PWM for buck converter

[johnsorious]

I've been trying for hours on getting my buck converter right. I've seen the pdf for the buck converter from microchip and am trying to build the same buck converter but am trying to simulated it on multisim. I can't get the pwm at the gate of the mosfet to work. Can anyone help me out? I'm trying to achieve a 12v dc input with a 5v dc output at 2amps. I've attached the multisim simulation that I'm working on and the pdf that I was going off of. 

[Danubio]

Hello johnsorious, 

Could you please describe the issue that you're having in more detail please?

Regards,

Daniel C.

[OStep]

You do not have any signal driving your mosfet gate. You just have a floating gate. Why would expect that Mosfet to turn on/off?

Also, assuming you can get 5V on the output, if you want to to have a 2.5A flowing through your load, you need to actually provide the load.

Finally, you're using a P-Channel power mosfet. N-channel mosfets are typically used due to their lower On-resistance.

Attached is modified circuit that generates close to 5V/2.5A on the output.

Notice the following:

• The gate voltage must exceed the source voltage in order to keep the N-Channel MOSFET in fully turn-On. That is why I have 17V amplitude in the pulse source
• The duty cycle is set to 41.66%. In an ideal buck converter, this would lead to an average voltage of 12V x 0.4166= 5V on the output. But since there are some voltage drops in the diode and the mosfet, the voltage is slightly lower
• To gain some insight into what affects the operation of the circuit, I suggest you play with values of: Inductance, Capacitance, Switching frequency, and Rload. You should monitor various voltages and current as you do this, most importantly the output voltage, and inductor current. You will notice how the ripple is affected. You will also notice that the average voltage will stay more or less 4.5V, until the inductor current ripple waveform starts hitting 0Amps. This is when the converter enters the Discontinuous conduction mode. 

Have fun with it. You can learn a lot even from this basic circuit.

[johnsorious]

---

@OStep wrote:

You do not have any signal driving your mosfet gate. You just have a floating gate. Why would expect that Mosfet to turn on/off?

Also, assuming you can get 5V on the output, if you want to to have a 2.5A flowing through your load, you need to actually provide the load.

Finally, you're using a P-Channel power mosfet. N-channel mosfets are typically used due to their lower On-resistance.

Attached is modified circuit that generates close to 5V/2.5A on the output.

Notice the following:

• The gate voltage must exceed the source voltage in order to keep the N-Channel MOSFET in fully turn-On. That is why I have 17V amplitude in the pulse source
• The duty cycle is set to 41.66%. In an ideal buck converter, this would lead to an average voltage of 12V x 0.4166= 5V on the output. But since there are some voltage drops in the diode and the mosfet, the voltage is slightly lower
• To gain some insight into what affects the operation of the circuit, I suggest you play with values of: Inductance, Capacitance, Switching frequency, and Rload. You should monitor various voltages and current as you do this, most importantly the output voltage, and inductor current. You will notice how the ripple is affected. You will also notice that the average voltage will stay more or less 4.5V, until the inductor current ripple waveform starts hitting 0Amps. This is when the converter enters the Discontinuous conduction mode. 

Have fun with it. You can learn a lot even from this basic circuit.

I tried this and it works. Is there a n channel mosfet available in multisim and on the market that could operate on a pic chip outputing 5v? 

---

[OStep]

You would not typically drive Power MOSFETs directly from a digital/logic devices like microcontrollers. The output voltage levels on them are often not sufficient, as you pointed out. More importantly is that the pins do not have sufficient current sink/source capacity to turn on/off the MOSFET quickly enough.

A gate driver chip between the microcontroller and the N-channel mosfet should be used. For example:

http://www.microchip.com/wwwproducts/Devices.aspx?product=MCP1401

[johnsorious]

I know the pic chip I want to use is capable of 8MHz and I would be using 400KHz as shown above. When I used the gate driver chip paired with my pic chip would the chip be providing the correct frequency or would the gate driver chip? 

[OStep]

The gate driver does not set the switching frequency. You can think of it more as an amplifier. You supply the PWM waveform and the gate driver outputs a similar waveform but with typically higher voltages and higher currents in order to correctly and quickly drive the gate of the MOSFET.  

You will need to configure your PIC to output a PWM at 400Khz.

[johnsorious]

Is there a reason why it takes some time for the buck converter to reach ~5V? I was expecting it to be almost instanteously ~5V at the output when turned on.

[johnsorious]

---

@johnsorious wrote:

Is there a reason why it takes some time for the buck converter to reach ~5V? I was expecting it to be almost instanteously ~5V at the output when turned on.

---

Ignore that I figured out why I thought it took a while to get to ~5V on the oscilloscope because it was measuring the beginning.

But is there any reason why when I do a 120VAC input ==> 10:1 ideal transformer ==> full wave rectifier==>filtering cap then into the buck converter designed above that it doesn't work? Did I size my input capacitor wrong? I'm seeing many different equations for sizing filtering cap for full wave rectifier. I've tried multiple but it doesn't work when I plug it into my buck ckt.

[OStep]

It takes time for the converter to reach ~5V because the output capacitor and output inductor are initially discharged; they need some time to reach steady-state. 

Your AC-DC circuit chain sounds logical. However, I am unable to say why it "doesn't work". There could be many reasons for it.

You might want to simualte the circuit in stages. First simulate your AC-DC transformer and rectifier stage. Ensure to add some load resistor and ensure that the voltages are as you expect them. Then add your DC-DC buck converter.

Please attach your Multisim files to help with the inspection.