Saturday, March 9, 2013

Slightly less than epic fan reviews

So, I ordered some fans.

I'll give a quick synopsis of my fans and then explain why nearly all fan data is irrelevant.



So I recently got a fractal design R4 case with 2 fractal design silent series R2 fans. One of them ticks slightly, which mostly disappears at 7v, but actually gets worse at 5v. The other one ticks so bad, I can't tolerate it from the next room.

I initially ordered 2 140mm nf-a14 FLX fans from noctua

Even at lowest fan speeds they hum! I can hear both of them, at 5v, from the next room. Skip these.

I then ordered 2 120mm cougar 800-1500rpm PWM fans. They also hum and tick. Even at 5v.

then I ordered 3 140mm thermalright x silent 900rpm fans. They aren't exactly 25mm, more likely 27mm thick. They don't fit properly into my 140mm slots because of that, except the back one. 2 of the 3 also ticked and hummed. The third *slightly* ticks, but not to the point where it's irritating. I'm keeping the third one, the rest are going back.

I tried out a pair of scythe kama flow 2 900rpm fans. One of them is near silent at all voltages, while the 2nd one is near silent only while undervolted. If I run them at 12V, it's too loud with the one humming, and if I run them at 7v, they're far too slow to even matter.

The CPU fan that I have, is a noctua nf-f12, which I can't hear at all, when I let it PWM scale with CPU temperatures, with default asus standard scaling. No ticks, no hums, no airflow noise.


One thing you rarely see in fan reviews on the internet is redundancy tests, and tests with fans from real retailers. Fans that have been sent to reviewers are typically hand chosen from a batch by the manufacturer, after the manufacturer tests them, to make sure there are no flaws. The quality control on fans is pretty abysmal. Having the box dropped once can make the fans no longer work properly. Bearings can be crushed very easily.



Even if you do magically get a quality fan, how do you know if the reviewers are measuring correctly?


Every measurement has margins of error, environmental test issues, and a simple DB rating doesn't tell you what kind of fan noise it is. If it's a hum, then it'll be really annoying... if it's a whoosh, it's kind of relaxing.


What conditions do you measure DB in? What tools? What distance? Open case? Closed case? What case? Open bench is unrealistic. On a heatsink isn't relevant for case fans. On a case isn't relevant for heatsink fans. Radiator fans needs depend on fin spacing.


Note, this is just benchmarking fans. How do reviews work for more complex things!?!?

No comments:

Post a Comment