that SMR will be cheaper than renewable in 10 to 20 years and wait and pray...

SMR is not a viable option right now. That is an interesting possibility we should study and work on, that could be a good surprise later on, but right now 1) it is not ready 2) we don't know if it will be cost competitive... Putting aside nuclear risks and radioactive waste problem (that are minimized by SMR)

SMRs are being touted not because they are particularly promising, but because they are the last option nuclear supporters have. Traditional reactors have run out of chances, so now it's either the Hail Mary Reactors or give up.

Are there any known problems or difficulties with them? I guess the materials science aspect is extremely difficult as it is already a huge problem in conventional designs...

The big problem is diseconomies of scale in construction and operation. They require more material per MW of power, and more staff too. The story is that economies of scale of manufacture, and changes to regulations about operation, can counter this, but that remains to be demonstrated.

Indeed. We need to keep rolling out renewables until nuclear is ready.

> But the fact is we have now cheap existing clean technologies, with a fast decreasing cost (decreasing cost of energy, and decreasing cost of managing intermittence)

The cost of solar and wind energy when the sun shines or the wind blows has certainly decreased a lot in recent years, you're right about that.

But decreasing cost of managing intermittence? I haven't heard of any big recent advances in that (aside from the successful use in Australia of batteries to damp fluctuations on a timescale of minutes, which is great but doesn't address timescales of hours to months). What am I missing?

Battery storage has got way cheaper. 4x cheaper per kWh since 2013, according to [0].

They're still not cheap enough to make solar + batteries price-competitive in most markets, but their cost absolutely has been decreasing fast.

[0] https://cleantechnica.com/2019/12/04/powering-the-ev-revolut...

Considering the fact that battery storage would need to undergo several orders of magnitude in price reduction to provide true grid-scale storage, does this really matter? Battery storage is, at best, going to provide municipal load for minutes or maybe hours, but it is unlikely to cover a 24-48 hour power ebb anytime in the next few decades.

If battery storage keeps getting cheaper at the same rate, then yes - two orders of magnitude would be seen in twenty years. Absolutely an important consideration if you're considering a big upfront investment in a nuclear plant whose returns will be realized twenty to seventy years from now (or whatever reasonable build time and operating lifetime you come up with).

Needing "several orders of magnitude" improvement is also pessimistic. There are installations that are borderline economically profitable right now (Tesla battery farm in South Australia): another factor of 5 or 10 makes those attractive in many more places.

Those installations are borderline profitable because either they are a small drop in the bucket of the required storage capacity (c.f. Tesla battery farm) or are heavily subsidized to make renewables appear viable. You are also making some large assumptions regarding battery improvements that may turn out to never be realized -- there may be interesting new chemistries and techniques that work for smaller scale but are infeasible for one reason or another at large scale. It is also possible that the opposite is true (e.g. like for fuel cells) where bigger is better in terms of efficiency and production costs, but so far I have no seen any indication of this.

Battery prices (for different battery technologies) are currently in free fall, and the market is booming, so we are 100% sure to see important economy of scale there. This in currently not the case at all of SMR...

When it comes to technology at pilot scale, price and capacity improvement may turn out to never be realized... But there are dozens of promising tech, so it seems unlikely that all tech will "fail"... and SMR exhibit higer level of risk cause it is even not at the pilot scale project, and it is mostly one technology (with 2 or 3 varieties)

And there are many more tech at the lab stage

And there is also thermal storage, with many proven and cheap technology, and many different innovation at different stage in the pipe

And there are all the innovation to reduce the intermittence and help manage it

And there are all the innovation and intervention to allow to make energy demand more flexible

SMR is based upon well-known technology that we have spent decades working on and improving. Other than pumped hydro and batteries there is no potential grid-scale energy storage mechanism that exists beyond desktop models and toy pilots.

Battery prices are dropping but the easy low-hanging fruit has been plucked so now the rate of price decline is dropping -- 83% last decade but only 33% predicted over next five years. Production is also a long way off from where it would need to be. There are a couple of GWh plants coming online, but that is still only a fraction of what is needed if battery is going to scale up to handle grid storage.

Lots of hand-waving examples you have there, very little in production. A more likely long-term outcome will be to push the storage burden on to the renewable production plants -- only accept power from a producer at a rate they can provide over the next 24 hours with the grid only needing a small surge capacity built into it to smooth out demand spikes.

SMR are not even at the toy pilots stage. We know the tech and we know it can work, but we don't when, what investment is required and if it will be competitive. A lot of uncertainty, but one tech only (with few different variations). There are dozens of storage tech with the same or lower uncertainty level... So dozens more chances a tech will work...

About storage tech working now, you forgot possibly the most used right know and the cheapest one: thermal storage (giving up to 9,000MW to the French grid for example!). With a huge and cheap potential and mature cheap tech.

There are a combination of different things.

Firstly, better capacity factor for onshore wind, offshore wind becoming cost competitive and the emergence of floating offshore wind reduce a bit the intermittence, then the cost of managing it

Secondly renewable is becoming so cheap that installing more than peak demand and setting up curtailment is cost competitive and reduces the intermittence and the cost of managing it. There are also solar project not optimizing for maximum production, but for producing when needed.

Solar-plus-storage is becoming more and more common for new project, with cost falling. Generally all the batteries cost are in free fall while we are just at the beginning of the market expansion: massive economy of scale are on the way, with proven technologies

More electrical storage are coming... Flow batteries, pumped hydro on isolated reservoir (eg Gordon Butte), liquid air, underground compressed air, stacked blocs, hydrogen... And there are many other tech in the lab at the same level development of SMR. Of course all the promising tech wont be a success or be cost competitive, but with SMR you bet on one tech, here you bet on more than a dozen tech.

Virtual Power plant and Demand-response project and market are blooming (and that is a start)...

This last point open a wider market for thermal storage, with existing cheap and mature technologies, and many developing project (relatively low tech) like inter seasonal heat or cold storage

Why do you frame it as a decision between the two? The market is beyond huge, there are bound to be use cases where one or the other is clearly better.

Also a 10x cost improvement on nuclear would blow everything else away (in a good way). I don't think renewables have the same potential - for one thing, they're a more mature technology (weird, I know, but they've been the focus of global development for quite some time now).

And also there's only so much energy you can get out of solar and wind - you just can't get more than 1.3 kw from a square meter of solar, you have to cover more square meters. On the other hand there's no theoretical reason why you wouldn't have a clean, safe reactor in each car. It's just a technological challenge.

Reply on news.ycombinator.com