Hacker News new | past | comments | ask | show | jobs | submit login

> if successful LK-99 would be a watershed moment for humanity easily on-par with invention of the transistor

Is this really true? I've seen comments that "it would make a lot of things quite a bit more efficient, but it's not a revolution".

edit: Reading through the linked material now: https://nitter.moomoo.me/Andercot/status/1685088625187495936...




I think that this is kinda like the lamp => transistor change. The immediate usages are not revolutionary, moving from lamps to transistors in amplifiers didn't change everything overnight. First products with transistors were barely better than the ones before. But oh boy did it create an infinite new horizon of opportunities on the long term

I don't really know whether the analogy holds, but I find reasonable reasons to think it does (we could finally get our low-voltage DC home \o/ maglevs \o/ routine/cheap IRMs \o/. that's just the "simple" applications that we can already foresee right now)


As far as I'm concerned, I enjoy the rich, full sound of silicon semiconductors. I won't be putting any of these LK-99 abominations near my ears any time soon. /s


What do you mean by lamps? Electron tubes?


Yes, they're talking about https://en.wikipedia.org/wiki/Vacuum_tube s


I think so. In some languages electron tubes are called lamps.


Vacuum or Electron tubes describe how they work, lamps are what they look like, and valve describes the function that they provide.


Is this the same as “valves” that were always talked about in the distinctive sound retro (now) guitar amplifiers?


I have always called them vacuum tubes (in US) but I think they are valves in the UK.


Yes


My guess would be that those letters are an even more shortened form of linear amp(lifier). That term does not really imply anything about transistors vs tubes, but the way GP uses it suggested some jargon exists where it's exclusively used for the tube incarnation of the concept.


> My guess would be that those letters are an even more shortened form of linear amp(lifier).

No, they were colloquially called (void) lamps in at least French and Slavic languages, because they looked like incandescent bulbs, that were called lamps (lampe/лампа) as well.


And tubes glow like a lamp so it is natural to call them so.


So they are lamps, just very inefficient.


I don't think so, it seems to me it is because its semblance with incandescent bulbs. In Spanish are actually also known as "bulbos"[1] (bulbs).

[1] https://es.wikipedia.org/wiki/V%C3%A1lvula_termoi%C3%B3nica


thermionic valves I would guess.


Weapons.

Energy weapons are still mostly science fiction right now. This changes that. Energy storage increases by an order of magnitude and with no resistance you can charge and discharge batteries instantaneously.

Probably some other really interesting uses, but energy weapons is what we'll figure out first. And not just giant ship mounted rail guns—if true this will start a new arms race for personal handheld weapons. No more gunpowder. Think rifles with almost no maintenance. Outrageous magazine capacity with shot capacity limited only by energy storage capacity… which can easily and quickly be resupplied if you are behind a supply line.

If this is real you'll seen an untethered Boston Dynamics Atlas with a 7-minute runtime and a placeholder handheld railgun by year's end. That will kick off a series of RFPs, and in 3-5 years… real life Terminators baby. While we take refuge under the rubble we'll hear them chanting above us, "Hello, I'm calling about your car's extended warranty."


Wtf are you talking about? Superconductors are limited to 40kJ/L energy density.

For reference, gasoline is 33,000 kJ/L.


That's about an order of magnitude more than current batteries.



We could do solid state current switching before transistors, with diode valves and whatnot. Transistors "just" made it more efficient.

Room temperature superconductors would not just just make a host of things more efficient, it would also enable devices that rely on high-strength magnetic fields to become much more widespread. For example, if you could remove the cooling requirements for MRI scanners then they could become much more usable "in the field".


This wouldn't have any effect on MRI scanners generating such magnetic fields.. they would be too dangerous to use in a non-prepared environment. They'd be small and use less power but still require a prepared environment wouldn't they?

Use it on a battle field and it'd be like Magneto was throwing shrapnel around I'd think.


Building a room from copper RF shielding panels [1] and soldering them together is a lot easier in the field than supplying a whole liquid helium or nitrogen cooling system. The former is well within the capabilities of a forward operating base - they can just ship the shielding in on plywood and the electrical technicians are more than capable of soldering copper.

