Tool/software:
Hi,
I am trying to design a buck converter with an input voltage between 33V and 51V. The output voltage is 20V. The output current is between 100mA and 400mA. The converter works for input voltage between 37V and 51V. But under 37V, the output regulation is not very good. The problem is independant of the current. Also at higher input voltage than 37V, I find the duty not really really stable. It slightly move as if there was some noise on the sense resistor ?
The converter has been made through SIMPLIS simulation and thank to the
LM70XX0-CALC — LM70XX0 quickstart design tool calculator. I attached the file if it helps.
The test have been made on the demoboard. I modified the demoboard according to my needs. It allows me to change of converter if it does not work correctly on the final board.
The initial inductor is 47uH. The output ceramic capacitor are about 50uF. The frequency is 450kHz and it works in FPWM (synchronized through a waveform generator). I suspected at the beginning that it could be something like spread spectrum enabled but when it is synchronized, the LM70XX0 is automatically working in FPWM and the DRSS function is disabled. The sense resistor is equal to 39 mOhm
When I take a look at the duty cycle, it is not stable and move from low duty cycle to high duty cycle and it really looks like subharmonic oscillation. So as I have not access to the value of the slope compensation, I decided to modify the quality factor Q of the transfer function by modifying the value of the inductor. So I decided to increase the inductor to reduce the resonnance responsible of subharmonic oscillation. I confirmed this by a simulation and it effectively reduce the resonnance. I placed a 68uH and then a 150uH. The problem is still the same and it seems that the inductance has no effect on the problem. I also desperately tried to place a 22uH. The problem is still the same.
According to the quickstart design and through SIMPLIS, the converter is stable, the phase margin is around 75° and the crossover frequency about 10 kHz.
I decided to look at the demoboard to see if the problem was also present on the demoboard and it is also present on it but the problem is attenuated. We can see the problem when the input voltage is equal to 6V and the output voltage is equal to 5V @ Iout = 100mA. I do not know if the problem change by increasing the ouput current.
Do you know what could be the problem ? If you need any more information do not hesitate.
