How much power does a cellphone use?
March 27, 2006 8:48 PM Subscribe
For that matter, how about an IPod? It's not a battery life issue, but a question about the about of power the various server farms consume while supporting your widget's content needs. I learned today that "Computers in the average corporate data center consume the energy equivalent of 80 barrels of oil each day" So I wondered how much power your typical digital device gobbles per day?
80 barrels of oil = 3712 gallons of gas = 137,344 kilowatt hours.
Per day?
I call shenanigans on that statistic.
posted by frogan at 9:25 PM on March 27, 2006
Per day?
I call shenanigans on that statistic.
posted by frogan at 9:25 PM on March 27, 2006
Perfect link, tellurian ... from that site, we get this report, which concludes that things aren't terribly bad now, and they'll get significantly better over time:
Minicomputer and mainframe unit energy consumptions have been falling for years, as more and more of the functions previously handled by peripherals are integrated into fewer and fewer chips. Heat is an especially important issue in these machines (it reduces equipment lifetime), and the manufacturers have for this reason pushed to reduce energy use.
...
While energy use for this equipment has grown rapidly in recent years, this growth is likely to slow in the next decade because the US commercial sector market is becoming saturated (especially for PC CPUs and monitors).
posted by frogan at 10:39 PM on March 27, 2006
Minicomputer and mainframe unit energy consumptions have been falling for years, as more and more of the functions previously handled by peripherals are integrated into fewer and fewer chips. Heat is an especially important issue in these machines (it reduces equipment lifetime), and the manufacturers have for this reason pushed to reduce energy use.
...
While energy use for this equipment has grown rapidly in recent years, this growth is likely to slow in the next decade because the US commercial sector market is becoming saturated (especially for PC CPUs and monitors).
posted by frogan at 10:39 PM on March 27, 2006
Wait, I'm confused. Are you asking how much energy it takes to use a cell-phone, in aggregate? As in, including the power dissipated in the phone, all the cell towers, and all the networking infrastructure needed for that call to get from point A to point B? Or are you just asking about the power consumption of a cell phone in isolation? It's really not clear what you're asking because you mention datacenters in the same question.
Power consumption of a cell phone is not exactly a trivial thing to calculate. They are designed to be as miserly as possible, to make the battery charge last as long as possible. This means that they continuously regulate the amount of RF power they transmit so that at any given time they are transmitting only the smallest amount of power necessary for the signal to make it to the tower. And it should be obvious from the specifications that a cell phone in standby mode takes vastly less power than one that is actually making a call. So you would really have to fully specify what you mean here.
Alternatively, maybe you're just asking how much power power is stored in one charge of a typical cell phone's battery. That might be easier to calculate. Just about all batteries will have a capacity specified in Ampere-Hours (or milliamp-hours, same thing.) If you know the output voltage (and you're willing to make the simplifying assumption that the output voltage doesn't vary, which is not really true) then you can just multiply amp-hours times volts to get watt-hours which is directly comparable with other figures.
posted by Rhomboid at 10:57 PM on March 27, 2006
Power consumption of a cell phone is not exactly a trivial thing to calculate. They are designed to be as miserly as possible, to make the battery charge last as long as possible. This means that they continuously regulate the amount of RF power they transmit so that at any given time they are transmitting only the smallest amount of power necessary for the signal to make it to the tower. And it should be obvious from the specifications that a cell phone in standby mode takes vastly less power than one that is actually making a call. So you would really have to fully specify what you mean here.
Alternatively, maybe you're just asking how much power power is stored in one charge of a typical cell phone's battery. That might be easier to calculate. Just about all batteries will have a capacity specified in Ampere-Hours (or milliamp-hours, same thing.) If you know the output voltage (and you're willing to make the simplifying assumption that the output voltage doesn't vary, which is not really true) then you can just multiply amp-hours times volts to get watt-hours which is directly comparable with other figures.
posted by Rhomboid at 10:57 PM on March 27, 2006
The full numbers are at the bottom of this page:
"Did you know? "The average corporate data center burns through 80 barrels of oil per day. Calculation based on a data center size of 2 MW that burns an equivalent of 3.3 barrels of oil per hour."
