Hacker News Reader

Comment by Robotbeat | Hacker News Reader

There's nothing quite like trying to get fusion to work for a while to make you appreciate how awesome fission technology is.

If we applied even half the cleverness needed for fusion to making better fission technology, we'd probably be way better off.

...and then there's solar. Why even bother with producing the energy, just capture it with a very thin solid state device! Just need to automate the planting of solar panels in the desert, and we could produce all of our electricity from the Sun. Using just a fifth to a tenth of the land (and much crappier land that nothing much can grow on) that we use for /ethanol/ production alone.

(Yes, storage is tough, but is getting cheaper, and we can just plant more solar panels so there's enough power even during cloudy days... Although this is mostly a thought exercise. The best plan for deep decarbonization by far is to operate with a mix of clean power sources optimized for high capacity factor, including at least our current nuclear fleet... They help provide a reliable baseline which drastically reduces the amount of storage and over-installation required. That last 20% of power produced by nuclear is worth its weight in gold and should be protected at least until all fossil power production is ended.)

View on HN

How exactly do you propose to get that electricity from the desert to say Finland?

Furthermore, you need to plaster a vast area in the desert with solar plants, which brings along all sorts of concentrated security risks. The alternative is to have many smaller plants dispersed, which results in a big loss of efficiency.

I'm all for clean energy, but you seem to be handwaving away pretty huge issues as "thought exercise". Our inability to store energy is not just some side concern that will go away any time now. There is a reason carbon fuels are so prevalent - their energy density is far above what we can store in batteries. Without scalable and high density energy storage, unpredictable energy sources like solar or wind are much less useful because often there is no sun or wind at high demand and on the contrary there can be an overproduction during low demand. Germany already causes serious headaches for the European grid when it floods the network with electricity on sunny/windy days. They actually have to pay other countries to take over the extra capacity.

Solar, wind, tidal etc... energy are all awesome and clean (if we disregard the land that they take up), but there are many difficult issues that remain to be solved before they can become a core part of the energy production.

> Furthermore, you need to plaster a vast area in the desert with solar plants, which brings along all sorts of concentrated security risks

This sentence seems to be at odds with itself. How does needing a large area concentrate security risks? If anything it seems to disperse them.

Unlike a coal or nuclear plant, you can't just walk up to a critical part of a solar farm and blow it up. Any part that you damage is relatively isolated from the parts you don't damage. Thus, it takes considerably more effort to take the entire plant offline.

Sorry I worded that sentence poorly. Let's assume two scenarios:

a) We create a 200 square mile solar field in the Sahara, from which cables run to Europe to supply electricity. To take this plant out, one just cuts the cables. There, you just plunged Europe into darkness.

b) We create a 20x 10 square mile solar fields in the Sahara, all dispersed over a very large area. Well, this is an engineering and construction project on the scale that dwarfs even the previous project. It is truly vast. Nevermind all the political issues again.

I'm surprised to see this kind of simplistic reductionist thinking on HN.

"Just cut the cables" isn't trivial. Have you ever seen a high voltage tower? Not only are they not accessible without professional gear, it's also going to require even more expensive gear to cut multiple inch thick conductors.

In all likelihood he power to Europe will be transmitted via multiple undersea cables, the effort of disrupting which would be unavailable to all but state agents, in which case the security issue is irrelevant, as you are likely already in a state of war.

> the effort of disrupting which would be unavailable to all but state agents

It's a good thing then that there is a long history of all state agents in northern Africa maintaining friendly and stable relations with European governments.

(Yes, said Europeans also carry a great deal of responsibility for this situation, but that doesn't make the situation any more tractable).

Almost like the natural gas pipelines that come into Europe from our best buddy Vlad.

Wouldn't creating a partnership then help to stabilize that relationship?

Partnerships with whom? Several different warring tribes that want to gain control over the other or achieve independence? Morocco or Jordan maybe, the rest of them most likely not.

I'm not 100% positive that this is the right video, but I think it is. As part of the "overthrow a nation" plan, he talks specifically about destroying critical infrastructure, like electricity.

The solution for taking out high voltage powerlines? A pair of Skilsaws hanging under a quadcopter. Drop down onto it and... something bad's going to happen!

edit: I also missed the link: https://www.youtube.com/watch?v=m1lhGqNCZlA

edit 2: I scrolled around a bit and it's near the 30 minute mark.

Absolutely. The infrastructure is very fickle, but also enormous on the scale of countries.

Since we're talking about hypotheticals, I could say these things are also true:

1. There are counter-drones deployed and patrolling air space above the towers.

2. There are hundreds of independent power lines with 2N redundancy, so cutting 5-10 high voltage lines would do nothing.

3. The undersea cable points are protected by state agents and have good security for amateurs to compromise.

4. Europe will have 24-48 hours worth of storage even if ALL the lines are cut from the Solar Generation plants.

5. Europe will have backup power generation capacity in the form of locally stored and operated fossil fuel plants.

____

In other words, there's absolutely no evidence that the future hypothetical solar installation is going to be more vulnerable than the infrastructure that already exists today, and surprisingly Europe doesn't plunge into darkness every week.

I don't think the main problem here is random terrorists cutting power to Europe by blowing up a power line tower. The real problem is that nation states that control the Sahara haven't proven to be stable (Syria, Egypt), or don't always have great relationships with Europe (Algeria in the 90s).

As to points 4 and 5, although storage and some backup capacity is reasonably feasible, this would mean that Europe could very well be forced to go to war in short order to get their full power supply back if a rogue organization gained power in North Africa.

It might motivate Europe to improve relations with the region

We'll be better off building nuclear reactors, thank you. The social and political challenges are much lower and so are the risks.

(I obviously meant Libya, not Syria)

Burry the cables. Won't protect you from everything, but makes it a bit harder to compromise them. You can even have "cages" around them that alarm a nearby response team of any manipulation. Of course, this adds siginifcant cost, but then you have to take the total economical cost into account if an attack is successful.

