So I got my hands on this for a week now.
Before this I was using thermaltake fans on my rad, rated 2000rpm 78cfm.
This scythe G is rated similar.
First is the noise, noiselevel is similar to the TT, but not as high pitched.
Silence is not BIG concern, this post is really about performance of the 2.
The thermaltakes actually moved more air, and pulled more air through the radiator fins.
I was a bit dissappointed in the scythes, the fans I have heard so much about with every fanboy raving about them immediately anytime there is a mention of fans.
These scythes were also 2X the price of the TT fans, moved less air and had poorer static pressure.
I don't have andy REAL testing methods so please do take my experience with a grain of salt.
What i did was put a thin paper behind the rad and let the fans go at max, one after another.
Consistenly the TT had stronger pull THROUGH the rad fins than the scythe G. The tt was able to lift the paper from farther distances and pulled the paper in faster and stronger.

Noise wise, because the TT is higher pitch, they dial down more noisely, the scythe G dials down much quieter.

i am still surprised that nobody ever mentions TT fans as rad fans when they are actually good performers for rads.
And for those who like LED fans, the TT 78cfm models come in LED and non LED, and they can still be found for less than they flex's.

One other thing I noticed was at full speed, the humming from the scythe G is audible, but with the TT since it's pitch is higher is blends in with the extra air that it moves, and you can't tell if the noise is because of the motor or the whoosh of air.

Both fans have pros and cons.
quiet being a factor the flex dials down quieter, airflow and static pressure with the TT wins.
If both fans were to be used at max speed at all times, i would pick the TT. The motor whine blends in with the whoosh made from the extra air it moves over the scythe.

Slightly old post but tt fans are great for rads but their lousy reputation shoots them in the foot, at least for more serious people. I can recommend the tt-1225 1300 rpm fans as well, 54 cfm I believe, while loud not fighter jet loud and at 5v still spins enough to cool an idle 4870 and overclocked q9550. They do however have sleeve bearings so horizontal mounting isn't the best idea.

if you want fans with high static pressure you better look for some 38mm fans

San Ace 109R1212H1011 is known to have great static pressure and undervolts quietlly.

http://cgi.ebay.com/ws/eBayISAPI.dll?Vi ... 0321137303

i was going to send a couple in for review to SPCR but only have about 40 left now

For static pressure the rule of thumb is that the thicker the better.
With the same free flow rating the noise (at free flow) will be higher though.

Cheers
Olle

Any idea what the slowest speed/voltage these fans will run at?

The S flexes just don't handle high impedance that well.

For that look at the Scythe Gentle Typhoons, Akasa Apache and the Noctua CPU fans.

If you have a high impedance situation, it seems to me the best solution is to find an alternative without the impedance. Even with a good fan, back pressure is only going to increase the noise to air volume ratio.

There is not any good testing out there, but I suspect that the GT fans are as good as it gets for quiet high impedance cooling.

If you need to go to 38mm, there aren't any quiet fans out there at that size that I am aware of. If you need 38mm, why not use the 55mm fans jointly designed by Feser and Noiseblocker.

But if you need that kind of horsepower, why not look for a radiator with looser fins and less backpressure?

Who makes the quietest radiators?

No one seems to publish CFMs at pressure.

So none of the published specs are any good.

When manufacturer's publish pressure, they don't say at what volume. I think it is normally the pressure at stall.

Neither of those numbers is useful.

Why don't you try ignoring the published specs and buy one Gentle Typhoon and see what it does in real life. Despite its published specs (which are presumably accurate - Scythe seems to be generally honest) see what it does when operating outside the conditions those specs are measured at.

Just take a look at those blades. Compare them with the Noctua case fans and then to the Noctua CPU fans.

There is not a lot of room for air to slip backward in the GT. In free air it runs hundreds of more rpm more to generate the same cfm of lesser fans when measured it is measured in free air.

But it looks like it was engineered to keep delivering 80% of that cfm when a less fan is down to 20%. Sort of like a diesel hauling a heavy load. Not as fast the the sports car, but when some torque is required to pull weight up a hill, its there.

See what it does.

Most published stats about fans are crap... CFM and db are usually a joke (there are exceptions), so expecting accurate static pressure stats? ah we can dream! hehe.

I do find it a little amusing that the OP has posted in SPCR about using 2000 rpm fans! 2000rpm tends to be very loud in my books!

