Fan Testing Round 6, Potential Fan List/Methods

[Martinm210]

Hmm, how about doing what skinneelab guys do - several readings, dropping best and worst, averaging rest?

BTW, what different causes might cause different fan performance? If we theoretise/analise about it, maybe we can come up with better ideas how to counter those causes for more objective results?
Eg. how about ambient temp or humidity impact on fan rpm-s, storage circumstances (eg. storing time, vertical or horizontal) impact on bearing, bearing type specifics (eg. how do result repeatability looks for other ball bearing fans, do they fluctuate same as GTs?) and many other possible causes i forgot to imagine/think about ..

The big difference I see that I do have control over is warm up time (I noticed that back when I was testing rads on the crystalfontz, it's darn near impossible to hold a fixed RPM). There is a pretty dramatic change in the first few minutes where RPM increases with temperature, this is why I've been doing the whole 5 minute warm up thing. But I'm sure there are all sorts of variables that affect the result such as humidity, temperature, ambient noise level, altitude, and sample variance. Overall it seems noise levels are fairly controllable and repeatable, it's the air flow and fan RPM variances that's hard to pin down...but I've seen that by using the crystalfontz. Probably has alot to do with the temperature and run time and viscosity of the grease/oils used to lubricate the fans. In general I'm seeing more consistency is the sleeve or FDB sleeve like bearing systems that float on fluids. The fans like the GT with actual ball bearings and/or fans that have harmonic issues are the trouble makers for repeatability. I may try retesting some of the top fans, but I have no plan to retest them all many times...that would simply become "Work"..

I'm probably just being overly anal about repeatability, but I'm looking for some way of checking my setup as opposed to the actual fan. I want to make sure I'm comfortable with testing over multiple days that I'm getting the same thing. My anemometer is actually specified to read within 3% accuracy, so 1% repeatability (relative) is really good.

If you read fan specs, they all say +-10% on just about everything, that's probably about right if you started testing for sample variance.

I think we just need to accept the 10% spec and understand the limitations of testing one sample in a less than perfect environment. I could retest every fan 5 times, but then what have we done to test sample variance? You'd probably have to test 5 samples, 5 times, at sea level, in a fix ambient, fixed humidity, 0dBA anechoic chamber over 5 mounts to remove all the variables, but we have to be realistic at hobby/recreational level. This will be more than good enough I think and no I'm not retesting each fan 5 times.....I am going to use my fan check process though at least once per day run the D12SM12 to ensure nothing on the test rig has changed. This is also why I wanted to keep the fan testing rounds small, then they all get tested at relatively the same baseline, it's when you try to create a big database that all these little variables become very important. We'll see how it goes..I may have to break it up into several smaller rounds if I start seeing some check in variances I can't pin down...

Most of the fan testing/reviews I'm seeing on youtube is someone plugging in the fan sitting on a table feeling the air with the palm of their hand and subjectively saying "Wow, it's completely silent, and moving A LOT of air".

I'm not doing vertical though, I'm just not setup for that right now, but I might consider it later and run a few of the top picks after this round is complete.

I'm happy with the new anemometer, even with the GT +- 10 FPM is really not that bad, that's only about 1 CFM either way. With the old anemometer I had seen some variances as high as 4CFM, so this is a huge step in the right direction.