2x80mm quieter than one 120mm?

Nightwatch · Post by **Nightwatch** » Thu Nov 13, 2003 5:31 pm

I've been wanting to quieten my pc for a while now and have been surfing these forums and others in an effort to gain some knowledge on what would be the best way to do that. I have now started in on my pc and am in the process of replacing the noisy fans with quieter ones. Somewhere in my surfing I recall reading an article which stated, essentially, that 2 x 80mm fans running at the same cfm as one 120mm fan would be quieter than the 120mm fan. Basically I want to know if this is true, or only true under certain conditions, and if so, what those conditions are.

Later....

Wedge · Post by **Wedge** » Thu Nov 13, 2003 7:48 pm

Different people will tell you different things. Unless he's had a change of view on this, MikeC, the site administrator, favors 2 80mm fans over a single 120mm.

Dorothy, a popular UK vendor here at SPCR Forums, favors the use of a 92mm fan in lieu of a 120mm.

They both have had a lot of experience with this, but that is not to negate the other opinions here that differ.

I tend to think the 2 80mms undervolted would be quieter than the undervolted 120mm, but I cannot tell you this from experience as I have never tried 2 80mms.

However, my personal view is that a quiet 92mm (Panaflo) is an awesome air pusher for the noise level -- a nice compromise between the two choices. I have this in my 3700amb and it is surprisingly quiet when undervolted, and the air flow is great.

dukla2000 · Post by **dukla2000** » Fri Nov 14, 2003 12:29 am

OK - I am an admitted 120mm junkie and my 'spreadsheet review' (which I admit is no substitute for the experience of Dorothy & MikeC) says the 120mm is better IF you can work with a single fan solution. (The benefit of multiple 80mm is you can spread them around to better target the air.)

I don't have all the relevant formulae to hand, but my 'results' are:
Papst 4412FGL (120mm)
55.3 cfm, 26dBA (spec @ 12V in free air)
20.6 cfm (@10Pa pressure - to cause an argument, @ 12V)
8.6cfm, 7dBA @5V @10Pa

Panaflo L1A (80mm)
24cfm, 21dBA (spec @ 12V in free air)
5.7cfm (@10Pa pressure @ 12V)
2.375cfm (@10Pa @5V)
so you need qty 4 for similar airflow which is
9.5cfm, 15dBA @ 5V @10Pa Qty 4

The yawning gap in this 'paper analysis' is I can fully accept the formula that calculates dBA at reduced voltage just has to have limitations at low fan speeds or somewhere, but I have never seen the assumptions behind it. It is an absolute fact my Papst 4412FGL @ 5V is audible to me from about 50cm, and sure as hell my home environment may be quiet but my ears are old and ...

For reference the dBA formula is
New dBA = Old dBA - 50 * Log (Old rpm/New rpm)
where base 10 logarithms are used.

Now in practice I am absolutely convinced 1 * 120mm at low voltage is plenty of cooling for an 'average' (todays 70W CPU, 1 hdd etc) case: my signature system has the Papst as the only case fannage, my second system has the 120mm in a SeaSonic psu as the only case fannage. Whether I could cool adequately with only 2 or 3 low speed 80mm I don't know.

ps - BTW - Welcome. And is 5 Commissioner St still serving food?

[Edit]
Correction - I was trying to work out where I got my numbers from and discovered 1 deliberate error. The issue is how to sum the noise produced by Qty 4 fans, my initial 15dBA is of unknown provenance. (i.e. completely wrong and indefensible!) Referring to fanspecs.xls and the related blanket-insulation link the sum of Qty 4 Panaflo L1A at 5V is 8dBA. So bottom line, 4 * 80mm L1A are pretty equivalent (noise & airflow) to 1 * Papst 120mm on paper. [/edit]

silvervarg · Post by **silvervarg** » Fri Nov 14, 2003 5:35 am

dukla2000:
the sum of Qty 4 Panaflo L1A at 5V is 8dBA. So bottom line, 4 * 80mm L1A are pretty equivalent (noise & airflow) to 1 * Papst 120mm on paper.

This numbers does seem fairly accurate.
I'm not really sure how the pressure in the box will affect things. I guess we all 3 setups present in SPCR. (underpressure / almost ambient pressure, / overpressure). It then depends if we talk about input fan or exhause fan.