So they're saying 1 barrel is 1,650 KWh, or about the same as frogan used. And 2MW is a huge data center.
Taking the numbers from here an iPod battery holds 3.5 Wh, so that's half a million charges per barrel, or allowing for the inefficiency of charging, 250-400,000.
Cellphone batteries are usually a bit bigger than iPod batteries, but you get the idea.
posted by cillit bang at 11:02 PM on March 27, 2006
"Did you know? "The average corporate data center burns through 80 barrels of oil per day. Calculation based on a data center size of 2 MW that burns an equivalent of 3.3 barrels of oil per hour."
So they're saying 1 barrel is 1,650 KWh, or about the same as frogan used. And 2MW is a huge data center.
Taking the numbers from here an iPod battery holds 3.5 Wh, so that's half a million charges per barrel, or allowing for the inefficiency of charging, 250-400,000.
Cellphone batteries are usually a bit bigger than iPod batteries, but you get the idea.
posted by cillit bang at 11:02 PM on March 27, 2006
You could get a Kill a Watt, Electricity Usage Monitor and find out. I've actually wanted one of these for a while.
posted by fenriq at 11:15 PM on March 27, 2006
posted by fenriq at 11:15 PM on March 27, 2006
Response by poster: I'm trying to understand the invisible power consumption associated with say, an iPod, or cellphone. Eg if there are a certain number of iPods out there and Apple's iTunes server consumes X Kilowatts of power, what is the average user's share of that power overhead?
Likewise, say a phone: E.g. If T-Mobile has a given number of subscribers and servers what is the total actual energy footprint associated with owning a cellphone per year?
posted by marvin at 11:28 PM on March 27, 2006
Likewise, say a phone: E.g. If T-Mobile has a given number of subscribers and servers what is the total actual energy footprint associated with owning a cellphone per year?
posted by marvin at 11:28 PM on March 27, 2006
I think the original poster was actually trying to find out how much power is consumed by the iTunes Music Store, per customer or per customer-downloaded song.
Even thought the iTunes music store is certain to use a ton of electricity, comparitively, it's quite a bargain.
Imagine the amount of energy that would instead be required to make those millions and millions of little plastic and metal foil discs and then ship them to all corners of the world....
posted by jaded at 5:06 AM on March 28, 2006
Even thought the iTunes music store is certain to use a ton of electricity, comparitively, it's quite a bargain.
Imagine the amount of energy that would instead be required to make those millions and millions of little plastic and metal foil discs and then ship them to all corners of the world....
posted by jaded at 5:06 AM on March 28, 2006
Response by poster: About 17 percent of the nation's server farms are located in the San Francisco Bay Area and Silicon Valley, requiring about 80 megawatts of power to run
I guess I could dig deeper and do the math.: 80 Megawatts support how many PCs, cellphones, blackberrys, and iPods etc in the Bay Area?
Someone must have done a breakbown on the total power overhead for these devices somewhere..
posted by marvin at 11:25 AM on March 28, 2006
I guess I could dig deeper and do the math.: 80 Megawatts support how many PCs, cellphones, blackberrys, and iPods etc in the Bay Area?
Someone must have done a breakbown on the total power overhead for these devices somewhere..
posted by marvin at 11:25 AM on March 28, 2006
This thread is closed to new comments.
I work at a place with a supercomputer, and I once overheard one of the PR people who shows it off boast to some potential investors that it uses the same amount of electricity/day as 4 families in a year. I guess average families, and probably a non-leap year.
It does generate a tremendous amount of heat, and I imagine just the air conditioning costs to keep all those processors from melting is appreciable.
How many kilowatt hours in are in a barrel anyway? I tried to convert by way of rods to the fortnight but I got confused and gave up, but I don't think 80 barrels is out of line or that Sun would lie to you about the issue.
posted by moift at 9:16 PM on March 27, 2006