Every now and then a new fan arrives with much hype. Sometimes it proves to be a decent fan, and sometimes it doesn't. So far, not one of them have lived up to anything close to the full measure of the hype.
The new 9-blade NMB with (shudder) ball bearings has barely appeared - the reviews so far are in furrin languages - and yet another contender has come along. Paper-announced earlier, this one apparently is available for purchase in some EU countries now. I refer to the Noctua NF-S12-1200 fan, which is mentioned in this HSF review. Like the Scythe S-Flex announcements, this review included a CAD drawing of the insides of the hub, where the magnet appears to have been located in a very different place from standard fans, if I understand the drawing.
This one uses sleeve bearings (much improved, natcherly), not ball bearings, so I'll probably buy one to try out.
The HSF review predicts that the fan will appear in the US of A "in a few weeks", which is reasonable if the fan can be bought in Europe now (at 15.8 to 19 EU). Google "Noctua NF-S12-1200 fan". Prices on page 2 of the result (in EUs).
Yet Another Wonderful Fan (YAWF): whoopee
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
Felger Carbon
- Posts: 2049
- Joined: Thu Dec 15, 2005 11:06 am
- Location: Klamath Falls, OR
From that Bjorn3D review:
http://www.fsec.ucf.edu/bldg/baihp/pubs ... /index.htm
That's weird, because in another technical paper I read they said you had to decrease tip clearances to reduce noise:Furthermore, among other measures to prevent vortices like the rounded rotor hub, the distance between rotor and stator (tip clearance) was increased in order to reduce the problem of tip vortices which plays an important part in the noise emission of axial fans.
http://www.fsec.ucf.edu/bldg/baihp/pubs ... /index.htm
Another very interesting tidbit in that paper is the following:Reducing Tip Clearances and Sound Control
The functionality of an air conditioner condenser exhaust is essentially analogous to a ducted fan. Research done over the last twenty years within aeronautical engineering has shown that tip clearance of ducted fan blades to diffuser walls is critical to performance (Rajagopalan and Zhang, 1989; Abrego and Bulaga, 2002).
Unfortunately, low tip clearances are practically difficult in manufacture due to required tolerances. Should fan blades strike a solid diffuser wall, the fan blades or motor may be damaged or unacceptable noise created. Thus, in air conditioner fan manufacturing, the fan blades typically have a gap of 0.3 to 0.4 inches (0.8 - 1.0 cm) to the steel sidewall diffuser. This large tip clearance has a disadvantageous impact on the ducted fan’s performance.
which suggests to me that the dimples on the Sharkoon Silent Eagle are there for marketing purposes only.Note the improvements in air moving efficiency. As the shaft power requirement increases between the square and the cube of the flow quantity, the advantage of the foam strip for A5 (2,110 to 2,300 cfm) represents a measured improvement in the air moving efficiency of nearly 23%. Moreover, at three feet (1m) away from the condenser, we measured sound reductions of at least approximately two decibels (15% more quiet to the human ear). In contrast, we had previously attempted a number of other suggested improvements (forward swept blades, dimpled air foils and winglets) which did not produce any measurable sound reduction.
-
Aleksi
- Friend of SPCR
- Posts: 889
- Joined: Sun Dec 05, 2004 11:34 pm
- Location: Finland -- Folding For SPCR
Hi guys,
Felger, the NMB is not really a new fan. Scythe just bought some surplus (that's actually just a guess, but having information about the NMB-MAT valueline production schedules on the particular spec it seems they are surplus) and are selling them in their Scythe labeled plastic bags.
Minebea has three levels of ball bearings, lowest being the Valueline Bearing and the best being the Reference Bearing.
The Reference Bearing is shielded and closed. The specific Valueline Scythe is selling is VB2, which means it is NOT shielded or closed (more noise) and only has 35% of the oil fill in a Reference Bearing. Also the Valueline also has a "lowest cost non-Panasonic" chipset, which is not acoustically good.
Also, Scythe is pulling some marketing moves with that fan. Lifetime of 180K hours? That's BS... They're using MTBF / MTTF to get a bigger figure, just because they know the valueline isn't that long life.
The L10 specs for the NMB-MAT are...
- Reference bearing H spec (B50) 100k hours OR Reference bearing L spec <135k hours
- Half filled valueline bearing (the VB2) is only 35k hours.
And about the Noctua? Well, reading about the Austrian fan institute (or what ever it was) got me laughing hysterically
That's my rant for the month.
