Question about temp and fan speed

  • Thread starter Thread starter News Grouips
  • Start date Start date
N

News Grouips

Hi,
This is my first question to this group, hope I include all the needed info
about my equipment to get your help.

My questions are related to fan speeds and CPU/MoBo temperatures. I know
all fans are running by viewing them and air flow feel. I wonder if the CPU
might run cooler if the rear fan ran faster (there are no adjustments). Is
the CPU at an expected temperature for a C2D? Also, since the front fan is
actually running, I assume the temp sensor is open (see data below). Does
anyone know how to determine this? I know the mobo plugs are in but the fan
connections are not visible. Does anyone know what the resistance should be
on the fan leads? I don't have an appropriate fan for substitution.

My setup at this point has:
ASUS P5B-E motherboard
Intel Core 2 Duo E6400 2.13GHz Dual
1 GB PC6400 DDR2 ram
NVidia 256MB GeForce 7300GT PCI-E video
Enermax 420W EG425P-VESFMA power supply
A floppy
2 hard disks
1 CD-RW optical drive
CoolerMaster Centurian CAC-T05-UB case

The motherboard voltages, temperatures and speeds as follows:
Vcore = 1.12v
+3.3 = 3.28v
+5 = 5.15v
+12 = 12.14v
CPU = 40 to 45 c
Motherboard = 30 to 38 c
Rear fan = 746 rpm
Front fan = 0 rpm
Power = 1530 to 1591 rpm
All these reading are for a quite more-or-less quiescent idling over an hour
or two.

Thanks for any suggestions or information.
Ray
 
News said:
Hi,
This is my first question to this group, hope I include all the needed info
about my equipment to get your help.

My questions are related to fan speeds and CPU/MoBo temperatures. I know
all fans are running by viewing them and air flow feel. I wonder if the CPU
might run cooler if the rear fan ran faster (there are no adjustments). Is
the CPU at an expected temperature for a C2D? Also, since the front fan is
actually running, I assume the temp sensor is open (see data below). Does
anyone know how to determine this? I know the mobo plugs are in but the fan
connections are not visible. Does anyone know what the resistance should be
on the fan leads? I don't have an appropriate fan for substitution.

My setup at this point has:
ASUS P5B-E motherboard
Intel Core 2 Duo E6400 2.13GHz Dual
1 GB PC6400 DDR2 ram
NVidia 256MB GeForce 7300GT PCI-E video
Enermax 420W EG425P-VESFMA power supply
A floppy
2 hard disks
1 CD-RW optical drive
CoolerMaster Centurian CAC-T05-UB case

The motherboard voltages, temperatures and speeds as follows:
Vcore = 1.12v
+3.3 = 3.28v
+5 = 5.15v
+12 = 12.14v
CPU = 40 to 45 c
Motherboard = 30 to 38 c
Rear fan = 746 rpm
Front fan = 0 rpm
Power = 1530 to 1591 rpm
All these reading are for a quite more-or-less quiescent idling over an hour
or two.

Thanks for any suggestions or information.
Ray
Click to expand...

If you are worried about fan RPMs, not all fans have an RPM signal. There
are two types of fans. A fan with three wires (+12V, GND, RPM) has an RPM
signal. A fan with two wires (+12V, GND) has no RPM signal. Many computer
case fans are the two wire variety (because it is a few cents cheaper to
make). That is why not all the case fans may report their RPMs.

Your temperatures are not all that out of the ordinary. You
need to load up your processor cores, with some 100% CPU loading
programs, and measure the temperatures again. Then post back with
the results of your testing.

Paul
 
Paul said:
If you are worried about fan RPMs, not all fans have an RPM signal. There
are two types of fans. A fan with three wires (+12V, GND, RPM) has an RPM
signal. A fan with two wires (+12V, GND) has no RPM signal. Many computer
case fans are the two wire variety (because it is a few cents cheaper to
make). That is why not all the case fans may report their RPMs.

