Quieter Case Fans

[Johnf1285]

After inspecting the intake, exhaust, and PSU fan in my 2010 Mac Pro, I have found the following:

Stock Mac Pro exhaust and intake fans are basically this same model, but with a 4pin connector and gray.
http://www.newegg.com/Product/Product.aspx?Item=N82E16835213005

It appears that the stock Mac Pro PSU fan is also a 120mm fan, but of the standard depth, also with a 4 pin connector. I could not find any similar product based upon the model number stamped into the stock fan. It looks like your run of the mill 120mm case fan.

However, the intake and exhaust fans are a bit beefier, with a slightly more aggressive blade orientation. The stock fans are attached to their plastic fixtures with rubber fasteners that need to be replaced whenever a fan is swapped within the plastic fan fixture. Short of their unique fixture, its a regular computer case fan.

I am kicking around the idea of installing three of these, 1 as the intake, 1 as the exhaust and 1 as the PSU fan:

http://www.newegg.com/Product/Product.aspx?Item=N82E16835426015

The GELID fan connector matches that of the Mac Pro Delta fans (hopefully no splicing and soldering of connectors). The GELID seem to be very quiet yet still effective. The GELID fan also includes 4 rubber fasteners which is a big plus, AND the cable is sleeved, so it will look pretty in the Mac Pro.

I have another set of stock Mac Pro intake and exhaust fans, complete in their fixtures, and a stock PSU fan, so I think I might be willing to try this out. The biggest concern is how the GELID interfaces with the Mac Pro and whether or not it will provide adequate cooling at the right RPM.

Under full load in Prime 95 (OSX), I never saw intake, exhaust, or PSU fan spin over the default 600 RPM. Obviously the GELID fan's have a different design and move less air per RPM compared to the Deltas. However, the question is at what RPM is the GELID air movement rating assessed at, and at what RPM is Delta's air movement capabilities assessed at? Does the Delta fan provide so much cooling at 600 RPM that it does not need to increase its speed? Will the GELID at 600 RPM also provide enough cooling too or will the RPM's increase accordingly? Or will they run full speed because their incompatible with the logic board?

I think this will be an interesting experiment. I foresee the GELID being a direct replacement to the PSU fan as the stock one already looks extremely basic. However, the air movement capabilities of the GELID versus the Delta has me guessing what the result will be like.

Other side project includes seeing what the PCI fan is all about and a potential way to passively cool the HD5770.

The sound doesn't bother me, but I just cannot stop tinkering with this thing!

Will post updates this week with pictures.

 

[handheldgames]

Reading the specs, it looked like there are 2 cables coming off the fans. After 1 year my 2009s are starting to get louder. Cheaper quality than 2008. The 80mm pci bay fan is the loudest.

 

[ValSalva]

am kicking around the idea of installing three of these, 1 as the intake, 1 as the exhaust and 1 as the PSU fan

I am thinking about replacing the front intake fan with something much quieter. That has been the main source of annoying noise on my 2010 Hexacore Mac Pro. It's not loud but it is clearly louder than most of the home built quiet PC's I've made for work.

Do you know of any guides on how to replace these fans and how much disassembly it requires? I've never worked extensively inside of a Mac Pro. Thanks!

 

[cherry su]

What about Noctua fans? Those have a very good reputation of being quiet, although they're more expensive than average fans.

 

[Transporteur]

Given the height of the PSU, the fan in there can't be a 120mm.

My bet is on 80mm.

 

[DanielCoffey]

But are the Apple fans PWM fans even though they have a 4-pin connector?

 

[Johnf1285]

PSU is 120mm.

I hope the pwm GELID fans work. I am willing to try this out, and if I need to return the fans so be it. Today I am going to investigate the PCI fan and see what I can do about that and maybe add to my order. That and the booster fan on the heatsink are the loudest. I don't feel like reapplying thermal grease, so that is staying.. for now. But the PCI fan may need some changing too!

 

[Johnf1285]

Here are some pictures. The first picture is all four fans (PCI, PSU, Intake, Exhaust). Next is the information that is on the label of the Intake/Exhaust fans. Then the information on the PCI fan. The next picture is the only information I could find on the PSU fan. The last picture is the following connectors from left to right. Exhaust, Intake and PSU. The PCI fan has a strange proprietary 6 pin connector that mates right into the logicboard. Replacing this fan looks like some splicing of wires and the original conncetor is required. I plan to try replacing the three 120mm fans first, then take a stab at the PCI fan depending upon the results of the 120mm fan replacements.


