[anleva]

I have 2 fans in my HTPC. The CPU fan and a system/case fan blowing air out the rear of the case. Both are connected to 4 pin headers on my motherboard which is an Intel DH67CF.


The system fan is a Noctua 4-pin PWM.


I've been using the Bios to control both fans. Up until now fans have been running great. The bios has been increasing the fan speed based on temperature and for the most part both have been spinning at a fairly low RPMs as the component temps have been running in a nice temperature range.


This morning I noticed my system fan was running very fast/high rpms. Nothing out of the ordinary temperature wise as shown in the Intel Desktop Utility. Approximately the same temps as always. That has happened once or twice before and a reboot will fix it and get it back to normal. Not this time. After every reboot the fan stays spinning a high rpms. I've gone back into the Bios and played around with the fan settings, gone from automatic to manual and back, changed back to defaults but the fan continues to spin at a high rate at every reboot.


The CPU fan is still running at a lower speed and the Bios is managing it just as before. Just not the system fan.


So do I have a bad header now or has my fan gone bad? Or any other ideas on what it could be and how to troubleshoot and fix it?

 

[Sammy2]

It would seem the fan is going out and running at full speed or even beyond. You may want to try a different fan.

 

[anleva]

Could be. I think I still have the original case fan so it would help me narrow down if it is a fan issue or not.

 

[Sammy2]

Yeah.. Start there..


Maybe that GT 430 is causing it to go full speed all the time?

 

[anleva]

My temperatures have been fine. And it was working fine for the last 4-5 days I've had the GT 430 up until this morning.


Other thing I've noticed is that I occasionally will see a temp of 0 via Intel Desktop Utility. So that gets me wondering if it is a bad sensor or something is messed up in the bios. I think I will also take out the battery, clear CMOS, and maybe doing a bios restore as well.

 

[Sammy2]

Oh that could be some other issue then.


My case only has the 120mm fan on the powersupply and it is a little stuffy in there. I had the GT 430 OC'd yesterday just to see how it ran and it hit 85C for a stretch until I clocked it back then it dropped down to around 70C. I'm not too worried about it. These things are designed to run hot and everything else in the case was around 45C. I might put the squirrel cage slot fan back in the case, though.

 

[ChairTwo]

After you reset to defaults update the bios to the latest version http://downloadcenter.intel.com/Product_Filter.aspx?ProductID=3283 . Also make sure Intel Desktop Utilitles is up to date.

 

[anleva]

I updated my BIOS with the recovery, jumper method. Problem remains.


I have two 4 pin pwm fans, the CPU fan and System fan. So I swapped the connection. PWM doesn't work for whatever fan is connected to the system fan header but does work for the fan connected to the CPU fan header. The fan connected to the system fan header stays on at 100%. So both fans work.


So this has to be a hardware issue right? Of the system fan header?


I'm outside the NewEgg 30 day RMA so I'm guessing I would need to work directly with Intel to do a warranty replacement?

 

[dksc318]

Looks like the MB system fan PWM function failed. The fans have internal pullup to keep fan at 100% in absence of a PWM input signal.


I am using temperature control instead of PWM control for my system fans. These have a small thermistor on the fan or on a longer cord and use three pin instead of 4 pin connectors. Gelid TC series is of this type of fans. If you search for "sensor" on the CWC-group website cooling category, you will find many temperature controlled fans.

 

[anleva]

Thanks. So basically if PWM fails the system fan is running at 100% voltage? The temperature control fan you mention sounds like an interesting option. What I've done for now is to use the LNA for the Noctua fan to knock it down from around 1800 rpm to 1300 rpm. Actually its pretty good at this speed anyway. Good airflow and very quiet.


Any idea what could cause the PWM function to fail on a header?

 

[dksc318]

Could be as simple as a solder joint failure on the temp sensor or PWM output. The temp sensor (thermistor) is just a small chip mounted on the board. A slip of the screw driver could knock one off. I always have this fear when I work on a MB.

 

[kesawi]

After updating the bios as others have suggested try speedfan ( http://www.almico.com/speedfan.php ) and see whether you can control the fans that way.

 

[anleva]



Quote:
Originally Posted by kesawi  /t/1419492/system-fan-issue#post_22205244


After updating the bios as others have suggested try speedfan ( http://www.almico.com/speedfan.php ) and see whether you can control the fans that way.

