Apple Powering Nevada Datacenter With Solar Farm 104
Nerval's Lobster writes "Apple's Nevada data center has been in the works for quite some time: a 2,200-acre plot outside of Reno will host a 90,000-square-foot datacenter that, in turn, will support the tech giant's cloud services. Apple will reportedly spend $1 billion over the next decade on the facilities, in return for significant tax abatements at the city, county and state levels. It will also fund and build a 137-acre solar farm, managed in conjunction with NV Energy, to power the datacenter (it will generate approximately 43.5 million kilowatt hours of electricity). The Reno datacenter will be the third Apple cloud facility in the U.S. that is powered largely or entirely by solar power. Sixty percent of the power for Apple's North Carolina datacenter comes from an existing solar-power farm near the facility; an Apple datacenter in Oregon uses solar power for part of its power load, but also uses power from wind and hydroelectric sources."
Wait, what? (Score:5, Funny)
They're powering their iCloud with solar panels? I thought clouds blocked the sun?
Re:Wait, what? (Score:5, Funny)
It's the basic principle of Conservation of Cloud: when there are clouds over the datacenter, the cloud in the data center goes away.
Re:Wait, what? (Score:5, Funny)
Jonny Ive added a translucency effect to the new iClouds, so they should fine.
Re: (Score:2)
They're powering their iCloud with solar panels? I thought clouds blocked the sun?
The spirit of Jobs will continue to shine down upon them and power them. Ye of little belief.
Well yes... (Score:2)
What do you think the sun is doing on top of the clouds exactly....
And it's always been true that solar energy is responsible for clouds. Otherwise how would water evaporate?
Re:Wait, what? (Score:4, Funny)
They're powering their iCloud with solar panels? I thought clouds blocked the sun?
Its Oracle not Sun now.
Re: (Score:1)
I may not understand basic weather but I do understand C++ weather.
Re: (Score:2)
Cool! (Score:2)
Re: (Score:3)
The major downside would be apple.com being down once the sun sets.
Re:Cool! (Score:5, Funny)
Re:Cool! (Score:5, Funny)
Not user-replaceable.
They'll just have to ship their datacenter to an authorized service provider.
Re: (Score:2)
Then again with all the AC needed just to survive in NV it is hard to imagine solar ever outstripping peak daytime demand.
Hoover does not power Reno (Score:1)
Are you kidding me? Have you seen the size of Nevada? Do you know how far Reno is from the Hoover Dam?? 2 seconds on Google would have shown you that Reno is powered by the natural gas fired Frank A. Tracy Generating Station. Furthermore, Clark County has FIVE nat-gas plants AND a coal burner to power Vegas and friends. The vast majority of Hoover power goes to CA and AZ.
Re: (Score:2)
That seems like the definition of "cross purposes" to me. Go figure, them wasting all the transmission losses.
Re: (Score:2)
When they built the Hoover Dam, Las Vegas was a small town and Nevada had a tiny population so the setup was that the majority of the energy goes to Southern California. I think Nevada gets only about 25%. Yes it hardly makes sense now that Vegas is right next door but almost all Vegas power comes from coal and natural gas.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Re: (Score:3)
This problem is either handled by thermal storage or by just simply using other power source (commercial, wind, etc) when there's no sun.
One common approach is liquid sodium. You use an array of moving mirrors to heat sodium being circulated in a tower in the middle of the mirror farm. That circulates with a large, heavily insulated tank of liquid sodium under the facility. It heats up during the day, cools off during the evening, but is always hot enough to boil water to run turbines. (unless you get li
Re: (Score:2)
Except that a data center of this size and reputation can't afford to go down, hence there'll be some big BIG BIIIIG backup generators, and I imagine there'll also be a system (some batteries or big caps) to buffer grid fluctuations. I didn't read the article, but are they planning to run their machines on DC?
Re: (Score:2)
I suspect it goes through an inverter just like everything else.
My computer is hooked up to a UPS, and the UPS is feeding it AC power. I'm pretty sure this is a solved problem.
