Talk:Computer cooling

This  level-5 vital article is rated B-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Computing Low‑importance This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing Low This article has been rated as Low-importance on the project's importance scale.

To-do list for Computer cooling: edit · history · watch · refresh · Updated 2018-03-12 Here are some tasks awaiting attention: Expand : Describe some methods of reusing waste heat from data centers

this topic name should change to liquid cooling. not always water is used. actually most times is used a special coolant. calling it water is kinda dangerous. try to drink cooland and you'll see what happens. — Preceding unsigned comment added by 85.240.64.53 (talk) 16:40, 10 June 2012 (UTC) actually, it shouldn't even be liquid cooling because all liquids are fluids but not all fluids are liquids. Cooling has elevated over some years and water cooling is impossible unless there are tubes involved. Mineral oil cooling on the other hand is efficient without tubes.Reply

Someone has created a CPU cooling wikipedia article that is similarly related to this one. I believe that it should be merged with this article, or at least linked too. --Farnkerl 15:44, 5 August 2005 (UTC)Reply

I agree. I'm editing the Computer cooling article to include the info in the CPU cooling article, applied to computer components in general (where appropriate), as much of it applies equally to other components. I'll see how it works out. If it doesn't, I'll just put a link in an appropriate place. --FrostyBytes 18:53, 21 March 2006 (UTC)Reply

May I introduce a new computer cooling concept, Ice-Cooled-Computer-Cooler:

http://www.ice-cooled-computer-cooler.homechoice.co.uk

Suggest someone add a phase change section.

I just merged CPU cooling into this article, we need to remove the redundant information.

Well, that was a bit too hasty, in my opinion. Anyway, I've finally merged the CPU section into the article. Additional clean up may be required, I'd appreciate it if someone could check it out, I'm probably too invested in the article to see its flaws. --FrostyBytes 01:00, 15 June 2006 (UTC)Reply

I think that computer fan should not be merged with Computer cooling but instead only be listed in short form. Computer fan should be a seperate part from Computer cooling because it is a different thing. For example, although there is a Compressor in the Air conditionig system, but they are seperated article due to Compressor can be written in detail, similar to Computer fan article, and may make the Computer cooling article too long if is merged.Carbonferum 00:46, 27 May 2006 (UTC)Reply

What about mentioning the sandia impeller cooler?

---

Hey, I think the "Heat Sink" article should be merged with this too, it is pitiful by itself and really belongs here... I don't know how to do that though. :(

I currently disagree. Computer cooling is an objective; heat sinks are an object. Each has their place as a separate article. As an analogy, you will find articles on hunting and on arrows. --Charles Gaudette 21:10, 23 June 2006 (UTC)Reply

[]I completely Disagree. The purpose of a heat sink is to cool. Thus making it apart of the computer coolant process.-AKiLi

"...but there is no anecdotal evidence of rounding cables reducing performance"

I do not believe this is the correct usage of the word "anecdotal"; the proper word is most likely "empirical". Anecdotal evidence is useless, while empirical evidence has actual value. However, I do not know the author's original intent. Please advise. -- June 2006

With recent changes, the computer cooling article has seen significant improvement. As of this date, the following modifications would help greatly in improving the article:

• Citations, citations, citations! That is to say, cite your sources (or even find sources for others, if you're in a helpful mood).
• The addition of more photos and graphs, especially a picture showing typical airflow paths inside desktop tower enclosures.
• Content order optimization
• The change of descriptions regarding propietary hardware to more generalized archetypical hardware types.

--FrostyBytes 01:25, 24 June 2006 (UTC)Reply

Not sure if I am doing this correctly, but the link to Leonard Peltier leads to a Native American activist, not the intended individual. http://en.wikipedia.org/wiki/Leonard_Peltier -Jascal

I was just reading the article as a novice and was thinking that a sentence listing the range of temperature limits should have been included somewhere with designations for normal and critical values. Thanks, Ben —Preceding unsigned comment added by Bbentrup (talk • contribs) 22:32, 3 July 2008 (UTC) http://www.heatsink-guide.com/ would this work? SWMn (talk) 17:46, 7 January 2013 (UTC)Reply

The heat-pipe-section is very short and old.

