Probably pwm fan share will slowly grow by time (regarding current best being 3-pin analogue voltage regulated ones), just as pwm header count on yer average motherboard compared to old 3pin ones count.
Probably pwm fan share will slowly grow by time (regarding current best being 3-pin analogue voltage regulated ones), just as pwm header count on yer average motherboard compared to old 3pin ones count.
As pwm versions of the best fans come to market I believe MB makers will add more pwm to their boards
The pic based controllers should prove interesting ....If the cost is reasonable. I am thinking that resistance based control for 3 wire fans is reliable until we solve the issues. even a switched resistor network would do for most people.
![]()
My Biggest Fear Is When I die, My Wife Sells All My Stuff For What I Told Her I Paid For It.79 SB threads and 32 IB Threads across 4 rigs 111 threads Crunching!!
Martin,
There are a few advantages of PWM.
Not the least of which is the fact that the RPM is regulated the same on all fans regardless of how they're made (it doesn't matter if the fan turns on at 4v or 7v - both will have about 50% RPM at 50% PWM duty cycle).
Also, PWM lets you use MOSFET transistors, which usually have a considerably higher power rating than BJT transistors with the same size/package.
BTTB - Gigabyte Z87X-OC - WCed I7 4770k - 2x8gb Ballistix 1600mhz - Zotac GTX 780
Asus Xonar Pheobus - OS -> Toshiba Q 256gb - Games -> 2x Agility 4 256gb Raid0
Corsair HX850 - Tecnofront HWD BenchTable - Asus VE278Q 5760x1080
Serveur - Asus Z77m PRO - 2500K - NH-C12P - 4x4gb G.Skill Ares 1600mhz
Agility 4 128gb - Corsair CX430M - 1TB Black - 2TB green - 2TB Red
KatPat - Gigabyte 990FXA-UD3 - FX-8320 - PH-TC14CS - 2x4gb Viper 1600mhz - GTS 450
Samsung Evo 120gb - Corsair HX750 - Bitfenix Survivor White - Asus VE247H
Would you be willing to test one more 120mm fan? I'd be willing to have a 120mm and 140mm Be Quiet! USC fan sent to you via Aquatuning.us. Send me a PM if interested!
Imho it's the fan to beat for very low noise levels. I have many fans (see below) but this is honestly the only one that I can say is virtually inaudible at 500-600 RPM and close range.
I have tested it against many fans, like the Gentle Typhoon (great 1k+, but I just can't get them anywhere near inaudible), Nexus, Noctua P-series, Enermax Magma, Arctic F12, all the Noiseblockers (want more of those?) for comparison but they just don't match up. I would be really interested to see if it's only the noise level that's really good compared to the RPM or if the noise/CFM is also great.
And thanks for doing this!
Good discussion.
I guess I did miss the lower RPM advantage of PWM. This PWM controller does give me a slight advantage in have a lower RPM floor (Maybe .5 to .7 "Volts" worth). I don't quite understand why yet, the PWM signal is almost converted to lower voltage using the large capacitor but it does seem to have a better ability at going a little lower in speed..
So far I haven't noticed any benefit to "Less motor Tick" with this particular PWM controller, but it also appears there are several types and ways of doing PWM control.
When I play around with my Crystalfontz PWM controller, it actually has an obnoxious "Kick" if I dial it down really low. The fan will actually torque on off at a very slow cycle rate (maybe 4times per second) and you can even feel and watch a strong fan move at each kick. The designer of the PWM circuit I used to build the video test actually notes he had this same problem until adding the large cap to smooth out the PWM pulses. The only good thing about my Crystalfontz controller is that it maintains RPM readout, so it must be reading during each pulse and calculating or something along those lines.
Regardless, so far I haven't found "3 PIN" PWM to be superior to voltage as I had hoped.
Either you get the "PWM KICK" noise without the cap, or "Voltage TICK".
Seems like 4pin PWM is the only real good PWM setup, but even that is a bit "Fan Specific". Some of my 4 pin PWM fans have a large range of control, nearly 0-100%, while most seem to only turn down a little bit...and some make some noise, while others seems to be pretty good.
