[hunson_abadeer]

This is precisely what put me off in these discussions. Not the idea that we might have found a room-temperature superconductor - that part was exciting. It's the part where people confidently talked about its applications without realizing that they probably wouldn't revolutionize CPU performance (Josephson junctions don't seem to work well as non-cryogenic temperatures for reasons unrelated to superconductivity), power grid transmission (transmission lines are already pretty efficient and we already choose less efficient materials for cost), or energy storage (LK-99 would likely have a fairly modest current limit before it stops superconducting).

LK-99 would have interesting applications, known and unknown, but we have a pretty good understanding of superconductors based on 100 years of practical research, and I find this kind of instant punditry pretty tiresome.

[raphlinus]

Amen. When someone does the math and adds up the winners and losers in all this, one clear winner will be this video from Asianometry, entitled The History of Superconductors (Before LK-99)[1]. It only lightly touched on LK-99 itself, but did an excellent job going through the actual science-based history of superconductors, covering in particular detail previous hype waves. A major point is that the YBCO superconductors, while an amazing scientific discovery, haven't had revolutionary applications, and have only lightly displaced lower temperature (niobium-titanium metal alloy) superconductors in applications requiring generating strong magnetic fields, including MRI machines. For the curious, [2] goes into considerable detail on potential applications and challenges for HTSC in MRI.

[1]: https://www.youtube.com/watch?v=wUczYHyOhLM

[2]: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5472374/

[DennisP]

REBCO is revolutionizing fusion reactors. Several companies are using it to build tokamaks with the same performance as ITER, but in a tenth the size.

REBCO supports stronger magnetic fields, and conveniently, tokamak output scales with the fourth power of magnetic field strength.

[pfdietz]

> same performance as ITER, but in a tenth the size

So instead of being 400 times the volume of a PWR with the same gross power output, they're just 40 times the volume. It's no panacea to the economic challenges facing fusion.

The other way to get high volumetric power density is go with a configuration of higher beta, the ratio of plasma pressure to magnetic pressure (fusion power at a given magnetic field scales as beta^2). Helion isn't using superconductors at all.

[u320]

REBCO was a bigger deal for fusion than LK99 would have been. We can't make tokamaks smaller, the magnetic forces would rip them apart.

[kerkeslager]

> transmission lines are already pretty efficient and we already choose less efficient materials for cost

You're correct, and this highlights a problem I often see in discussions: "efficiency" just is a measure of benefit/cost. Without knowing the units of benefit and cost, people aren't making meaningful statements when they say "efficient". The important efficiency of transmission lines is capacity per dollar, not capacity per material, and no material requiring lab crystallization is going to be remotely competitive in capacity per dollar.

[criley2]

This is an absolutely disingenuous point that compares the cost of full-economy-of-scale tech to literal one off R&D prototypes.

Maybe new technology made in a lab can one day scale up and compete against current low-cost high-scale solutions. Crazy idea, I know.

However, trying to artificially limit all discussion about R&D and future tech by claiming "it's more expensive than fully scaled solutions" has got to be full luddism. This loom prototype is too expensive! I can hire a man for a shilling a day!

[kerkeslager]

> This is an absolutely disingenuous point that compares the cost of full-economy-of-scale tech to literal one off R&D prototypes.

No, even at scale, materials that you can extract from ore are inherently going to be cheaper than materials you have to extract from three different ores and then crystallize, even in a manufacturing lab. These just aren't comparable processes, and no amount of scale is ever going to fix that.

Instead of assuming I'm making a disingenuous point, you might have asked for clarification.

That's setting aside the problems others have brought up, which is that the materials in question have other properties besides conductivity which make these materials inappropriate for transmission application.

[u320]

No, switching from raw aluminium to an obscure synthesised compound is not going to be worth it for a few % efficieny gain. We've had centuries of "scale up" with copper and it's still not worth it.

[benj111]

Still isn't going to work.

>material requiring lab crystallization

How are you going to string a crystals between towers? The material properties are all wrong for this application.

[kergonath]

Aluminium and copper in cables are crystals. The crystal bit is not the problem.

[benj111]

I'm not sure of the correct scientific language here.

As far as I'm aware this is a brittle /inflexible material so my point about the mechanical properties still stands.