It is nor true anymore, that heatpipes are only used for small form factor and silent cooling. AMD even uses heat-pipe coolers as boxed coolers for its high end Athlon 64 X2 models (I think X2 4800+ and upwards).

Also I think it should be mentioned, if it is a problem, when you install heatpipe-cooler motherboards upside-down, which must be done in some cases like this one.

H₂C (for short) is added as a hybrid spot cooling technique that mixes thermoelectric and liquid cooling, along with temperature sensors and control logic to manage fan and pump speed and TEC voltage to prevent frost and humidity condensation. Dell invented H₂C and is the first company to use this technology. A short article on H₂C links to related materials, including Dell's technical paper. I tried to write the article without marketing hype and keep it neutral, but mentioning Dell is a must so people know where to find additional info.

-Wcaswell

Liquid Cooled Peltier devices are nothing new, nor is electronic temperature regulation. H₂C just sounds like another out-of-the-box extreme cooling product. --711groove 13:46, 21 March 2007 (UTC)Reply

Actally the DELL H₂C unit was produced by an external contractor owned by CoolIT (now known for making "whitelabel" water cooling products for Corsair and others - example: the Corsair H60). It is produced with components from CoolIT and Delphi and while DELL certainly have let the marketing drums roll, it is doubtful they have done more than merely paying an outside contractor for a ready-to-use PC cooling kit. On a side-note they are not offering service on the units and documentation is not available so sadly they are virtually impossible to maintain for non-DELL personnel SplatMan DK (talk) 13:19, 5 November 2012 (UTC)Reply

The link to Toms hardware guide gives a 404 error. —The preceding unsigned comment was added by 83.70.48.143 (talk) 16:29, 4 February 2007 (UTC).Reply

The article fails to mention IHS lapping, this should be mentioned in the lapping section, as it is equally, if not more common than heatsink lapping among overclockers. --711groove 13:46, 21 March 2007 (UTC)Reply

It is also worth noting that as the surface gets smoother, there is less and less ability for TIMs (such as thermal grease, or a standard thermal pad) to fill microscopic gaps from thermal expansion, thereby reducing over-all effectiveness of such materials and in some cases inhibiting thermal transfer.

Is that actually a concern? It seems to me that making the surface smoother implies a reduction of those microscopic gaps, and thus a smaller amount of thermal grease is necessary -- if the grease particles have no gaps to cling to, so much the better. VanishingUser 03:50, 8 April 2007 (UTC)Reply

Wow. Talk about misunderstanding. ok. a few theory basics then the application. The most efficent heat transfer due to conduction between adjacent atoms. Diffrent atoms form diffrent pure lattus structures which conduct heat at diffrent rates. ( The nunber of free electrons helps, so good electical conductors become good heat conductors ). The distance between the nuculi ( or density of the material ) influnces it also. Pure copper is a great heat conductor, but soft, so a tiny bit of impurity is added to gain strength, while maintaining the thermal conductive properties. But when you want to transfer heat away from metal wires in glass you have an interface that has a gap. You have to fill in that interface with something. Air works pretty good, but Silicon Dioxide works a lot better. ( you can also use plain toothpaste with a good effect ). When the interface is sooth from polishing, there is less room for a bridge compound, and thus more coduction of heat takes place between pure materials, glass to copper.

This has been shown in practice for thousands of examples. The closing of microsopic gaps with conductive material ( copper ) improves the heat transfer. Microscopic gaps are NOT closed by thermal expansion, due to the low coefficent of the copper alloy subratate. I do not know the source of your informaiton, but clearly in both disagrees with materials science, and the practice of lapping.

Can you show a few examples of this phenomon? Artoftransformation 19:13, 7 August 2007 (UTC) "Encycloperic content must be verifiable"Reply

I'm not understanding your argument. Are you claiming it's better to have more gaps and more transfer compound or less gaps and less compound? How is glass relevant to this? Ham Pastrami 08:30, 8 August 2007 (UTC)Reply

Its better to have LESS of a GAP, and as a result, less combound. Glass is relevent, because most CPUs are made of silicon, and you have to be extremly carefull when you put the heat sink on. Artoftransformation 09:15, 11 August 2007 (UTC) ( excuse me? your name is Ham Pastrami? Really? )Reply