Damage that takes a few hours to fix is not going to destabilize a country. If you can scale it to keep up 24/7 disruption you already have an army and far more useful approaches like destroying ports.

> "Just cut the cables" isn't trivial.

Well, how many towers do you have to knock down using explosives before the lines fail?

Just 1.

The feet of the towers are simply accessible, so there's nothing stopping people from even just using a metal file. It wouldn't be safe, but you'd get the job done.

Or just plow a truck into their feet. You only really need to bend one of the feet and the wind will do the rest.

> I'm surprised to see this kind of simplistic reductionist thinking on HN.

The thinking is in fact correct. The parent should have used a better, more serious, and guaranteed inevitable scenario: war. Bomb the concentrated infrastructure and take down all of Europe's power supply. It wouldn't take very high precision and it wouldn't take a vast number of hits: it would require a few, clumsy hits, which would be nearly impossible to stop.

The present system, which tends to be far more distributed than not (to take down all the power in Europe today, you need to hit a lot of spread out targets), is vastly superior to extreme concentration.

Why cut it? Explosives would be much easier (think IEDs), and could be improvised from simple sources like gasoline and petroleum jelly. No need for a tool to "cut" a power line, regardless of it being above or below ground.

I also think the location thing may be addressed in later solar technology. i.e. For our currently tech the sun really needs to be out. But as far as I remember, there's still a good amount of energy available on cloudy days (assuming it's not pitch black). As solar technology progresses so too will the locations it can be installed increase.

> it's also going to require even more expensive gear to cut multiple inch thick conductors.

I agree with most of what you've said, but cutting a few inch thick conductors doesn't seem all that hard. I think a single stick of dynamite would do it pretty easily. Two for good measure. This would still be an act of war of course.

The undersea cables don't need to be cut, they're perfectly capable of breaking on their own.[0]

[0] https://en.wikipedia.org/wiki/Basslink#2015-2016_outage

Then "evil state" bombs cables, or power plant. The point is that the cost of defending such a crucial part of our infrastructure would not be trivial, and it's still hosted on foreign soil, so... politics.

It has been done in Crimeea last year. Pro gear to cut HV cables = C4. How do you run an undersea cable through the desert?

Or a single car bomb?

Renewable power is very much going to be a distributed more than our current power generation. Further, you want redundancy in power transmission so you can do maintenance anyway and many countries are already dependent on foreign sources for a significant chunk of there daily energy demand for economic reasons. Any actor that can destroy thousands of cables can cut the power systems we have now.

Further, outside of city's you can generally get solar panels from a roof to provide power for the building under it making the grid a lot more redundant vs cutting cables.

Exactly, while Sahara is obviously sunny and the land is cheap, it might not be the best place to place solar plants, due to the logistics.

I think that roof top solar is just as viable in more or less the whole Europe - even in the Nordics, at least the southern parts.

The only thing that would make desert-sized solar plants a really good business would be really cheap low efficiency cells.

> ...one just cuts the cables

Isn't this a problem with any centralised electrical power source? In fact I've always thought that terrorists could cause far more disruption by taking out a couple of transmission line towers than blowing people up. And it would only take a decent sized spanner in many cases.

The only solution to this (and to the massive resistive losses over power lines) is localising power production. And that's where solar, wind, small-scale hydro and local battery storage look very effective.

Seems a lot easier to fix a cable cut than a sabotaged nuclear power plant. I mean if you are the point where a nation-state is detonating fuel-air explosives over solar farms then there's probably no power plant design that is going to survive.

"just cut the cables"

Not that easy. Furthermore, it's the same for any large power plant of any type. This is a very poor argument against solar. It's also a stretch to imagine that Europe would rely on a single power plant for its sustenance.

The commenter is literally suggesting that it rely on a giant solar deployment.

You can't just walk up to coal or nuclear plant and blow it up either.

If you wanted to take a power plant offline, solar included, you would target the grid connection wiring or the high volatage transformers.

High latitudes (above 50 degrees) have a big problem with solar power. Luckily, almost the only people who live that far from the equator are Northern Europeans, who tend to have access to hydro and nuclear. (this is why it's so stupid for Germany to shut down their nuclear power... they're replacing it with coal, brown coal, and gas!)

The vast, VAST majority of the world's population lives FAR closer to the equator.

Again, you seem to have ignored the part of my post where I mentioned installing enough solar to produce power even during cloudy days. Cost of utility solar is super cheap now. If it continues reducing in cost, it'll be plenty cheap enough to just install 3-5x as many panels as you need to provide power during cloudy days. Yes, solar CAN get cheap enough for that to be a good idea.

I haven't ignored the part where you mentioned overprovisioning. Where do you propose to put the excess power during sunny days? Stable baseline sources like nuclear cannot just be shut down or spun up on a moment's notice to accomodate a vast number of solar energy entering or leaving the grid. At the same time, we don't have a capability to store electricity at sufficient scale so we do need the stable baseline to cover a high enough percentage of our production.

The fact of the matter is, as long as we don't have a scalable storage solution, power sources where we can't regulate the power output cannot take up base load and can at best be a small overlay that can pick up load if the beneficial weather happens to coincide with higher load requirements.

This really is not a big technical problem. You simply don't output the power. This is trivial with any kind of decent power-optimizing inverter setup commonly available today. Solar PV isn't spinning a turbine that needs to find a load. If you disconnect a solar panel, it doesn't damage itself, it just sits there.

But I'm sure we'd eventually find uses for this extra energy. Like producing ammonia for agriculture.

No one I know is opposed to renewable energy, but advocates really do everybody a disservice when they try to argue that an intermittent power source without storage is a reasonable replacement for base load power. As Bill Gates said in an interview "…They have this statement that the cost of solar photovoltaic is the same as hydrocarbon’s. And that’s one of those misleadingly meaningless statements. What they mean is that at noon in Arizona, the cost of that kilowatt-hour is the same as a hydrocarbon kilowatt-hour. But it doesn’t come at night, it doesn’t come after the sun hasn’t shone, so the fact that in that one moment you reach parity, so what? The reading public, when they see things like that, they underestimate how hard this thing is. So false solutions like divestment or “Oh, it’s easy to do” hurt our ability to fix the problems. Distinguishing a real solution from a false solution is actually very complicated."

https://www.theatlantic.com/magazine/archive/2015/11/we-need...