[1] https://www.nelcoworldwide.com/medical-shielding-products/rf...


Sounds like a good idea for supermagnetic landmine or a grenade. Throw one into a trench and all enemies lose their weapons.


It sounds absolutely absurd but I'm thinking of how the military generates EMPs (explosive forcing a puck through inductive coil to generate a huge current which then generates the EMP), and maybe you actually could get some serious "deployable", very strong magnetic fields. Basically take the explosive EMP, and instead of directly trying to maximise EM induction in a pulse, you somehow try and stretch that pulse and apply it to a superconducting coil, generating a brief instant of strong magnetic pull.

The "pull weapons" thing is silly, it's a magnetic field so if you're that close the same weight in explosives should kill them, and with the forces it would need it's not going to be non-lethal.

I could maybe see this as an active defence system. Have a little turret, if you detect an incoming slug you fire your little grenade at the slug, it sets off a pulse that imparts substantial impulse to anything that's magnetic, diamagnetic, paramagnetic (lead, steel, uranium all included). Maybe it could send a high speed slug tumbling, or off course, not sure.

In general I think RTAPS is much more likely to have civilian uses than military ones, aside from banal military uses like "the generator works more efficiently because parts are superconductive now". Maybe railguns, but I don't think electric resistivity is the limiting factor there, more friction or plasma confinement.


> Transistors "just" made it more efficient.

Sort of, the thing about transistors isn't the efficiency as much as the scalability.

You'd be very hard-pressed to make a modern processor out of vacuum tubes. It would be enormous, tedious to build (couldn't use modern lithography), and also consume tons of power.


> . It would be enormous, tedious to build (couldn't use modern lithography), and also consume tons of power.

Sounds to me like you're saying that Transistors are more efficient than vacuum tubes, both in terms of space and power consumption.


Many orders of magnitude so.

The last generations of vacuum tube processors were the size of large multi-story office buildings and had 50k tubes.

Compare to your smartphone, it would take billions of tubes to duplicate it.


Later research into thermionic devices did produce miniaturized versions, but nothing close to the scale of silicon transistors today.

If the same R&D might went into thermionics, we might just have devices of similar scales.


Yes. But at some point a quantitative difference becomes qualitativ.


If you're simply using efficient as a synonym for better, then every improvement is tautologically more efficient than what it immediately replaces. But solid state transistors were pursued specifically for their reduced power consumption while their scalability, which at the time was an afterthought, wound up making them revolutionary.


Tubes also burn out regularly (literally every day) when you have thousands of them.


> [...] tedious to build [...]

It (or major parts of it at least) could be built by a fully automatic / robotized manufacturing plant. That isn't "tedious". Even for a one of its kind processor that would be much cheaper than the manual way.


> For example, if you could remove the cooling requirements for MRI scanners

But MRI is almost the only tech we know today to be affected by this (and we could count maglev as well, but I don't think it'll change the landscape too much here, as the deployment of high speed trains un general is less technologically limited than it is by limited politcal will).

And even for MRI, room-temperature SC is only a big deal if you can find zone that works for high current, wich isn't the case here son far.


Petrol is just "better" coal. Except that you can't make flight with steam engines.

When room-temperature SC becomes possible, previously unthinkable applications suddenly become viable.


Just to be the pedantic one… I did some searching and found one single example of a steam-powered aircraft that actually flew. A bunch of other attempts are listed on wiki, all unsuccessful.

https://books.google.com/books?id=AygDAAAAMBAJ&dq=Popular+Sc...


Sweet! I found a video clip at https://archive.org/details/turner_video_11514 at timestamp 54:47 for about 20 seconds.

(Found that reference via https://web.archive.org/web/20030203103015/http://www.flyste... which I found via https://archive.org/details/australian_model_engineering-iss... )


And thanks, now I am researching small steam boilers that I could install in a Pietenpol homebuilt aircraft that a friend and I have been talking about building for a few years now. All I need is 30 kilowatts.

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


I knew it!

Yeah, I really meant coal-powered steam engines.


