The Official Cooling Thread!

Arctic Silver V, hands down.

 

how well does it cool?...very well...

lol this is funning yesterday i'm cleaning my computer and i checking if theres still thermal paste...so i took of the fan, but just couldn't take off the bloody heat sink...its stuck so i pull it out and the cpu stuck on it and i'm like oops...the paste is so dry...

 

What you probably have there is thermal epoxy. That can be an absolute @&^*&^! to get rid of if you want to put new stuff on.

 

GTR35,

Arctic Silver 5 or Arctic Silver Ceramique. I've done tests on 4 different platforms now and they work equally well. If you choose the Ceramique, put it in the Microwave for 4-5 seconds as it flows easier. It's sticky as hell! LOL!!

Best Regards,
theone :>)

 

hi all

running a Gigabyte 965P-DS3, bios F5. My 'System Temperature' is running at between 45-50'C. I was just wondering if this is 'safe', as I remember back in the days of Athlon XP's hearing something baout a max of 40'C for the system temps.

If my current temp is fine, anyone got any insight into the max possible safe temp?

Cheers in advance

 

stueyt,

Buy yourself a Silenx 40x20x20 6cfm fan and you will be able to hold it down with a couple of screws. Make it blow down into the heat sink. If you want to do it right, you may have to pull the motherboard as it's almost impossible to get it off without doing that, and clean the old hardened paste. I use my fingernail on the chip and a small jewelers screwdriver on the heat sink. Clean both with some Isopropal Alchohol and put some Arctic Silver 5 or Ceramique on the chip and re-install it. My MB temp dropped 13C that way with the same MB. I went from 47C to to 34C and it even got lower as the AS-5 cured!

Part of the problem is how close the heat sink is to the CPU socket. An Arctic Freezer 7 Pro overhangs it a bit and you get a lot of heat migration that way. With the fan blowing down it keeps the heat from rising and adding to the overall heat already in the CPU cooler The CPU temp will go down about 8-10C too!

What Revision motherboard do you have? I had Rev. 1.33 and the F-10 bios was the best for me!

Best Regards,
theone :>)

 

yo bros...i've been researching on fans...its most likely to be case fan...

Noctua NF-P12
Noiseblocker 120mm XL1

i haven't heard of these brands before tho...i'm planning to use one of these fans as CPU fans...for thermalright ultra 120...what do you guys think

 

GTR35,

I know Sam's gonna scream but I'm just not sold on the Noctua. I don't like the teeth on the fan blade for one and why don't the give the CFM of the fan instead of that fancy formula they use. In a very dusty environment like where I live, those little teeth would quickly collect dust and it would get off balance because dirt would eventually be in the gap of the teeth and fly off in spots. That's why I dumped the Zalman 9500! The round design was it's fatal flaw for me. Dust would rapidly collect in the "V" of the fins and spread outward, blocking the airflow where you needed it the most. Looked like a freekin carpet in there! LOL!! Besides, if you are doing any serious overclocking the Noctua's speed will have to be raised anyway, so what's the point!

I'm from the school that says get it as cool as you can get it, within reason. Mine is no where near as quiet as Sam's is, but considering it sits to my right on top of my desk and is only 3 feet away from my right ear, I don't even really notice it. I couldn't settle for 10C warmer when I know that it's so easy to drop that CPU temp below 30. It's all about getting the heat out of the case. Even though I'm 63, I have excellent hearing and my high frequency hearing ability is over 21 thousand cycles! I wouldn't buy an aluminum case knowing what I know now, so I don't get any annoying vibrations or funny noises and surges. I think the fact that the fan speeds are steadier than with Sleeve bearing fans and the computer always makes the same noise helps too, as you get used to the sound and don't even notice it! I tried at least 3 different fan setups before settling on the Silverstones. I played with all sorts of fan speeds and directions of airflow, and I think I've gotten it right. It should last a long time without any problems related to heat. Like I said, it runs at 4.0gHz but I'll choose to run it about 3.8gHz. Keeps the voltage down and heat is not an issue!

Just my opinion. It's like i said to Sam the other day! We're both right! For him it's silence, for me it's better cooling. Both work real well!

Best Regards,
Russ :<)



I play Delta Force Extreme, All 3 Quakes and A.O.E. and everything remains cool. With the new MB it hits 42-44 while encoding with DVDRB/CCE and barely hit 50C with Orthos

 

@theonejrs so which fan you talking about?...

all i know is that Thermalright Ultra 120 Extreme is good...but i don't know any good 120mm fan to put on it?

