Hacker News Reader

Comment by dsq | Hacker News Reader

For a non physicist, what in room temp superconductors would be the holy grail? That is, what would open up that is currently off limits?

Is it “only” better transport of electricity, or would completely new things become possible?

View on HN

MRIs or anything with very high magnetic field requirements become a lot easier to build once you don't need to cool them down as much. Inversely, Generators etc. would benefit.

Magnetic levitation (Meissner-Ochsenfeld) would also not require such low temperatures.

Even Quantum computing using superconducting qubits might become easier (although there, superconductivity is not the main reason for low temperatures).

"Only" better conductivity is a big deal, after all.

Along the same line of thought, but flywheel energy storage would become quite appealing when the rotating mass can maintain stable levitation without any energy input (in the form of electromagnets or refrigeration as needed today).

better electric motors in general, they can be smaller when they have less waste heat to worry about, and higher torques. The navy was playing around with miniaturization of vessel motors more than a decade ago[0] but I haven't run into how that projects gone in the intervening years.

[0]https://www.powermag.com/superconductor-motor-for-navy-passe...

Flywheels main problem is already just energy density, they're more similar to a capacitor than a battery. There are situations where they would certainly be better, but as it is for the situations where they are useful I don't think the very low rate of power loss the best ones currently have is that big a deal.

What form would magnetic levitation take? Could we have silently cars, floating houses, flameless rockets, etc?

Maglev trains and the like.

It would be nice if MRIs weren't so fucking loud. It's like a Chemical Brothers concert inside one of those things.

Yes although I find it strangely comforting. I have fallen asleep inside an MRI machine more than once.

I remember that listening to all the harmonics was very interesting. I thought I could hear the vertical and horizontal refresh frequencies. Maybe I was imagining that though.

Just a note that even if falling short of room temperature, liquid nitrogen (77 K) is a lot cheaper and easier to work with than helium, and is not a scarce resource. A lot of useful technologies would open up if we get past the 77 K threshold, at manageable pressure.

huh? we have LN2 temperature superconductors. In this clip I show flux pinning in YBCO, cooled by LN2 on my dining room table: http://nt4tn.net/random/superconductor.mp4

There are many limits in current HTS-- like it's hard to build coils out of brittle ceramics-- so having alternatives would be very useful, so perhaps that's what you meant?

Yes. Practical magnets. But still, I'll check out your link at intermission. ;-)

I wonder how much RTS research could be expected to produce better materials for making electromagnets.

Even Niobium-tin only gets there via clever manufacturing techniques (e.g. the internal-tin process; https://en.wikipedia.org/wiki/Niobium%E2%80%93tin#Niobium_ti... ).

Is ductile HTS a major research target too, or even believed to be possible? Or is the hope that by studying RTS' we'll stumble into something that can be manufactured by chance or gain some deeper understanding of their behavior that let us design more useful materials? It seems the media spills the most ink on room temp but I agree than LN2 temp but able to be formed would be more important-- I think if YBCO worked at -70C it still would not see that much commercial use.

No expert by any means, but SQUIDs[1] rely on superconductivity to detect tiny magnetic fields. One use is for recording neuron activity[2], so perhaps room-temperature superconductors could allow for a portable setup.

This article[3] goes into some of the advantages of a less restrictive setup, but the laser-pumped sensors they rely on still require a lot of additional hardware.

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

[2]: https://en.wikipedia.org/wiki/Magnetoencephalography

[3]: https://hebergement.universite-paris-saclay.fr/supraconducti...

I'd love to have a SQUID of my own to directly detect the |A| field that physics courses usually describe as "merely an abstraction" (Electromagnetic Vector Potential). I have some experiments I'd like to try.

You might enjoy the [Aharonov-Bohm effect](https://en.wikipedia.org/wiki/Aharonov%E2%80%93Bohm_effect). Quantum mechanics says we can in fact detect `A`. A similar effect exists for `phi`.

One application could be to join up the world's electricity grids, though I suspect that our land-mass politics could well be a barrier to that. It could smooth the supply/demand requirements between the dayside and the nightside of the earth.

It could likely be done with ordinary cables, but the power loss would be significant.