Rule 1: Provide Refrences:

Please See -> http://reference.techpowerup.com/Thermal_Paste [[1]]

That is fine, but you seem to be agreeing with the comment while having an argumentative tone so it wasn't clear what your position was. My real name is not Ham Pastrami, that is just my user name. :) Ham Pastrami 17:38, 14 August 2007 (UTC)Reply

Image:Pcprobe.png is being used on this article. I notice the image page specifies that the image is being used under fair use but there is no explanation or rationale as to why its use in this Wikipedia article constitutes fair use. In addition to the boilerplate fair use template, you must also write out on the image description page a specific explanation or rationale for why using this image in each article is consistent with fair use.

Please go to the image description page and edit it to include a fair use rationale. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to insure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page.

If there is other fair use media, consider checking that you have specified the fair use rationale on the other images used on this page. Note that any fair use images uploaded after 4 May, 2006, and lacking such an explanation will be deleted one week after they have been uploaded, as described on criteria for speedy deletion. If you have any questions please ask them at the Media copyright questions page. Thank you.

BetacommandBot (talk) 18:52, 5 December 2007 (UTC)Reply

There are a lot of statements in here that aren't cited that seem kind of like they should be verified.

Poor airflow (including turbulence) due to friction that reduces the amount of air flowing through a case, possibly causing stable whirlpools of hot air in certain areas.

This seems kind of vague, at least the part about stable whirlpools!

Laptops are typically made to rest on a solid surface. Unfortunately a flat surface is the least desirable angle to dissipate heat, lower temperatures are achieved by a chimney effect when a laptop is set at an angle from horizontal.

How does the 'chimney effect' apply to this? Is it saying that hot air gets trapped under the laptop?

Most older PCs use flat ribbon cables to connect storage drives (IDE or SCSI). These large flat cables greatly impede airflow by causing drag and turbulence. Overclockers and modders often replace these with rounded cables, with the conductive wires bunched together tightly to reduce surface area.

Cylindrical shapes are pretty awful when it comes to drag and turbulence, maybe this was referring to something more specific but I'm not sure what. RadioYeti (talk) 16:14, 23 April 2008 (UTC)Reply

The section on "Peltier cooling or thermoelectric cooling" seems very similar (almost identical to) content on http://www.heatsink-guide.com/peltier.htm I don't know the correct policy in this case, (does it need a cite? is this plagiarism?) but I thought I would note it as I was searching for info on Peltier cooling with regards to computer cooling and came across both of these sources. —Preceding unsigned comment added by 209.144.248.167 (talk) 16:18, 23 June 2008 (UTC)Reply

In the watercooling paragraph there is almost zero information and there is a link to a main article on water cooling which covers other uses, so why not put the part on computer water cooling in here, since it isn't very long? Same on the heatpipe cooling, you would expect an article related to this one, but you get a scientific article with no information about computer heatpipe use in it. And if you click on the watercooling link from here you would expect something to do with this article, after all you are reading the computer page.

'Heat Pump' A heat pump is similar to the thermodynamic cycle of the refrigerator, but the point of interest in a Heat Pump is the condenser, not the evaporator (Refrigerator). Suggest the use of the term 'heat Pump' is dropped, as cooling based on Peltier etc. decidedly does NOT use the heat transfer mechanisms associated with a heat Pump! —Preceding unsigned comment added by 82.195.166.195 (talk) 12:25, 7 April 2010 (UTC)Reply

124.177.184.228 (talk) 04:13, 12 August 2008 (UTC)Reply

Is there anything wrong with opening the case and aiming a $50 vornado at it? Seems like dinky case fans can't do jack squat to fix cooling problems, but I'm wondering if there is damage a house fan can do to electronics. — Preceding unsigned comment added by 99.45.98.137 (talk) 06:41, 29 June 2011 (UTC)Reply

Descriptions of cooling systems for mainframes would be of value here; not every computer that needs cooling can be lifted by one person. --Wtshymanski (talk) 20:32, 20 June 2010 (UTC)Reply

I believe that there are at least two notable technologies that should be mentioned.

1. Closed loop water cooling linked by a heat exchanger to an external supply of chilled water.
2. Pressurized Helium in a Thermal Conduction Module (TCM).