 

GTR35,

I was talking about the Silverstone FM121
http://www.newegg.com/Product/Product.aspx?Item=N82E16811999344

Extremely good airflow at low rpm. No need to run them over about 1200 rpm for what you want it for. I will move more than enough air, for sure I know Sam's gonna yell at me but I don't care! I have some of the lowest temps on both forums I belong to and I like it. Should be perfect for the Thermalright Ultra 120. I've tried Sythes, the Thermaltake Thunderblade and these Silverstones and these win hands down! Just don't use them for a case fan in an aluminum case unless it has the Nickle plated rear where the fan mount is. I am so glad I didn't buy that aluminum case I was looking at a year or so ago. I should have known better anyway as I once had an aluminum bodied Jaguar years ago and it always had mystery noises and weird vibrations! Aluminum cases are a pain in the A$$ when it comes to noise and strange vibrations! Comes with the speed control electronics built into the fan, so no controller needed. There's a tiny Potentiometer (like a volume control knob) to adjust the rpms. Comes with a bracket for a floppy bay but I prefer to either mount it in a 5.25" bay plastic blank or drill a hole in the back somewhere. Once it's set, you usually don't have to touch it. I use one as my rear case fan as well!

Best Regards,
Russ

 

Theone and GTR: Amazingly I'm not going to scream. I dislike 120mm Noctua fans. I HEAVILY dislike Noctua fans. The reason why is that whilst the fan blade noise is second to none for a mid-speed 120mm fan (the 1200 model anyway), a 'glitch' with the bearing system means that after about a week, any 120mm Noctua fan will produce a 550Hz whine at a similar volume to the fan itself. That is a high pitch, and very annoying. You will do much better with a Nexus RealSilent case fan (those are VERY quiet) or a Scythe S-Flex. I use a combination of both of these, and on their own they are inaudible in a near-silent room. The only noise my PC makes that's audible is the HDDs (and barely) and the PSU fan.
The Thermalright Ultra 120 is a fabulous heatsink, and will work wonders even with a silent low-speed fan, but avoid the Noctuas. With regard to Noiseblockers I believe most of them are rebadged Coolink fans, which are OK, but often very prone to vibration (they certainly are if they have the translucent frame rather than plain black). Without question, Nexus and Scythe have got it sewn up in the silent fan arena.

For reference, a Scythe S-Flex fan at 1200rpm on the Ultra 120 will push around 80% of the airflow of an FM121, but be pretty much silent at doing so. If you're after quiet, you'll have to turn the silverstone down far lower than that, and you'll end up with less airflow overall. In a vibration-free environment, they're about equals (assuming you run the 121s at a lower speed), but the higher pitched noise of the silverstones, coupled with the much increased vibrations means they aren't a competitor in a realistic quietPC arena.

 

sammorris,

I tried one and the results were not good. I had good cooling and all that but the rpms changed a good bit varying up and down and the resulting speed change was annoying, a problem that got worse as you lowered the speed! This is what I bought:S-FLEX SFF21F and it is loud! Granted, at lower rpms it runs quieter as well it should considering it's 28.0 dBa at 1600! It doesn't push 80% of the of the FM121 at 1600rpm (max)! It doesn't push quite as much air at 1600 than the Silverstone does at 1200 rpm! At 63.7 cfm at top speed, it also didn't cool as well as the Silverstone at 1400 rpm, but it was also about 10 cfm less and was loud doing it!

To date I've tried the stock Cooler Master fan, the Thunderblade and the highest airflow S-Flex and a Generic I had (just for grins) and nothing has worked as well as the FM121. The Cooler Master was reasonably quiet, but didn't move enough air. The Thunderblade roared. The S-Flex doesn't move enough air even at full chat and makes enough noise to get anyone's attention (maybe even some dead peoples)! The FM121 will make more noise at top speed but there's no reason to ever run them that high, ever! The generic was a joke, although it was extremely quiet. It should be as it was less than 20 cfm! LOL!! I got a big kick out of the model number, POS-1! ROFLMAO!!

I'm going to do some experimenting with lining the case with acoustic foam. It's what we use at the office to line the sound chamber, and I'll let you know how that works out.

My air intake inlet is on the bottom of the computer and I cut up strips of the soft foam that usually comes with motherboards to keep the Anti-Static bag from getting holes in it. Since the opening is just a simple rectangle, it works great. It has a grill with holes in it so it can't go anywhere or be sucked into the fan!

