Optimal 120mm fan setup in case for HDD cooling etc?

[ThaCrip]

My computer case is Antec SLK3000-B (pictured below(just a picture from Newegg's website for reference)) and i have Scythe S-FLEX SFF21F 120mm fan as a exhaust fan for the past 9 years now (it replaced the stock fan, which came with the case, which died less than 3 years later. i leave my PC on 24/7 pretty much.) and it still works fine.

 

but just today i received my order for a couple of COUGAR CF-V12H 120mm fans and currently have one of those two fans installed in the front of the case(prior to this i only had the exhaust fan(120mm Scythe)), in the purple 120mm clip shown in the picture below, so it blows air across three hard drives i have installed. i currently have one hard drive at the top, then center, then bottom on the hard drive holder in the picture below (the HGST 5TB is in the bottom of the hard drive rack which runs the hottest(see more below)). so basically all are spaced apart which i figure is probably better than sitting them right next to each other as it can hold five hard drives.

 

the Cougar fan is rated at 60.4CFM at 1200rpm and the Scythe fan is rated at 63.7 CFM at 1600rpm. i got those because they seem to be pretty quiet and have a long life span as the Scythe is rated at 150,000hours (which means i am not much over half way through it's rated life currently) and the Cougar's claim 300,000 hours of which even if i get about half of that 300k ill be more than happy.

 

according to CrystalDiskInfo v7.0.5 my 5TB HGST hard drive is generally floating in the 35-37c range at a idle (current room temp is 74-75f(in summer it will peak around 85f)(based on the clock i have in my room)), which is the hottest of the hard drives i have installed (the rest are 28c and 32c that the Cougar fan is pointed towards).

 

but basically my question is... would i benefit from better cooling of my hard drives, mainly the 35-37c idle temp drive as mentioned above, if i removed the purple plastic clip that holds in the front 120mm fan(which is what one of the Cougar fans is in right now as i type this) and then used the clips (or screws) they give me to hold it to the case more directly? ; or would it not really be worth my time etc? ; i just wonder if there is any general guideline you want to keep the hard drives at in terms of temps, like say no more than 45c etc? ; also, while 35-37c at a idle sounds solid enough i just wonder how hot the temp will get in more of it's worse case scenario thing which is when my room shoots up to 85f or so and see's some decent drive usage as just off the top of my head i would imagine i would like to keep it under 45c etc.

 

i ask that because earlier on the other Cougar fan, which is not currently installed in anything, i took it out of the box and connected it to one of my old computers just to see airflow when nothing is restricting it and it seems solid enough but i noticed as it was vertical (like how they typically set inside of a PC case) and then slowly moved my right hand towards the back of it where air gets sucked in to the fan you could feel a noticeable air flow drop off when my left hand was there to feel the air coming out of the fan. like i slowly moved my right hand towards the backside of where the fan pulls in air and within a few inches or so of the back of the fan you could feel a noticeable decrease in the air output of the fan on my other hand. so that kinda has me wondering whether that purple 120mm fan clip (shown in the picture below in front of the case) is potentially restricting the air flow of the Cougar fan a bit. hence, raising temps of my 5TB HGST drive a bit. so would removing that purple 120mm fan clip be worth my time to try or do you think my current temps are more than good enough?

 

any thoughts or suggestions?

 

SIDE NOTE: prior to receiving the Cougar fan, maybe the last week or so, i had some old cheap 80mm fan sitting in the general purple clip in front of the case blowing air over mainly the HGST drive etc and that seemed to shoot down the temps to maybe 32c or so but it was not a proper installation and it's a little noisy where as my current fan is silent.

 

here are some Newegg.com links to the three things described in the topic... my case (which i had since March 2006) = https://www.newegg.com/Product/Product.aspx?Item=N82E16811129152 ; Cougar fan = https://www.newegg.com/Product/Product.aspx?Item=N82E16835553001 ; Scythe fan = https://www.newegg.com/Product/Product.aspx?Item=N82E16835185006

 

p.s. just to state the obvious... i made sure the Cougar fan is spinning in the case, and in the proper direction, and everything is properly seated in the purple 120mm clip inside of the case.

 

[DevTech]

You are overthinking this.

 

Blow cool air on things you want to be cool is all you need to know. Unblocking air flow restrictions such as that plastic clip will help but only you know if it is worth the trouble and sometimes screwing a fan to the case introduces vibrations that can cause noise depending on the case design (unlikely)

 

The big mistake everyone makes is to do anything else besides blowing cool air on the things that needs cooling.

 

For example exhaust fans are rarely very effective in comparison because they rely on a mental fantasy model of "suction" that does not exist in the case itself. I can't believe how hard it is for people to understand this simple situation of any "normal" case being a "leaky boat" from the point of view of the fan.

 

Blow everything. Suck nothing. Use large quiet slow moving fans. Have a fan blowing  within 4 inches of anything you want to cool.

 

Simplest config is blow on hard drives, blow on chipset, blow on CPU cooler intake, blow on GPU intake. Maximizes cooling where it is need for least number of fans. Attempt to configure so nothing is cooled by hot air inside case - hard to do for power supplies and many integrated water radiators, but always possible with some ingenuity.

[ThaCrip]

so in other words, to keep it short... you think i am pretty much fine as is, or maybe it's worth attempting the removal of the 120mm plastic clip to connect the Cougar fan directly to the case.

 

Thanks for your time.