(Gates is investing in 4th gen nuclear and energy storage companies so he is putting his money where his mouth is.)

The term Baseload power is not a goal, it's a Problem with some types of power not being flexible. In that context, soar is closer to Baseload power than peaking power plants. Yes, it currently better matches demand than Baseload power, but as you ramp it up the supply / demand miss mach will simply shift around the day.

Turning on some peaking power plants at night vs the day is not a significant issue. Further, the solution is excess capacity not long term storage.

You are handwaving away huge issues here. As Bill Gates said in the same interview I quoted from before:

>"...It’s kind of ironic: Germany, by installing so much rooftop solar, has it that both their coal plants and their rooftop solar are available in the summer, and the price of power during the day actually goes negative—they pay people to take it. Then at night the only source is the coal, and because the energy companies have to recover their capital costs, they either raise the price because they’re not getting any return for the day, or they slowly go bankrupt. There are many people working on storage—batteries are a form of storage, and there’s a few others, like compressed air, hot metals. But it’s not at all clear that we will get grid-scale economic storage. We’re more than a factor of 10 away from the economics to get that."

Bill Gates is very much part of the problem.

Having fossil fuel plants that only operate when needed, but get paid more per kWh than they do now when they run all the time (since as well as fuel costs they have up front construction to cover) is not even close to being a problem. It is in fact part of the solution. What is he doing complaining about this?

Negative prices are also not the big drama he makes them out to be. Sad that he's got a reputation as some kind of genius and he's scared of negative numbers as if it's some kind of witchcraft to have too much of something and use pricing and markets to decide which providers should scale back.

Translating for those scared of negative prices: someone wants to keep providing power for some reason (maybe their setup doesn't like rapid changes in output, which would cost them money in repairs), they want to pay you $X to shut down your system instead. Do a calculation and figure out if shutting down your system is cheaper than that, if so, take the money. Market forces at work! The inflexible system don't make as much money, and the flexible systems gain it.

Or, someone with flexibility in when they need electricity says "I'll agree to heat my water tanks at the point that helps the grid most, how much will you pay me for doing that? The amount we'll pay you is the negative price, except not negative anymore".

Coal and nuclear are huge steam engines. They cool down when you turn them off which causes thermal stress and you then need to heat them up again before they can generate power. Thus Germany's problem is they have coal power plants in the first place.

Gas turbines also have thermal stress, but they are designed for this and they loose vastly less power when you turn them off. With enough of them you can have zero storage and zero problems. Because storage is simply one solution to production / demand mismatch not the only one.

Now yes, peaking power costs more but energy costing more at night is rational behavior and without subsidies you will minimize overall costs.

On the plus side, the CO2 from burning natural gas is less than coal and it doesn't emit mercury or the other particulates and doesn't have a waste issue like coal does. Unfortunately for those who care about climate change, (and if you don't, you should) there are inevitable methane releases from fracking and from distribution of natural gas and those are now known to be much worse than previously thought:

>...Back in August, a NOAA-led study measured a stunning 6% to 12% methane leakage over one of the country’s largest gas fields — which would gut the climate benefits of switching from coal to gas. We’ve known for a long time that methane is a far more potent greenhouse gas than carbon dioxide (CO2), which is released when any hydrocarbon, like natural gas, is burned. But the IPCC’s latest report, released Monday (big PDF here), reports that methane is 34 times stronger a heat-trapping gas than CO2 over a 100-year time scale, so its global-warming potential (GWP) is 34. That is a nearly 40% increase from the IPCC’s previous estimate of 25. ...The IPCC reports that, over a 20-year time frame, methane has a global warming potential of 86 compared to CO2, up from its previous estimate of 72. Given that we are approaching real, irreversible tipping points in the climate system, climate studies should, at the very least, include analyses that use this 20-year time horizon. Finally, it bears repeating that natural gas from even the best fracked wells is still a climate-destroying fossil fuel. If we are to avoid catastrophic warming, our natural gas consumption has to peak sometime in the next 10 to 15 years, according to studies by both the Center for American Progress and the Union of Concerned Scientists.

https://thinkprogress.org/more-bad-news-for-fracking-ipcc-wa...

As we use more and more natural gas, we can expect more and more methane disasters like the leak from Aliso Canyon in CA which was the largest leak in US history. This released over 100,000 tons of methane into the atmosphere and required 11,000 residents to be evacuated.

http://www.bbc.com/news/science-environment-35659947

100,000 tons of methane with a half life of less than 10 years. The math is complex and 1 ton of methane becomes more than one ton of CO2, but the actual long term impact is fairly minimal vs the savings from not using Coal. Remember that's ~100,000 tons / ~10,000,000,000 tons = 1/100,000 so yea methane is bad, but in this context it's 0.001% so it's still a rounding error even if it's 100 times that bad.

Anyway, I agree it's not without costs. But, if we got to the point where peaking power from gas makes up ~10% of total electricity supply and the rest of that is renewable power then we will have made massive progress. In other words don't let perfection stand in the way of progress.

>100,000 tons of methane with a half life of less than 10 years. The math is complex and 1 ton of methane becomes more than one ton of CO2,

You are downplaying the impact of methane. As I quoted in the last post, the IPCC estimate is that over 100 years a given amount of methane is 34 times more potent as a heat-trapping gas compared to CO2. Over a 20 year period (which is probably more relevant considering where we are in climate change) methane is 86 times more potent.

>...but the actual long term impact is fairly minimal vs the savings from not using Coal.

That's the problem - there is enough methane being released by our new reliance on natural gas there is concern that we may be having just as much of an impact on climate change as if we went and burned coal.

>...Remember that's ~100,000 tons / ~10,000,000,000 tons = 1/100,000 so yea methane is bad, but in this context it's 0.001% so it's still a rounding error even if it's 100 times that bad.