It does not seem that realistic in most cases to use 4*80mm fans on one side (e.g. exhaust) due to space limitations.
Area or 2*80mm fans with a hypotetical zero mm hub does give 10048mm^2 area. Same for a single 120mm fan with zero mm hub is 11304mm^2. This should probably be a lot closer in area than 4*80mm fans.
I claim that an equal total area should have the same possibilities to push the same amount of air with equal sound generated from the pushed air (=excluding bearing sounds).
However you should withdraw the hub size from the area. I was just too lazy to do that. Also I did not take into account that the actual size of the fanblade radius is slightly less than the fan size.

Nightwatch · Post by **Nightwatch** » Fri Nov 14, 2003 3:59 pm

Thanks guys, for your input.
I s'pose that in the end it all comes down to what the individual actually hears, some will find the single 120mm better than 2 80mm and vice versa, altho by your calc, dukla2000 and silvervarg, one would need 4 80mm fans to equal the airflow of the one 120mm. At the moment I have the Pabst 4412FGL on order and it should arrive at Papst here in about 10 weeks or so from Germany. In the meantime, I'm using a Sunon KDE 1212 PMS which is their version of a low noise fan, altho at 34 db (12v), its not exactly a whisper. Unfortunately I havn't been able to source any Panaflo fans here in South Africa, it seems we dont get them here, and any contact I have tried to make with
Panasonic S.A has met with the very thing Im trying to get my pc to do:... silence!

Nightwatch · Post by **Nightwatch** » Fri Nov 14, 2003 4:10 pm

Oh, almost forgot,
Thanks for the welcome, Dukla2000, these are the first forums I've posted to in all my years of surfing, this is a really brilliant site. As far as 5 Commisioner street is concerned, I remember it was a really nice Chinese restaurant, but sadly, I havent been there in years...That part of town is going downhill at a rapid rate of knots.

later....

silvervarg · Post by **silvervarg** » Sat Nov 15, 2003 11:43 am

Nighthawk:
altho by your calc, dukla2000 and silvervarg, one would need 4 80mm fans to equal the airflow of the one 120mm

I think you got my message a little bit wrong. I tried to say that you can get almost what airflow you want from whatever fan setup you like if you are just prepared to accept a certain speed and sound level.
However my personal experience tells me that for low airflow and normal airflow setups 2*80mm will often give a slightly better result that a single 120mm fan.

No lets check some numbers to back that up (everything fetched from the fanspec2.xls seen in the same link as dukla2000 posted).
Note that I use the fanspec2.xls that has much less errors than the fanspec.xls.

First start with the 120mm fan suggested by dukla2000:
Papst 4412FGL (120mm)
@12V 55cfm, 26 dBA.
@5V 23 cfm, 12 dBA.

So what we want to get is something close to 23 cfm with 80mm fans so we can compare.

First we check the Panaflo L1A 80mm (FBA08A12L1A):
This fan has exceptionally bad numbers in this excel chart.
@12V 24cfm, 21dBA
@5V 10cfm, 9dBA
2 fans @ 5V 20 cfm, 12 dBA.
Result is slightly worse than 120mm Papst according to chart. I believe reality would be different.

Next try is the Noiseblocker S2 (80mm):
The fan has exceptionally good number in this excel chart.
@12V 50cfm, 19dBA
@5V 21cfm, 6dBA
4 fans @5V 84cfm, 12 dBA
2 fans @7V 58cfm, 13 dBA
Results show that the S2 beats the 120mm Papst many times. I believe reality would be different, the S2 is not even close to this good.

Lets check up on a few 80mm fans that seem to have ok numbers in the chart:
Adda AD0812DX-A7OGL (80mm):
@12V 21cfm, 14dBA
@5V 9cfm, 3dBA
3 fans @5V 27cfm, ~7dBA.
Results for this fan seems a little bit better than I think it performs IRL.

Another fan that seems to be rather good:
YS Tech FD8125 (80mm):
@12V 26cfm, 19dBA
@5V 11cfm, 5dBA
2 fans @5V 22cfm, 8dBA

And to be real fair to the Papst fan we also compare with a standard Papst 80mm fan. 8412NGL:
@12V 20cfm, 12dBA
@5V 8cfm, 1dBA
2 fans @7V 23cfm, 7dBA

Conclusion:
Out of the 5 setups of 80mm fans 4 of them did better on paper than the single 120mm fan did on paper.
4 of the setups used 2*80mm and one used 3*80mm fans.
Numbers might not be accurate in the excel sheet, but they are probably probably off by almost the same number for 80mm and 120mm fans.

