Connect with us

Fan EMF Circuit

Discussion in 'General Electronics Discussion' started by fatman57, May 31, 2013.

Scroll to continue with content
  1. fatman57

    fatman57

    110
    0
    May 27, 2013
    Hi all,

    This is my first post and first attempt to solder a circuit that I designed onto a board. I have tested it on a breadboard but just want to make extra sure what I am doing is correct.

    I have a circuit (see image below) that runs a 12V PC fan, the signal comes from an Arduino. I know that with an inductive load (fan) I need to guard against back EMF. This can be done with a diode, I have also added a capacitor as I will be using PWM (not the topic of the question here, it is there to make the fan run quietly). In case it is unclear the semi-circle object that the fan negative links to is a 3 pin transistor.

    Is what I have put together correct?

    Thanks in advance for any help!

    [​IMG]
     
  2. Harald Kapp

    Harald Kapp Moderator Moderator

    11,319
    2,595
    Nov 17, 2011
    This circuit looks o.k. under the assumption that the transistor is of the NPN type, base is connected to the resistor and emitter to negative.
    Note that "negative" actually should be ground (GND). Otherwise the transistor will be on even at Vbase=0V. Since the circuit works on the breadboard, i assume that's how you connected it anyway.

    I'm concerned about the capacitor. It will look like a short circuit to the transistor causing a high inrush current. This may destroy the transistor after a while. Normally you don't need a smoothing capacitor in a pwm motor control application. The fan's mass will provide mechanical smoothing. You may have to tweak the pwm frequency if the fan is by chance operated at a resonant frequency where it will emit a loud tone at the pwm frequency.

    If you feel insecure about removing the capacitor, make it at least much smaller. Try 100nF ceramic, for example.
     
  3. fatman57

    fatman57

    110
    0
    May 27, 2013
    Thanks!

    It is an NPN, and it might be my terminology that is wrong concerning the use of 'ground' and 'negative' - what is the difference?

    If I was using a transistor to create a signal and wanted more of a sign wave, surely I would have to use a capacitor then? If so surely the setup would be similar, no?

    I haven't tried much as I have been busy but it could just be that I hit the resonance frequency, it does go away at higher speeds (if I remember correctly), the lowest speed at which the fan starts turning is when the sound is at its most.

    EDIT: if I remove the capacitor, is the general rule of thumb that the diode be in parallel with the inductive load and (almost obviously) not allow electron flow from the positive to the ground?
     
    Last edited: Jun 1, 2013
  4. Harald Kapp

    Harald Kapp Moderator Moderator

    11,319
    2,595
    Nov 17, 2011
    Sorry the term sign wave is unbeknowst to me. Do you mean sine wave? If so, a simple capacitor does not suffce. You would then need a filter, probably an LC filter. But from my point of view you don't need a sine wave to drive the motor. Plain DC or unfiltered PWM will suffice.

    At what frequency do you operate the pwm? Have you tried higher frequencies to get rid of the noise?

    The diode, as connected, will not allow current to flow from positive to ground. However, it will short circuit the emfgenerated by the motor when no power is applied. It therefore is absolutely correct as used here.
     
  5. fatman57

    fatman57

    110
    0
    May 27, 2013
    Thanks!

    'Sign' was a typo - thanks for the correction I looked up LC filter, I was confusing the fact that a capacitor can/will smooth out a signal.

    I operate the PWN at 15% of 1kHz (if that makes any sense, the arduino loops to the tune of 1kHz and I choose what percentage of that to use according to sensor input) at its lowest. Any value lower than that and the fan will not turn, I found a big capacitor silences this and IIRC it did diminish and eventually disappear (without the use of a capacitor) the higher it went (I will have to check this in a few weeks time though, I don't have time at the moment to work on this fully).
     
    Last edited: Jun 2, 2013
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-