Using a PC as a heater

[79wjd]

The heat in my apartment just stopped working and so I started to wonder.....how efficient is a computer at converting electricity into heat? Is it less efficient than a traditional electric space heater? Would 300w of computer put out the same amount of heat as a 300w heater (I know space heaters are generally >1 kw)?

 

also, how well would 300w heat a room that's 10'x10'x8' with no windows?

[DocSwag]

I'd imagine most of the energy would go into heat. I can't think of any other major energy source the electricity is being turned into.

[Mira Yurizaki]

A computer chugging 1KW is comparable to a 1KW space heater, at least according to Puget Systems: https://www.pugetsystems.com/labs/articles/Gaming-PC-vs-Space-Heater-Efficiency-511/

[dalekphalm]

1 minute ago, DocSwag said:

I'd imagine most of the energy would go into heat. I can't think of any other major energy source the electricity is being turned into.

Err - what? For a PC? Most of the energy goes into... you know... powering the computer.

 

Using a PC as a space heater would be very inefficient.

[dalekphalm]

1 minute ago, M.Yurizaki said:

A computer chugging 1KW is comparable to a 1KW space heater, at least according to Puget Systems: https://www.pugetsystems.com/labs/articles/Gaming-PC-vs-Space-Heater-Efficiency-511/

Incredibly hard to believe - that basically means how inefficient the components are at powering themselves (generating lots of excess heat).

 

I'd be curious to see if others have tested this.

[79wjd]

 

4 minutes ago, wrathoftheturkey said:

Very, very inefficient. 

depends. On a 7970. any day. On a 780, it'd be meh

2 minutes ago, dalekphalm said:

Using a PC as a space heater would be very inefficient.

Apparently not: 

2 minutes ago, M.Yurizaki said:

A computer chugging 1KW is comparable to a 1KW space heater, at least according to Puget Systems: https://www.pugetsystems.com/labs/articles/Gaming-PC-vs-Space-Heater-Efficiency-511/

[DocSwag]

Just now, dalekphalm said:

Err - what? For a PC? Most of the energy goes into... you know... powering the computer.

 

Using a PC as a space heater would be very inefficient.

? In what form of energy?

 

Most of the electrical energy from the wall is being converted into thermal energy while doing work (in the form of calculations). The only other forms of energy I can think of are kinetic (air moving by fans), sound (sound of fans), and light (monitor). By definition energy is conserved so...

 

Yes, it is going into powering the computer, but most of that energy is converted into heat in the process of doing calculations.

[DocSwag]

1 minute ago, dalekphalm said:

Incredibly hard to believe - that basically means how inefficient the components are at powering themselves (generating lots of excess heat).

 

I'd be curious to see if others have tested this.

It's called physics.

[dalekphalm]

Just now, djdwosk97 said:

 

Apparently not: 

Mind boggling. According to that graph, the space heater is actually even worse compared to the PC. Something doesn't seem right. I'd be interested in seeing a comparison of multiple small electric heaters, not just a single random model.

[Mira Yurizaki]

1 minute ago, dalekphalm said:

Incredibly hard to believe - that basically means how inefficient the components are at powering themselves (generating lots of excess heat).

 

I'd be curious to see if others have tested this.

When you consider P = IV and you can play with ohm's law all day, and resistance is converting electrical energy into heat, you can make a case that consuming 1KW of electrical power does indeed create 1KW of heat, regardless of how it's being used.

[DocSwag]

1 minute ago, dalekphalm said:

Mind boggling. According to that graph, the space heater is actually even worse compared to the PC. Something doesn't seem right. I'd be interested in seeing a comparison of multiple small electric heaters, not just a single random model.

Then, according to you, what is the energy from the wall being converted to? Energy is conserved, so it can't be being converted into "calculations" or "work."

[dalekphalm]

1 minute ago, DocSwag said:

Then, according to you, what is the energy from the wall being converted to? Energy is conserved, so it can't be being converted into "calculations" or "work."

I would have thought the energy (depending on the specific component, but let's assume GPU and CPU) is being used to "flip" the actual transistors. I don't think you can consider that kinetic energy though, as as far as I'm aware, the transistors don't actually move.

 

I'm not saying that a computer shouldn't produce any heat, but a device which it's sole purpose is to create heat (a space heater), should be more efficient at the job, then a device that is doing other work.

