Q-fan settings and buying a variable speed detectable and variable fan: WTF

I just bought an Asus P4C800-E deluxe, but don't have a cpu or ram yet. So:

I need to buy a fan with 3-wire; I've got a couple 2-wire already.

A 2-wire type fan has power supplied only, and the power is connected not to
the mobo, but via a molex connector to the wires that come directly from
the power supply, on the same power chain that powers the CD/HD/Floppy. If
you have a connector for CHA_FAN on the mobo it is probably a 3-wire
type,which allows mionitoring of the fans speed as well as supplying the
power, all on the same 3-pin plug, at the mobo.

I am going to try to get to the bottom of the fan speed monitoring problem
with respect to variable speed fan type options. They are cheap. For
instance,
Vantec has an 80mm that has the following specs: 23.8-41.6 cfm,
25.5-37.5dBA, 1950-3400 RPM.

These are the things you have to look for
because Asus has alluded to Q-fan as " The Asus Q-fan technology smartly
adjusts the fan speeds according to the system loading to ensure quiet, cool
and efficient operation".

I don't know if this is just alarm/ set it yourself, etc...
Probably hocus pocus. There is nothing on the web site I can find. I could
read it a thousand times and not know the first thing about what they're
talking about. Anybody got any info about this?

Are all fans adjustable? i.e. just change the power supplied and that
changes the speed. Are the resulting speeds always determinable? Do you
need a special variable adjustable, and variable monitorable fan type? In
theory I assume that at least all of the fans I will discuss here are in
fact monitorable using pcProbe, or whatever it is called now, whether in
bios, or the os, or both.

If the problem arises, how does a layman determine the correct heat setting
to use, e.g. Intel site? Anybody got any pointers. e.g H65C, M55C, L35C.
How do you get the specific information to apply to your specific processor.
Especially 2.4C, 2.6C, or 2.8C(don't know yet): are all similarly oc'able?


This could mean many things: when the computer shuts down, when the alarm
goes off, when the fan comes on, when the fan goes off, the minimum
allowable fan setting, the maximum fan speed setting, or any countless
combination... I can think of a similar number of good reasons for having
this technology.

I think the ideal possiblity is, since the mobo has cpu and mobo temp sensor
settings, and obviously (some?)fan speeds are adjustable, that you use the
feature settings this is designed for to set the max allowable temp that you
want to allow, before (i) the fan speeds up, or turns on, (ii) the settings
causing the excess heat are changed, (iii) a warning is given, (iv) the
computer shuts down... You see I'm back to the beginning again.

The bios options under Power for Q-Fan are enable/disable
When enabled select Fan Spped Ratio; any one of: 11/16 12/16 13/16 14/16
15/16. 11/16 is the minimum fan speed ratio. Select a higher ratio if you
installed additional devices and the system requires more ventilation.

Can someone who knows please clear this up


BTW, the POWMAX LP-8800C 350W ATX psu I have says on the box "supports fan
speed monitoring". Other than the connections for the disk drives, it only
has a 4-pin and 20-pin connector, which I think would be standard for this
type anyways. Will I get a power supply fan speed with this? Does the
signal come through the 20- or 4- pin connection. Alternatively, there is
also a 3-pin PWR_FAN, as well as the CHA_FAN and CPU_FAN. The CHS and PWR
fans obvoiously get the job done of determining their fan speeds through
these 3-pin connectors, which also supply the power to the fans. Is it
possible that this would connect into the psu's fan. If this is the case,
could I have this option, and what would I have to do to enable it. What
were they talking about when they said "supports fan speed monitoring"?

kgs wrote:

I just bought an Asus P4C800-E deluxe, but don't have a cpu or ram yet. So:

I need to buy a fan with 3-wire; I've got a couple 2-wire already.


There are very few problems with properly mounted fans, regardless of
their location. In all of our recent installations, the heatsink/fan
that is supplied with the retail Intel P4 CPU's have done quite well,
although a little noisy, depending on the computer case. These Intel
fans are 3-wire, the third wire is the tachometer lead, but also have
their own thermal control. Consequently, the Q-Fan setting is disabled
with these fans. It is Winter in Southern California but the CPU's run
between 25 to 35 deg. C. under normal load at around 2,900 RPM.

