Intel DP35DP, CPU fan should automatically speed up?

J

John Doe

I have the same problem with the Q9550 stock CPU fan as I did with
the E6850, the fan stays at 1160 rpm while the CPU temperature goes
to 75°C. Maybe my mainboard/motherboard is faulty? Something obvious
in the BIOS that I am missing? Last time, I clipped the blue fan wire
(one of four) and that seemed to help, the CPU fan increased when the
CPU temperature increased.

Thanks.
 
P

Paul

John said:
I have the same problem with the Q9550 stock CPU fan as I did with
the E6850, the fan stays at 1160 rpm while the CPU temperature goes
to 75°C. Maybe my mainboard/motherboard is faulty? Something obvious
in the BIOS that I am missing? Last time, I clipped the blue fan wire
(one of four) and that seemed to help, the CPU fan increased when the
CPU temperature increased.

Thanks.

Is there a setting in the BIOS for QST ? That is a fan control system
which is inside the chipset. On older motherboards, fan control is
done totally by the SuperI/O chip (a chip separate from the chipset).
Intel added QST and uses the Management Engine, a small micro inside
the Northbridge/Southbridge chipset.

There is a note in your manual...

"INTEGRATOR’S NOTE

Regardless of the memory configuration used (dual channel, single channel,
or flex mode), DIMM 0 of Channel A must always be populated. This is a
requirement of the Intel Management Engine in ICH9R."

That is the DIMM slot nearest to the processor. If you're using two sticks
of RAM, one stick would go in that slot, plus a matching stick installed
on the other channel. The Management Engine steals a small amount of RAM
from DIMM 0, to hold code to control the fans. Your BIOS chip would have two
code images, one loaded by the Management Engine, the other by the main processor.

Since the motherboard manual doesn't describe the BIOS screens, I don't know
what options are available to you. On a Gigabyte board, they have an option
to disable QST, and I think that makes the fans run full speed.

Paul
 
J

John Doe

....
Is there a setting in the BIOS for QST ? That is a fan control
system which is inside the chipset. On older motherboards, fan
control is done totally by the SuperI/O chip (a chip separate
from the chipset). Intel added QST and uses the Management
Engine, a small micro inside the Northbridge/Southbridge
chipset.

There is a note in your manual...

"INTEGRATOR’S NOTE

Regardless of the memory configuration used (dual channel,
single channel, or flex mode), DIMM 0 of Channel A must always
be populated."

It is, all memory slots are used.
Since the motherboard manual doesn't describe the BIOS screens,
I don't know what options are available to you. On a Gigabyte
board, they have an option to disable QST, and I think that
makes the fans run full speed.

According to my PDF document "BIOS Settings Glossary,
biosglossarybymenu_v09.pdf", there should be "Unlock Intel(R) QST"
in my DP35DP BIOS under "Advanced > Fan Control Configuration
Menu", but there is no mention of QST.

I have a temporary workaround or two, so it is no big deal. Also, I
might try the Intel utilities "itoolkit" and the "IIA", whatever those
are, apparently they provide some lower level mainboard management. I
was contemplating taking a look at them anyway.
 
P

Paul

John said:
According to my PDF document "BIOS Settings Glossary,
biosglossarybymenu_v09.pdf", there should be "Unlock Intel(R) QST"
in my DP35DP BIOS under "Advanced > Fan Control Configuration
Menu", but there is no mention of QST.

I have a temporary workaround or two, so it is no big deal. Also, I
might try the Intel utilities "itoolkit" and the "IIA", whatever those
are, apparently they provide some lower level mainboard management. I
was contemplating taking a look at them anyway.

I have a Gigabyte motherboard manual (GA-P35-DS3) in my collection, and this is
from their BIOS page.

*******
Smart FAN Control Method

Specifies how to control CPU fan speed.

Auto Lets BIOS control CPU fan speed. (Default)

Intel QST Allows CPU fan speed to be controlled by the
Intel Quiet System Technology (QST). This feature
requires the installation of Intel Host Embedded
Control Interface (HECI) driver from the motherboard
driver disk.

Legacy Allows CPU fan to run at different speed according
to the CPU temperature.

Disabled Forces CPU fan to run at full speed.
*******

I have no clue, as to why you'd need a "HECI driver". That implies
the control method depends on the host CPU. I thought QST relied
on the Management Engine, which should need no driver. Unless
the HECI driver allows controlling QST from the OS ?

