Hardware Analysis Community Forums http://www.hardwareanalysis.com/content/topic/33965/ 2009-01-07T23:50:56-05:00 http://www.hardwareanalysis.com/content/topic/33965/ http://media.hardwareanalysis.com/halogo.gif text/html 2008-08-13T21:47:25-05:00 Dave Ingram http://www.hardwareanalysis.com/content/topic/33965/?o=40#542928 Dublin... Yes i know its unconventional.<br /> However, if you look at the rig, you will see the machine in question is a 1 sq foot black magnesium cube, with sealed sides and top and only a very small vent in front.<br /> <br /> Please also note in this rig the PSU is over the video card.<br /> <br /> So dumping in cool air from the back (Rig is less than 1 cubic foot) and exhausting out the PSU is no problem for such a small area If you look at a next cube. you will see the only venting possible(3 sides sealed) the only possible venting options are rear, bottom and front. Hot air pours out of every vent, the bottom the front the back. Dual 2Ghz with less than 20 db noise at 100%x2 cpu utilization, for this case it works great!<br /> <br /> Taking this to a standard case, ala my year 2000 top of the line Antec Performance II, i've found that you can eliminate alot of noise by not having any front fans whatsoever<br /> bring in fresh air thru the 2 80 MM in back and hot air will exhaust out the PSU.<br /> (Note this case door is solid, without venting or fan adapter)<br /> text/html 2008-08-13T15:49:23-05:00 Beavis Khan http://www.hardwareanalysis.com/content/topic/33965/?o=20#542905 DublinGunner said: <div class="quote">The idea is to have a FLOW of air THROUGH the case, not dump lotf os air inside.<br /> <br /> Genreal rule of thumb - front= air in, rear= air out.<br /> <br /> You need to have the air come in to the case, absorb some of the heat, and then exhausted.</div><br /> <br /> Yep. I do prefer to have a little more fan airflow on the intake than on the exhaust though - having a slight positive pressure in the case keeps it from sucking air through gaps in the case, drive bays, etc. At least that way you can install filters on intake fans and catch a fair bit of the dust before it accumulates inside. text/html 2008-08-13T15:17:39-05:00 DublinGunner http://www.hardwareanalysis.com/content/topic/33965/?o=20#542904 Mr.INgram - that is not very good advice at all.<br /> <br /> The idea is to have a FLOW of air THROUGH the case, not dump lotf os air inside.<br /> <br /> Genreal rule of thumb - front= air in, rear= air out.<br /> <br /> You need to have the air come in to the case, absorb some of the heat, and then exhausted. text/html 2008-08-13T05:58:56-05:00 Dave Ingram http://www.hardwareanalysis.com/content/topic/33965/?o=20#542876 I'm a strong believer in having the case fans blow IN<br /> <br /> (Constantly dumping in fresh air) Given adequate venting, will work well.<br /> <br /> However you do need to be more adamant about dusting<br /> <br /> My new Quadcore ( a q6600) with factory fan dosent go over 54 C at full 100% CPU load for all 4 processors.<br /> <br /> my old casemod rig<br /> <br /> NeXT cube gutted with an Athlon Dual Core 3800+ and Nvidia 7600 GS has run at 100% Cpu utilization for a period of 2 weeks without crashing.<br /> <br /> Note the pictures show the original build, thermaltake ductingmod (stealth 80 MM) turned all the way down on vantec nexus fan controller (5v? 7v?), I like a quiet computer and even the CPU fan is on the controller (forbidden i know) but theres a giant hunk of copper, and aluminum and a 80 mmfan dumping onto another 80 mm fan on the sink. try running SETI command line or (BOINC at 100%x2) with YOUR cpu fan turned all the way down.<br /> <br /> Fans Blowing IN... works dandy till it gets really dusty!<br /> <br /> Replacing NB fan the original Shuttle Mobo broke, wouldent post any video thru built in or any known working video card. Replaced with an ECS Mobo and Athlon 3800+, runs quiet and runs well-- very well air cooled w fans blowing IN<br /> <br /> Case mod stats here<br /> <a class="ext" href="/action/r/http://morgothsring.freeforums.org/next-cube-project-t8.html" target="_blank">http://morgothsring.freeforums.org/next-cube-project-t8.html</a><br /> <br /> Note* since i made this post i have finished off the back. its not so ugly anymore!