Re: (Score:1)
Very simple. It takes far more energy (as in burned coal and oil) to make the PV cells, the panels, the frames, the wiring, and the charge controllers than the panel will EVER gain back in its usable lifetime.
Apple could go with the technology of "burning" aluminum to alumina and be better off with less oil/coal burnt, but solar provides a nice illusion of being green with the toy amounts of energy gained, when in reality, it takes at least 2-3 watts of energy to get one watt of solar out.
But it is a "gree
Re: (Score:2)
Very simple. It takes far more energy (as in burned coal and oil) to make the PV cells, the panels, the frames, the wiring, and the charge controllers than the panel will EVER gain back in its usable lifetime.
An AC spouting FUD. Who would have thunk it.
Re: (Score:2, Informative)
I'm a physicist and software engineer, and although I agree that the idea is fantastic, I'm skeptical of the execution. Photovoltaics, as I understand, are economically less viable than concentrated solar power (even Concentrated Photovoltaics, which are more efficient than your run-of-the-mill solar panel, aren't quite there yet), particularly in the form of Solar Power Towers (http://en.wikipedia.org/wiki/Solar_power_tower). I'm not sure what the obsession is with solar panels: they're not only resource-i
Re: (Score:2)
It's simple really. Concentrated solar needs a big tower and an array of mirrors. PV just needs a roof, which every building already has. If the building is designed to last 5 years or more and you can afford the up-front cost the solar PV will always pay off, so why wouldn't you do it?
Re: (Score:2)
If the building is designed to last 5 years or more and you can afford the up-front cost the solar PV will always pay off, so why wouldn't you do it?
The national average home uses 940 kWh/month. The national average for 1 kWh is $0.099. Therefore the national average electricity cost is $93.06/month.
In order to break even after 5 years the cost of the solar power system must be no greater than $5583.60 and at that price it must cover 100% of the homes energy needs.
In what dream world are you living in that such a system can be installed for that price?
Re: (Score:2)
You don't seem to understand basic economics.
Your consumption is reduced when you use energy that is generated. You get paid for energy fed back into the grid when you are not using it. In some places you get paid regardless of if you used it or you fed it back into the grid.
At the moment the pay-back time for a medium sized (say 2-3kW) installation is about 5 years in most developed countries. Some a bit more, some a bit less, depending on the cost of electricity and the feed-in tariff.
Re: (Score:2)
You don't seem to understand basic economics. Your consumption is reduced when you use energy that is generated. You get paid for energy fed back into the grid when you are not using it.
No, you don't seem to understand basic economics.
I realize that you think vague general ideas trumps actual numbers, but in reality they do not.
In effect, you have made it even harder to recover the costs over 5 years because you used more panels than you needed in order to have that surplus electricity that you think is a magic bullet, so your hand waving about it means absolutely nothing to people with a brain because they know that having a surplus costs fucking money.
Do some math, find a supplier
Re: (Score:1)
You don't seem to understand basic economics. Your consumption is reduced when you use energy that is generated. You get paid for energy fed back into the grid when you are not using it.
No, you don't seem to understand basic economics.
I realize that you think vague general ideas trumps actual numbers, but in reality they do not.
In effect, you have made it even harder to recover the costs over 5 years because you used more panels than you needed in order to have that surplus electricity that you think is a magic bullet, so your hand waving about it means absolutely nothing to people with a brain because they know that having a surplus costs fucking money.
Do some math, find a supplier, find an installer, then show us some numbers. if thats too hard for you, then maybe you shouldnt be lecturing anyone about basic economics. Here, i'll start this off for you: The cheapest system that would produce a monthly surplus costs $12,515 [wholesalesolar.com] from this supplier.. and that doesnt even fucking include installation cost or roof racks. To produce a 100% surplus, that would cost $23,770 but of course only get you back $5583.60 over 5 years and save $5583.60 over 5 years, for a total loss after 5 years of $12,602.80.. again not including the cost of fucking installation.
Don't assume he was going for 100% of household needs. Rarely is the sweet spot on the bell curve of costs/rebates/credits anywhere near 100% (and apple is surely getting incentives to max this out).