Edit -

PCI Fan looks as if its a 92mm fan according to its model number after a quick google search. Information on PCI fan http://www.delta.com.tw/product/cp/dcfans/download/pdf/AFB/AFB92x92x38mm.pdf


Information on exhaust/intake fans - intake/exhaust http://www.delta.com.tw/product/cp/dcfans/download/pdf/AFB/AFB120x120x38mm.pdf

The PSU fan looks like its made by RAIDMAX. The only thing I could find that was close was a RAIDMAX fan with the model number PBT-GF30. The Mac Pro PSU fan is PBT-GF25. Could be a totally different manufacturer though.

 

[Johnf1285]

Turns out I had a PWM 4 pin connector fan that was used in an old pc box. I pulled it out and tested it.

Unfortunately it ran at 1300rpm and speedfan couldn't do anything to change it. Oh well, looks like I ought to cancel my Newegg order asap.

 

[handheldgames]

Turns out I had a PWM 4 pin connector fan that was used in an old pc box. I pulled it out and tested it.

Unfortunately it ran at 1300rpm and speedfan couldn't do anything to change it. Oh well, looks like I ought to cancel my Newegg order asap.

I've been using these fans for years. Silent and should be controllable. Double check the pinout on Theo connectors. Perhaps apple made them "special".

http://www.newegg.com/Product/Produ...rder=BESTMATCH&Description=Scythe+fan&x=0&y=0

 

[Garamond]

I've been using these fans for years. Silent and should be controllable. Double check the pinout on Theo connectors. Perhaps apple made them "special".

http://www.newegg.com/Product/Produ...rder=BESTMATCH&Description=Scythe+fan&x=0&y=0

I'm also thinking of replacing my MP's case fans. Any idea which one of those linked would fit?

 

[DanielCoffey]

From what I have read elsewhere, Apple case fans may not be PWM but use the 4th line for speed regulation in a different way.

 

[Johnf1285]

From what I have read elsewhere, Apple case fans may not be PWM but use the 4th line for speed regulation in a different way.

Supposedly the 4th lead on the Delta fans that Apple uses, uses an analog signal to adjust the speed, as opposed to these fans using a "digital" signal for PWM. I have a feeling my experiment is ending before it begins lol!

 

[philipma1957]

I have run into various problems when trying to put in better fans. Apple may have switched the controller wires. you may need to cut and splice wires to make it work. Pos and neg leads are two wires and the other two wires for control of fan speed may simply be switched.

I have had this happen two or three times runs into money to test because you need to cut and splice to test out the new fan. My money is they reversed the controller wires. Of course I may be wrong and the apple fan may have a chip built into the motor that will only read a special code.

 

[Johnf1285]

I have run into various problems when trying to put in better fans. Apple may have switched the controller wires. you may need to cut and splice wires to make it work. Pos and neg leads are two wires and the other two wires for control of fan speed may simply be switched.

I have had this happen two or three times runs into money to test because you need to cut and splice to test out the new fan. My money is they reversed the controller wires. Of course I may be wrong and the apple fan may have a chip built into the motor that will only read a special code.

Yeah I am thinking, much like most hardware Apple uses, it may appear like a typical PC component but has some special firmware or addition to it. The other thing I can think of is using maybe a 3 pin fan that has its own speed control switch. As the intake and exhaust have ALWAYS ran at 600rpm on my machine, even under full load, this could be a very viable option! I will try this out if the GELID fans do not interface well. Luckily I have a few of the fans with their own speedcontrol switch lying around from a former Antec case

 

[ActionableMango]

According to this thread (http://www.silentpcreview.com/forums/viewtopic.php?t=58212), the Apple fans use voltage on the 4th pin to control the fan speed, not a digital signal.

It is probably too complicated to convert the signal.

I've seen fans with built-in temperature sensor and speed control. There is a pot to adjust the base speed and they go up from there depending on the temperature.

 

[j2048b]

Yeah I am thinking, much like most hardware Apple uses, it may appear like a typical PC component but has some special firmware or addition to it. The other thing I can think of is using maybe a 3 pin fan that has its own speed control switch. As the intake and exhaust have ALWAYS ran at 600rpm on my machine, even under full load, this could be a very viable option! I will try this out if the GELID fans do not interface well. Luckily I have a few of the fans with their own speedcontrol switch lying around from a former Antec case

not sure if it even works correctly but on my mac pro 2008 version i am using smc fan control and my specs show this;

exhaust=1214 rpm's
cpu memory=1300 rpm,

and they only seem to be loud if i up them, but run pretty quiet as of now, but also wanted to see about changing mine out to silent ones!

good luck with this quest!

