| Author |
Message |
nobody
Joined: Mar 09, 2008 Posts: 1687 Location: Not here
|
Posted: Tue Jan 26, 2010 8:04 am Post subject:
How to fool a mobo Subject description: And make it believe the non-extant fans are extant |
 |
|
This would go into an equipment or DIY thread except it's not a musical equipment. The computer I'm looking to do this on is my Linux music workstation, though.
I figured someone here might at least have an idea, because we've got a lot of people working with circuit boards. I need to somehow fool my computer's motherboard into thinking that its own internal fans are present when they are not (trust me, I have an ample, custom external cooling system that works wonderfully). At the moment, I need to trick the PSU and/or the mobo into believing the PSU fans are present when I disconnect them. Without the PSU fans, the machine will not even bother to start and will put up an amber LED warning. It will NOT be a danger to run the computer without the PSU fans (I know what I'm doing). I just need the computer to believe the fans are there so I can disconnect the last two internal fans - the PSU fans - and have a silent audio workstation.
If you want more info on what I'm doing: http://blog.audiodef.com/2010/01/adventures-in-creative-computing-vol-1.html |
|
|
Back to top
|
|
 |
seraph
Editor


Joined: Jun 21, 2003 Posts: 12398 Location: Firenze, Italy
Audio files: 33
G2 patch files: 2
|
Posted: Tue Jan 26, 2010 9:17 am Post subject:
|
 |
|
I thought a mobo was some kind of monkey  _________________ homepage - blog - forum - youtube
| Quote: | | Don't die with your music still in you - Wayne Dyer |
|
|
|
Back to top
|
|
 |
DrJustice

Joined: Sep 13, 2004 Posts: 2112 Location: Morokulien
Audio files: 4
|
Posted: Tue Jan 26, 2010 11:02 am Post subject:
|
 |
|
I thought it might be about fooling a molbo For what it's worth, they're said to be easy to fool. But molbos don't really have a lot of fans...
How about replacing the fan with a resistor, if they're two-wire ones? AFAIK the three-wire variety are monitored for actual rotation and may not be so easy to fool. Don't some BIOS'es have options to ignore temperature and fan alarms?
DJ
-- |
|
|
Back to top
|
|
 |
nobody
Joined: Mar 09, 2008 Posts: 1687 Location: Not here
|
|
|
Back to top
|
|
 |
nobody
Joined: Mar 09, 2008 Posts: 1687 Location: Not here
|
Posted: Tue Jan 26, 2010 12:57 pm Post subject:
|
 |
|
| DrJustice wrote: | I thought it might be about fooling a molbo For what it's worth, they're said to be easy to fool. But molbos don't really have a lot of fans...
How about replacing the fan with a resistor, if they're two-wire ones? AFAIK the three-wire variety are monitored for actual rotation and may not be so easy to fool. Don't some BIOS'es have options to ignore temperature and fan alarms?
DJ
-- |
From THAT story, I think I could figure out how to fool a molbo on my own.
I like the resistor idea. I'm going to research that! I don't think the BIOS for this machine has the option of ignoring sensor warnings, which, when the machine is used as it is "supposed" to be used - as a high-load server - makes sense. But that sense-making is not in my favour! |
|
|
Back to top
|
|
 |
dmosc
Joined: Jun 23, 2003 Posts: 298
|
Posted: Fri Mar 05, 2010 3:27 pm Post subject:
|
 |
|
| Any fan monitoring is generally done through the middle wire, the yellow wire, which the motherboard will read a voltage from. It interpolates that voltage into an RPM. The actual appropriate RPM for a given fan varies greatly so generally they have settings in the motherboard. If your motherboard is unhappy with a 0RPM value no matter what you tell it, you can fake any given fan speed by supplying a small voltage at the pin in question. I suggest using the +3.3V line you already have from your power supply. I forget the conversion but it should be like 10k RPM's or somesuch. |
|
|
Back to top
|
|
 |
nobody
Joined: Mar 09, 2008 Posts: 1687 Location: Not here
|
Posted: Sat Mar 06, 2010 8:55 am Post subject:
|
 |
|
I got the main fans taken care of. Of six, one was left inside to tell the mobo that something is there. What I'm trying to do now is get the PSU fans out (since sufficient air flow is generated elsewhere). I tried them in series, but that was not acceptable to the PSU. I believe the PSU, when you push the power button, ramps its fans up to max speed so that it knows that capability is available, and in series, they won't reach max speed. So I will next try one of two things: either insert a 555 timer that tells the PSU what it wants to hear, or leave them in but insert a switch so that x seconds after boot, the PSU fans can simply be turned off.
Now that I think about it, I'm in favour of a 555 timer, since for all I know, the PSU continuously monitors its fans and will shut the system down if they aren't running at all. A 555 timer will always inform the PSU that an acceptable RPM exists. |
|
|
Back to top
|
|
 |
dmosc
Joined: Jun 23, 2003 Posts: 298
|
Posted: Thu Mar 11, 2010 11:07 pm Post subject:
|
 |
|
| It might help if you were more specific on your system components. I've never used a power supply that cared at all about fans. |
|
|
Back to top
|
|
 |
nobody
Joined: Mar 09, 2008 Posts: 1687 Location: Not here
|
Posted: Fri Mar 12, 2010 6:22 am Post subject:
|
 |
|
I don't know how to identify individual electronic components, but it's an HP ProLiant DL145 G2. I'm assuming that because this machine was factory-designed to be a high-load device, the parts were designed with fail-safes. While that's good for it's intended purpose, I'm hardly using it for that, so the fail-safes are just fails for me.  |
|
|
Back to top
|
|
 |
|