Best 80mm 4-pin CPU fan?

[Shrek]

I am wondering what 80mm 4-pin CPU fan replacement to get; long life being a big issue for me.

I was thinking about a Noctua till I realized they only consume around 1W, so thought to seek advice here.

 

[Bill_Bright]

I was thinking about a Noctua till I realized they are only around 1W

The wattage it consumes is irrelevant.

Just to clarify, you are only replacing the fan - not the heatsink too?

Assuming compatible voltage and physical dimensions, what matters most with a cooling fan is how much air it moves in cubic feet per minute (CFM). Next (at least to me - because I hate fan noise) is how much racket it makes when spinning at full speed. So check the dB levels too.

As for long life - with a quality fan, it really is immaterial these days. It is much more likely you will retire the CPU due to obsolescence long before then. For this reason, bearing type is of much less importance these days too. Even a quality sleeve bearing fan can provide very quiet (important since I hate fan noise), vibration free service, and long life.

Last - do note that box fans come in different thicknesses too. If your case is a slim model, you may need to get a lower profile fan. But if you have lots of space, I recommend the thickest (tallest?) 80mm fan you can get (that still meets your other requirements - like low noise). A taller fan will have wider blades. A wider blade can scoop up and push more air at the same RPM as a thinner fan. This means a quality wider blade can move more air much more quietly (did I mention I hate fan noise?) than a thinner fan.

I hope you noted that fan noise is important to me! But note too I also kept saying to get a "quality" fan.

 

[mechtech]

Hard to go wrong with a noctua fan.

More important than noise, life or power, could be application.

If this is going on a cooler, want to make sure it has the rpm (typically 3000rpm for 80mm) If it's a case fan then it's not as critical.

 

[Bill_Bright]

RPM is not really that important either. It IS a factor, but CFM is, by far, the most important.

Depending on the shape (pitch) and width of the blade, you can easily have a well designed fan spinning at 2500RPM move a LOT more air than a cheap fan spinning at 3000 or even 4000RPM. And it is the amount of air being moved that determines the cooling factor of the fan.

 

[Shrek]

Hard to go wrong with a noctua fan.

More important than noise, life or power, could be application.

If this is going on a cooler, want to make sure it has the rpm (typically 3000rpm for 80mm) If it's a case fan then it's not as critical.

The 80mm Noctua maxes out at 2,200 rpm

 

[eidairaman1]

Look at Scythe, Bitfenix maybe

 

[Bill_Bright]

Again - what matters is the maximum CFM, not RPM.

I have used several Scythe fans over the years - always been happy with them. Never used Bitfenix.

That said, I don't normally replace CPU cooler fans so I have to say, my actual experience is with case fans - though 80mm box fans should physically fit in either application.

 

[eidairaman1]

Again - what matters is the maximum CFM, not RPM.

I have used several Scythe fans over the years - always been happy with them. Never used Bitfenix.

That said, I don't normally replace CPU cooler fans so I have to say, my actual experience is with case fans - though 80mm box fans should physically fit in either application.

I use the bitfenix on my case.

My Cooler has 2 Scythes on it, I could actually pull 1 off and it be fine.

 

[Bill_Bright]

My Cooler has 2 Scythes on it, I could actually pull 1 off and it be fine.

That is probably because you have good case cooling!

 

[eidairaman1]

That is probably because you have good case cooling!

That I do lol

Scythe Ashura, 2×BitFenix 230mm Spectre Pro LED (Blue,Green), 2x BitFenix 140mm Spectre Pro LED

 

[The red spirit]

RPM is not really that important either. It IS a factor, but CFM is, by far, the most important.

Depending on the shape (pitch) and width of the blade, you can easily have a well designed fan spinning at 2500RPM move a LOT more air than a cheap fan spinning at 3000 or even 4000RPM. And it is the amount of air being moved that determines the cooling factor of the fan.

That ain't true. There isn't nearly as much delta between rpms and cfm between generic and "premium" fans. It's very hard to fuck up that much and difference between generic 7 blader and equivalent Noctua fan might be just 300-500 rpm tops. Most likely there isn't that much delta anyway. And you can't really design a significantly different fan in same space constraints either, so they are all more or less the same. You only have cfm and loudness metrics that matter. There's no way to make cfms significantly quieter, when you basically create turbulent flow.

