I have always had a huge fascination with electronics, but never the time to devote to learning it beyond a couple of quick tutorials here and there (and the basics you learn in school). So, sorry if this is too dumb a question for this forum, but here goes anyway.

I have a (SATA) harddrive caddy which I use for backups and moving these from one computer to another. It works well but is notorious for using bad (ie, loud and pretty useless) fans. The device is currently connected so I cannot check, but I believe the fan is 5V (I doubt it is 12V, but it is possible). I can disconnect this fan, and the device works fine, but it causes one of the LED indicators at the front to remain on.. which is also pretty annoying (it's the same LED used to signal when the disc is being accessed, so if it is always on, that renders this function useless too).

But this got me thinking, "what if I just connect an equivalent resistor there instead of the fan? This might mean I can remove the fan, and the noise, and not have that LED permanently lit up." So, am I thinking correctly here? I have a multimeter, so I can check the resistance of the fan, and find a resistor of the same resistance. Would this work? And, what should I have in mind when trying to do this? Thanks in advance.

Edit: A few of the answers have been asking this, so I thought I would just clarify that this is a 2-wire fan. I don't want to replace the fan, I just want to know (also out of curiosity and to learn more about electronics) if this is possible (which, given some of the answers, it looks like it is) and how to go about doing this (safely).

In case anyone is curious, this is the item: KingWin SATA Aluminum Mobile Rack with Single Fan, KF-811-BK

  • \$\begingroup\$ is that fan sounds so noisy. most probably it's due some damaged blade of the fan. When you make it balanced then it won't be a problem any-longer. \$\endgroup\$ – Standard Sandun Apr 4 '14 at 19:31
  • \$\begingroup\$ The thing is that reviews for the product all have people complaining of noisy fans, so it's a product defect. It doesn't even need the fan at all, and I would just remove it, except for the LED that stays on permanently. \$\endgroup\$ – insaner Apr 4 '14 at 20:18
  • \$\begingroup\$ could you record what you call noise? May be listening to it I could guess whether it's familiar or not. \$\endgroup\$ – Standard Sandun Apr 5 '14 at 1:31
  • 1
    \$\begingroup\$ I can't record the noise at this point, but the best way I can think to describe it is that it sounds like a cranky cat being tortured with humanly inaudible noise. Yes, it sounds like a tortured cat. \$\endgroup\$ – insaner Apr 5 '14 at 8:39
  • \$\begingroup\$ A picture of the circuit board would be useful in helping you. \$\endgroup\$ – Passerby Apr 5 '14 at 9:02

If the fault detection depends on current variation with rotation, the resistor isn't going to fool it at all. After all, current will not drop if the fan stalls with dog hair, but air flow will, so simply detecting the current is about useless for detecting fan failure (and is seldom used). Tachometer fans have an extra lead for this purpose, but it can be achieved without extra bits.

You could probably make an oscillator circuit to fool it that would be close enough, but it might be easier to search out the controller and disable it or just replace the fan with a quiet one.

For an example of a current detecting controller, the Microchip TC64x series uses a current sense resistor on the low side:

enter image description here

I would suggest replacing the fan.. the rule of thumb is every ten degrees C of heating halves the life of the product.


Do not simply measure the resistance of the fan. This will be incorrect, because the motor changes resistance as it rotates. Instead, measure the current draw of the running fan, and use that to calculate the necessary resistor.

  • \$\begingroup\$ So resistor needed will be 5/current? \$\endgroup\$ – insaner Apr 4 '14 at 20:16
  • \$\begingroup\$ If you've measured the voltage and it is indeed 5 volts then yes. 5V/current = resistance. \$\endgroup\$ – horta Apr 5 '14 at 2:52

I'm not sure what kind of fan you're using. If it's a 3 or 4 wire PC fan, you might have issues due to the tacho line not returning any info to the computer.

See this: http://www.pcbheaven.com/wikipages/How_PC_Fans_Work/

Otherwise, if you have a 2 wire one you could replace it with a resistor, but if there's no tacho line, then you don't even need a resistor since no process should care about whether or not it's sucking power.

  • \$\begingroup\$ It's a 2 wire for sure. But removing it still causes the LED to stay permanently lit for some reason, so something is definitely going on. \$\endgroup\$ – insaner Apr 4 '14 at 20:19
  • \$\begingroup\$ Then you'll want to replace it with a resistor of equivalent power capabilities and resistance. To find the resistance of a fan, you'll have to check while it's running. Do this by checking the voltage across it running. Then separately, put an ammeter in series with the fan to determine how much current it uses. Then use Ohm's law to determine the live running resistance. (Volts/Amps) \$\endgroup\$ – horta Apr 4 '14 at 20:35

A 5vdc Fan 40mm has .29FLA(Full load amps)(random fan I looked up) 5v /.29amps=17.24ohms. That gives you the resistance you need. Then you need to do 5v*0.29Amps=1.45Watts (the resistor has to be rated at that wattage). I suggest looking up the fans FLA if your going to fool a current sensor.

Look up ohm's law and power dissipation here: http://en.wikipedia.org/wiki/Electrical_resistance_and_conductance

  • \$\begingroup\$ The resistance is voltage divided by current, not multiplied. Correct result, wrong operator. \$\endgroup\$ – alexan_e Apr 4 '14 at 19:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.