Airflow vs static pressure fans. Are they a scam?

[HTC]

I meant as they age... not as they are new.

I wouldn't believe how many ball bearings go to scrap PER DAY because they fail noise testing, and that's THE DAY THEY ARE ASSEMBLED.

As long as it's protected (the covers) and not damaged in some way, a ball bearing WILL perform correctly for many MANY years.

 

[The red spirit]

I think we would get better results from the counter rotating fans on the Fuma 2, had MajorHardware tested them on the same slot. He tested with an extra push-pull-pull triple fan setup, yet he never considered push-push-pull like optimum tech did. Long story short, at same noise corrected performance, this enables 0.5 degrees better cooling even not considering counter rotating impellers. Which was my idea for gradual air acceleration for high laminar flow at even less axial pressure gradient.

I doubt it would make a difference. He said that anything beyond stock config barely does anything differently, which would mean that Fuma 2's heatsink isn't very restrictive and there's no need for more static pressure. Instead, maybe higher air velocity may help.

I wouldn't believe how many ball bearings go to scrap PER DAY because they fail noise testing, and that's THE DAY THEY ARE ASSEMBLED.

As long as it's protected (the covers) and not damaged in some way, a ball bearing WILL perform correctly for many MANY years.

By correctly, do you mean that they won't get any louder?

 

[80-watt Hamster]

I don't wanna hear about Arctic P12s, P14, Pwhatevers ever again. Bought a 5 pack, some P14s as well, and all of them had the worst motor noise ever.

My NF-A12x25s on my cooler and the Silent Wings 3 I use as case fans are so quiet that the coil noise in my PC is more audible.


I'm over here trudging along with 60 cfm because I've found that anything above 1000 rpm on my 140mm fans make 0 difference to cooling but a lot of difference to my hearing.

I just bought a P12 to see what all the fuss is about. Not impressed. Motor noise isn't great as you said, and mine is noticeably out of balance.

 

[HTC]

I doubt it would make a difference. He said that anything beyond stock config barely does anything differently, which would mean that Fuma 2's heatsink isn't very restrictive and there's no need for more static pressure. Instead, maybe higher air velocity may help.


By correctly, do you mean that they won't get any louder?

Likely, yes. Unless ofc affected by external forces during it's lifespan.

For an applicable example, say you use a can of compressed air to clean your fans and, in so doing, you make the fan spin @ a much higher than rated speed: this MAY cause problems for the bearing because it was never designed to work @ such speed.

 

[The red spirit]

I just bought a P12 to see what all the fuss is about. Not impressed. Motor noise isn't great as you said, and mine is noticeably out of balance.

Oh well... That's disappointing for sure, but from all data they seem to be good and sometimes chart topping fans. I have bought some other Arctic stuff and its quite solid, fan being out of balance seems like manufacturing defect. I hope that you can RMA them.

Likely, yes. Unless ofc affected by external forces during it's lifespan.

For an applicable example: say you use a can of compressed air to clean your fans and, in so doing, you make the fan spin @ a much higher than rated speed. This MAY cause problems for the bearing because it was never designed to work @ such speed.

Out of curiosity, what can happen to bearing if it is spun too fast?

 

[HTC]

Oh well... That's disappointing for sure, but from all data they seem to be good and sometimes chart topping fans. I have bought some other Arctic stuff and its quite solid, fan being out of balance seems like manufacturing defect. I hope that you can RMA them.


Out of curiosity, what can happen to bearing if it is spun too fast?

I'm not entirely sure because i work in the manufacturing rings section and not in the assembly quality control section, though i did work in the bearing's rings assembly in the past (placed the spheres or the cages in the bearings).

Cage picture:

Think of it this way: what happens when you rev your car's engine OVER the red line's max?

 

[The red spirit]

I'm not entirely sure because i work in the manufacturing rings section and not in the assembly quality control section, though i did work in the bearing's rings assembly in the past (placed the spheres or the cages in the bearings).

Think of it this way: what happens when you rev your car's engine until the red line's max?

I dunno, because I don't drive, but technically it should heat up beyond manufacturer engineered limit (sorry, I'm talking about overreving, not redlining). Depending on how bad overheating is, it can bend engine's internal components and thus ruin the engine. Due to high pressure inside cylinders, if something isn't able to contain it there, engine will "blow" up. But then again, it depends on engine and on particular weak spots of engine and of left tolerance by manufacturer. Some cards rev very high without problems. Honda S2000 or EK9 R, can rev over 8000 rpm without problem. Some cars have rotary engines and those usually are designed to be revved a lot and for those cars it's advisable to rev them one per month, so that apex seals don't get flooded. Last rotary car, RX8 can rev to over 9000 rpm and it's totally fine with that, also it sound really nice at those revs. Motorcycles and F1 cars can rev a lot more.