As a robotics engineer I would love the tiny and extremely powerful motors it would allow. Also I suppose you would not need gearboxes (or the need for them would be reduced) which would lower cost and mass even more.

Not as much "off limits" but "horribly expensive because of cooling requirements"

I'm also not an expert by any means, but wouldn't this be used for free computation? like a CPU that can compute without consuming any power?

if anyone knows better I would like to know why or why not this is an possibility

>like a CPU that can compute without consuming any power?

Erasing bits consumes power. You'll probably want to start investigating conservative logic:

>Conservative logic is a comprehensive model of computation which explicitly reflects a number of fundamental principles of physics, such as the reversibility of the dynamical laws and the conservation of certain additive quantities (among which energy plays a distinguished role). Because it more closely mirrors physics than traditional models of computation, conservative logic is in a better position to provide indications concerning the realization of high-performance computing systems, i.e., of systems that make very efficient use of the "computing resources" actually offered by nature. In particular, conservative logic shows that it is ideally possible to build sequential circuits with zero internal power dissipation

https://www.cs.princeton.edu/courses/archive/fall06/cos576/p...

...and also reversible computation:

https://spectrum.ieee.org/the-future-of-computing-depends-on...

We'd finally have laptops that don't overheat (well, not heat at all) and are not constrained by thermal limits! Imagine putting a desktop CPU into your laptop! (Battery lifetime issues aside)

Not so fast, just because something becomes available doesn't mean it will be immediately used for everything it _could_ be used for. Cost remains a factor.

I very much doubt room temperature superconductors would ever be used in consumer electronics. They probably wouldn't even be widely used in data centers outside of truly cutting edge "no expense spared" systems.

Sure but that's true for many (if not all) things that superconductors might (in theory) enable. I do think, though, that especially in computing there'd be enough economic incentive to use superconductors everywhere if the energy savings are substantial enough.

No heat means no power usage (unless we also find a way around the laws of thermodynamics), so no superconductors aren't going to make your laptop magically not heat up. It could help a bit with power lines going to transistors, but to do actual computions we need to consume power and thus generate heat.

You're right, I was being overly sloppy with my wording here. I was only referring to the uniform electric resistance part of energy losses that would be gone. My understanding had always been that the latters makes up a significant portion of the total energy CPUs consume because we're still many orders of magnitude away from the Landauer limit, https://en.m.wikipedia.org/wiki/Landauer%27s_principle .

To get there we wouldn't just need a superconductor, but a supersemiconductor that could switch instantly between infinite and zero resistance. I don't think we're anywhere near that, let alone one at room temperature.

From the top my head, most power is dissipated in switching, then leakage (for that we would actually need higher resistance in the transistors) and only then resistive losses in powerlines.

> From the top my head, most power is dissipated in switching, then leakage (for that we would actually need higher resistance in the transistors) and only then resistive losses in powerlines.

Interesting, I stand corrected then!

Most of that heat is generated in the silicon, not the wiring. Unless we discover a semiconducting superconductor with which to replace the silicon junctions, we'll still have plenty of heat dissipation in the CPU.

The article mentions resistance. Things are needed to overcome resistance, and that introduces wear and tear. Magnetic levitation would likely alleviate those problems.

Example: You could install solar panels to Sahara and Australia and enjoy all-yea-round 24h solar energy in Germany.

Maglev trains might change from prohibitively expensive to reasonably affordable.

Fusion power would basically be solved.

That's interesting. I didn't know that whether fusion power would be available depends on room temperature superconductors. Is there a simple explanation for that?

The most common and promising fusion experiments today (JET, ITER, Wendelstein-7X, CFS/MIT SPARC) are magnetic confinement fusion: use big, powerful magnets to shape and compress a hot deuterium-tritium plasma until fusion happens. In the Sun, gravity does the job, but at an extremely low power density; on Earth, we need to use magnets. These electromagnets draw a lot of current.

The big problem in fusion is containing and shaping the plasma without any instabilities. If we had room-temperature superconductors, the smallest current through a superconducting magnet would generate a very large magnetic field, enough to shape any plasma.

That’s awesome. Not OP, but thanks for the TIL.

Reply on news.ycombinator.com