Microsoft-backed Veir is bringing superconductors to data centers
1–10 of 12 posts
Re: Microsoft-backed Veir is bringing superconductors to data centers
#2[flagged]
Re: Microsoft-backed Veir is bringing superconductors to data centers
#3[flagged]
Since my work is vaguely related to superconductors, I saw this comment and was excited to dig into all the errors in the article, but actually couldn't find any in the parts discussing the superconductors specifically. (I don't know data centers and can't comment on that bit.) 77 K is indeed an appropriate temperature for LN2 coolant for high-temperature superconductors like they're using. What errors did you see?
Re: Microsoft-backed Veir is bringing superconductors to data centers
#4[flagged]
Since my work is vaguely related to superconductors, I saw this comment and was excited to dig into all the errors in the article, but actually couldn't find any in the parts discussing the superconductors specifically. (I don't know data centers and can't comment on that bit.) 77 K is indeed an appropriate temperature for LN2 coolant for high-temperature superconductors like they're using. What errors did you see?
Well the third paragraph implies that "low-voltage" is a factor against having lots of heat and size, when the opposite is true.
Otherwise nothing pops out to me.
Re: Microsoft-backed Veir is bringing superconductors to data centers
#5[flagged]
Since my work is vaguely related to superconductors, I saw this comment and was excited to dig into all the errors in the article, but actually couldn't find any in the parts discussing the superconductors specifically. (I don't know data centers and can't comment on that bit.) 77 K is indeed an appropriate temperature for LN2 coolant for high-temperature superconductors like they're using. What errors did you see?
The very first sentence is confusing. "Power demands of data centers have grown from tens to 200 kilowatts in just a few years". I assume they're talking a single rack here, not "power demand of data centers".
Re: Microsoft-backed Veir is bringing superconductors to data centers
#6I casually research this topic once in a while, and so far, have not seen high temperature superconductors emerge from either the laboratory or short haul pilot projects. Is this technology close to widespread commercial use?
Re: Microsoft-backed Veir is bringing superconductors to data centers
#7I casually research this topic once in a while, and so far, have not seen high temperature superconductors emerge from either the laboratory or short haul pilot projects. Is this technology close to widespread commercial use?
The technology is widespread, you can buy spools of wires and wind your own coils. A spool might still be in the range of 3k$+, probably due to the manufacturing costs of all those single wire strands.
However, as we already know, the superconductivity currently starts only at low temps and I think the wires are limited by a maximum current (max field density?)
If you handle the spool, NEVER lose the beginning of the wire, always fixate it, otherwise there might be a knots after some unspooling, then you might need to buy a new one.
Re: Microsoft-backed Veir is bringing superconductors to data centers
#8Dumb fact: the layer of Uranus’s atmosphere where it’s 1 bar is also around 75 K, the temperature high-temperature superconductors like to live at.
Re: Microsoft-backed Veir is bringing superconductors to data centers
#9I casually research this topic once in a while, and so far, have not seen high temperature superconductors emerge from either the laboratory or short haul pilot projects. Is this technology close to widespread commercial use?
High temperature super conductors is a term of art. It does not mean what it sounds like. They still have to be pretty cold.
Re: Microsoft-backed Veir is bringing superconductors to data centers
#10I casually research this topic once in a while, and so far, have not seen high temperature superconductors emerge from either the laboratory or short haul pilot projects. Is this technology close to widespread commercial use?
I'm not sure what your criteria are, but high-Tc superconductors are used in a bunch of commercial applications. Magnet windings, very high-Q RF filters, and some exotic switching applications. "High temperature" still means liquid-nitrogen cooled but compared to liquid helium that's swelteringly hot.