Hacker News Reader

Comment by epistasis | Hacker News Reader

Fervo isn't just trying, they are succeeding in bringing the drilling advances used in fracking, along with other innovation to get through the harder rocks typically encountered when not drilling for fossil fuels, to deliver dispatchable and storable energy. It's actually much better quality of power plant than nuclear because it can be scaled up and down throughout the day economically, whereas nuclear becomes even more uneconomic if it is forced to match the needs of the grid, and the reactors typically require far different designs than what is usually built (France has done a bit with this).

I would bet on geothermal over nuclear in a second for future electricity generation. Its so much more promising, has a tech curve, and has far more innovation and advanced tech adoption.

View on HN

> It's actually much better quality of power plant than nuclear because it can be scaled up and down throughout the day economically...

It's not that nuclear reactors can't be built to vary their output. It's that nuclear power is almost all fixed cost. If the average power level over a year is 50%, the power cost doubles. Geothermal has similar economics.

Not quite, with geothermal the storage is built into the system by simply limiting the intake, which builds pressure and can shift energy production throughout the day.

The cost of enhanced geothermal is roughly the exact same cost as nuclear today (if you exclude the very high cost of failed build attempts for nuclear), and it shares similar economics of a very very low OpEx to CapEx ratio. However the economics differ massively in that enhanced geothermal is getting cheaper as we build more, but nuclear tends to get more expensive as we build more. Geothermal is a technology, nuclear is a monument.

> Not quite … The cost of enhanced geothermal is roughly the exact same cost as nuclear … shares similar economics of a very very low OpEx to CapEx ratio.

It can’t be both not quire and the same.

All else being equal, a five billion dollar geothermal plant running at 50% has the same problem as a five billion dollar nuclear plant running at 50%: where’s my profit?

Sure, load following is trivial with geothermal, but nuclear generally isn’t trying to compete in that space, so we can discount the difference there.

Unlike nuclear, that time running at 50% isn't thrown away capital expense, it is merely delayed power output that can be utilized with slightly higher turbine capacity, which is the cheaper part of the capex, and would be needed for higher power output anyway.

For nuclear, adding thermal storage for time shifting would be the most equivalent to what's happening with geothermal storage, but with geothermal there's no additional capex or engineering needed.

> that time running at 50% isn't thrown away capital expense

I don’t understand how you think capex works?

You outlay x on capital expenses to get the plant running, that’s all the plant and equipment. You run the plant at 50% it takes you twice as long to recoup your capex.

Always happy to be corrected.

Geothermal is limited by the flux* of what you can pull from the ground, which sort of scales with capital costs.

Reducing the draw from the well at time X, leaves more available at time Y. In this way, the capital cost is reasonably preserved and your correction is offered.

--It's not exactly the same thing as flux ---I haven't verified the relative losses here of delaying utilizing the available heat, but rather am assuming the OP is correct about what I believe to be their point*

I still don’t follow.

If the geothermal plant isn’t ran at or near capacity all the time, the capex takes longer to recoup.

You can’t just, say, run it at 50% for a while, and then at 150% later, because a) ya typically can’t run plant at 150%, and for power generators ya can only run them at whatever you’re contracted to supply, so you’d typically want to run geothermal at capacity all the time.

This is true of anything that requires capital expenditure.

Drilling the well is the expensive part. It heats at a relatively constant rate depending on the geometry. You could size the plant to be exactly matched to the well output--but the generators and such are relatively cheap. So instead you make that part of the plant slightly oversized, so you can run at over the well's capacity when electricity is expensive, and under the capacity when it's cheap. The thermal mass of the rocks allows you to average this out over time.

So there are two capacities, that of the well and that of generation (oversized with respect to the well). On average this varying scheme utilizes the well at 100% of capacity (the expensive part to increase), and the turbine generation at less than 100% capacity (not expensive to oversize).

You can run it at 400% by building more generators and heat pumps. They aren't the expensive bit.

The expensive bit is drilling and that gives you roughly X heat per month. You can use that at a constant rate or all in one week.

Like someone below said:

> They just let pressure build up higher than normal for 6-8 hours and then the turbine generates more power than normal until the pressure falls back to normal levels again.

So they can run at more than 100%.