 

[mechtech]

And if you want to get more technical on fans

Fan Affinity Laws

The affinity laws can be used to calculate resulting volume capacity, head or power consumption when speed or wheel diameters are changed.

www.engineeringtoolbox.com

also
you could look up forced convection in heat transfer text as it's a function of velocity, but that's getting overkill for a cpu cooler/fan I would think

Just look up the specs of the original fan and get something in the ball park and call it a day.

 

[68Olds]

I put four of these Arctic 80mm PWM fans in my very old Antec full tower. Nice improvement using the motherboard PWM to control speed over the old rattling three wire 80mm fans. They are only rated for 31cfm, but two intake & two exhaust keep the case nice & cool. A lot more affordable than Noctua.
ARCTIC F8 PWM PST - 80 mm PWM PST Case Fan with PWM Sharing Technology (PST), Quiet Motor, Computer, Fan Speed: 300-2000 RPM - Black : Electronics (amazon.com)

 

[Bill_Bright]

That ain't true.

It is totally true - and you would know this if you studied aerodynamics.

A fan blade is essentially a propellor blade or a rotating wing. Pitch, width and length all dictate how much air is moved. Yes, of course, RPM matters too. But once again just because this fan is spinning at some RPM, that does not mean it automatically is moving more air (has a higher CFM) than that fan.

If what you said was true, then these two 120mm fans would move the exact same amount of air.

Bgears b-Blaster 120 Case Fan - Newegg.com

Rosewill RFTL-131209B Blue LED cooling fan/Blue LED - Newegg.com

Now which of those two 2,000 RPM fans will provide the most cooling?

There isn't nearly as much delta between rpms and cfm between generic and "premium" fans.

Totally NOT the point of anything I was saying. My point for getting a quality fan had nothing to do with CFM or RPM. Rather that was about noise, vibration and longevity.

 

[The red spirit]

It is totally true - and you would know this if you studied aerodynamics.

I studied fans, so I know what I'm talking about. No it doesn't really matter much, when you normalize then by cfm. You always get mostly the same noise level and mostly small delta of rpms.

A fan blade is essentially a propellor blade or a rotating wing. Pitch, width and length all dictate how much air is moved. Yes, of course, RPM matters too. But once again just because a this fan is spinning at some RPM, that does not mean it automatically is moving more air (has a higher CFM) than that fan.

There's not much you can tweak, most fans use very similar blades or amount of blades.

If what you said was true, then these two 120mm fans would move the exact same amount of air.

Bgears b-Blaster 120 Case Fan - Newegg.com

Rosewill RFTL-131209B Blue LED cooling fan/Blue LED - Newegg.com

Now which of those two 2,000 RPM fans will provide the most cooling?

Without any testing done, I won't say anything about their actual performance, but that bgears is seriously inflating cfm metric. The thing about cfm or db metric is that there isn't one standard to measure them and they can measure almost anything and plop down bigger or smaller numbers how it fits them. Those specs are basically a scam and it has been true for very long time, especially with those less legit brands.

But just from looking at those fans, Rosewill's design seems to be more pressure oriented, so it might generate a bit more active pressure, therefore less cfms at higher rpms, but not more noise. Since Rosewill fan seemingly uses thin translucent plastic, blades are likely not as sturdy as on bgears fan and might warp or vibrate at full speed.

Either way, both fans are overpriced by at least two times. You can get Xilence, SilentiumPC, Fractal or Arctic 120mm fans for around 5EUR in EU or for 3 EUR some DeepCool fans. 12 dollars for some knock-off crap is way too much. Generic Gembird 120mm crapola is 3 EUR in EU, but I would just get something better than them.

Totally NOT the point of anything I was saying. My point for getting a quality fan had nothing to do with CFM or RPM. Rather that was about noise, vibration and longevity.

Noise is cfm mostly, unless you have failing bearing that grinds. Longevity is non-issue with fans since even basic sleeve fans last for over 14 years. Vibrations are hard to predict and you can't just tell from picture if it will vibrate or not, but generally, fans don't vibrate at lower rpms.

 

[DrCR]

Bill, don’t forget about whether the fan is LED or not. That can improve FPS by 5-10%, just like going from a beige to black case. DYI window mod can improve performance too, but not as much as factory tempered glass.

