3-PH PLL for 400Hz Supply

Hi,

I am using Matlab R2020 and TI F28388D to implement a PLL for 3-Ph grid connected converter. My grid is 400Hz. I have attached my block diagram. It does synchronize to the grid, however I am getting high frequency noise. Unfortunately I cannot attach a video of my pulses, but I can send via email if someone is intrested. Can someone please give some advice how to reduce the noise? I am using GaNs as my switching devices, so even a SC for 100ns would be catostrophic.

My sample time is 3.33e-6

PID Controller Parameters are

P= 40

I= 5000

D= 20

Thank you

Regards

Bilal

Answer 1

[Joel Van Sickel]

Hello Bilal,

I don't have the particular support package installed, so its possible I missed it, but it looks like you perform no filtering on your inputs. However, maybe you are filtering on the TI block in hardware, in which case I apologize. You can either implement a PLL that naturally fiters out this noise, or you can just filter the inputs, and take the phase shift into account. I didn't take a close look at what PLL approach you were using as there are so many, but you could probably just put a low pass filter on the inputs and take into account the phase shift you inject from that. This is especially true since you have a fixed line frequency and the noise from a GaN converter should be pretty far away from 400 Hz.

I would recommend including a plant model in your simulation so that you can test new ideas by injecting noise into the simulation and vetting your approach before attempting it in HW. Maybe you have done that, but the shared model shows the controller only.

Regards,

Joel

[Bilal Ahmad]

Hi Joel,

Thank you for your reply. Yes you are right I am not using any filtering at the moment. I tried using FIR filter at the input but it kind of messed up my model and I guess I gave up on that idea too quickly. I will try it again and update it and if still does not work would love to get your insights.

Yes my converter is not this model but I have another model with converter in it where I tested it before using but still this jitter mucked up my actuall rectifier and blow up GaNs.