Spank you very much,
Aleksi
Felger, the NMB is not really a new fan. Scythe just bought some surplus (that's actually just a guess, but having information about the NMB-MAT valueline production schedules on the particular spec it seems they are surplus) and are selling them in their Scythe labeled plastic bags.
Minebea has three levels of ball bearings, lowest being the Valueline Bearing and the best being the Reference Bearing.
The Reference Bearing is shielded and closed. The specific Valueline Scythe is selling is VB2, which means it is NOT shielded or closed (more noise) and only has 35% of the oil fill in a Reference Bearing. Also the Valueline also has a "lowest cost non-Panasonic" chipset, which is not acoustically good.
Also, Scythe is pulling some marketing moves with that fan. Lifetime of 180K hours? That's BS... They're using MTBF / MTTF to get a bigger figure, just because they know the valueline isn't that long life.
The L10 specs for the NMB-MAT are...
- Reference bearing H spec (B50) 100k hours OR Reference bearing L spec <135k hours
- Half filled valueline bearing (the VB2) is only 35k hours.
And about the Noctua? Well, reading about the Austrian fan institute (or what ever it was) got me laughing hysterically
That's my rant for the month.
Spank you very much,
Aleksi
The Noctua is pretty nice, comparing it directly with an S-FLEX it seems to push more air at the same rpm, at about the same noise level. Both are 1200rpm. At 7V it's the same thing. Obviously my hands and ears are not calibrated instruments, and i don't have a hetsink that takes 120mm fans to check if they ahve any difference in cooling ability.. Both of these fans seem to have only slightly more motor/bearing noise than the nexus, can't hear anything but airflow from about 20-30cm away.. No idea how the sample variance is on these fans.
If you read this carefully they are talking primarily about the performance of the fan blade. The noise they are refering to is the side of the blade striking the shroud, not air turbulence. Increasing the tolerance of the shroud may result in lowered turbulence noise (unknown), but fan performance will suffer (fact).tip clearance of ducted fan blades to diffuser walls is critical to performance (Rajagopalan and Zhang, 1989; Abrego and Bulaga, 2002).
Unfortunately, low tip clearances are practically difficult in manufacture due to required tolerances. Should fan blades strike a solid diffuser wall, the fan blades or motor may be damaged or unacceptable noise created.
-
NeilBlanchard
- Moderator
- Posts: 7681
- Joined: Mon Dec 09, 2002 7:11 pm
- Location: Maynard, MA, Eaarth
- Contact:
Greetings,
Hmmm, this fan seems to break all the rules that I know of: straight blades (in that that they are not swept forward), steep angle of attack (though the leading edge does flatten out a fair bit), small surface area on each blade, and lots of space between the blades (which is usually used on high RPM fans, not low).
Interesting.
Hmmm, this fan seems to break all the rules that I know of: straight blades (in that that they are not swept forward), steep angle of attack (though the leading edge does flatten out a fair bit), small surface area on each blade, and lots of space between the blades (which is usually used on high RPM fans, not low).
Interesting.
<puts BEng Graduate Mechnical Engineer (3rd class) hat on>
I would say that the conclusions of the guy from the Austrian Institute for Heat-Transmission and Fan Technology (Österreichisches Institut für Wärmeübertragung und Ventilatorentechnik, ÖIWV)are about correct.
The tip geometry of the fan is such that the max arm is only 1200rpm. The seems to be fan is designed to take advantage of all the air that does through it, leading to the tight tip clearance and the straight walls of the fan frame.
The hub design is definatly designed to reduce the degradation caused by the affect of magnetic fields on lubrication fluids (there is a paper by one of my Lancaster Uni lecturers - a Proff Roger Kemp from his time at GE Traction about this).
The additional magnet reduces the degradation affect by keeping the lines of magnetic flux uniform, reducing the genertion of edy currents with the lubricant that would degrade it. The upside is that the fan balences itself quicker. <takes hat off>
I wish that they had implimented a similar system on the YSTech TMD II fan as that fan was just terrible for being unbalenced (becouase of the hub/motor design) and notorious resonating my case and its 92mm fans and 7200rpm hdds.
I would say that the conclusions of the guy from the Austrian Institute for Heat-Transmission and Fan Technology (Österreichisches Institut für Wärmeübertragung und Ventilatorentechnik, ÖIWV)are about correct.
