Does 4-pin cpu fan connector imply PWM ?

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View
It's me again ...

Motherboard = ECS model RC410/800-M, LGA755 socket
CPU = Intel Pentium-D, 805, 2.66ghz

Both of these items come supplied with a 4-pin cpu-fan connector, which
to my understanding, implies there is PWM fan-control functionality.

So far I see nothing that indicates any kind of fan control. The CPU
fan spins at ~3000 rpm and the temp stays at ~30(C) both during rest
and during normal work load (surfing the web, working with Excel or
Word and other everyday tasks). Should I not see a change in rpm's and
temp at some point in time to indicate fan control is indeed working?
Especially at rest when cool ... shouldn't the fan be running very
slowly or completely off?

Or ... isn't that how PWM fan control works?

I thought about kicking off one of my a 3-d rendering engines as a
brute test and attempt to drive up the rpm's and temp but I'm fearful
of stressing a system that may not be running correctly.

Any help is appreciated.


Re: Does 4-pin cpu fan connector imply PWM ?

Quoted text here. Click to load it

Get a copy of Speedfan and see what it reads.

It could be that the labels for motherboard temperature
and CPU temperature are mixed up.

Speedfan has the ability to adjust the fan speed
controller, if one is present. What Speedfan cannot
tell, is whether the fan speed controller is actually
connected to the fan header. So you have to observe the
results for yourself, to be able to tell whether
Speedfan is working or not.


Re: Does 4-pin cpu fan connector imply PWM ?


Quoted text here. Click to load it

Not necessarily, it only implies they followed Intel's spec
on the kind of fan header to use, it can instead be only the
power and RPM with PWM function unused.

Quoted text here. Click to load it

Well if the CPU temp remains constant, why would the RPM
change?  That would imply correct function of PWM, that RPM
changes when temp does.

You need to put a reasonable load on the system, as basic
tasks don't do that.  Run Prime95's Torture Test, large
in-place FFTs setting for at least 30 minutes (though temp
should begin rising, if it will, almost immediately, else
RPM changes to control it but perhaps not linearly enough so
it should still be noticable.

Quoted text here. Click to load it

No need to be fearful when you have the temp feedback you
can monitor.  3000 RPM isn't exactly slow either, not as if
your fan had stopped, personally I wouldn't want a system
that had to run the fan at > 3000 RPM to keep the CPU cool.
As Paul noted you might have the fan plugged into wrong
header, or in worst case the board's PWM feature is
defective or a bios bug has interfered with it's correct
function... it IS an ECS motherboard, people don't always
pay more for better boards just for the heck of it.

Re: Does 4-pin cpu fan connector imply PWM ?

Besides the other suggestions you've received, I'll offer the following

1)  Check your BIOS settings.  My Asus board has BIOS settings for managing
(or not) the CPU fan.

2)  With my 830D processor, the fan typically runs right around 2000 rpm
when I'm not doing anything fun, and a CPU temp of about 40C.  Under heavy
load, the MB reports a speed of about 2600 rpm, with temps up to 55C.  I've
never seen a speed as high as 3000 rpm, and I've never seen a temp as low as

As someone else said, if you're monitoring the temp, running a "stress test"
shouldn't cause you great concern.  I probably wouldn't kick it off just
before you go to bed and let it run though. :)  I don't think you're going
to see the fan go any faster though, as it would start to lift off the
processor.  FWIW, I can quite clearly hear the difference between the fan
running at 2000 rpm and 2600 rpm.  The former being much more conducive to a
"quiet" office environment, and the latter being barely acceptable when
playing a game, or encoding video files.


Quoted text here. Click to load it

Re: Does 4-pin cpu fan connector imply PWM ?

You can probably tell this is my first walk on the wild side of fan
control and PWM so I greatly appreciate all the replies.

Clint's post is interesting in that his processor appears to also be a
Pentium-D (same as mine, albeit somewhat faster -- 3.0 ghz Vs. my 2.66
ghz) so it's apples to apples and going by what he posted, I'm inclined
to believe either my motherboard does not honor PWM fan control or I
don't have a setting correct in the bios, which BTW, does not show any
obvious setting or toggle for fan control -- if it's there ... it's
well disguised. Looks like it's time to make a call to the MB vendor
and ask a few questions.

Agsain, thank you for the help.

Jorge /

Clint wrote:
Quoted text here. Click to load it

Site Timeline