No announcement yet.

Can't regulate fan speed on P55 Deluxe3

  • Filter
  • Time
  • Show
Clear All
new posts

  • Can't regulate fan speed on P55 Deluxe3

    I simply can't regulate the fanspeed of my chasis fan's. I already tried to do this in the BIOS and with Asrock OC tuner. I'm able to change the level of the fanspeed but after the changes have been aplied the rpm just stays the same :s

    I have been able to change the CPU fanspeed but just can't change the chasis fanspeed

    Win7 32-bit
    Asrock P55 Deluxe3


  • #2
    Re: Can't regulate fan speed on P55 Deluxe3

    Hi Wtr,

    When it comes to adjusting the speed of fans by any means, it's sometimes important to understand just how that specific fan operates and how the desired change is meant to be put into effect.

    Some fans, such as those that used to be bundled with Intel retail boxed cpu heatsink and fan combinations, and some models of case fans produced by Thermaltake and other manufacturers, have a built-in thermistor ("a type of resistor whose resistance varies significantly with temperature") that is meant to control the operation of the fan RPM based upon the perceived "need" for cooling; hot air cools less effectively, and thus more airflow is needed. In the case of thermistor case fans, the thermistor itself typically looks like a black, silver, or green "blob" of epoxy-ish substance on the end of two wires. It may be positioned directly behind the fan hub or elsewhere within the casing, or the wires may be sufficiently lengthy to allow you to position it nearest the component whose temperatures you are attempting to regulate with that fan. A good picture of what to look for is right here: , though obviously colours may vary. :)

    Other fans, such as the 4-pin type that come with newer Intel retail boxed cpu heatisnk and fan assemblies, can communicate with hardware (the motherboard) regarding the incoming air temperature, the rate of heating and cooling of crucial components, etc. and have their speed dynamically adjusted to try to keep pace with anticipated changes in temperature.

    Finally, the most basic type of fans simply have no throttling systems whatsoever, and run their motors at RPM rates approximately proportional to the voltage given to them vs. their designed voltage times their designed voltage (New RPM ~ old RPM * New voltage/Old Voltage). This is, by far, what is most often used in the inexpensive case fan world, and is also the technology behind many fan speed control systems; in a good system, separate inputs (such as dials) for each connected fan allow you to adjust the speed of each, by regulating the individual voltage sent to each connection. Experienced users of such systems will typically tell you that they often run in "steps," where speeds can vary by as much as 25-33% of fan RPM per level - not exactly the smoothest operation in the world. It is also true that some fans will not start spinning below a minimum applied voltage threshold - but it's entirely possible they can be reduced below that speed afterwards.

    Here is a bit of an arbitrary example, using a fan controller designed to avoid the "initial stall," and using a 2200 RPM fan:

    Setting 0: off
    Setting 1: ~1100 RPM
    Setting 2: ~1465 RPM
    Setting 3: ~1830 RPM
    Setting 4: ~2195 RPM

    Another controller may offer a range like:

    Setting 0: off
    Setting 1: ~550 RPM - WILL NOT START, but will operate at this speed once the fan is spinning
    Setting 2: ~1100 RPM
    Setting 3: ~1650 RPM
    Setting 4: ~2200 RPM

    Note that I am using GREATLY simplified examples, but the general concepts do apply. So what am I trying to get at here? Basically, it's this:

    If your fan is the 4-pin type and fully supports the standards for fan control and regulation, as does the corresponding device to which it's connected, great! You should see a great level of control and adjustability over its performance. However, if the fan supports four pin control and the header it's plugged into does not, or vice-versa, you lose the advantages of the system.

    If your fan is a thermistor-regulated fan and it is not yet running at full throttle, and it is told to reduce speed via a decrease incoming voltage, temperatures in its vicinity may increase sufficiently to cause the thermistor to adjust, resulting in an increase in amperage and total power to the fan (when I said other fan speeds were roughly proportional to the applied voltage, that was because you could assume their physical resistance stayed within a given range because the motor was physically the same, and thus that the amps and power were roughly proportional to the voltage - in this case, however, the thermistor upstream of the motor in the series circuit means that the total system resistance can vary pretty greatly) which can cause it to speed right back up to where it left off - albeit at lower voltage and higher amps, which is *not* a desirable condition.