The problem is that EVERY natural gas well, storage facility and pipeline leaks methane. This was just the largest one. As I quoted in the last post, "...Back in August, a NOAA-led study measured a stunning 6% to 12% methane leakage over one of the country’s largest gas fields — which would gut the climate benefits of switching from coal to gas."

>... But, if we got to the point where peaking power from gas makes up ~10% of total electricity supply and the rest of that is renewable power then we will have made massive progress.

Yea no kidding - 90% power from intermittent sources would be pretty remarkable. The problem is that 90% power from intermittent sources won't be possible unless there are major advances in grid storage technologies.

In terms of having to use natural gas, nuclear power plants in France and in Germany operate in load-following mode, so there is no reason this can't be done elsewhere. Minimizing our use of natural gas when when you consider the climate changing negative externalities of natural gas, would be a good thing.

That's just lying with numbers. 86 / 5 = 17.5 so average from year 20 to year 100 is 34 - 17.5 = 16.5x. And again the majority of that 16.5 is over years 20-40.

Further, because of this decay increases in release rate don't stack at 86x year over year. You get a magnified increase in the first year and a smaller net increase every year after that until your down to the carbon content of methane.

Under 10% total gen from peaking power is viable with over production of wind and solar and zero grid storage. It's not going to be 10% 24/7 but 0 most of the day, regular generation at part of the day, and rare spikes up to 50%. You can drop that under 5% with moderate grid storage of around 1h average demand, but you start needing vastly more storage the lower you want to take that number.

>That's just lying with numbers.

No, those numbers are from the IPCC which represents the mainstream consensus. Take it up with them if you think you are right and they are wrong. Using the 20 years number is reasonable considering how far along we are in climate change. As the IPCC report states: "There is no scientific argument for selecting 100 years compared with other choices (Fuglestvedt et al., 2003; Shine, 2009). The choice of time horizon is a value judgement since it depends on the relative weight assigned to effects at different times."

>...Under 10% total gen from peaking power is viable with over production of wind and solar and zero grid storage.

Yea I hear advocates make such claims all the time. Unfortunately I never hear the same thing from people who actually run a major utility. We might get there someday, but for now I have to agree with Gates: "...What they mean is that at noon in Arizona, the cost of that kilowatt-hour is the same as a hydrocarbon kilowatt-hour. But it doesn’t come at night, it doesn’t come after the sun hasn’t shone, so the fact that in that one moment you reach parity, so what? The reading public, when they see things like that, they underestimate how hard this thing is. So false solutions like divestment or “Oh, it’s easy to do” hurt our ability to fix the problems. Distinguishing a real solution from a false solution is actually very complicated."

Ed: Make that 5/4 * 16.5x or 20.6 However, again it becomes more than 1 CO2 by weight because the 2 O's in CO2 make up a larger fraction of total weight than 4 H's.

Nobody ever mentions varying the price of electricity to reduce demand when the sun isn't shining. We need to get past the assumption that electricity prices should be fixed.

This won't solve the baseload problem, but it will reduce it.

Electric prices have never been fixed at scale. It's really only small scale consumers that get a fixed but often higher rate. See Enron for example. Changing when energy is cheap does not break the model.

My electric rates have been fixed rates 24/7 for decades.

Are you using KW, MW, or GW of power?

So yea, residential customers get a fixed rate but that's because there is no point in demand based pricing. They simply charge you enough that it makes little difference what time of day you use slightly more power. In the future the same is likely to be true.

My own usage is inconsequential, but not the aggregate of all the consumers.

You can use the excess to create fuel, such as cracking H out of H2O, and then use that to reduce the nighttime problem. I don't think it'll be hard to find a use for excess power.

Desalination of seawater seems a safe bet for excess solar power.

That's a horrible solution. Water is even less portable than energy, even the driest desert can't afford to import water. You're trading a highly liquid but hard to transport commodity for a highly illiquid and hard to transport commodity.

Scenario 1: Your solar power is used near the site of the power installation. In this case, there must be people living near the site in order to use all that power. Those people require freshwater. Therefore, there will be a means to transport water to the site.

Scenario 2: Your solar power is used elsewhere. In this case, there must be a means to transmit power to the people who will use it. Those people are likely to live near seawater. The closest such people are also likely to have a limited supply of freshwater.

You are already either transporting water to the solar site, or transporting power away from it to somewhere else. So the idle capacity of the solar plant will be usable for desalination and/or wastewater reclamation. And I think aqueducts, particularly seawater aqueducts, are cheaper than you realize. The majority of energy required is in increasing the elevation of the water, and that can be done with surplus power, too.

> Water is [...] highly illiquid

Do you mean ice or steam? ;-)

Ah! I was going to write water pump reservoirs or at some price point electrolysis + fuel cells, but making cheap water in the daytime is also good.

We already have a solution to all our energy needs. Funny to see people come up with outlandish suggestions. I am not sure what I am missing. Because logic points to only one direction.

I don't think there is a viable solution for buffering solar or wind (or nuclear) with renewable sources, unless you happen to have access to a lot of hydro power.

Dynamic remote cut-off of heaters, chargers, and other non-critical consumers is one suggestion, but the equipment cost is probably at least 100 Eur per consumer.

Well... Outlandish is to think that replacing a huge amount of infrastructure of a type to another is as simple as your sentence makes it look like.

By no measure is reverse osmosis or flash distillation of water cheap even if the energy is free.

Imagine building a huge water desalination plant and only running it when there is excess power available on the grid.

I can't see anyone wanting to invest in that, except maybe a financially myopic government, and they tend not to last very long.

If the energy is free, you could just do electrolysis with huge carbon electrodes and immediately ignite the gases, bubbling the exhaust through your freshwater reservoir. You can knock together a setup like that in your own backyard from junkyard parts.

How you do it depends on whether you prioritize capital costs or operating costs.

You can do things with the excess power that otherwise would be un-economical to do during a cloudy day. Things like smelting aluminum, desalinating water, etc take massive amounts of power.

