Does high RPM have that much impact on Cooling?

Toggle sidebar Toggle sidebar
C

charmer

Honorable
Aug 18, 2013
477
0
10,960
Hello everyone,
I was wondering if a high rpm fan will cool down my system better then a mid rpm fan.
I ask it because i have x2 12'' 2000rpm reference fans on my close loop water CPU cooler (its a ThermalTake water 3 performer), BUT they are noisy as hell! so i'm thinking of getting x2 Noctua NF-F12 but they run at 1500rpm so i'm afraid it will affect my cooling.

Can anyone please tell me if that extra ~500rpm will have that much impact on my system or do you have any other solution?
 
Sort by date Sort by votes


I disagree, a fan with high CFM in open air may not have sufficient static pressure to push air through an object like a radiator. Static pressure is of significant importance too.

http://www.overclock.net/t/737319/static-pressure-vs-cfm

The blade and structure design, determines the noise for a given rpm/pressure/flow.
 
Upvote 0 Downvote
There are 3 basic types of fans.
1. High CFM. These are typically case fans, their blades are designed and pitched to move a lot of air, usually very low static pressure. Usually 9+ thin blades.
2. Average fan. These fans move less air than a High CFM, however, have higher static pressure. Usually 7-9 mid sized blades.
3. High Static Pressure. These fans are typically radiator cooling fans, and their blades are designed and pitched specifically to push a limited amount of air with more force. Usually 5-7 very wide, cup shaped blades.

All fans have several things in common.
1. Noise. The faster a blade moves the more turbulence it creates, creating more noise. Many of the better fan companies specifically design their blades to eliminate as much of the turbulence as possible, special ridges, planes and angles, spikes.
2. Speed. The faster a blade spins the more air will move over its surface, creating higher CFM and more pressure, up to a certain point. After that point is reached, the turbulence from the blade tips actually starts to hinder the airflow and pressure, creating a backfeed and lowering the output. This is why some fans spin faster than others.
3. Architecture. Pretty much most fans of the same basic shape, do the same basic job. Only manufacturers who go above and beyond that will have better performance. This is usually indicated by better bearings, performance reviews, quality of manufacture etc.

The NF-F12 is a high static pressure fan designed for very high static pressure. Primary purpose is radiator cooling.
The NF-P12 is an Average fan designed for good static pressure and good CFM. Primary purpose is CPU tower radiator cooling.
The NF-S12 is a High CFM fan designed for maximum airflow at low noise. Primary purpose is Case fan.
 
Upvote 0 Downvote
Pretty much so yes, if you are trying to cool a radiator or tower heatsink. At 1500rpm max, it will also be considerably quieter. I have an NF-F12 on my Corsair H-55 single radiator. It keeps my 3570k OC@4.2GHz to about 32*C at idle and 64*C under prime95 small fft. At idle it spins @300rpm. Full CPU load and it runs @1050rpm. It doesn't max out. That's what static pressure will do for a radiator.
 
Upvote 0 Downvote
Pretty much all the information in this thread is wrong and should be disregarded. OP, the short answer to your question is that yes, by far the single biggest determination in the cooling performance of a rad fan will be the RPM. The best fan in the world running at 1000RPM will never cool as well as a crappy fan at 2000RPM (assume same size fan). That is just how fluid dynamics work.
The long answer is that the cooling performance of a particular fan is determined solely by how much air it moves. ALL the numbers you see are only important in so far as they affect this number. And the overwhelming majority of that number is determined by how fast the fan spins. "But Deuce65, what about this static pressure everyone keeps talking about? They all tell me this is important to look at for rad fans". And it is, but the number printed on the box is almost always useless. The rated static pressure of a fan tells you the maximum pressure the fan can overcome (to be precise, how high it can force a column of water) . But you aren't going up against the maximum pressure (which would be a brick wall), you're going up against a radiator with vent holes in it. Likewise, the max cfm tells you the airflow with no resistance at all, but there will be resistance (we are putting something in front of the fan and even if we didn't, there is air resistance). The REAL airflow of a REAL fan in a REAL situation can't be described by one or two static numbers; it is a non linear equation and the value lies somewhere along a plot of that equation. And because it is non linear the two endpoints don't tell you anything about what numbers are between those two endpoints.
To give a real world example, i'll use two fans I own, a cougar vortex pwm and a scythe gentle typhoon ap-15.
The Cougar has a rated max airflow of 70.5 cfm and a rated static pressure of 2.2 mm h2o
The Gentle Typhoon as a rated max airflow of 58 cfm and a rated static pressure of 2.03 mm h2o.
Looking at that one would think the Cougar is a much better rad fan, having better values in both. And yet the Cougar is a terrible rad fan (great case fan though), while the Gentle Typhoon is widely regarded (correctly I should add) as the best rad fan out there. How can this be? Simple, If you look at a P-Q graph of actual real world airflow values the Typhoon is much higher everywhere along the graph EXCEPT at the two endpoints (which are not real world values, as explained earlier). Unless you are going to take the time to track down the actual P-Q curves of the fans you are considering, looking at the numbers is completely pointless and does not give any indication at all of how well they will perform.
Anyways, what that has to do with your original question is, yes. If you swap out to 1500RPM fans, even very nice, expensive ones like the Noctuas, you will have worse cooling performance. I can guarantee you that, and I don't need to see the static pressure or max cfm rating of your stock fans to tell you that.
 