On a serious note, you could consider Phantek as a brand as well. My last experience with 80mm was Japaniflows (Japanese Paniflow), but that was forever ago.

You always get mostly the same noise level and mostly small delta of rpms.

That made me snort my coffee and I’m not even drinking any. Just for reference, those who study science tend to make use such assertions as absolutes. Best regards regardless to you though.

 

[Bill_Bright]

Bill, don’t forget about whether the fan is LED or not. That can improve FPS by 5-10%

LOL There you go!

when you normalize then by cfm. You always get mostly the same noise level and mostly small delta of rpms.

Total obfuscation just to argue against a point I never made. I NEVER said anything about "normalize by CFM". In fact, the "differences" in CFM values is my whole point. So you don't even know what this discussion is about.

There's not much you can tweak, most fans use very similar blades or amount of blades.

Total nonsense. There is lots you can tweak, and I have already mentioned them. They include width and pitch. Length matters too, of course, but Shrek has already stated this is an 80mm fan so not much wiggle room there. and to make the claim that most fans use similar blades just illustrates your lack of knowledge here.

Those specs are basically a scam

Yeah right.

I repeat what I said above - CFM is the most important factor to consider when it comes to a cooling fan. Other factors may be important but when it comes to moving heat away from a heat sink, it is all about how much air you can move across and through the fins of that heat sink. And the amount of air moving is measured in cubic feet per minute (CFM), not RPM.

thin translucent plastic

Oh wow! I totally missed that. Obviously translucent plastic blades must be hugely inferior and very noisy! And forgive me for not realizing you have the skill to look at a picture and tell everything there is to know about the fan blades - having studied aerodynamics and all.

 

[The red spirit]

Total nonsense. There is lots you can tweak, and I have already mentioned them. They include width and pitch.

And yet most fans basically clones or don't deviate much from that sacred 120mm size.

Length matters too, of course, but Shrek has already stated this is an 80mm fan so not much wiggle room there. and to make the claim that most fans use similar blades just illustrates your lack of knowledge here.

Yeah right, becsue in predefined fan sizes you can fit random size blades

I repeat what I said above - CFM is the most important factor to consider when it comes to a cooling fan. Other factors may be important but when it comes to moving heat away from a heat sink, it is all about how much air you can move across and through the fins of that heat sink. And the amount of air moving is measured in cubic feet per minute (CFM), not RPM.

And I don't argue with that. I just say that similar rpms usually mean similar cfms. Very simple concept too. You quoted those two fans, that's why i replied about them. Lying about cfm is decades old problem.

That made me snort my coffee and I’m not even drinking any. Just for reference, those who study science tend to make use such assertions as absolutes. Best regards regardless to you though.

It's not an absolute, it's just a common tendency, due to 120mm fans mostly being the same. That's due to universally accepted generic 7 blade fan design, which is roughly optimal fan for computer use.

 

[Bill_Bright]

And I don't argue with that.

What??? You started out this exchange by saying, after I said CFM is what matters most,

That ain't true.

I am glad you finally come around, contradicted yourself and now accept the true facts.

I hope you realize how pointlessly obnoxious you are being here.

Ummm, that was me being sarcastic - marveling at your advanced skills when you, just by looking at the pictures, determined the one (but apparently not the other) manufacturer totally lied about their CFM specs and how you quickly noted that, because the blades were translucent, they must be inferior and noisy. Sorry you didn't pick up on that.

It's not an absolute, it's just a common tendency, due to 120mm fans mostly being the same. That's due to universally accepted

This is more nonsense from you. First, you said "you always get..". That "IS" speaking in absolutes. So more contradictions.

And you don't speak for the universe - or most experts. It is NOT universally accepted that 120mm (or any size computer case fan) fans are mostly the same because they are NOT mostly the same - EXCEPT that they all (and yes, "all" is an absolute) fit standard 120mm fan mounting spots and have nearly identical blade lengths. Beyond that, there are vast differences - many of which have already been pointed out.

Now I will give you that many, if not most budget/cheap/generic fans probably are "mostly the same". But I have been saying all along, starting in post #2 above to get "a quality fan".

@Shrek - sorry Andy, for all the misinformation and nonsense posted in this thread. I hope you have culled out the hogwash.

