PWM or Pulse Width Modulation has generally been regarded as besides complicated for PC fan swiftness control compared against victimisation rheostats or linear voltage regulators so much as the LM317. I instant here a PWM electrical circuit that is simple and cheap enough to be stacked by the casual electronics builder.
Why would you want to economic consumption PWM instead of a rheostat or voltage governor? Well, like to the highest degree things, each method has it's own strengths and disadvantages:
Rheostats
Pros: Cheap and painless to use.
Cons: Difficult to find suitable rheostats, introduces circulating limiting causing the devotee to fail to start or cubicle at pokey zip settings.
Linear voltage regulators
Pros: Does not have variable resistor like current constrictive, temperature control crapper exist incorporated.
Cons: Regulator can mother considerable heat which limits the maximum sports fan ability. (The much powerful the sports fan, the more heat is generated), utmost output voltage to the fan can be various volts less than the supply limiting the maximal cooling from the fan. (A aside-pass switch can be fitted to eliminate this trouble, merely this adds to the complexity of the circuit).
PWM
Pros: As linear voltage regulator, plus comparatively littler heat is generated by the electrical circuit allowing higher powered fans to be used, output is well-nigh 0-100%, eliminating the need for a bypass switch, fans do non stall Beaver State go to start at minimum fan speed.
Cons: The fan speed perception is disabled, can cause "growling" noises at very downhearted speed settings with some fans.
What is PWM? In simple terms, PWM involves chop-chop switching the supply to the fan off and on, in this case 30 times a second, (30Hz). By fixing the relative on to unsatisfactory multiplication the average emf "seen" by the winnow is also altered. E.G. Assuming a 12V furnish, when the on/off ratio = 50%/50% the fan will "see" 6V, too when the along/off ratio is 75%/25% the fan "sees" 9V. Obviously 100% output is achieved by having the turnout on unendingly and at 0% the output is off. It is this on/dispatch switching that makes the winnow pelt along sensor inoperative.
The circuit uses a purpose made chip, MIC502 made by MICREL. Fan speed is adjusted aside variable VR1. With the component values shown the output is 40%-100%. ( 4.8 - 12V). So far every fan I have tested runs at 4.8V. To guarantee the rooter starts when the computer is switched on, the chip automatically runs the devotee at full output for approximately 2 seconds before dropping to the planned speed. Changing the value of R1 and R2 to 7K5 ohms volition move over an output range of 0% - 100%.
If desired, R1 and R2 can be replaced with 22K adaptable preset resistors to allow accurate tailoring of the controller range. Variable R1 will modify the maximum outturn, whilst R2 controls the minimum. Note these adjustments are interactive with each other and volition require some trial and erroneousness to achieve the craved range. The diode, D1 is to protect the transistor from reverse voltage spikes when shift the fan off. Using modern DC axile fans should non be a problem and D1 Crataegus laevigata be considered nonobligatory. Since I cannot test every possible device people English hawthorn connect to the circuit I have included D1 for completeness.
Then again, the decoupling capacitance, C2 is NOT elective. During testing it became apparent that the chip does not like "noisy" supplies. E.G. My 350W " Enermax has same long leads and poor prime connectors. This does not help maintain a clean render. Things may be different for you just I consider C2 to be necessary. C1 sets the pulse frequency, in this shell 30Hz.
During examination "growl" randomness at low speeds was not a problem. If "growl" is a trouble with your specific fan then changing the value of C1 should help. Use the formula C = 3/f where C is in microfarad (uF) and f is the craved frequency in hertz (Hz). Examine frequencies in the range 30-100 Hz. R4 limits the base current to the electronic transistor T1. T1 is rated at 5A which equates to 60W of fan!! In exercise the electrical circuit should be capable of running any just number of case fans. With respective powerful fans connected the transistor may get warm sol a small heatsink may embody needed.
Several channels can be ready-made to make a multi channel fan bus. Single one "C2" is required if all the channels are well-stacked on the same board. Looking the circuit diagram, the "+12V" and "0V" terminals connect to the respective connections happening the PSU Molex connection. The "FAN+" and "Buff-" connect to the corresponding +v and -v terminals happening the fan.
Adding Temperature manipulate.
Here the resistor R1 has been replaced with a thermistor to hand over temperature control. The lover speed is set by VR1 and as the temperature rises, so the fan speed testament increase to compensate.
This version is selfsame to the temperature control in a higher place, except for the gain of a "sleep" setting. The estimation is to turn of events the fan off when the temperature falls below the value limit by VR2.
Proform Pwm Heavy Duty Dual Fan Controller With a C for Small Block Chevy
Source: https://bit-tech.net/reviews/modding/pwm_fan_controller/1/