How can I possibly reduce the power dessipation

[opins]

MPPT

Hi every one,

I have just built a 40V, 5A, Maximum Power Point Tracker, using a buck topology, and it works considerably well, especially for low sun conditions.

I used P Mosfet for the switching, and complimentary transistors (BC639 and BC640) as the mosfet gate driver.

My problem the Mosfet gets very hot at mid switching condition.
How can I possibly reduce the power dessipation?

I also want to build another circuit for 50V, 10A load
My problem is how to get a about 3mH, that can withstand morethan 10A?

can I use double inductor?

e.g If i have a 4mH, 6A dual inductor (please see attached), can i parallel the leads to get 2mH, 12A inductor?

I have attached a picture of the inductor I used for the previous circuit

 

[permute]

MPPT

why is the inductor getting hot? typically it will only get hot if there is a lot of current flowing when there is still voltage across the device.

this will either be due to I^2 R losses (Rds_on too high), or due to a fast switching rate -- on each cycle there is some amount of overlap between voltage falling and current rising, which means some amount of energy lost per cycle.

P-FETs tend to do worse in both regards. Perhaps you should look into using an N-FET. look up high-side gate drive schemes.

for inductors, yes, you can use 2 discrete inductors in parallel.

 

[FvM]

MPPT

You are showing a common mode choke. it's not suited as DC converter inductor, because it's getting saturated. Generally, I don't understand why you want to use such a high inductance. It would be only reasonable for low kHz switching frequencies.

 

[opins]

Re: MPPT

thanks for your response,

N mosfet for low side switching buck converter? the issue is that i have not seen a comprehensive literature on that. but I will still search further.

I may probably try using a mosfet gate driver IC, will that reduce switching losses (switching freq is about 15Khz)?

I use large inductance because i want it to operate in continuous conduction mode, and also, the switching frequency is about 15Khz (I use low freq bcos the mppt tracking circuit utilises ripple correlation)

 

[permute]

MPPT

N-FETs have been used for that purpose for a long time -- its also common to see the buck converter used with SCRs/Triacs (for AC), with IGBT's, or with other devices.

The large issue is with ground loops -- the output is connected between Vcc and the FET's drain. thus if the output's negative terminal is connected to a system ground by accident, it essentially shorts across the FET, removing all control from the circuit.

at 15khz, switching losses should be very low for many commercially available FETs. Rds_on and the I^2 R losses in the inductor will likely be the large sources of loss.

and as mentioned, you can also make a high-side driver for an N fet.