And when people refer to growing crystals, that generally refers to a particular kind of crystal. Ive never heard of anyone growing aluminium crystals, except if it's a compound, and then you get a crystal like we think of when we say crystals.

[kergonath]

> As far as I'm aware this is a brittle /inflexible material so my point about the mechanical properties still stands.

Yes. You want them to be ductile (malleable, or that can be deformed permanently in less-technical language). Although they could also be flexible (meaning that they can deform, but go back to their natural shape if we stop applying a force), as in the case of fibre optics cables, which are actually not crystals but quite brittle.

The interesting twist is that a solid pretty much has to be a crystal to be malleable. Almost all the metals you can think of are in their crystalline state.

> And when people refer to growing crystals, that generally refers to a particular kind of crystal.

I don’t know. From my experience people equate crystals with shiny things without really thinking about it. But this is HN, and we should try to be a bit better than a random person on the street. After all, most people don’t know a web browser from an OS, but I would be ridiculed if I make that confusion here.

It is a wonderful community where you are almost certain to discuss with some experts in pretty much any given field, it is a great opportunity to learn and grow.

> Ive never heard of anyone growing aluminium crystals

If you’ve seen solid aluminium, then you’ve seen it as a crystal. It is pretty much impossible with common techniques to get non-crystalline solid aluminium.

> except if it's a compound, and then you get a crystal like we think of when we say crystals.

That’s the thing, I don’t know what you think of when you say “crystal”. In actual fact, a crystal is a state of condensed matter in which atoms or ions are aligned in a 3-dimensional pattern that can be replicated to fill the space. In the case of aluminium, you can actually see how the atoms are arranged in a periodic way in articles such as this one (figure 3): https://www.researchgate.net/publication/323423565_Anomalous... . There are many other examples, and it is absolutely fascinating. We have the tools to count atoms and see the structure of the material!

And it is undoubtedly a crystal.

[rubylark]

In this context, they are speaking of electrical efficiency, i.e. the amount of power lost to system impedance during transmission, not some abstract concept like effectivity. The efficiency of a transmission line is expressed as a ratio of power received at one end of the line over the power sent at the other.[1]

[1] https://en.m.wikipedia.org/wiki/Electrical_efficiency

[AlexandrB]

The cost still matters because if the losses cost less to replace than the superconducting material costs to install, no one will use it. So parents point still stands. It doesn't matter how high the electrical efficiency is, what matters is cost efficiency.

[ABCLAW]

This is only true to a point. Evaluating incremental cost benefits on the basis of the delta of energy loss along specific lines ignores the state change that occurs when main trunk elements of the grid become lossless and energy generation and storage solutions can be deployed in a near-location agnostic manner.

As with all toy models being applied to the real world, there are important factors to model in that aren't immediately obvious.

[kerkeslager]

Who's "they"? If anyone is talking about electrical efficiency they shouldn't be because cost efficiency is what matters for a transmission line. Transmission maintainers have no reason to care if a wire transmits with 100% electrical efficiency if it's cheaper to lose some electricity than pay for the perfectly electrically-efficient wire.

[drdeca]

I had heard the parts about "probably wouldn't be a big deal for CPU performance" and "probably wouldn't be great for energy storage", but I hadn't heard the point about "we use less efficient materials for power grid transmission than we could, because of costs".

I suppose I didn't expect that we necessarily had like, the "absolute most efficient that could be made" (if that is something substantially more complicated at a materials-science level than "some simple-to-make-alloy"), but I hadn't imagined that it was a substantial difference. (I think I had imagined that they were... copper wires with like, surrounding metal tubes, or something? I hadn't thought much about it.)

Could you either say, or give my a search term I should look up in order to read, a little more about the trade-off being made between materials cost and efficiency of transmission lines?

[svetb]

Am not the author of that comment, but the fact that comes to mind is that aluminum is used for virtually all transmission and distribution lines - for price reasons - even though copper has better conductivity.

If we did discover a room-temperature superconductor, I suspect it would be a while before the cost to produce it in the bulk quantities required for electrical transmission are economically attractive compared to what’s already available.

[tsimionescu]

> it would be a while before the cost to produce it in the bulk quantities required for electrical transmission are economically attractive compared to what’s already available.

