[Moved] Dc to Dc Buck Converter and Mosfet Driver

[Tony_Ulp]

I am trying to design dc to dc buck converter to be controlled by micorcontroller or fpga.

The problem i am having is that the gate resistor is getting hot quickly. (please see attachment - diagram excatly as
connected on the stripboard).

1. Can you please have a look at the diagram and advise me whats wrong. TC4451 inverter / TC4452 non inverter. (only had these types available)
Resistor between gate and source. Do i need one for this application? how to calculate the value?
2. I want to test buck converter with external pwm, e.g. 555 timer just to be sure is working ok beforeconnected to micorcontroller / fpga board.

Please help me with this project i really appreciated.


Thank you very much

Tony

 

[crutschow]

Re: Dc to Dc Buck Converter and Mosfet Driver

I don't offhand see any reason for R2 to get hot so perhaps there's a miswire.

 

[Tony_Ulp]

Re: Dc to Dc Buck Converter and Mosfet Driver

Thanks crutschow for reply. I double checked the wiring, found dry joint on pin2 of Mosfet driver.
I am also investigating pwm side.

Cheers

Tony

 

[Tony_Ulp]

Re: Dc to Dc Buck Converter and Mosfet Driver

R2 was getting hot due to zener diode being soldered to the gate of mosfet. Resoldered as in diagram, that sorted the problem.

However, the mosfet is getting too hot, after few seconds from start up. The mosfet is ok, wiring is ok, i used function generator (square w/f) to apply 100kHz to pin 2 (input) of TC4452 Mosfet Driver. Reduced the frequency, still too hot. It seems like mosfet is in linear mode rather than on saturation mode. Can someone please help me with this problem.

Thanks

Tony

 

[FvM]

Re: Dc to Dc Buck Converter and Mosfet Driver

It seems like mosfet is in linear mode rather than on saturation mode.

As expectable with the shown driver circuit.

Driver voltage must be at least 6V above Vin to switch the MOSFET fully on, needing an additional supply and stressing the MOSFET driver up to the maximum rating.

 

[BradtheRad]

Re: Dc to Dc Buck Converter and Mosfet Driver

Yes, it's possible your switching device is operating in a middle region between going full on, full off.

You must make certain that your gate voltage is going sufficiently high so it will turn the mosfet full on.

Another point.
Your coil is 220 uH. If you are operating it at 100 kHz, that is probably too fast.

According to my simulation, you can slow it to 20 or 30 or 40 kHz. It may even be essential to do this, so your mosfet does not spend as much time in the linear region.

 

[Tony_Ulp]

Re: Dc to Dc Buck Converter and Mosfet Driver

Thanks guys for your useful information.

Checked the voltage at pint 7 of Mosfet driver and is reading only 9V, 9V less as FVM suggested (6v above Vin).
I am thinking of using LM317 as an additonal supply to set to 18V for gat drive.
BradtheRad, i still want to run it at 100 kHz, possible 200kHz if i can, what value of inductor would you suggest, any calculations (formula) will be helpful.

Thank you

Thank

 

[BradtheRad]

Re: Dc to Dc Buck Converter and Mosfet Driver

BradtheRad, i still want to run it at 100 kHz, possible 200kHz if i can, what value of inductor would you suggest, any calculations (formula) will be helpful.

Thank you

The LR time constant enters into play here. In the time between switch-Off or On, you want some excursion in the waveform through the coil. More than 10 percent.

Example... Here is a simulation when the coil value is 30 uH, with a switching frequency of 100 kHz. (I made an assumption that your desired output voltage is 5V.)



For 200 kHz frequency, you can use a coil between 5 and 30 uH.

The above also assumes you can get full-on and full-off switching action from your mosfet at these speeds.