2
\$\begingroup\$

I'm new at electronics (my new hobby) and got a lot of good responds from you guys so hope you can help me out with this one as well.

I got this an USB fan that I wanted to modulate by controlling it with some PWM.

First I did the following

schematic

simulate this circuit – Schematic created using CircuitLab

D5 is the Arduino digital pins outputting the PWM. It gave a lot of noise - so thought that it might have something to do with the capacitor (C1). Just for fun I moved it around and it worked! The LED was not flickering anymore so I added another LED to see the output (like a control).

schematic

simulate this circuit

Working fan

closer look at the capacitor

It works and I should be happy (I am :D ) - but I have some questions I hope you can help me with.

  1. Why didn't my first attempt work _____ ? (would it work with a MOSFET?)

  2. Is the capacitor adjusting my PWM to a linear voltage that partially opens the transistor _______?

  3. Is the diode (D2 1N4007) placed correctly to protect induction _? (noticed that the swapping the wires of the fan does not change the rotation direction - it just don't rotate..)

Suggestions are very welcome since next step is to put it on a pref-board

\$\endgroup\$
4
  • 1
    \$\begingroup\$ Do you intend to drive the fan with PWM or with a more or less analog voltage/current? Because the first drawing looks more like a PWM drive and the latter looks more like an analog control. In both cases you don't have a fly back diode across the fan, that is bad. It should be reverse biased in normal operation. In PWM mode your transistor shouldn't get all too hot, in analog mode it probably will. In PWM mode your fan will probably whistle slightly at 490Hz, in analog mode it won't. Not sure why the top diagram doesn't work, but you could remove the series LED. Place a fly back diode! \$\endgroup\$ – jippie Jun 15 '13 at 17:37
  • \$\begingroup\$ Added the fly back diode :-) Tried replacing the LED with wire and it didn't work. I also tried switching the wire and resistor - so the resistor was before the capacitor - that didn't work either. Yes I have a heat sink on my transistor due to the heat generated. \$\endgroup\$ – Norfeldt Jun 16 '13 at 7:42
  • \$\begingroup\$ @jippie I want to convert the PWM to an analog signal AFTER the transistor. I got stuck with this design - since after some time it became unstable. So I have taken an new attempt electronics.stackexchange.com/questions/74520/… \$\endgroup\$ – Norfeldt Jun 30 '13 at 15:11
  • \$\begingroup\$ @Norfeldt If my answer worked, could you please mark it as correct? Thanks! :) \$\endgroup\$ – haneefmubarak Aug 4 '13 at 21:02
1
\$\begingroup\$
  1. I'm thinking that for some reason PWM was faster than your transistor (which means it would behave the same with a FET)... So as you suggest in #2...

  2. From your schematic, there is reason to believe that the capacitor and the LED are just holding the voltage at (VCC - VLED), which just turns on the transistor at the maximum current allowed by your resistor. If you want to be able to control it via PWM, use an RC filter to smoothen out the voltage. In simpler terms, switch the places of the resistor and the LED.

  3. Yes, the diode has been placed correctly to prevent inductive kickback (at least, to the battery), however for full inductive flyback protection, you should move the diode to being across the fan.

Here's an image of what it should look like once you've made all of the changes:

CircuitLab Schematic 694upd

But otherwise, good work! :)

\$\endgroup\$
3
  • \$\begingroup\$ Thank you very much for your answer. Tried switching the LED as you suggested and it didn't work. I added the flyback diode :-) \$\endgroup\$ – Norfeldt Jun 16 '13 at 7:43
  • \$\begingroup\$ Updated the picture. The LED needs a large forward voltage, which may not be possible in this circumstance. That's why you have the other LED. You could actually connect that to the same pin though. :) \$\endgroup\$ – haneefmubarak Jun 16 '13 at 18:57
  • \$\begingroup\$ I actually ended with posting a new question electronics.stackexchange.com/questions/74520/… which solved my problem \$\endgroup\$ – Norfeldt Aug 5 '13 at 7:10

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