This data center will be Europe’s first with hydrogen backup power Dutch Data centre biz NorthC is replacing its backup power generators at the company's facility in Groningen to run on green hydrogen, saying it's a European first for data center design. The H2-powered 500KW hydrogen fuel cell module will run on hydrogen created by renewable power and is intended to reduce consumption of …
That would be equivalent to 0.00028% of the car emissions (considering there are about 8500000 cars in the NL). That is still an extremely small change and will have no real measurable effect. The 500kW power generation for the datacenter would be equivalent to about 263000 kg of CO2 per year (at 60g/kWh). That would calculate to a 30% reduction for this particular datacenter if the hydrogen system would be a zero sum system and the calculation would be this easy (which it is not). The actual cost is the datacenter itself, its content/hardware and the permanent power usage.
As commendable I find these small changes, I do not think we can continue with these small changes. This particular datacenter may attract 10..30% more customers/capacity with all the attention and its power usage will increase. Therefore, the reduction is compensated by higher usage and we are no better off than before. We need to break the pattern to do better.
You need a lot more than small changes at 8.5 tons CO2 emissions per person per year in the Netherlands.
You are right, many small changes can add up. But we need to reduce at least 50% in the next 10..20 years. We need to be at zero emissions in 30 years and negative emissions in 40 years. So, just reducing at small scale is a start, but not a solution.
How would you half the datacenter emissions within the next 20 years? Halving in 20 years means 3.4% reduction per year for the next 20 years, each year.
I'd halve DC 'emissions' by increasing the cost of energy. Seems like politicians agree because that's current energy policy.
Rest is politics as well. So the 'need' for Net Zero is political goal seeking. 2050 is a convenient marketing deadline, and the 'need' to hit that adds time pressure, which increases costs. The science doesn't demand 2050, because the science is based on CO2 accumulation in atmosphere and what that might mean for radiative transfer.
That theory is plugged into climate models, and then used to scare the gullible. This policy has been very successful over the last few decades. What isn't talked about as much is models vs reality, which generally show the models overestimate CO2's effects. But modelling produces projections, not predictions. This is convenient because predictions are easily falsified, and that's just not part of climate 'science'.
There are also problems with scale. This will save 78t of CO2 a year. In context, natural carbon cycle is around 210 gigatonnes a year. And climate dogma requires it's followers to believe in a mythical 'equilibrium' where the homeopathic amount of CO2 added by datacentres will make the world burn. Which also leads to inconsistent policies. So stuff grows, dies, decays releasing CO2 and CH4. So the EU decides wetlands restoration is a good thing, even if that increases carbon emissions.
A more modest proposal would be to orbit a few MASERs that could sterilise km^2 of land and stop soil microbes from killing school kids.
> How would you half the datacenter emissions within the next 20 years?
Unlike a politician, I'll give you an answer... ;-)
I've always been pro nuclear but clearly we've left it too late for nuclear to be a solution in a 20 year time frame. Maybe shutting down Facebook would save enough? Make the web text only for a period of time?
I would say hydrogen is the obvious only easy step - renewable to hydrogen to fuel cell. It's inefficient but it works. And that buys time for a better solution later. Either that or we're planting fuck tons worth of trees, chopping them down and burying them in old mineshafts, hoping against hope that they don't spontaneously combust.
What we mustn't do is be trapped into the line of thought that any given solution won't solve the problem therefore we should keep looking for others but there seems to be quite a lot of that happening already.
And a variation on the above, is that we mustn't allow ourselves to be trapped by bean counters: it's fine to build a reactor now even if it ends up being shutdown ahead of its potential maximum lifetime and is therefore somehow 'too expensive'. It's not too expensive if it buys time.
It's not a perfect answer - but if I could think of that then someone else would have already.
Oh, are you prepared to pay a *cost* to meet your declared goal? The following suggestions will *work*, they are all independent, let’s see what you are really prepared to *pay*…..