 

with that said... i might consider leaving things as is and just keep a eye on the temps over the next few months as that should give me a general idea of where things would peak as i don't want to obsess over the hard drive temps (especially given i never had one die inside my case so far even with their elevated temps) but i just wonder if there is any general guideline on hard drive temps as it just seems like if you touch a hard drive, when it's running, it should be comfortable to the touch and not to where it's somewhat uncomfortable to hold in your hand. so say maybe low 40's is the max comfortable temp for a hard drive? ; because i know without no fan blowing on the 5TB HGST drive (which is 7200rpm) the temps were quite a bit higher than they are currently (if i recall correctly i think it was something around 50c and might have even been higher when under a decent load) as i want to say there was more than a 10c increase in the temp without no fan blowing on that drive and i am pretty sure it was cooler in the room than it is as i type this which says about 79f and the 5TB is 36-37c at the moment. even a 2TB 7200rpm Hitachi reads 34c, which is the center drive in the rack, which makes me think it's probably not worth my time to swap drives around and either just leave things as is or remove the plastic clip and connect the Cougar fan directly to the case, like i mentioned at the top of this post, with the rubber pieces they gave me (they got screws but i would rather avoid using them even though it will be easier than installing the rubber pieces that hold the fan to the case) but those will likely be fairly difficult to install given ill have to install them on the case side first and then, with the rubber pieces installed inside of the case, ill have to somehow get the four rubber pieces pulled into place on the fan which, the more i think about it, i am probably going to avoid it unless idle temps of the hard drive start to get a little high as currently at a idle i am sure they are clearly on the safe side of things as, at least from what i can make out, without knowing anything definitive, HDD temps probably are sorta like this for a general rough guideline (in Celsius temps)...

 

30's and cooler = green (optimal)

40's = yellow (caution)

50's and hotter = red (probably not good)

 

but that's only a rough guess as it don't appear there is any definitive correlation between hard drive temps and their failure rates so i guess i should not be too worried about it but i just figure, like i mentioned above, that something comfortable to the touch seems more likely to last longer than something a bit uncomfortable to the touch.

 

For example exhaust fans are rarely very effective in comparison because they rely on a mental fantasy model of "suction" that does not exist in the case itself. I can't believe how hard it is for people to understand this simple situation of any "normal" case being a "leaky boat" from the point of view of the fan.

 

Blow everything. Suck nothing. Use large quiet slow moving fans. Have a fan blowing  within 4 inches of anything you want to cool.

 

That sounds like some good general advice overall but it seems like it would be a good idea to have one exhaust fan to get the hot air out of the case that builds up otherwise your going to be blowing that hot air inside the case over the components etc, right?

 

because i would imagine if you just strictly had fans only blowing on components (CPU/GPU/HDD etc) that you want to cool that the general temp inside the case would rise, at least some, correct?

 

so i don't really think people are relying on an exhaust fan(s) to keep the components inside of their case cool but that it seems like a good idea to keep the air temp inside of the case about as low as can be, so that the fans blowing directly on the components like CPU etc can work more optimally, which i figure one exhaust fan should do this job well enough and then everything else, do basically what you said, which is blow air on the components you want cooled.

[goretsky Supervisor]

Hello,

 

My very general rule of thumb is to push cooler air into the case from the lower front towards the bottom, and pull warm air out from the back rear towards the top.  This should create a stream of air inside the chassis which brings a constant stream of cool air in, and exhausts heated air in the rear.  Of course, this assumes that you're not running with the side panel(s) off, in which case not too much of anything's going to happen, unless you place fans right over hotspots.

 

Regards,

 

Aryeh Goretsky

 

[ThaCrip]

@goretsky

 

That's pretty much what i got right now as you can basically see in the picture above where that black fan is(i.e. exhaust fan), is where the Scythe fan (63.7 CFM/1600rpm) is located for the past 9+ years (it's still going strong as it's rated for 17+ years. i leave my PC on pretty much all the time.) and only very recently i installed the Cougar (60.4CFM/1200rpm) fan in the front (intake) to mainly help with hard drive cooling which claims double the life of the Scythe fan. but like i said, if i even get about half of that (i.e. similar to the Scythe fan life span) ill be more than happy. but i generally expect to get at least 10 years or so out of case fan, which is why i generally stick to the Fluid Dynamic Bearing type of fan, as i don't want to be replacing fans all of the time as i figure if a fan can't get at least 5+ years of life out of it, with the PC running all of the time, it's just not worth buying at all. the price on the Cougar fans are solid at a little over $12 each which from doing a bit of research seems like they are a solid choice with noise/air flow/longevity/price balance. i think i paid close to $20 for that Scythe fan back in early 2008.

 

but anyways, the side of my case looks like this...

 

so as you can see it does have a couple of vent holes in the side of it which i can't imagine would interfere much with the general airflow. also, i typically blow out my computer case thoroughly about twice a year with a air compressor, especially the CPU/GPU heatsink and fan areas. the PSU does not seem to have much dust in it (had it since Nov 2012 and it's warrantied til Nov 2017) as the fan barely moves as i just don't think it gets hot enough to spin up all that much.

 

Thanks for your input as i kinda figured one exhaust and one intake fan is a good standard to go by for general case airflow which i figure with a couple of those with decent CFM output, like i got, i should not have any obvious cooling issues as you can clearly feel decent air being pushed out the back of the case.

[DevTech]

7 hours ago, goretsky said:

My very general rule of thumb is to push cooler air into the case from the lower front towards the bottom, and pull warm air out from the back rear towards the top.  This should create a stream of air inside the chassis which brings a constant stream of cool air in, and exhausts heated air in the rear.  Of course, this assumes that you're not running with the side panel(s) off, in which case not too much of anything's going to happen, unless you place fans right over hotspots.

 

 

This is a false fantasy view of what air molecules are actually doing although you are 100% in the orthodox camp here by promoting an unquestioned model of air flow that had heavy promotion by major oems and others around the turn of the century.

 

The key to understanding the problem is actually your mention of the side panels being off for which you immediately see the obvious solution is to place fans right over hotspots. That is the only reliable and correct way to deal with cooling for the vast majority of people that lack infrared heat gun monitors and smoke analysis chambers.

 

Because from the point of view of an air molecule, the sides of a typical computer case are essentially already off. We see a square box in which fantasy air flows can conveniently take place but the air molecules see a leaky container full of holes and cracks and joints and obstructions that they must obey over and above what we wish would be happening.

 

Fans that exhaust air will suck in air wherever they can easily get it which translates to whatever is within a few inches of the fan. It will almost always not be that "gulf stream" of fantasy air that was sucked in at the bottom of the case.