We have also played with the ThermalTake P4 Xaser heatsink/fans under
normal speed and variable speed control. Normal speed is high and loud
but same temperatures are achieved at around 3,100 RPM.

Bottom line is that Q-Fan is effective mainly for high-RPM HSF's that
do not have its own thermal control...just to minimize the noise as
the HSF's tend to be effective in a well-ventilated case. A good CPU
temp for a P4 CPU can range from 25 to 50 deg C., depending on the
load conditions. And optimum fan speed for HSF's with manual controllers
is whatever obtains minimal fan noise at acceptable operating temp.

As for the other motherboard fan leads, there is no speed control for
fans connected to PWR or CHA. The third pin is for the tachometer.
And the PSU fan is normally independent of the motherboard.

In article , Mistoffolees
wrote:

kgs wrote:

I just bought an Asus P4C800-E deluxe, but don't have a cpu or ram yet. So:

I need to buy a fan with 3-wire; I've got a couple 2-wire already.


There are very few problems with properly mounted fans, regardless of
their location. In all of our recent installations, the heatsink/fan
that is supplied with the retail Intel P4 CPU's have done quite well,
although a little noisy, depending on the computer case. These Intel
fans are 3-wire, the third wire is the tachometer lead, but also have
their own thermal control. Consequently, the Q-Fan setting is disabled
with these fans. It is Winter in Southern California but the CPU's run
between 25 to 35 deg. C. under normal load at around 2,900 RPM.

We have also played with the ThermalTake P4 Xaser heatsink/fans under
normal speed and variable speed control. Normal speed is high and loud
but same temperatures are achieved at around 3,100 RPM.

Bottom line is that Q-Fan is effective mainly for high-RPM HSF's that
do not have its own thermal control...just to minimize the noise as
the HSF's tend to be effective in a well-ventilated case. A good CPU
temp for a P4 CPU can range from 25 to 50 deg C., depending on the
load conditions. And optimum fan speed for HSF's with manual controllers
is whatever obtains minimal fan noise at acceptable operating temp.

As for the other motherboard fan leads, there is no speed control for
fans connected to PWR or CHA. The third pin is for the tachometer.
And the PSU fan is normally independent of the motherboard.

It is hard to tell from the photo of the motherboard in the manual,
but the board could be using a Winbond W83627THF SuperI/O chip, which
includes a hardware monitor. The datasheet I have was downloaded from:

http://www.winbond-usa.com/products/...CIC/627thf.pdf

On page 115, it says this chip only has PWM fan control outputs for
up to two fans, so three motherboard fans cannot be controlled
independently. Also, this chip is a "dumb" fan controller, in that
software has to update the "duty cycle" setting every time the fan speed
needs to be adjusted. Maybe this is done by the BIOS code - I have
no way of knowing that for sure. It is also difficult to guess at
what the control algorithm would be in the BIOS. The BIOS has the
ability to monitor some temperatures, and based on this the BIOS could
adjust the duty cycle - the BIOS could even monitor the fan RPMs recorded
by the Winbond chip, and adjust the duty cycle to achieve a fixed fan
speed at a given motherboard temperature. To know more, I would have to
connect a meter to the fan voltage and watch what happens when the processor
heats up.

Note that when using the retail Intel HSF provided with your processor,
the Intel fan also has a temperature control feature. There is a thermistor
on the Intel HSF that measures air temperature. The thermistor is monitored
by a control circuit inside the Intel fan, and the voltage to the fan
is adjusted based on the ambient temp in the case.

With the two control features enabled, sometimes you can get extreme swings
in fan speed, such as the inability to get a good maximum fan speed when
the processor is running hot. To fix this, you could either buy a CPU fan
that doesn't monitor the air temperature, you could short out the thermistor
on the Intel fan to get it to run at higher speed, or you could disable
Q-fan and rely on the Intel HSF thermistor to make the necessary adjustment.

Because a fan without the tachometer output is a few cents cheaper to make,
the market is flooded with two wire fans. This is unfortunate, as if all the
fans had tacho output, you wouldn't have to do a lot of searching to find
a good fan with RPM monitor.

