|
|
Post by macguy4u on Mar 21, 2010 14:08:01 GMT
Hey all
I know I have been posting alot but Im almost done with my mod -everything fired up last night except for the original MAC Pro case fan and the USB (which I dont have connected yet)
The fans connect perfect on my ASUS Rampage board but dont work-
How can I get these to work because I have 3 brand new MAC case fans that I would like to use.
Matt
|
|
|
|
Post by pirloui on Mar 21, 2010 14:12:19 GMT
"Mac"
The fans are not PWM fans, they use 5 and 12V and I'm not sure how the speed is modulated, but you will find thread about it. It's easier and probably more silent to use a regular 120mm fan.
|
|
|
|
Post by macguy4u on Mar 21, 2010 14:24:08 GMT
HEy Pirloui
Thanks --Thats what I was kinda thinking to do as well -less hassle if I just put new fans in there
Matt
|
|
|
|
Post by blackknight on Mar 21, 2010 16:49:05 GMT
Pirloui is correct, the fans are not PWM. The way that they are varied is adjusting the 5V wire attached to where the PWM signal is usually sent to a normal fan. Basically they are adjusted using analog voltage and not digital PWM. If you send 2.5V to that wire, you'll get about 1000RPM (5V is about 1500RPM), 3.3V is about 800RPM. I have found several ways to adjust their speed, but not using the temp sensors on the motherboard. They are nice fans though as they are quiet and move lots of air. I made a circuit that has a variable resistor on it so I can manually adjust the speeds, but I don't have to change it much.
|
|
|
|
Post by jedric on Mar 22, 2010 20:22:24 GMT
|
|
|
|
Post by blackknight on Mar 23, 2010 21:56:27 GMT
Jedric, you're right, it is good reading, but unfortunately I don't know if it applies to this problem. The issue is that the PWM controller on motherboards is unable to control the fans Apple uses. They are both 4 wire fans, but it's possible that normal PWM fans are low frequency and Apple's fans are high. I am just guessing here though.
|
|
|
|
Post by jedric on Mar 24, 2010 15:04:48 GMT
|
|
|
|
Post by blackknight on Mar 24, 2010 18:15:29 GMT
The problem is the fans, not the motherboard. From what I read about the Mac Pro motherboard, the 4th pin (should be PWM) is an analog 0-5V signal from the temperature sensors used for HDD and something else. These sensors are the ones mounted behind the HDDs next to the SATA connectors. The fans that Apple used were made specifically for Apple and nothing else, so this doesn't surprise me. I'm trying to figure out a way to reduce the voltage from a standard motherboard header to 0-5V instead of 0-12V so that our motherboards can still control them.
My idea is to put a rheostat inline on the 12v pin of a motherboard header and plug the other end of the wire into the 4th pin of the fan. I would use the rheostat to reduce the maximum 12V down to 5V before connecting it to the fan though as 12V will fry the fan (Shanky I think already did this). The main power and ground pins of the fan would be fed power from the PSU and not the motherboard as 0.75A/9W is too much for it. According to the specs on these fans, they are good for 130 CFM at their full speed of 3300 RPM and 44 dB.
|
|
|
|
Post by blackknight on Mar 25, 2010 18:32:06 GMT
Hey, I just realized that I may have a solution. I was reading an article about a new Noctua CPU cooler and remembered something about Noctua fans... they have a U.L.N.A. adapter that reduces the voltage to the fan. I looked up some other articles on Noctua fans to see how much voltage is sent to the fan with the ULNA adapter and it looks to be about 5V. I just ordered another Noctua fan (I lost the adapters for my other two fans) to get another ULNA adapter to see if I can use it on the 12V output of the motherboard header. If it works, it should reduce the 12V down to 5V and when the motherboard slows the fan down it should reduce to less that 5V. If this works, I believe I can run a wire to all the apple fans' control pins to vary their speed. I'll update once I find out (should be tomorrow).
|
|
|
|
Post by jedric on Mar 26, 2010 15:54:30 GMT
What probably would be of interest would be whats on the ULNA cable that help reduce the voltage. It would be expesive to buy the fans just for the ULNA cables.
If you can share the pinouts for the Apple fans as well that would be appreciated.
|
|
|
|
Post by blackknight on Mar 26, 2010 18:18:21 GMT
Ok, I did some experimenting today. I got the new fan because I needed one anyways, but now I have some things to play with. So, first of all, I don't know how Shanky fried his Apple fan (posted in the Hack Pro Mod thread) unless he put 12V to both outer pins of the connector. This is wrong as the pinout looking at it from the end of the connector looks like this:
|______|
| 1 2 3 4 |
where the top spikes are the key tabs on the connector. Pin 1 is ground, pin 2 is 12V power, pin 3 is the Tach sens., and pin 4 is the control pin. I supplied 12V to pin 2 and slowly increased the voltage to pin 4 up to 12V without frying the fan at all. This means that if you make a connector that is exactly the same as a 3 pin fan (a normal 3 pin to 4 pin molex adapter) and connect it to the PSU molex connector (these fans draw a lot of current so don't power it from the motherboard) and run two more wires to the tach and voltage wire from the motherboard to the tach and control pins of the fan, you would have control.
That being said, these fans run really fast at a full 12V and even down to 5-6V and are very loud. That's where the ULNA adapter comes in. These are simply just a 3 pin extension with a resistor inline on the power wire. There are two adapters that came with the fan and one is 150 ohm, the other is 50 ohm. These will reduce the 12V output of the motherboard header to about 5-6V or 3-4V depending on the adapter. This is a much more reasonable speed.
|
|
|
|
Post by blackknight on Mar 26, 2010 18:46:40 GMT
Ok, so I found another problem. The resistor that comes with the Noctua fan is just not capable of handling the wattage of more than one of these fans (I used the PSU fan to test with first, now I'm using the monster Delta fans. I need to check my current figures and probably use a higher wattage resistor to handle the heat. These mini ones that Noctua included are designed for a fan that uses 1.20W. I'll do some more experimentation and let you know how it goes.
|
|
|
|
Post by blackknight on Mar 26, 2010 19:13:00 GMT
I'm starting to think that these fans are just too powerful to try to control. They just draw too much current. I honestly am just going to give up on them I think.
*Edit*
Unless they draw all of their current on what I've thought was the control pin. I wonder if it's the other power pin...
|
|
|
|
Post by blackknight on Mar 28, 2010 15:59:09 GMT
|
|
|
|
Post by aquamac on Mar 28, 2010 16:23:45 GMT
Hi Blacknight,
I came to the same conclusion, though it's pretty frustrating as these fans look great and are powerful too. I am still using them on one mod that I run at work with a +5v input as opposed to 12v.
|
|