Best Regards,
Russ :<)

 

I use the SFF21F and its speed seems pretty much dead-on. Of course I don't know that, because I can't hear it. The S-Flex produces 31dB of noise at 1600rpm according to SPCR, and on that scale, I'd say the FM121 must be a good 35dB at 1600rpm. It produces far more air, but to me, that level of noise is unacceptable. Of course, the noise of the Scythes at full speed is also unacceptable, but you know me...
In order to get the Fm121s inaudible, I had to drop the speed to only 500rpm. At that speed, the efficiency of the blade design starts to drop away, and I could barely feel the air coming out of it. I can run the S-Flex at 750rpm and not hear a thing. There is certainly more air coming out of that than there was from the FM121s. I'm alarmed that you found the noise of the S-Flex loud, since you run your FM121s at a speed that produces the same noise level (1400rpm).
I still have my fan test rig set up, so I'll set them to the relevant speeds and upload the result.

 

sammorris,

Maybe it's just your case! If I pull the cover off and put my ear about 6" away all you hear from the FM121 is a gentle rush of air. I didn't even use the Silicone mounts.

Unless you are saying that you were comparing the S-FLEX and the Silverstone both at 750 rpm. The FM121's slowest speed is 800 rpm and makes all sorts of noises when you use an external fan controller. I dug out an old fan controller I had the other day after you mentioned your experiment and both of the Silverstones sound horrible using it. Very annoying. Right now with the new MB the loudest thing is the Freezer 7 because it now runs about 1 to 1.1 thousand rpm instead of the 700 to 800 it used to idle at! The price of progress! LOL!! Still, considering it sits right at ear level on my desk (which should make a pretty good sound board), it's reasonably quiet. I had all 3 of my neighbors kids move away from the computer until the couldn't hear it and it's dead silent at about 9 feet and there is almost no vibration at all. Considering that there is 7 fans running inside it, that's remarkable. If I fill up a soup bowl with water and set it on top of the computer, there isn't so much as even a faint ripple and you can barely tell it's on by touch anywhere on the case surfaces. Very Smooth! The most noticeable sound is the air rushing in the static air vent on the side cover, Since it will hold a piece of bond paper without falling, I can see why! LOL!!

Best Regards,
Russ

 

I ran the FM121 outside (no turbulence) and I can tell you now, the Freezer 7 pro is pretty much inaudible at that speed. It's certainly inaudble compared to the FM121 at that speed.
What rpm do you run your FM121 at? I think you must be mistaking the noise it makes for something else.

I gave some scientific thought as to why you prefer the FM121, and it now makes more sense to me. Though the airflow out of the S-Flex is lower (by around 20% I'd say for the same rpm), having 7 big blades rather than 9 smaller ones means the air pressure is much higher. Consequently, with a standard fan grille like the one on the back of your Coolermaster case, the fan will make much more of a 'rushing' sound as the air is blown through the mesh, which is interrupting the airflow. Through a traditional concentric ring fan cover, that noise would be halved. It's also worth noting the vibration issue. Unless you use a steel case with anti-vibration gromets, the FM121 is always going to appear louder because the frame vibrates more (On a desk, the Fm121 is louder at 800rpm than the S-Flex is at 1300rpm because of this). Concluding that therefore, I think the S-Flex is the better application for a PC case. At a given airflow it's quieter in most people's cases, probably partly because of the Fluid Dynamic bearing. The rpm of mine doesn't vary in the slightest. The higher air pressure means it's also suitable for applications like heatsinks whilst running at low speeds (and making less noise).
Don't get me wrong, I like the idea of the FM121 (why do you think I bought three of them rather than just the one), as a fan thats adjustable from near-silent to blow-your-face-off airflow, and including its own fan controller for if you haven't got one. However, for me, and for a lot of others, I simply think the S-Flex is of more use.
I ran the FM121 outside (no turbulence) and I can tell you now, the Freezer 7 pro is pretty much inaudible at that speed. It's certainly inaudble compared to the FM121 at that speed.
What rpm do you run your FM121 at? I think you must be mistaking the noise it makes for something else.

I gave some scientific thought as to why you prefer the FM121, and it now makes more sense to me. Though the airflow out of the S-Flex is lower (by around 20% I'd say for the same rpm), having 7 big blades rather than 9 smaller ones means the air pressure is much higher. Consequently, with a standard fan grille like the one on the back of your Coolermaster case, the fan will make much more of a 'rushing' sound as the air is blown through the mesh, which is interrupting the airflow. Through a traditional concentric ring fan cover, that noise would be halved. It's also worth noting the vibration issue. Unless you use a steel case with anti-vibration gromets, the FM121 is always going to appear louder because the frame vibrates more (On a desk, the Fm121 is louder at 800rpm than the S-Flex is at 1300rpm because of this). Concluding that therefore, I think the S-Flex is the better application for a PC case. At a given airflow it's quieter in most people's cases, probably partly because of the Fluid Dynamic bearing. The rpm of mine doesn't vary in the slightest. The higher air pressure means it's also suitable for applications like heatsinks whilst running at low speeds (and making less noise).
Don't get me wrong, I like the idea of the FM121 (why do you think I bought three of them rather than just the one), as a fan thats adjustable from near-silent to blow-your-face-off airflow, and including its own fan controller for if you haven't got one. However, for me, and for a lot of others, I simply think the S-Flex is of more use.