Everyone is overthinking this. They just let pressure build up higher than normal for 6-8 hours and then the turbine generates more power than normal until the pressure falls back to normal levels again. This would not take 50% and make it 100% again, but it does give you something.

Many turbine are running in extreme high pressure (for efficiency). There are no much margin for "build up higher than normal".

Of course we can build something less efficient to allow "pressure build up", but that's another trade off.

It's not pressure on the turbine that builds up, it's pressure at the well head.

> Sure, load following is trivial with geothermal, but nuclear generally isn’t trying to compete in that space, so we can discount the difference there.

Isn't this logic flawed?

> Sure, Metric A is better with Option A, but Option B is so bad in that space they are avoiding it, therefore we can discount the difference there.

Geothermal isn't cheap. Trimming those fixed costs means siting on fault lines/earthquakes and higher opex (insurance).

Fervo/Google got dogged for announcing their plant in UT because they avoided disclosures about the capacity [0]. It's more of a very small scale pilot of a couple MWs, but they buried key facts about the project assumedly on purpose due to lack of significance.

[0] https://blog.google/outreach-initiatives/sustainability/goog...

Fervo's initial demonstration project was next to an existing power plant in Nevada which previously failed to produce at it's stated capacity over time (Battle Mountain) so they were able to tie in extra MWt capacity to an existing ORC turbine. Fervo's technology has to be located somewhat near existing traditional hydrothermal geothermal resources because it's the convection along an exiting fault for hundreds of thousands of years that produces an above background thermal gradient near enough to the surface for it to be economical. That is true for their demonstration area in Utah which is located near the existing Blundel geothermal power plant in Milford Utah.

Sure. Exposing these situational constraints and free benefits (third-party sunk costs) aligns with my stance.

I don't agree with the above comment:

> Fervo isn't just trying, they are succeeding

I think as a tech demonstration project it was successful because they were a bit conservative in some ways that will make the economics look worse. I agree it's far from "geothermal everywhere" which seems to be the hype. You can't extrapolate that from one successful EGS well literally right next to an existing geothermal power plant.

> they were a bit conservative in some ways that will make the economics look worse

Or they simply ran into headwinds on a speculative project. I'll "take the under" when your PR is cagey about basic project attributes.

They do a good job of publishing their results in technical industry publications (advancing the field overall in a surprisingly open way) but I agree can be misleading in their marketing.

It will be interesting to see the results of the Cape project once they do multi-well laterals from a single pad power plant with larger diameter wells. That is really more a demonstration of power plant economics beyond the technical feasibility of creating a horizontally fracked reservoir that can be operated for a year.

Ah!

Cape Station does look much more significant. [0] 400MW of power plant capacity with 2028 COD and mostly contracted with SoCal Edison? Good job.

[0] https://www.utilitydive.com/news/cape-station-enhanced-geoth...

Can't you just burn up the power, e.g. with water->hydrogen conversion, water desalination, pumped hydro, etc. Whenever the grid isn't demanding 100% for your power you throw it at some other profit-generating venture.

Granted you then have to built of of those power-using-things and only run it when grid demand drops.

> Can't you just [store the power]

If you find a cheap solution to power storage, you can make a lot of money. This is the key enabler for a 100% renewable grid.

We already do electricity -> hydrogen generation, the hydrogen is then stored in underground pumped storage (empty gas wells), and can be used in place of natural gas.

Although presumably if you set up one of these facilities you need it running 24/7 to be profitable not just when there is a lull in demand elsewhere.

you'd be the first trillionaire.

> e.g. with [...] pumped hydro

That's done a lot in France, but there's a limited amount of pumped hydro available in the end.

Until recently, countries with lots of available nuclear energy didn't really need to produce fresh water, and non-gas-produced hydrogen is still a WIP.

Relatedly, I've read a couple of neat articles on using mine shafts as gravity batteries: https://bigthink.com/the-future/coal-mines-gravity-battery-e...

Major footnote here that I'm wayyyy out of my wheelhouse on this stuff, so there may be reasons that this doesn't work. I invite correction if that's the case so we can all learn some stuff :)

Gravity batteries have horribly low energy densities. There's usually a better option.

I'd say that there is always a better option. Can someone point to a case where a gravity battery would definitely be better than all the alternatives?