 

 

 

 

[DevTech]

3 hours ago, ThaCrip said:

so as you can see it does have a couple of vent holes in the side of it which i can't imagine would interfere much with the general airflow. also, i typically blow out my computer case thoroughly about twice a year with a air compressor, especially the CPU/GPU heatsink and fan areas. the PSU does not seem to have much dust in it (had it since Nov 2012 and it's warrantied til Nov 2017) as the fan barely moves as i just don't think it gets hot enough to spin up all that much.

 

Thanks for your input as i kinda figured one exhaust and one intake fan is a good standard to go by for general case airflow which i figure with a couple of those with decent CFM output, like i got, i should not have any obvious cooling issues as you can clearly feel decent air being pushed out the back of the case.

"a couple of vent holes in the side of it which i can't imagine would interfere much with the general airflow"

 

You have a "Failure of Imagination"

 

In addition to those holes are all the unseen holes from the typical construction of a mass produced metal case.

 

Imagine - putting your mouth on that exhaust fan hole at the top rear of the case and trying to suck air through two cubic feet of leaky case from the bottom front. Right. The fan can't imagine doing that either.

 

[DevTech]

11 hours ago, ThaCrip said:

so in other words, to keep it short... you think i am pretty much fine as is, or maybe it's worth attempting the removal of the 120mm plastic clip to connect the Cougar fan directly to the case.

 

Thanks for your time.

 

with that said... i might consider leaving things as is and just keep a eye on the temps over the next few months as that should give me a general idea of where things would peak as i don't want to obsess over the hard drive temps (especially given i never had one die inside my case so far even with their elevated temps) but i just wonder if there is any general guideline on hard drive temps as it just seems like if you touch a hard drive, when it's running, it should be comfortable to the touch and not to where it's somewhat uncomfortable to hold in your hand. so say maybe low 40's is the max comfortable temp for a hard drive? ; because i know without no fan blowing on the 5TB HGST drive (which is 7200rpm) the temps were quite a bit higher than they are currently (if i recall correctly i think it was something around 50c and might have even been higher when under a decent load) as i want to say there was more than a 10c increase in the temp without no fan blowing on that drive and i am pretty sure it was cooler in the room than it is as i type this which says about 79f and the 5TB is 36-37c at the moment. even a 2TB 7200rpm Hitachi reads 34c, which is the center drive in the rack, which makes me think it's probably not worth my time to swap drives around and either just leave things as is or remove the plastic clip and connect the Cougar fan directly to the case, like i mentioned at the top of this post, with the rubber pieces they gave me (they got screws but i would rather avoid using them even though it will be easier than installing the rubber pieces that hold the fan to the case) but those will likely be fairly difficult to install given ill have to install them on the case side first and then, with the rubber pieces installed inside of the case, ill have to somehow get the four rubber pieces pulled into place on the fan which, the more i think about it, i am probably going to avoid it unless idle temps of the hard drive start to get a little high as currently at a idle i am sure they are clearly on the safe side of things as, at least from what i can make out, without knowing anything definitive, HDD temps probably are sorta like this for a general rough guideline (in Celsius temps)...

 

30's and cooler = green (optimal)

40's = yellow (caution)

50's and hotter = red (probably not good)

 

but that's only a rough guess as it don't appear there is any definitive correlation between hard drive temps and their failure rates so i guess i should not be too worried about it but i just figure, like i mentioned above, that something comfortable to the touch seems more likely to last longer than something a bit uncomfortable to the touch.

 

 

 

 

That sounds like some good general advice overall but it seems like it would be a good idea to have one exhaust fan to get the hot air out of the case that builds up otherwise your going to be blowing that hot air inside the case over the components etc, right?

 

because i would imagine if you just strictly had fans only blowing on components (CPU/GPU/HDD etc) that you want to cool that the general temp inside the case would rise, at least some, correct?

 

so i don't really think people are relying on an exhaust fan(s) to keep the components inside of their case cool but that it seems like a good idea to keep the air temp inside of the case about as low as can be, so that the fans blowing directly on the components like CPU etc can work more optimally, which i figure one exhaust fan should do this job well enough and then everything else, do basically what you said, which is blow air on the components you want cooled.

1. Your concerns about the plastic clip are ridiculous. Ignore it or try it but enough thinking about it. It is of almost no effect on your overall cooling.

 

2. You have already researched that there is no good data to support temperature correlating with hard drive longevity.

 

Any sort of air flow at all across the hard drive will prevent the creation of an extreme heat pocket. The urgency of airflow depends on how the drives are mounted. If there is any kind of vibration reducing plastic grommets etc or plastic mounting rails preventing metal to metal heat sinking of the drive case, then you need some sort of airflow and the amount of airflow simply does not matter.

 

3. I have much doubts that anyone reading this will change their minds on "exhaust fans" because it is firmly locked in the category of "everybody knows" but what the heck, I will tilt at windmills. https://www.nytimes.com/2017/05/29/opinion/in-praise-of-lost-causes.html

 

For the same reason that exhaust fans are unable to SUCK anything due to leaky cases, if you have a few fans blowing wonderful cool air from outside the case onto hotspots, the "hot air" will not build up inside the case but instead will simply leak out the nearest crack or hole.

 

4. From my measurements, it is harmless most of the time to leave the exhaust fan in place if you don't mind the slight additional noise it causes. Removing the exhaust fan and leaving its empty hole unfilled will be just as effective in letting any hot air escape the case. Sometimes the exhaust fan actually gets in the way and raises internal temps a bit and in very rare cases it actually reduces temps a bit but either way it is not a significant factor other than psychological perhaps.

 

5. If you want to improve on my "rules of thumb" within the specifics of your case, then be scientific and take actual measurements. Purchase a infrared heat gun and try different configurations. Or take a giant load of silicon caulking to seal everything and see if you can construct the Mythical Case where Suckage actually works

 

 

[ThaCrip]

@DevTech

 

Imagine - putting your mouth on that exhaust fan hole at the top rear of the case and trying to suck air through two cubic feet of leaky case from the bottom front. Right. The fan can't imagine doing that either.

 

Yeah, i don't expect it to do much there as it seems like it would just draw air from the points of least resistance etc.