The tip geometry of the fan is such that the max arm is only 1200rpm. The seems to be fan is designed to take advantage of all the air that does through it, leading to the tight tip clearance and the straight walls of the fan frame.
The hub design is definatly designed to reduce the degradation caused by the affect of magnetic fields on lubrication fluids (there is a paper by one of my Lancaster Uni lecturers - a Proff Roger Kemp from his time at GE Traction about this).
The additional magnet reduces the degradation affect by keeping the lines of magnetic flux uniform, reducing the genertion of edy currents with the lubricant that would degrade it. The upside is that the fan balences itself quicker. <takes hat off>
I wish that they had implimented a similar system on the YSTech TMD II fan as that fan was just terrible for being unbalenced (becouase of the hub/motor design) and notorious resonating my case and its 92mm fans and 7200rpm hdds.
Airplane turbofans have very little space between the blades, and they run at very high rpms. You need the blades close together when they will face a lot of backpressure (as in a turbofan); in PC fans it's not such a big problem. The steep angle of attack is probably because it only runs at very low speeds.lots of space between the blades (which is usually used on high RPM fans, not low).
-
Aleksi
- Friend of SPCR
- Posts: 889
- Joined: Sun Dec 05, 2004 11:34 pm
- Location: Finland -- Folding For SPCR
Jaganath... that is a surprising question from a person who has taken part in various fan, blade design etc discussions, but to put it shortly: an enclosure creates a certain resistance to an airflow, so fans need pressure capability to work with the enclosure resistance.
And got confirmation, the Scythe fan is not supplied by NMB-MAT, so indeed it is surplus.
And got confirmation, the Scythe fan is not supplied by NMB-MAT, so indeed it is surplus.
What do you mean tight tip clearance? There's about 3-4mm or room, compared to about 1mm on the S-FLEX. And the walls of the frame are straight only on the intake side, the exhaust side is as any other fan.
Compared to a normal fan the blades are about 2/3 of the width.
The blades are not straight, they are swept forward and the angle of the sweep is larger than the angle on normal fans, the blades are at a 45° angle, the "front-edge" of the blade is pointing almost directly in the direction of the rotating fan blade so there it makes a very sharp 45° turn on the backside of the blade, the inside of the blade has a slightly smaller angle because of the profile of the blade. The back edge of the blades have a larger "curve" than the front edge, and the tip of the back edge is bent at a noticeably larger angle.
Im not saying if its going to work good or bad in an actual system, im just telling you what i see when i look at it up close.
If anyone has any interest i could take some close-ups of the fan and post them here.
Compared to a normal fan the blades are about 2/3 of the width.
The blades are not straight, they are swept forward and the angle of the sweep is larger than the angle on normal fans, the blades are at a 45° angle, the "front-edge" of the blade is pointing almost directly in the direction of the rotating fan blade so there it makes a very sharp 45° turn on the backside of the blade, the inside of the blade has a slightly smaller angle because of the profile of the blade. The back edge of the blades have a larger "curve" than the front edge, and the tip of the back edge is bent at a noticeably larger angle.
Im not saying if its going to work good or bad in an actual system, im just telling you what i see when i look at it up close.
If anyone has any interest i could take some close-ups of the fan and post them here.
-
NeilBlanchard
- Moderator
- Posts: 7681
- Joined: Mon Dec 09, 2002 7:11 pm
- Location: Maynard, MA, Eaarth
- Contact:
Hi Nici,
I realize that the angle of attack changes aggressively. This is related to the smaller surface area, as well. This fan accelerates the air much more aggressively, and the trailing edges look to be pretty close to straight down.
The forward sweep that most fans have (that this one is missing) is along the leading edge: normally the outer tip is "ahead" of the rest of the leading edge. Like this:
[BTW, this Scythe is a very quiet fan indeed -- slightly better than the GlobalWin NCB, IMO]
This sweep (I think) helps prevent the air from "spilling" off the end of the blade, but instead sorta' scoops air in; and it moves it inward towards the hub instead.
All of the compound curves, and the larger surface area, allow the acceleration of the air to be more gradual, than it appears to be with this Noctua fan -- like I said before: it seems to break all the rules of fan blade design, for quietness anyway.
Another thing is does is the trailing edge of the blades are almost parallel with the support struts; causing an abrupt "shadow" in the air flow, all along the blade. Many fans (but not all) tend to angle the support struts so that the trailing edges only intersect wit them at one point at any given moment -- so the "shadow" is much smaller and is not abrupt. My best guess is that this will cause a choppiness to the sound of the air flow.
