Microsoft signs a power purchase contract for fusion power in five years

  • Active since 1995, Hearth.com is THE place on the internet for free information and advice about wood stoves, pellet stoves and other energy saving equipment.

    We strive to provide opinions, articles, discussions and history related to Hearth Products and in a more general sense, energy issues.

    We promote the EFFICIENT, RESPONSIBLE, CLEAN and SAFE use of all fuels, whether renewable or fossil.

peakbagger

Minister of Fire
Jul 11, 2008
8,845
Northern NH

I bet there is some interesting fine print in that contract. I was looking for the article date to be 4/1

The ability to make Helium 3 is also interesting, helium 3 is the McGuffin of many a science fiction story and the reason cited many times as a reason to colonize the moon to mine Helium 3.as its claimed to be far more abundant.
 
My financial advisor once told me that the most certain way to lose money would be to invest in the energy tech companies that Gates invested in.
 
  • Haha
  • Like
Reactions: Ashful and woodgeek
But this is not Bill, it is Microsoft...

I think the He-3 issue is big. It is produced in nuclear reactors, but in such small quantities, and its abundance is so low that it costs more than $1000 per gram. If I remember correctly, a 1000 MW fusion plant would need about 100 kg per year, aka 100 million dollars.

Their production better be low in energy cost lest it eat up all the power (so far fictionally) generated by its fusion with deuterium...
 
Data centers use lots of power. Azure ( MS owns them ) is the second largest cloud computing vendor behind AWS. My guess is, they want to use fusion so that they can brag about using “clean energy” to power their DC’s.
 
Data centers use lots of power. Azure ( MS owns them ) is the second largest cloud computing vendor behind AWS. My guess is, they want to use fusion so that they can brag about using “clean energy” to power their DC’s.

We've had more than a few renewable energy projects built in the last few years where most of their electricity is being sold on contract to these large data centers.

On the electricity market, energy generated from renewable sources sells for about double that of electricity generated from fossil fuels.
 
  • Like
Reactions: sloeffle
We've had more than a few renewable energy projects built in the last few years where most of their electricity is being sold on contract to these large data centers.

On the electricity market, energy generated from renewable sources sells for about double that of electricity generated from fossil fuels.
Yes, I'm sure Microsoft's interest is cloud related. Unfortunately, a truly obscene amount of this power is going to cryptocurrency. According to this report it is more than all US refrigerators, combined and more power consumed than countries like Norway and Ukraine.
 
Yes, I'm sure Microsoft's interest is cloud related. Unfortunately, a truly obscene amount of this power is going to cryptocurrency. According to this report it is more than all US refrigerators, combined and more power consumed than countries like Norway and Ukraine.
I don’t get cryptocurrency. I think it’s dumb. Fusion is cool. Fusion powered cryptocurrency won’t happen any time soon. Even if MS has a contract.
 
  • Like
Reactions: stoveliker
There are crypto mining operations in a box that are located where power is the cheapest. There is a lot of well gas (methane) being vented all over the world at remote locations. Some cryptominers install generators that run off gas that would be flared anyhow and set the cryptomine in box next to the generator. They just churn 24/7,
 
  • Like
Reactions: sloeffle
I am still quite pessimistic for using fusion for stationary power generation on the earth (but might be good for high Isp rockets in the far future). Only bc power on Earth is going to be plentiful and dirt cheap by the time this stuff gets to working.

I also think NONE of the current fusion startups are going to get to net power generation, after looking at them, they are all old ideas relabeled by snake oil salesmen.

I do think ITER has a decent shot of getting there (high enough Q_plasma to allow theoretical Q_total>1 in the next plant) in a decade or so. But no idea if that >$10B plant will ever be built (bc it will not be profitable in $$ even if it is profitable in Watts).

The He3 fuel cycle is a loser compared the D-T, which is higher power at lower temps, and BOTH cycles really make neutrons (since D-D will occur as a side reaction in the He3 cycle at reasonable power densities) and require a breeder blanket.

Good to see that MS management is still as clueless as ever. We should invest in something else.

ETA: I wouldn't be surprised if some quantum computing breakthrough in the next 10 years ends up devaluing all crypto.
 
  • Like
Reactions: begreen and EbS-P