Home     Contact     Projects     Experiments     Circuits     Theory     BLOG     PIC Tutorials     Time for Science     RSS     Terms of services     Privacy policy  
   
 Home      Projects     Experiments     Circuits     Theory     BLOG     PIC Tutorials     Time for Science   

23 March 2010
Author: Giorgos Lazaridis
Simple Ways to Make Fans Silent

From time to time, i have publish circuits to control a DC fan speed. Other circuits change the voltage that powers the fan, others use PWM to control it. Some use only discrete components, some others use microcontrollers. Some are easy to implement, some are harder. While i was absorbed in the complexity of more advanced circuits, suddenly i saw light! "Hey, why not presenting the easiest method EVER of reducing the RPM speed of a fan"? I think that there are many people out there that will love this method, as it is easy, cheap - cheaper than you can think - and yet it works! Moreover, it needs of NO circuit board or other complicated staff!




The basic idea

Well, the basic idea is that, a DC fan's rotation speed depends -among other characteristics- on the voltage that the fan is powered. Lower voltage will cause the fan to draw less current ( I = U / R ). Less current generates weaker magnetic field. Weaker magnetic field produces less torque on the shaft. Less torque on the shaft means simply less rpm!

But first, let's see the circuit that will follow in action:





How can someone reduce the voltage?

No! I will hold my self! Although i am seduced by the idea that i can extend this paragraph to pages, i will only explain -as promised- the easiest way to do it. In a word, this way is called "Resistor"!. The only thing that has to be done, is to connect a resistor in series with the power supply of the fan. The resistor will cause a voltage drop across its leads. This voltage drop will be subtracted from the overall voltage on the fan, and therefore the voltage across the fan will be reduced.

If you read the theory about connecting resistors, you can see that a resistor will cause a voltage drop across its leads and can be calculated as follows:


U = R x I

Where R is the resistor value in ohms, and I the current that flows through the resistor. Now, it is not THAT easy to calculate the resistor that you need for your fan. That is because each fan has completely different characteristics. For example, other fans draw 290mA, others 400mA, others 120mA. Some fans can operate with lower voltages down to 5 volts, others down to 6volts, others down to 7 volts. Bellow these voltages they may stall. I really, really discourage you trying to figure out the fan characteristics and calculate the resistor value. Instead, i encourage you to get a set of resistors between 10 and 100 ohms and try each one. After all, each one will not cost more than $0.20.




Be careful with the resistor wattage!!!

Before you go and get your resistors, you need to know this. There are resistors from 1/8 watt up to several watts. Your resistor must be able to dissipate the amount of power that needs, otherwise it will be burned. Again, there is a proper formula to calculate the wattage:


P = V x I

Where P is the wattage, V the voltage drop and I the current. For example, if you want to operate your fan at 7 volts, that is 5 volts less than the power supply (12V),and you want to provide max current 200mA, you need the following resistor:


R = V / I = 5 / 0.2 = 25 Ohms

P = V x I = 5 x 0.4 = 1 Watt

Now what? You will go and ask of a resistor 25 Ohms 1 Watt? NO! First of all, you will not find such a resistor. The resistor values are standard (you can find the resistor standard values here). And also, 1 Watt is the lower limit. So...

CONCLUSION:

Provide yourself a batch of resistors between 10 and 100 Ohms, ALL rated at 5 Watts. Maybe 2 watts are also enough. This is something that you have to find it out yourself. If you are asked about the tolerance, just answer "5% is ok!". If you do not want to look "completely noob", here are some resistor values that DO exist: 10, 12, 15, 18, 22, 27, 33, 39, 47, 56, 68, 82 and 100 Ohms, all rated at 2 and 5 Watts with tolerance 5%. I would suggest you try with 12 Ohms, 22 Ohms, 33 Ohms, 56 Ohms and 82 Ohms. But do NOT rush. Read the following article as more interesting ideas are yet to follow.




And how am i going to connect this resistor???

