120 mm Fan Questions

[jcrouse]

Well, In the continuing eductaion process that seems to be WC'ing and in line with my other post on 140mm case fans, I have a question about 120mm fans.

The Noiseblockers have been mentioned more than one place as a great choice. I will be using one of these as a case fan and more as radiator fans on XSPC RX series radiators. The confusion sets in since there are three 120mm models.

Noiseblocker NB Multiframe S-Series
Noiseblocker NB-BlackSilentPro
Noiseblocker NB-BlackSilentFan

Which one would you guys recommend? Also which speed would be sufficient to cool 3 Way SLI GTX 470's on the XSPC RX480 rad, bearing in mind that I am looking for extreme quite and don't overclock the video cards?

Thanks yet again,
John

[IanM]

Noiseblocker NB Multiframe S-Series
Noiseblocker NB-BlackSilentProI believe these are the same, other than the fancy frame, but that frame is unhelpful because it will be more complicated to seal against a shroud. Hence the BlackSilentPro is the better choice of these for use with a radiator. That said, look at the guides posted:

http://www.xtremesystems.org/forums/showthread.php?t=202394 (scroll down to fans in order to find many threads and a wealth of test data)
Also Martinm210's testing now has a 'round 8' thread: http://www.xtremesystems.org/forums/showthread.php?t=263748 You may find something that appeals to you more.

[jcrouse]

I believe these are the same, other than the fancy frame, but that frame is unhelpful because it will be more complicated to seal against a shroud. Hence the BlackSilentPro is the better choice of these for use with a radiator. That said, look at the guides posted:

http://www.xtremesystems.org/forums/showthread.php?t=202394 (scroll down to fans in order to find many threads and a wealth of test data)
Also Martinm210's testing now has a 'round 8' thread: http://www.xtremesystems.org/forums/showthread.php?t=263748 You may find something that appeals to you more.

I am thinking not sice the tests don't indicate such. I had seen and looked through the charts but the BlackSilentFan was not in the test and was one that interested me due to blade color (of all thing, the blue).

I am also not sure how much CFM (or RPM) I need for the above application.

Thanks,
John

[Martinm210]

I think they are great case fans.

Pay attention to the test scenario used in reviews. An open air test is no good for evaluating how a fan will do on a radiator and a radiator test is no good for evaluation an open case fan application.

The NB are nice quality fans, but I think their blades are too thin and made of materials that don't block radiator noises very well. The best radiator fans have thick and dense blades which seem to resist the noises that are otherwise amplified in the rad itself. The radiator works as a speakerbox of sorts...

I've also noticed some differences in the materials used, softer plastics are better for rads.

[IanM]

I am thinking not since the tests don't indicate such.Which tests?

Read the marketing spiel on the BlackSilentPro: http://www.noiseblocker.de/en/Multiframe120.php# then compare the blade profiles from the pictures:
http://www.noiseblocker.de/Pressefotos/M12-S1_l%C3%BCfter.jpg
http://www.noiseblocker.de/Pressefotos/PL-1_l%C3%BCfter.jpg

I'm not sure what else to say to persuade you, they are the same aside from the frame. Email Noiseblocker to get confirmation?

and I think the BlackSilent is the same blade profile, but different bearing:
http://www.noiseblocker.de/Pressefotos/XL1_l%C3%BCfter.jpg

Read again all the X-bit 120mm fan comparisons I posted in your other thread, you have to look at the old 'Be Cool' article and the newer <1350 rpm comparison. I don't think there is any significant difference in performance, especially at truly low noise speeds, and the SPL set up they use places the mic at only 150mm from the fan hub - at the 'eye of the storm' i.e. it's picking up hub noise rather than air turbulance noise. If the Multiframe is quieter than the BlackSilent Pro at the hub, with the same bearing, then it's only because of the frame decoupling system. However, the difference is trivial, and the blade profile is identical - so the significant noise from turbulent air will be generated with the same pattern.

Anyway you want the original BlackSilent? If the Scythe Gentle Typhoon is still the king of the 25mm radiator fans, compare to that and look at the old data: http://www.xbitlabs.com/articles/coolers/display/120-140-fans-roundup_28.html There is nothing much in it:

http://www.xbitlabs.com/images/coolers/120-140-fans-roundup/13_diagr_33dba.png

Someone else will have to say if 4 of these will be adequate on your system and radiator.

[jcrouse]

I think they are great case fans.

Pay attention to the test scenario used in reviews. An open air test is no good for evaluating how a fan will do on a radiator and a radiator test is no good for evaluation an open case fan application.

The NB are nice quality fans, but I think their blades are too thin and made of materials that don't block radiator noises very well. The best radiator fans have thick and dense blades which seem to resist the noises that are otherwise amplified in the rad itself. The radiator works as a speakerbox of sorts...

I've also noticed some differences in the materials used, softer plastics are better for rads.

Martin ... thanks for your reply, much appreciated and respected. Was your approval for the BlackSilentFan (blue blades) ? If so, which speed, the 1000 or 1500 RPM's do you think would be best for the above app? Do you think the 1000's would be sufficient with no overclocking? Also, what other comments can you make on RPM / CFM requirements for the application?

Thanks a TON,
John

[Martinm210]

I was trying to suggest a different brand. I really like the NB fans as open air case fans, but I wouldn't recommend them for a radiator. They have really good motor noise quality, but that basic 7 blade fan style just doesn't perform that well on a radiator scenario. The Gentle Typhoons pretty much rule the cfm/noise ratio from about 1000 to 1800 RPM. There are other good ones, but nothing in the same league when installed on a radiator.

Read into the details of how Xbit does their noise measurements. They test air flow in a chamber, then they test the noise with the fan sitting completely in the open air.

