Hello! I want to make an electronic circuit to control a 36V BLDC motor (three phases, hall sensored). I made three half-bridges (like the ones in the picture) for the three connection wires from the motor. I control the bridge with the help of an arduino through the IN_LO and IN_HO pins. At the beginning I fed the circuit with VCC=22V. The motor turned very well, the switching sequence was good. Everything was fine, until I got up the courage and powered the circuit with a VCC=36V battery. The circuit did not like the 36V. Q1 and Q3 from each half-bridge are gone.
How can I improve these half bridges? I would like to try this without IC drivers for each transistor.

 

those drivers do not look quite right. and the problem is two things. First, the top part of the driver is not switching fully into saturation, and that results in a much greater power dissipation and heating in the upper driver transistors. the second problem would seem to be that there is no provision for assuring that one half of a driver section is switched off before the other section is switched on. That will result in a shorted circuit across the supply voltage, and something will fail.

 

I replaced the motor windings with rezistors to the output of the half-bridges. I measured with the voltmeter if the high side mosfet opens to maximum. It has the same output as the input voltage. From the programming of the microcontroller, I make sure that there is no short circuit across the supply voltage, ensuring a dead time in the code. Maybe is not enough.

 

Hello! I want to make an electronic circuit to control a 36V BLDC motor (three phases, hall sensored). I made three half-bridges (like the ones in the picture) for the three connection wires from the motor. I control the bridge with the help of an arduino through the IN_LO and IN_HO pins. At the beginning I fed the circuit with VCC=22V. The motor turned very well, the switching sequence was good. Everything was fine, until I got up the courage and powered the circuit with a VCC=36V battery. The circuit did not like the 36V. Q1 and Q3 from each half-bridge are gone.
How can I improve these half bridges? I would like to try this without IC drivers for each transistor.

You're over driving the gates of the Mosfets if you take them above their maximum allowed. Check the datasheet, its almost certainly +or - 20 Volts.

An easy way, if you don't mind using an additional supply, is to move diode D1 to the lower VCC point, run that from 10 or 12 Volts, and then the upper Mosfet drain from whatever you require. The upper gate will then be driven to 10V above the Mosfet drain supply.

See below post #199

 

You're over driving the gates of the Mosfets if you take them above their maximum allowed. Check the datasheet, its almost certainly +or - 20 Volts.

An easy way, if you don't mind using an additional supply, is to move diode D1 to the lower VCC point, run that from 10 or 12 Volts, and then the upper Mosfet drain from whatever you require. The upper gate will then be driven to 10V above the Mosfet drain supply.

See below post #199

Thank you for your advice! You are right. It's the first time I'm working on a schematic from scratch. I wouldn't really want to add another voltage source, but I'll try anyway. I found another solution to drive the high side mosfet, by using a transformer with a 1:1 ratio at the gate-source, in this way galvanic isolation is also done. If anyone has tried this method, please let me know if it works well.

 

Thank you for your advice! You are right. It's the first time I'm working on a schematic from scratch. I wouldn't really want to add another voltage source, but I'll try anyway. I found another solution to drive the high side mosfet, by using a transformer with a 1:1 ratio at the gate-source, in this way galvanic isolation is also done. If anyone has tried this method, please let me know if it works well.

Yes, that works as well, and is a standard method for high voltage isolation, but needs relatively high frequency drive to keep the transformers from becoming too large! I had my circuit in post #199 working properly down to a few Hertz. It's horses for courses!