Can you spot the bogus fan specification?

[Felger Carbon]

Yate Loon lists the D22SL-12H fan on their website:
220mm 11 standard blades 600RPM 135CFM 25.5dBA

Globe lists the following fan on their website:
220mm 13 thin blades 500RPM 104.48CFM 21dBA
(104.48CFM = 177.5CMH)

Thermaltake sells side panels with this fan spec:
220mm 13 paddle blades 600RPM 67.69CFM 15dBA
(67.69CFM = 115 CMH)

I would guess that most of you would pick out the 15dBA Thermaltake spec as being bogus, but you might pick others as well. It's my keyboard, so I get to toss my opinion in the ring first.

I believe the Thermaltake specs (this time) were not written by TT, but are being passed along verbatim from Power Cooling. TT did thoughtfully convert the CMH (cubic meters/hour) figure to CFM for us.

I also believe none of those specs were invented by the three fan makers. And I know that the 15dBA spec is the noisier fan when CFM is equalized. Which is as it should be; low-pitch wide paddle blades should make more noise at a given CFM in free air. They also might work better in a case - which is why they're in production; nobody makes inferior fans on purpose.

The 15dBA fan is louder? At equal CFM it certainly is. Here's the calculated noise for the three fans if they are all pushing 67.69CFM:

Fan specifications in free air
Yate Loon 300.84RPM 67.69CFM 7.51dBA
Globe 323.94RPM 67.69CFM 9.69dBA
Power Cooling 600RPM 67.69CFM 15dBA

To a first-order approximation, CFM is directly proportional to RPM. dBA is proportional to 60*LOG10[RPM]. For a more detailed explanation, check with Jaganath; he's good at this stuff.

The first thing you should notice is that it takes almost twice the RPM for the paddle-blade fan to produce the same CFM. That's because it uses very low-pitch blades, which is bad in free air and good in a PC case because it produces higher pressure. It should be no surprise that the fan with the standard blades (YL) produces the (normalized) CFM with the least noise, because standard blades are used on almost all fans because almost all fans are optimized for maximum CFM at minimum noise in free air.

So, which fan would produce better cooling/noise in a PC case? I'd cautiously favor the paddle blade. I recently did an experiment that came out a wash, but that used a paddle-blade fan with ball bearings against standard blades with sleeve bearings. The 600RPM (nominal) Power Cooling fan is now available (on a TT side panel) with quiet sleeve bearings. I now own 3 of them.

[Bluefront]

If I read you correctly, the TT-type fan should push more CFM at a given rpm, when sucking through a filter, or when blowing into a computer case.....and this is something you have tried without a conclusive result so far? I am interested in this subject, but I am at a loss as to how to confirm any result of testing. The variables would make any test results suspect anyway. And I wonder if the differences between these various types would even matter in the real world? Since I always filter intakes on every major project, using the right blade design would be beneficial to me, although difficult to judge.........

[Felger Carbon]

If I read you correctly, the TT-type fan should push more CFM at a given rpm, when sucking through a filter...

Every hair of my engineer-being screams that a low-pitch paddle-blade fan is what's needed to suck air through a filter. However, I've made no experiments with filters, nor do I ever plan to make any such experiments.

Why? Because PCs are my hobby now that I've retired, and I don't keep any one case long enough for dust to be a problem. Dust, to me, is totally irrelevant.

[Bluefront]

Well too bad......when I get my main computer set up completely, I don't like to touch the thing unless something breaks. I might go a few years without taking the thing apart. And I can do that if I use good filters that are easily cleaned. Computer filters are very rare around here.....I almost feel like I'm the only one who uses them. None-the-less, I will continue to design setups with filters in mind, even if I don't run that particular computer very much. When I build a new computer, it might turn out to be my favorite. You just never know......

[Otter]

dBA is proportional to 60*LOG10[RPM]. For a more detailed explanation, check with Jaganath; he's good at this stuff.

Now that's a very useful formula if it works in the real world. I think you mean 60^LOG10[RPM], though.

Of course, the problem with comparing noise specs from different manufacturers is that it's unlikely they took their measurements using the same methods. Even if everyone is honest (which would be a bit out of the ordinary for Thermaltake), the lack of measurement standards makes comparisons meaningless.

[Felger Carbon]

Now that's a very useful formula if it works in the real world. I think you mean 60^LOG10[RPM], though.

No, I meant multiply (*) and not power (^). However Jaganath has posted a Comair link which states that the first number should be 50 and not 60. If this is true, doubling/halving the RPM would produce a 15dBA change, not 18dBA.

As my initial posting in this thread pointed out, the TT noise spec is actually consistent with the YL and Globe spec. So if the TT spec is bogus, so is the YL and Globe spec. I was under the impression that Yate Loon didn't play the bogus-spec game (Globe, I don't know about). I don't know what to think at this point - if YL has gone over to the dark side, then fan specifications are all useless!

[Rusty075]

I was under the impression that Yate Loon didn't play the bogus-spec game (Globe, I don't know about). I don't know what to think at this point - if YL has gone over to the dark side, then fan specifications are all useless!

General rule:

The people who actually make the fans tend to have reliable specs, for legal and liability reasons. If an engineer spec's a fan for a piece of equipment, and that equipment then dies a fiery death due to bogus fan specs from the fan manufacturer, you can guarantee that the engineer's company lawyers will be very interested.

But the people who resell the fans are free to lie, cheat, and steal to their hearts' content, because there are no consumer protection labeling laws regarding fans or PC components.

[Otter]

Okay, the actual equation is
dB2 = dB1 + 50 log10 (RPM2 / RPM1)

http://www.comairrotron.com/acoustic_noise.shtml

This is not the same as saying that the SPL is proportionate to 50 log10(RPM). I knew that couldn't be right so I tried some possible typo corrections, and using SPL proportionate to 60^log10(RPM), I came quite close to your results for the D22SL-12H.

Applying the correct equation to the big Loon, we get 10.5dB (Yate Loon lists dB on their site, not dBa.)
25.5dB + 50*log10(300.84rpm/600rpm)=10.5dB
for the Globe: 11.6dBA
and of course the 250mm Thermaltake/Power Cooling fan is unchanged at 15dBA

As my initial posting in this thread pointed out, the TT noise spec is actually consistent with the YL and Globe spec. So if the TT spec is bogus, so is the YL and Globe spec. I was under the impression that Yate Loon didn't play the bogus-spec game (Globe, I don't know about). I don't know what to think at this point - if YL has gone over to the dark side, then fan specifications are all useless!

I agree with your initial assessment that these numbers seem to be consistent. As you pointed out, the paddle-bladed fan has to spin roughly twice as fast as the Yate Loon and hence makes more noise. Factor in varying methods of taking SPL readings and that Yate Loon's noise specs aren't A weighted, and the suprising thing is that the numbers are this close. However, for all those reasons, I agree that comparing noise specs from different manufacturers is useless.

Noise specs can be useful for comparing fans from the same manufacturer, though, especially if they're straight from the engineer's mouth and haven't been translated into marketspeak. I suspect Thermaltake often quotes CFM for 12V and dBA for the minimum starting voltage. If this is true, they aren't actually lying, but without qualifiers these numbers are very deceptive. In this case, the performance of the big fan is so much better than the usual 120mm fans that TT probably didn't feel the need to obfuscate.