Your temperatures are not all that out of the ordinary. You
need to load up your processor cores, with some 100% CPU loading
programs, and measure the temperatures again. Then post back with
the results of your testing.

Paul
Click to expand...

Thanks for your reply. Yes, both of my fans provide rpm monitoring. The
back fan originally read in the 1200 rpms then it went to the 700's. At
this moment it is back to a reading of 1240 rpm. The front fan originally
read about 1200 then went to 0 and then back to around 1200 again for a
while. It now reads 0 again. I don't know if the fan sensors, the wiring,
the monitor or software is at fault. My concern is that the fan monitors
are no help in monitoring computer health - they can't be trusted. If
someone has some experience in this type of problem, maybe with this same
case and fans, perhaps I can learn an easy way to fix the problem without
just tearing out the fans or ignoring them. Yes, I will monitor the CPU
temp under load but I was trying to resolve this question before I did
anything else. Under limited testing so far I have not seen a CPU temp over
about 46 c. I'm not into game playing so don't expect too much loading.

Ray
 
News said:
Thanks for your reply. Yes, both of my fans provide rpm monitoring. The
back fan originally read in the 1200 rpms then it went to the 700's. At
this moment it is back to a reading of 1240 rpm. The front fan originally
read about 1200 then went to 0 and then back to around 1200 again for a
while. It now reads 0 again. I don't know if the fan sensors, the wiring,
the monitor or software is at fault. My concern is that the fan monitors
are no help in monitoring computer health - they can't be trusted. If
someone has some experience in this type of problem, maybe with this same
case and fans, perhaps I can learn an easy way to fix the problem without
just tearing out the fans or ignoring them. Yes, I will monitor the CPU
temp under load but I was trying to resolve this question before I did
anything else. Under limited testing so far I have not seen a CPU temp over
about 46 c. I'm not into game playing so don't expect too much loading.

Ray
Click to expand...

OK. In fact there is no need to panic. This is actually quite normal.

The monitor chip has a minimum speed it can measure. Let us say it is
600RPM. If the fan was running at 601 RPM, we would see 601 on the
display. If the fan dropped to 599 RPM, the display will read 0. The
reason for this, has to do with the overflow of a counter in the monitor
chip.

Some of the third party tools, like Speedfan (almico.com), have an
option to adjust a divider in the monitor chip. The divider feeds the
fan monitor. Depending on whether the divider is set to x2, x4, x8
and so on, that changes the minimum fan reading. The minimum speed
might be 1800RPM, 900RPM, 450RPM, in response to changing the monitor
divider.

Some BIOS are pretty stupid, and set the divider such that the fan
speed minimum is quite high. That increases the odds that the fan will
read zero, and alarm during POST. I have a P4 motherboard that does
that. My power supply fan is monitored. The power supply fan starts
at below 1800RPM. That trips up the BIOS when the computer starts.
I have to press F1, before I can finish booting the computer. Once
the power supply heats up a bit, the fan runs above 1800 RPM and
there are no further problems.

So, the answer to your monitoring problem, can be to use Speedfan
or a similar program. The motherboard maker may provide a fan speed
monitor program. The BIOS may have an option to set the divider
(unlikely, but we can still hope).

What you are seeing is a "normal annoyance".

Now, assuming the minimum is being met, the next question is whether
the speed variation you are seeing, actually correlates with what the
fan is doing. If the "sound" from the fan is at a constant speed,
and you see the display dip from 1200 to 700, or read 63000, then
that is a problem. On older motherboards, two fan monitoring programs
could fight over the SMBUS, and when one program attempts to take a
reading from the monitor chip, it can upset the other program. That
leads to garbage data. I've stopped seeing those kinds of complaints,
because I think the monitor chips are no longer on the SMBUS.

You can also occasionally have a hardware failure, like a bad signal
from the fan to the motherboard.

