February 15, 201016 yr Anyone know if it would be possible to control disk fan speed based on actual S.M.A.R.T. temp of the disks and not based on mainboard sensors? In a depth Norco 4220 case all the mainboard sensors happen to be way in the back where everything runs much cooler than in the front of the case with all the disks loaded. I thought about putting some additional temp sensors in between the disks and making an independent speed control of the drive fans but it would be much easier to change fan speed control via software somewhere in ACPI. Any hints highly appreciated.
February 15, 201016 yr Yes, I do it..... but you have to write a script, and you have to have a mobo that supports PWM can control.
February 16, 201016 yr Author bubbaQ, thanks for your reply. I have a mobo with PWM control. Can you please share some more insights, maybe your script? Thanks much in advance!
April 13, 201016 yr BubbaQ, does that script effect all other fans in the case other than CPU and GPU? Are thos rpm's for some megablower, they seem quite low?
April 13, 201016 yr why dont you just buy fans with a thermal sensor you can stick to ya hdd, or hotswaps with temp sonsors
April 13, 201016 yr why dont you just buy fans with a thermal sensor you can stick to ya hdd, or hotswaps with temp sonsors Why do that when you can write a script that will do that for free?
April 13, 201016 yr Author why dont you just buy fans with a thermal sensor you can stick to ya hdd, or hotswaps with temp sonsors Because sensing the temp inside a drive is much more precise and allows for finer grain speed control. temp sensor is just on/off, using a script you have about 100 other possibilities for adjusting speed. No wires needed, no bulk sensors that are dumb devices and cost money. I would anytime prefer using 4-pin PWM control from *within* software than a dumb sensor.
April 13, 201016 yr Are thos rpm's for some megablower, they seem quite low? Those are not RPM's... they are PWM settings (possible values 0-255). at PWM-255 you get full speed. Less than 255, you get some fraction of full speed.
April 13, 201016 yr Author bubbaQ, if a PWM fan is spec'ed at 1000-2000RPM is it able to rotate slower than 1000RPM is this a hard limit and it won't be able to rotate say with 200RPM?
April 13, 201016 yr Author Curious to know if the X7SBE mainboard would be able to drive high CFM fans like the DELTA AFC1212DE-PWM as it draws 1.6A??
April 13, 201016 yr bubbaQ, if a PWM fan is spec'ed at 1000-2000RPM is it able to rotate slower than 1000RPM is this a hard limit and it won't be able to rotate say with 200RPM? It all depends on the fan. I've seen some fans able to run below with only 5V, and I've seen some that won't run below 9V.
April 13, 201016 yr Author bubbaQ, if a PWM fan is spec'ed at 1000-2000RPM is it able to rotate slower than 1000RPM is this a hard limit and it won't be able to rotate say with 200RPM? It all depends on the fan. I've seen some fans able to run below with only 5V, and I've seen some that won't run below 9V. I thought that PWM fans are constant 12V and digitally speed controlled via a 4th pin?!
April 14, 201016 yr I thought that PWM fans are constant 12V and digitally speed controlled via a 4th pin?! There are 2 types. (old) PWM of the 12V line on pin 2, and (new) constant 12V on pin 2 and a control signal on pin 4. The old way, the mobo does the PWM. WIth the 4-pin fans, the fan itself does the PWM.
July 19, 201015 yr http://lime-technology.com/forum/index.php?topic=2857.msg23752#msg23752 I have the following situation: PWM works on 2 headers - PWM1 (used already by CPU with 4pin, thus 12V and pwm on Pin4) and PWM2 - that is available, but uses PWM on Pin2 - Pin4 is unused. In general, it works fine to control a "classic" fan on PWM2 (a true 4-pin-pwm-fan doestn't work of course) with a script controlled by the HD temperature. My question is: I would like to control all my HD-fans with this script. They are currently connected to 7V - which is too low for a paritycheck and too high if the drives are spun down. Obviously I should not connect all 6 fans to a single header - so my question is: Is there a solution to drive 6 fans by a script via PWM2, assuming all fans are the same brand and I use the speedpin just from one of them? Thanks in advance, Guzzi Script
July 19, 201015 yr Nope. Not without building a custom power circuit that replicated the PWM. You can safely split 2 or even 3 fans on one header in many cases... perhaps you cam run 3 on PMM and 3 on 7v.
July 19, 201015 yr Author Curious to know if the X7SBE mainboard would be able to drive high CFM fans like the DELTA AFC1212DE-PWM as it draws 1.6A?? To answer my own question after trying it out, perhaps someone else may be interested: No problem with those fans on the X7SBE. The mobo drives three of them, each on its own fan header without any problem, while the BIOS controls their speed. It is not possible to control RPM from software though (unRAID or full Slackware Linux).
July 19, 201015 yr Author Nope. Not without building a custom power circuit that replicated the PWM. You can safely split 2 or even 3 fans on one header in many cases... perhaps you cam run 3 on PMM and 3 on 7v. He may want to check out the Arctic Cooling F8 or F12 PWM fans with PWM sharing technology. They allow connecting up to 5 fans to a single PWM connector, of course depending on the actual mobo, but check those fans out. They move a lot of air for their size and noise they develop.
July 19, 201015 yr I use them and they are great. But they are still drawing all their power from one mobo header, and can exceed the wattage ratting of the mobo if you are not careful.
July 19, 201015 yr Author bubbaQ, you are right of course. So, carefully check the fan header specs on the actual mobo.
July 19, 201015 yr Nope. Not without building a custom power circuit that replicated the PWM. You can safely split 2 or even 3 fans on one header in many cases... perhaps you cam run 3 on PMM and 3 on 7v. Thanks for the quick replay, BubbaQ. Hmmm, any hint on such a power circuit - maybe a "proven one"? I could imagine it would be easy to to that for "true 4-pin-fans" - just take the 12V (pins 1&2) from the powersupply directly and provide pwm signal (pin4) to all fans - the current then would be supplied by the powersupply instead of the fanheader ... but this doestn't work on my pwm2, that uses pwm on pin2 and thus has to provide the full load. Would it be possible to use that pin2 signal for pin4 signaling and the 12V from atx directly when using "true 4-pin-fans" - or are the signals different?
July 20, 201015 yr You may try poke around here to get some ideas: http://www.analog.com/library/analogdialogue/archives/38-02/fan_speed.pdf http://forum.allaboutcircuits.com/showthread.php?t=32565
July 21, 201015 yr You may try poke around here to get some ideas: http://www.analog.com/library/analogdialogue/archives/38-02/fan_speed.pdf http://forum.allaboutcircuits.com/showthread.php?t=32565 Thanks, did some reading, but I think I will not start building my own circuits - any suggestion for something available to buy? The optimum would be a fancontroller, that can be controlled via the script and has enough power to drive 6 fans in parallel. RPMs could be taken from the MoBo, but it should somehow be accessible via .../hostx/pwmy, right?
July 21, 201015 yr Sure, but you better brush up on your programming skills... all require custom programming, and not just a script to write to a pseudofile. Matrix Orbital MX2 Crystal Fontz CFA633 T-Balancer? bigNG MultiFunction Fan / Pump Controller
Archived
This topic is now archived and is closed to further replies.