 

Now I fully admit I'm not a physicist. So you're saying that by definition, all Integrated Circuit/processor calculations are thermal energy? Do you have any links to documentation that I could read up on this?

[Electronics Wizardy]

17 minutes ago, dalekphalm said:

Incredibly hard to believe - that basically means how inefficient the components are at powering themselves (generating lots of excess heat).

Then where is the energy going? Its not making light, noise, or movement. 

[Electronics Wizardy]

7 minutes ago, dalekphalm said:

I would have thought the energy (depending on the specific component, but let's assume GPU and CPU) is being used to "flip" the actual transistors. I don't think you can consider that kinetic energy though, as as far as I'm aware, the transistors don't actually move.

Think of it as the transitors are shorting the circuilt, or leaving it open. They energy is being turned into heard due to the restance

[dalekphalm]

1 minute ago, Electronics Wizardy said:

Think of it as the transitors are shorting the circuilt, or leaving it open. They energy is being turned into heard due to the restance

Yeah but isn't the entire point of modern electronics and Processor material technologies that they are low resistance? Isn't lower resistance desired?

 

And a heater would essentially want to maximize the resistance since that creates the heat.

[Electronics Wizardy]

Just now, dalekphalm said:

Yeah but isn't the entire point of modern electronics and Processor material technologies that they are low resistance? Isn't lower resistance desired?

The restance doesn't really matter, that 1v for the cpu still has to do something, and it will naturally generate heat. 

[WereCat]

I have radiator in my room turned off for years..even when there is -10°C outside. My PC keeps my room very pleasantly warm even if I have window sligthlt opened to get fresh air.

 

About 300W for PC in typical gaming load+2 monitors+ router so about 400W max.

Room is 18 square meters.

 

For whole apartment it may be possible with more power hungry PC with water cooling and TV running somewhere in the background.

[Automobili3XF]

Back when i had my AMD machine, i'd kept it next to my feet during winter time when playing some games, it would knock the chill out of my feet in seconds, and keep then very, very warm.

[TheRandomness]

4 minutes ago, Andydandyfiddy said:

You could use a Pentium 4 and a HD 5970 or GTX 295

Or, you know, a GTX 480. c:

[Elerek]

10 minutes ago, dalekphalm said:

Yeah but isn't the entire point of modern electronics and Processor material technologies that they are low resistance? Isn't lower resistance desired?

 

And a heater would essentially want to maximize the resistance since that creates the heat.

Yes, they are getting more efficient, that's why you can get a 4ghz cpu that runs on ~60watts. Efficiency means uses less power, not makes the power somehow vanish from existence. If you use the electricity in a computer, it goes to heat. Modern parts are more efficient because they do more with less power, thus run cooler.

[79wjd]

14 minutes ago, WereCat said:

I have radiator in my room turned off for years..even when there is -10°C outside. My PC keeps my room very pleasantly warm even if I have window sligthlt opened to get fresh air.

 

About 300W for PC in typical gaming load+2 monitors+ router so about 400W max.

Room is 18 square meters.

 

For whole apartment it may be possible with more power hungry PC with water cooling and TV running somewhere in the background.

interesting.....I've been trying to heat my (9sq meters) room with my pc (300w~) for about 30+ minutes and I haven't noticed much. 

[mariushm]

No matter how you look at it, using a computer to heat your room is stupid. You could run some superpi or some benchmarking tools to keep your cpu hot, and you could use some bitcoin mining software or some benchmark like furmark to heat the video card, but there's much easier ways to produce heat.

 

You can make your own heater using simple resistors and a power supply. (could be computer power supply, could be laptop adapter, could be anything)

 

You have 12v generated by the computer power supply, so you can simply connect resistors to that 12v output of your power supply and let them heat.

 

We have the basic formula V = I x R  (voltage equals current times resistance )  from which we can determine P = I²xR  ( power dissipated in resistor equals current squared times resistance). We just have to be careful when picking  a resistor value careful so that the power dissipated in the resistor will be at most around 80-85% the power rating of the resistor (otherwise it will melt down, overheat etc) 

 

So for example a good match for our 12v voltage that's available in computer power supplies is 56 ohm resistor.   Since V = I x R , this means through resistor there's going to be I = V / R = 12v / 56 ohm = 0.215 A , and so the power dissipated in the resistor will be P = 0.215 x 0.215 x 56 = about 2.58 watts. A 56 ohm 3w max resistor is 12 cents, if you buy 100 of them :  http://www.digikey.com/product-detail/en/yageo/PNP300JR-73-56R/56AECT-ND/2059166

