Voltage Ripple Buck converter

Discussion in 'General Electronics Chat' started by vultac, Aug 28, 2010.

  1. vultac

    Thread Starter Active Member

    Mar 2, 2009
    Hi, So i have a buck converter circuit, and in simulation I have assumed the voltage ripple to be 0.1%, in simulation the ripple is very small, 6.56V~6.57V. However, when i built it in the circuit the output voltage ripple is around 6.5V to 6.9V... What are the drawbacks of having a large output voltage ripple?
  2. tom66

    Senior Member

    May 9, 2009
    The drawbacks are that you have introduced another noise source into your system and that if you intend to operate say an audio amplifier off this, you'll find that the noise will get into your output. Also, an output like that is not going to be stable enough to run something like a microcontroller off, even with a voltage regulator, the output will have ripple on it.
  3. hgmjr


    Jan 28, 2005
    Can you post the schematic of your circuit so that we can see if there is anything that can be done to enhance the design?

  4. vultac

    Thread Starter Active Member

    Mar 2, 2009
    i see, thanks...

    this is is the schematic, well in simulation it is working fine... when i built it onto the breadboard...i realized that my output ripple is 500mV... and also im supposed to buck down to 6V

    the one label V_17V is the actual graph from an oscilloscope from the breadboard ... the one in yellow is the output voltage ripple

    The other is from the simulation..
  5. Wendy


    Mar 24, 2008
    35µF? For a filter cap? Try something a bit larger, along with a 0.1µF in parallel.
  6. Potato Pudding

    Well-Known Member

    Jun 11, 2010
    That isn't that bad actually. An SMPS can be fussy and require some fine adjustments to get it working smoothly.

    What size are your output filter choke and Capacitors?
    That 6.7 ± 0.2V level, you could possibly reduce it with filtering into something useful but not .1%.

    To get .1% you might need to put some type of smoothing regulation in but that will cost you a volt for an LDO type, or some waste current for a shunt type.

    First step is to go over everything in your whole switcher and look at it all critically.

    Are you building this or breadboarding it? Breadboarding a switcher and getting only that much extra ripple would be very lucky and good.

    Capacitance and inductance in a breadboard can hash your frequencies. A breadboard that has seen lots of use or abuse just won't make tight enough connections for testing a switcher.

    Do you have all the listed components with no substitutions or possibly abused components? That level of ripple for example could be from trying to use a silicon diode instead of a Schottky. A leaky Capacitor could also do it. Try replacing all the electrolytics one at a time. Make certain that you are looking for things like voltage ratings and wattages, not just component type values.

    Do you have short as possible wiring in the high frequency circuits? You could have extra inductance for using a few extra inches of wire that would change phasing and that changes the negative positive feedback balance, and where you are supposed to be getting smoothing you start to get rippling.

    Check the component arrangements. Don't put your switching coil next to your output filter choke or you could get transformer action and add ripple to the output.

    Most times it is something basic. A loose wire or a floating ungrounded circuit can cause this type of problem.
    vultac likes this.
  7. SgtWookie


    Jul 17, 2007
    Why do you keep starting new threads with your same crappy circuit?

    You're trying to use an N-channel MOSFET on the high side without a proper driver. It's never going to work well.

    In your simulation, you are using ideal inductors and ideal capacitors; they have no parasitics. In the real world, your inductors and capacitors will have LOTS of parasitics.

    Your inductor is way too large, and your output cap is far too small. No wonder you have excessive ripple.
  8. marshallf3

    Well-Known Member

    Jul 26, 2010
    Don't know where a 35 uF cap would come from anyway, and a common 33 uF isn't going to have the lowest ESR in the world.

    I didn't even notice the way that FET is being driven but that could certainly contribute as well.
    vultac likes this.