Google goes geothermal to power some bitbarns Google has added a novel form of geothermal power to the list of carbon-free energy sources fueling its data-crunching empire. This week the ads and search behemoth, working in collaboration with startup Fervo Energy, brought a "first-of-its-kind" enhanced geothermal power plant online in Nevada to supplement more traditional …
And more importantly: stuff you use continuously.
Outsourcing is more beneficial the more intermittent or fluctuating your demand is.
The more it fluctuates, the better it is to share with others to smooth out demand, divide up the capex (which is set by peak demand) and spread the risks.
Contract manufacturing for example, is largely a product of this.
"Same reason that a crypto mining firm bought an old coal plant. The more you have under your control the less susceptible you are to outside forces. "
If the mining operation is on the grounds of the power plant, transmission losses are next to nothing. That increase in efficiency can mean profits by not paying 'the grid' for power where they have to take those losses into their pricing calculations.
You already know the answer to this from the stateside perspective. Any organisation of moderate size or above that values uptime will have it's own generation in preference to reliance on the networks, which offer little to nothing in the way of guarantees.
European networks typically incorporate more redundancy but they are more expensive per kWh delivered. They do have the benefit of lower thermal loads and so tend to last a lot longer with all the equipment in parallel.
Some types of customers the cost of downtime is such that the cost of adding a generator even if used only once every 20 years for 20 mins the cost-benefit still favours buying the genny regardless of which side of the pond you reside.
If you want the worst of both worlds the UK has you covered; we have shortfalls in capabilities AND sky high bills, backed up by a clueless BEIS/OFGEM/EA triumvirate of conflicting objectives such that very little gets done to address the causes.
"If you want the worst of both worlds the UK has you covered; we have shortfalls in capabilities AND sky high bills, backed up by a clueless BEIS/OFGEM/EA triumvirate of conflicting objectives such that very little gets done to address the causes."
Minor correction that BEIS is now DESNZ (colloquially Des-nes), and EA involvement in energy is negligible. Both Ofgem and EA are both administrative quangos, so whilst they might be useless and Ofgem are useless, neither hold policy responsibility. Defra get consulted on environmental matters by DESNZ. But the overall assessment, yes we have a system that doesn't work well for what it now has to do, is poorly designed as a starting place for the magical world of Net Zero, and has created vastly expensive energy costs due to a poor system design and an over-reliance upon "markets".
Don't forget that you've not mentioned one the biggest contributors to the chaos - an endless stream of poorly qualified fools for ministers. The current post holder failed to even graduate, after studying Philosophy and Law at Cambridge, so is qualified in nothing. The previous post holder had a degree in history, and so it goes on through the twelve makeweights who have occupied the minister's chair since the 2010 election and that number includes "talent" like Hancock and Kwarteng. Lest anybody think Labour were any better, in the five years prior to the 2010 election, they motored through five energy ministers, mostly with similar useless degrees in history, sociology, law, politics.
So ultimately the problem is not the administration of policy (poor thought that often is) but that the politicians don't understand their brief, are poorly educated, don't stay around to see anything through, and always put short term party needs before the needs of the nation.
Oh you English just love to pretend you are world leaders in everything.
I see your minister and raise you NZ's last Minister of Energy and Resources, The R.Hon. Megan Woods PhD (NZ History).
Wasting oxygen that could have been smouldering damp peat to keep our bothys warm.
"I asked the electricity companies if they were ripping the public off , and they said they weren't " (to paraphrase her electricity industry review)
"mostly with similar useless degrees in history, sociology, law, politics."
A doctorate in playground supervision is the gold brick road to government service posts in addition to the similar degrees above.
Too many degrees are in the sort of things that the children of wealthy parents would get. It required them to attend college where they'd be exposed to things, but what they ultimately be doing is research into subjects with no proper jobs in the real world. It wouldn't do for them to get a degree in something very work-a-day. This is not to say that some very informative and enlightened work hasn't been done, it's just that it isn't the sort of work with lots of job openings that pay a living wage. Perfect for some scion that needs something to focus on and keep from besmirching their parent's good name.
"Too many degrees are in the sort of things that the children of wealthy parents would get. "
I believe Politics, Philosophy & Economics is the gold standard of uselessness. Ideally from Oxford, then back to a job that daddy's friends find for you, and thence into politics. That's how we ended up with that toffee nosed pillock Cameron, who gave us Brexit, who ran away, leaving May (geography), Johnson (classics), Truss (PPE) and now Sunak (yet another PPE waster).
"I believe Politics, Philosophy & Economics is the gold standard of uselessness."
I wouldn't go that far, but I would say that a degree in those subjects by themselves is not as useful as they would be paired with something else. If your goal is to gain a government job analyzing and creating policy on fiscal matters, a degree in Economics would be expected. I would also like to see sheepskins in sociology and at least some coursework in science/engineering. Formulating fiscal policy isn't going to be be relevant if it doesn't consider how people react and behave in the real world. One doesn't have to look very hard to find academics with such a narrow focus that they disconnect from reality. In my world, I might use an ideal transistor as a first approximation when designing a circuit, but I have to understand how a transistor deviates from that ideal when I go to build a real device.
DTI/DECC/BEIS/DBEIZ/Whatever they are called this week are, as you say, an outfit that had been subject to rack and ruin by a string of hopeless cabinet ministers.
The lack of well defined authority and weak and conflicting objectives have predictable results : failure for consumers.