I realize that the angle of attack changes aggressively. This is related to the smaller surface area, as well. This fan accelerates the air much more aggressively, and the trailing edges look to be pretty close to straight down.
The forward sweep that most fans have (that this one is missing) is along the leading edge: normally the outer tip is "ahead" of the rest of the leading edge. Like this:
[BTW, this Scythe is a very quiet fan indeed -- slightly better than the GlobalWin NCB, IMO]
This sweep (I think) helps prevent the air from "spilling" off the end of the blade, but instead sorta' scoops air in; and it moves it inward towards the hub instead.
All of the compound curves, and the larger surface area, allow the acceleration of the air to be more gradual, than it appears to be with this Noctua fan -- like I said before: it seems to break all the rules of fan blade design, for quietness anyway.
Another thing is does is the trailing edge of the blades are almost parallel with the support struts; causing an abrupt "shadow" in the air flow, all along the blade. Many fans (but not all) tend to angle the support struts so that the trailing edges only intersect wit them at one point at any given moment -- so the "shadow" is much smaller and is not abrupt. My best guess is that this will cause a choppiness to the sound of the air flow.
-
Tibors
- Patron of SPCR
- Posts: 2674
- Joined: Sun Jul 04, 2004 6:07 am
- Location: Houten, The Netherlands, Europe
For people who can read German:
Das Institut
The "Österreichisches Institut für Wärmeübertragung und Ventilatorentechnik" is not some independent Austrian research institute. It's an organisation that is co-founded by the company that owns the Noctua brand. At least that's what their own website says.
Something else that I think is a bit strange, is that the institute doesn't seem to have an official adress. The only way to contact them is by a webform. Is there by any change an SPCR visitor who's located in Vienna, who can check with the local chamber of commerce if it actually officially exists? Because to me it sounds like a marketing tool and not something real. Or maybe it's just a fancy name for their R&D department
Edit:
Another dead givaway: Google only finds their own homepage and news/reviews of the new Noctua fan.
Why do companies think they can get away with this kind of nonsense in the internet age. (To answer myself: The stupid masses might actually eat this.)
Das Institut
The "Österreichisches Institut für Wärmeübertragung und Ventilatorentechnik" is not some independent Austrian research institute. It's an organisation that is co-founded by the company that owns the Noctua brand. At least that's what their own website says.
Something else that I think is a bit strange, is that the institute doesn't seem to have an official adress. The only way to contact them is by a webform. Is there by any change an SPCR visitor who's located in Vienna, who can check with the local chamber of commerce if it actually officially exists? Because to me it sounds like a marketing tool and not something real. Or maybe it's just a fancy name for their R&D department
Edit:
Another dead givaway: Google only finds their own homepage and news/reviews of the new Noctua fan.
Why do companies think they can get away with this kind of nonsense in the internet age. (To answer myself: The stupid masses might actually eat this.)
-
Felger Carbon
- Posts: 2049
- Joined: Thu Dec 15, 2005 11:06 am
- Location: Klamath Falls, OR
To return to something technical: with conventional fans, the ones we all know and love/hate, there's a tradeoff between air flow (CFM) and back pressure. The commonly advertised CFM figure (if not actually invented) is based on zero backpressure. On the other hand, if the backpressure is high enough, the airflow drops to zero. Panasonic, for example, publishes graphs of this.
While the real graph is not a straight line, to a first order approximation it is. Let's (for discussion purposes) normalize a straight-line graph so it goes from pressure = 1, airflow = 0 to pressure = 0, airflow = 1.
We have all worked with fans enough that we "know what they look like". Thus a common reaction to the new Noctua fan is that the blades are thin and have a steep "pitch". Not many of us have seen the new 220mm fan that is commonly (and falsely) advertised as "25 cm", but I have one, and its blades have a very "flat pitch" - the reverse of the Noctua.
Turboprop commuter aircraft have variable pitch propellors, and shorter takeoff distances than jets. A "flat pitch" produces less airflow for a given rpm, but more pressure. A "steep pitch" would produce more airflow for a given rpm, but be very intolerant of backpressure.
The Noctua promotional material emphasizes that its design produces more airflow. I believe they are absolutely correct, but at the cost of not tolerating any significant air resistance (which produces backpressure).