He is correct, a 5-year payback is the norm. You size the array to maximize your bang-for-the-buck.
The economics are basic, and they work.
Re: (Score:2)
He is correct, a 5-year payback is the norm. You size the array to maximize your bang-for-the-buck.
The AC that replied to him cited otherwise. Seems that electricity has to cost $0.50/kWh in order for it to pay back in 5 years.
$0.50/kWh is not the norm. Its 500% of the norm.
Re: (Score:2)
I live in CA. 4 years ago I installed a $70k ($50k after rebates) 8kW solar system on my roof and my garage's roof.
Prior to solar, my original electricity bill peaked at ~$1100/month, more commonly about $600. This is due mainly to my own choices, no doubt, but still that's what we are dealing with - in Summer, the AC is on quite a bit (it's been 106F this last week) and the pool pump needs to run 8 hours a day for good cleaning. There's also the 2 pond pumps which run 24/7 and the reef tank pumps which als
Re: (Score:2)
Solar thermal power stations like the SEGS in the Mojave desert require lots of water to operate since they use a Carnot cycle, producing steam to drive turbines which then needs condensing on the other side of the loop like any thermal power station (coal, gas, nuclear). The SEGS generating facility evaporates 3.5 tonnes of water for every MWh of electricity produced, pumped out of a local aquifer which is not being replenished.
Thermal solar power stations tend to be situated in deserts where sunlight is a
Re: (Score:2)
but they're still quite inefficient (commercial units now have ~20% efficiency, only recently has research broken the 30% limit).
So what? The sun can be considered to be an infinite resource with regard to organic existence on earth. And it can't be turned off. If the solar cells aren't there, 100% is wasted rather than 80%.
So for solar farms, other things can easily become more important than simple efficiency per unit area. An electricity utility might want to run a solar power tower. A computer company, probably not.
"43.5 million kilowatt hours" (Score:4, Insightful)
Over its lifetime? Per year? Per what?
Re: (Score:2)
Probably 43.5 GWh over its lifetime. The article is badly written, and so is the summary: classic Slashdot style.
Re: (Score:2)
Re:"43.5 million kilowatt hours" (Score:4, Interesting)
137 acre plant != 137 acres of solar panels
Large solar setups need roads to access the panels, and if they are tilted it'll need space between panels to avoid wasted panel area from shadows.
Here's a similar but older plant:
http://en.wikipedia.org/wiki/Nellis_Solar_Power_Plant [wikipedia.org]
14 MW, 140 acres, 30 GW*h/year, built in 2007
Note in the photos how much sun hits the dirt (i.e. not on panels within the 140 acre plant).
So it's the right ballpark for a newer plant of the same size (but with better solar panels or packing) to be 18MW in 137 acres. I think you are right that it isn't optimistic, 43 GW*h/year sounds perfectly reasonable for a plant built 6 years later.
Re: (Score:2)
Note in the photos how much sun hits the dirt (i.e. not on panels within the 140 acre plant).
Thanks to the magic of Pythagoras, there may be more than 140 acres of panel surface in the 140 acre plant, and still be light that spills on the ground. And it may be at a more advantageous angle to the sun at the same time.
Re: (Score:2)
Your assumption is that peak insolation lasts for 6 hours a day which is wrong. Peak insolation only occurs around noon when the sun is high in the sky and the airpath to the panel surface is minimal. Early in the morning and late in the evening the sun is low in the sky, the airpath is long and the blue and UV light which drives photovoltaics gets absorbed by the air, attenuateded by clouds, diffracted by dust etc. and the output of the cells drops off significantly even if sun-tracking arrays are used (mo
Re: (Score:2)
It's true that you only get peak at solar noon, but your power doesn't drop to zero at 3PM. My estimate at 6 hours is equivalent to saying that the year round average is sort of close to 25% of peak power. To get an exact solution, you have to integrate a complicated equation, for which I do not have the time nor inclination. (That's too much like serious work that I won't do for free.) However, this site has a different method of estimation: http://www.solar-estimate.org/?page=solar-calculations [solar-estimate.org].