Note that there is no guarantee that that would ever happen. Electrical resistance is not the only thing you need for something to be an economically efficient power line. While superconductors are by definition excellent in terms of electrical resistance, there is nothing to guarantee that they wouldn't be too brittle, or too heavy, or too hard to mould into the required shape, or simply require materials that are too rare on Earth. And all of these would not be things that can just be worked around with better production processes or smart engineering - they would be fundamental limitations of the specific material, just like the low temperature requirements of currently known superconductors will never be improved with more research.

So this isn't a matter of when they would reach the point of being better economically, it's also very much a matter of if they would ever reach that point. Hopefully, we'll get lucky one day and find a material that is superconducting at room temperature and above, that is study and light and easy to make into wires and made out of abundantly available elements. LK-99 certainly wasn't most of these things. Even if it had been superconducting, it wasn't a good candidate for any of the other properties we want anyway, so it likely wouldn't have been much better than other known materials for most applications.

[pclmulqdq]

The pace of development of computing seems to have trained people to think in terms of "when" for science and engineering problems. The normal paradigm is to think in terms of "if," and that aligns well with most non-computing inventions.

There is a good chance that they never reach the exponential breakpoints that everyone likes to fantasize about.

[jonathankoren]

Yeah. There’s a lot of wishful thinking about science and sciencing up solutions to the world’s problems — especially here. The fact is, most progress is slow, and even if there is progress, it’s not necessarily economical in either financial or energy perspective.

[u320]

In theory, we could have had a much better power grid with more transmission. The reasons we don't have nothing to do with the price of aluminium, or the resistive losses of it. It's just difficult to build large-scale infrastructure. Transmission projects typically spend longer in court than actually building them. Superconductors would not change a thing, unless it changed that.

[Tuna-Fish]

Aluminum vs Copper is not that simple. Aluminum has worse conductivity for the same area, but area is in no way fixed. And aluminum has actually better conductivity than copper for the same weight. You just have to make the cables a bit thicker.

[anamexis]

I think the relevant metric here is conductivity for the same cost.

[II2II]

It may come down to cost, but other physical properties enter the picture. For example: thermal expansion is an issue for overhead power lines, along with how ductile it is.

In other cases it is more important to reduce resistance, not so much because of the power loss but because of what the power loss means: the generation of heat that may be difficult to remove.

Of course you can get around those problems at extra cost, but it is more than a straight up comparison of the material cost of the conductor.

[Joker_vD]

In some desperate places, people would cut down aluminum power lines and sell them to scrapyards for some quick buck. But copper power lines? Those would be in a similar danger in many more places.

[Roark66]

Not only in desperate places. I heard last year (or the year before) someone stole few km of train wire in Germany. Although to this day some people think it was a Russian sabotage rather than genuine theft. Previously (for example in Poland) I used to hear about things like this all the time until maybe a decade ago.

[pbhjpbhj]

They steal buried copper cables in rural locations (UK) by attaching one end to a truck and driving off. Mostly seems to be communication lines.

[lazide]

Tying a high voltage power line to a truck is a recipe for an exploding/melting truck, long before someone could pull it down.

Communication/low voltage is a different matter of course.

[chias]

Silver is even more conductive than copper!

[dgoldstein0]

And gold too.

Very expensive to build anything sizable out of it

[RF_Savage]

Gold (2.44x10-8 Ω•m) is worse than copper (1.68x10-8 Ω•m), but better than aluminium (2.82x10-8 Ω•m).

It does have excellent anti-corrosion properties.

I wonder what kinds of alloys we will see in the potential future with asteroid mining and thus comparatively cheap gold. Imagine replacing lead with gold in industrial applications. Or the stainless steels with a gold component in them.

[dgoldstein0]

The trouble with asteroid mining is that getting anything there and back is expensive, let alone any heavy equipment needed for large scale mining.

My guess is the main application will be for space missions that find it cheaper to carry mining/manufacturing equipment rather than all the materials they need. Even that seems potentially a ways off. I suppose we could mine asteroids for science sooner, but that's quite a bit different than any mission plan which includes mining as a part of the required logistics. Maybe if there's some materials needed for extending life support capabilities? But still I'd have to wonder why not just take the extra supplies with you.