Upvote 0 Downvote
It's funny, I swapped out the crappy 1600rpm stock corsair fan for an expensive noctua running at 1000rpm, and my cooling went from high 80's to low 60's, and yet I lost 600 rpm. So it makes me wonder how it is that crappy 2000 would be better than expensive 1500 that is supposedly designed for rad cooling?
 
Upvote 0 Downvote


I would really want to know the answer to that as well.. Noise is still an issue and the db' is a real measured number, Noctua NF-F12 has around 19db' under load while my current 2000 rpm Thermaltake fans run around 32db'.

The conclusion is that 500rpm add 70% more noise..

EDIT: I forgot to mention that my rad has 2 fans on each side so it's a push and pool job, I don't know if it matters so I'm just saying
 
Upvote 0 Downvote


No you didn't. You're simply making up numbers to try and prove a incorrect point. Anyone here who has a corsair AIO can confirm this. If you were truly running upper 80s celcius with the stock fans turned up, you were exceeding the limits of the cooler. You didn't magically drop 20 degrees by swapping out for SLOWER fans pushing less air.
Ive tested the corsair h50i fan and it pushes just under 500fpm through the rad it comes with. Noctuas at max speed (to say nothing of the reduced speed you ran them at) push.....a little under 500fpm through the same rad. Same amount of airflow through the rad = same temperature. Incidentally, those corsair fans that come with their rads are actually quite good with airflow, they're just noisy as crap. Now in that regard the noctua is much better. But if he is concerned about noise quality he would still be better off getting AP-15s or AP-14s, both of which push more air through a rad then a Noctua, do it quieter, and cost half as much.
 
Upvote 0 Downvote


Well noise quality is where the expensive fans outshine the cheap fans. They will typically push much larger amounts of air at a lower noise volume then cheaper fans. If that is your goal I would suggest the AP 15s I mentioned, noiseblocker B-12 3s are also very good. Both will run near the same speed as your current fans, and will push similar volumes of air, but do so much quieter.

I would like to add too, that you mentioned you have a push pull arrangement. That makes the choice of fan even less important when it comes to strictly airflow (noise still would be improved with better fans). When you use a push\pull arrangement, what you are doing is increasing the "pressure" of the fans, i.e. their ability to get air through the rad. Making the already useless static pressure number even less useful, if that is possible.

Incidentally, if you have a chance, I suggest you google martinslab. Not sure if it still exists, but he basically took each and every fan he could get his hands on and actually measured the actual real amount of air each fan pushed through various radiators, at various speeds, noise levels, etc. So you can get some really accurate numbers from there.
 
Upvote 0 Downvote
Since you started spouting MartinsLabs as a source, is suggest you go back and actually finish reading the postings there. It was stated quite plainly that in lower rpm applications (below @1500 rpm) pull was a more effective method in regards to single fans on a radiator, and push (above @1700) was better. This also was only true for fans with a high enough SP to actually do the job.
When it comes to 'fluid dynamics' as applied to airflow around an obstruction, its also a well known fact that high cfm means crap without the corresponding SP. You can have very high CFM in an unobstructed path and it will work fantastic as a cooling medium, throw a radiator in front of it and it'll do nada. This is because the fin density creates more turbulence than the fan has ability to over come. It cavitates, and does nothing, no mater how much rpm you throw at it. Static pressure is needed to force the airflow through the turbulence, especially onthicker radiators like the h80 uses.
As far as stock Corsair fans go, the fans you are using in your argument are the ones rated onthe h100i, which in fact are quite respectable, if loud. This is not the same fan that comes on the h55 I have. The older h50 also uses a different radiator, with lower fin density, so you can throw that argument out too.
As for calling me a liar and accusing me of making up numbers to prove a point, you may want to re-think that. I am no 'punk kid' still in high school who thinks he knows everything. I do hold a degree in electronics engineering and have done so for many years. I was tinkering with computers when you were probably still in diapers. My first computer, if you can call it that, was a Vic20 with space invaders on a tape drive that took 10 minutes to load. My father in law was a WW1 vet 'prop man' whose sole job was the maintenance of propellers on aircraft, so, I have a somewhat unique perspective on what a fan does and will do. I can read. It's quite easy. If I felt inclined, I would tear apart my pc, hook up the old corsair fan and post the results, 87*C under prime95 small fft after 1 hr @4.2GHz, vrs the Noctua at 64*C, both ran under am ambient room temp of 73*F after the computer was idle for several hours. I don't feel that inclined.
Conclusion: Don't assume that just because you have a little knowledge, that everyone else isdumber than you are
P.s. sorry for the typo's, this is a pretty sorry excuse for a cell phone
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom