DC-DC SMPS - Where does the voltage boost come from?

Discussion in 'General Electronics Chat' started by MrSoftware, Oct 9, 2015.

  1. MrSoftware

    Thread Starter Active Member

    Oct 29, 2013
    501
    123
    I'm trying to figure out where the voltage boost comes from in the SMPS side of this audio amplifier.

    In this case, the output stage is capable of 180w RMS into a 4-ohm load with a 1kHz sine wave. If my math is right this requires about 38Vpp. The input to the SMPS is 14v DC. I get that it uses a oscillator and some MOSFETs to create AC, and I was expecting to see a step-up transformer before it's rectified back into DC, but I don't see any sizable transformer at all? Can someone fill me in on how the voltage is boosted in this case? Is it likely that the larger toroid is wound like a transformer? Is it possibly charging the large caps in parallel, then switching them to series to get a voltage boost?

    Here's a picture of the SMPS side of it, I unfortunately don't have an accurate schematic which is part of my confusion.


    ms250_ps.jpg
     
  2. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,290
    1,255
    The big ugly inductor on the right might be a transformer. Does it run off of AC or DC?
    Schematic?
     
  3. MrSoftware

    Thread Starter Active Member

    Oct 29, 2013
    501
    123
    It runs off of 12v-14v DC. Unfortunately I don't have a schematic, wish that I did! I thought that big ugly toroid might act as a transformer, but I wasn't sure. It's got 9 wires coming off of it so it can't have many wraps around..
     
    Last edited: Oct 9, 2015
  4. bumba000

    Member

    Oct 7, 2015
    72
    2
    not sure how many different ways a boost can be accomplished but from all my research and experience thus far I can say that boost happens by energizing one or more inductors from a single power source using a mosfet. The mosfet is essentially engaged and disengaged repeatedly. When this happens a small amount of power is left in the inductor and the next time the mosfet is engaged it adds the power source to the power that has been retained in the inductor. This creates the boost.

    Hope I've helped to answer your question :)

    edit** I was so excited to be able to answer a question here that I didn't finish reading your question. If I had I would have realized that you probably already know what I just explained. I think you're looking for a much more advanced answer. Good Luck!
     
  5. Mosaic

    New Member

    Aug 2, 2010
    8
    1
    Boost supplies rely on the voltage boost when the magnetic field collapses in the inductor.
     
    MrSoftware likes this.
  6. #12

    Expert

    Nov 30, 2010
    16,295
    6,807
    You seem to understand how transformers work, so I'll skip that. Switched capacitor power supplies aren't used for high power, but you seem to understand that, too. The part you seem to question is where a mosfet grounds the far side of an inductor to start current flowing. That forms a magnetic field in the inductor. When the mosfet switches off, the magnetic field collapses and that is the energy which is forcing current to flow. The results obey Energy=1/2 LI^2 and that energy is stored in a filter cap obeying Energy = 1/2 CV^2. So, LI^2 = CV^2 and that's how you convert current to voltage.
     
    MrSoftware and GopherT like this.
  7. Lestraveled

    Well-Known Member

    May 19, 2014
    1,957
    1,215
    Inductors and capacitors are wonderful things. They are great for transforming one voltage/current ratio to a different voltage/current ratio. Consider a capacitor, it can be charged from a low current source, yet when discharged it can output much higher amperage. Consider an inductor, it can be charged with a voltage, yet when discharged, can output much higher voltage. It is not required to use a transformer to convert from one voltage/current ratio to another, you can use a simple inductor/capacitor to make your conversion. These features are inherent in the components, and are the basis of buck and boost power converters.
     
    MrSoftware and Sinus23 like this.
  8. SoftwareGuy

    New Member

    Oct 29, 2013
    29
    1
    Ahh thanks for the excellent descriptions, that makes perfect sense. I didn't think of using the inductor that way, I guess I didn't expect shorting the far side to ground to build up the fields in the inductor since it sounds so wasteful. Then again, current doesn't flow immediately, so if you switch off the grounding transistor before a lot of current flows then you get enough field strength to work with before a lot of current is wasted to ground, do I understand that part right?
     
  9. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,539
    1,251
    The smaller inductor to the left is a common mode choke, part of the electromagnetic interference filter. The fat boy to the right is the power transformer. The SG2525 IC to its left is the pulse width modulator for the boost converter.

    A typical boost converter uses inductive kick to produce a voltage spike, then filters the spike into a capacitor that deliver the energy more evenly for a longer period of time. Changing the amount of time the inductor is charged varies the output voltage.

    ak
     
    MrSoftware and GopherT like this.
  10. #12

    Expert

    Nov 30, 2010
    16,295
    6,807
    Yes, it does. At first glance, every electron that flows out of the source of power is, "used up". Shunting them to ground looks exactly like a waste to me, but switching converters achieve some of the best conversion percentages available.
    That time delay is exactly when the energy is being used to create a magnetic field. Most of that energy is recovered when the mosfet is switched off and the magnetic field kicks current through to the filter capacitor.

    So, yeah, it is not intuitive, but it works better than most other methods.

    Look at it this way: If the left side of the inductor is at Vcc and the right side is grounded, where is all that (P=IE) energy going? A little bit of it is wasted heating the resistance of the coil, but 99% of it is spent building the magnetic field. As long as the current is switched off before the inductor saturates (and becomes merely a resistor), almost all that energy is spent jamming the energy onto the filter cap. Inductors don't care how much voltage is required to discharge the energy, and that is the principle that makes it all work. The inductor will create however much voltage is required to dump the energy.
     
    Last edited: Oct 9, 2015
    MrSoftware and GopherT like this.
  11. MrSoftware

    Thread Starter Active Member

    Oct 29, 2013
    501
    123
    Ok I think I got it. So those power MOSFETs simply ground the inductor to build field strength then turn off. The collapsing field causes the voltage across the inductor to jump, forcing current through the rectifier diodes and into the caps for storage. This repeats at high frequency to keep the caps charged up at higher voltage. The "on" duration of the MOSFETs can be changed to tune the field strength, and therefore the peak voltage potential. Since they are npn transistors, the gate is hit with positive voltage to turn on, and 0v to turn off.

    Cool I learned something today, thanks everyone for teaching me! :)
     
  12. Lestraveled

    Well-Known Member

    May 19, 2014
    1,957
    1,215
    Love it :):)
     
    #12 likes this.
  13. #12

    Expert

    Nov 30, 2010
    16,295
    6,807
    That's what the, "like" button is for.

    The bottom line is energy transfer. Electrons are, "wasted" to ground when the mosfet is, "on", but the energy used to get them there is almost entirely spent building the magnetic field. That magnetic field is what drives the energy into the filter cap. The wasted part is in the resistance of the coil, the resistance of the mosfet, and the voltage drop across the diode(s). There is even a method to use mosfets for the diodes to reduce that energy loss from 0.6 volts to millivolts. I can't remember the name of it right now. Something about synchronous switched rectifying.
     
    MrSoftware and GopherT like this.
  14. crutschow

    Expert

    Mar 14, 2008
    13,014
    3,234
    Synchronous Rectifier is the phase you are looking for, I believe.
     
    MrSoftware and #12 like this.
  15. Dodgydave

    Distinguished Member

    Jun 22, 2012
    4,986
    745
Loading...