Maybe a moon or Mars base could change some calculus. As I suspect the break even point of such a plan may require lots of use of any such equipment.

[yetihehe]

Probably the most useful mtal from asteroid mining will be platinum for use in catalysts.

[elihu]

Aluminum vs copper is a good example. Another is that we already do use superconducting transmission lines in a few places. We could do more of that, but presumably it's expensive to install and/or maintain otherwise we'd be using it everywhere. I'm not sure what the longest or highest capacity superconducting links currently in existence are.

[SamBam]

That have to be kept cool with liquid nitrogen, so it would have to be pretty darn short.

[XorNot]

Actually no: they have to be insulated well. People forget that it doesn't actually take energy to stay cool, just to remove the heat. The issue is what's your heat gain from insulation inefficiency per length - and it does get better then thicker your cable gets, because volume increases more rapidly then surface area.

[klempner]

If you're dealing with usecases that need to be cooled anyway, you may well be better off with the tradeoff of needing liquid nitrogen cooling and better insulation in exchange for entirely eliminating resistive heat.

[kergonath]

High-temperature ones can be cooler with liquid nitrogen. Standard ones, the ones most commonly used, require liquid helium.

[nathan_f77]

I wonder if we can use these superconducters on spacecraft and probes. Maybe we can place superconducting links on the outer hull of a spacecraft heading to Mars, or a probe heading into outer space.

[krisoft]

But why? What is the problem you are trying to solve by placing superconducting links on the outer hull of spacecraft?

[RF_Savage]

Cooling them would still be a problem. The sunny side might not be the best place for them.

They might find a niche in some instruments in probes, but for wiring it does not make sense. The rest of the probe electronics don't like being that cold.

[elihu]

Well, I think it comes down to whether the energy cost of active cooling is better or worse than resistive losses. Which one is better doesn't depend on cable length.

[MobiusHorizons]

We frequently use aluminum wires with a higher thickness to make up for the lower conductivity as compared to copper. It’s not as simple as cost vs performance though, as aluminum is substantially less dense than copper. Gold and silver are also better conductors than copper, but of course are very expensive, and still have resistance. Zero resistance may be with it on some cases. For instance in projects that currently use high voltage dc it may be worth it due to safety and complexity wins, but that all would depend on how hard (expense and complexity) the superconductor is to deploy.

[cf141q5325]

>We frequently use aluminum wires with a higher thickness to make up for the lower conductivity as compared to copper.

Aluminum wires even made it into residential housing when copper was expensive/rare. https://en.wikipedia.org/wiki/Aluminum_building_wiring

[stoobs]

Can confirm, my parents had an aluminium telephone line in the UK until it failed and had to be replaced. Moot point as it's replaced with a fibre optic cable now though.

[cf141q5325]

The problem is with their usage as mains power. I think they are considered a fire hazard in older German homes.

[stoobs]

Not surprising, any degradation in the connection leads to intermittent connection/high resistance fault = heat and poof, there it goes.

[ithkuil]

No only silver is a better conductor than copper. Gold is worse.

[cogman10]

The crux of the problem for superconductors used as power delivery is the "critical field" problem. [1]

Super conductors are superconductive to a point. Once that point is crossed they turn into regular conductors. (I've seen ~1A cited. For context, EVs charge at around 500A).

To make them useful for power transmission, you'd have to up the voltage to insane levels to avoid collapsing the field.

[1] https://en.wikipedia.org/wiki/Critical_field

[floxy]

Superconductors have a critical current density (Ampere/m^2) that varies with temperature and external magnetic field[0]. So if you want more current, you need to use a bigger wire (and/or make it cooler). YBCO HTS tapes have enough current density for power transmission[1].

[0] https://en.wikipedia.org/wiki/Yttrium_barium_copper_oxide#/m...

[1] https://www.amsc.com/comed-and-amsc-announce-successful-inte...

[Eduard]

or just go the straightforward way and use several transmissions in parallel, as it is already done for existing superconducting lines in production.

The AmpaCity project in Essen, Germany, gives insights about the implementation details, as the involved parties were required to publish their work.

https://www.enargus.de/pub/bscw.cgi/?op=enargus.eps2&q=%2201...

for the specific aspect under discussion, the Karlsruhe Institute of Technology report is of interest:

https://www.tib.eu/de/suchen/id/TIBKAT:872231372/Ampacity-10...