For example, I found some software here. Maybe this will work. (I checked
inside this, and it is just a driver. No utility with GUI is provided.)

"Intel ME: Management Engine Driver for Intel 3 Series Chipset-Based Desktop Boards"

http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=13477&ProdId=2782&lang=eng

DP35DP

HTH,
Paul
 
P

Paul

Paul said:
I have a Gigabyte motherboard manual (GA-P35-DS3) in my collection, and
this is
from their BIOS page.

*******
Smart FAN Control Method

Specifies how to control CPU fan speed.

Auto Lets BIOS control CPU fan speed. (Default)

Intel QST Allows CPU fan speed to be controlled by the
Intel Quiet System Technology (QST). This feature
requires the installation of Intel Host Embedded
Control Interface (HECI) driver from the motherboard
driver disk.

Legacy Allows CPU fan to run at different speed according
to the CPU temperature.

Disabled Forces CPU fan to run at full speed.
*******

I have no clue, as to why you'd need a "HECI driver". That implies
the control method depends on the host CPU. I thought QST relied
on the Management Engine, which should need no driver. Unless
the HECI driver allows controlling QST from the OS ?

For example, I found some software here. Maybe this will work. (I checked
inside this, and it is just a driver. No utility with GUI is provided.)

"Intel ME: Management Engine Driver for Intel 3 Series Chipset-Based
Desktop Boards"

http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=13477&ProdId=2782&lang=eng


DP35DP

HTH,
Paul

Another reference to HECI here. Speedfan.

http://www.almico.com/sfhistory.php

"added preliminary detection of the HECI driver"

Paul
 
J

John Doe

....
Intel QST... Allows CPU fan speed to be controlled by the Intel
Quiet System Technology (QST). This feature requires the
installation of Intel Host Embedded Control Interface (HECI)
driver from the motherboard driver disk.

Legacy... Allows CPU fan to run at different speed according to
the CPU temperature.
I have no clue, as to why you'd need a "HECI driver". That
implies the control method depends on the host CPU. I thought
QST relied on the Management Engine, which should need no
driver. Unless the HECI driver allows controlling QST from the
OS ?

Personally, QST (whatever it is) makes no sense to me anyway. I
would think that everyone would want their CPU fan to run
according to the CPU temperature. If it were too loud, then I
would do something other than stopping the CPU fan from running
according to CPU temperature.
For example, I found some software here. Maybe this will work.
(I checked inside this and it is just a driver. No utility with
GUI is provided.)

How did you do that and tell that it has no GUI? Mainly curious.
"Intel ME: Management Engine Driver for Intel 3 Series
Chipset-Based Desktop Boards"

I recall trying that. I will post any experience with the
mentioned utilities.

Thanks.
 
J

John Doe

I will post any experience with the mentioned utilities.

Now posting from my secondary PC.
:(
At first, naturally I thought the BIOS was broken. Symptoms seemed to
get worse over time and trials, down to the point that the CPU (fan)
was spinning up when the PS on/off switch was flipped on, and then it
went off, and then it repeated the same over and over again. Of course
the PC was stripped down to the bare essentials. Fired up my spare PC.
Moved the DVD drive into it, to see if a long standing problem with its
door not properely closing would occure. It closed just fine. Now I am
wondering if the problem could (coincidently) be the PS instead of the
mainboard. Wish I could test the older PS on the new MB, but it has
only an old 20 pin MB connector. I could test the newer PS on the old
MB, but not sure I want to risk it.

Anybody know if the same power suply output that powers the DVD drive
door also powers the CPU? Both 12v?

Thanks.
 
P

Paul

John said:
Now posting from my secondary PC.
:(
At first, naturally I thought the BIOS was broken. Symptoms seemed to
get worse over time and trials, down to the point that the CPU (fan)
was spinning up when the PS on/off switch was flipped on, and then it
went off, and then it repeated the same over and over again. Of course
the PC was stripped down to the bare essentials. Fired up my spare PC.
Moved the DVD drive into it, to see if a long standing problem with its
door not properely closing would occure. It closed just fine. Now I am
wondering if the problem could (coincidently) be the PS instead of the
mainboard. Wish I could test the older PS on the new MB, but it has
only an old 20 pin MB connector. I could test the newer PS on the old
MB, but not sure I want to risk it.

Anybody know if the same power suply output that powers the DVD drive
door also powers the CPU? Both 12v?

Thanks.

Do you have a multimeter you can test with ?

