P5AD2 Premium

L

Leigh-Anne Mills

I have a P5AD2 Premium with a intel 3.6GHz on it, the procressor fans runs
always 2567rpm and when the processor is at 100Percent the temp is 158F,
should the fan run at a higher speed, when the processor runs hotter, if so
what am i doing wrong.

Thx

Leigh-Anne
 
P

Paul

Leigh-Anne Mills said:
I have a P5AD2 Premium with a intel 3.6GHz on it, the procressor fans runs
always 2567rpm and when the processor is at 100Percent the temp is 158F,
should the fan run at a higher speed, when the processor runs hotter, if so
what am i doing wrong.

Thx

Leigh-Anne
158F is 70C, and that is a magic temp for the processor. Your processor
is thermal throttling - in other words, it is not running at full speed,
and won't give a good benchmark in its current condition. The
temperature would have risen even higher, if the processor hadn't
reduced the percentage of cycles doing actual computing. (If throttling
is not enough to keep it cool, eventually the processor will turn off
the computer, so there is another level of control it hasn't used yet.)

Looking at the manual, you have one complicated piece of machinery.
The CPU fan header has four pins. Three of the pins are like any
old fan header - they consist of +12V, GND, tachometer. In the
good old days, you would vary the voltage on the +12V pin, to change
the fan speed. This could be done with a linear control (i.e. a
rheostat, fanbus, or similar names), or via PWM (pulse width
modulation). PWM is a square wave signal, where the width of the
pulse is varied. When a pulse is integrated (smoothed with a capacitor),
a voltage results, and as the pulse width varies, the fan speed would
vary because the filtered voltage is varying. In other words, a PWM
method sources +12V to the fan, and can vary the voltage to the fan.

So, those were the old methods for controlling the +12V pin.

Also, in the good old days, Intel had a thermistor inside the fan body.
Intel controlled the fan speed based on the air temperature passing
through the fan body. The idea here is, the fan provides "constant
cooling effort". If the air temperature rises, the fan has to move
more air, to maintain the same die temperature. I've never seen those
terms in print, so that is my best guess as to their misguided thinking.
In this case, fan speed is minimal at 30C, and maxed at 38C or higher
air temperature.

Now, back to the present day. Your new CPU and its headsink/fan, have a
fourth pin on the fan. It, too, is a PWM pin. The PWM feature works the
same way as it would have on an old fan. The motherboard can send a
square wave to the fan, and inside the fan, the square wave is used to
vary the fan voltage. It is just the smoothing and switching that is
being done inside the fan now, as the switching component can be kept
cool as it lives inside the fan.

The BIOS gets in on the fun at this point. The BIOS has three options.

If you select "Q-fan Disabled", then the PWM signal is turned off, the
fan gets full +12V power. The Intel fan, when not receiving a PWM
signal, falls back to using the thermistor. If the air temp is 30C
or less, the fan speed runs at min. If over 38C, the fan runs at max.
That is probably what is happening to you right now - Q-fan is off
and air temp is too low to make the fan run at full speed. Your
computer case must be very well ventilated.

When Q-fan is enabled, the BIOS has options for "DC" or "PWM"
control. The DC option, means the motherboard does something to vary
the +12V signal to the fan. The PWM signal, at this point, would be
turned off. At the fan end, the fan thinks it is dealing
with a "dumb" motherboard, so the fan is using its thermistor to control
the fan speed. The fan runs slow, if the air temp is below 30C. The fan
runs at full speed, if the air temp reaches 38C. The motherboard
measures the CPU die temperature, and if the die is over 50C, the fan
voltage will be ramped up. In this case, you are really at the mercy of
the air temperature again, as at 70C die temperature, the fan receives
the full +12V from the motherboard, but the thermistor still controls
fan speed as well. (Or stated another way, there are two temperature
control methods at work at the same time, and the Intel method
dominates the other method.)

If you select the "PWM" option in the BIOS, when Q-fan is enabled, then
the 12V signal gets the full 12V, and the PWM signal is turned on. As
in the "DC" control case, the motherboard measures the die temperature,
and the fan speed is controlled via the PWM signal. The difference in
this case, is the Intel fan stops using its thermistor, so if the
die hits 50C, the fan should ramp up, and be running faster than it
currently is.

Based on reading the manual, and the datasheet for an LGA775 retail
processor, I would say "enable Q-Fan, select PWM control" to fix
your problem.

I wish this had occurred to me before, because other people have
complained about high processor temperatures, and it never
occurred to me that the Asus defaults would be so stupid. I
suspect a non-default setting is best.

HTH,
Paul
 
L

Leigh-Anne Mills

Thx for the info Paul.

Maybe next time i also should read the manual (RTFM).

Just need to treak the Temp to let Q-Fan to be enabled, had it set to high,
look like i have it ok now.

My Mobo always stay at 95F, the only fans i have in it, are the intel stock
fan, a 120mm chassis fan, that came with the case, and a coolermaster 80mm
fan near the HDDs.

The case i have is this

http://www.cluboverclocker.com/reviews/cases/foxconn/diabolic/

One of the best i have ever owned.

Leigh-Anne
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top