<br /> <br /> I like how Xbitlabs declares a system stable after 30 mins. RUN IT 2 weeks than tell me if its burned in!<br /> <br /> <br /> text/html 2008-06-20T16:58:30-05:00 Tam the Bam http://www.hardwareanalysis.com/content/topic/33965/?o=20#538527 <br /> Am gonna use toilet paper! Pmsl! <img src="http://media.hardwareanalysis.com/smilies/smile5.gif" width="14" height="14" border="0" alt=":P" title=":P"> text/html 2008-06-20T14:34:36-05:00 Dragon, Peaceful http://www.hardwareanalysis.com/content/topic/33965/?o=20#538513 While John's post above gave very good methods to tell fan direction via arrow indicators, left hand rule, and the curvature of the blades, here are two more methods.<br /> <br /> One, the label. For a typical PC case fan, usually the side where there is a label covering the center of the fan (and usually under the label there's the access for putting oil to the shaft), is the side which air will be blow out. However, as I said &quot;typical&quot; and &quot;usually&quot;, there are exceptions out there where the fan is made the other way around, so this is not a 100% guaranteed method.<br /> <br /> Two, paper test. Our hands are NOT the best tools to tell direction of air flow. Trust me, I did one time tried to feel which way air goes for a case fan, both sides felt about the same. Until I held a piece of paper near each side, then I finally determined the direction. Holding the paper on the top edge with the surface of paper parallel to the fan. If the lower edge of the paper bents toward the fan, that's the intake side. If the lower edge of the paper bents away from the fan, that's the exhaust side. You need a light piece of paper of course. Try not to use toilet paper, though, since it tends to be sucked in. <img src="http://media.hardwareanalysis.com/smilies/smile4.gif" width="14" height="14" border="0" alt=";)" title=";)"> And obviously, you need to power on your fan to do this.<br /> text/html 2008-06-20T11:51:06-05:00 DublinGunner http://www.hardwareanalysis.com/content/topic/33965/?o=20#538507 The easiest way is to look at the fins. It will be quite apparent which way the air flows.<br /> <br /> But yes, you can give it a quick spin if you like, but be very careful not to damage the fan - I wouldnt usually suggest that as being the best course of action, as you could damage the motor text/html 2008-06-20T09:13:02-05:00 Meats_Of_Evil http://www.hardwareanalysis.com/content/topic/33965/?o=20#538505 Such a simple way to find out is wet your hand and spin the fan manually just give it a good whirl then put your hand on either side to see which side it's blowing air towards.<br /> <br /> So easy a caveman can do it! text/html 2008-06-20T08:17:46-05:00 john albrich http://www.hardwareanalysis.com/content/topic/33965/?o=20#538503 Right On, ~Vel !!!<br /> <br /> When you're doing those experiments, don't forget working with and trying to optimize temp v. rpm profiles as well. A parametric change in one item can cause differences in temp readings elsewhere. text/html 2008-06-20T05:58:52-05:00 ~Vel http://www.hardwareanalysis.com/content/topic/33965/?o=20#538494 I've noticed that toying around with the fan direction is best. Take a notepad &amp; mark down the temps with which a complete list of which way each fan is blowing in your case with that temp. Do that for every possible combination, choose the best one. The best way is usually case &amp; hardware placement dependent and one arrangement isn't universally the best. text/html 2008-06-20T05:48:59-05:00 john albrich http://www.hardwareanalysis.com/content/topic/33965/?o=20#538493 bdubtubs said: <div class="quote">...If you have too much coming in... then the air pools. It really doesn't matter, as long as you have more air moving out than in. That is the best air flow movement....</div><br /> Sorry, but that is absolutely meaningless.<br /> <br /> You cannot have &quot;more air moving out than in&quot; in a typical computer case.<br /> <br /> What is important is that the computer has sufficient total cool air volume moving through (e.g. into and out of) the case to remove the generated heat. The higher the total air volume per unit time, the more heat can be removed.