*******

Yes, it is possible the same rail powers both. An experienced power supply
person, can tell by looking at the innards. Sites like jonnyguru.com sometimes
delve into the details of individual designs. There is one other web site
that takes apart power supply for analysis purposes.

A single output rail can be connected to all 12V loads (pcpower.com approach).

A single output rail can have current limiter circuits inserted in each
12V output leg, to simulate independent outputs. I expect that is a common
design technique now, as it is cheaper to make and has most of the
desired safety features. It seems to be the design intent of the ATX 2.2 spec.

A power supply can have multiple, independent outputs. A giveaway is the
power supply chassis may be a lot longer than normal. If you look at the
average power supply today, they're back to "normal" size, which implies
a return to one monstrous output, plus some current limiters for protection
purposes. Some guys can tell you exactly how it works, by looking at the guts,
but I can't always pick apart the different types. "Larry", who posts here
regularly, knows them a lot better than I do.

*******

The main reason for wanting to plug a 24 pin main power connector
into any board, is for SLI or Crossfire video card configurations.
(And SLI or Crossfire implies a PCI Express motherboard.)
The motherboard 12V rail may be heavily enough loaded there, to want
the second yellow wire a 24 pin connector gives you.

For single video card or no video card applications, a 20 pin connector
can be used on a 24 pin motherboard. And even the vice-versa case is allowed,
you can plug a 24 pin into a 20 pin, as long as you can figure out the pinout.
A lot of 24 pin connectors, can be split into two pieces, at the hinge between
the two sections. For those supplies that don't hinge, as long as there aren't
tall components in the way, you can plug a 24 pin into a 20 pin motherboard.

This picture, is of a 24 pin, plugged into a 20 pin motherboard. As long as
there is no electrolytic cap right next to the 20 pin, to bump into the four
"overhang" pins, it will fit.

http://www.playtool.com/pages/psuconnectors/24in20.jpg

This picture is a 20 pin power supply and a 24 pin motherboard, valid
as long as there is just one PCI Express video card installed (rough rule of thumb).

http://www.playtool.com/pages/psuconnectors/20in24.jpg

In neither case, is an adapter cable required. An adapter cable doesn't solve
any electrical problems. An adapter cable is good if you need additional cable
length, and the harness on the power supply happens to be too short.

So most of the motherboard requirements are met just fine, with the lowly 20 pin
connector.

*******

If you need specs, this is my spec list. But this doesn't describe in detail,
how the power supply company must design what is inside. They have a lot of
flexibility in what they can do. And what is printed on the spec label, isn't
guaranteed to elucidate the architecture inside. (When it says "12V1" and
"12V2", they could both be coming from the same output side thing.)

http://web.archive.org/web/20030424...org/developer/specs/atx/ATX_ATX12V_PS_1_1.pdf

http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf

http://www.formfactors.org/developer/specs/ATX12V_PSDG_2_2_public_br2.pdf

Paul
 
P

Paul

John said:
umm, nevermind, probably the clip on the connector

Yup, line up the clip.

The clip is important, in the long run, because there is a
risk of "thermal walkout" if it wasn't there. The expansion
and contraction from one session to the next on the PC,
would be enough to loosen the connector, if the clip wasn't
there to keep it secured.

Paul
 
J

John Doe

....
The clip is important, in the long run, because there is a risk
of "thermal walkout" if it wasn't there. The expansion and
contraction from one session to the next on the PC, would be
enough to loosen the connector, if the clip wasn't there to keep
it secured.

My clip holder/ridge on the socket (DP36DP) runs all the way along
the socket. But the order was easy enough, especially since
apparently the odd honeycomb shape of the connectors forces
correct mating.

I have determined that both of my power supplies (Antec SmartPower
350 and TruePower II 380) produce the same behavior with both
motherboards. That means I wont make the mistake of buying the
PS... THANKS. I guess, somehow the (suspected) defective MB caused
the DVD drive to fail to close, before it eventually totally
failed. The DVD drive door closes fine connected to the old
motherboard even using the previously suspect PS.
 
J

John Doe

maclark7029 said:
I'm having the exact same problem with this MB. Same behavior
with two new PS. Have also swapped memory twice. Funny thing
is I can take the memory out and put back in and the stupid
system will work, but then randomly BSOD and the problems starts
all over...

I just ordered a Gigabyte P45 "ultra durable" motherboard for $90
US from Newegg. It has two BIOS chips, so that should stop
concerns about messing up the BIOS here.

Good luck.
 

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