 

but i can say this... with only that exhaust fan going (like i had before i installed the intake fan recently, which was running like this since March 2006 basically) it is pulling in air from the outside and from the front of the case, at least some, purely because dust builds up on the front panels dust filter. so that has to prove there is at least a little air being pulled into my case from the front otherwise the dust would not be there.

 

1. Your concerns about the plastic clip are ridiculous. Ignore it or try it but enough thinking about it. It is of almost no effect on your overall cooling.

 

Yeah, i am going to leave things (i.e. the front Cougar fan with the clip) as is currently as temps are not high enough for me to be of any real concern at the moment. ill just keep an eye on things through the next few months as this way room temp will be at it's peak and if things are okay then they always will be okay.

 

Any sort of air flow at all across the hard drive will prevent the creation of an extreme heat pocket. The urgency of airflow depends on how the drives are mounted. If there is any kind of vibration reducing plastic grommets etc or plastic mounting rails preventing metal to metal heat sinking of the drive case, then you need some sort of airflow and the amount of airflow simply does not matter.

 

The hard drive sits on the steel rack is all and have rubber pieces between the screws to dampen noise etc.

 

I have much doubts that anyone reading this will change their minds on "exhaust fans" because it is firmly locked in the category of "everybody knows" but what the heck, I will tilt at windmills.

 

From my measurements, it is harmless most of the time to leave the exhaust fan in place if you don't mind the slight additional noise it causes. Removing the exhaust fan and leaving its empty hole unfilled will be just as effective in letting any hot air escape the case. Sometimes the exhaust fan actually gets in the way and raises internal temps a bit and in very rare cases it actually reduces temps a bit but either way it is not a significant factor other than psychological perhaps.

 

Makes me wonder if turning that exhaust fan around and make it a intake will be worth trying?

 

i ask that because it will be pointing towards the general CPU area's heatsink/fan. but even if it does a little it probably will largely not matter especially because it's not like my CPU is running 100% all that often. so i guess if it can drop peak CPU temps when under full load it 'may' be worth trying but i think it tops out in the low 60's (Celsius) and that's when being really hammered by Intel Burn Test or Prime95 and the like which hammer CPU more than typical real worth programs. but right now with typical usage it's usually in the 30 something to maybe 50-ish range as i got RealTemp v3.70 running and it shows it's been running for 41 hours and the highest temps it recorded on the two cores of my i3-2120 CPU is 48c and 50c and the low was 28c and 30c.

 

but it makes me wonder if i could use that Scythe fan that's currently exhaust and turn it around as a intake, like i was thinking about considering at some point, and then see if it blows enough air over the CPU to maybe even consider passive cooling (i.e. no more Intel CPU fan blowing on the stock heatsink) as i figure the less fans etc in the case the less noise. but speaking of noise i wonder if the bulk of my general PC noise is from the hard drives as the PSU fan barely moves, which i imagine is pretty much silent, and i got the one exhaust/one intake fan, and then the stock i3-2120 CPU fan and then my tiny GPU (Radeon 5670 512MB single slot) which i can't imagine makes much noise, especially since it's fan speed does not ramp up unless it's really being taxed.

 

so you NEVER use any exhaust fans in your PC's?

[T3X4S]

Guys, look @ it this way:

OEM computers sometimes only have 2 fans, and they run for years with nothing more.  They use crappy, generic fans and crappy thermal paste.
Granted, they usually dont have high end hardware, but there is no mention of high end hardware in this post - it is mainly discussing HDDs.

@DevTech is correct - overthinking 

If you are getting reports of your HDDs failing or errors being thrown due to heat - then address it, otherwise - stop.

A fan on your CPU, 1 intake is all that OEMs usually have - and many run 24/7 and only die because of a crappy PSU, or a HDD failing near the MTBF - and there are millions of those PCs in use.

[DevTech]

2 hours ago, ThaCrip said:

Makes me wonder if turning that exhaust fan around and make it a intake will be worth trying?

 

i ask that because it will be pointing towards the general CPU area's heatsink/fan. but even if it does a little it probably will largely not matter especially because it's not like my CPU is running 100% all that often. so i guess if it can drop peak CPU temps when under full load it 'may' be worth trying but i think it tops out in the low 60's (Celsius) and that's when being really hammered by Intel Burn Test or Prime95 and the like which hammer CPU more than typical real worth programs. but right now with typical usage it's usually in the 30 something to maybe 50-ish range as i got RealTemp v3.70 running and it shows it's been running for 41 hours and the highest temps it recorded on the two cores of my i3-2120 CPU is 48c and 50c and the low was 28c and 30c.

 

but it makes me wonder if i could use that Scythe fan that's currently exhaust and turn it around as a intake, like i was thinking about considering at some point, and then see if it blows enough air over the CPU to maybe even consider passive cooling (i.e. no more Intel CPU fan blowing on the stock heatsink) as i figure the less fans etc in the case the less noise. but speaking of noise i wonder if the bulk of my general PC noise is from the hard drives as the PSU fan barely moves, which i imagine is pretty much silent, and i got the one exhaust/one intake fan, and then the stock i3-2120 CPU fan and then my tiny GPU (Radeon 5670 512MB single slot) which i can't imagine makes much noise, especially since it's fan speed does not ramp up unless it's really being taxed.

 

so you NEVER use any exhaust fans in your PC's?

1. Turning exhaust fan around makes no sense since there is nothing in that area you want to cool. You can't depend on fans to blow or suck any more than 4 inches before the air gets a mind of its own

 

2. Don't wonder where the noise is coming from - put your ear up real close it is is almost always obvious what is making the most noise. If in doubt, unplug all your fans and plug them in one at a time.

 

3. You are making torturous mental knot twisting about fans and cooling when you have nothing to cool! You don't have any high performance equipment with a large heat dissipation and certainly nothing worth overclocking.

 

4. If you did in fact have anything worth cooling, you would blow air from outside the case right on the component that needs cooling as close as possible to the item. Typically this means cutting holes in the case sides in just the right location and mounting fans in the holes you just cut.

 

5. It is correct that I do NOT ever use exhaust fans because based on actual measurements that anyone can choose to make they do not lower temperature of anything that needs a lower temp and just generate noise for no good reason. I just leave the hole empty and if any air wants to walk its way out the convenient hole then it is welcome to do so.