While we have been accustomed to promotional hyperbole, it's refreshing to find a product design which (in my opinion) is described correctly, and is designed to achieve a specific objective (superior airflow). The fact that you need a large, virtually empty case to get the low backpressure needed is something you cannot expect Noctua to emphasize.
My mini-atx case(s) are crowded inside because I like my cases as small as can be reasonably achieved. This is not how you get low backpressure!
What my small case(s) need is the opposite: a flat-pitch fan to produce more backpressure, with a larger diameter to make up for its shortcoming in airflow. This is exactly what the 220mm fan gives me. Too bad the only way to buy it is to buy a Thermaltake side cover.
While the real graph is not a straight line, to a first order approximation it is. Let's (for discussion purposes) normalize a straight-line graph so it goes from pressure = 1, airflow = 0 to pressure = 0, airflow = 1.
We have all worked with fans enough that we "know what they look like". Thus a common reaction to the new Noctua fan is that the blades are thin and have a steep "pitch". Not many of us have seen the new 220mm fan that is commonly (and falsely) advertised as "25 cm", but I have one, and its blades have a very "flat pitch" - the reverse of the Noctua.
Turboprop commuter aircraft have variable pitch propellors, and shorter takeoff distances than jets. A "flat pitch" produces less airflow for a given rpm, but more pressure. A "steep pitch" would produce more airflow for a given rpm, but be very intolerant of backpressure.
The Noctua promotional material emphasizes that its design produces more airflow. I believe they are absolutely correct, but at the cost of not tolerating any significant air resistance (which produces backpressure).
While we have been accustomed to promotional hyperbole, it's refreshing to find a product design which (in my opinion) is described correctly, and is designed to achieve a specific objective (superior airflow). The fact that you need a large, virtually empty case to get the low backpressure needed is something you cannot expect Noctua to emphasize.
My mini-atx case(s) are crowded inside because I like my cases as small as can be reasonably achieved. This is not how you get low backpressure!
What my small case(s) need is the opposite: a flat-pitch fan to produce more backpressure, with a larger diameter to make up for its shortcoming in airflow. This is exactly what the 220mm fan gives me. Too bad the only way to buy it is to buy a Thermaltake side cover.
Or you could switch to squirrel-cage/centrifugal fans, which produce more pressure for a given airflow, although it seems to be hard to find a quiet one; Bluefront's DIY slot fans also seem to offer a good combination of pressure and airflow.What my small case(s) need is the opposite: a flat-pitch fan to produce more backpressure, with a larger diameter to make up for its shortcoming in airflow.
+1Tibors wrote:I wouldn't have expected, that even the critical minds at SPCR would ignore this and rather talk about techicalities.
And at the same time, I'm wondering what would have been the amount of noise, if Tt had done the same.
The reviews, that I have read about these Noctua fans are quite promising. Very quiet and still pushing more air then other fans at similar noise levels. I also like about the fact that the fan ships with screws, fan decouplers, 3-pin to molex adapter and their U.L.N.A cable to drops voltages. Pretty good bundle for a system build, especially considering the fan is cheaper then nexus. My motherboard has only 2 fan headers, which are used by cpu cooler and vf900 so these cables could be really handy for people like me. Noctua seems to have good faith with the fans since they come with 6 year warranty.
I hope the Noctua has sent SPCR samples, so they can be included in the upcoming umpteen fans review.
-
Forumlurker
- Patron of SPCR
- Posts: 7
- Joined: Fri Oct 13, 2006 11:12 am
I think your Scythes are at least as good as the Noctuas:McBanjo wrote:I think they do blow a bit more than my S-Flex 1200rpm, atleast it feels so on my hand, and soundwise they are about the same. So they are pretty good
http://www.au-ja.de/review-noctuanfs12-4.phtml
http://www.au-ja.de/review-noctuanfs12-5.phtml
(Compare the results of Scythe@1200rpm and Noctua@1288rpm.)
Yeap read those and my test wasn't very scientificpaapaa wrote:I think your Scythes are at least as good as the Noctuas:
http://www.au-ja.de/review-noctuanfs12-4.phtml
http://www.au-ja.de/review-noctuanfs12-5.phtml
(Compare the results of Scythe@1200rpm and Noctua@1288rpm.)
Just my hand and my ear.
But if they cost the same I would probibly go Noctua but otherwise the brand that is cheaperst. Doesn't really mather