#1 Centralise all email to a single state-sanctioned provider. The purpose is to minimise the transmission cost. There is on-site caching; the absolute central authority primarily holds pointers as to where the data is, not the data itself. As a sub-goal, no email is ever really “sent”….it gets uploaded once, and downloaded once. But mostly, all the email threads where the same documents are attached again and again, are all de-duplicated once in a central location
#2 Citizens are given a data-budget. They have to live within their means. Instead of taking 100 photos a day, leaving them all on their iPhone to be archived by iCloud forever never to be looked at again, it’s a civic duty to check through the ones you want to keep more than 90 days, and you are allowed to archive no more than 10 per day.
#3 From an energy-use perspective, it’s just insane to individually stream movies. I know you enjoy Netflix, but stop it please. For a very small price indeed, we can divide the energy use by x100. Do this via round-robin central broadcast, over the radio spectrum. It doesn’t need to be like the old days, you can still have a roster of 1000 films. But broadcast round-robin on a single 100MHz wide multiplex, they can all come round once per day. You can’t decide *one day ahead* which movie you want to watch, to save maybe 1% of the total energy usage of planet? Screw you.
#4 Ah, this is an easy one. Outlaw all crypto. Nuff said.
Do we need to go on?
Everyone knows how to save the planet *perfectly well*.
I'll take example my teacher used about littering.
"It's only a crisp wrapper, what's the problem? "
"It's not a problem, it's just when 55 million people have the same attitude as you"
An individual makes little difference, but
a thousand or a million of individuals making the same difference has a huge effect.
At least they are using a fuel cell rather than switching to some sort of hydrogen ICE.
This may only be a tiny dent in thing but as a concept it has to be better. Yes, they hydrogen may not be entirely green to start with however will all these things we have to start somewhere.
Hydrogen may be inefficient however if you and goal is zero CO2 emissions, it is still better than fossil fuels. It can also be stored and overall if probably far less damaging to the environment than battery technologies.
The problem with hydrogen storage is that its critical point is -240C, so it can't be liquefied at normal working temperatures, and that in turn means that you can only get reasonable mass and therefore reasonable energy density is to compress it ii lot. Typically 700 bar. Which in turn means heavy and expensive infrastructure.
Solid state storage, like manganese hydrides, gets re-invented every few years (I am pretty sure I saw Raymond Baxter wax lyrical about it on Tomorrow's Word) and may be ready about the same time as commercial fusion power.
It does seem odd to choose hydrogen as an energy transport for backup applications, given the difficulty and expense of storing it. I can see benefits in using it for daily power, but keeping tanks of it around for years "just in case" doesn't seem efficient. A liquid fuel like (bio) diesel would be more practical.
Will we be reading an On Call article 10 years from now, about the DR failover that didn't happen because someone forgot to refill the H2 tanks?
This post has been deleted by its author
I would assume that testing requirements would still remain in place (as it's primarily the testing they are trying to clean up by using H2), so hopefully someone would notice if the tanks were getting low on H2.
Obviously won't help if they don't actually bother testing!
The problem with hydrogen is that it’s neither energy dense nor safe to transport. Given that the only way to make it (by definition) is electrolysis of water, focusing on the fuel completely misses the point. It’s not only a *small* effect, it’s *no* effect on Greenhouse emissions. The only sensible way to do this *for Greenhouse* is:
1) Synthesise $WorkingFluid using renewable electricity, from raw environmental materials, as an energy storage medium. Usually at a central location.
2) Transport $WorkingFluid
3) “Burn” $WorkingFluid where it’s needed.
Chemically, these are always “Reduction”, and “Oxidation”, which technically mean adding and subtracting electrons. Oxidation doesnt absolutely require Oxygen. But in practice on Earth it’s dumb not too. Given that #1 and #3 are the exact inverse, from the point of total CO2 emissions alone, it absolutely doesn’t matter what $WorkingFluid is. It can contain as much carbon as you like, just means you are absorbing CO2 in step #1. It’s an engineering choice, not an environmental one.
