Hello, Plexim community I’ve been wondering how to properly run a simulation of a basic DC/DC converter topology for a while. The first thing I noticed is that the inrush current is a real deal in the application. I tried looking at the powerstage protection block, but I’m not pretty sure how to properly configure it, I want to run in open-loop at the moment and most of the information and demos all include a controller…
Hi Erick, are you able to share your model that way the community could provide some tips/tricks specific to your model.
Inrush is definitely a big issue and there are many different ways to handle this. I’m no expert and hopefully someone else in the community could help expand on this but from my experience strategies that have been used include:
soft-start:
for a variable frequency DC/DC converter you could slowly ramp the frequency
for duty-cycle control you could have a controller that slowly ramps the duty cycle
here is an example from one of the demo models in the TI TSP that does a combination of both
DC/DC converter with CV-CC controllers where the outer voltage loop will ramp up the current reference at startup → you could classify this as a soft-start mechanism as well
The power stage protection is used to force the PWMs to go to safe state when a configured fault is detected (either on a digital input or an analog input). It could be used in addition to the above strategies to help protect your converter. For instance, you implement a CV-CC controller that enables your converter to reach steady state under normal operating condition but under a fault condition (e.g. inductor short ) or a heavy load startup condition the powerstage protection block will help protect your switch and keep it within it’s safe operating region.
Hi Munadir, thanks for your time answering my question. The thing is here I use the ePWM for the TI in two boost converters, one of them have the advantage of reduce the inrush current, nevertheless I wonder how to soft start it from zero without any control scheme.
Hi Erick, in your model there is nothing a controller can do to prevent inrush. This is because you are basically step changing the input voltage and the inrush will be dictated by the impedance in the path between your voltage source and the capacitors. In fact, the step change from 10V to 30V at 200ms can also not be mitigated by any control strategy. You will need to design HW to be able to mitigate the inrush. These are classic challenges that lots of engineers face and have to make design tradeoffs.