I believe that bearing noise are normally in favour for the 80mm fans, especialy when running at 5-7V.

JVM · Post by **JVM** » Sun Nov 16, 2003 8:02 pm

From what I've read here about the Nexus Real Silent Case Fan (80mm) running at 12 volts and 1500 RPM, it would appear to be fairly quiet. So would that fan running at 12 volts produce the same air flow or better than a 120mm fan running at 5 volts and about 1200 RPM?

silvervarg · Post by **silvervarg** » Mon Nov 17, 2003 1:34 am

JVM:
So would that fan (Nexus 80mm) running at 12 volts produce the same air flow or better than a 120mm fan running at 5 volts and about 1200 RPM?

No.
An estimate is to count the speed the tip of the fanblades will move:
Nexus: 80mm*3.14*1500 RPM
120mm fan: 120mm*3.14*1200RPM

Nexus/120mm fan gives ~83% blade tip speed.
So the Nexus will move no more than 83% of the air that the 120mm fan moves.

The real important thing is to know how much air you really need to move and where that air needs to move. Once you figured that out you can pick fans with very little problem.
Running any fan at 12V is usually not the best way to achieve a quiet computer. If you are thinking about 120mm fan or 80mm your best choise is usually either a single 120mm fan or 2*80mm fans.
If fans are for the back of the box don't forget to count the PSU fan.
So you might end up with 1*80+1*120mm depending on you box and what PSU you choosed.

dukla2000 · Post by **dukla2000** » Mon Nov 17, 2003 2:41 am

Nightwatch - hopefully what you have got from this thread is the basic problems of doing a paper comparison

The real bottom line is you have to try some fans in your case with your mounting methods with your setup. Starting with 'good potential low noise fans' (Panaflo 80mm L1A, Papst 120mm 4412FGL ...) is helpful: you will tune from there and collect a set of 'spare' fans like the rest of us. I doubt fan perfection is ever achieved, or at least, not for me!

Below I highlight some of the deltas between 'my spreadsheet' and gmjames fanspec2.xls. I am NOT trying to score I am right (I admitted in my first post I doubted my 5V 4412 was 7dBA); maybe just emphasize doubts whether ANY spreadsheet can give a real answer (and thus come back to 'you have to try ...').

silvervarg wrote:Numbers might not be accurate in the excel sheet, but they are probably probably off by almost the same number for 80mm and 120mm fans.

OK - fanspec2 has 12dBA for a 4412FGL at 5V, I have 7dBA. The reason for this is I used the 'plain' log formula, fanspec2 (and the related article and discussions here) draws a curve through the published specs for all the fans in the 4412 range, and then extrapolates that curve.

Now previously I posted what I saw as a potential gotcha in this technique - my menagerie of fans includes 8412NGL and 8412NG, and optically the blades are different. Whether this is because of the batches of my fans, factory of manufacture, of indicative of other fan ranges from other manufacturers I do not know.

The second gotcha is static pressure. As I said 'to cause an argument' I assume 10Pa. In reality very few people know what static pressure they have around their PC fans. The best argument/debate (critcism I faced) was in this thread with SpyderCat. But bottom line from that thread is SpyderCat concluded his 9V L1A was effectively pushing 6.7cfm, which is significantly below the free air prediction most of us make for 5V. The real point here is that as far as I can see on paper, low flow 80mm fans are handicapped much more by static pressure than low flow 120mm fans. Certainly my Papst 8412NGL is useless at 5V and my 3412NGL (92mm) are next to useless at 5V.

Last - as per silvervarg at low airflow, the inherent fan bearing noise becomes a big issue. Hence the comparative rankings here at SPCR or recently at a Finnish site are golddust.

silvervarg · Post by **silvervarg** » Mon Nov 17, 2003 6:17 am

Thanks for a very interesting note.
It sure was a lengthy reading on the thread about how many cfm fans push in a real in box environment. Very interesting, but perhaps a very short summary of the findings might be in place.

Unfortunately the information in the comparative rankings here at SPCR does not give much help in this area. The Finnish review is great in this matter, but the major problem is that it only did review 80mm fans, so it does not help in comparing 80 and 120mm fans.

I believe we have come to the point where we have agreed that the solutions will be rather close (1*120mm vs 2*80mm) in sound level and that we agreed that the bearing sounds will be the major factor for what solution is most quiet.