You can't just do those things when the weather is good.

Not without huge tax breaks to incentivise load following, because intermittent production of large scale industrial processes is hard and not cheap.

They already do this without any gov intervention. The local steel recycling plant only operates when power is cheap. While they do need some advance notice to make sure they have enough people to run, they probably can get a good enough idea from a weather forecast (power here is mostly hydro with a dollop of wind and fossil).

Some processes are better fits than others. If you have 2-5x extra capacity that is sold cheap on sunny days, someone will figure out how to use it.

Do we have any numbers on what it means that they run only when the power is cheap?

Does that mean they can operate in 4hr windows? Or do they run have to run the furnace for a minimum period? A large quantity of steel takes a large amount of energy to melt.

Interesting question. Brief research says the cycle time is ~1 hour (or less):

http://www.steel.org/making-steel/how-its-made/processes/pro...

It could also be a plant for shredding scrap, where I guess it wouldn't be a time issue so much (and if it doesn't have the material inflows to run full time, timing cheap power would make sense).

Thanks for the link, looks pretty detailed. I work in steel fabrication, so it'll make interesting reading for me.

> Where do you propose to put the excess power during sunny days

Heat up some molten salts or any phase change material, then run the steam through a generator. Or pump water uphill. Not very efficient but doable.

> How exactly do you propose to get that electricity from the desert to say Finland?

You don't. No one said it has to be all solar everywhere.

> there are many difficult issues that remain to be solved before they can become a core part of the energy production.

Wind and solar are already core sources in some places.

In many places, a decent solution to the night-time solar problem is natural gas power stations. We built them already.

Finland will use Finland-appropriate solutions. That has very little to do with the appropriateness of solar in Mumbai or Phoenix.

Are even thinking about pollution from storage technologies.

Finland net imports of electricity in 2009 were 15%, so about the same as it is today. They have a lot of hydro power and little reason to stop using it. Countries are ok with risking 15% of their generation from external sources because you can get by on 85% supply with minimal disruptions and you can ramp up supply in the short term.

Solar in the Sahara is an example, but countries already spend a few percent of GDP on safety, being less efficient but more independent is generally an excepted trade-off. Further, Sahara solar > desalination > farming is probably a better use of that space vs transmitting it.

> How exactly do you propose to get that electricity from the desert to say Finland?

High voltage DC transmission works fine for this problem. All of the engineering challenges have been credibly worked out and priced (https://en.wikipedia.org/wiki/Desertec). The main problems with this approach are political, in that it requires massive cooperation between many countries over very long time scales (decades).

Why is there a security risk with putting solar in the desert? It seems like a relatively easy to secure location, and solar cells are hardly high value-density targets for theft or vandalism.

So you think northern African nations are going to let European nations station troops there and take over a big chunk of their land? That doesn't seem realistic, and seems rather imperialistic actually.

Letting the northern African nations provide the security is just plain folly, and also makes Europe completely dependent on these economically backwards and frequently hostile nations.

Why wouldn't an African nation want to charge rent to some European country for vast amounts of land that are otherwise literally worthless[1]? It's not like they're strangers to private company security forces that are basically just extrajudicial mercenaries. History says any company that wants to do this just needs to apply some bribes to the appropriate officials.

[1] In the sense that nobody is willing to buy it so it has no price.

Yeah, I'm sure it'll look just great that Germany or whoever is stationing the Wehrmacht in northern Africa, and no one will complain about that... /s

You really think that's a viable way to maintain the electric power supply for all of Europe? "What could possibly go wrong?!"

It won't be uniformed European troops. It will be Blackwater type mercenaries. They'll have an implicit license to kill and/or abuse any locals. That's how you do security for European companies in Africa.

You really think you should put the entire welfare and existence of the European continent in the favor of some hired Blackwater mercenaries? Are you f'in serious?

Yes, destabilized countries always act rationally on the global market. /s

The danger is political risk, not necessarily theft or vandalism. Any dependency on territory not under your own societal control can be used to hold you hostage and cannot be depended upon.

Forget the solar farm, easiest is to cut power lines. (explosives?)

Europe imports around 30% of its oil and gas from Russia. The pipelines between the east and west were originally setup during Soviet times.

Given how stable that has been it seems unlikely that a small African country, who will be getting a big chunk of cash from Europe, would do anything stupid.

They haven't been perfectly stable, though. We've seen since 2005 a few cases of oil and gas prices surging upwards as part of the Russia-Ukraine dispute (in 2005, 80% of all gas from Russia to the EU went through Ukraine).

Europe is currently highly dependent on Russia's natural gas for their electricity; I don't see this as worse. Indeed, better to diversify your power imports.

(Better still to generate locally, but not every country has the land available to do so).

https://en.wikipedia.org/wiki/Russia_in_the_European_energy_...

So we resolve all regional conflicts and unify the world under one country. It's not hard.

Not sure why you're being downvoted. The Chinese are actively researching in the area of ultra high voltage DC transmission so that power can be distributed long distances from remote renewable resources. Even if the problem were as simple as a direct link from the Sahara to Finland, China has already built UHVDC transmission lines half that length. UHVDC technology is only improving with time and it's an exciting field to watch.

The economist had a reasonable article, though light on technical detail: http://www.economist.com/news/science-and-technology/2171432...

> Furthermore, you need to plaster a vast area in the desert with solar plants, which brings along all sorts of concentrated security risks.

I mean, most of the same security risks appear in oil pipelines and extraction.

And the vast area is still very small compared to agriculture. And it can be done in chunks. The electrical grid is concentrated by its very nature, but we still manage just fine.

(Solar is also one of the few truly distributed power sources, but I expect most power in the future will come from vast solar farms due to the ease of automation.)

True, and this is a issue today as well. Which is why I think that proposing a solution that compounds this issue even further should not be taken lightly.

Edit: put differently, why should we adopt a new solution that is even more exposed to the already known risks?

A large solar field makes for a pretty diffuse target. You might be able to target the connection to the grid, but that's no worse than it is today (and would be easily repaired).