The engineering requirements on $WorkingFluid are: Energy Dense, Not Explosive to transport, non-toxic oxidation products. Note how Hydrogen is a *terrible* working fluid. Actually diesel is *perfectly fine* from CO2 perspective . Petrol (aka hexane/octane) is better. The thing about that, is that you can transition to a renewable economy without incurring *any of the costs* of changing your consumer-side gear. There’s nothing at all wrong with on-site generators using diesel or petrol. Sorry, there just isn’t. It *looks* bad, but that’s all. We would need to use *pure* without sulphur or nitrogen (see non-toxic products), but that’s a relatively minor point.
And yet, that pointless transition is what the politicians have been pushed to spend their entire Green budget on. The technically *necessary* part, synthesising a good $WorkingFluid from renewable electricity, we’ve spent precisely zero on. Genius.
It’s almost as if leaving engineering problems in the hands of pressure groups driven by technical illiterates, is counter-productive, and produces “answers” with precisely the opposite effects than they would have wanted, if they had really understood the problem. Which they don’t. And they will shout down anyone who tries to actually make progress, because it is far more important to them to be right than to achieve the outcome they say they want.
Sorry…
The only way to make it *which is in any shape or form green or renewable”. I agree that is how it *is* made in practice. Which is environmental lunacy. Hydrogen is significantly carbon *positive* end-to-end compared to a diesel generator, even if you ignore transmission losses. It manages to emit CO2 as a by-product of its feedstock, even before you account for the use of fossil-fuel (more methane) to heat the steam. I was at least giving hydrogen the credit of “it could theoretically be no more environmentally disastrous than diesel if we did this properly….although currently it is worse”
There is one sense in which hydrogen manufacture is non-disastrous. A little appreciated fact by the Greens is that hydrocarbons don’t come out of the ground pre-packaged into “things we like” and “things we hate”. You dig the hole, a mush of different things comes out in random proportions, and you either find a way to use exactly the random proportions that come out, or you just spill the unused fractions to poison fish. In the case of methane, the flares of methane being burnt off at well-heads as waste, can be seen from space to easily outshine the total brightness of all electrical lighting on the planet. An alien telescope pointed at the Earth would conclude that we never discovered the electric light, and still use gas mantle lighting.
And yes, again, I am aware that hydrocarbons are cracked from one form to another to change them to the proportions we actually use. 40% of the total primary energy goes into that process.
So, for example, refusing to use jet fuel on the basis of environmental consequences is yet another example of causing the opposite effect to intended. Jet fuel is kerosene. If you don’t use it as it comes out of the ground, you either spend 40% of it to convert it to something more useful (*increase* net CO2 per energy use). Or, and this is the *vile* kicker, what actually happens: increase its usage in cooking in poorer countries. Kerosene stoves are *appallingly* polluting, inefficient and toxic. More people will die this year from the poor air quality due to their kerosene stoves, than from Covid. About 4 million deaths annually.
Maybe, just maybe, someone could focus some Green funding on sorting *that* problem out? No. Thought not.
.
The problem is how it is made.
If you're burning coal to transform H2O into hydrogen/oxygen, then you're not getting the benefits of hydrogen because you've burned coal.
The only way hydrogen can be justified is with nuclear/solar/wind energy.
We have to shut down the coal-based power stations. All of them. Of course, we can't do that in one go right away, we first have to get ITER to work and, concurrently, get Thorium stations installed everywhere.
Between the two, things just might work out.
The thorium cycle isn't just a scifi fantasy. There was a successful trial with one built in the 1960's by Oak Ridge National Laboratories. The prototype was successful and did produce energy, but the whole "cannot be used to make weapons grade nuclear material" (which the US DoD wanted at the time) meant it never received funding to go past the prototype phase.
