P5GD1-VM fan speed problems

[Guest]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

I'm hoping somebody can help me with this problem because I think I've
tried everything.

The problem is that when I plug the CPU fan into the CPU_FAN1 socket on
the motherboard, the fan runs at a very high speed and makes a ton of
noise.

Here are the specs for the machine:
-Antec Sonata case w/ 380W PS
-ASUS P5GD1-VM motherboard
-P4 3.2Ghz CPU
-1GB (2X512MB) PC3200 RAM
-LiteOn DVD-ROM
-NEC 16X DVD +/- RW
-Seagate 120GB SATA
-Seagate 200GB SATA

I'm currently using the Cooler Master Hyper48 Silent fan. This fan is
supposed to be silent, but when plugged into the motherboard, it sounds
like a vacuum cleaner. It is anything but silent. When I check the
bios, it reports the fan running at 3400+ RPM. The fan is supposed to
be a 1400RPM fan. It also reports the CPU temp at 42C, which should be
low enough to not cause the fan to be throttled up so high.

If I use one of those adapters that allows you to plug the fan directly
into one of the power leads, then it runs at the advertised 1400RPM.
The problem with this, is that I lose the fan monitoring/throttling
from the mobo, and I also get the annoying beep warning when starting
the computer up.

Here is what I've tried (without success) to fix this problem:
-Swapped out the motherboard for another P5GD1-VM
-Tried a different fan. A Thermaltake. Same fan speed in BIOS
-Tried a different CPU.
-Tried enabling/disabling Q-Fan in the BIOS
-Flashed the bios to the latest version: 1007
-Made sure the FANPWR1 jumper was set to PWM (also tried DC just for
kicks)
-Verified that I am plugged into the CPU_FAN1 socket NOT the CHA_FAN1
socket
-Appropriately applied thermal grease to the CPU (not too much/not too
little)

Any help/advice anyone can supply would be greatly appreciated. As I
said, it runs at a normal speed when plugged into a power lead via an
adapter, but that is a work-around, and not a solution that I want to
use long-term. The CPU runs at 48-49C when plugged in with that lead,
but I really want to have the mobo monitoring that fan.

Thanks,

Indy

 

[Paul]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

In article <1103415817.243060.209160@c13g2000cwb.googlegroups.com>, "Indy"
<cmmcorp@yahoo.com> wrote:

> I'm hoping somebody can help me with this problem because I think I've
> tried everything.
>
> The problem is that when I plug the CPU fan into the CPU_FAN1 socket on
> the motherboard, the fan runs at a very high speed and makes a ton of
> noise.
>
> Here are the specs for the machine:
> -Antec Sonata case w/ 380W PS
> -ASUS P5GD1-VM motherboard
> -P4 3.2Ghz CPU
> -1GB (2X512MB) PC3200 RAM
> -LiteOn DVD-ROM
> -NEC 16X DVD +/- RW
> -Seagate 120GB SATA
> -Seagate 200GB SATA
>
> I'm currently using the Cooler Master Hyper48 Silent fan. This fan is
> supposed to be silent, but when plugged into the motherboard, it sounds
> like a vacuum cleaner. It is anything but silent. When I check the
> bios, it reports the fan running at 3400+ RPM. The fan is supposed to
> be a 1400RPM fan. It also reports the CPU temp at 42C, which should be
> low enough to not cause the fan to be throttled up so high.
>
> If I use one of those adapters that allows you to plug the fan directly
> into one of the power leads, then it runs at the advertised 1400RPM.
> The problem with this, is that I lose the fan monitoring/throttling
> from the mobo, and I also get the annoying beep warning when starting
> the computer up.
>
> Here is what I've tried (without success) to fix this problem:
> -Swapped out the motherboard for another P5GD1-VM
> -Tried a different fan. A Thermaltake. Same fan speed in BIOS
> -Tried a different CPU.
> -Tried enabling/disabling Q-Fan in the BIOS
> -Flashed the bios to the latest version: 1007
> -Made sure the FANPWR1 jumper was set to PWM (also tried DC just for
> kicks)
> -Verified that I am plugged into the CPU_FAN1 socket NOT the CHA_FAN1
> socket
> -Appropriately applied thermal grease to the CPU (not too much/not too
> little)
>
> Any help/advice anyone can supply would be greatly appreciated. As I
> said, it runs at a normal speed when plugged into a power lead via an
> adapter, but that is a work-around, and not a solution that I want to
> use long-term. The CPU runs at 48-49C when plugged in with that lead,
> but I really want to have the mobo monitoring that fan.
>
> Thanks,
>
> Indy

