Reducing voltage to a fan

[audioguru]

A resistor in series with the fan limits the current so the fan might not start running.
Use a proper 12V power supply.

 

[snobrder27]

A resistor in series with the fan limits the current so the fan might not start running.
Use a proper 12V power supply.

isnt that why the capacitor would be added?

 

[snobrder27]

alright im rather confused at the explanation of the ohms law stuff. the fan needs 12 volts at .85 amps (Current(I)) and is supplied 20 so the voltage (V) is 8 volts therefore 8=.85R so the resistor (R?) needs to be 9.5ish ohms(8/.85)?? and then the power (P) is found by multiplying the voltage by the current, .85*8 so the resistor needs to be 6.8 watts? now what do i do with these numbers because last time i bought a resister for an LED i was obviously ignorant to these laws and their meaning and only followed directions so am i using the laws correctly?

 

[kchriste]

isnt that why the capacitor would be added?

No, the capacitor prevents the voltage across the 12V fan from spiking up to 20V as it runs.
I actually tested this today with a 12V 4watt computer fan and a 20Ω resistor. My setup ran with 18V input, so there was apx 6Vdc across the resistor. A scope showed a low frequency waveform varing between 8 to 18V as the fan's electronic controller fired the coils in the motor. It was a very nasty looking waveform. Adding a 1000uF 25V capacitor brought the ripple down to apx 0.5Vpk-pk and the DC voltage was apx 12V. The fan started fine with the cap and resistor in circuit.

EDIT:
Yes, the resistor works out to 9.4Ω and 6.8watts. Use a 10Ω 10watt resistor. It will get hot, so place it in the airflow to keep it cool.
Since your fan draws so much current, I would also upgrade the cap size to 2200uF or so if you are going to use a resistor/cap.
At this point, I am leaning more towards using a regulator IC on a heatsink instead because of the higher current and increased ripple current in the cap. What parts suppliers do you have available to you?

 

[westom]

alright im rather confused at the explanation of the ohms law stuff.

Your calculations are correct. I though you were using a standard 80mm or 120 mm fan. A fan of single digit watts to cool a computer. This is a 10 watt fan.

The 10 ohm 10 watt resistor is (was) available in Radio Shack. It will be about as warm as a standard 7 watt Christmas tree light.

You could also so same with a 7812 single chip power supply. Might also still be sold in Radio Shack. But due to high current, the supply chip must be mounted on some metal for cooling. Otherwise would go into protective overheat protection mode. Again, to learn. Routine are for many electronics devices (power supply chips, Intel CPUs, etc) to protect themselves from overheating with no damage.

Apppreciate that your numbers have completely changed the whole problem. Yes you could use the 10 ohm 10 watt resistor. But your problem is approaches limits, for example, on that single chip power supply solution. You will feel things get warm (and well within safe operation even though it feels warm). And to appreciate how much heat is 6 watts.

If a 12 volt fand creates 20 volts spikes, that is perfectly normal and acceptable operation. No acceptable fan will create spikes that destroy itself.

Assuming Radio Shack still sells that 10 ohm 10 watt resistor, then a quick purchase can result in a quick test.

BTW, getting confirmation of your math is simple. More fun is to discover math errors by creating smoke.

 

[audioguru]

The current of a DC motor is extremely high when it starts because it is "stalled'. When it has a resistor in series then its max current is limited and it might not start running.

 

[kchriste]

If a 12 volt fand creates 20 volts spikes, that is perfectly normal and acceptable operation. No acceptable fan will create spikes that destroy itself.

The fan alone didn't create the spikes. The 18 volt peaks came through the resistor when the motor drive transistors shut off and there was briefly no load after the resistor. This happened apx 100x a second as the fan was running. In snobrder27's case, they will be 20V peaks across his 12v fan. The peaks will not be there if the fan is run directly from a 12V supply. That is why I added the capacitor to the crummy design.

 

[Hero999]

A fan is not an ordinary DC motor. It's really an AC motor with an inverter so it can run from DC. I doubt that a DC fan motor will draw a large inrush current.

 

[snobrder27]

Just wanted to let everyone know that I finally got to radioshack. After wiring the 10 watt 10 ohm resistor into the circut the fan runs perfectly. Although after i finished I did try touching a 25 volt 1000 cap across the fan and all that happened was i got a couple of sparks then the fan would stop spinning and I'd have to unplug it and plug it back in to get it to run again so I suppose I'll leave the cap out.

 

[westom]

After wiring the 10 watt 10 ohm resistor into the circut the fan runs perfectly.

Good. Sometime in the future (and look forward to this), you will make a math error and see smoke. We call that fun - or learning from your mistakes. Appreciate how easy using math is to design simple solutions. And why a solution requires both the math and a successful test before that solution is complete.