In series of course! Well, each fan comes with 2 or 3 wires. There are also the modern PWM fans that have 4 wires. We will not discuss about them though. So, in each case, there should be a RED wire. This is the POSITIVE power supply of the fan, and this is the one that you have to chop. If there is no red wire, then you have probably a 3-wire fan. In this case, the POSITIVE wire is the middle one. The resistor must be placed on this wire. For this reason, you need to cut the positive wire and solder the two ends of the resistor in each side. Remember to use heat-shrink tubes! You don't want your nice motherboard to be short-circuited with +12 volts do you?


Provide yourself the fan and the resistor If the connector has 3 wires, the middle one is the POSITIVE. If it has 2, the RED one is the POSITIVE. Cut the POSITIVE wire... ...and shorten it according to the length of the resistor
Shorten also the resistor leads, should be about 1cm (half an inch) Cut 2 pieces of heat shrink tube, about 5mm (1/4'')longer than the resistor leads. Solder the POSITIVE wire, one on each resistor lead. Have the heat shrink tube first in the wire!!! Your fan is ready!


Some brands comes with a pre-made resistor kit. Two molex connectors are bypassed between the fan connector and the motherboard connector. The positive wire between these molexes is cut and a resistor is connected in series. No difference than before, yet easier to install:


There are pre-made kits with series resistors. This one uses a 57 ohms 2 Watt resistorfor the reduction. The installation is very easy. You simply put it between the fan connector and the motherboard connector.




So what? My fan will be lazy for the rest of it's life?

Ehmmmm, if you stop here, the answer is YES. But there is a better way. You can solder a simple switch parallel to the resistor. When the switch is open, the current will go through the resistor and the fan will rotate at low speed. But when the switch is closed, the current will follow the "easiest" way, the one with the less resistance, and that is through the switch! The voltage drop across the resistor is almost zero (theoretically it is zero). The fan will then rotate at full speed! So, you can simply have 2 wires soldered parallel to the resistor and a selector switch in front of your PC case. With the switch you select the "Turtle" or the "Rabbit" speed! So simple!


You need a simple switch, with 2 or 3 pins, and 2 wires. Solder the wires to the switch. In case there are 3 pins, you will use the middle one and another one, whichever you want. Solder the other sides of the wires to the resistor, one on each side. Use heat shrink tubes for both the resistor leads and the switch pins! It will be more secure and it will give a professional look


This is the final "circuit" that you will have!





Circuit Ideas

Admit it, it was too easy! I should include some other circuits more advanced! Actually, what i have in mind, is...

The easiest way to have adjustable speeds in a DC fan

I have tried, and i will show you 3 circuits that does the same thing: Providing more than one speed selection of a DC motor. All circuits use a switch with multiple positions. I will start with the simplest:

Multiple speed selection circuit for DC motors using resistors


Watch a video of the circuit

Once again, this circuit will use ONLY resistors. The idea is similar to the first circuit described with the resistor. But now, i will use several "small" resistors in series. When you connect resistors in series, the total resistance is the sum of each resistor value. Thus, if i connect 8 resistors each one 10 ohms, i will have a total resistance of 80 Ohms. And now comes the tricky part: I will use a rotary selector with 4 positions. Each position, is connected in a different node of the resistors. For example, the first position of the switch will be connected at the end of the resistor series. The second will be connected between the 6th and the 7th resistor node, the third between the 3rd and the 4th, and the last will be connected on the first resistor, directly on the positive wire. The schematic diagram is as follows:





As you can see, the 4-positions selector each time uses a different number of resistors. On the first position (most right), all resistors takes part. On the second position, 6 resistors takes part. On the third position 3 resistors takes part and on the last position there is no resistor connected. The more the resistors that takes part, the more the voltage drop generated, thus the less the speed of the motor.

