MOSFET Is not completely on..!

Discussion in 'General Electronics Chat' started by Rydaholic, Apr 9, 2014.

  1. Rydaholic

    Thread Starter New Member

    Feb 4, 2014
    15
    0
    Hi every one,
    I am stuck in a big problem. My MOSFET driver didn't work properly, em using 2110 as driver and IR3205 Mosfet. A PWM signal of almost 5v is applied at Hin Pin of 2110 from Function Generator. When I varied the Duty cycle the load voltages also varies, it seems to be fine with duty cycle, but at 100% duty cycle the voltage of source is not the same as the drain voltage, means MOSFET is not completely on, I think that Driver is working with some problems. Em attached the circuit, but some component values are different which I used in my circiut which are as follows:
    Frequency of operation= 20kHz
    C1=C3= 1uF
    C2=C4= 0.1uF
    D1=D3= 1N4148
    Gate to source resistor= 113 K-ohms
    Drain Voltages is 30v.
    My main problem is MOSFET is still not completely ON, in ON condition it gives me of only 18 to 20v, I think, it should be near 30v.
    My second problem is when I connect a simple DC Motor or Big load to the source of the mosfet, the 2110 is burned after 2 to 3 seconds. I destroyed more than dozen of it.

    I am making it to drive a buck converter.
    Please answer me because I have stuck in it by two weeks.
    Sorry for bad English.
    Thanks..!
     
  2. richard.cs

    Member

    Mar 3, 2012
    162
    31
    The source voltage will not rise above the gate whatever the drain voltage. For this application you need to either change the gate drive or use a P-channel MOSFET
     
  3. joeyd999

    AAC Fanatic!

    Jun 6, 2011
    2,693
    2,765
    Yes, Richard is correct. You have designed a source-follower, which operates in its linear region. It does not behave as a switch in this configuration.
     
  4. MaxHeadRoom

    Expert

    Jul 18, 2013
    10,571
    2,382
    If this is a simple single drive, why not use a low side drive?
    Then you don't need to be concerned with the Vboost circuit.
    Max.
     
  5. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    sorry guys but this is not a source follower. If the circuit is done properly and the bootstrap works correctly then C1 and C2 get charged 12V above the 300V supply.

    First things first, 1N4148 is only good up to 150v so get something better like 1N4007.
    Second thing, duty cycle must ALWAYS be less than 100%, otherwise the bootstrap caps stop charging and the mosfet eventually falls back into linear region.
    Third thing, to allow the bootstrap caps to charge when the mosfet is off, there needs to be a suitable load, that brings the source of the mosfet down to 0V when the mosfet is off, so it needs to be resistive or at least you need a diode and a resistor to allow the bootstrap to charge.
     
    Rydaholic and shortbus like this.
  6. joeyd999

    AAC Fanatic!

    Jun 6, 2011
    2,693
    2,765
    I neglected to read the datasheet. My bad. :p
     
  7. Rydaholic

    Thread Starter New Member

    Feb 4, 2014
    15
    0
    Thanks kubeek. I changed the diodes and my duty cycle is always less than 100%. And Load is also connected. After changing my diodes, my readings become some useful. When Vd is at 30 volts, I get almost 24 volts. One more thing when I reduced the switching frequency of PWM to 5 kHZ, the voltage becomes higher upto 28 to 29 volts. Is there any suggestion to get almost equal voltages to drain voltages, what should I do?
    Another thing is that, suppose this 24 volts is fine for my circuit (which is not :( ) and when I connected a simple buck circuit constructed of a inductor, diode and capacitor, the output I get is equal to the input which is 24 at minimum duty cycle, when I increased DUTY cycle I get voltages near 30v, which is not my input. As theory says it must step down, or low volts than input, but it becomes a Boost application :p
     
  8. Rydaholic

    Thread Starter New Member

    Feb 4, 2014
    15
    0
    @MaxHeadRoom
    I have to make this high side switch because after that I also used low side switch. Basically I want to make a synchronous buck. So first I am trying to make a high side switch with minimum power dissipation and voltage drop.
     
  9. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,293
    1,262
    Take out R3, it is not needed. Then recalculate the boost capacitor. It will be much smaller. Not sure how much this has to do with your problem... but...
     
  10. Rydaholic

    Thread Starter New Member

    Feb 4, 2014
    15
    0
    Thanks ronv.
    I am confused. I think that gate to source resistor is necessary, isn't? Otherwise MOSFET got burn some times, and it should be a big value like in mega-Ohms. And also tell If other thing in schematic is wrong? What say?
     
  11. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,293
    1,262
    The only time you might need a resistor is if you disconnect the driver, otherwise the output of the driver will either be pulled high or low. The low value of resistor will discharge the boost capacitor very quickly dropping the turn on voltage. The diode across the gate resistor may not be needed unless you are trying to turn the FET on slow and off fast.
     
  12. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    What do you mean by Vd? What is your duty cycle? Do you have an oscilloscope?
     
  13. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,293
    1,262
    Like this:

    PS The 3205 is only good for 55 volts.
     
  14. richard.cs

    Member

    Mar 3, 2012
    162
    31
    Same here :rolleyes:
     
Loading...