I ran the tests on a carpet...

My PC has two S-Flexes in it, though. I turned them up to full speed and walked away from it. At a distance of 10 feet they were inaudible.

 

sammorris,

It's always set to 1400! I get my best temps that way!

My turn to teach a little! LOL!! Air pressure is a function of cfm vs restriction. The more restriction, the higher the air pressure. Higher pressure means lower cfm and more noise because if you restrict the airflow, it creates backpressure. That's why some of the cheap cases with the fan mounting holes for 80 and 120mm fans don't work well, especially the ones that only have vent holes for the 80s. The cfm determines how fast the air moves and how well it can remove the heat from the case. The number of blades doesn't really matter much although the higher the blade count, the smoother it should turn. The reason is that all fan blades pull on the hub. The more blades the more evenly that pull is distributed within the 360 degree circle.

I could make mine much more silent than it does, but the trade off would be about 10C higher temps. When I determined the fan speeds that cooled best, I ran OCCT and adjusted the fan speeds down until the temps started to rise and then raised them slightly to the next thousand mark above where it had started rising. If you adjust them at idle, there isn't enough airflow to remove the heat fast enough when things start heating up so that's why some people get very high temps while gaming. Enough airflow for low temps, but not near enough when the temps get high!

Mine will never be as silent as yours, but it's still a great improvement over the thunderblade and the Noctua. At 100+ cfm there's a considerable amount of airflow through the side vent in the cover. I'm still going to try the Acoustic foam and see what it does. Can't hurt! LOL!! I already know that the air noise is here to stay! LOL!!

Best Regards,
Russ

 

The phrase "Pulling on the hub" suggests vibration. However, the frequency most fans produce vibrations at is not the blade frequency (rpm *7 for the s-flex, or rpm*9 for the silverstone), it is in fact the pole frequency (rpm * 4 for each pole of the magnet). If you manually try and turn the blades of a fan very slowly, you'll find it naturally rests in only 4 positions each revolution, with a noticeable 'thump' as the blade assembly gets pulled to the next 'stable state'. Consequently this 'shakes' the frame, or creates a pulse of vibration. Since this happens four times every time the fan blades go round, the main frequency of Vibration (excluding the harmonics) for the silverstone, or indeed any fan at 1400rpm is 93.3Hz.

As for the pressure argument, as a designer of devices that work solely on the basis of air pressure you hold the information high ground. However, I always assumed that pressure is what overcame blockages, not what formed from them. I figured since what a fan does is push air, the larger the blade is, the more violently the air gets beaten through the frame, thus keeping the pressure up.

As far as silence goes, I know you live in a hotter climate than me, but your PC puts out significantly less heat. I'm sure you'd be able to get your PC quieter than current and still have perfectly reasonable temps. However, the low fan speed of your Freezer 7 may be holding you back. Although the E6750's lower heat output gives you an advantage, I found that with my overclock, keeping my Freezer 7 at anything below 1500rpm resulted in high temps (55C or above at load). With my Thermalright, I can run the fans at any speed I like and not have any issues, as long as they are actually turning round.

 

sammorris,

Maybe I said that wrong. More blades spread the load on the hub around so the hub doesn't distort as much as the overall load on any individual part of the hub is reduced. 9 lighter blades do not cause as much stress as 7 heavier ones. There are limits that affect efficiency as with too many blades you couldn't get them out of the way fast enough to keep the trailing blades from smacking the air coming off the blades in front of it, making noise. The whop whop whop sound of a helicopter is made by just that. A few years back some fan manufacturers actually made blades with a solid circle of plastic around them to try and cut down on vibration. The magnetic effect you mentioned doesn't really come into play as noise, as it doesn't take very many rpms to negate it. All motors work that way in one way or another. Fans for computers are DC shaded pole motors. Some motors have brushes that run on an armature that has segments separated by thin Mica strips and do essentially the same thing as to magnets in a fan motor do. Brushless induction motors use the 60 cycle (50 cycle in England) frequency by the reversing of polarity to accomplish the same thing. What does affect motors like the Noctua is the magnetic seal as it runs in contact with metal, even though it's coated with something like ceramic or teflon. They suck in a real dusty environment. You will notice that they are rated for 150,000 hours. Problem is they are not guaranteed for that long so what good is it! Theoretically they should last for 17+ years, so why is the warranty 1 year! It sounds good but it is just an advertising ploy. Especially since there's only 8,760 hours in a year! LOL!!