 

There's also resistors rated for higher power, for example the bigger ceramic resistors which are rated for 5w or 7w or 10w. For example here's a 33 ohm 5w rated resistor, 18 cents each if you buy 100, 32 cents if you buy 25 : http://www.digikey.com/product-detail/en/yageo/SQP500JB-33R/33W-5-ND/18667

Each resistor would dissipate about 4.3 watts of power, so 100 such resistors would produce 430 watts of heat

 

So with just 12 dollars, you can make your own 258 watt heater (or 18$ for 430 watts), excluding shipping which adds maybe an extra 8$ or so.

 

Each resistor would heat up a lot if it has to dissipate an amount close to its maximum rating (well over 100 degrees C), so it would also be a good idea to put the strip of resistors in a duct or a tube and have a fan on one side of the tube blow air through the tube to constantly move colder air over the resistors. The air would warm up from the resistors and in turn heat your room.

 

Each molex connector or sata connector or 6pin pci-e connector on your computer can provide up to around 8A of current safely or about 100 watts, so you could separate the resistors in sets of 50 or somethign like that and give each strip its own connector to the power supply.

If you add an on/off switch to each strip, you have your own variable heater.

Another example : 18 ohm 10w resistor , would dissipate 8 watts if connected to 12v .. and they cost 35 cents each in 25pcs, 24 cents if you get 100: http://www.digikey.com/product-detail/en/yageo/SQP10AJB-18R/18W-10-ND/18741

 

 For over 5w per resistor you really have to push air over the resistors to move the heat from the surface of the resistors into the air and heat the air in your room... so factor in the cost of the 80mm - 120mm fan pushing air through a box or tube or something that holds your resistors (you want some shape that would make the air from fan move over all resistors)

 

 

[79wjd]

5 minutes ago, mariushm said:

No matter how you look at it, using a computer to heat your room is stupid. .......but there's much easier ways to produce heat.
 

 

Running Heaven/Valley sounds a lot easier then going to the store and buying all of that and then putting it together seeing as I have a PC next to me. (but that's not even the point)

[DocSwag]

19 minutes ago, dalekphalm said:

Yeah but isn't the entire point of modern electronics and Processor material technologies that they are low resistance? Isn't lower resistance desired?

 

And a heater would essentially want to maximize the resistance since that creates the heat.

Yes, the process technologies are meant to make the transistors themselves use less energy to flip, BUT at the same time, we are also using the increased power and thermal headroom to use more transistors and increase clock speeds. So yes, modern transistors at emu church more efficient from ones from, say, 10 years ago, but the power savings are "invested" into more transistors for higher ipc and/or more cores or higher clocks.

 

As for how the transistors generate heat:

Quote

A transistor (FET, in modern ICs) never switches instantly from full OFF to full ON. There is a period while it's turning on or off where the FET acts like a resistor (even when fully ON it still has a resistance).

As you know, passing a current through a resistor generates heat (P=I2RP=I2R or P=V2RP=V2R).

The more the transistors switch the more time they spend in that resistive state, so the more heat they generate. So the amount of heat generated can be directly proportional to the number of transistors - but it is also dependent on which transistors are doing what and when, and that depends on what the chip is being instructed to do.

Yes, manufacturers may position specific blocks of their design (not individual transistors, but blocks that form a complete function) in certain areas depending on the heat that block could generate - either to place it in a location with better heat bonding, or to place it away from another block that may generate heat. They also have to take into account power distribution within the chip, so placing blocks arbitrarily may not always be possible, so they have to come to a compromise.

I got this answer from here: http://electronics.stackexchange.com/questions/160249/why-does-a-processor-get-hot

 

Naturally there is some electricity that gets through the transistor to the next one and so on, but a lot of the electricity is converted to heat. This is why people often use the TDP to somewhat estimate the power draw of a processing unit-nearly all of the electrical energy is converted to heat.

 

This stuff can be quite interesting. It's always cool to not just use the stuff but actually (semi) understand what's going on behind the scenes!

[DocSwag]

2 minutes ago, djdwosk97 said:

Running Heaven/Valley sounds a lot easier then going to the store and buying all of that and then putting it together seeing as I have a PC next to me. (but that's not even the point)

Don't run heaven; do F@H... For science!