I am building a solar charge controller. I have based the design on pwm5 by Julian illet (circuit diagram attached). I am trying to convert this PWM charge controller in to a mppt one.

The buck converter I am using is shown on the second picture. Switching frequency is 31.5 kHz. I have no idea of the inductor I'm using (see the third picture).

My problem was that both IRF3205 are getting very hot. But the IRF3205 I'm using as a diode in the buck converter stopped heating up when I changed the switching frequency to 31.5kHz.

Switching MOSFET is not heating up when it is operating in PWM mode, but the addition of the buck converter circuit makes it toasty. There are no heat sinks on any of the MOSFETs.

enter image description hereenter image description hereenter image description here

  • 1
    \$\begingroup\$ It is unclear to me why you included the first schematic (in green/red, looks like made in KiCad) and the hand drawn schematic. A number of things might be happening in your "hookup". For example currents go in loops, the loop for the current through your MOSFET, inductor, capacitor and (via ground) back to your MOSFET is a very long one. You need to connect the capacitor - node directly to the source of the MOSFET to keep the loop small. Look at how Julian builds his in his videos, he keeps loops small and solders everything, that prevents the issues you see. \$\endgroup\$ Commented Feb 22, 2019 at 7:41
  • 1
    \$\begingroup\$ First schematic is that of the pwm5 controller.hand drawn schematic adds to the first one.in my version the source of the switching MOSFET goes to LC circuit to complete the buck converter instead of going directly to the battery.i have avoided the use of long wires to connect the components but the issue still persist.thanks a lot for your answer \$\endgroup\$
    – jery john
    Commented Feb 22, 2019 at 8:01
  • 2
    \$\begingroup\$ The top N-FET can't ever turn fully on without a system to drive it properly, such as a driver with bootstrapping, causing it to get hot. \$\endgroup\$ Commented Feb 22, 2019 at 8:10
  • \$\begingroup\$ The inductor could also be an issue, if it does not have enough inductance (which depends on the material of the core) it will saturate and that will increase the current dramatically which will heat up your MOSFET. @Unimportant is correct, a common mistake all beginner make is that an NMOS connected like that will act as a switch when the gate is connected to the input voltage (same voltage as the drain). That is not so! The gate needs to be at a higher voltage, so the solar panel voltage + 5 V for example. \$\endgroup\$ Commented Feb 22, 2019 at 8:11
  • \$\begingroup\$ The driver for the fet is the same charge pump circuit for the pwm5 charge controller.are you suggesting it may not provide enough current for the gate drive?.and it is my understanding that voltage at the gate mustbe atleast 10 volts higher than source to turn on the fet.or am I incorrect and voltage at gate need to be 5 volt higher than voltage at drain? \$\endgroup\$
    – jery john
    Commented Feb 22, 2019 at 8:30

1 Answer 1


The upper MOSFET is N channel and requires a voltage at the gate several volts greater than the solar panel voltage in order to turn it on properly. This then ensures that you get a low volt drop between drain and source when it's conducting: -

enter image description here

The lower half of the picture above will be a nightmare when you fix the upper MOSFET gate drive because any switching regulator needs tight control of interconnections between components or it will not perform adequately.

The next problem you might face is saturation of the inductor when operating at such a low operating frequency given the perceived requirements for performance of a battery charger.

  • \$\begingroup\$ Yeah the lower half is a temporary setup to see if my idea works.but isn't Vgs the determining factor as to turn on the fet rather than gate to drain voltage \$\endgroup\$
    – jery john
    Commented Feb 22, 2019 at 8:47
  • \$\begingroup\$ If you want the MOSFET to turn on properly, you want source to be connected to drain and that means the source will be dragged up to the drain voltage and that means the gate needs to be several volts higher than the source and that means it needs to be several volts higher than the drain. Think about it. \$\endgroup\$
    – Andy aka
    Commented Feb 22, 2019 at 9:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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