    Finally, for "conventional" fans with no "intelligent" internal throttling systems, as demonstrated in the above over-simplified examples, if a request for an RPM change has been issued but does nto actually take place, it may simply be a result of the controller not having the available precision to comply with the request. Try using values that differ even further from stock, and see if they take hold. If not, it is entirely possible that speed control of the requested function may not actually be supported for the chosen power source. In that case, your best bet if you really wish to have specific control over the given fans may be to invest in an aftermarket fan controller. There are some out there with very nifty features, including connectivity to your motherboard's internal USB headers for displaying information on an external LCD or controlling fan speeds via software and such. Others may prefer the simple "Three knobs in a row" approach, where they can physically crank up/down the juice on specific fans as they wish, at any time. I personally own and use a Vantec fan and light controller for the computer I am writing this on right now, but for the sake of imbiased recommendations, I neither fully endorse it nor discourage you from checking them out. :P

    DISCLAIMER!!! I make absolutely no guarantees about the technical accuracy of any information provided above. While due diligence has been attempted to ensure the factual correctness of all given information both through publicly-accessible resources such as Wikipedia, Intel design specifications and documents, and personal experience testing and using an array of devices from different manufacturers, some concepts may be over-simplified or slightly inaccurate for the sake of simplicity or time (I was a former Electrical Engineering student, so at one time I probably possessed "above average" knowledge of the working of such systems, but as I later switched into Aerospace Engineering, going back and trying to relearn anything I might have forgotten from my earlier texts for a forum post seems a tad much :P). Feel free to add any additional details or useful information as you see fit, but try to ensure you do so in the spirit of educating the community rather than as any effort at personal discreditation.

    Now that that's out of the way... Good luck in identifying what types of components you have, and hopefully reaching a solution you find desirable. :)

    Edit: In my original post, I forgot to discuss the concept of PWM (Pulse Width Modulation) fan control that can be employed by some motherboards, video cards, etc. From Wikipedia:

    "Unlike the linear methods above that are based on voltage loss, PWM switches the input voltage between (nearly) fully on and fully off. This means there is practically no voltage or power loss and associated heat output. PWM controller can be a relatively small, low-power and cool-running, albeit complex, component that doesn't require heavy duty resistors, diodes or transistors and associated heatsinking."

    Essentially, by rapidly and frequently applying and removing power going to the fan, it experiences brief bursts of acceleration and freewheeling (continuing to spin due to its own inertia while receiving some deceleration due to involved friction, magnetic effects within the motor, etc.) . In theory, most fans should perform acceptably in this case unless they had absolutely zero internal friction, blade friction, and magnetic effects (in which case any bursts applied force would result in higher and higher rotational velocities in perpetuity) or had so much drag or inertia that the applied bursts couldn't get the fan started in the first place.

    And yyyyeeeaahhh... As Wikipedia mentions, it's basically the preferred method of fan speed control by modern motherboards, so almost all of the prior information I provided may be entirely redundant (though hopefully still interesting!). :P Long story short: it may be that your PWM controller for chassis fan headers has gone wonky or is non-functional, lol. Still, it never hurts to get informed, right? heh...
    Last edited by JWK; 09-19-2010, 12:48 AM. Reason: Oops! Forgot to mention PWM fan speed control!


    • #3
      Re: Can't regulate fan speed on P55 Deluxe3

      First of all thx for the quick reply.

      I have checked the connectors and the CPU fan has a 4 pin connector and the three case fans have 3 pin connectors.

      I think that's why I can't regulate my case fans, but I think it's strange that there's an option to regulate the fan speed in the BIOS if there are only 3 pin connectors on the motherboard.


      • #4
        Re: Can't regulate fan speed on P55 Deluxe3

        Originally posted by WtrV View Post
        First of all thx for the quick reply.

        I have checked the connectors and the CPU fan has a 4 pin connector and the three case fans have 3 pin connectors.

        I think that's why I can't regulate my case fans, but I think it's strange that there's an option to regulate the fan speed in the BIOS if there are only 3 pin connectors on the motherboard.
        Hi again,

        One thing that perhaps I should have elaborated on is that the 4th pin is a specific speed control signal pin that allows the speed to be set without forcing the use of any of the other described methods (thermistor, voltage regulation, global PWM), but that does not mean that only 4 pin fans can be regulated; that's precisely the role of the other methods.

        For three pin fans, the third pin (which some fans don't even bother connecting a wire to) is the RPM detection wire that simply provides feedback to your motherboard on the current state of the fan. I've had at least one motherboard that seemed smart enough to provide a reading of "NO FAN" for the speed on a header where no fan was connected, and "RPM ERR" when a 2-pin fan was drawing power without returning an RPM signal. A motherboard employing PWM fan control might check for a signal from the third wire to ensure a fan is at least running before permitting it to be throttled down to a lower setting - heck, upon booting most computers, you will hear all of the fans run to full speed as the BIOS begins to initialize, before regulation kicks in.

        If you have absolutely no luck in regulating the speed of your chassis fan via the BIOS, after first determining you don't have a thermistor fan, you can try using soemthing like SpeedFan which is a software solution that could theoretically override BIOS settings, and if that fails, you could try something like <a href="">this</a>, which allows you to manually adjust the speed of a single fan connected to the motherboard, while still receiving monitoring by the board. Alternatively, you could go full-out with a fan controller supporting multiple devices, though if you wish to continue monitory speeds, you will likely need either adapters to run signal wires to the motherboard (e.g. a single wire on the 3rd pin of a 3 pin header that directs the signal from the fan to a header) or to ensure your fan controller provides some sort of monitoring capability. Check out <a href="">this video</a> for a demonstration of a high-end implementation of that sort.