 

[Johnf1285]

According to this thread (http://www.silentpcreview.com/forums/viewtopic.php?t=58212), the Apple fans use voltage on the 4th pin to control the fan speed, not a digital signal.

It is probably too complicated to convert the signal.

I've seen fans with built-in temperature sensor and speed control. There is a pot to adjust the base speed and they go up from there depending on the temperature.

I'm going to try out a few things tomorrow night. I have a feeling the GELID fans will not be working. I will try out the other fans I have laying around. All this tinkering in the case makes me wish I had another pc build

 

[mif]

Apple is trying, heavens .. Propellers are better than the model 1.1. Kiitos.

 

[Garamond]

Apple is trying, heavens .. Propellers are better than the model 1.1. Kiitos.

Well, not much better than the 2,1 models anyway. My 5,1 is noisier than my old 2,1 from '07.

 

[Johnf1285]

Expected results with the GELID fans. The machine booted and cranked the fans to 1300rpm. They weren't too loud at that speed! lol.


Oh well. At least we know that the Delta fans aren't so cheapo ones.

 

[nobiggiestudio]

This might be a late reply but anywho...from a Dell Dimension 5100 I got nmb-mat fans 120X120X30. And the the airflow is 201 cfm rather than the mac pro cpu fan...was it 121cfm or something?.
These fans are expensive but great. These fans has raw power!
The mac pro's Delta fans is powerful but in a gentle way compared to NMB-MAT raw power fans at full speed.
But I don't know if you could buy them per piece. Or you would need to bulk order 100 lots per shipping or similar.

 

[ScottishCaptain]

Supposedly the 4th lead on the Delta fans that Apple uses, uses an analog signal to adjust the speed, as opposed to these fans using a "digital" signal for PWM. I have a feeling my experiment is ending before it begins lol!

This is true. The 4th pin on the Apple fans is an analogue voltage, and not a digital PWM signal.

The Delta fans inside the Mac Pro are custom built units. They use the 4th pin to vary the voltage across the rotor coils, rather then as a trigger that rapidly turns the coil on and off at a varying frequency. This eliminates the whole knocking issue you'll get with a crappy PWM system where the drive frequency is lower then it should be.

I would not recommend replacing the Mac Pro fans- unless you're replacing them with the same part (the Apple Delta units). The SMC won't be able to vary the fan speed which will cause your AHT tests to fail. If your fans don't spin as fast as they need to spin to handle the full load of the machine, then the computer will overheat and the system will probably shut itself down (it should, anyways).

It is probably too complicated to convert the signal.

Depends. I know someone who did a conversion mod and landed up building his own uC-based PCB. It took him about 3 days to design the PCB, write the uC code, and build 3 of the units. They translated between the minimum and maximum analogue voltages of the Mac Pro logic board, and threw out a PWM signal that a normal 4-pin fan would accept. As I recall the other pins were pass-through, the replacement fans were mostly a direct replacement for the Mac Pro units (min/max RPM wise) so the tach signal didn't need to be fiddled with, so long as the fan achieved maximum RPM when the Mac Pro told it to the AHT/ASD tests would pass with flying colours.

-SC

 

[misanthrophy]

Sorry for opening again this old thread, but I am sure there are still people interested in optimizing the cooling of their MP's.

I do not understand anything from another forum, because I am not a technician, but someone said, changing the wires could resolve the issue with controlling the fans:

https://www.tonymacx86.com/threads/macpro-fan-pinout.122584/#post-917813

Could someone tell me if that makes sense and should work?

 

[siox23]

i ordered some of these pcb's:

http://www.icstation.com/voltage-co...cle-adjustable-conversion-module-p-12430.html

i think this could work in a 5.1 without the need to design a custom pcb.

according to these pinouts:

https://pinouts.ru/Motherboard/#Apple

the control voltage is 0-5V.

the board i ordered will convert 0-5V (or 0-10V -> jumper) into a 0-100% pwm signal and make a standard pwm 4 pin fan controlable by apple's logic including speed readings.

Also one should be able to provide the necessary supply voltage for the board from the motherboard fan connector.

will give u feedback after testing.