Instead of going into the area of bearings I look a little bit into the area of static pressure.

Dukla2000:
The real point here is that as far as I can see on paper, low flow 80mm fans are handicapped much more by static pressure than low flow 120mm fans. Certainly my Papst 8412NGL is useless at 5V and my 3412NGL (92mm) are next to useless at 5V.

Now we get to the point where I start to disagree a lot with Dukla.
I have not seen anything presented that tells me that 120mm fans react better to static pressure than 80mm fans. Please provide info in this area if you have some.
I have seen vauge information that thinner fans are more handicapped by low airflow.

You mention a single fan used by SpyderCat and seem to generalise this to how 80mm case fans will react compared to 120mm case fans.
The fan SpiderCat used had narrow ducting (quite long) on the intake side and pulled all air through the CPU heatsink. He still reached 37% efficiency (according to calculations in that thread).
The same fan would have provided a lot higher efficiency if used as a case fan. Perhaps close to 60% (speculation).

Doroty:
That is realised cfm, and it is usual to de-rate your fans by 50%.

Considering Dorotys statement it seems that the 80mm L1A is probably going to do above average when dealing with static preassue.
Most people here use 25mm thick fans for both 80mm and 120mm fans. Iff you where thinking about 38mm thickness for 120mm fan that will make thing a bit different.

Nightwatch · Post by **Nightwatch** » Mon Nov 17, 2003 2:56 pm

Ok, thanks for the breakdown silvervarg, it seems this is a topic that one can debate about for a long time, but in the end, its pretty much what you hear from your box that will decide the matter. As for the fan collection, Dukla, mine is well and truly on the way, with 3*120mm (anyone want to buy a papst 120mm variofan?!!)and 5*80mm fans already in the drawer gathering dust as I had to try various fans that were readily available here to see if they were quiet enough. As i said previously, atm im using a sunon 120mm (38mm thick) fan which is the quietest ive found so far. On order I have the papst 4412FGL and I think Im going to contact Dorothy and see if she can ship some of the panaflo's to S.A. Btw, dukla, yours and gmjames' spreadsheets was how I found Papst here in S.A. Thanks again guys for your help.

dukla2000 · Post by **dukla2000** » Tue Nov 18, 2003 1:20 am

silvervarg wrote:
Dukla2000:
The real point here is that as far as I can see on paper, low flow 80mm fans are handicapped much more by static pressure than low flow 120mm fans. Certainly my Papst 8412NGL is useless at 5V and my 3412NGL (92mm) are next to useless at 5V.
Now we get to the point where I start to disagree a lot with Dukla.
I have not seen anything presented that tells me that 120mm fans react better to static pressure than 80mm fans. Please provide info in this area if you have some.

It is a flippant (non-scientific) comment that perhaps I should withdraw - certainly I feel the argument is impossible to win

. For a start, we would have to debate a baseline to compare 80mm and 120mm fans. 12V in free air is easy, but same airflow, or rpm, or dBA and/or blade tip speed ...???

But to explain why I made it: I have a spreadsheet of fan data that is for a start biased: for 80 mm I have mostly fans I have a sample of, for 120mm I have a lot of slow/quiet fans because of my search for a case fan. I chose 10Pa as some static pressure curves draw this on the axis, except others use mm of water or inches or ..., so many of my 10Pa readings are based on a ruler on a small graph! I have no firm opinion whether 10Pa is representitive of cases, power supplies, heatsinks or ducts - it was chosen merely as an 'easy' line to read on (some) graphs! (Somewhere in the Sunon or NMB or Comair Rotron or ??? docs there is a number that is higher than 10Pa quoted as typical for electronic casing but I was content to stick to 'my' 10Pa because of my neat cables, I figured a mid-tower case was a lot less cluttered than a stereo amp or ....

). OK - enough excuses, here is some data! The first 5 fans (listed in lowest quoted dBA order) I have 10Pa values for are:
80mm (cfm free air, cfm @ 10Pa)
Papst 8412NGL 19.4 0.0
YS-Tech FD1281252S-1A 26.2 8.0
Panaflo FBA08A12L1A 24.0 5.7
NMB 3110-KL-04W-B19 P50 25.0 8.0
ARX FD1280-S2000E (CH80S12M) 29.8 24.0

The average of these is 37% of their free air flow.

