Hi,
In my model, I am using a voltage source inverter that is controlled by a TI F28379D launch pad. My controller monitors the voltages and currents and switches a GPIO from 1 to 0 when an error is detected. In real life, this flag turns off all gate signals. I want to simulate this in my model.
I’ve used a DI input and so far have multiplied it by all the PWM capture signals. I think this is not a really good way to go, and it seems that it doesn’t work either. So my question is: does anyone have any idea how to solve this problem?
So my question is: Has anyone an idea how to solve this problem?
Thanks a lot, Ben
Is it somthing about the PWM CAPTURE and average model?Maybe using the power module especilly designed for RT-BOX can help???
To clearify my setup:
I would like to simulate a 3 Phase, 3-Level T-Type NPC Voltage Source Inverter for grid injections on the RT-Box and using an external F28379d Controller to fire the IGBTs. For the VSI I use 3 times the Power Module “3-Level Half Bridge (T-Type)” with the PWM Capture Block for its pulses.
Now, I would like to set a GPIO on the F28379D wich is read by a DI (or something else?) on the RT Box to block all firing pulses at once.
Thanks
Below is a screenshot from one of the TI C2000 TSP demo models. The model uses the approach you describe - in the bottom left of the schematic a digital input is used to toggle the PWM signals going to the switch. If the signal is low, then all PWM signals are zero, and the average on-time for all switches is zero is as well. If the signal is 1, then the PWM signals connect to the component gates as expected.
Multiplying by the digital input signal would have a similar impact. Can you be specific on why that approach doesn’t seem to work for you?
You can download the TI C2000 TSP from the Plexim website (TI C2000 Code Generation | Plexim) and explore the demo models - most of them implement a similar logic.
“Multiplying by the digital input signal would have a similar impact.”
Yes, thats right. It works for me now. There was one bigger mistake I made before I asked this. Thank you for your answer - i think, that the solution with that switch seems a little bit more “professional” than multiplying the signal… but again - that also works.
Thanks Benjamin