Now days it's just inertia and the fallacy of sunk costs that keeps the uranium based reactors as the dominant type. Even though weapons grade nuclear material is no longer desired too many companies have invested too much into uranium to change direction. Companies without uranium reactors and governments are investing in fusion instead. From a scientific point of view there's still plenty of unknowns in that area and people can make a name for themselves and get their names in the history books
@Pascal Monett
"We have to shut down the coal-based power stations. All of them. Of course, we can't do that in one go right away"
Why do we have to shut down coal anyway? Plenty more coal plants are being planned to be built worldwide. Its a cheap and stable technology that works.
@druck
"Coal plants around the world are burning dirty low quality high CO2 lignite coal"
And Germany the glorious land of the green shut down their nukes and fell back on such coal generation too!
"climate change activists want it to say that way."
They want to relearn construction of mud huts the lunatics.
They can add that capability down the road. Even if they use grid electric to make it, it might be made with renewable energy depending on their utility. They might even take advantage of some sort of a special tariff to use power when the utility has an excess due to peak renewable energy production on a sunny & windy when the temperature is moderate so most customers aren't heating or cooling buildings, since they don't need to have a schedule for making hydrogen.
It is a step along the way, but even if they are using grid electricity generated by natural gas it will result in less CO2 than an on-site diesel generator. If you wait for the perfect solution (like your dream of thorium reactors which aren't in commercial use ANYWHERE yet, let alone everywhere) you will be waiting a long long time.
Of *course* we can shut down coal right away.
This may surprise you, but 75% of all *global* CO2 emissions from coal come from just 5 power stations, which produce only 5% of the power produced by coal. Most of the problem comes from just a few *very inefficient* coal stations. Not that we shouldn’t shut them all down, but just the top 5 would get you most of the gains. I’ll give you the names of three of those top five:
Belchatow, yes that’s really its name. Poland
Kozienice, yes that’s really it’s name. Poland
Neurath, Germany.
Shut down those three, that’s 50% of global coal CO2 right there.
You may spot the common strand….I just can’t be arsed any more.
This is a very interesting first step. It's more than a proof of concept, it shows it works in the real world.
Storing electricity is not easy, but storing electricity as hydrogen is much easier, can be made locally and reused at the same place.
It would interesting to have more details on how NorthC produces (or gets) the H2
The UK produced more than 50% or its electricity through "green generation" last year (or the year before?) Yet, increasing the wholesale price of fossil fuels causes the cost of 'green' electricity to the customer to go through the roof ... Either companies producing green electricity do so at a loss or minimal profit under normal conditions (neither of which seem to be the case based of financial reports) or they are massively profiteering on the current tidal wave of of tax-payer subsidies and a state-sponsored, monopolistic market.
"Green" is the new "oil" - just watch how many politicians are elected to boards over the next few years ...
"It would interesting to have more details on how NorthC produces (or gets) the hydrogen
Yes. Wouldn't it just. My guess is that it will be "green" roughly until the first or second tank full is exhausted, and then it will quietly be obtained from the "grey" market and hence no better than the diesel it's replacing - especially after taking into account the cost of the kit.
This whole thing is an exercise in greenwash - pure and simple.
Why go for "number of kilometers a Dutch person drives per day" when you can go for "how fast a Dutch person can cycle in a proper headwind"?. That is vaguely measured on an annual basis.
I know the answer is, er, blowing in the wind..
:)
That would be a speed measurement and we already have sheep.
But ElReg definitely needs a longer length descriptor since the longest is one fatburg which is only 64 meters.
At 32 km the Dutch motorist seems reasonable since it has an official source.
But in a nod to Good Omens I would suggest the 188 km M25 Car Park.
Storing electricity as hydrogen is generally a monumentally stupid and inefficient thing to do. The electricity (electrolysis) hydrogen (fuel cell) electricity round trip is at best 50% efficient, and that's without including the significant energy costs of compressing it for storage.
The only time it might possibly be justified is when you need mobile power - on trains, for example - but even then batteries are better in almost every respect. In this case even that excuse fails. Greenwashing at its finest.