I plan to do some 4pin PWM testing after I get my last order in using my 4 pin PWM controller Bing designed.
Last edited by Martinm210; 11-30-2010 at 03:33 PM.
Martin, fyi, at that diy controller of yours, with just an additional of 3 poles switch and a resistor, it will cover full range with 3 zones at 0% , 2-98% and 100% like this here :
That resistor values is not critical, as long above 100 Ohm, just look at your components bin, any resistor that has a resistance in hundreds of ohm should be fine, heck, even 1K should be ok too.
PWM controller mod.png
Of course this minor mod will not have as the other circuit that has a clean and smooth range from 0 to 100%, but think this should be enough to serve your purpose.![]()
Last edited by bing; 11-30-2010 at 07:35 PM.
Great, btw do you have any plan to include Thermalright TY-140 in your 140mm fans test ?
Really curious how it perform and the outcome if its one of your 140 test, cause I own few of them and imo it beats Noctua P14 and Yateloon's 140 in air hs at the noise/cooling performance ratio, subjectively of course. Not sure bout it's cfm though.
No, I only have the TR X-Silent, it's the only TR 140 with a full 140mm frame. I looked at the TY, but since it's a 120mm fan mount, I'm not sure how I would go about mounting it to the 140mm radiator. I have the 120mm to 140mm adapter that should allow me to mount some 120s on the 140mm radiator, but that adapter is pretty tight to the 120mm spec. If anything the adapter is a touch small. The TY140 would overhang and loose a bunch of pressure.
Hehe... just TAPE (double-sided) that TY-140 on there!
I dunno... is that too much of a variable as to invalidate the results?
But almost no-one gonna use such piece of wood on real rads/real builds, and existing 120-140mm fan adapters (eg. BP's or Koolance's ones) have limited inner size (for 120mm fans), so only way to go is to make some custom adapters or use only on rads that have both 120mm & 140mm mounting holes (eg. tfc monsta and aquacomputer's one). So while yes, TY-140 shown very good cfm/db in Jordan's test, it's non standard mounting holes & fan casing makes it hardly usable on LC rads and most probably it will be seen (if ever) in veeery small amount of LC builds.
TR X-Silent shown good cfm/db as well, but was little set back by it's max rpm-s of only 900 IIRC.
Good ideas. I don't mind entertaining any sort of side experiments, but I haven't been sponsored any of the 140mm types with120mm frame mounts. I also spent the last of my fan fund, so no more unless someone wants to step up.
It might be better to wait until I get the 140s I have on hand done first anyhow...hoping to do that this coming weekend. Then we'll see if there are any tends or stonger types to focus more on.
Martin, What pwm fans do you have or have ordered so far?
Could add funds for
http://www.crazypc.com/products/afc1212de-58162.html and http://www.crazypc.com/products/afc1212d-58170.html (the second one is an af but 25mm which makes me curious....
Let me know if interested
EDIT it might be interesting to try and find the PLA12025S12M-4 fan from the Xigmatek aegir (a 2000 rpm pwm model)
Last edited by OldChap; 12-01-2010 at 02:38 PM.
![]()
My Biggest Fear Is When I die, My Wife Sells All My Stuff For What I Told Her I Paid For It.79 SB threads and 32 IB Threads across 4 rigs 111 threads Crunching!!
OldChap: For Delta AFC1212DE: "5pin / 4 wire PWM (DELL Style Connector)" - what's that? :/
I couldn't tell what the 5th pin is but dell and HP and others have built MB's with this format of pinout for cpu fans and on some you need to be wary of which pin is which. I first saw it on an optiplex desktop board and later on a HP 9400 workstation board. Other than that the fan works just fine. The connector may need the lugs cutting off to use it though
![]()
My Biggest Fear Is When I die, My Wife Sells All My Stuff For What I Told Her I Paid For It.79 SB threads and 32 IB Threads across 4 rigs 111 threads Crunching!!
