Why do fan RPM monitoring?

Control: management of fans, temp/rpm monitoring via soft/hardware

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
rtsai
Posts: 261
Joined: Sat Jun 19, 2004 6:49 am
Location: Boston, MA

Why do fan RPM monitoring?

Post by rtsai » Sun Feb 27, 2005 8:46 am

If you are already doing temperature monitoring of components (CPU, MB, GPU, HDD), is it that important to do fan RPM monitoring at all?

It seems to me that it is the temperatures that are important, and not fan speed. I can understand why one would want to *control* fan RPMs (because you thereby indirectly control temperature), but I don't really understand why *monitoring* fan RPMs is important if you are already doing temperature monitoring. Any fan failures should show up as increased temperatures, right?

Do thermistors break down or become de-calibrated over time, so that fan RPM monitoring kind of provides some redundant measure of cooling (e.g., RPM > 0 means the fan hasn't failed)? Am I missing something really obvious?

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Sun Feb 27, 2005 9:11 am

Of course it's not really necessary......but it does give you an immediate indication if something is wrong. I have my systems setup with a fan control that varies rpm with temp changes. So if I see a temp rise without an RPM increase, I know something is going on that requires action.

Whether you monitor fan RPMs is up to your own personal choice....I do it for peace of mind, and because it requires no effort.

joshd2012
Posts: 92
Joined: Thu Feb 24, 2005 1:24 pm
Location: USA

Post by joshd2012 » Sun Feb 27, 2005 2:15 pm

Temperature is not something you can control, it is something that you control by changing fan speed. Its a very easy way to adapt to different system stresses. You can set a fan speed for idle, for gaming, and for hard processing, and just dial in those RPMs before you begin your task to ensure that you don't melt your cpu.

Sure, you can see that your temperature is rising while you are doing some task, and then increase your fan speed to compensate. But wouldn't it be easier to just dial up your fan speed to compensate for possible temperature rise without having to check temperature every 5 minutes to see what your speed changes have done?

Really just a personal preference, but it does come in handy. Its not just for show if that is what you are asking.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Sun Feb 27, 2005 2:48 pm

Manual fan-speed control is one of many ways to handle temperatures. But it's crude, requires user input every time you do something intense, and can easily cause problems if you forget one time to tweak the speed.

I usually setup my cooling to work on automatic, using temp controlled fans, software like speedfan, or some type of automatic fan controller, with maybe some extra fans that turn on at high temps. It's more difficult to setup......but once running properly, requires little monitoring.

rtsai
Posts: 261
Joined: Sat Jun 19, 2004 6:49 am
Location: Boston, MA

Post by rtsai » Sun Feb 27, 2005 6:21 pm

The original motivation for my question was because my Yate Loon fans have both 3-pin and 4-pin connectors. I thought that by undervolting the Molex connectors I'd have to forgo the 3-pin connector, so I wanted to know exactly what I'd be missing without RPM monitoring.

Then I realized (from elsewhere in this forum - thanks!) that I can just take the power cables out of the 3-pin connector still get my RPM monitoring.

Post Reply