 

6. If I ever had a reason to suck air through the cracks in the case 4 inches from the exhaust fan so I could get a breeze out the back of the case then I would use an exhaust fan. It is almost always harmless to leave the exhaust fan in place so I do have one computer that has an exhaust fan with very pretty lights in it that I just left there.

 

7. Please stop thinking and start testing. Try things out. Put your fingers on things and see how hot they get. Put some cheapo plastic thermometers on various places etc. There is a Science to all of this, not VooDoo. Also maybe think about buying a kick-ass component that generates a lot of heat!

 

 

 

[T3X4S]

44 minutes ago, DevTech said:

 

 

3. You are making torturous mental knot twisting about fans and cooling when you have nothing to cool! You don't have any high performance equipment with a large heat dissipation and certainly nothing worth overclocking.

 

 

 

 

 

All that needs to be said

[ThaCrip]

@DevTech

 

Quote

2. Don't wonder where the noise is coming from - put your ear up real close it is is almost always obvious what is making the most noise. If in doubt, unplug all your fans and plug them in one at a time.

 

Well there is no real noise that stands out in a negative way, just the general noise a PC generates, which is expected on some level, which either has to be from fans (not the new Cougar since it sounds the same with or without that fan installed) or the hard drives since nothing else moves. not a big deal though as my main PC is much more silent than a old Alienware PC i had in 2001 (i still have this but it rarely see's any real use) which sounds a lot louder.

 

Quote

3. You are making torturous mental knot twisting about fans and cooling when you have nothing to cool! You don't have any high performance equipment with a large heat dissipation and certainly nothing worth overclocking.

 

Yeah, just something to play with a bit is all

 

i don't have any fancy equipment as it's just a 'ASUS P8H61-M LX' motherboard which i bought that along with the i3-2120 CPU and 8GB of DDR3 RAM back in May 2012 for a solid price which was a big upgrade for general system performance over my previous mobo/cpu/ram i had since March 2006.

 

it's possible i could upgrade my video card (preferably a single slot design) to play some newer games though but that's about it with my current system as i figure anything beyond that and i am probably better off investing it into a whole new PC build. but even this i am not in a hurry to do.

 

Quote

4. If you did in fact have anything worth cooling, you would blow air from outside the case right on the component that needs cooling as close as possible to the item. Typically this means cutting holes in the case sides in just the right location and mounting fans in the holes you just cut.

 

Given this info, i may disconnect the power to the Scythe exhaust fan just to see if i can hear any decibel decrease and see what happens to general temps of the PC as if things are the same then there is pretty much no real reason to have it running.

[goretsky Supervisor]

Hello,

 

Interesting reading.  Do you have any videos you can share showing this? 

 

The only studies I ever saw were at my last employer, which was an embedded systems manufacturer, and they did some studies where they pushed air across the chassis from right to left in a case designed to be racked, but that's the only thing I recall seeing.  This was a completely custom setup (all boards and layouts custom designed) and the only commodity parts were a pair of SCSI HDDs, so the design wasn't applicable to the rest of the market.

 

Regards,

 

Aryeh Goretsky

 

 

17 hours ago, DevTech said:

This is a false fantasy view of what air molecules are actually doing although you are 100% in the orthodox camp here by promoting an unquestioned model of air flow that had heavy promotion by major oems and others around the turn of the century.

 

The key to understanding the problem is actually your mention of the side panels being off for which you immediately see the obvious solution is to place fans right over hotspots. That is the only reliable and correct way to deal with cooling for the vast majority of people that lack infrared heat gun monitors and smoke analysis chambers.

 

Because from the point of view of an air molecule, the sides of a typical computer case are essentially already off. We see a square box in which fantasy air flows can conveniently take place but the air molecules see a leaky container full of holes and cracks and joints and obstructions that they must obey over and above what we wish would be happening.

 

Fans that exhaust air will suck in air wherever they can easily get it which translates to whatever is within a few inches of the fan. It will almost always not be that "gulf stream" of fantasy air that was sucked in at the bottom of the case.

 

 

 

 

 

[DevTech]

4 hours ago, goretsky said:

Interesting reading.  Do you have any videos you can share showing this? 

 

The only studies I ever saw were at my last employer, which was an embedded systems manufacturer, and they did some studies where they pushed air across the chassis from right to left in a case designed to be racked, but that's the only thing I recall seeing.  This was a completely custom setup (all boards and layouts custom designed) and the only commodity parts were a pair of SCSI HDDs, so the design wasn't applicable to the rest of the market.

 

 

1. I realized that air flow was a complex thing after I saw videos of smoke flowing through a PC case a long time ago. And that was when we didn't have CPUs at 200 Watts and Huge Chipsets and power conversion circuits on the motherboard delivering over 40 amps (like a car battery starting a motor except 24/7) to just the CPU chip. And then multiple GPUs each more transistors than the CPU!

 

2. So I decided to actually measure stuff. My advice to the average person without an infrared heat gun is you can't predict air flow and you can't figure out SUCTION but if you blow cold air 4inches or less from a component that needs it, then that is PREDICTABLE and deterministic.

 

3. Copying the OEM model has never made sense if people had stopped to consider the resources available to an OEM. They have the measuring tools, the smoke chambers,etc  and based on all that the OEM will make plastic shrouds and ducts and metal surfaces part of the case that direct airflow based on a measured model. You might have noticed that most OEM cases are very tight and the sides actually SEAL and there are no random holes and grids to make the case look sexy.

 

4. Every so often I provide this information and a few people read it maybe and think to themselves that the classic diagram of airflow they google from some random website makes more sense to them since that's what their uncle or brother etc told them. I guess I made the yearly educational attempt here since the OP was a detail freak but on the flip side he had nothing that needed cooling. And if Google delivers this to somebody 6 months from now, the fancy diagram explaining the chaotic flight of the bumblebees around the inside of the case will be missing. Optimizing this stuff probably doesn't matter anymore since we have probably reached a peak in overall power dissipation of high end systems and power and cooling requirements will start trending downwards. I used to recommend 850 W PSU (for high end CPU and 1 High end GPU) but now I can see 650 Watt starting to be enough.

