Use MC33063 with NMOS in Buck configuration?

[rentner]

Hello, is there any way, to use the MC33063, to make a buck converter using an external transistor, which would be a N-Channel MOSFET, to allow:
-High current
-High input voltage
-High efficiency - or as high as you'd expect it to be in that configuration.

Well. There must be a way to do that, but I got no Clue at all. I was going to simply use an emitter follower (PNP+NPN), but somehow the mosfet does not at all, what it would have to do. Also, i want to use a NMOS, not PMOS, which would be the actual transistor for a buck converter.


Please help me with that problem. I need a circuit, that gives me a starting point. It is ok, if it uses just a 12V supply. Keep it easy. I will edit it on my own later and show the results.

I just don't know, how to get it done. PNP bipolar transistor was all. I did not come any further.

[c4757p]

If you insist on using an N-channel FET you'll need a gate driver with charge pump, because you've got to get the gate up above the supply voltage. Much simpler to use a P-channel FET or a PNP transistor.

[rentner]

The term you are looking for to search is called "High side mosfet driver" or "isolated mosfet driver". The logical thing to do is get a controller that has all this integrated.

Maybe with some sort of GDT? Gate Drive Transformer? That'd be possible for me. But how? Can somebody give a circuit example?


I'd use PNP or PMOS, but aren't they more expensive for the same of NPN or NMOS complementary? Also there are more NMOS out there, so I can choose between more transistors. Would love to use MC33063, because it is the only Converter available to me. Germany has a very, very bad Electronics Part Market. MC33063 is available very easy, but more advanced IC's are only available for high prices or imported from China etc, which takes years to ship... Still waiting for a Class D Amplifier from China, for about 1 month.

I'd love to stay with my MC33063, because in almost ANY situation, it will be the cheapest solution available to me.

[c4757p]

Yes, PNP and PMOS are more expensive, but what happens when you combine NMOS + driver?

[lgbeno]

I'm not a pro with power supplies but it might be possible to boot strap a nmos transistor with that chip.

Here's what I'm thinking:

Connect gate of nmos to pin 2 then a cap between the nmos source and pin and pin 1.  Also put a diode between pin 1 (cathode) and vin (anode).  Do not connect pin 1 to vin...

When q1 is on then your nmos is on, when q1 is off then nmos is off and the cap is charged to vin.

A few issues that I see with this is that your nmos will have to be rated for VGS equal to your vin and I really don't know about stability.  Also you need some gate resistance on the MOSFET.

Be ready to blow up some parts if you try this but I think it could work.

[rentner]

I was searching for some good PMOS. I found one perfect with VDS=200V, which will be enough for my low side. But for the High side with 170V this is not enougth. The only Mosfets, I can find, resisting that voltage, are with RDS of over 200mOhm, which is way to high. More than 5W heat are not ok for me. Actually, they dissipate over 15W!

Is it better, to use a PNP Transistor for that? Since the voltage is so high, the voltage drop shouldn't realy matter anyway. I found some PNPs with 300V+! Dissipating about 50-75W of heat... Should realy be enough. Would you recomment them, since good P-Chan MOSFETS in that Voltage class are expensive as hell?

[lgbeno]

I think pnp is better for this device, depending on how you drive the pmos you'll either need a pretty strong pull up or pull down to turn the pmos on/off when q1 is off.  Can also use a helper transistor externally which will greatly help efficiency.

Also think current through the device not vds for considering power dissipation.  I. Think the rated voltage for the controller is also 40v...

What's the end goal?

Vin=?
Vout=?
Iout=?

[rentner]

What's the end goal?

Vin=?
Vout=?
Iout=?

Galvanic isolated:
Vin1: 60V
Vin2: 60V
Vin3: 160V
Vin4: 160V
Vin5...X: 12V Low current Driver Supplies (Just a small toroid, 1 Diode, 1 Cap and that's all. Not much for that at all!)

Output - Always referenced to ground, Amps as max RMS rating:
Vout: 36V+ 6A
Vout: 36V- 6A
Vout: 130V+ 5A
Vout: 130V- 5A
Vout: 12V+ 1A

Supply for a Home Reciever. The supply is about 100 times more difficult, than the Amplifier section. That power classes...


Also, I want to rund Buck mode for lower Currents through the inductors. That's, why I don't use 30V or so, to Boost them up. no, I want to buck it down.

So you would recomment PNP Transistors? Well, maybe you aren't that wrong, actually...

Still need a way, to drive this.

I was trying to use Photocouplers. That worked at the beginning. But the Current amplification failed at 2 things:
Higher Frequency was lower amplitude on output. 60-70kHz is my goal.
Higher or lower currents to swich also killed the signal. Could you make me a simple circuit for connecting the 2 sides of the circuit example in the attachment?

It just shows the basic idea of my converter. There was an NMOS - Ignore it an think of, if there was simply a PNP Transistor instead.

I'd love to have a circuit for that, because I fail for hours now.