120mm
Commonwealth FP-108MS3 54.0 28.2
Papst 4412FGL 55.3 20.6
NMB 4710NL-04W-B10-P00 58.6 24.7
Nidec D12T-12PL 53.0 17.7
Sanyo Denki 9G1212M402 58.6 44.1

On average these are still pushing 48% of their free air value at 10Pa.

Regarding 38mm fans: I have also heard (or expect) they should stand static pressure better, except (in an even less scientific 'dukla survey' - probably of Papst 120mm cause I have the catalogue) I have not really found data to support it. I would not be surprised if the data exists: I am only saying my 10 minute paper chase did not obviously support the theory!

silvervarg · Post by **silvervarg** » Tue Nov 18, 2003 8:38 am

At the moment I lack any good facts, so I'll give this matter a rest, at least for now.
Great to see some numbers, even though the Papst 80mm can hardly go down to 0.0cfm... It just seems unrealistic.

The numbers point in the direction that 80mm fans should loose almost 2/3 of the cfm at this pressure and 120mm fans would "only" loose half the pressure.
I think it does matter wich way the fans are turned (over vs under pressure).
Also we might have the wrong pressure as you mentioned. Does anyone have equipment to measure pressure in the box while running?

I was almost certain that thicker fans should do better with pressure, but after reading your note I am not so sure anymore. Perhaps I will just work a bit extra to make sure I have as open intakes as possible to reduce the underpressure in my boxes.

Gooserider · Post by **Gooserider** » Tue Nov 18, 2003 5:28 pm

silvervarg:
At the moment I lack any good facts, so I'll give this matter a rest, at least for now.
Great to see some numbers, even though the Papst 80mm can hardly go down to 0.0cfm... It just seems unrealistic.

The numbers point in the direction that 80mm fans should loose almost 2/3 of the cfm at this pressure and 120mm fans would "only" loose half the pressure.
I think it does matter wich way the fans are turned (over vs under pressure).
Also we might have the wrong pressure as you mentioned. Does anyone have equipment to measure pressure in the box while running?

I was almost certain that thicker fans should do better with pressure, but after reading your note I am not so sure anymore. Perhaps I will just work a bit extra to make sure I have as open intakes as possible to reduce the underpressure in my boxes.

No solid numbers especially on undervolted stuff, but some observations based on published PQ data. My assumption is that static pressure capability would roughly track RPM or CFM ratings, so I would expect a fan's undervolted SP would be about the same as it's % of full power rating as it's undervolted CFM.

1. I have noticed that most of the PQ curves I've seen for 120mm fans showed that thick fans had a higher max static pressure rating listed than the thin models from the same vendor. Unfortuneately, it also showed higher pressures from the higher CFM / noisier models in the same line. In addition, I observed that the outputs of the thin fans dropped much faster than that of thick fans as SP increased.

2. I didn't spend as much time looking at 80mm curves, or 80 vs. 120 curves since my design uses only 120mm's, but it appeared that the 80mm models didn't have as much pressure potential as the 120mm units.

3. BillA of ProCooling and Overclocker's fame did some fan testing to see how they stood up against pushing air through rads. His data seemed to show that thick 120mm fans did best in terms of delivering the highest %age of free air rated CFM. He said both the CFM delivery of both 80mm and thin 120mm's tended to fall off badly as soon as they hit any sort of backpressure. BillA is one of the more scientifically oriented test people out there, and I would consider his work to be highly reliable and authoritative.

I'm not sure how a thick 80 would compare to a thin 120, but IMHO if I had any kind of back pressure / flow resistance concerns, I would go for the thick 120mm.

Gooserider

dukla2000 · Post by **dukla2000** » Wed Nov 19, 2003 1:10 am

silvervarg wrote:... even though the Papst 80mm can hardly go down to 0.0cfm... It just seems unrealistic.

No it is true - see the bottom of page 1 here! This is when I started being concerned about static pressure: I had just bought some 8412NGL, read 'static pressure' somewhere and suddenly realised my '12dBA' fans came complete with a near inability to move air when confronted by any obstacle!

silvervarg wrote:I was almost certain that thicker fans should do better with pressure, but after reading your note I am not so sure anymore.

Me too (expect thicker fans to cope better). Except I have just done another in-depth (10 minute paper

) review and figure the 120mm Papst range has some 'fair' attributes for comparison, in that there is a 4300N range (32mm) and 4200N range (38mm). (At 80mm and 92mm the thicker fans are not N.)