You can use a selector with more positions for more speed adjustments. As for the resistor wattage, here comes the good news. The wattage of each resistor, is the result of the total power divided by the number of the resistors. But be careful! If for example the total wattage is 2 watts, this does NOT mean that each resistor can be 2/8 = 0.250 Watts!No! Because when the selector is on the 2nd position, only 3 resistors are used, and they should be 2/3 = 0.66 Watts! I suggest, you have the first 4 resistors 1 Watt, and all others 1/2 Watt. You could have for simplicity all resistors rated 1 Watt as well! No problem. It all depends on the current that the motor will draw and the voltage reduction. For example, my fan that is small and draws only 290 mAmps, runs with no problem having all the resistors rated 1/2 Watt. I NEVER use less than 3 resistors in series though!




Multiple speed selection circuit for DC motors using diodes


Watch a video of the circuit

My favorite! This method works exactly like the previous one - with the resistors. A number of diodes are connected in series. Each diode has about 0.8 Volts voltage drop across its leads. 10 diodes for example would create 8 volts voltage drop. This method is much more power-efficient than using resistors, as diodes do not generate too much heat as the resistors. Moreover, it is easier than the resistors to calculate the total voltage drop, when you know the drop of each diode. I use for example the 1N4001 that have (according to the datasheet) typical voltage drop around 0.93 Volts. No matter which fan i use, the total voltage drop will be the same.

You should pay attention when connecting the diodes. Unlike the resistors, diodes have polarity! There is a strip near one end of the diode. This strip indicates the cathode of the diode. The diode is properly connected (forward biased), when the anode (the other end) is connected to the positive of the power supply and the cathode is connected to the negative. A reverse biased resistor will not allow any current to flow within, until the break point where the diode is destroyed.





The method with the diodes is by far better than the resistors. With the diodes you can have more steps of speeds and they are more stable and power efficient. Also, the 1N series can handle up to 1A current. For a PC fan, the 1N4001 diode is perfect, and almost have the same price as the resistors. Almost. But it worth its weight.




Multiple speed selection circuit for DC motors using Zener diodes


Watch a video of the circuit

I don't really like this method. it is expensive and you cannot have many speed steps. But a circuit with 2 speeds (+ full speed) can be very easily implemented. Moreover, a circuit like the first one (with the single resistor and the switch) would be more efficient and more stable if a Zener diode had been used instead. A 5 Watts Zenner may cost up to $1.

A Zener diode is a remarkable component. If a Zener diode is reverse biased, it will generate a very specific voltage drop across its leads. This voltage drop is pre-defined and is fixed, no matter how much current goes through. There is of course a power limitation for Zener diodes. For PC fans, you need to consider 5 Watts Zener diodes and not less! Even a small fan that draws 300 mA, will need 3.6 Watts!

The circuit that i tested had 2 Zener diodes with voltage drop 3.9 Volts each. I am not sure, but i think that there are no Zener diodes rated up to 5 Watts with smaller voltage drop. This is a major disadvantage if you want many speed selections. It is impossible. If you connect for example 3 Zener diodes in series and each one has 3.9 voltage drop, then the fan will stall as the voltage drop overall is too much (11.7).

Finally, pay attention when connecting Zener diodes. They MUST BE CONNECTED REVERSE! Like simple diodes, Zener diodes have also a small strip to indicate the cathode. But unlike simple diodes, the cathode MUST BE CONNECTED TO THE POSITIVE!!! Otherwise the diode will be immediately and irreversibly damaged.









Relative pages
  • Learn how to connect resistors together
  • The Ohm law
  • The resistor theory
  • Learn about the most popular PC Cooling methods
  • An intelligent self-tunned fan PWM controller
  • A 2-speed PWM temperature fan controller
  • Control a 3-wire Fan with PWM pulses and use the Pulse Stretching method to get clear rpf feedback
  • Dr.Calculus: Ohm law calculator
  • Dr.Calculus: Standard resistor values calculator





  • Comments

      Name

      Email (shall not be published)

      Website

    Notify me of new posts via email


    Write your comments below:
    BEFORE you post a comment:You are welcome to comment for corrections and suggestions on this page. But if you have questions please use the forum instead to post it. Thank you.