You think a *500kW* hydrogen fuel cell is better than batteries?
A Tesla P100 can stack out maximum battery power 600kW. The peak power requirement can be served from something that fits in the boot of a car.
*Hydrogen* however……Well, here’s a state-of-the-art hydrogen fuel cell from Panasonic
https://news.panasonic.com/global/press/data/2021/10/en211001-4/en211001-4.html
It’s…..5kW, and is exactly designed for the sort of commercial uninterruptible power supply being discussed. It weighs 200kg, and although it’s 0.6m3 in size, when you look at the clearances specified it’s basically 4m3. Scaled up to 500kW, this is 20 tonnes, and takes up a 14x10 meter fenced compound (they have pictures).
Fuel-wise…..maybe you want 2hours @ 500kw = 1000kWhr?
I won’t go through the calculation, but the long and short of it is that the 1000kWh battery still fits in the Tesla if you chuck the occupants out, whereas the hydrogen requires a 20m3 tank at 700psi.
Overall, it’s battery: small caravan trailer, versus fuel cell: substantial part of site real estate.
The current norm for a diesel generator of this capacity is roughly one standard Container.
The real point is that “the direction of travel” (whether it’s sensible or not), is that the 70kWh battery for the average family car is going to have to come down to less than £2k in price. So a 1MWhr backup generator would cost £28k, which is maybe double a current diesel generator, but not outrageous when you consider maintenance and running checks. Basically, the consumer automotive requirement is going to force batteries to undercut every other solution. It’s a solved problem.
Most hydrogen is produced as a byproduct of gas production, as a byproduct of oil extraction.
Also, how much more volatile is hydrogen than diesel to store, long term? Much more irritating to handle, I'd wager.
Diesel is fine as a fuel for a backup generator, it's not regularly used anyway. If it is commonly used, then there's a general issue in that country.
Diesel is fine as a fuel for a backup generator, it's not regularly used anyway. If it is commonly used, then there's a general issue in that country.
And in any case you can run your backup generator on 100% bio-diesel (been there, done that).
It's rather difficult to follow what savings they claim will result from what, but my best guess is that if every data centre in Europe went over to hydrogen, they'd save a collective total of 78t CO2 emission per yer as a result of not having to fire up their generators to test them. Since the EU produces around 2.5bn tonnes of CO2 per year, this would be a reduction of around 0.003%.
Whoop de friggin' doo.
Hydrogen is an incredibly light gas and disperses upwards into the atmosphere very quickly, in fact too quickly for combustion to occur.
There are crash tests carried out on hydrogen fuel cell car tanks where they do not explode because of the speed with which the hydrogen disperses.
In the Hindenburg accident it is not clear whether the hydrogen was the main fuel of the fire or the fuselage itself. Hydrogen fire doesn't produce a visible flame (see the Space Shuttle engines), so the intense flames in the Hindenburg footage would have been caused by the fuselage.
In other words, hydrogen is incredibly safe – we understand its properties very well and have been safely using it in the fertiliser industry for decades
So easy to be negative. Of course it's new, inefficient and not set up yet to be compared to all existing tech. As was the car in its early days, not handling the horse tracks very well with all the not existing infrastructure.
You have to start somewhere. All I have to say...
You aren’t understanding the difference between limitations of physics vs engineering.
Your argument is the one used by every inventor of a Perpetual motion machine. Look, I start it up and it goes for the whole length of this video without visibly slowing down! All it needs is a bit of scaling up and free energy!
The hydrogen economy is *exactly* a perpetual motion machine.
My home in Herne Bay, Kent is surrounded by off shore wind turbines and solar farms where farmers' fields used to be. And very soon, a 'green hydrogen' factory a few metres away from our estate.
There are already some concerns about safety - leaks, detectability of leaks, and safe transport.
But what I find strange is how the process is that green at all. Solar and Wind energy, to electric, to make hydrogen, which is then transported by trucks to London (some 60 miles west), to fill 20 London Buses.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