Wish there was more testing of fan pressure... one of the few tests I know of was done by Martin:
http://www.xtremesystems.org/forums/showthread.php?t=223391
http://martin.skinneelabs.com/FanTesting.html




EDIT: Notice where S Flex G sits? near the bottom... not ideal rad fans, assuming the testing is accurate.

The Thermochill PA120.3 series and PA140.3 series are designed for low pressure fans. They have about 10 fins per inch. The Black Ice 480GTX has about 19 double stacked fins, and the Feser Monsta 120.3 has about 12 per inch. There's pic in this link: http://martin.skinneelabs.com/HWlabs480GTX-Review.html

Was this aimed at me? Due to my 2000rpm fan speed comment? I am aware that some rads are designed for low speed (well, more accurately, as you say, low pressure fans), and some for higher speed fans. I was simply saying that I find 2000 rpm fans very loud! If anyone wants to use them, more power to them.

Yes, seems like the speed of the fan is that single largest determinant of the volume of sound it generates.

It is a larger factor than even the brand of the fan (as long as you are talking about reasonably good brands known generally for being quiet).

The GT fans seem to be an exception. All things equal it seems like they run 400 or so more RPM to produce the same cfm at the same noise.

For the life of me I can't read that chart. Can you translate it. Which fans look most interesting and why, especially at the low rpms where they are going to be quiet?

Yeah its a bit nuts with the number of fans on it... and its also not entirely accurate to compare fans based on RPM, as some fans blow more air at the same RPM.... although, as a crude generalisation it works.

Given the topic of this particular forum, this graph is probably most useful:


Given we want the lowest possible noise and the highest possible air flow, we want fans which are in the upper left corner of that graph (low noise, high air flow). Each fan is measured at different speeds, which obviously affects both factors, hence the line.

The gentle typhoons (brown filled square) are excellent based on that information.

All the usual testing caveats apply here, especially as I doubt Martin's equipment is on the same level as SPCR's... however, it does provide an interesting comparison of fans.... not to mention videos to listen to. The reason videos are good is that pure dB doesn't actually give you an idea of how annoying the sound will be. The other thing to note, is that the two different gentle typhoon fans tested are slightly different... whether this is due to testing inaccuracies or quality variances I cannot say. They do both test pretty well tho!

EDIT: The reason I included the other graph above is that when you buy fans, you buy xxx rpm...

EDIT2: If you don't read anything else from Martin's report, this is important to note:

1. Which ones in your opinion were number 2, 3 and 4?

2. I trust someone's measurement of rpm more than their measurement of db.

3. Also, what exactly was the resistance they were measuring? It must be more realistic than free air measurement, but what exactly were they measuring?

1. From what I can read off that graph (and I had nothing to do with this, so this is purely my interpretation), I would say that the Gentle Typhoon, Yate Loon, San Ace and Panaflow all perform well at the low end of the scale -- ie with all fans under volted to varying degrees. The caveat here is that the 38mm fans have a different noise signature, which may be more annoying if its audible.

2. Sure, especially at low noise levels. However, comparing RPM vs CFM is not entirely fair... especially with a fan like the Gentle Typhoon which actually spins a fair bit faster than other fans for the same CFM. Definitely take the results with a grain of salt.... as Martin commented, at the low end he was at the limits of his equipment.

3. The resistance used in the forum post is an XSPC RS120. This is a medium-low density radiator (approx 13 FPI). A different radiator will obviously affect the results, although I would expect the comparative outcome to be roughly similar. The actual pressure testing done on the other link is in some ways more valuable... but testing against a rad gives a more realistic test scenario.

Hope this helps! I wish we had more good testers of the likes of skinee/martin and SPCR!

It was aimed at ces, he asked if any radiators were designed for low restriction air flow a number of posts back.

RPM and dBa aren't really the deciding factors here. Noise character is. A smooth noise character lets the fan spin faster and still be unobtrusive.

I hope Mike's going to review one of the Gentle Typhoons some time soon. There are too many factors that influence noise characteristics for a simple dBa rating to cover.


+1 I totally agree.

Ah sorry. In that case, yes, I completely agree with your post.



Completely agree, I did mention this in my long rambling posts above, but it may have got lost in the noise. RPM is a factor only in that fans are sold as a certain RPM, and really no other specification about the fan can be believed (well size too I guess).

FWIW however, the gentle typhoons did rank very highly on the noise vs airflow chart, even without taking into consideration the noise character.... and their noise character is supposed to be better than the average fan.

I was repeating it for emphasis. I find separating the main point of the post slightly gives it more wheight.

I do hope the GTyphoons are good fans, but more blades make more noise, though other design points can negate that. Time will tell.