And I don't think our current electrical grid has a major problem that even a slight attention to infrastructure couldn't fix.

Ok so you are saying that the new solution isn't any worse in terms of security than the current one, so it's fine?

In any case, you cannot compare this to oil pipelines directly because we can stockpile oil. Even if all the pipelines are taken out, many countries stockpile oil that can bridge them over until an alternative solution is found. If a considerable part of the electricity comes from some remote site in a desert and the connection is taken out, that'll immediately plunge the country into darkness.

I like your optimism about all the problems being trivial, but with your mindset every problem we currently face is trivial.

At this point, why don't you just propose that we go and harvest Helium-3 from Jupiter? I mean, we already sent some probes that far, it's just some engineering issues, right?

There are a whole host of political, engineering and social issues that need to be sorted before something like you propose can be done. It can work for individual countries in certain regions, but it is not in a state where it can fix the energy problems of the world or liberate us of the carbon dependence.

"Ok so you are saying that the new solution isn't any worse in terms of security than the current one, so it's fine?"

Yes. You're blowing up the security argument when really it isn't the security of it that is the problem -- our current infrastructure isn't more secure, and we have had only minimal problems with it thus far. The problem is the imminent and looming destruction of various thousands of ecosystems, and the continuing habitability of the planet.

It could be done. We got to the moon in about twenty years, which was much more of a technological leap than this will be. With the concentrated effort and enough propaganda, we could make the switch from coal/gas/oil. We are very rapidly running out of runway.

I'm the classic US-centric American. Northern Europe can solve its own problems, they have all the tools (nuclear, hydro, wind, geothermal). The rest of the world (i.e. the vast majority of people) is closer to the equator and doesn't need a cable running to the Sahara.

About a decade ago, there was a book written which looked at the ability of Great Britain to become self-sufficient entirely in renewable energy (https://www.withouthotair.com/ looks to be the URL).

It concluded that there is literally not enough renewable energy potential in Great Britain (e.g., cover every square millimeter of space with solar panels) to do so at then-current energy consumption rates. It also pointed out problems with approaches like "giant solar fields in the Sahara" (the requisite solar fields are, well, roughly the size of Germany in terms of required area). Nuclear is technically not a renewable energy resource, but it does have requisite power generation capabilities--and very staunch anti-nuclear activism that makes expanding nuclear difficult.

All in all, quite an eye-opening read.

"It concluded that there is literally not enough renewable energy potential in Great Britain (e.g., cover every square millimeter of space with solar panels) to do so at then-current energy consumption rates."

That is false, though. Just a little math shows why.

The UK is 242,500 km². Nameplate solar capacity on that would be on the order of 150MW/km², so 36 Terawatts. UK average electrical consumption is 32 /Giga/watts. Assume solar produces 1/4 of that 36TW on average when it's sunny during equinox, so 9TW. Even on a pretty cloudy day on the winter solstace in London, you're still going to produce at least 1/10th to 1/20th of that much, so worst case of worst cases (cloudy day on solstice) you're looking at 450GW average if you cover the entire UK in solar panels. That's still over ten times the UK's average electrical consumption.

(I used UK instead of Great Britain, but that doesn't make much difference.)

I mean, the overall point that you wouldn't want to power the UK on solar alone is totally and completely valid (and I fully support the Hinkley Point C nuclear power plant, by the way), but the idea that it's literally not possible is BS.

Even if you include TOTAL energy instead of electrical energy (in the US, electricity is responsible for 40% of energy use, so I'll use that as a guesstimate), and assume 100MW/km^2 instead of 150 (even though I already accounted for that in the cloudy winter day part), you're still producing about 3.7 times as much energy on that cloudy winter day than the total UK energy use. And that's before we use higher efficiency solar panels or deep sea hydrogen seasonal storage (the latter of which is one of the very few sensible seasonal energy storage methods).

It's an absurd claim to say it's impossible, and there's almost nothing more annoying than a source claiming to be "without the hot air" actually being full of plenty of hot air on absolutist claims like this. You don't correct BS by using equal and opposite BS.

(Most energy advocates I know do this... both renewable advocates and nuclear advocates, exchanging BS about the opposite technology. It's a clean energy circular firing squad, with the climate in the middle.)

Energy consumption here does not mean just current electricity usage. It also includes things like fueling cars and planes. Also, the 150MW/km² is apparently roughly the summer expected solar insolation; year-round average insolation comes out to ~100MW/km² per the book's numbers.

The total chart at the end came out to <https://www.withouthotair.com/c18/page_103.shtml>, with solar making up about 68 kWh/d per person.

And fortunately, we've got those covered.

I'm not the OP, but here's a few thoughts...

Highly distributed rotating masses would handle a large part of the storage, turning it into more of a power smoothing problem.

Secondly, I presume that at some point we'll be covering every building (rooftop and window) with some sort of solar film that converts electricity.

Why don't we have those today in large scale production? Probably for the same reason that the US builds more roads instead of mass transit - it's a bias to build what we've alway built, not a real technological limitation.

There's definitely a link between where it's economical to place a panel and panel efficiency. As the technology matures it will make sense to place it more places. I think at some point, ultimately, it will be in the paint of cars/buildings.

Exactly. Just like Moore's law drove the price of computing power to put it practically everywhere, there is a cost curve for solar that will do the same.

Solar PV is great distributed power source to install on roofs and maybe that's it. You can combine it with partial storage for household consumption, but even that is a burden.

>How exactly do you propose to get that electricity from the desert to say Finland?

Microwaves

http://www.economist.com/node/18864324

i read an article that the nordics actually gets as much sun as southern europe, if you count the whole year. Its as good a place for solar as anywhere else.

The summers are sunny all day and night. The problem becomes how to store the energy until the dark winter months.

(There are solar panels that doesnt need heat just light)

Edit: http://www.nordicenergy.org/article/solar-power-at-the-arcti...

The seasonal variation of solar power is a big problem if you want to rely mostly on solar power, though. Seasonal storage is insanely expensive with any of the usual methods. I mean, let's say to get daily storage, you need 8 hours of batteries. To get 4 or 5 weeks of storage, you need more like 800 hours of batteries. It's literally 100x the cost.