The original posters questions:
Are all fans adjustable? i.e. just change the power supplied and that
changes the speed. Are the resulting speeds always determinable? Do you
need a special variable adjustable, and variable monitorable fan type?

Fan speed changes in proportion to voltage. Many fans are guaranteed to
spin between 7 volts and 12 volts supplied. Since the fan speed changes
in relation to the resistance to air flow, the speed is not regulated and
there can be large variations from unit to unit. To regulate the fan speed,
there are regulator chips (Maxim makes one) that watch the tachometer
output and adjust the voltage to achieve a constant speed - I've only
seen one product that uses a chip like that. Fans don't list this voltage
versus speed thing as a feature as such, but motherboards or other devices
vary the voltage to the fan, if a noise control feature is desired. As
the original poster noted, not all fans have the tacho output, so neither
hardware nor software closed loop control of speed is possible with a dumb
two pin fan.

Fan voltage can be adjusted via a variable resistor (many drive bay
controllers do that) or a variable voltage can be achieved by using
pulse width modulation (PWM) and a capacitor as an integrator. The PWM
method is efficient and doesn't heat up the transistor used to switch
the current appreciably - this is why PWM is used on a motherboard, to
avoid yet another source of heat. Since the transistor is small, there
is usually a limit to the current that can be controlled via PWM -
exceeding the current limit burns out the transistor and leaves you with
a dead fan header.

HTH,
Paul

"Paul" wrote in message
...
It is hard to tell from the photo of the motherboard in the manual,
but the board could be using a Winbond W83627THF SuperI/O chip, which
includes a hardware monitor. The datasheet I have was downloaded from:

http://www.winbond-usa.com/products/...CIC/627thf.pdf

Just as a footnote, i own this board, and CPU-Z confirms that the Winbond is
in fact a W83627THF.

"As for the other motherboard fan leads, there is no speed control for
fans connected to PWR or CHA. The third pin is for the tachometer.
And the PSU fan is normally independent of the motherboard."

On the p4c800-e dlx there are 3-pin power & monitoring connectors for
cpu/cha & pwr fans.

There is also a thermal thermal sensor cable to conect to the psu.

The key word here I guess is CONTROL. I have yet to discover if there is a
control for the cpu fan. Is this in fact true, and is this adjusted by the
asus BIOS and/or software?

My psu box POWMAX LP-8800C says "supports fan speed monitoring" ! Does this
mean it allows others to achieve this?

So for case fan options, now that I am about to buy, and everything is
riding on this as of now, to connect to the mobo CHA_FAN all I need is a
3-wire fan and that will allow tach monitoring, but no control, which is not
not an option w/ asus bios, or o/ s/w with anything other than the CPU_FAN
for the purpose o running a cool and quiet pc. All I need to find out is
what bios, or o/ s/w does exactly. These fans are pretty cheap even in red.

I'll need to figure out what this means:
I want to know what monitoring
is done by asus, and what the result is when selecting a different fan.
Fans have differnt cfms/rpms/dBAs/VAs, and POSSIBLY the ability to monitor
and change these values. One fan will not behave the same as another. p2-27
of the p2c800-e deluxe manual says the "3-pin CPU_FAN, PWR_FAN, CHA_FAN fan
connectors support cooling fans of 350mA~740mA (8.88Wmax.) or a total of
1A~2.22A (26.64W max.) at +12V". I don't know if is the variable
output(controllable range), or if any value in between is the max fan rating
to connect, and if it makes a diffrence if the fan types allows supports
monitoring and adjustment, or if all 3-wire fans do all of the above. What
about adding other built-in fan functionality inline with the 3-pin at the
mobo. I want cool, but I don't neccesarily want to bypass an already
existing technology, especially since I am ignorant as to the cause and
effects, and needs. What about attaching a reostat & tach inline on the
CHA_FAN jack on the mobo.

Can you split the fan load at the CHA-FAN connector, (and monitor) for (i)
output fan @ rear top, and (ii) input fan @ front (don't get me started
about HD cooling in the process).

I'm just upgrading to this mobo and soon will get el4200 ram, and a 2.4C-~3C
processor, and will probably want to see what it can do, in comparison to
published results. Don't know what heat to expect.