I would hope so! Look at the cooling surface area of the monster! LOL!! The bottom line is that the Freezer 7 does a great job of cooling. Remember my temps on the 965P with the E4300? Gina's temps weren't like yours with the same MB and CPU! Something wasn't right. I don't know if you got a bad Freezer 7 or what but I set Gina's up with the stock Intel cooler with no temperature problems at all. She installed the Freezer 7 after I shipped it to chicago!

Case design means a lot too, even down to the location of the fans in relation to the CPU. On mine the side 80mm Silverstone is almost directly on top of the Freezer 7 if the case was laid on it's side. blowing out it draws heat away from the Cooler itself without interfering with the HSF's airflow. Makes all the difference in the world. The big problem is that there are no standards for airflow in cases. I've seen tons of cases where the cover fan is smack in the middle of the window, where it doesn't do much good. A lot of window cases also lack static vents, so the video cards suffer. My case is far from the most expensive and isn't as pretty as others, but it is more functional than many when it comes to running cool. When buying a new case these days I've learned what kind of fan layout a case needs for good cooling so when I buy one, I know what to look for!

Best Regards,
Russ

 

Hmm, the Noctua fans didn't do very well in my system. They were quiet as fans, for about a week, then the high pitched whine put me off.
As for the pole frequency being negated at higher speeds, not so I'm afraid. Testing this by tacking a fan to my desk to hear the vibration clearly, producing a sound using a signal generator at rpm*4 caused a heavy superposition effect. The loudest noise of the fan was rpm*4, which can only be the vibrations caused by pole movement. I wouldn't have thought it previously, but evidence suggests otherwise!

As for my Freezer 7's performance, it was probably due to the setup I had. I was still using the low speed fans I have now, and the cooler pulled hot air off my fanless graphics card and the hothead chipset from the P5N-E, exhausting it into the power supply. There was hot everywhere! lol
Now with the Ultra 120, the CPU cooling is isolated from the other components as a strict barrel of airflow!

The fan arrangement in my case works rather well, the side fan is at the right position to exhaust away from (or intake to, if you're so inclined lol) the graphics card. The front fan cools the hard drives well, and the rear fan cools the CPU et. al. The top and bottom fan slots can be used for additional ventilation.

 

Sam,
This is what I'm trying to explain. All DC motors run on the same principles! Some are run using magnets,field windings and an armature and some are run using just the field windings and an armature with brushes doing in effect, the same job as the magnets do. They all work the same. Most computer fan motors have fixed magnets, fields and an armature and are brushless! That's why the the fan staggers when you turn it by hand. The design has nothing to do with vibration if run at the proper voltage! It doesn't sit there going bup bup bup! I do know that all DC motors work on different variations of the same theme! All of them depend on their momentum to run smoothly just like a flywheel on a car engine. They wouldn't run at all without it. The only real difference is how that magnetic pull is created. with a brush motor, it will spin freely when turned by hand and the power is off. Put power to it and it has the same stagger but is more precisely controlled at any given speed because of the multiple segments on the armature, they still have their power points fixed at specific points in the rotation!

That's why voltage controllers for most case and computer fans are junk! They lower the voltage and the lower the voltage the more vibration you get. That's why we have PWM control for most cpu cooler fans these days. The voltage remains the same but the point that the pulse occurs changes, but the change is even, as the electronics switches the power between the different field windings to smooth it out. That's where the electronics like in the Silverstones comes into play. That's also why most PWM motors won't run below 800 rpm!

When you lower the voltage on a regular fan motor you create more vibration because you just put a tiny hitch in the pulling parts. It's not as strong a pull as full voltage from the fields, yet the magnets, being what they are don't lose any power so their pull is fixed. It's a little bit like spark timing in a car. The electronics change the point where the pull of the fields occurs to smooth out and counter the pull of the magnets and keep it smooth. This allows the motor to run slower with minimal vibration. With conventional brushless motors it does everything at a fixed point in the rotation regardless of the speed. This causes a slight jerk in that rotation where the motor and the magnets are fighting one another for a fraction of a second when you lower the voltage, which translates out to vibration. It's that slight change in pull that does it as the magnets pull one way and the fields pull the other. The forces fight one another and gets worse as the voltage is lowered.

Best Regards,
Russ