Thanks. Very interesting. Speedfan does work to change that's fan's RPM while the PWM BIOS method does not any longer. So I assume Speedfan is controlling fan speed using a different method than PWM?


To use Speedfan what changes do I need to make in my BIOS for fan control? Currently it is set to automatic. The other option is manual with the ability to set a duty cycle (%) minimum RPM. Do I set to manual 100%? How do you keep Speedfan and the BIOS from fighting each other for control of fan speed?

 

[dksc318]

I don't think SpeedFan changes the fan header voltage. May be PWM is working but the temp sensor part is faulty. Without a valid temperature, the MB set PWM to 100%.

 

[anleva]

I am getting realistic temperature readings and changes via the BIOS, Speedfan, Intel Desktop Utilities and HW Monitor. So temperature sensing is working. I was basing my fan speeds off of the CPU temperature and the CPU fan PWM control is working. Just not PWM control of the System Fan.


But RPM control via temperature sensed of the system/outlet fan is working via Speedfan. Just not via the BIOS. Not sure why that would be then.


Any insight on what changes I need to make in the BIOS for fan control if I use Speedfan instead?

 

[kesawi]



Quote:
Originally Posted by anleva  /t/1419492/system-fan-issue#post_22205963


I am getting realistic temperature readings and changes via the BIOS, Speedfan, Intel Desktop Utilities and HW Monitor. So temperature sensing is working. I was basing my fan speeds off of the CPU temperature and the CPU fan PWM control is working. Just not PWM control of the System Fan.

But RPM control via temperature sensed of the system/outlet fan is working via Speedfan. Just not via the BIOS. Not sure why that would be then.

Any insight on what changes I need to make in the BIOS for fan control if I use Speedfan instead?

You shouldn't need to make any changes to the bios when using speedfan.

 

[anleva]

Solved it. By accident. Bios bug with the update that required me to also relabel my fan. Damn you Intel. Who'd a thunk changing a LABEL: on a fan would effect PWM.

 

[Mike99]

I also have 2 fans, the CPU fan is connected to the 4 pin CPU cooling fan connector on the MOBO and the case fan is connected to the 3 pin “SYS_FAN” connector on the MOBO.


I’ve been using HWMonitor to check the temps & fan speeds & it does display the CPU fan speed. But nothing displays for the case fan even though it is running.


HWM has a heading for Fans PWM with 3 fans listed under it but they are all 0%.


What does it take to display and/or control the 3 pin case fan?

 

[Mike99]

I just found this comment on an AnandTech review of an ECS H55 mobo. Maybe the same holds true for the ECS H61 that I'm using.


A couple of onboard fan headers are provided for use and comprehensive fan control is offered for the CPU fan header via the BIOS. Unfortunately there are no control options for the SYS fan whatsoever. ECS do not provide OS level fan control or monitoring software with the H55H-M V1, and third party tools like SpeedFan don't work with this board either.

 

[arnyk]



Quote:
Originally Posted by anleva  /t/1419492/system-fan-issue#post_22203340


I have 2 fans in my HTPC. The CPU fan and a system/case fan blowing air out the rear of the case. Both are connected to 4 pin headers on my motherboard which is an Intel DH67CF.

The system fan is a Noctua 4-pin PWM.

I've been using the Bios to control both fans. Up until now fans have been running great. The bios has been increasing the fan speed based on temperature and for the most part both have been spinning at a fairly low RPMs as the component temps have been running in a nice temperature range.

This morning I noticed my system fan was running very fast/high rpms. Nothing out of the ordinary temperature wise as shown in the Intel Desktop Utility. Approximately the same temps as always. That has happened once or twice before and a reboot will fix it and get it back to normal. Not this time. After every reboot the fan stays spinning a high rpms. I've gone back into the Bios and played around with the fan settings, gone from automatic to manual and back, changed back to defaults but the fan continues to spin at a high rate at every reboot.

The CPU fan is still running at a lower speed and the Bios is managing it just as before. Just not the system fan.

So do I have a bad header now or has my fan gone bad? Or any other ideas on what it could be and how to troubleshoot and fix it?

Any changes to the ambient temperatures?