Comparing 2 fans that are 2450rpm, 5.1 bel (@12V in free air),
4312NM (32mm) is 136m3/h free air, 125 @ 10Pa and 50 @ 30Pa
4212NM (38mm) is 144m3/h free air, 128 @ 10Pa and 45 @ 30Pa
Although the thicker fan is better at 10Pa, it is also better in free air - indicates to me the Papst design (at least) is that thicker fans are targetted at higher cfm rather than higher static pressure. And at 30Pa (again an arbitrary choice, figured these bigger fans may be better at higher static pressure) the thinner fan seems better

Comparing 2 fans that are similar cfm (97.7 v 97.1) @ 12V
4312NN (32mm) is 166m3/h in free air, 155 @ 10Pa and 85 @ 30Pa
4212NN (38mm) is 165m3/h in free air, 150 @ 10Pa and 68 @ 30Pa
Again the thinner fan seems better!

Obviously the data is wrong! Does anyone else have enough curves from some other manufacturer that would support a theory that thicker fans cope with static pressure better! (Please

)

silvervarg · Post by **silvervarg** » Wed Nov 19, 2003 2:14 am

As I saw how the numbers just didn't make any realistic sense to me I started to wonder just how much underpressure 10 Pa is.
My initial guess was that 10 Pa was a realistic underpressure in a computer box, but seeing facts that a 12dBA 80mm Papst should move no air at that pressure made me ponder. In reality many people use that fan as a case fan and it does work fine.
Since I didn't doubt the facts presented for the Papst fan the number that had to be way off is the 10 Pa static pressure.

The fan is rated 19.4cfm@12V.
The most reliable source I could find is the Finnish review of 80mm fans (the best fan review I have ever read).
That test is in a controlled environment where fans suck air through a tube and an airflow meter. My guess is that the resistance from the tube plus the airflow meter is fairly equal to what we get in a very well ventilated box (that hopefully most entusiasts in this forum has).
In that test the 12dBA 80mm Papst moves [email protected] and [email protected].

I believe that the main reason is that we will not get much of static pressure difference at all in a good case. All we will get is restricted airflow due to obstacles, but that is not exactly the same as static pressure.

So I do agree with the numbers that slow RPM fans react poorly to high static pressure, but it does not seem like we have any high static pressure difference in our boxes.

Gooserider · Post by **Gooserider** » Thu Nov 20, 2003 8:49 pm

I just did some poking around in a few fan spec PDF files that I had handy to compare 38 and 25mm fans. It was interesting, and seems to show that 38mm fans do put out higher pressure. Not sure how this will relate to undervolting, but...

I looked at the ADDA 12025 (25mm) and 12038 (38mm) series of 120mm fans. The PDFs didn't have PQ curves, so all I have to go by is the max pressure specs. I didn't include dB specs, but it is worth noting that few of these fans are ones that SPCR would consider tolerable. I include the higher noise / CFM models on the theory that the fans may spin the same blades at different RPM's, and thus looking at the different speed models may give a glimpse of what might happen with undervolting.

In the following table, I'm listing each pair of fans, first the 25mm, then the 38mm in each speed letter. I include the CFM ratings, and the max static pressure, in mm of H2O. If you'd have preferred different units, I'll let you do your own conversions

Code: Select all

Speed  RPM    CFM   SP (mmH2O)
U 25   2500   98.6  4.48
U 38   3200  125.1  9.50 *
H 25   2200   85.2  3.45
H 38   2800  104.8  6.86
M 25   2050   80.5  2.97
M 38   2540   94.0  6.20
L 25   1800   72.0  2.41
L 38   1950   72.3  3.43
(* 24V model, not listed for 12V, but other models were same for both voltages)

Note that in every case, the 38mm model gave much higher static pressure than the 25mm unit. In the case of the L models, which had similiar RPM and CFM ratings, the difference was smaller, but still quite significant.

(Next post, comparison with Delta fans)

Gooserider · Post by **Gooserider** » Thu Nov 20, 2003 9:49 pm

Continued from previous...

Doing the same style of chart, this time using the Delta AFB120120254 (25mm) and AFB12012038 (38mm) series fans. These did have PQ charts on them, and it appeared to me like the 38mm charts didn't have as steep of a slope to them, implying that they would hold the CFM longer in the face of increased static pressure.