Theoretically you could get a seperate temperature sensor to connect b/t the
mobo and a placement in the hottest part of the psu (the highest number is
probably the one you want). This value may enter the ASUS equation.

But in order to respond to the mobo/cpu/psu temp readings by controlling the
CHA, PWR, and OTHER.... fans, (apparently the cpu fans are monitored and
adjusted based on the equation), you need to disconnect the fans from the
the mobo. If you disconnect them then you have no assurance then are
running, and you don't know their speed. If they were plugged into the
mobo, and the speed monitoring option was on, an alarm beeps, and flashes,
and you see speed 0, where normally you would be ok.

Unless you can control the speed inline (how about one of the many
faceplates). Can you then still have your speed read monitored by
bios/software, and have your alarm if threshold rotaion monitored this way?

At this point I'm going to need 476 componentsm, in place of an answer.

These things come with temp sensors for each of, say, 4 fan speed control
knobs. Sometimes the temp sensor is in the fan itself. Don't know if it
must be for this to work. Could you get a unit with individual temp sensors
that are not part of the fan, and could I use one of these to connect to
psu. This would't be advantageous in the equation, but at least I'd know
the temp. And I could get a bunch of other fans and put them all over the
pc. My psu has over temp and over power safeguards. Good thing b/c I
wouldn't know how hot to set my sensor before it e-mails the CIA.

BTW, if using a dual channel kit of EL4200 RAM with an ASUS P4C800-E
deluxe, and using the (not add~$300.00) 2.4C/2.6C/2.8C/3.0C cpu, for example
@1:1 @~280/560/1160FSB @3-3-4-7 @2.85V (w/ appropraite voltages and timings
supplied by anandtech.com) considered the best, as many sites say it is, is
this not considered overclocking, if it is indeed the best (don't try to get
better), and does that not indicate that it is going to get hot, to the
point that I may be prepared to need extra cooling.

Another site with similar results w/ma obo and ram, which may be better for
me, for cooling, for a change are at ...

fyi, I needed to find out about these fan thingies anyways.

winbond.com, no legitreviews.com

I have read all the reviews, and I'm gonna copy-cat as much as possible,
because there is a known means to an end. I will be preparing to tackle
any cooling issues if req'd at any time. I have 2 questions: (1)How much do
I
need for my specific purpose, and if you can (2), should I get the
specific processor.

I am upgrading because here in Toronto, using Rogers hi-speed cable
internet, I have recently discovered the newsgroups. These contain upwards
of 10, 000, 30,000, 100,000 messages in each group. I get a six pack
dispered throughout about eleven groups each day. I mark for download, and
then finally
combine and decode them, currently using OE. My old 192 MB, cusl2-c w/
eleron600(I know ) just crashes, hangs, waits, etc.. to excess.

Q1

How much EL4200 does a serious homegrown OE nntp user need: (2x256), or
(2x512)? I don't have the chance to play around. I can only do things once
at a time.

I am upgrading to an ASUS P4C800E-deluxe. I plan on also getting the 2.6C
box cpu, because it seems like a sweet spot, and is not the baby in the C
series. I assume the 2.4, 2.6, 2.8, and 3.0 behave the same. From reviews

I assume the 3.2 behaves somewhat differently, maybe more ocable. It is
also

lot more money. If recommended I could also get the 2.4C(~$???),
2.8C(add~$50), or 3.0C(unlikely)(add ~$150). I just tried to fill in this
2.4C(~$???) value above from a www site to get my facts all in a row, and
found the 2.4 and 2.6 similarly priced. I have read a couple times in
reviews that the 2.4C has been used and they said it was used because it
was very overclockable. I assumed this would be true similarly about any of
the C b/t 2.4C-3.0C. I also assumed the articles I was reading may be older,
and that the 2.4C was probably the sweet spot when the article was written,
and they would have achieved similar(and therefore better) with a 2.6C.

Q2

Any comments on the 2.4C vs 2.6C question.



don't worry, the best is yet to come, wait for ALL-IN Wonder post.

I am forbidden past http://www.winbond-usa.com/products/, and will proceed
from there.

It will take me a while to digest this.

Have you aswered my question?

I'll be back.