They
Re: (Score:1)
Over its lifetime? Per year? Per what?
The second quoted article says, "While the total cost of the proposed solar project was not disclosed, it’s ultimately expected to generate 43.5 million kilowatt hours of solar energy — the equivalent of taking 6,400 cars off the road each year, the company said in a statement."
So I guess the savings is 43.5 million kilowatt hours per year, but it's not clear.
Re: (Score:2)
Let's see if this car-off-the-road unit makes any sense. Cars are usually powered with gasoline. A gallon of gas has 33.4kWh of energy, according to the DOE. 6800kWh therefore sound like about 203 gallons of gas. If an average car gets 24 miles per gallon, that represents about 4800 vehicle miles. Most people drive
Re: (Score:3)
Night (Score:1, Flamebait)
Too bad it will have to shutdown at night when the sun goes down.
Re: (Score:1)
Uhh, battery backup storing such a huge amount of power that isn't being used? You know, these things called uninterruptable power supplies?
43.5 million kilowatt hours ?? (Score:1)
And then what - they'll dismantle it? Normally such things are spec'ed in wattage
Re: (Score:3)
The wattage is by itself meaningless. How much power is produced varies wildly based on the duration and intensity of the sunlight, which varies by geographic location and local weather patterns.
Measuring it in watt hours, on the other hand, is a practical measurement, since it's both how power is actually paid for, and gives you an idea of the real output of the system.
For example, TFA indicates this is an 18 megawatt installation, but the 43.5 "million kilo" watt hours (or 43.5 gigawatt hours if you don't
Re: (Score:3)
The wattage is by itself meaningless. How much power is produced varies wildly based on the duration and intensity of the sunlight, which varies by geographic location and local weather patterns.
Measuring it in watt hours, on the other hand, is a practical measurement, since it's both how power is actually paid for, and gives you an idea of the real output of the system.
For example, TFA indicates this is an 18 megawatt installation, but the 43.5 "million kilo" watt hours (or 43.5 gigawatt hours if you don't use bullshit units) indicates an average of a bit under 5 megawatts.
43.5M KWh by itself is meaningless, unless, as the parent posted wrote, they are planning to tear down the plant after it hits that level.
You're assuming 43.5M KWh per year which wasn't stated in the summary and only implied in the article.
Re: (Score:2)
43.5 million kilowatt hours ?? And then what - they'll dismantle it?
Yes. Then all the data has been processed. In that power budget all programs should have finished and returned the Final Result which is then archived and the computers will be shut down. Then the building is transformed into a generic concrete slab factory.
Not too hot? (Score:1)
Nevada and solar (Score:3)
The Burning Man [burningman.com] festival noted that with all the Nevada rebates on solar panels, net was effectively the cost of installation.
Burning Man has access to a large amount of volunteer labor, so they can effectively put up solar panels for free. They setup panels to power parts of the event (the man), then move them to Gerlach once the festival is over. As I recall, the goal was to provide all the power [cnet.com] for the towns nearest the festival.
I wish other states were as forward-looking. At this point the benefits (to the state) of encouraging the infrastructure probably outweigh the costs.
(Yes, I know. Just getting to burning man uses an enormous amount of fossil fuels. What's your point?)
Re: (Score:3)
I wish other states were as forward-looking. At this point the benefits (to the state) of encouraging the infrastructure probably outweigh the costs.
They aren't so forward looking. The subsidies are for very large installations, effectively eliminating residential installations - keeping us beholden to corporate central generation. In fact, the dollars available for residential solar are so small that only about 70 installs per year get any money at all.
http://solarpowerrocks.com/nevada/ [solarpowerrocks.com]
At least you don't have to pay property tax on your solar equipment, but Nevada's got one of the lowest property tax rates in the country anyway so it ain't all that b
Re: (Score:2)
Or are you one of those people that believe that "subsidy" is a magical source of free money?
Are you one of those people who think that the government would give back the money if there was no place to spend it?