Code: Select all

Speed   RPM   CFM   S.P. (mmH2O)
L 25   1900   62.9  3.93
L 38   2000   84.8  4.56
M 25   2200   73.8  5.15
M 38   2300   95.0  6.00
H 25   2500   82.7  6.25
H 38   2600  105.9  7.60
HH 25  2800   95.1  7.73
HH 38  2900  120.1  9.00
VH 25  3100  103.8  9.43
VH 38  3200  130.0 10.70
SH 25  3400  113.1 10.92
SH 38  3700  151.9 14.50

These are a bit more confusing. In any given speed, the 38mm has higher airflow and pressure, but also more RPM.

If you attempt to get comparable CFM's, the 25mm will have the higher pressure, but at a MUCH higher RPM.

If you go for comparable pressures, then the 38's give more volume at lower RPM.

This set of data doesn't exactly confirm the '38's give more pressure' theory, but it does seem to imply that they might. It certainly appears that the 38's move more air volume per fan RPM, which presumably means higher potential pressure per RPM, which is a good thing IMHO.

Gooserider

Gooserider · Post by **Gooserider** » Thu Nov 20, 2003 10:01 pm

silvervarg:
I believe that the main reason is that we will not get much of static pressure difference at all in a good case. All we will get is restricted airflow due to obstacles, but that is not exactly the same as static pressure.

This is one that I disagree on totally... Static pressure is CAUSED by restricted airflow and NOTHING ELSE! How much pressure is seen will depend on the amount of restriction.

If there is no restriction, then you are by definition in 'free air' and will have zero static pressure and get the free air rated output.

If there is total restriction, then you will have zero volume, and the max rated static pressure.

If there is only partial restriction, then you will see some reduction in volume from the free air rating, and some increase in static pressure from zero.

The reason we don't get much static pressure in a good (or even a bad) case is only partially due to low restriction, more importantly it is due to the simple fact that a rotary fan is not capable of producing much static pressure, but instead simply moves less volume.

Gooserider

Gxcad · Post by **Gxcad** » Thu Nov 20, 2003 11:08 pm

Pardon my ignorance for not reading the entire thread, I basically wanted to just drop down my own opinion and that is that 2x80mm is better than 1 120mm fan. Why? Perhaps my reason makes it easier for me to be definitive in my answer because my reason is not superior airflow vs noise but rather than 80mm fans simply can be quieted further than 120mm fans. There is no 120mm fan in existance than can be as silent as 2 panaflo 80mm's at 5v. I have a 120mm fan setup and a few quiet 120mm fans, yet even at 5v they are not quiet enough for me, wheras my 80mm L1A's are easily inaudible at 5v and even at 7v are very quiet.

-Ken

JVM · Post by **JVM** » Thu Jan 01, 2004 11:02 am

A Panaflo 80mm at 7V produces 9.3 db and 14.0 cfm. A Papst 4412 120mm at 7V produces 14.3 db and 32.3 cfm. Looks like one Panaflo at 7 volts is hardly moving any air. So if you use 2 Panaflos at 7V and get 12.3 db and 28 cfm, then you get slightly less noise with less cfm. But I think the real issue then becomes how loud is your PSU? If your PSU fan is producing 20-23 db, then that will likely overcome any noise by the 120mm making 14.3 db while giving better airflow and perhaps making life easier for the PSU.

silvervarg · Post by **silvervarg** » Fri Jan 02, 2004 7:46 am

JVM:
A Panaflo 80mm at 7V produces 9.3 db and 14.0 cfm. A Papst 4412 120mm at 7V produces 14.3 db and 32.3 cfm.

These are calculated DB values, and possible the CFM values also. They do not take clicking noises etc into account, so they might not be accurate. 120mm fans tend to have a bit more bearing noise when undervolted a lot due to the heavier weight of the fan hub and blades.
So if like Gxcad you think a 120mm fan @ 5V is to loud you need to go for smaller fans. I assume he has done something to quiet down his PSU a lot as well.
Most people will think a 25dBA computer is very good, but a few people in this forum want to go a lot lower before they are content (if they ever will).

JVM · Post by **JVM** » Fri Jan 02, 2004 9:44 am

Fan clicking can occur with any size and so I don't know how much more of a problem it would be with any particular size but maybe fan quality has more to do with it.

