Problems with High Side Driver

Status
Not open for further replies.
I want to design a Push Pull Class D amplifier with a switching frequency around 200kHz. The Triangle wave is high precicision archieved with super fast OpAmps build out Complementary (NPN) BC547C and (PNP) BC557C. This oscillator works at low side voltage.

Than I offset the IN signal by around 0.3V. Now I put this in a dual DIY High Speed Comparator. The result of the Voltage offset gives me the precice Dead Time I want. The Dead Time can be changed linearly with a 100k pot.


I do not Invert the low side Signal, it just gets into a dual totem pole driver.

BUT HOW TO DRIVE THE HIGH SIDE MOSFET?

There is a big big big problem with this Circuit! The Suply Voltage is -165V/+165V. 1kW into 8 Ohms, 2kW into 4 Ohms and 2 Ohms is not yet planned.



Since the Mosfets used for this Job need to work at 200kHz, widthstand the Supply (2x165V is 330V in total!) and the high currents:
I do NOT WANT to use any P Channel Mosfet, since good P Channel Mosfets for that purpouse are VERY expensive. Over 10€ each. And I need at least 2-3 per side, to lower the losses of RdsOn.




The Problem is: Even thought the Low side switches perfectly, the High Side does NOT. I need a super reliable high side driver for 330V+12V Supply difference.

I think it is time to mention the hardest problem: 100% discrete components!!


Please help me with that. The High Side driver is all that is missing from the Amplifier. The Feedback, Preamplification, Modulation is all done realy perfectly now, but the High Side Mosfet driving does not work.


I am getting shoot throughts and giant dead times and all sorts of stuff. I just want a realy realy fast reliable high side driver. And if it only pushes 1A into the gate I am still quite happy!


What I do not want: Gate Driver Transformer.

Why? Well, they are very unreliable. I tested many circuits for a buck converter, where PWM also is important. And... It did not work at all. Or at least as bad as my solutions for bootstrapping high side drivers.

Max. Gate Charge will be 43nC per Fet, Drivers will be used for a single mosfet only, so multiple Mosfets means multiple drivers.




Well, enougth said: Please help me with that. I am working on this project since last year and failing on a single Mosfet driver...
AIsJFAhKTKc8PAYMFHhoCyBJboYEGJbEECEGCjAEi1MMzyYvlVREIRKG06tRj3YwEtjozw1caCaY0FQiEoEYimoMdLXI1IUlSBD04ECqCRSBWYCACCABAIACBZxSWgDLhhVIqNq82ZBpDRshDYzcclTnl6qBB3T18fHA1ocxQQwsFFglw6RclmRZ+IFC4UBWoc6k8qCjiBIWCaRQYYKoWCJPpSLwWdXrjhlfdHaRoKDFCQ07LdTUmsJLhANiMvR0eXIpB4xWR16UgVNBlIZTwnAg6LDnSzFBIcCMgOCHzIo0c7AQGIRpILAZcgIJ+CTgBC0us1zMFXjgSqwUheQuAgwIADs=
 
Posted below. This represents the equavilent Circuit. The real one will be kept secret. The Signal on the High Side Mosfet gets delayed and also is unstable.


Why even post a Circuit? I don't want a fix of my Circuit. I want the Circuit meeting all my requirements.

330V supply range
200kHz Operation
Super fast turn on and turn off (switching speed of at least 5mHz)
High Side Mosfet Driver for IRF740, if that matters


P Channel CANNOT be used. To expensive and everything in my budget is by far to bad.
 

Attachments

  • circuit.jpg
    circuit.jpg
    340.6 KB · Views: 114
And what do you need it for? All I need was said. The circuit will be released if everything is working in reallife, since the potential difference is 330V! I will not release a not working High Voltage circuit. If it is working, ok. I could sent you a PM, that is ok.
 
Twice the Frequency? This will cause even more Gate Driving issues. The higher the frequency the more difficult it will be to switch mosfets easily.


Just one question: For a bootstrapping High Side driver, I would need a super, super, super fast Mosfet. That means a Mosfet with a gate Charge as low as possible. Are there payable Mosfets out there with a Gate Charge of 2nC or lower? That would be good for my purpouse. It needs Vds 400V. Quite difficult. The slower, the mosfet is, the more time it takes also it takes harder driving which means switching Delays. 🙁
 
Status
Not open for further replies.