[ThaCrip]

@DevTech

 

I guess I made the yearly educational attempt here since the OP was a detail freak but on the flip side he had nothing that needed cooling.

 

Well i did with the hard drives as while, like i was saying, i did not find any clear cut correlation online between hard drive temps and failure rates it just seems like as a general guideline that the cooler something runs the better chance of it having a longer life span especially with something that's only warm to the touch vs uncomfortable to the touch (this is especially noticeable in a external hard drive docking station i got with a 120mm USB powered fan cooling it with that warm to the touch vs uncomfortable thing).

 

but with that said... i realize that what i was asking in the topic is not something i should have any major concern over, but it's just more of a personal precaution. a bit more peace of mind is all.

 

2. So I decided to actually measure stuff. My advice to the average person without an infrared heat gun is you can't predict air flow and you can't figure out SUCTION but if you blow cold air 4inches or less from a component that needs it, then that is PREDICTABLE and deterministic.

 

While blowing cold air 4 inches or less on a component that needs it is always a good idea if possible, ain't there a possibility that just having the air freely flow out of the case on it's own (since you don't use a exhaust fan) has a potential to build up case heat a little bit and having one exhaust fan, preferably at a higher point on the case(since heat rises), would help guaranteed case temps won't rise? ; since it would pull the air out of the case faster than just having it creep out on it's own? ; or am i totally mistaken?

 

because i realize that if a component always has a constant flow of cool air outside of the case on it will obviously be fine i just wonder if the top portion of the case may build up hot air a bit, like if your PSU is there could it potentially stress that a little more with the added heat or is this a non-issue?

 

but even with that direct cold air thing... it might not always be easy to get cold air directly outside of the case direct to the component your trying to cool.

[goretsky Supervisor]

Hello,

 

I wish I had the equipment to measure this.  I have been slowly running my hand across the various openings in my computer's chassis (Fractal Design R4, default fans installed in default locations) and don't feel any air escaping.  I was hoping I'd feel something but nothing other then where the fans are located.

 

Regards,

 

Aryeh Goretsky

 

[DevTech]

7 hours ago, ThaCrip said:

While blowing cold air 4 inches or less on a component that needs it is always a good idea if possible, ain't there a possibility that just having the air freely flow out of the case on it's own (since you don't use a exhaust fan) has a potential to build up case heat a little bit and having one exhaust fan, preferably at a higher point on the case(since heat rises), would help guaranteed case temps won't rise? ; since it would pull the air out of the case faster than just having it creep out on it's own? ; or am i totally mistaken?

 

because i realize that if a component always has a constant flow of cool air outside of the case on it will obviously be fine i just wonder if the top portion of the case may build up hot air a bit, like if your PSU is there could it potentially stress that a little more with the added heat or is this a non-issue?

 

but even with that direct cold air thing... it might not always be easy to get cold air directly outside of the case direct to the component your trying to cool.

You are basically trying to say "please let me stick my head back in the sand and pretend that exhaust fans do something critically useful because I just love that simple diagram of cool air sucking in the bottom and then rising with heat like sunlight on the ocean to be exhausted by my nice exhaust fan at the top rear"

 

1. Totally mistaken. Nothing wrong with some hot air in the case as long as it is not being used by any cooling source, since every component that needs it is getting blowing cold air. Don't forget that electronics have a normal temp range that would feel uncomfortably hot if that was your room temperature. Anyways no need to pull air out faster since it is air that isn't doing anything anyways and the most likely scenario with a exhaust fan is that it will just pull air INTO the case within 4 inches of it's intake only to shove it out again.

 

2. HOT air on PSU is only a concern if you have violated the cool air rule. You must position or adjust your PSU of course so it gets cool air from outside the case like any other component that needs cooling. If you examine modern cases (at a local store etc) the PSU is mounted at the bottom of the case and the PSU fan sucking in air is facing downwards where there will be a grill on the bottom of the case to deliver cold air to it.

 

If you are stuck with a top mounted PSU, turn it upside down so that the fan intake is facing upwards. In most cases there will be about 1/2 inch gap between the fan intake and the top of the case. Drill holes in the side of the case next to the fan intake and presto you have cool outside air for the PSU! (Some tiny metal work at rear to reverse PSU might be needed depending on quality and era of case - most good cases provide flexible PSU mounting)

 

3. It is ALWAYS easy to get direct outside air to every component that matters. You simply need to cut fan sized holes in the side of the case at the right location.

 

4. For typical retro case such as yours:

 

A. Turn PSU upside down and drill vents holes in each side of case.

 

B. Fan on front blowing across hard drives

 

C. Fan on side of case blowing at chipset and CPU power area of motherboard.

 

D. Fan on side of case blowing air where the GPU intake is located.

 

E. Fan on side of case or sometimes top front blowing air where CPU intake is located although for CPU a simple integrated water cooler should be used and the radiator is typically mounted where the exhaust fan is located and you then reverse the radiator fans to blow cool air into the case across the radiator and really go nuts worrying about how you are creating monster hot air inside the case. (if you understood the above, cool air to the CPU takes precedence over everything and everything that needs it is nice and cool)

 

5. The reason you are obsessing over the "fantasy" model of air flow is that it worked well enough at the turn of the millennium when nothing inside a computer was particularly hot. The model just falls apart when you have 200 Watt heat sources everywhere!

 

6. For a fun project you could do all the changes in 4 A to 4 E which should take a month or so and then just as you finish, the new Intel i9 18 core CPU should be in stock for you to install with a nice NVIDIA 1080 Ti. If you keep your hardware "as is" then your cooling was just fine and needs no attention at all.

 

[DevTech]

5 hours ago, goretsky said:

I wish I had the equipment to measure this.  I have been slowly running my hand across the various openings in my computer's chassis (Fractal Design R4, default fans installed in default locations) and don't feel any air escaping.  I was hoping I'd feel something but nothing other then where the fans are located.

OK. So SUCKAGE does not work since there is no air tight SEAL in a typical case or even a vague hint of a sealed air flow.

