1
\$\begingroup\$

what I'm recently occupied with is trying to create a model and later build a synchronous buck converter. I've been doing some troubleshooting to try to resolve my issue, but honestly I just got stuck. The output voltage is supposed to be a half of the input voltage and I've achieved that. My problem is associated with huge switching power losses that would cause the MOSFETs (they are both to be changed, I'm not going to use the shown model) to explode. I've been thinking about implementing soft switching, but I have little experience and don't really know how to tackle that. Maybe someone brighter in that matter than me could offer a piece of advice? enter image description here

enter image description here

\$\endgroup\$
2
  • \$\begingroup\$ The two FETs are on together during the 10ns switching time. This results in an almost short across the 50V supply. You need to arrang a "dead-time" where both switches are off before the other turns on. \$\endgroup\$
    – RussellH
    Commented Apr 8, 2023 at 21:23
  • \$\begingroup\$ The position of V2 is completely wrong and will wreck the upper MOSFET in a nano second. The gate needs to be driven with respect to the source. As for your main problem, add dead-time between the MOSFETs being activated. \$\endgroup\$
    – Andy aka
    Commented Apr 8, 2023 at 21:23

1 Answer 1

2
\$\begingroup\$

Shoot-through seems to be the issue. When both MOSFETs are switched at exactly the same time, there will be a time frame in which both MOSFETs are turned on which will result in high currents and power dissipation. You will need to implement a dead time in which both MOSFETs are turned off before turning on one of the MOSFETs. Typical dead times are tens to hundreds of nanoseconds depending on the switching speed.

Another problem I see with circuit is the driving of the high side MOSFET with a 65 V gate voltage. You will have to find a way of driving it with a stable gate-source voltage in a practical circuit. For a simple simulation like yours I would use a voltage source from gate to source rather than gate to GND.

\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.