So yeah, if you want to run on natural gas for half the year, then solar in the nordics isn't bad.

But luckily the nordics have lots of hydro and don't need any gas or even solar.

But on the equator, the seasonal variation is small. The Global South will have a big energy cost advantage as solar costs keep dropping.

https://en.wikipedia.org/wiki/Hydrogen_fuel

Spoken like someone who has never visited a desert.

As a Southern California native, I am intimately familiar with the Mojave, Anza Borrego, and Sonora deserts. I will tell you that these deserts are far from "crappy land that nothing can grow on" as you put it.

The North American deserts are areas of huge ecological and geological diversity. They are also some of the most beautiful and awe-inspiring landscapes that exist on this planet.

We do utilize a lot of area in the Western Mojave for solar and wind power. For instance, there is a HUGE field of wind turbines in the San Gorgonio pass near Palm Springs at the intersection of the 10 and the 62.

That said, I am no way in favor of scarring any more of our deserts by covering "a fifth to a tenth" of them in solar panels. However, I am not against concentrated solar facilities several square miles in area as long as they remain contained.

I have thought about how we can "disperse" a solar panel network into a system of smaller panels and batteries and the best idea I can come up with is requiring all commercial buildings to cover their roofs in solar panels and maintain on-site battery banks. Also requiring all homes and apartment buildings to have at least one solar panel per household. These ideas come with their own trade-offs although.

I said a fifth or a tenth of the land in the US used for ethanol production. NOT a fifth or a tenth of our deserts. More like 1%.

And my point is we can install solar on land that has no other use. Like land contaminated by industrial waste, etc.

1% of all deserts is only correct if you are talking about only today's electricity use. Even moderate growth and actually solving energy, not just the electricity subset drives these percentages much higher. To get US centric again, NREL published an optimistic 2016 report suggesting that the maximum rooftop solar capability could ever meet is 40% of electricity. That's 15% of today's energy use in an electricity heavy country.

>1% of all deserts is only correct if you are talking about only today's electricity use. Even moderate growth and actually solving energy, not just the electricity subset drives these percentages much higher.

U.S. electricity usage has been flat for the past 10 years[0] despite a cumulative GDP growth of 34% over the same period. I'm not saying the same trend will continue to hold, but you are overestimating future electricity usage growth.

[0] https://www.statista.com/statistics/201794/us-electricity-co...

Ah. My misunderstanding.

  That said, I am no way in favor of scarring any more of 
  our deserts by covering "a fifth to a tenth" of them in 
  solar panels.

Tough luck. We live in a democracy. When we run out of oil, the number of people who want electricity are going to outnumber the number of people who want to look at dirt.

Thanks for writing this. I am not sure from where this solar lobby sprung up suddenly. If people are so eco conscious, that they support solar. I am not sure how they are ok with have huge parcels of land be covered with panels. We are not even talking about pollution from storage technologies.

Turns out those deserts may not be so useless after all I'm afraid, for example dust blown from the Sahara fertilizes the Amazon: https://www.nasa.gov/content/goddard/nasa-satellite-reveals-...

Solar has become more efficient and economic in recent years but remains only a small part of the energy puzzle for various reasons.

Lots of people are trying to apply that cleverness to fission. The problem is that governments make life much harder for fission projects than for fusion projects. There's some justification for that; fission done well is great, but fission done badly can cause much more serious accidents than fusion done badly, and proliferation is much less of a concern with fusion.

Even so, the regulation is more of a problem than it needs to be. The NRC currently requires a near-complete design up front, costing several hundred million dollars, before they even hint at whether they'll approve your project. If they turn you down, you're done. If they approve, you've still just got paper; it's only then that you get to start experiments with nuclear material. If they just had a more incremental process it'd be a big help.

Meanwhile, people build deuterium fusion reactors in their basements and nobody bats an eye.

Re solar: not just deserts (there aren't many in the UK) - https://www.fastcompany.com/3053998/these-sky-high-balloons-...

Yes, I was being US-centric. I know most about the energy situation in the US, and I happen to live here. :)

Thinking deeper into the future, floating solar farms could also be a thing. The oceans are huge, and most nations are not land-locked. They'd have their own engineering challenges, for sure, but those challenges look like patty-cake compared to fusion power.

I have to think there are major ecological concerns with blocking sunlight over large areas. They may not be negative impacts, but surely it isn't a no-impact situation.

There is always some impact.

There's a disease in our energy discussions where if something has ANY effect or can be measured at all, it is considered a big no-no. We lack a sense of proportion. Radiation is SUPER easy to measure even tiny amounts, for instance, since you can look for the signature of a certain decay tree, so people hear that you can measure radiation from Fukushima and lose their minds. Likewise, people make a big deal about solar using up a lot of land, but then forget that agriculture uses up on the order of 100 TIMES the land that solar would take up if we relied entirely on solar for electricity. The sense of scale is entirely lacking.

"The sense of scale is entirely lacking."

Agree and it extends to risks. Better to kill 100K people per year with coal, than possibly maybe hundreds might die (but almost certainly zero actually will) in a nuclear accident. Your diet and your car are far more likely to kill you than global warming.

> Your diet and your car are far more likely to kill you than global warming.

True for me. True for my (entirely hypothetical and never likely to exist) grandchildren? Less clear.

An example of this which always gets me with nuclear power is the waste problem. The fact that the waste remains dangerous for tens of thousands of years is always held up as a huge problem, completely ignoring the vast quantities of extremely poisonous chemical waste produced by other industries which lasts forever.

"This waste is dangerous." "OK, guess we'd better be careful with it."

"This waste is dangerous for the next 10,000 years." "Holy shit! We need to build a high-tech underground fortress to contain it!"

> "This waste is dangerous for the next 10,000 years." "Holy shit! We need to build a high-tech underground fortress to contain it!"