If you built a coal gasifier, or maybe used a powdered/pelletized form of coal, I could see it being plausible. (but still dumb)

Having just returned from a week at the Experimental Aircraft Association’s yearly gathering, I’m kind of interested in using up some of my employer’s time to see if I can make it work, at least on paper. Luckily I even have a budget line available for silly studies like this.


But not visiting Andromeda, the next star system. Which would be the only such application I can think of now.


Maybe, but the same was already said when cuprate SC appeared in the late eighties with their relatively high critical T°, but they haven't really fulfilled their promise…


Are they really the same though? I mean we are talking about room temperature, not the"(relatively) high temperature", a term that is arguably misleading to a layperson.

I'd imagine most people who are informed of the difference would recognize that cuprate SC is nowhere near as useful.


Every article older than a couple weeks talking about superconductors goes on and on about how useful they are, except for the logistics of keeping them cold

Example: https://en.m.wikipedia.org/wiki/Superconducting_magnetic_ene...


Superconductors suporting high current are incredibly useful, but that's a big caveat. There's a reason MRI scans still use the super tedious helium-cooling instead of a much simpler nitrogen cooking allowed by cuprates: the cuprates SC aren't good enough for MRI.


Iirc the LHC experiment would not be possible without those superconductors. That’s a pretty big deal.


As if the US is the only country. Plenty of EU countries, China, and Japan would invest in developing even higher speed rail options.


The speed of trains is generally not limited by engineering. It’s limited by corners, i.e. obstacles. There are already very fast trains in existence that don’t use maglev.


You're assuming they'd replace existing routes rather than build new routes that would suit the higher speeds.


That's the main problem, building a rail line zigzagging around the country is arguably a ton easier than having to legislate, seize, steal and forcibly bulldoze a straight path for a maglev.


…and sometimes there are obstacles that you can’t bulldoze. Mountains, rivers, towns, etc. You can build tunnels and bridges but it’s very expensive just to get people to their destination a little faster. The temptation to add a little bend is strong, but the smaller the minimum turning radius, the slower the maximum speed, and the higher the chance that you don’t need maglev at all.

It’s very dependent on geography. Ironically the US is a much better candidate for maglev than Europe with its wide unpopulated expanses.


I'm French, and France has one of the biggest high-speed train network, and the limitation to build additionnal lines isn't a technological one…


I'd agree that it's not technological: it's cost and practicality.


trains that float above the rail means less maintenance. the high cost of trains is partly due to the maintenance of the rail network.


Ah, but how many countries would develop superconducting maglevs just to fund their defense program that's using similar research to make superconducting rail guns?


Probably zero. The problem with current railguns is not that they have too much electrical resistance but that their barrel wear is much too high for practical use. Superconductors wouldn't solve that.

In any case countries that want to spend money on military research just do so, using the budget they allot to military R&D. There's no need to try and hide that you are researching railguns.


SMES systems offer grid-level energy storage at about 98% efficiency. You could also move nuclear reactors to the middle of the desert because no one wants them in their backyard, then losslessly transport their power back to the grid. Same for bringing power from solar and wind installations. These are just two examples with immediate commercial application. The economics of energy production shift when you eliminate 15% of the cost and distance limitations. That would have a big impact on decarbonization. Longer term, josephson junctions replace transistors in processors.


You could put nuclear reactors in the middle of the desert today (ignoring cooling for now). The reasons we don't typically do it has nothing to do with resistance.


Superconductors allow creation of non-chemical energy storage whose capacity is only limited by storage's mechanical resistance from being crushed by its own magnetic field and which has nothing like "finite charge-discharge cycle life limit"

Additionally it has no internal resistance, so you can charge such battery virtually immediately, but you can also discharge it immediately. This will be important for energy based weapons.


The storage capacity is also limited by the superconductor's critical current. I imagine everything is fine until you add that extra bit of current that causes the material to switch out of superconductivity and all the concentrated energy has to find an escape real quick...


This (unplanned loss of superconductivity) sometimes happens to accelerator magnets in LHC. To prevent stuff from melting/blowing up, there's an elaborate protection system integrated ([1]).

[1] LHC Magnet Quench Protection System https://cds.cern.ch/record/259538/files/P00021565.pdf


Sounds like an energy weapon to me!


That's a physics-based assumption, not an engineering-based one.