So, play around with Speedfan a bit, and see if the results are
more reasonable. I cannot remember which program does it, but
one of the third party programs is smart enough to "auto adjust"
the divider, until the fan reading is within measurement range.
The author of that program, knows his stuff. The other utility
writers should be shot, for using a fixed value, and not hiding
the details of this issue from the end user. With properly written
software or BIOS firmware, the fan should always be readable. I
think there is enough dynamic range in the monitor chip, for
everything except the slowest of fans.

I had some fun with those programs, here:
http://groups.google.ca/group/alt.c..._frm/thread/f943fc7a3a12fa97/b0bae4450470898d

Paul
 
Paul said:
OK. In fact there is no need to panic. This is actually quite normal.

The monitor chip has a minimum speed it can measure. Let us say it is
600RPM. If the fan was running at 601 RPM, we would see 601 on the
display. If the fan dropped to 599 RPM, the display will read 0. The
reason for this, has to do with the overflow of a counter in the monitor
chip.

Some of the third party tools, like Speedfan (almico.com), have an
option to adjust a divider in the monitor chip. The divider feeds the
fan monitor. Depending on whether the divider is set to x2, x4, x8
and so on, that changes the minimum fan reading. The minimum speed
might be 1800RPM, 900RPM, 450RPM, in response to changing the monitor
divider.

Some BIOS are pretty stupid, and set the divider such that the fan
speed minimum is quite high. That increases the odds that the fan will
read zero, and alarm during POST. I have a P4 motherboard that does
that. My power supply fan is monitored. The power supply fan starts
at below 1800RPM. That trips up the BIOS when the computer starts.
I have to press F1, before I can finish booting the computer. Once
the power supply heats up a bit, the fan runs above 1800 RPM and
there are no further problems.

So, the answer to your monitoring problem, can be to use Speedfan
or a similar program. The motherboard maker may provide a fan speed
monitor program. The BIOS may have an option to set the divider
(unlikely, but we can still hope).

What you are seeing is a "normal annoyance".

Now, assuming the minimum is being met, the next question is whether
the speed variation you are seeing, actually correlates with what the
fan is doing. If the "sound" from the fan is at a constant speed,
and you see the display dip from 1200 to 700, or read 63000, then
that is a problem. On older motherboards, two fan monitoring programs
could fight over the SMBUS, and when one program attempts to take a
reading from the monitor chip, it can upset the other program. That
leads to garbage data. I've stopped seeing those kinds of complaints,
because I think the monitor chips are no longer on the SMBUS.

You can also occasionally have a hardware failure, like a bad signal
from the fan to the motherboard.

So, play around with Speedfan a bit, and see if the results are
more reasonable. I cannot remember which program does it, but
one of the third party programs is smart enough to "auto adjust"
the divider, until the fan reading is within measurement range.
The author of that program, knows his stuff. The other utility
writers should be shot, for using a fixed value, and not hiding
the details of this issue from the end user. With properly written
software or BIOS firmware, the fan should always be readable. I
think there is enough dynamic range in the monitor chip, for
everything except the slowest of fans.

I had some fun with those programs, here:
http://groups.google.ca/group/alt.comp.periphs.mainboard.asus/browse_frm/thr
ead/f943fc7a3a12fa97/b0bae4450470898d

Paul
Click to expand...

Thanks very much for the information. It may be why sometimes I get a 0
reading all the time for one fan. One fan monitor reads low (700s) most of
the time and then sometimes it starts higher (1200s) or at 0. The monitor
shows the rpms even when it is lower than the Probe II minimum of 800.
There is no provision for changing the minimums in Probe II or my BIOS.

When the monitor's value changes I have not been able to feel or hear any
difference in the actual fans. Seems that what ever the actual speeds are
the CPU is always ok.

I will get Speedfan and try it. I also will read your reference URL info.
Thanks Paul.

Ray
 
Back
Top