Time is a major risk factor. Look where we were 10,000 years ago. There is relevant chance that 10,000 years in the future no one will easily understand the danger of radiation anymore, so we have to prepare for that possibility.

You seem to have missed the other half of the point I was making, which was that there's lots of non-nuclear hazardous waste which lasts forever and we don't get nearly as upset about those.

Energy storage is easier than fission or fusion. We've just been working on the other two problems for ten times as long. Grid energy storage became financially viable maybe five years ago and there are already startups in compressed air and sodium-ion batteries (which can be potentially much larger/cheaper than lithium-ion at the cost of being less space-efficient). Nuclear energy has been financially viable forever and it took an existential crisis for the richest nation in the world to devote all of its resources to it for years to make it happen.

Energy storage for a few hours is easier than fission. Seasonal energy storage, on the other hand, is much more difficult.

This is why I think a mix of clean power sources (including some nuclear) is going to be optimal.

I am of the firm belief that there are MULTIPLE feasible ways to solve the clean energy problem that we, as a society, could probably do if we put as much effort into it as we did a world war. Examples:

<>nuclear France (ultimately using breeders of uranium or thorium)

<>hydro/geothermal Iceland

<>over-built intermittent renewables buffered with grid storage and perhaps either a little hydrogen seasonal storage or a mega-grid connecting north to south

even exotic solutions like

<>inertial or magnetic confinement fusion

<>space-based solar power (which is easier than fusion)

But the optimal (both in time and money) solution is to use a mix of clean energy sources (solar, wind, hydro, geothermal, nuclear) and some storage and some demand-response (smart meters are a pretty obvious solution, here) and some increased geographical grid interconnection (UHVDC power lines).

Really the only reasonable world wide solar plan I've ever seen is solar power satellites. There has been a lot of work done there and they beam the energy back to the planet as microwaves where it is needed into the existing grid system. But so far the cost of getting all of that material into orbit to build them is something of a deal breaker.

https://dothemath.ucsd.edu/2012/03/space-based-solar-power/

>At this level, our 3.6 km diameter collecting area would generate about 40 GWh of energy in a day, at an assumed reception/conversion efficiency of 70%. By comparison, a flat array of 15%-efficient PV panels occupying the same area in the Mojave Desert would generate about a fourth as much energy averaged over the year. So these beaming hotspots are not terribly more concentrated than what the sunlight provides already. Again, I find myself scratching my head as to why we should go to so much trouble.

Grid delivery. You cannot easily move power from the desert into urban centers.

Well, you sure can't point the power delivery beam at downtown LA either.

And the footprint of the rectenna array is enormous-- Tom assumes a 100GHz transmit antenna 30m wide, which means the downlink antenna has to be 3.6km wide! And this is the minor axis-- the solar array has to be geosynchronous orbit, which means any receiving antenna not on the equator has be an ellipse.

Not just deserts. You could also float them in the ocean, no? We have 70% of our planet we're not using.

Spread them far enough apart and add some movement and shadows wouldn't hurt wildlife much.

The ocean environment isn't easy on equipment, plus storms are a major issue. Then there's cabling...

That all looks trivial compared to the challenges of fusion power, though.

Are you sure? This is a recurring theme in your arguments but I don't necessarily agree with you.

What are the reasons for you to think that?

Read the article that this conversation is attached to for a list of very good reasons. Getting fusion to work is most certainly possible, but incredibly hard. Engineering for salt water is way easier than engineering for neutron activation, achieving high neutron economy to breed the tritium, managing tritium, managing a high temperature reactor next to a very cold magnetic containment system, a high vacuum system, oh and the very basic problem that has yet to be solved (achieving even just breakeven fusion in a non-H-bomb) etc, etc. It really does look trivial to just properly seal stuff against saltwater.

We'll probably need fusion to help reach the stars, so I think it's definitely something that should be pursued, but making it an economic power source for Earth seems very unlikely to me.

We already have things that float in the ocean for long periods of time (ships, oil rigs) and which withstand ocean storms, so it seems like a solved problem compared to fusion.

And people could...idk...work on solar rigs to maintain these systems!

Truthfully in the medium term next generation modular fission reactors seem like the way to go, as a complement to intermittent and variable solar/wind power.

If the LPPFusion approach eventually works out, it will be a giant win compared to the Tokomak or implosion fusion reactors.

By the time you mount those panels on a floating structure the lifetime energy yield may well drop to zero or (most probably) negative. Building a ship has a huge energy/emissions cost as the raw materials have to be mined, refined, transported, formed and assembled.

I'd recommend this site for a good overview on where our energy really goes and what might be achievable with renewables: https://www.withouthotair.com/

> You could also float them in the ocean, no?

Yes, offshore wind and solar farms do exist.

Bump to Robotbeat. Fission reactor designs, with many clever advances in safety, have never been tested because of the lack of support from U.S. DoE. Pilot-scale reactors are expensive, but would really help to advance our working knowledge of these new technologies. Overall, nuclear fission has gotten a bad wrap.

As for the cost of storage, that was worked out 20-30 years ago via the Yuca Mountain facility. If one of the local managers hadn't shifted to a "organic-based storage medium', that was not pre-approved for that application, the fire wouldn't have happened and everything would be fine.

Thorconpower is a case in point. They tried very hard to get their experimental reactor off the ground in the US but their pilot project seems to have moved to Indonesia instead now. That's a shame - Thorium and molten salt reactor experience and technological mastery would be extremely useful for solar system space colonization efforts past the orbit of Earth (insolation levels fall dramatically and a lot of basic industrial chemical engineering off planet will encounter abundant raw materials but will also need easy access to enormous quantities of energy). Fissiln is so worthwhile pursuing and it's reputation so bad it will probably need an Elon Musk type persona to tackle it.

But that is kinda how everything goes with nuclear. You can't plan for the "if everything goes right" scenario. Shit happens. Humans are very prone to errors. But nuclear proponents always want to pretend that isn't the case and explain away very real things that have happened as being anomalous- when in reality, errors and mistakes are the way the universe works.

More power to desert people! What a great idea!

Reply on news.ycombinator.com