H-Bridge Inverter breaks after a while

Hello everyone
I have designed and implemented a one phase variable frequency inverter using SPWM.
The inverter works for a few minutes and then breaks down. It has happened several times and Mosfets got short circuit and I replaced them. I have attached the circuit. Could anyone give me help on this issue? is there anything wrong with my design?

Here is some detail:

1. SPWM is generated by Atmega 32 AVR
2. Output frequency of the inverter changes from 8 Hz to 55 Hz by a Potentiometer.
3. The delay between series Mosfet is 2us.

hoseink08 said:

Hello everyone

The inverter works for a few minutes and then breaks down. It has happened several times and Mosfets got short circuit and I replaced them.

Here is some detail:

1. SPWM is generated by Atmega 32 AVR
2. Output frequency of the inverter changes from 8 Hz to 55 Hz by a Potentiometer.
3. The delay between series Mosfet is 2us.

Click to expand...

when it fails, you change the FETs. Do you have to replace anything else?

please show the gate drive signals, PB0, PB1, PB2 and PB3

the source of Q1 and Q3 should be tied to the ground of their respective low side drivers
(per the IR2110 drive data sheet, typical connection, page 1)
it looks like the sources of Q1 and Q3 are not connected said pins

sources of Q0 and Q2 appear to be connected to their respective low sides of the driver output, signals N1 and N2

Regarding the gate drive signal: I dont have any picture of the implemented circuit for gate signal.
Common Ground: Yes, I did this in implemented circuit with a wire.
N1/N2 signals: They connect Vs to the source of upper Mosfets, As stated in Datasheet.

It has failed on several occasion. first time when I turned the circuit off, Mosfets got short circuit. When Turn Off command is given, I send zero to output of micro. I dont know whats wrong with it.
The second time when I increased the speed quickly it has occurred again

There are many possible failure causes in the circuit layout, insufficient DC bus bypassing, succeptibility of the control circuit to interferences from the power stage.

Hi,

As mentioned already a couple if time in the forum: A Mosfet fails in case of
* overvoltage, even very short spikes
* overcurrent
* overtemperature

Thus check all of them..

Klaus

What is connected to the output, P0 and P1?

what is the capacitance of C400V Power Cap?

how much current is in that 680 nH inductor feeding the bridge rectifier?

I built a switch mode power supply once that was current fed instead of voltage fed
when the FETs switched, they kept blowing up

the drive was designed for a voltage fed system, so all of the FETs were off briefly
when switching legs (from Q0/Q3 to Q1/Q2, and back, etc)
turning off all of the FETs removed the path for the current and it blew up

when we changed the drive so that all of the FETs were ON when switching legs,
it worked fine.

i don't think that's your problem, because of the capacitor on output of the bridge
but if its a high current feed, and the capacitor is small, then the voltage can rise enough to blow up the FETs

your gate drive is not strong enough to hold the upper device OFF when the bottom device turns ON, this gives minor shoot through ( short of Vin ) eventually the mosfet fail due to this. Try 47E turn on resistor and a 40V, 1A schottky diode in reverse across 47E for a fast turn off ...

A slower turn on of mosfet gives less gate rise pulse on other device => less minor shoot through ...

this is a common problem for newbies...

Easy peasy said:

your gate drive is not strong enough to hold the upper device OFF when the bottom device turns ON, this gives minor shoot through ( short of Vin ) eventually the mosfet fail due to this. Try 47E turn on resistor and a 40V, 1A schottky diode in reverse across 47E for a fast turn off ...

A slower turn on of mosfet gives less gate rise pulse on other device => less minor shoot through ...

this is a common problem for newbies...

Click to expand...

Thank you very much, Should I use this resistor and Schottky Diode for all of Mosfets?

FvM said:

There are many possible failure causes in the circuit layout, insufficient DC bus bypassing, succeptibility of the control circuit to interferences from the power stage.

Click to expand...

if the problem is that the power circuit influences control circuit, what should I do? Should I separate control part from power? because now all of them have a common ground.
Should I separate 5 V from 15V Mosfet Driver and 300V Power part? Does microcontroller get noise?

Hi,

There is no need to seperate power from control, but you need a good circuit ... that means schematic as well as PCB layout.
All good schematics is useless if you have a bad PCB layout.
But I've not seen any PCB layout till now.

Klaus

slower turn on for all mosfets - yes - this will get you going - make sure your heatsinking is good..!