It's is doubly infuriating that real expertise exists in this field, casually ignored. The war on experts continues. It's right here on our doorstep. Let's use it!
"Some types of customers the cost of downtime is such that the cost of adding a generator even if used only once every 20 years for 20 mins the cost-benefit still favours buying the genny regardless of which side of the pond you reside."
I had a girlfriend that worked for an automaker and she explained why it's cheaper to pay for a private jet to deliver parts to the factory than to wait for a truck to show up. The down-time on the line was so expensive that hiring a private jet, while expensive, was much less than shutting down.
It's a common thread in risk analysis. It doesn't matter if it's adding backups to things or deciding whether it's a good idea to outsource. Some MBA's don't consider the consequences of outsourcing if the vendor fails to deliver. If you're an eTailer and use an outside service for all of your computing, will you go out of business if that service goes down and people can't click "buy" for a day or two during the busiest holiday shopping period? Are you a big enough customer of that vendor that they will even take your call in the event of an outage?
"Can anyone think why big companies are looking to build their own power generation instead of relying on the grids? "
Reasons vary - but if they can build and connect their own capacity then they can avoid or minimise paying distribution and transmission charges (and greenification policy costs added to the distributions costs) that can be up to a third of the total cost of grid electricity, and they can avoid peak demand pricing on grid connected generation. Peak demand wholesale pricing can be around 3x the lowest system cost, so it's a potentially big win, but a huge amount depends on how much the DC can and will rely on the grid, available generation technologies and land to build them.
But bear in mind that so far DC operators are only playing with generation - for all the puff about renewable powered DCs and small nuclear reactors, the overwhelming majority of DCs are grid connected. Being 100% self-reliant for juice is possible, but it's rarely viable, and there's plenty of places such as most of the UK where it would be impossible.
"Can anyone think why big companies are looking to build their own power generation instead of relying on the grids?"
They get a government grant/credit/tax abatement to build said plant and save money year after year by using something they got for free, or nearly free. As long as they aren't building a coal fired plant and they are big enough to not really need the money, governments are keen to throw money at them.
Being picky, is the heat generated by the Earth's crust or the Earth's core.
I was under the impression that geothermal actually obtains heat that is moving through the crust (the solid rock) having come from the core (or perhaps that should be the mantle).
Also, if the data centre uses water for cooling, is that 'pre-heated' water then being used as what is pumped into the ground?
So far as I know, the heat is thought to be generated by the decay of a variety of radioactive isotopes diffused throughout the planet. I don't think we currently know for sure what isotopes, nor in what quantities, nor or how they are distributed throughout the mass of the planet. Ask again every few decades. Someday you'll get an answer.
While Google's foray into geothermal energy is commendable for its renewable energy initiative, it raises several ecological concerns that seem glossed over in this "happy-clappy" narrative. The disruption of unique subterranean microbial ecosystems, potential thermal pollution, and the increased risk of seismic activity due to fracturing processes are significant issues.
Moreover, the substantial water usage and possible contamination in arid regions are worrying. These factors, combined with the impact on local land use and habitat disruption, suggest that this project might not be as environmentally benign as portrayed.
Sustainable energy solutions should not compromise the very ecosystems they aim to protect.
@elsergiovolador
"While Google's foray into geothermal energy is commendable for its renewable energy initiative, it raises several ecological concerns that seem glossed over in this "happy-clappy" narrative."
The happy clappy narrative glosses over ecological concerns. Energy generation comes with risks and tradeoffs, its the only way to do it.
Are we really supposed to worry about subterranean microbial ecosystems? So you would ban all drilling, for anything - including water - in the name of saving them?
A small scale project like this will tell us whether this is a viable strategy in the long term. Will less heat be extracted over time, or does the surrounding rock move heat efficiently enough that any lost heat is easily replaced? Does a single well cause a statistically significant increase in detectable seismic activity?
Not sure why you think this would use a lot of water? They would recycle the same water through the well, not use fresh water at the inlet and dump it at the outlet.
"Moreover, the substantial water usage and possible contamination in arid regions are worrying. "
In Northern California, a geothermal area ran out of natural supplies of local water and ran pipelines to sewage treatment plants (the output, not input) and used that water to inject into the system. The very high temperatures killed off any nasties still extent.
Any power generation is going to have some sort of effect different than what occurs naturally. The question is one of which method will have the least impact. Can we delete some coal plants by adding some geo-thermal plants and wind up with fewer/less damaging emissions?
"you mean like the wind turbines cluttering up the landscape, eye pollution, and chopping up birds?"
No. Aesthetics isn't "pollution" or Tesla's new pickup would be banned. The issue with birds being killed is a big concern (unless it's sparrows and pigeons. The doves that hang out around my house can piss off too).
I've had to power profile subroutines in the past to keep my power consumption undercontrol on a microcontroller....
Why aren't the middleware suppliers and application programmers checking for power optimizations, or even the compiler suppliers? Would help with this sort of problem if they used power efficient code, middleware, hardware, and operating systems....
...but they are throwing cash at power generation instead? I think I need to retire.
I was really happy to read about this one.
While this Google system is limited because of its depth big companies building these systems will likely lead to further advances. No matter where on the planet you are if you dig deep enough it's HOT. Deep geothermal could be, and should be, a major power source.
It's non polluting, no pesky fuel to worry about, and could even be located at the site of existing plants to take advantage of the already established electrical infrastructure. In some locations deep geothermal could directly replace coal fired boilers and power the same turbines.
This is good stuff!
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
