Buck converter - low bootstrap capcitor voltage

Discussion in 'The Projects Forum' started by steve135, Jul 21, 2015.

  1. steve135

    Thread Starter New Member

    Jun 29, 2015
    11
    0
    Hey everyone,

    I decided to make my own buck converter using an Atmel 328au uC, NCP5901 MOSFET gate driver, and an op amp for current feedback (which I am not using currently). I attached two images, one of the board layout and one of an oscilloscope image of the voltage at the gate (CH1) and the voltage at the source (CH2). Vin is = 12V so I expected the bootstrap capacitor to be close to 24V - some losses. Both oscilloscope images are referenced to GND. I'm sure there are a lot of mistakes here but this is my learning process, haven't done too many designs so far. What may be the reason the board is not operating as I had intended? Also I have many other oscilloscope images at various test points but I don't know which ones I should post.

    Thank you for your time,
    Stephen

    imageBuckCurrentControl.png mosfetVgs.png
     
  2. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,290
    1,255
  3. steve135

    Thread Starter New Member

    Jun 29, 2015
    11
    0
    buckSchematic.png
    I apologize for forgetting the schematic.
     
  4. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,290
    1,255
    It should work in FET mode although there will be a lot of shoot thru current in the FETs.
    In diode mode it will need a load so the bootstrap cap can charge.
     
  5. steve135

    Thread Starter New Member

    Jun 29, 2015
    11
    0
    The above oscilloscope image is operating with two FETs. The data sheet has an "Adaptive Anti−Cross−Conduction Circuit," according to the data sheet but here is a comparison of the gates of both FETs:
    gateComparison.png
    I removed the R11 (100 Ohm series resistor for bootstrap circuit) and there was no effect. Lower frequency was the only change I could make to have an impact. This circuit is intended to run at 120KHz.

    So it appears there may be shoot through. Could that be due to too slow of a gate charge? Perhaps lower gate resistors? But I am still primarily curious as to why the bootstrap behaves this way in FET mode?
     
  6. DickCappels

    Moderator

    Aug 21, 2008
    2,653
    632
    The ATMEGA328's maximum power supply is 5.5 volts. Running it from 6 volts ±some percentage is not a good idea.
     
  7. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,290
    1,255
    @steve135
    I don't have that driver, but here is a simulation of one with enough gate resistance to prevent shoot thru. As you can see the gate signals are pretty ugly so maybe 120Khz is to fast.
    What FETs are you using.
     
  8. steve135

    Thread Starter New Member

    Jun 29, 2015
    11
    0
    I replaced a the bootstrap capacitor and all was well. What are the chances I damaged that component somehow with hand soldering? Or maybe I grabbed the wrong bag... I am using a NVTR4503NT1G
     
  9. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,290
    1,255
    Ahh yes, much lower gate capacitance than the one I had in the simulation. Hard to tell about the cap. Seldom find a bad one out of the bag, so maybe to hot.
     
Loading...