The bottom line is simple. More air means better cooling. So look at the CFM ratings. If fan noise is a concern, by all means, look there too. RPM ratings are meaningless when choosing between two fans. There is absolutely nothing to suggest a faster spinning fan will cool better just because it spins faster. Again, it is about how much air it is moving.

But remember, it is the case's responsibility to provide a sufficient supply of cool air "flowing" through the case. With good case cooling, odds are any "quality" CPU fan you pick will rarely ever need to spin at full speed (and loudness) anyway - especially if it has a "quality" heatsink to work with.

 

[tabascosauz]

Stop bickering with each other. If you don't have anything meaningful to contribute about comparing 80mm fans, pack up and take your disagreements elsewhere.

To get back on track:

I am wondering what 80mm 4-pin CPU fan replacement to get; long life being a big issue for me.

I was thinking about a Noctua till I realized they only consume around 1W, so thought to seek advice here.

Unsure as to the concern over the power consumption (would you like it to draw more power?). NF-R8 is about 0.7W rated, NF-A8 is about 1W rated. I have both and they perform well enough that I'd be confident to use them on a small CPU cooler. They push a lot of air but can get a bit noisy at the top end (2200rpm for A8, 1700 for R8).

 

[bobbybluz]

Take a look on Ebay for 80mm Delta server fans. Performance ranges from mild to insane, quality is high and prices are low. Most are double ball bearing and designed to take years of 24/7 industrial abuse. Some are actually quiet while others sound like coffee grinders depending upon the cfm you desire. I use Delta fans nearly exclusively ever since I was building servers years ago.

 

[Shrek]

Unsure as to the concern over the power consumption (would you like it to draw more power?). NF-R8 is about 0.7W rated, NF-A8 is about 1W rated. I have both and they perform well enough that I'd be confident to use them on a small CPU cooler. They push a lot of air but can get a bit noisy at the top end (2200rpm for A8, 1700 for R8).

Most CPU fans seem to take 2-3W, then again quiet seems good to me and 150,000 hrs even better; might risk it with the 2,200-rpm unit.

Take a look on Ebay for 80mm Delta server fans. Performance ranges from mild to insane, quality is high and prices are low. Most are double ball bearing and designed to take years of 24/7 industrial abuse. Some are actually quiet while others sound like coffee grinders depending upon the cfm you desire. I use Delta fans nearly exclusively ever since I was building servers years ago.

Much appreciated.

 

[eidairaman1]

Take a look on Ebay for 80mm Delta server fans. Performance ranges from mild to insane, quality is high and prices are low. Most are double ball bearing and designed to take years of 24/7 industrial abuse. Some are actually quiet while others sound like coffee grinders depending upon the cfm you desire. I use Delta fans nearly exclusively ever since I was building servers years ago.

I had a 60mm delta, it was loud as hell

 

[Shrek]

Loud is not good

I also realized that while a case fan does not experience much blockage, the same may not be true for a CPU fan
and so the pressure may be more important than flow rate (Fans (noctua.at))

NF-A8 PWM

• 55.5 m^3/h (32.7 CFM)
• 2.37 mm H2O

NF-R8 redux-1800 PWM

• 53.5 m^3/h (31.5 CFM)
• 1.41 mm H2O

so the second has almost the same flow rate but much less pressure ability.

There is a black version of the NF-AB PWM, but it costs a good bit more than the brown version. I actually find the grey NF-R8 nicest, but I'll take function over form any day.

 

[eidairaman1]

Loud is not good

I also realized that while a case fan does not experience much blockage, that same may not be true for a CPU fan
and so the mm H2O may be more important than flow rate (Fans (noctua.at))

NF-A8 PWM

• 55.5 m^3/h
• 2.37 mm H2O

NF-R8 redux-1800 PWM

• 53.5 m^3/h
• 1.41 mm H2O

so the second has almost the same flow rate but much less pressure ability.

The more obstructed a fan is the higher static pressure is needed. So most cases with drives the front fan should be High Static and the Rear High Air Flow

Im fortunate that mine doesnt need moar cooling lol.

I could always get a thicker tower cooler or go custom loop for the gpu mono block along side the Chipset, VRMs, Ram, CPU, back side of cpu socket.

I'm not big on tempered glass fronts due to air restriction.