The most important thing about noise to me, aside from loudness, is the type of noise. Clicking and high-frequency type noises are the most annoying to me. Sometimes I could say louder is better if the lower volume produces a more annoying noise.

silvervarg · Post by **silvervarg** » Fri Jan 02, 2004 2:18 pm

Fan clicking can occur for all sizes, and usually it does to some extent on almost all fans when you turn the voltage low enough.
Both from my own experience and from what other have written in this forum the tendency seems to be that bigger fans click more and have more bering noise if the fans are the same brand.
In fact I don't think anyone has complained about 40mm fans that click.

Some brands tend to click a lot more than others and some brands have a lot bigger quality spread than other brands.

This comes from the fact that the bigger mass that is moving the more will inperfect balance cause bearing noise since a larger force will push against the bearing. You can see that even heavy aluminum fans use light plastic in the moving parts to reduce this effect, but they use plastic in a metal-lookalike-color.

JVM · Post by **JVM** » Fri Jan 02, 2004 3:50 pm

The thing I've noticed from reading on this website is that cases with 120mm rear exhaust fans are preferred. Even the top-rated Evercase 4252 is recommended for 120mm over the 80mm model. So, it appears to me that this issue of 80 to 120mm for exhaust is a simple matter of the 120mm fan being the way to go.

silvervarg · Post by **silvervarg** » Sat Jan 03, 2004 12:38 am

There are very few cases that take 2*80mm fans as rear exhaust. The ones that do often have other drawbacks. E.g. more poorly buildt etc, so it is hard to find a 2*80mm exhaust case that you don't need to modify a bit.
Once you start to modify cases they become even harder to compare.
So it is often easier to get a case with 120mm exhaust.
One good option is to use 1*80mm rear+1*80mm in the top blowing up.

Another bonus for the 2*80mm is that you get some redundancy. Even if your fan fails you will still have some cooling left. So for the same reason people get RAID systems for redundancy 2 fans is preferable over a single fan.
This analogy is not perfect, since you still have other fans that you depend on, so you still have single point of failure.

Post by **MikeC** » Sat Jan 03, 2004 12:47 am

Partly depends on the system you're wanting to cool. 1 or 2 quiet 80mm fans is almost invariably quieter than 1 quiet 120mm fan, but if you don't need the cooling airflow of the 120mm, then why bother?? My hottest system is a P4-2.8; a single Panaflo at 6-7V for exhaust and a 5V Panaflo in the PSu is enough to keep it perfectly stable all the time.

JVM · Post by **JVM** » Sat Jan 03, 2004 6:14 am

There are a few cases with 80mm fans for exhaust that I strongly considered: Lian-Li 6070 and SilentMaxx ST-11. But both cases have significant drawbacks - airflow! The Lian-Li is aluminum and looks like people are complaining about vibration along with poor airflow. Sounds to me like steel is a better way to go. The SilentMaxx looks great but airflow is a problem and plastic fan mounts don't help. I am not into modding cases so that doesn't leave much of a choice. I can get the Lian-Li 6070 with rear case grill cut out for just $15 dollars additional and it was very tempting.

I don't doubt 80mm fans are quieter than 120mm fans. But I really want a case with a door - oh that Lian-Li 6070 is THE case with a door - but had to settle on the Sonata because I prefer air intake on the bottom as opposed to front. Yes, I know front intake is better for airflow, but I feel noise could escape the front intake. Another thing for me is I have options with cases having intake on bottom and that includes an acoustical box with front door that I could use to further reduce noise while not interferring with airflow since air gets in from bottom. I also like having a door to keep my cats from turning on the computer and getting their fur in the intake. You should see the bottom of my room air cleaner with all the cat hair that gets trapped on the intake and how I have to dig it out.

So, got to have a door. Even better with my acoustical box covering the whole front - and that means air intake on bottom.

I would have been very happy if Lian-Li made that 6070 with steel instead of aluminum, although I'm not crazy about 2 front intake fans. But the Nexus 80mm fans undervolted would probably be quiet enough to make that situation okay. However, the thin aluminum side panels on the Lian-Li just kind of scared me away. And the "foam" used in that case doesn't appear to be significant in terms of acoustical properties.

Still, the Lian-Li 6070 looks so good...?

One other thing is while 80mm fans are quieter than 120mm fans, the PSU fan (80mm) is going to produce something in the 22-23 decibel range and so would most likely drown out a 120mm Papst undervolted to 5 or 7 volts.