Pumped hydro is a form of gravity battery. It doesn't have great energy density, but it has fantastic power density and responsiveness. That's where its strength lies. We have also probably already built most of the ones that could possibly be built.

Closed loop pumped hydro can be built anywhere there's a hill, there's tons of available expansion capacity there

There are 86 large pumped storage sites in the world. It does work, but you need the right geography. You need two good reservoir sites at considerably different levels close to one another. That's somewhat hard to find.

https://www.volts.wtf/p/whats-the-deal-with-pumped-hydro

David Roberts

Yeah, we have a lot of hills. And that's it. So, other than that, you could plop one of these down almost anywhere.

Erik Steimle

That's correct, yeah. You need some proximity, obviously, to transmission and load.

Have you heard about xenon poisoning? Load following with a nuclear power plant is much more complex than it seems

The LWR is a difficult case for xenon poisoning (compared to other reactors) because of the thermal spectrum and the lack of homogenization (it's not just a temporal problem but a spatial problem) but if you add reactivity swing it can be managed. It's a problem for going from 0-100% quickly but not a problem for following loads across the day, see

https://www.oecd-nea.org/upload/docs/application/pdf/2021-12...

This already makes a large assumption about the type of reactor you are using. For example, a liquid fluoride thorium reactor (LFTR) does not suffer from the xenon poisoning issue, as do many other designs.

I don’t believe there are any commercial LFTR plants?

It's an interesting thing that pro-nuclear people always talk about "Gen X" power plats with no issues and just a big fat gogo stamp on them, but when you ask about any commercial examples they all come up short.

I'm pro nuclear, but I don't think using hypotheticals and futures is how to convince people nuclear is good, it's good because it works and it doesn't poison the planet (Yes there have been accidents and they have been dramatised but count deaths and it becomes as irrational as being afraid of flying)

It takes decades to turn on a new plant which is why you won't see the latest and greatest commercially deployed anytime soon.

Molten Salt reactors are designs from the 60s-70s and afaik no one ever built on commercially. No one was willing to take a gamble on building a new design when we had working reactors already and approval for those was already difficult enough.

AFAIU the corrosion problems are still not solved from a commercial PoV. Ie, without having to shut down and replace piping too often for it too be viable.

I had not, but TIL about xenon poisoning and that it was a major contributing factor in the Chernobyl accident (the tl;dr is they did a test where they reduced the fission rate; there was a buildup of Xenon 135 because of that, which caused the reaction to not start back up as fast as the operators thought it should. They removed the control rods almost completely, and when the Xenon stopped doing its thing they had a runaway reaction on their hands.

The xenon poisoning is why they took the control rods all the way out, but the runaway reaction was probably caused by the graphite tips on the ends of the control rods. As the control rods were scrammed these tips passed through areas of high neuton flux and caused a spike in power which probably caused the explosion.

There were a lot more factors at play in the chain, but the burnoff of the xenon wasn't itself the proximal cause of the explosion.

Complex or not, it's doable as the nuclear power plants in France are doing. Germany used to do it as well, until they shut down all their nuclear plants.

Though due to the xenon poisoning, they apparently use the reactors which at the moment have the freshest fuel for the load following, as they have more excess reactivity available to overpower the xenon. They can ramp at something around 5% of rated capacity per minute between around 30-100% of full power.

Most other countries with nuclear power plants have a much lower share of nuclear in their grids, so they haven't needed to do it, as due to the economics of nuclear it makes the best sense to run flat out as much as possible in order to recoup capital costs.

>Geothermal has similar economics.

Doesn't the ground itself act as an energy reservoir for geothermal? I haven't looked this one up, but it definitely seems like geothermal should be very dispatchable.

My understanding of that most nuclear power plants currently operating can already scale up and down to match demand.

They don't because I fuel costs are so low that the power is essentially free when compared to idling.

That's correct. There's nothing technically special about the French nuclear plants that load follow. It's just that France has so much nuclear power in the grid that some of their nuclear plants do load following.

Nuclear reactors can indeed be made to adjust their power level very quickly, but this typically requires the use of highly enriched uranium which introduces new costs and problems.

> I would bet on geothermal over nuclear in a second for future electricity generation.