          

  • At 28 January 2016, 19:38:27 user Mahdi wrote:   [reply @ Mahdi]
    • I have a 5 Volt fan with 0.5A. which kind of resistors can decrease 1/3 and 2/3 of the speed of the fan.
      in addition which kind of switch is appropriate of bypassing the resistor.
      please Help me


  • At 6 January 2016, 19:07:13 user RyanR wrote:   [reply @ RyanR]
    • Do you have a place that you recommend to buy such cheap resistors as you have mentioned?


  • At 15 October 2015, 21:20:32 user A Mok wrote:   [reply @ A Mok]
    • This tutorial is exactly what I needed coming from a non-electronics background.
      I really appreciate the tutorials you are making. They're simple, well explained, low cost and saves me money in the long run. The net could do with more content like this.


  • At 21 January 2015, 17:00:48 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @Harry sure


  • At 20 January 2015, 14:36:57 user Harry wrote:   [reply @ Harry]
    • Could this be done for usual 4 pin CPU fan?


  • At 9 November 2013, 14:28:17 user Ken wrote:   [reply @ Ken]
    • I found this:

      Original fan voltage: 12V Original fan current: 0.27A

      V=IR (Ohm's Law), so 12 = 0.27R => R = 12 / 0.27 = 44.444
      Fan resistance: 44.444 Ohms

      Target voltage: 7V

      V=IR (Ohm's Law), so 7 = 44.444I => I = 7 / 44.444 = 0.158
      Target current: 0.158A

      Voltage drop across resistor: 12V - 7V = 5V
      V=IR (Ohm's Law), so 5 = 0.158R => R = 5 / 0.158 = 31.746

      Target resistance: 31.746 Ohms
      Power dissipated by resistor: 5V * 0.158A = 0.788W

      http://www.blackfiveservices.co.uk/fanspeedcalc.php?Voltage=12&Current=0.27&Target=7


  • At 9 November 2013, 8:20:51 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @Ken This is the equation but it is not exactly correct, because as voltage drops so does the current. You can start with a 15 to 20 Ohms resistor and run tests. This is the simplest method


  • At 9 November 2013, 1:23:22 user Ken wrote:   [reply @ Ken]
    • I got a 12v 0.27A PC case fan and I want to know what resistor is needed to run at 7v. 5v is the voltage drop, Is the equation R=5v/0.27A R=18 Ohms ???
      Thanks


  • At 8 May 2013, 14:59:41 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @Chas Normally should be ok


  • At 7 May 2013, 6:41:16 user Chas wrote:   [reply @ Chas]
    • With the speed control circuit using the two "Zenar" diodes above, would there be any problem using two fans, 12v 300Ma
      regards Chas


  • At 28 April 2013, 20:05:42 user Mortimer Rhind-Tutt wrote:   [reply @ Mortimer Rhind-Tutt]
    • I just want to thank you for an excellent site.
      I have found your circuits and ideas, along with your very clear explanations, very helpful.


  • At 26 February 2013, 21:08:18 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @Tenfa Sorry, i do not do circuits on demand


  • At 23 February 2013, 15:07:44 user Tenfa wrote:   [reply @ Tenfa]
    • Hello mister/misses,
      I have a problem with my woodstove, It got 2 fans that runs with 24 volts supply, Bud the issiue is that the controller is broken so I fal without controller.My question to you is as next. Can you sent me a idea how i can make another controller so I can use my woodstove again?
      Thank You.

      Nice regards. Tenfa.


  • At 6 January 2013, 12:56:43 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @Allan IMO 85mm is enough. But the bigger the fan the more silent it is. use a carbon filter for the fumes.


  • At 3 January 2013, 23:10:52 user Allan wrote:   [reply @ Allan]
    • Very clever idea, Giorgos. Thank you for sharing it.

      What size fan would you recommend to make for my solder station--to suck the solder fumes away from my nose while I solder ? rpm (or cubic feet/min) and fan diameter

      I'm not sure whether to vent the output air outside or just farther away in the same room. Haven't figured that out yet.

      Allan


  • At 17 September 2012, 4:28:45 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @brunob it depends on the fan you us. A 2w fan is a small one, i experimented with larger fans. anyway, it is better to play safe. but if you calculate that a smaller diode will do, it is ok.


