400V power supply for experiments

[uA741]

Hi,
I would be interested in designing a working power supply capable of delivering 400V DC in order to power a few Geiger Muller Tubes for a scientific experiment. My idea is to power such a device from batteries so I would like to have a reasonable efficiency to keep the experiment running in the field for a reasonable amount of time.

I have to say that I am only a hobbyist with little experience so I will explain my idea and the results of some early experiments that I have done.

My starting idea was to try to drive a small transformer via an H-bridge MOSFET driver, something like the schematic that I have attached. After searching online I selected the transformer ATB322524-0110-T000 and I have experimented a bit using two SI1029X-T1-GE3 as switching elements and MMBT2222A as NPN transistor.

All the resistors have been set to 1Kohm and the circuit has been powered by 13.6V from a DC-DC converter based on an MT3608. Experimenting a bit however I had not very satisfying results. At low frequencies, less than 300-350KHz, the power consumption of the circuit was very high resulting in overheating of the MOSFETs and breaking down of the DC-DC converter output. At 300-350KHz the output of the transformer became greater than 150Vpp (I have not constructed the diode multiplier yet). Probing the gate of the MOSFETs with an oscilloscope reveals however that, despite the falling of the driver signal on channel A or B (corresponding to the switching OFF of the corresponding NPN transistor), the voltage at the gate start to rise only near the end on the OFF time of the driver signal itself. This gets worse with higher frequency since the MOSFETs have no time to turn ON.

I would like to have your experienced opinions about my idea. Is the idea of using an H-bridge good or it has something inherently wrong? Do you think that I should employ a different circuit? Is the problem of the delayed MOSFET turn on due to the charge/discharge of the gate through the resistor going to VCC? If yes, can I substitute the resistor with a PNP transistor in order to have a lower impedance path of charging?

At the moment I am thinking that solving the delayed turning ON of the MOSFET issue should allow me to run at higher frequencies further reducing the power consumption keeping the MOSFET cooler. Am I correct or I am falling in some trap for young players?

Thank you a lot

[Marco]

Some of your compatriots already made a nice, elegant and safe regulated power supply for Geiger tubes.

https://www.theremino.com/technical/schematics#geigeradapter

Given the low power required and the use of a battery a bridge driver with transformer is more trouble than its worth. The efficiency of the driver/control circuitry starts to be such a large part of the equation that getting a little more efficiency in the switches and passives isn't all that meaningful.

[David Hess]

A boost or flyback converter can drive a voltage multiplier just as well.  Using a push-pull transformer is more suitable where higher output currents are required.

[uA741]

Thank you very much for the pieces of advice. I surely experiment with other design as you suggested.

[Gyro]

A relevant thread from a few years back...

https://www.eevblog.com/forum/beginners/looking-for-hv-schematic-400v-easy-to-build-from-scrap-parts/