MC33063 buck scheme from 12V to 3,3V @ 750mA

Hello Gianni,

Welcome to the forums!

I don't see any issues with your schematic. You can reduce the number of resistors that you have where you have the 5 in parallel. Ideally, you would have a single resistor, but I am guessing you are using what is available. You can use 3 in parallel if youd like, which will set the peak current at 900mA. Since you are saying that it should be 750mA, this should be ok.

Also, when doing the layout, be sure to put the capacitors and inductors very close to the device.

Best,
Michael

Hello Michael,

thank you for your reply.

The resistors are 5 because this is a topology that I use in many boards with different output current.

I'm just testing the dc-dc converter with a resistive load of 4,3 Ohm (Iout 770 mA about) with Rsense of 0,33 Ohm so Ipeak = 900mA.

I see an abnormal switching wave and a sawtooth wave on the timing capacitor (Ct). I report below the waveforms to understand better.

The Ct (fuchsia wave) has three different charge time so the ton is different each time for the transistor.

I think that the reason of this behaviour is the peak current limit to 900mA that modify the rise timing of Ct in the oscillator.

Which are your considerations about? Is dangerous use the Mc33063A in this way? It has also a Tcase of 70°C about.

Regards

Hello,

You are right. When the current exceeds the "acceptable" current for the sense resistor, this will trigger the capacitor charge current to increase so that the current is limited.

This app note gives an excellent overview of the operation of the device: http://www.ti.com/lit/an/slva252b/slva252b.pdf

Even though the sense resistor is set so that it will trip with 900mA, the device is not perfectly efficient. If we assume 85% efficiency, which may be slightly high, then an output current of 765mA would be enough to reach 900mA on the input.

It is not dangerous to use the device in this way as it is this is the expected behavior based on the sense resistor. You could make the sense resistor smaller, and I would expect that you don't see this behavior.

As for your mention about the 70C. Is that your case resistance? Certainly thermal dissipation is important in any power design and needs to be considered.

This device is guaranteed to 85C, so it is not recommended to go above that.

Best,
Michael

Hello,
ok for the input current behaviour but for the output current of 765mA i'm not sure that is ok with only 900mA of input.
Infact the Ipeak=2*Iout (from the application note slva252b) so to reach 750mA I need to set an input current of 1,5A.
regards

Hello,

Yes, I agree with your assessment. You are absolutely correct, and that was my mistake before.

The peak current is 2*Iout and the peak current is limited by the sense resistor.

Best,
Michael