Separating whoosh from fan bearing noise

Felger Carbon · Post by **Felger Carbon** » Sat Jul 14, 2007 10:13 am

I've demonstrated that "whoosh" noise is not caused directly by the fan, but indirectly by creating a fast-moving air column that will inevitably encounter "lazy" air, with noisy turbulence resulting that we call whoosh.

I propose that whoosh be separated from noises that are indisputably tied to the fan, notably bearing noise, but perhaps some fan blade noise as well. An SLM cannot distinguish between any of these; even a recording will have fan effects mostly masked by the high level of whoosh.

How do we separate the whoosh from the fan? By not letting the column of fast-moving air encounter any lazy air until it's 20 to 50 feet away. We do this with a tube; for 120mm fans we'd need about a 5" tube and for 220mm fans an 8.2" to 8.7" ID tube (220mm fan blades are a hair less than 8.2" in dia).

I hear the laughter: a 40 foot tube in my place? Where do I find the 10 weightlifters to move this into place, and later remove it? How much does it cost?

While a big heavy tube would work, I had in mind a 40 foot tube you could fold up and tuck into your shirt pocket, the one that holds a pack of cigarettes! Ah, the wonders of modern plastics!

How about .004" thick clear plastic "disposable air duct hose" in 25, 50, or 100ft increments at $.14 to $.24 per foot, in 4", 6", 10" etc diameter (but folds flat). Go here, Find "page 197", and look near the bottom of the page.

A recording of the true fan noises without the masking whoosh sound would be interesting... especially in comparisons with other fans recorded similarly.

tehfire · Post by **tehfire** » Sat Jul 14, 2007 2:11 pm

My principal concern would be what sound-isolating effects would the tube have? You chould place the microphone at the intake side of the fan, I'm unaware of where exactly SPCR places the microphone currently.

At the end of the day, I think air turbulence is an unavoidable and still important factor when considering a fan. I don't care how quiet the actual bearing is - if I can hear sound coming from my computer, it's too noisy no matter where the sound's coming from.

Bluefront · Post by **Bluefront** » Sat Jul 14, 2007 3:40 pm

I normally can separate fan bearing noise from the woosh sound, by the use of a mechanic's stethascope. You will absolutely eliminate the airflow/fan blade noise, by placing the probe directly on the motor. Works for me...

Das_Saunamies · Post by **Das_Saunamies** » Sun Jul 15, 2007 1:17 am

Just like you can hear vibrations by placing a probe or suitable rod on the suspect, certainly works.

Most reviewers should be able to distinguish the bearing noise of a fan from the rest of it, and also be able to tell if there are annoying aspects to a fan's noise. I don't feel I would benefit greatly from a separate bearing noise test, the overall observation of sound is sufficient.

It's a good idea, and I don't doubt its applicability, but what would be the use for an average reader of such comparisons, compared to overall noise analysis at such low RPM that 'whoosh' doesn't occur much?

Bluefront · Post by **Bluefront** » Sun Jul 15, 2007 2:37 am

There is a benefit when using a stetascope, as opposed to a single rod of some sort. Outside sounds are blocked, and you get the benefit of using both ears to judge a single noise source. For identifying fan bearing noise, you can't beat the thing.....about $3 at Harbor Freight, other places.

Brian · Post by **Brian** » Sat Oct 06, 2007 3:13 pm

You wouldn't be able to hear anything over the rattle of your plastic duct. Also, a duct of that size is a huge flow restriction that would alter the fan's behavior.

I bet a diverging nozzle, like a giant cone, instead of a giant duct, would quiet the sound of air slowing down (by slowing the air down gradually), while allowing you to pick up on the airborne sound of fan blades hitting air.

Felger Carbon · Post by **Felger Carbon** » Sat Oct 06, 2007 4:35 pm

Brian wrote:You wouldn't be able to hear anything over the rattle of your plastic duct.

When I placed my last order with McMasters, I bought 25ft at 15 cents a foot of that long, thin, disposable duct. I'm gonna scissor off 8 feet or so and give it a try when I get a chance. We'll see if it rattles or not.

If you want to know how many teeth a horse has, you can go two ways: listen to expert opinion, or open a horse's mouth and start counting. I often favor the latter approach.

spookmineer · Post by **spookmineer** » Sat Oct 06, 2007 6:15 pm

Empirical evidence FTW!

Felger Carbon · Post by **Felger Carbon** » Sun Oct 07, 2007 1:30 pm

The 25 feet of duct weighed 12.2 oz. I whacked off 6 feet and taped to to a 40dBA Yate Loon D12SH-12. Shortened it up to about 4.5 feet which was more manageable.

The plastic is much heavier than I expected. Hard to keep it open the full length because of its weight. I'm gonna have to rig a double-sided tape jig to get a clearer picture of what's going on, but here's an interim report:

The air that emerges from the end of the tube, several feet from the fan, is completely non-turbulent. Pure constant wind sound yes, whoosh no. With my ear next to the exhaust end of the tube, the fan motor and blade noise, several feet away, was louder.

Holding the tube vertically with the fan next to my ear, I could hear some whoosh but not nearly as much as I remember from my first tests of the fan. My first tests, using the SLM microphone to "map" the whoosh, found some whoosh in front of the fan but most of it behind. It seems to me that what was in front is still there, but with a 4.5 foot tube, that portion of the whoosh that was behind the fan has not just moved, but entirely disappeared.

This obviously suggests further tests with progressively shorter tube lengths.

Sometimes experiments produce unexpected results.

Oh, yes: No flutter at all, but that may be because of the weight of the plastic.

Anybody want to try this, you want a high-RPM fan so you have some noise to listen to or (if you have an SLM) measure. 25 feet of this tubing (the 10" wide when flat version) costs only $3.75. Have at it!