This is what I've got lined up for the 4 pin PWM test, not very many...but a place to start.
Sure, I'm game. You sponsor & pick, and I'll test...either one looks like a good option assuming it follows the AFB1212VHErev3 qualities.. I have the non PWM AFV1212HE in route, but that'll be the standard voltage control test. I could play around with my 3 pin PWM controller some more too, but I'm not seeing any noise differences with that one yet.
Last edited by Martinm210; 12-01-2010 at 03:16 PM.
That is one hell of an assumption but there is one way to find out
I'll get some more funds to you tomorrow. I think you saw that dell delta somewhere cheaper so I'll leave that part up to you.
![]()
My Biggest Fear Is When I die, My Wife Sells All My Stuff For What I Told Her I Paid For It.79 SB threads and 32 IB Threads across 4 rigs 111 threads Crunching!!
So are you interested in me sponsoring you some fans? I can have some 140mm Noiseblocker and Be Quiet fans sent to you, and a 120mm Be Quiet as well. Or I can actually and try have them send you some for free, I think I know a guy who might know some guys![]()
Aquatuning offer to sponsor some fans too a while back. I turned them down at the time because I was feeling a bit overwhelmed with the growing pile of fans I had untested at the time, but I could ask if they are still willing. I'm starting to get caught up and feeling a little better now.
I really have no idea what to try, but I'm open to anything.![]()
Last edited by Martinm210; 12-01-2010 at 04:36 PM.
Yes, that 5 pins Dell's connector at this fan can be replaced with the standard 4 pins pwm connector, here is mine :
P1020274 AFC1212DE 2.jpg
Good fan that covers wide range of rpm while very quite when slowed down to minimum speed.
Recommended for Martin to test it.![]()
If you were to take #AT up on that offer, and you had the time, and didnt mind the extra work, i would be very interested in the YL 140m 2300rpm
I am going to need a whole pile of these soon and i am curious to know what i am letting my self in for noise wise.![]()
Last edited by affiliate13; 12-02-2010 at 12:55 AM.
ASUS Rampage II Ex 1504
Intel i7 920 D0 3850A795
Sapphire HD5970 EK FC 1Ghz/1150Mhz 1.237v
OS Intel X25-M RAID0
Data 500GB Seagate Barracuda 7200/32Mb
OCZ Reaper 6GB/1800/C8
BeQuiet Dark Power Pro 1200w
1X 18w DDC3.2 w/XSPC Res Top
1X 18w DDC3.25 w/XSPC Res Top
1X XSPC Submersible Pump/Res w/RS120mm
HW Labs GTX480 4x120mm
TFC Monsta 6x140mm
EK Supreme LT 3/8" Feser Tubing
EK IOH/SB and Mosfet Blocks
Lian Li V2110 w/EX-34
26" Iiyama Prolite
With blades like that that delta probably good pressure fan?
Last edited by Church; 12-02-2010 at 01:54 AM.
my motherboard already has 4-pin PWM fan headers. but as you said, right now we are bound to 3-wire fans as there are no real alternatives in PWM land available. may be TY-140 but unfortunately its not available until now. so i stick with gt's for now ...
yes pic based controllers are the way to go. this allows full control of the fan by software. i'm not interested in having this fancy lightning knobs somewhere on the front of my pc which in fact is located under the desk. prefer to control everything from the desktop. so that's the way i go now.
don't think the problem with PWM control for 3-wire fans can be solved in any way. as fans are build of brushless dc motors, they all have some sort of circuitry built in which switches the the magnetic field in order to allow the fan to rotate. this circuitry gets confused by applying a PWM signal instead of a steady voltage. some fans have more problem with PWM than others.
agree to 100%: for 3-wire fans, voltage control is the way to go.
Processor: Intel Core i7 990X
Motherboard: ASUS Rampage III Extreme
Memory: Corsair CMT6GX3M3A2000C8
Video Card: MSI N680GTX Lightning
Power Supply: Seasonic S12 650W
Case: Chieftec BH-01B-B-B
Bookmarks