I'm experiencing problems with the fan control. The CPU fan works as designed; it runs around 550 rpm while Windows 8.1 is idle. It speeds up a little, if the CPU load is high for a long time.
The two other fans (front, rear) are running with approx. 1300 rpm, while completely ignorig the cpu load: The never speed up or down, even regardless what is set up in the BIOS.
So, please tell me how to get the fan control working for all of the fans.
The fan speed control algorithms are designed to keep your system as quiet as possible while still addressing thermal conditions. Only if the temperatures reach concerning levels is the fan response going to be increased. Are you seeing temperatures reach levels that you think they shouldn't? Please provide additional information...
thanks for your response. It's the statement "fan speed control algorithms are designed to keep your system as quiet as possible" that made me opening this thread. Why, you might ask? Even if the cpu is idle (just presenting the Windows 8.1 desktop) the front and rear fan are running with approx. 1300 rpm - that's not what I'll call "quiet". The cpu fan runs at this point of time with 550 rpm (THAT is quiet)
I ran a cpu load test to get some stress into it and some heat from it. After a while (~ 30 mins), temperature is rising up to 75 degrees Celsius: The cpu fan goes up to 580 (which is still quiet) the rear and front fan keep running at 1300 rpm (which still isn't quiet). I assume, the cpu fan will go even faster if I'll continue the test, won't it?
Could you tell me how to set the cooling option in the BIOS to get the rear and front fan quiet (when teh sysem idles)?
MartinEdit: All of these fans are 4pin (the rear/front fans are Arctic Cooler)
Because we support both 3-wire (voltage-controlled) and 4-wire (PWM-controlled) fans on the chassis and auxiliary fan headers, we had to choose a minimum duty cycle for these headers that would work for both types. We also designed the voltage-control circuit so that the response from 3-wire fans would be similar to that from typical 4-wire fans. This is not going to be completely successful with every single fan, however; there is simply too much variance in their implementations. In this particular case, you have 4-wire fans that are a lot more responsive than we expected at the minimum duty cycle level. In order to use these fans, you will need to lower the minimum duty cycle for the headers that you are plugging them into. This is done in the Cooling scene in BIOS Setup (Visual BIOS). Click on the entry for the fan header in question and then lower the Minimum Duty Cycle Parameter. Based upon what you have said, a setting of 20% will likely be more appropriate. This happens to be, by the way, the default minimum duty cycle setting that we use with the Processor Fan header. It was chosen to work with the fans in the integrated heatsink-fan units included with boxed processors. In this case, for some third-party processor cooling fans, the minimum duty cycle actually has to be raised.
Each and every processor has a particular temperature, called its Tcontrol temperature, that it can sustain, without degradation, for its warranted lifetime. The Tcontrol temperature can vary from one individual processor to another. I have not seen one below 80 degrees Celsius -- at least not for many, many years. Bottom line, 75 degrees, while high, is not concerning when seen during a sustained stress test and the processor fan's response sounds appropriate. But, if you truly want to hold your processor's temperature to a lower level during a test like this, you have a number of choices at your disposal:
In the case of the processor fan, the default is for all temperature inputs to be equally weighted. That is, any increases in response requested by any of the temperatures will be applied fully. For example, if the response algorithm says increase the duty cycle by 10%, then the full 10% would be applied immediately. In the case of the chassis fans, however, the default is for the processor fan to be weighted at only 50%, lowering (or, in some cases, eliminating) the rate at which the increase if applied. This is to avoid these fans speeding up and slowing down in direct relationship to the processor temperature. For chassis fans that can be individually heard by the user, this could be irritating. On the other hand, if there is a situation where the processor fan is having to speed up too much, you can increase the weighting of the processor temperature at each chassis fan. Making the chassis fans run faster sooner in the process can result in the processor fan not having to work as hard. Depending upon the acoustic cost of each fan, this can raise or lessen the overall acoustics of the system. You have to decide based upon the acoustics of your fans - and even issues like where you sit in relation to the PC chassis, etc.
I hope that's enough of an introduction. In the 8 Series Intel® Desktop Boards we included the most sophisticated fan speed control system ever provided in a desktop motherboard. With capability comes complexity, however, and many folks have a tough time handling its configuration. Let me know if you have any additional questions...
it's me again - after "some" time I'fe found the spare time to test the fan control with the information given by you. To make it shot: No change at all
CPU fan still responses to the CPU temp as it should, but the front and rear fans running still at approx. 1400 rpm: They both "ignore" everything I've done
increasing the duty cycle, changing the influence of the processor's temperature on the fans etc.
How can I change the algorithm at all? I can't find an appropriate menu itom in the Visual BIOS?
And, _the_ question, that let me rotate like a fan: Why are they running all the time at the same speed?
Edit: Even if I set the fan control to "manual", the rpm does not change when I change the duty cycle :-(