 

Conversely, without a sealed environment you can't get POSITIVE AIR PRESSURE either, so you won't feel air escaping.

[ThaCrip]

@DevTech

 

Quote

If you examine modern cases (at a local store etc) the PSU is mounted at the bottom of the case and the PSU fan sucking in air is facing downwards where there will be a grill on the bottom of the case to deliver cold air to it.

 

Yeah, i have noticed that here and there online. i imagine that's pretty much the standard for cases for years now?

 

Quote

If you are stuck with a top mounted PSU, turn it upside down so that the fan intake is facing upwards. In most cases there will be about 1/2 inch gap between the fan intake and the top of the case. Drill holes in the side of the case next to the fan intake and presto you have cool outside air for the PSU! (Some tiny metal work at rear to reverse PSU might be needed depending on quality and era of case - most good cases provide flexible PSU mounting)

 

Not a bad idea

 

i bought my Antec SLK3000-B case in March 2006. you can basically see the PSU mount in one of the pictures i posted above. maybe it's something i can consider especially if i hang onto this case with any future upgrades. but if i get enough spare cash built up i would imagine ill probably just build a entirely new PC (without a monitor) from scratch with a mobo/cpu/gpu/ram/psu. but to save some $$$ it's possible, especially if i upgrade my GPU on my current system, that i could end up transferring the upgraded GPU to the new build and reinstall the current Radeon 5670 into the current PC. it seems lately, for years now, you can get a pretty powerful PC for not all that much $$$. not all that long ago it seems you had to pay a arm and a leg for strong gaming performance (i don't play much games anymore though but there is maybe a handful of more recent games i would not mind playing if my GPU were faster).

 

also, i don't know if this matters much but the fan moves pretty slowly on the PSU (might suggest the PSU is not stressed/hot(?) ). the PSU i have is this one... https://www.newegg.com/Product/Product.aspx?item=N82E16817151094 ; the price of that has came down since i bought it in Nov 2012 which says i paid $74.98 (i think was free ship but even if it was not add about $5 on top of that) for it back then and it's currently $49.89 which is quite great for a PSU with a 5 year warranty and Seasonic brand which at the time i got mine it appeared it's one of the most reliable PSU brands in general. it even currently has a $15 rebate which would put it about $35 which is quite great for a PSU of that quality and decent wattage. even the 620 watt version is not much more at basically $57.

 

Quote

The reason you are obsessing over the "fantasy" model of air flow is that it worked well enough at the turn of the millennium when nothing inside a computer was particularly hot. The model just falls apart when you have 200 Watt heat sources everywhere!

 

i would imagine my setup is pretty minimal on power usage as my GPU is powered by the motherboard and is quite compact as i prefer those single slot designs as they take up less space.

 

plus, my PSU fan RPM's are minimal which probably means it's not even close to being stressed and it's a 520watts PSU (the one in the link above).

 

and my current GPU does not have any heating issues either as i just blow out the computer about twice a year, with a air compressor, and blow out the CPU/GPU areas which keep heat down quite a bit especially when under full load with the FurMark stress test program as after some dust builds up in it you can see a solid increase in peak GPU temp vs once the dust is cleaned out it drops back quite a bit and the GPU fan does not ramp up it's RPM's as much etc as at a idle it's fan runs at 28% and only when the dust builds up and you run sometimes that taxes it, does it shoot up the fan speed etc. basically when the RPM's of the GPU fan shoot up i can notice a decibel increase from the case. but this is pretty much a non-issue.

 

but in general... i try to avoid all of those super power hungry components as when i upgrade i just generally try to find the solid price/performance thing which usually i can get solid CPU/GPU that does not need a nuclear power plant to run

 

Quote

For a fun project you could do all the changes in 4 A to 4 E which should take a month or so and then just as you finish, the new Intel i9 18 core CPU should be in stock for you to install with a nice NVIDIA 1080 Ti. If you keep your hardware "as is" then your cooling was just fine and needs no attention at all.

 

Yeah, but speaking of that i9 18 core CPU... i seen that article on Neowin which is ridiculous at $2k as you can build a ENTIRE PC, and a really powerful one at that, for that price or less. anyone who drops $2k on a CPU simply has too much disposable $ as i figure even $300-400 for a CPU is already pushing it and that's five times that price.

 

but cost aside... that would be nice to see how quick it encodes x264 video which i imagine is like lightning since that encoder takes advantage of CPU's with more and more cores.

 

also, it's possible ill upgrade my GPU so i can play some modern games but it's kinda up in the air at the moment as i am in no hurry. but when(if?) i do get one i would prefer to keep it to $150 MAX and hopefully have a single slot design so hopefully i can keep my last PCI-E add-on slot open for any potential upgrades but i doubt ill be adding in anything else as currently i got two PCI-E addon cards which is a USB 3.0 card and the other gives me a couple extra SATA ports (along with one IDE port) so i got a total of six (4 on motherboard, 2 on the PCI-E card). i also need a card with a HDMI port on it since i currently have mine connected to the TV's HDMI port as it works great for movie playback etc.

Edited June 1, 2017 by ThaCrip

[ThaCrip]

Well i played around a little bit with things and currently i settled on one Cougar fan for intake, like i already mentioned above, and then the other Cougar fan as a exhaust (i just installed it to the back of the case with two of those rubber things(it comes with four but seems unnecessary to install the bottom two and while not super hard to install they are not easy either) and then i removed that plastic cone thing on the side of the cases door and taped those holes closed on the inside of the case door in a attempt to help it to draw more air from the front of the case etc.

 

before settling on the above setup for now... i even tried blowing air into the case from the back, so it was a intake instead of a exhaust (so basically i had two 120mm fans as intake with no exhaust and could feel air being pushed outside of the holes on the side of the case), and from what i could tell this 'may' have slightly lowered the CPU idle temp (but i am not going to claim this concretely as overall i would say it's negligible) but the 5TB HGST seemed to raise a little bit('maybe' 2-3c) in temps which means overall it seemed to get a little worse, if anything. so then i just flipped the fan over, so it was back to exhaust like usual, and then tapped up the holes in the side of the case like i explained above.