Unlike the effect on nuclear fusion, the idea of using SC as storage devices is pretty much pure theory.

Edit: it's currently leaving the theory part at MW scales as pointed out below. That makes using LK-99 much more likely. But using LK-99 in a SC storage device is still theoretical.


> But using LK-99 in a SC storage device is still theoretical.

Of course it is, any kind of use of this stuff is still theoretical. That's a content free statement. But GP was making the assumption that if it works it can be used for storage. But that doesn't really follow from the properties of the material as described so far. You'd need a lot more current carrying capacity for that to become a realistic possibility.


It is actually used and it exists. https://en.wikipedia.org/wiki/Superconducting_magnetic_energ... See applications


That's neat, but it misses the point. LK-99 is not even available for engineering, yet. You might be able to use one of the existing designs. Maybe you need to come up with a new design. But you cannot do that just yet.


Will batteries still explode like LiPo batteries?


No, they’ll explode unlike LiPo batteries. LiPo batteries merely catch fire.

An electromagnet, superconducting or otherwise, has large internal forces. If the support structure lets go, it will move. And, if part of the circuit becomes non-conducting, an inductive kickback will occur, generating enough voltage to (initially) sustain the original current.

You could cut a Li-ion battery in half, and the two halves will continue to store their chemical energy, at least until they burn up. If you cut an inductor in half (which is what this type of energy storage device is), that energy will dissipate very quickly whether you like it or not.


A battery that transform all energy it stored into heat sounds more like a bomb instead. Imagine someone accidentally heat up the 'super-conductive' battery to the point that it is no longer super-conductive.


When an MRI machine has a quenching event, it's a very serious deal. All the energy from the electromagnet gets dumped as heat, which boils off the coolant very quickly. MRI machines have to have a quench pipe which allows the now-gaseous helium to escape to the outside just in case of such a thing happening.

The LHC had a quench event in 2008, which explosively vapourised about 6 tonnes of helium, resulting in considerable damage, and it took more than a year for the accelerator to come back online.


That goes for any sufficiently dense energy storage system, including most batteries. Imagine someone accidentally shorted out a car battery with something that can withstand a few thousand amps. (Come to think of it, a room temperature superconductor would do nicely for that purpose...).


Well, for starters, batteries have their own internal resistance. Not a whole lot, but enough that whether you short out with copper wire or superconductor, the effect would be the same.


> Well, for starters, batteries have their own internal resistance.

Indeed they do.

> Not a whole lot, but enough that whether you short out with copper wire or superconductor, the effect would be the same.

Not necessarily, assuming a charged battery in many cases with a copper wire the wire will simply heat up to the point of evaporation and then break the circuit as it sprays molten copper bits all over the place. Some heat will be generated in the battery as well. Watch people mess up with starter cables for some ideas on how this tends to go (and do so from a distance...).

Using a massive copper connector that you some how instantly put across the terminal and manage to keep there would indeed make the balance of the resistance shift to the guts of the battery, which would heat up faster than that that energy can be shed and hence in all likelihood (violently) explode. Besides bits of molten lead and zinc for a car battery you now also have the joy of having to deal with spraying acid. Which depending on the state of charge of the battery can be really nasty stuff.

With a superconductor there would be no chance of the conductor evaporating first, there isn't any work done in the superconductor so it will stay cold, an explosion of the battery would be all but guaranteed.


Keep in mind that a superconducting energy storage device is kind of the opposite of a battery. An idle battery at full charge has some voltage and zero current, and it’s perfectly happy to stay like that.

An idle superconducting energy storage at full “charge” is not carrying a charge at all — it’s carrying a current. If you cut the wire (or blow a fuse), V = L dI/dt will generate an arbitrarily high voltage to keep that current flowing.

I imagine one would need some spark gaps and/or capacitors to limit the voltage.


It's a coil. You can just place another inductor around it and increase / decrease that current at will. This is how Kamerlingh Onnes injected current into his superconducting media during the original experiments and that is a two-way street.

https://www.lorentz.leidenuniv.nl/history/cold/DelftKes_HKO_...


I did something similar once with a high power lowish voltage battery array - my soldering iron's shroud literally vaporized with a deafening bang before even triggering the fuse.