Try as I might, I cannot come up with a hypothesis that covers
all the symptoms.

Hyper48 review:
http://www.insanetek.com/index.php?page=cmhyper48

Closeup of Delta fan used:
http://www.insanetek.com/reviews/cooling/Cooler%20Master%20Hyper%2048/images/Picture%20041.jpg

The fan is Delta model NFB0912L 12V 0.17A.
This datasheet says that fan is 2100 RPM at 12V, with no backpressure.
When the fan is bolted to the heatsink, the backpressure might
cause the fan to slow to 1700 RPM, so that is reasonable.

http://www.delta.com.tw/products/dcfans/pdf/NFBseries.pdf

Is it possible the monitor on the P5GD1-VM has miswired the
fan monitoring pins, such that the RPMs seen are from another
fan ? I guess you'd have noticed that by now.

I've had one fan "go crazy" on me. A while back, I bought a
couple of 40mm Northbridge fans. Ones with ceramic ball bearings.
I tested them before install, and the first fan behaved
normally. I plugged it in, and within 2 seconds it was running
at a stable speed.

The second fan, when connected to +12V, started spinning like the
first, but it continued to accelerate for as long as power was
applied. The fan was moving a tremendous volume of air (for a
40mm) before I killed the power. Now, pumping a volume of air
is like wind resistance - the load rises rapidly with speed -
and that fan at least doubled its speed before I disconnected
it. I expect the little transistors inside would probably burn
out in short order, doing stuff like that.

This seems to violate what little I know about how motors
work. Something in the fan is doing the equivalent of
increasing the apparent applied voltage to the fan, as I
thought back-EMF was an invariant property and you cannot
make back-EMF "go away". What that something would be, is
unclear to me.

So, my best guess would have been a defective fan.

Now, the Thermaltake fan you connected - is it running at
its rated speed, or is it way out of spec as well? I
cannot buy two fans "going crazy" at the same time. And
there is no convenient way to get more than 12V applied
to the fan (a short from -5V to the ground pin on the
fan header would do it, but that is far fetched, and it
is possible the motherboard doesn't even connect to -5V).

So, see if "crazy fan syndrome" fits your symptoms, and
see if you can get a replacement Delta fan.

The Delta fan is shown with a three pin header, so it isn't
using the PWM signal from the motherboard. It would be
controllable with the "DC" option, which would change the
+12V applied to 7 to 8 volts or so. But, I don't think the
"crazy fan syndrome" really cares about voltage too much,
as the fan still runs faster than it should, even on reduced
voltage.

There is a short discussion here, but it is for a three phase
motor, which is not exactly the same thing as the fan motor.
But the word "constant" appears in the text a few times, and
what I observed was not a constant behavior, but a runaway
behavior.

http://www.eece.maine.edu/motor/BPMM_Ch8.pdf (ph.2)

There is a posting by "radix2" at the bottom of this page,
describing what "back-EMF" is. When the back-EMF balances
the applied voltage to the motor, the motor should stop
speeding up, but mine didn't.

http://www.rcgroups.com/forums/archive/index.php/t-5585

How is the +12V on your PSU ? Does it read the correct
value in the BIOS monitor ? Is +12V much higher than it
is supposed to be ? 12.6V is the 5% point and higher than
that might indicate a PSU problem.

HTH,
Paul

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

Paul,

Thanks for the reply. The PSU voltages are all within reasonable
ranges. The +12V line is running at 12.04, so I don't see a problem
there.