Anyway, fans aren't cars and there's not much heat made by them. Also, I'm pretty sure that exactly the same bearings are used for quiet and faster fan models, so they probably can handle going over spec for a while (when vacuuming them for example). The bigger problem is that if you spin them fast, they become electricity generators and do start to send electricity the other way. I have seen this myself. If I blow hard on CM Blue LED fan, LEDs light up. Certainly not much , just a little bit, but maybe I generated 6 volts or so.

 

[HTC]

I dunno, because I don't drive, but technically it should heat up beyond manufacturer engineered limit (sorry, I'm talking about overreving, not redlining). Depending on how bad overheating is, it can bend engine's internal components and thus ruin the engine. Due to high pressure inside cylinders, if something isn't able to contain it there, engine will "blow" up. But then again, it depends on engine and on particular weak spots of engine and of left tolerance by manufacturer. Some cards rev very high without problems. Honda S2000 or EK9 R, can rev over 8000 rpm without problem. Some cars have rotary engines and those usually are designed to be revved a lot and for those cars it's advisable to rev them one per month, so that apex seals don't get flooded. Last rotary car, RX8 can rev to over 9000 rpm and it's totally fine with that, also it sound really nice at those revs. Motorcycles and F1 cars can rev a lot more.

Anyway, fans aren't cars and there's not much heat made by them. Also, I'm pretty sure that exactly the same bearings are used for quiet and faster fan models, so they probably can handle going over spec for a while (when vacuuming them for example). The bigger problem is that if you spin them fast, they become electricity generators and do start to send electricity the other way. I have seen this myself. If I blow hard on CM Blue LED fan, LEDs light up. Certainly not much , just a little bit, but maybe I generated 6 volts or so.

I edited my previous post but you quoted it before i did: i mean OVER the red line's max and NOT to the red line's max. I don't drive either.

Also, ball bearings DO heat up if you rotate them @ high speed, EVEN IF within speed tolerances: more so if outside speed tolerances. As with everything, use something incorrectly and it may break prematurely.

 

[claes]

Haha, he runs vortez.net. He isn't a rookie.

Are you sure? Looks like the editor is David Mitchelson and the (weird) fan reviews seem to be done by a Matthew Hodgson.

I've been doing fan reviews here as well, just search for "fan" or similar in the reviews section: https://www.techpowerup.com/review/

Oh yay great I hadn’t noticed

To be honest, there's no point in doing that. Generic 7/9 blader is pretty much the best shape of fan. There really aren't fans that are truly bad and fans that are way above others. They are all the variation of same or similar thing. Besides from catching some poorly performing fans, fan reviews are pretty much pointless now. It's just like power supplies. You could buy a clearly dangerous unit in the past, but today, pretty much anything is fine and there's not much point in paying more than you should.

Hard disagree — why’d you make this thread lol As a noise freak I miss SPCR dearly, and, like I had said before, most of these fan reviews have poor test methods that yield wacky results. Does a degree or two make that much of a difference? Not really. But 2dB does, as does motor noise, noise when faced with impedance, performance in free-air vs a radiator etc

 

[Shrek]

When hard drives used ball bearings, before fluid dynamic bearings, they could get VERY noisy with wear.

Hard drive ball bearing sound comparison - Bing video

and one can be sure this is not due to over revving

 

[The red spirit]

Also, ball bearings DO heat up if you rotate them @ high speed, EVEN IF within speed tolerances: more so if outside speed tolerances. As with everything, use something incorrectly and it may break prematurely.

Sure, but I'm afraid that it would take a lot of incorrect overreving of fans to actually damage them.

 

[HTC]

Sure, but I'm afraid that it would take a lot of incorrect overreving of fans to actually damage them.

Right away, probably not: over time, likely.

 

[The red spirit]

Hard disagree — why’d you make this thread lol As a noise freak I miss SPCR dearly, and, like I had said before, most of these fan reviews have poor test methods that yield wacky results. Does a degree or two make that much of a difference? Not really. But 2dB does, as does motor noise, noise when faced with impedance, performance in free-air vs a radiator etc

I read and watched tons of reviews and sorry, but almost all fans are pretty much the same in terms of performance and on top of that, if some "real world" testing is done, that will distort data, due to to certain fans being slightly better or slightly worse for certain conditions. Those people aren't aware of pressure/cfm curves it seems.

Even TPU reviews indicate that all fans are pretty much the same with some minor differences (although, their methodology is quite poor).

Right away, probably not: over time, likely.

I don't think that vacuuming or using air can, really counts as over time.