<br /> <br /> You do want to avoid obstructions, dead-ends, and abrupt directional changes where possible. &quot;Dress&quot; the cables (making sure they also don't obstruct any heat-dissipating device such as a heat-sink), ensure vents aren't blocked on either side, and that heat-sinks, fans, filters, etc. remain clean.<br /> <br /> It can also help to maximize cool air volume directly over higher-temperature surfaces, e.g. disk drives, CPUs, heat-sinks, chipsets, etc. If a CPU even with a decent air-cooled Heat-Sink Assembly requires further cooling, it will almost always help to duct additional cool air from an outside port in the computer case directly to the CPU's HSA.<br /> <br /> This topic is covered in numerous places with <i>lots</i> of arguments to keep people happy, and basic searches will direct you to them. text/html 2008-06-20T05:22:40-05:00 john albrich http://www.hardwareanalysis.com/content/topic/33965/?o=20#538492 <b>Most fans DO have 2 directional arrows. One arrow shows the direction of blade spin/rotation, and the other shows the direction of air-flow.</b> <br /> <br /> These 2 arrows on the shroud can be very hard to see as they are sometimes quite small, and consist of raised plastic the same color as the background plastic (think of black printing on a sheet of black paper). The arrows are usually located on the outer side surface of the shroud of the fan.<br /> <br /> <br /> In <i>general</i> the direction of air-flow is <b>from</b> the side without any permanent bracket on the fan rotor, <b>to</b> the side with the fixed brackets. However, always go by the arrows. Virtually all of the computer fans I've seen have the direction of air-flow moving relative to the direction of rotation like this:<br /> <br /> Rotation Arrow<br /> /\<br /> |-----&gt; Air-flow Arrow<br /> |<br /> <br /> <br /> Another way to visualize this, is hold your LEFT hand out, with your 4 fingers curled inwards, and your thumb sticking out. Your finger-tips show the direction of fan rotation, and your thumb shows the direction of air-flow.<br /> <br /> <br /> Yet another way to infer the direction of air-flow (especially if there is no shroud) is to look at the warp of one of the blades. Most computer fan blades have a warp across the width of each blade. Visualize that the end-on view of a blade looks a bit like a skewed &quot;(&quot; parenthesis. The air <i>generally</i> flows in the direction of the ---&gt; as the fan spins:<br /> <br /> ----- (----&gt; <br /> [cross-section view of blade &quot;(&quot; showing direction of air-flow ----&gt;]<br /> <br /> <br /> <br /> edit-highlight arrow information, add left hand rule, and blade cross-section air-flow info<br /> <br /> Note: descriptions are for typical +12V D.C. cooling fans used in computers text/html 2008-06-20T01:06:45-05:00 Michael ODoherty http://www.hardwareanalysis.com/content/topic/33965/?o=20#538485 How stupid is that.....if it blows in your face then thats blowing out..are you f@@king kidding me.....Do these fans have one central thing that tells you if it is pointing out or pointing in...without having to power it up!!!! I'm building my first computer and I don't know why fans just don't stamp an arrow, front &amp; back. I'm trying to build a negative airflow in my box and for the life of me I can't figure out which way these fans are going to be pointed when I power up. I know I'm missing something thats pretty obvious, so please point me in the right direction. text/html 2005-02-21T21:19:03-05:00 bdubtubs http://www.hardwareanalysis.com/content/topic/33965/?o=20#245056 You know guys... your making it harder than it is. negative air pressure in the case is your best bet. If you have too much coming in... then the air pools. It really doesn't matter, as long as you have more air moving out than in. That is the best air flow movement.<br /> <br /> Simplicity is better. text/html 2005-01-30T19:54:38-05:00 varun rao http://www.hardwareanalysis.com/content/topic/33965/?o=20#234655 I'm still not sure abou the side fan that sits directly over the processor.. should this blow in or out.. surely not in.. that would defeat what the processor fan was trying to do.