[TylerE]

A 2" diameter copper wire will have lower losses than a 1" diameter copper wire.

Copper is expensive so over hundreds of miles you may not want that.

[trzy]

Accelerationism has become a religion for many people working in tech. Social media is teaming with John the Baptists heralding the next messiah.

[Cthulhu_]

This is the curse of popular science websites hyping things up; most people, present company included, have no idea what the scientific language means - be it superconductivity, LHC results, or astronomic spectrography.

So popular science wraps it in a "what you could do with it. maybe. possibly." Or what it means. And commenters have latched onto it, but a lot is said with an air of confidence, of just-so. "Oh uh, superconductors, conducting is passing electricity from one end to the next, super is like really good, uuh uh uh... I know, what about power lines from the Sahara to Europe so they can build solar collectors down there!"

Same with exoplanets, the actual science is "yeah the luminosity of this star drops by 0.0003% at a cycle of 300 days and we're getting some photons that indicate there may be hydrogen molecules", pop sci turns that into "EARTH-2 TEEMING WITH LIFE DISCOVERED, GENERATION SHIP WHEN?"

[hardlianotion]

Funny you should mention the solar connectors and electricity interconnections. There is a Morocco -> UK interconnector project that is underway right now.

https://xlinks.co/morocco-uk-power-project/

[burnished]

Interesting, from what I saw a lot of people got informed on why those overly confident predictions were drek - I don't know that I have seen a claim go unchallenged.

Which seems ideal to me. Very educational.

[cogman10]

It was like stomping out weeds and it wasn't always well received.

I hope that those that got dashed (and observed the dashing) take a step back the next time something from "FuturistSuperScienceNews.com" or whatever pops up touting a revolutionary XYZ. Those sites are like 99% trash that train their readers to distrust science when their clickbate articles don't pan out. If I were conspiracy minded, I'd swear they exist to build out a mistrust in institutions.

[jboy55]

I felt a similar way with the news of the fusion 'breakthrough' around 6 months ago. "Fusion power is here! All we need to do is engineering!".

They achieved this fusion by creating a container of material that produced massive amounts of xrays when it was bombarded by a high powered laser. These xrays caused another container's surface to ablate at such a rate it compressed its interior to the point that fusion was achieved.

However, this being a weapons lab, they created the experiment to model the secondary device in an H-Bomb. The secondary is theorized outside the Top Secret world to be a cylindrical tamper of (enriched?) uranium. One hypothesis in the public sphere, is its the primary device's Xrays that cause this to ablate at such a rate and that the inside is compressed to achieve fusion. The purpose of the fusion is primarily for the neutrons it generates, which are used to cause a massive amount of fission in the tamper, producing the majority of the energy. For example, if replace the uranium with another non-fissile material, and you have a "neutron bomb".

The reason the breathless hype annoyed me is that at no point was usable energy the desire of the test. In fact, the test solely was to feed real world data back into the supercomputer models, so that we know how our existing stockpile of weapons would work or even perhaps to find optimizations. We know this mechanism of ablation causing fusion works, we've known for 60+ years, all we're doing is doing it in a lab.

I'm not sure why there is this need to hype these events, like fusion or LK-99 so much. It seems that being a naysayer is reacted to as if the naysayers are explaining a magician's tricks. As if we don't hype these events the public will lose interest, or even our children will drop out of STEM careers.

[EthanHeilman]

> They achieved this fusion by creating a container of material that produced massive amounts of xrays when it was bombarded by a high powered laser. These xrays caused another container's surface to ablate at such a rate it compressed its interior to the point that fusion was achieved.

You are telling me that a US weapons lab just announced a successful path to a laser triggered pure fusion bomb? Yikes!

Not actually sure if it can be used to ignite more fusion fuel, but if they using this to test secondaries then it sounds like it might.

I really hope we get fusion reactors before pure fusion bombs, as pure fusion bombs are going to be a nuclear non-proliferation nightmare. While it might not be easier to built pure fusion bombs than bombs with a fission trigger, controlling the precursors and knowledge is going to be very difficult.