 

[claes]

I don’t think you read or understood my post

Some fans are objectively better performers, some have lower variance between samples, all have different noise characteristics, etc. Joe YouTube might not notice (looking forward to GN though) but I sure do

 

[The red spirit]

I don’t think you read or understood my post

Some fans are objectively better performers, some have lower variance between samples, all have different noise characteristics, etc. Joe YouTube might not notice (looking forward to GN though) but I sure do

Objectively better by what? 5%? That stuff means nothing.

 

[freeagent]

Im sure set RPM limits are due to wear. Go over the limit and the wear pattern changes because of velocity. I lost a nice TY-147A due to over revs, but it is an FDB bearing. My double ball TY-143 had the same annoying high pitch for the 3 years it lived with me. I believe it died because of over revs too. But the bearing was fine, the board is cooked and wont rev over 650. My TY-147A wont rev over 250 and it feels like shit.

Edit:

So this is why I am changing my case fans, I am moving too much air.

 

[HTC]

I read and watched tons of reviews and sorry, but almost all fans are pretty much the same in terms of performance and on top of that, if some "real world" testing is done, that will distort data, due to to certain fans being slightly better or slightly worse for certain conditions. Those people aren't aware of pressure/cfm curves it seems.

Even TPU reviews indicate that all fans are pretty much the same with some minor differences (although, their methodology is quite poor).


I don't think that vacuuming or using air can, really counts as over time.

You misunderstood: it MAY shorten it's life span right away, depending on how much more than rated speed we're talking about. More likely though, the fan will work and APPEAR to be normal for some time but you'll begin to notice problems sooner than you would otherwise, and that's the part that takes time. In essence, instead of 6 years (as per Noctua's warranty), may last 5 or less, depending on the "out-of-spec treatment" it got.

 

[80251]

12cm Noiseblocker eLoops have the rotor stator ring which enable a 6 bladed design. I would say they do well.

I think we would get better results from the counter rotating fans on the Fuma 2, had MajorHardware tested them on the same slot. He tested with an extra push-pull-pull triple fan setup, yet he never considered push-push-pull like optimum tech did. Long story short, at same noise corrected performance, this enables 0.5 degrees better cooling even not considering counter rotating impellers. Which was my idea for gradual air acceleration for high laminar flow at even less axial pressure gradient.

Wouldn't the fact the air has to flow through the radiator end up straightening out the airflow so that contra-rotating fans would be irrelevant?

 

[VSG]

Even TPU reviews indicate that all fans are pretty much the same with some minor differences (although, their methodology is quite poor).

Why is my methodology quite poor?

 

[mtcn77]

Wouldn't the fact the air has to flow through the radiator end up straightening out the airflow so that contra-rotating fans would be irrelevant?

Says it word for word in the majorhardware review, too. It still works good with smoke test, though.

My point was trail edge separation that could be solved by two step acceleration of air. AF fans apply less pressure, sp fans more.

I know I said the opposite case previously before, but since stall is generated by pressure, dropping the pressure gradient at the second sp fan by a primary af fan could either dampen, or accentuate its noise pattern, depending on which edge the sound is generated from(accentuated at the front if it stems from too much charge at the compression side, or dampened at the trail if happening due to trail separation).
If majorhardware can do it, so can we.

Why is my methodology quite poor?

You know how things are. Your reviews aren't rgb flaired.

 

[formula383]

I'm not entirely sure because i work in the manufacturing rings section and not in the assembly quality control section, though i did work in the bearing's rings assembly in the past (placed the spheres or the cages in the bearings).

Cage picture:

Think of it this way: what happens when you rev your car's engine OVER the red line's max?

Thats not at all the same thing. Engines use fluid bearings, no balls in them. the valve train is what takes a shit first, and then possibly the pistions will seize as the oil gets too hot and gets burned off the cylinder walls.

 

[The red spirit]

Why is my methodology quite poor?

There are no tests done without radiator and using certain radiator may bias results, due to fitting that specific radiator better. I think that that particular Silverstone fan, which tops charts shouldn't be included in 120mm comparison, as it is 140mm fan with 120mm mounting holes. Which is quite unfair and it may not really fit into 120 mountings everywhere. For example, 240mm rad. It would be nice to see CFM/dBA chart too, which I feel is very relevant. Also there's no pure static pressure test, I guess radiator works as real world test, but I'm not too sure.

 

[mtcn77]

Also there's no pure static pressure test, I guess radiator works as real world test, but I'm not too sure.

Pressure test can be cheated by running louder.

 

[The red spirit]

Pressure test can be cheated by running louder.

Then dBA normalized static pressure test.

 

[mtcn77]

Then dBA normalized static pressure test.

*that is moving the goalpost.*
Don't censure someone else not recognising your own fault. Radiator tests should be the industry standard, not the other way around.