Let's spend government money on tangible efforts to improve society. If not, what do you think they will do with the money?
Re: (Score:2)
Cut taxes? People have a choice to invest in solar panels, or other things if they prefer. What is the advantage of the government taking that money and making the choice for them? I guess people are dumb and the government knows better, right?
Re: (Score:2)
no.. if a festival or a company wants free solar energy, you pay it(if you live there and pay taxes).
What happens after it generates 43.5 million KWh? (Score:2)
What happens after the plant generates 43.5 million kilowatt hours of electricity? Does it self destruct?
Assuming that they meant 43.5 million KWh per year, that's still only about 5MW of power on average [google.com], which is likely less than half what the datacenter will consume. And when the sun is not at its peak, it'll be drawing power from NV Energy's conventional fossil fuel plants.
Re: (Score:2)
Re: (Score:3)
...it'll be drawing power from NV Energy's conventional fossil fuel plants.
No, it will be drawing power from geothermal sources. Check here. [theregister.co.uk]
Re: (Score:1)
"43.5 million kilowatt hours should be enough for veryone", Tim Cook, 2013
Re: (Score:3)
That's less than a half-mile square (which is 160 acres).
They should go for at least a 1 mile square (640 acres) now that would be worth talking about!
The other (and more clear) way to say "half-mile square" (i.e. a square with each side being one half mile long) is "one quarter square mile".
Re: (Score:2)
804.672^2m is even clearer, or 880^2 yards if you like.
What are the other 2,060 acres for? (Score:2)
2,200 acres to house a 2.2 acre facility??? OK, add in the 137-acre solar farm, still less than 140 acres used?
Re: (Score:2)
Buffer space, so the panels don't get damaged from stray bullets.
A wind farm perhaps? (Score:2)
Anyone who has spent any amount of time in the desert knows two things: It's sunny most of the time and it's windy most of the time. Perhaps they're planning on covering their bases and are planning on installing some windmills to generate power when a storm rolls through and there is no sun. And on the sunny and windy days, any excess power they have can be sold to the local utility.
Re: (Score:3)
There are always batteries and even flywheels to store some power for use come night, which can help reduce the power bought from the local utility even more.
Solar is one of those things that is becoming a "why not?" as opposed to a "why?" Throw a couple panels on a roof or a shed, have a MPPT controller [1] for a set of batteries, then add a 3000 watt inverter and a 15A circuit to the house where all chargers and relatively low, parasitic items can plug into. That way, even though the larger items like t
Re: (Score:1)
Sold back to the utility? No, they are cut off [oregonlive.com]... Our energy/water management is absurd.
Re: (Score:1)
But still... 137 acres to power a 2 acre facility? The facility is just under 1.5% of the used space possibly up to 2% when the parking lot is factored in.
That sounds like a horribly inefficient use of land. So much for being green.
Re: (Score:2)
Wouldn't they then get a dressing down for not using panels made in the US?
Nellis AFB solar array (Score:2)
http://en.wikipedia.org/wiki/Nellis_Solar_Power_Plant [wikipedia.org]
Which company does what now? (Score:2)
Does Apple have permission from Sun to do this?
What happens when there are too many iClouds making the sky dark?
You would have to be an Oracle to sort all this out.
Basic information for.... (Score:2)
http://en.wikipedia.org/wiki/Solar_power_tower [wikipedia.org]
https://en.wikipedia.org/wiki/Concentrated_solar_power [wikipedia.org]
It always amazes me that people make the silly comment that solar power doesn't work when the sun goes down. I mean, seriously... these type of facilities have been in the movies for what...15-20 years now, and there's been working plants since 1984.
What is the temperature in Reno today? (Score:2)
ridiculous typo (Score:2)
Uh no, it will generate approximately 43.5 killowatts. That's how you measure power generators. I don't know of any solar arrays that produce 43.5 million killowatt hours and then just run out and stop permanently. Now if they gave a time period like it can generate 43.5 million KWh in one day, that's a valid but needlessly 2-dimensional unit.