  • At 16 September 2012, 21:41:19 user brunob wrote:   [reply @ brunob]
    • I think the 5W power requirement for Zener diode isn´t correct. See this example:
      "A 2W fan will pass 2/12 = 0.167A at 12V, and around 7.3/12 of that at 7.3V, say 0.102A. The 4.7V zener will then produce 4.7*0.102 = 0.48W of heat, which the 1.3W model will easily handle."
      http://www.pcsilencioso.com/cpemma/diodes.html


  • At 17 July 2012, 14:51:23 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @yllson what current each fan draws?


  • At 17 July 2012, 2:15:08 user yllson wrote:   [reply @ yllson]
    • Hi

      My name is Ylli.I live in Bulgari.
      I have 100 beehives and I want in summer time to put in each box of bees one fan 5 volt each.The volum of the box is 0.5 cm x 0.4 cm x 0.6 cm.
      The power sourse is 220 volt.I have no idea,how to do this job,this connection to work,to give fresh or cool air in each box.
      Total fans are 100 pcs,5 volt each fan.
      Please,I need help.

      Regards

      Yllson


  • At 10 March 2012, 14:03:35 user future eng. wrote:   [reply @ future eng.]
    • U is the diferentce in potential and its mesured in Volts
      Simble for the name , [Simble for mesuring units]
      U , [V]


  • At 1 January 2012, 2:27:45 user Donz wrote:   [reply @ Donz]
    • I have an xbox 360 and the cooling fan only puts out 5v so I want to run an external power supply of 12v. I don't want the fan to always run at 12v so I need a 5 watt resistor but I'm not too sure about the right ohm.any suggestion


  • At 24 November 2011, 21:01:55 user Kammenos wrote:   [reply @ Kammenos]
    • @Raitis maybe you need to put smaller resistor. If you put the 27 and the 47 ohms in parallel, you will have 17 ohms. Try this instead.


  • At 24 November 2011, 20:22:10 user Raitis wrote:   [reply @ Raitis]
    • I put 5w 27 om resistor but fan[12 v, 0.16A] does not work, I tried also with 47 om resistor, but nothing happens. why?


  • At 24 April 2010, 6:02:37 user Kammenos wrote:   [reply @ Kammenos]
    • To tell you the truth, i do not know the story. I know myself that "V" is the symbol for volts. I know also that "V", as well as "E" and "U" are used for voltage. I mainly use "V", except of the "Ohms law" and you know why? The story goes back to my school ages. I used the word "ROUI" in Greek. The letters "OU" are spelled like the double O for "room". I used this word to remember the R=U/I. V in Greek is spelled like N from "Never". Nevertheless, i never say that the voltage across a resistor is 10 U or 10 E! And if i do, then please tell me where it is to correct it. I always write the units with V.
      No matter what, haven't you ever seen another symbol? This guy (http://en.wikipedia.org/wiki/Voltage_regulator) uses U, ad this guy (http://www.allaboutcircuits.com/vol_1/chpt_2/1.html) uses E. This one (http://library.thinkquest.org/C0111209/parse.php?f=2%2Fsymbol) declares U as voltage and V as the unit volts.


  • At 24 April 2010, 4:51:22 user Robert Bryant wrote:   [reply @ Robert Bryant]
    • OK...now I am confused. When did someone change the symbol "V" for voltage, to the symbol "U" ?
      Back in the early 1970's, some manufacturers were using "U" for transistors (e.g. U6, U12) but for the forty+ years that I have been in electronics ohms law has always been; V=IxR, I=V/R, R=V/I. Things do change with time so please advise me.



    delicious
    digg
    reddit this Reddit this
    Faves



     HOT in heaven!


    NEW in heaven!



    New Theory: AC electric motor working principle



     Contact     Forum     Projects     Experiments     Circuits     Theory     BLOG     PIC Tutorials     Time for Science     RSS   

    Site design: Giorgos Lazaridis
    © Copyright 2008
    Please read the Terms of services and the Privacy policy