> "Fusion power is here! All we need to do is engineering!".

I agree with this statement and it has been true of fusion since at least the early 2000s. Don't underestimate the difficulty of engineering. Safe fission breeder reactors are an engineering problem as well, one which humanity has largely abandoned due to repeated failures.

[jboy55]

I think the most efficient means of delivering so much xrays that kilograms of material can fuse is with the primary stage of an hbomb, which is just an implosion fission bomb. I wouldn't be too worried about this test creating a new weapon.

However... In the early 80s, the SDI initiative aimed to have orbiting satellites that utilized x-ray lasers to shoot down incoming warheads. The theory of these were you had h-bombs in orbit, with long cylinders of a material that would amplify the x-rays from the bomb. You'd point these at the incoming warheads and trigger the bomb and (chefs kiss) you have beams of xrays that would destroy warheads.

One of the major reasons this was skuttled, was that the test they used to find a material they thought amplified xrays was flawed (see below).

With the test-ban treaty, they weren't able to test any other materials. Now we have a facility that tests materials to amplify x-rays...

Sidenote: The test was, explode a bomb in a tunnel, shut the tunnel down with explosives to trap the shockwave, then use the xrays to test materials to withstand x-rays as well as amplify them. Teller thought they had seen amplification and sold the military on the satellite idea. Another scientist, thought it was a secondary thermal effect on Oxygen. There is an interesting story about the back and forth, and the pressure to have another scientist lose his credentials for disagreeing with Teller, that is a good follow on to the Oppenheimer story. https://en.wikipedia.org/wiki/Project_Excalibur

[EthanHeilman]

> I think the most efficient means of delivering so much xrays that kilograms of material can fuse is with the primary stage of an hbomb, which is just an implosion fission bomb.

I agree with you, if you have fission triggers, you aren't going to want to use lasers. At least with today's lasers.

> I wouldn't be too worried about this test creating a new weapon.

My concern is that NNP has focused on controlling access to fissionable material, so potentially this is a path to h-bomb that doesn't require fissionable material. As lasers get better, secondaries that don't use controlled fission materials become a risk. At what point does the world start having to worry about controlling access to lasers? How does this impact the future research and funding of lasers?

Additionally if you can test h-bombs without tests. This also makes it easier to develop and test a h-bomb without revealing you have an h-bomb. Typically nuclear weapons tests are detectable via seismographs.

[Forgotthepass8]

It also wouldn't change much in MRI (formally NMR) -- it's also very limited on other factors

[pbmonster]

I mean, both NMR spectrometers and medical MRI machines would be a hell of a lot less complex without the cryostat.

If you remove that, those things become... really, just tubes wrapped in various coils connected to a software defined radio of average quality.

[Forgotthepass8]

The hardware for RF and Gradients alone isn't that cheap was my thought

Also you can't just write off the fringe field.

[KSteffensen]

A large part of the energy loss in electronics happens in switch-mode Buck-Boost DC-DC converters, as I understand it mainly due to internal resistance in the components used and due to the magnetic field not being directed enough to transfer 100% power between two inductors.

Would a cheap room temperature superconductor bring any benefits here?

[magicalhippo]

For "normal" DC-DC converters it's the losses in the semiconductor switches and diodes that dominate[1], unless cheap inductors or capacitors are used.

High-efficiency DC-DC converters often use a resonant tank circuit[1], which supports high-frequency operation and zero-current or zero-volt switching, which together significantly reduces switching losses.

In such a circuit I imagine superconducting inductors/transformers and superconducting capacitors could be beneficial to improving efficiency further.

Keep in mind though that resonant DC-DC converters can reach 98% (or higher) efficiency already[3] with current tech.

[1]: https://www.analog.com/en/technical-articles/an-efficiency-p...

[2]: https://www.monolithicpower.com/understanding-llc-operation-...

[3]: https://ietresearch.onlinelibrary.wiley.com/doi/10.1049/iet-... (random example)

[floxy]

>Josephson junctions don't seem to work well as non-cryogenic temperatures for reasons unrelated to superconductivity

Can you point me in the direction to learn more about this?

[Fatnino]

I think it would make MRI machines cheaper

[laserbeam]

Also... The material was always a ceramic, and you can't do much with other ceramic superconductors either.