Unfortunately I think it makes the Xbit results only good for a case fan scenario where you have no screen or other restriction. It misses the reverb and harmonics that are part of the real world and particularly misses how a fan behaves on a radiator. I also think there is something to how dense a fan blade is in holding and masking noises that are created within the radiator. I'm not quite sure, but that dense and very forward sweeping blade simply works the best on a radiator.

For the price of NB fans, you would be better off getting GTs. Better yet, buy one of each for yourself and try it on an actual radiator.


I would suggest GT15s in push pull on an RX360. From what I quickly pulled up, one article said something about 250 watts per card. 750watts is quite a load. Push/Pull GTs at their full 1850 RPM would probably get you under a 10C water/air delta.

No other fan would even come close to the noise level the GT15s would at that RPM range. They are really in a league of their own.

[Martinm210]

Look at the 30CFM level, that's the kind of fan power you need:

GT-15s measured a good 7dbA lower than the NB I tested...that's significant.
http://img98.imageshack.us/img98/71/112710b.png

[jcrouse]

I would suggest GT15s in push pull on an RX360. From what I quickly pulled up, one article said something about 250 watts per card. 750watts is quite a load. Push/Pull GTs at their full 1850 RPM would probably get you under a 10C water/air delta.

No other fan would even come close to the noise level the GT15s would at that RPM range. They are really in a league of their own.

Remember, if it matters, the radiator will be an RX480. Anyways, I will take your advice.

Thanks Martin,
John

[the finisher]

I have GT-13,14,15 & 29. These fans really are in a league of there own. Even for case fan duty, they are exceptionally quiet for the air they move.:)

[Martinm210]

Remember, if it matters, the radiator will be an RX480. Anyways, I will take your advice.

Thanks Martin,
John

Oh, ok. With the same fans, that would get you down in the 7.8C range push pull.

Or you could loose the push/pull which would bring you up to the 9.8C range.

And that's substituting a TFC480 rad. We don't have any good RX480 data to go by, but it should be in the ball park.

This is also assuming a 100% heat dump which is overly conservative. If it's an actual 80% your delta would be about .8X those numbers.

Unfortunately, that's some higher speeds where a more dense radiator would give you a little better numbers yet.

[IanM]

Xbit... They test air flow in a chamber, then they test the noise with the fan sitting completely in the open air.

Unfortunately I think it makes the Xbit results only good for a case fan scenario where you have no screen or other restriction.
I've thought for a while it's a far from perfect test, even for case fan testing. In the original test the airflow is constricted - literally forced into a bottleneck, so I think it artificially emphasises fan pressure (maybe this is OK?) Their revised test system is actually even more like a very low density radiator - with the bundle of straws inserted into the pipe. It would be better if the fans could be tested in a larger box so that the airflow could be undisrupted as it disperses in the natural cone described by the blade pattern, and the velocity measure at several points covering a wide angle.

I also find the noise test frustrating at the low end - the noise data curves flatten too much at low RPM and I don't think it's a true enough reflection of real experience.

At least the X-bit method is consistent and explained, and they also compare performance on a dense fin array air cooler, so the data is fair and useful, just incomplete.


I also think there is something to how dense a fan blade is in holding and masking noises that are created within the radiator. I'm not quite sure, but that dense and very forward sweeping blade simply works the best on a radiator.
Well, if you want to start learning about fan pressure it's more complicated than the single static pressure figure some manufacturers randomly quote on their specs. I found this article recently: http://www.esmagazine.com/Articles/Feature_Article/87328157de43a010VgnVCM100000f932a8c0____ It could give you an idea of what to research, though it all looks rather complicated.



I would suggest GT15s in push pull on an RX360.maybe more radiators are needed to make the system quiet, but leaving aside the OP's requirement to cool many components in this system...

The XSPC RX### is almost half the fin density of the Swiftech MCR###, so airflow should be less constricted. No doubt the GT's will still perform the best, but will the performance advantage v. other fans be less on this radiator? particularly in genuinely low noise / silent system where the fans are spinning 300-800 rpm.

I would expect different resonant frequencies and 'real world' noise patterns, with any radiator, not just the XSPC ones. But you can't test every fan on every radiator (or in every case)


No other fan would even come close to the noise level the GT15s would at that RPM range. They are really in a league of their own.Has anyone tried the Silverstone Air Penetrator's yet? The blade profile looks like it has some potential, click the photo to go to the product page...

http://www.silverstonetek.com/images/products/ap121/AP121-Front.jpg (http://www.silverstonetek.com/products/p_photo.php?pno=AP121&area=usa)

N.B. that's the 120mm version above, the 140mm & 180mm designs have more open blade profiles.

[Martinm210]

I originally had plans to do more testing on different radiators, but decided against it after the 50 some tests I ran on the mcr..I was burned out from that and left it there.

I did test the penetrator in an earlier test here:

http://www.xtremesystems.org/forums/showthread.php?t=259802&highlight=fan+testing+round

[IanM]

after the 50 some tests I ran on the mcr..I was burned out from that and left it there.I'm impressed that you did that many :)


I did test the penetrator in an earlier test here:

http://www.xtremesystems.org/forums/showthread.php?t=259802&highlight=fan+testing+roundsorry I overlooked that, I should've searched instead of just relying on the master list in the other round of testing :rolleyes: Not that special then...

[Church]

Yeah, it's wierd when sometimes our expectancies get smashed. I'd also expect good performance/high pressure from this fan's blade configuration. IIRC in this Martin's test (or was it another one about 140mm fans?) there was fan that did opposite. It's blade configuration had lot of open space between blades, yet it performed rather well.
Imho only way to get high air pressure for sure is to make fan's engine/bearing as silent as possible at higher speeds, so that it can simply rise air pressure with spinning faster then other fans at same noise level.