Ri of those is typically 1 Ohm or thereabouts, which at 500 A develops 500W in the battery itself and a large multiple of that in your 'load' (in this case your soldering iron :) ).

When working with large battery arrays I use tools that are taped in all the way except for the business end, just in case. All you need to do is drop a wrench in the wrong spot and it's party time.


Not 1Ohm, in my case the array had an internal resistance of 15 milliOhms, which is way more fun. Though I imagine the BMS and fuse were relatively effective resistors in that situation too.

Agreed about taped tools, I figured that one out right after replacing the fuse :)


Oh that's really danger territory.

A BMS typically has a small shunt that helps to figure out the state of charge as well as a large transistor in series with the current to allow switching the battery in and out of circuit.


Yes, I had an additional fuse in series, what I was referring to is that the BMS's own internal resistance is larger than that of the battery pack. Indeed in an overcurrent situation you can't rely on the MOSFET not to fail short.


I've seen fets in BMS in Bosch packs cracked in half, top blown off. Quite impressive.


I haven't taken a science class since I was 16 but could the battery be grounded to prevent an explosion?


How do you intend to keep the grounded battery charged ?


Well there wouldn't be fires that produce toxic fumes and burn even under water, but if the coil fractured in an accident all that energy would be released in a tiny fraction of a second instead of a combustion event that takes time. That seems bad...


> all that energy would be released in a tiny fraction of a second

Probably along with a large electromagnetic pulse...


Necessarily with a large EMP, as the energy is stored in the magnetic field.

Fortunately the energy density[0] is comparable with rubber bands[1] and supercapacitors[1], so while the EMP is large, it's not terrifyingly large.

[0] at least according to wikipedia, 4-40 kJ/kg: https://en.wikipedia.org/wiki/Superconducting_magnetic_energ...

[1] 1.6-6.6 kJ/kg and 10-30 kJ/kg respectively: https://en.wikipedia.org/wiki/Energy_density


The energy density isn't the scary bit, the power density is. The referenced paper in that Wikipedia article indicates ~10^5 higher power density than li-ion. So while it stores 10x less energy than li-ion it dumps it so much faster. The difference between a combustion and a detonation if you will. I'm guessing the vicinity of the thing would get turned to plasma.


A very small and localised plasma near the normal conductors that suddenly experience a rapidly decaying magnetic field; but the total energy is still (relatively) low compared to, say, the thermal heat capacity of the air of about 1.3 kJ/K/m^3.


That energy density number likely includes the cooling system.


That is what I thought too, but reading the paper referenced it seemed like they only included the coils, not even the support structure to keep them from collapsing. But it wasn't that clear.

Edit: they were clear that the limiting factor on energy and power density was the forces exerted on the coils.


A colleague recently showed me the coil of a broken superconducting magnet after it quenched. There was an easily visible rip in the structure, even though the whole thing was enclosed. It must have at least had enough kinetic energy to tear the metal apart.


I've had a relatively small (about 2" x 2" x 1") ceramic magnet explode on me. The shrapnel was embedded deep into the adjacent wall.


Can you elaborate on this please? How can superconductors be used for storage?


Basically, just push more and more electricity into a coil of superconductive wire. Because the wire has no resistance, the electricity just loops around forever.


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

For now, this is restricted to special applications due to great size and cost. The great size and cost are caused mainly by the cooling systems.


The magnetic field of a superconducting electromagnet contains a lot of energy because the field can be very strong with a large cross section. It’s the conjugate of a capacitor after all, it’s just that the energy density of normal inductors is much lower than that of capacitors. Using a superconductor also avoids resistive losses for (dis)charging.

(Inductors are very frequently used for very short-term energy storage (~fractions of a millisecond). For example, all energy output by a flyback converter was briefly stored in the transformer’s magnetic field. Unlike a regular transformer/forward converter, where the magnetic field is just a side effect of coupled inductors, so none of the energy is stored in it.)


But obviously you can't exactly carry multi-Tesla electromagnets around. So no car batteries, much less phone batteries. Large energy storage facilities for grid power regulation, possibly.