 

also, after installing that other Cougar fan i got in the exhaust, which is 1200rpm, i can definitely say it blows less air than the 1600RPM 120MM Scythe fan as there is a clear difference in air blowing out of the back of the case between the two as it's obvious when holding your hand in the general area of the air flow. the only thing i am not sure on is if me taping those holes on the side of the case closed potentially slowed air out the back of the case a bit on the Cougar fan or if the fan simply does not move as much air as the Scythe fan which i am personally leaning towards the Cougar fan simply does not move as much air as the Scythe fan as there is a clear difference in the amount of air they push which means one of those fans, likely the Cougar, is exaggerating their CFM claim (since both are rated around 60CFM) as there is no way those two are close (assuming any resistance on the Cougar fan is not slowing their air flow a bit) but it's probably not surprising given the 400rpm difference etc. but one little bonus for the Cougar fan is i think i can hear a difference in decibels between the two fans (but i am not going to say this definitively like i did with the air flow as there is obviously a difference there) which is a plus for the Cougar over the Scythe and i feel the Cougar cools 'well enough'. so given my setup, the Cougar is the overall better choice for me (assuming it lasts a long time like the Scythe fan).

 

i just thought i would post this little update on me playing around with things

 

p.s. hopefully things stay similar with temps once i get a video card upgrade as i am thinking ill eventually go to the 'Geforce GTX 1050 Ti 4GB' as it seems like a solid upgrade to what i got with solid power for current games and price seems reasonable, probably that sweet spot of price vs performance, and gets it's power from the motherboard so it's not drinking tons of power. from what i can tell it's the best all around video card you can get for $150 or less (seems to be $135 or so at the moment) as that's probably my best all around option short of building a whole new PC as it appears even with my underpowered i3-2120 CPU i can pretty much run modern games decent enough as i figure it's probably not worth investing into another CPU as that money (which is probably around $100 or so) is probably better spent on a whole new PC build and at that point, and depending on price etc of things, i might just shift the 1050 Ti card into the new build and reinstall my current Radeon 5670 back into the current PC. just some thoughts. if someone is still reading at this point... congrats on getting through my novel

[goretsky Supervisor]

Hello,

 

If you do not secure a fan by all four corner points, it will bang the unsecured corners against the chassis.  These micro-oscillations will cause the fan to fail prematurely.  My previously employer found this out the hard way when they mounted case fans using two screws diagonally to save on manufacturing and labor (assembly) costs and then began to have their hot-swappable fan trays begin to fail prematurely after a couple of years (they were supposed to be good for five).  Spares were sold to customers for $386, who were understandably upset when their included fan trays failed so early and they didn't have a hot spare.  I processed quite a few RMAs on them.  Funny thing is, the fan cards were actually fairly inexpensive (around $7-10 in parts, as I recall), but it cost about twice that ($15-20) to ship out replacement cards with overnight delivery. 

 

Regards,

 

Aryeh Goretsky

[DevTech]

https://www.youtube.com/watch?v=NoAzpa1x7jU     

https://en.wikipedia.org/wiki/Tears_in_rain_monologue

 

"I've seen things you people wouldn't believe. Attack ships on fire off the shoulder of Orion. I watched C-beams glitter in the dark near the Tannhäuser Gate. All those moments will be lost in time, like tears in rain."

 

Giving some consideration to changing my standard cooling advice to "build a smoke chamber" since it's a fun project which also cures meat and detects global warming...

[ThaCrip]

@goretsky

 

so your saying, even though i can't see any obvious issues with it holding the fan by the two top holes on the 120mm fan, it still will have issues down the road?

 

when i touch the fan i can't feel any obvious issues with vibration etc as when i push on the bottom of the fan it seems to be fairly securely resting on the PC case. plus, the fan touches the case has rubber pieces between the fans main plastic shell and where it touches the steel case(these are on the fan itself by default) which i imagine would dampen any vibration, right?

 

side note... after i removed the side of the door a moment ago to feel that exhaust fan for any noticeable vibrations i noticed that the air flow, from that exhaust fan, seems to increase out the back a noticeable amount and it's a more even flow out the back. given this info there must be some sorta resistance with the case being more sealed after i blocked those two openings on the side of the case with tape. makes me wonder if i should remove that or try removing that purple plastic clip and connecting the front case fan directly to the steel on the case (with the rubber pieces) as maybe that plastic clip interferes with intake airflow a bit and maybe chokes off that exhaust fan a little. or maybe i could consider removing some of the tape from the lower holes on the case etc. also, i even noticed it increases the fan output on exhaust when i have the door on but have the front of the case removed as it makes it easier for the exhaust fan to draw air in from some of the front side of the case that's more open.

 

right now i am leaning towards attempting to remove that purple clip and connecting fan directly to the case with those rubber pieces first and see what happens and then go from there.

 

any suggestions or is it one of those things i am just going to have to play with?

 

thanks for your time.

Edited June 8, 2017 by ThaCrip

[goretsky Supervisor]

Hello,

 

The body of the fan is going to be bouncing up and down against the chassis since two of the mounting points are not secured.  The distance is likely microscopic in nature, and the amount of force minuscule and it will take a long time before any problem begins to manifest.  But over tens of thousands of hours of use this will add up and the fans will fail before their rated lifecycle.  In our case, I think we were using ball bearing fans (I don't think sleeve bearing or brushless fans had 50,000 hour lifecycles a dozen years ago), and customers were getting about 40% usage of the fan trays before they died.

 

Our fans were the standard plastic shell screwed directly to a metal frame, no rubber padding.  Whether or not the rubber corners dampen any shock or amplify it probably depends on the fan being secured properly in the chassis.

 

Frankly, though, you're talking about a part that costs maybe $10-20?  If you don't want to go through the hassle of installing the fan correctly, you can just replace it when it dies.  These are not expensive parts, and replacements are readily available.

 

As far as for temperatures inside, my suggestion would be log the temperature over time before you make any changes, then make once change at a time, spending some time between each change so you can review the logged data.

 

Regards,

 

Aryeh Goretsky