The CDC 6x00 and Cyber lines are notable examples of computers using Freon® cooling.Shmuel (Seymour J.) Metz Username:Chatul (talk) 07:25, 14 November 2010 (UTC)Reply

On that note, the term supercomputer appears only once and refers to the Cray 2. A lot has happened since and the shift to air cooling in Blue Gene and the 2011 reversal to water in the Power 775 etc. is notable. Overall the article is pretty dated I would say. Not that I have time to fix it this decade, but just a note. History2007 (talk) 11:47, 14 August 2011 (UTC)Reply

Months later...added a bit on big iron and mentioned that new supercomputers don't use as much energy per MFLOP. --Wtshymanski (talk) 16:40, 20 September 2012 (UTC)Reply

because this has "cpu cooling" merged to it, i assume this is the right locaition to add this. the first p4 required holes in the case for scews are there any other processors that need holes added? this doesnt really come under the "atx" standard as intel added them (does it?). other categories it may go under is: Computer case screws or atx, well it can be good to know, so i know wheather my old cases are suitable for modern cpus.

Charlieb000 (talk) 00:46, 14 March 2012 (UTC)Reply

Are there any external heatsinks that are laptop sized and can be used to cool a whole laptop? Wsmss (talk) 11:27, 7 August 2012 (UTC)Reply

Is it just my English (which isn't my first language) or is there something fundamentally wrong with the following sentence from the Peltier section? "This results in an effective, albeit extremely inefficient, heat pump.". Can something be "effective" and "inefficient" at the same time? SplatMan DK (talk) 13:52, 5 November 2012 (UTC)Reply

Using a pile driver to hit a nail into a piece of wood might be effective (it achieves the nailing objective) but would be extremely inefficient in its gross over-utilization of resources (cost, power, etc). Effectiveness and efficiency are often discussed in information processing and management circles, but in a nutshell, effectiveness is to do with qualitative aspects, whilst efficiency is related to quantitative aspects. 203.129.23.146 (talk) 05:58, 8 November 2013 (UTC)Reply

Could someone give examples and reasons to malfunctions that occur in cold climate? (I noticed that RFID card didn't work properly at -20°C) — Preceding unsigned comment added by 81.197.190.212 (talk) 12:37, 29 December 2012 (UTC)Reply

I think that there should be more about air flow in Air cooling.SWMn (talk) 20:36, 6 February 2013 (UTC)Reply

The article talks a lot about how computer cooling is done, but it barely touches on the whys. It would be great if someone with the relevant knowledge and citations could add a section or paragraph about that. AFAIK, part of the "why" is in fact that resistance in the electronic components increases when they get hotter, and this increased resistance cause electronic signals to arrive improperly where they are needed, leading to potential calculation errors. Another part of the why is possibly thermal shock, i.e. thermal expansion of the components can cause them to 'crack'. I have no or very poor sources for either explanation though, so it would be good if someone could come with something with proper citations.--Anders Feder (talk) 17:01, 29 July 2013 (UTC)Reply

Perhaps these can be mentioned ? See http://www.silentpcreview.com/article114-page1.html KVDP (talk) 08:53, 24 July 2014 (UTC)Reply

Unsourced stub article, with no evident relevance (or use of that terminology) outside hobbyist liquid cooling of PCs. Andy Dingley (talk) 09:12, 14 August 2015 (UTC)Reply

The "Liquid submersion cooling" section should at least link somewhere with [Server Immersion Cooling] entry. --Dpope83 (talk) 11:29, 18 February 2016 (UTC)Reply

theres one drawback affecting overall design of computers inherent to air cooling: buildup of dust lumps on the heat transfer surfaces. it is especially problematic in computers with internal parts that are difficult to (and not designed for) being opened for maintenance. the passive parts of the cooling system, like the copper and aluminium heat transfer panels for graphic cards and CPUs trap dust that is carried by the air-flow and within a few (typically 2 to 5 years depending on how much dust is in the enviroment where the latop is used) years the effectiveness of the air-cooling downgrades sharply because the dust lumps are insulating heat and blocking air flow. this leads to the need to disassemble the outer case and manually remove the dust that is stuck inside, but this not encouraged by laptop manufacturers.