I don't understand that though; geothermal power is on paper much more low-tech, safer, cheaper, and deployable everywhere compared to nuclear. Why didn't it become the default way to generate power?

Or is deep drilling in fact more difficult or expensive than nuclear fission?

It is quite expensive, the heating reserves are finite and when you look closely at the details there are a lot of challenges. It isn't quite as easy as putting a pipe in the planet and pumping heat up (even though that is how it is portrayed).

> Why didn't it become the default way to generate power?

It's because unlike nuclear technology, drilling technology has advanced massively. It was spurred on by natural gas fracking, trying to get very hard to recover fossil fuels as the easy stuff ran out.

So there's an entire industry around this new technology that already exists but which has not yet been applied to geothermal. And for a long time there was little no no growth in electricity demand, which greatly suppresses the demand for new generation technology; basically everybody was competing to replace the natural gas, coal, and nuclear generation plants that were reaching their end of life.

Now, the hype of AI has made for a great excuse to build a bunch of new technology, which brings a flood of new investment money, political will at Pubic Utility Commissions, and end-users looking to sign PPAs to secure electricity which helps raise that money for new builds.

If you go based purely on when the technology was developed, sure, deep drilling is more difficult than nuclear fission, and in turn uses tons of new technologies at its base.

And until this new drilling technology was developed, fission was cheaper, but it won't be cheaper for long. Drilling is a technology with a learning curve which means that costs fall as we do more of it. Nuclear has not had a significant learning curve, and in fact in many cases it has gotten more expensive as construction labor gets more expensive over time.

"I would bet on geothermal over nuclear in a second for future electricity generation. Its so much more promising, has a tech curve, and has far more innovation and advanced tech adoption."

I'm glad that you are so bullish on a technology that still has very little to show for itself.

It should also be noted that the economics of geothermal installs also are the same ones as drilling for oil and gas -- so the cheaper drilling gets, the cheaper access to competitive resources.

Majority of electrical power - solar + nuclear is our best energy shot. Heating probably go with heat pump for 90% of the market.

That said - all of the above approach (including geothermal) - it shouldn't be this insipid argument of Geothermal vs Nuclear.

> It should also be noted that the economics of geothermal installs also are the same ones as drilling for oil and gas -- so the cheaper drilling gets, the cheaper access to competitive resources.

The economics of enhanced geothermal are vastly different in that geothermal has a learning curve where by merely drilling more wells, we make future geothermal cheaper by innovating technology. Nuclear reactors do not have a positive learning curve, and in fact are getting more expensive due to the ever increasing cost of construction labor (see Baumol's cost disease).

> insipid argument of Geothermal vs Nuclear.

First, expressing a preference for the future of one technology direction over the other is fundamental to discussing technology. It makes a huge difference in terms of getting better understanding of them, and is a core facet of any sort of planning. Sure, we should produce both, but we should also make predictions of what we think will happen in the future so that we can learn from the outcomes, and also do proper allocation of relative amounts of funding.

Nuclear has proven time and again to under deliver on all promises, results, and even ability to construct projects that have full regulatory approval. Geothermal is exactly the opposite. It has made realistic promises, overdelivered, and been fully transparent so far.

To not take into account these histories would itself be absolutely "insipid" and poor planning. That's not to say that nuclear should be fully excluded, but let's take into account the huge amount of risk with funding anything nuclear due to the inability of the industry to meet goals.

Any investment decision is a "vs." argument and must absolutely be looked at very very closely. And when we do that, every technology must be given a completely fair and honest assessment, we shouldn't put our thumb on the scale for nuclear just because we thought it wasn't given a fair shot in the past. We must look at it fully honestly without rose colored glasses, and I have yet to find somebody bullish on nuclear that doesn't wear extremely dark rose-colored glasses.

1. No it shouldn't be - it should be a look at Natural Gas/Coal generating facilities vs other technology.

2. How can you say Nuclear has underdelivered? Have you ever looked at the amount of nuclear deployment globally - does geothermal even get up to 0.1%? The cost of new Nuclear is a large function of the regulatory burdens especially in North America.

Don't read me wrong -- I'm all for geothermal but it certainly hasn't delivered on its promises and has only really worked out in uniquely favorable conditions (i.e. Iceland).

Reply on news.ycombinator.com