I don't think it's PSU voltage issue, because when I plug directly into
a PSU lead using the fan adapter, the fan runs at the expected RPMs.

I've tried two different P5GD1-VM motherboards, so I don't think it's a
motherboard issue, unless it is a design flaw, but I would expect to
see more people with these mobos complaining.

The thermaltake fan was running well out of it's range as well. It was
spec'd at 2100RPM, and was running at 3400+.

One thing I did notice in the BIOS under the CPU settings is that it
says that the ratio is unlocked, and it allows me to set the FSB ratio
to whatever I want. The ratio is currently set to 16 to get it to run
at 3.2Ghz. I've tried underclocking it down to 8, but he lowest it
will actually go is 14, so the chip runs at 2.8Ghz. This had no effect
on the fan speed.

Could it be that the motherboard is seeing the CPU is unlocked so it
assumes that I'm going to overclock it, and runs the fan at a higher
speed.

I don't know enough about electronics, but isn't the fan designed so
that when it gets maximum volatage (12V), then it will run at it's
maximum 2100RPM (per the spec). If that's the case, then how much
voltage is being applied to the thing to get it to run so fast?
Thanks again.

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

The adapter that I'm using to connect to a PSU lead just connects to
the yellow (+12V) and black (GND) lines off of the lead.

It doesn't make sense to me either. The fan connection on the MB
should be providing no more than 12V. I've been building computers for
a while now, and have never seen anything like this. I was certain
that it had to be a problem with the original MB that I had, so I
swapped it out for another (brand new) one. Same behavior.

The fact that it happens with 2 different makes/models of fans also
indicates that it's not an issue with the fan either.

I'm going to try another PSU and see if that changes the behavior at
all. I can't imagine that it will, since all of the voltages are
showing up properly in the BIOS.

Thanks for your help.

 

[Guest]

Archived from groups: alt.comp.periphs.mainboard.asus (More info?)

Indy wrote:
> Paul,
>
> Thanks for the reply. The PSU voltages are all within reasonable
> ranges. The +12V line is running at 12.04, so I don't see a problem
> there.
>
> I don't think it's PSU voltage issue, because when I plug directly into
> a PSU lead using the fan adapter, the fan runs at the expected RPMs.
>
> I've tried two different P5GD1-VM motherboards, so I don't think it's a
> motherboard issue, unless it is a design flaw, but I would expect to
> see more people with these mobos complaining.
>
> The thermaltake fan was running well out of it's range as well. It was
> spec'd at 2100RPM, and was running at 3400+.
>
> One thing I did notice in the BIOS under the CPU settings is that it
> says that the ratio is unlocked, and it allows me to set the FSB ratio
> to whatever I want. The ratio is currently set to 16 to get it to run
> at 3.2Ghz. I've tried underclocking it down to 8, but he lowest it
> will actually go is 14, so the chip runs at 2.8Ghz. This had no effect
> on the fan speed.
>
> Could it be that the motherboard is seeing the CPU is unlocked so it
> assumes that I'm going to overclock it, and runs the fan at a higher
> speed.
>
> I don't know enough about electronics, but isn't the fan designed so
> that when it gets maximum volatage (12V), then it will run at it's
> maximum 2100RPM (per the spec). If that's the case, then how much
> voltage is being applied to the thing to get it to run so fast?
> Thanks again.
>

I can't see how the fan should be running faster on the motherboard than
when connected right to the power supply, the motherboard won't supply
more than 12 volts. Unless the motherboard is supplying a lower voltage
and somehow screwing with the fan electronics, but I think that's unlikely.

Are you sure that adapter you used to connect to the power supply is
just a normal adapter delivering 12 volts and not one that's set up to
deliver lower voltage (like 7 volts) to the fan?

--
Robert Hancock Saskatoon, SK, Canada
To email, remove "nospam" from hancockr@nospamshaw.ca
Home Page: http://www.roberthancock.com/