indeed this may not belong here, perhaps rather to laptop design or designed product life-shortening/ planned obsolescence, but anyway mentioned it. 176.63.176.112 (talk) 16:38, 20 October 2016 (UTC).89.134.199.32 (talk) 11:26, 22 September 2018 (UTC).Reply

Hello fellow Wikipedians,

I have just modified one external link on Computer cooling. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20120813212211/http://www.hpcwire.com:80/hpcwire/2010-07-02/ibm_hot_water-cooled_supercomputer_goes_live_at_eth_zurich.html to http://www.hpcwire.com/hpcwire/2010-07-02/ibm_hot_water-cooled_supercomputer_goes_live_at_eth_zurich.html

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at {{Sourcecheck}}).

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 08:19, 29 November 2016 (UTC)Reply

Hello fellow Wikipedians,

I have just modified one external link on Computer cooling. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20140728123429/http://www.iceotope.com/the-science to http://www.iceotope.com/the-science
• Added {{dead link}} tag to http://driverstorer.com/oil_cooled_pc.html

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 21:05, 11 August 2017 (UTC)Reply

I have removed the following sentences from section Computer_cooling#Airflow (currently the text is part of a caption, but should be moved):

"For benefits of negative pressure, this configuration will tend to result in lower operating temperatures for the computer than positive pressure. This is because in positive pressure, warm air will build up inside the case. This warm air will be exhausted slower than it can build up, because the case's intake is greater than the exhaust. Meanwhile, in a negative pressure configuration, the computer will be able to exhaust the warm air faster than it can develop."

1) The first sentence suggests that lower differential pressure causes lower temperatures. I do not believe this. While it may be true in practice that negative pressure systems are cooler, if it is true, the reason is not the negative differential pressure. In theory, temperature is a function of cooling power (thermal energy transferred per time), which is mostly a function of airflow. If two systems differ only in differential pressure, but are otherwise identical, and in particular have the same airflow, I see no reason why their temperatures should differ much. I would even expect a system with a higher differential pressure to cool better, because higher pressure improves heat transfer from hot components to the air.

2) The next sentences talk of "build up of air". There is no "build up of air" worth mentioning. Yes, when starting the fans, for a brief time, I guess maybe a second or at most a few, there is a difference in airflow between intake and exhaust, but this does not affect cooling power. Once the pressure difference has been reached, intake flow exactly equals exhaust flow. Otherwise, the amount of air in the case would continuously fall or rise, which is impossible in practice.

Instead, pressure differences arise because fans create them. Well-documented fans come with specifications that state the pressure difference that they can create under certain conditions, for example their speed. Put simply, to arrive at the total differential pressure, one has to add the differential pressures of all intake fans, and subtract the pressures of all exhaust fans. If anyone disagrees, please explain and provide references. --RainerBlome (talk) 13:37, 17 December 2019 (UTC)Reply

A brief history in this article is kind of neeeded, imo, to give perpective, chronological order and orientate a reader. just overview: first time on the market, perfected, when it achieved teorethical limit, what type replaced it and then from the beginning for all types. this is a bit lacklustre. 95.178.152.20 (talk) 06:28, 27 April 2020 (UTC)Reply

a table would be nice:

	Technology	Underlying technology first patented	First year on the market	Min effective [W] output	Max effective [W] output	...
air cooling	Fans	for example thermosyphon was patented in 1937 but appeared in pc cooling systems around 2019
	Heat sinks
	Heat pipes
liquid cooling	Liquid nitrogen
	Liquid helium

This section has no citations for the discussion beginning with "A simple conservative rule of thumb for cooling flow requirements..." Using the approximation equations that follow, I've run a couple of trials and the math provides results that are difficult in the extreme to believe. For example, it calculates that the cooling airflow for a system that dissipates 500 W and is allowed a 40 °C temperature rise is 22 CFM. That is less than the airflow provided by a single 120 mm fan, many of which can source 150% of that flow or more. I've not been able to locate a reliable resource for this topic (was hoping to find one here) but there are plenty of demonstrations on the topic of air cooling PCs that show far more is needed. (See any case review in https://www.youtube.com/user/GamersNexus for examples of instrumented evaluations of cooling strategies.) It would be useful if the author of this section could provide citations so we could at least learn the source of the approximation math.