Could you shield them with mumetal? Maybe make tiny arrays of thousands of them each with smaller atorage surrounded by mumetal to let them be close. That way if they break it is only those along the break that would discharge instantly.


The quantities of energy released would heat up the material around it to the point that it may well lose its superconducting properties as well. Likely failure of any winding in such a setup would set off the rest as well unless there is a lot more separation. Mu metal has a relatively low saturation limit.

And that's before we get into the purely mechanical stresses created by such an event, which likely will destroy the vicinity of the carrier of the current.


Fermi check: Mu metal boxes are only good enough to reduce the influence of this planet's magnetic field, not eliminate, and that's a really weak field (in terms of flux density, it's of course also really freakin gigantic).


Well, they are used in electromagnetic shielding but you'd need an awful lot of it to overcome the kind of field that even a single winding of a superconductor puts out, my intuitive guess is that for any realistically available level of shielding that the magnetic field would punch right through it as if it wasn't there at all.

Just look at what happens if you leave something made out of metal lying around near an MRI machine when it is switched on and that's not for want of attempts to shield it.


If you will close it into ferrite cup as it is done today with coils, then all the magnetic field should be limited only on coil on ferrite material itself.


Vacuum tubes were quaint tools for labs, but then someone figured out how to program logic on them.

You can't a priori plan out a path to new technology, it's usually surprising. This new category of materials can be a platform for lots of surprising uses


The point contact transistor was rudimentary too.

https://en.wikipedia.org/wiki/Point-contact_transistor

that said, it was still manufactured and sold before being replaced by more "conventional" transistors.


consider that a big part of why we cant make better processors is because they get hot, and they get hot because of resistance. I'd say dramatically faster chips in smaller form factors (think enormous graphics card in your phone, most of the bulk in those is cooling) would be quite a revolution !


Depends on the details, as always.

I don't know the superconductivity mechanism here (does anyone?), but IIRC Cooper pairs in BCS theory can be separated by much larger gaps than the transistors in a modern CPU — hundreds, rather than single-digits, of nanometers.

Might make fully-3D processors much more viable though, from lack of heat dissipation; and if it does, that in turn might be able to make up for a coarser resolution.


Heat is a big factor, but part of limitations on speed really come from physical dimensions. At some point, speed of light travel becomes a limitation.

Also, the transistors themselves generate a lot of the heat... So you still have large amounts of heat to handle even if the interconnect traces are heatless.


> Heat is a big factor, but part of limitations on speed really come from physical dimensions.

undoubtably. We can't make things infinately fast. But being able to put the most powerful desktop processors and graphics cards we have today into a phone would definately be useful, not to mention laptops. It would also dramatically change how server farms work (their existance right now revolves around cooling, they could be so much more dense)


The transistors create a lot of heat because the current carrying capacity of the interconnect is severely limited which increases the time to switch the transistor from off to on or vv. This puts the transistor into a resistive domain where it consumes power and that leads to heat. If you use a superconducting interconnect that transition time will drop sharply and that in turn will lead to less power used by the transistor.


Oh yes… with more frequent and routine MRI we have a path to helping maintain longer human lifespan.


As one of the posters in that thread asked, "who is this guy?"


energy storage is the big one in my opinion.

you can just pump electricity into the superconductor capacitor and it wouldn't have any loss it would just stay there until discharge.


Capacitors are fairly good at keeping charge - there isn't much loss over time. In fact, capacitors used in high voltage equipment should generally have a discharge resistor wired across them to deliberately discharge them slowly, to prevent them being a hazard to anyone repairing the equipment.

Superconductors don't help with this (much - just maybe with the wires leading up to a capacitor). Superconductors allow much better inductors instead. You can also store energy in an inductor, but it's different because in a capacitor the charge stays put and in an inductor the current is constantly flowing.


It's not a capacitor. It's a coil.


Sounds like a bomb.


Is it a bomb?


I don't know.


It just grinds my gears when the response to a simple question, that is meant to be addressed to a layman, is peppered with acronyms and complicated diagrams/graphs.


Imagine how laypeople feel when they hear IT people talk.


Even inside tech we have a tendency to artificially inflate complexity: if only to self-aggrandize, increase job security, and/or source more capital.





Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: