Hi!
I designed a 300V DC out, 60kW (interrupted) SIC based power supply, the PI controller is based on the TI F28M35H52C1 microcontroller. It is a simple BUCK converter (a bunch of design consideration, and other constrains where taken into account to choose a simple BUCK at the end). The switching frequency is 100kHz.
I had really good results with the PS in the testing phase, with resistive load transients (up to 67kW for 6sec), but finally we got to a point, where we connected it to the system it should deliver power to. This system is a complex electrical equipment with various loads present.
There are two motors driven by an inverter in this system, one of them is a pump, the other one is a 3ph motor, both around 200-300W. The PS starts fine, powers the system on preload which is about 1kW. As soon as I turn on one of the motors, it just goes crazy, the PI controller acts really strange and the whole power output becomes very unstable (turn on of the 3ph ~300W motor):
I tried to adjust the PI controller parameters (I don't use the derivative gain yet, maybe I should?!), without any luck. Today I was experimenting with hardware problems, maybe some sort of ringing in the feedback or on the 24V DC power supply for the gate driver ant the controller, but I could not get any significant improvements.
What you can see on the above picture is the following:
- 300V DC output is stable on preload
- 3ph motor started, the inverter increases the frequency of the motor in 2sec intervals, this is pretty much visible on the PID and V OUT signals
- 3PH motor revs up, and on required speed, in this state the 300VDC is stable again
If I slow down the controller to a very concerning amount, which makes no sense at all, I can get a result like this at a certain point, slowing it further messes up everything again:

But this is extremely slow reaction time for this controller, imagine putting a 50kW load on this (I decreased the output voltage to 285V so high fluctuations of the output voltage don't ruin the subsystem)...
So I'm quite clueless at this point, if anybody has some advise, please let me know!
Cheers
I designed a 300V DC out, 60kW (interrupted) SIC based power supply, the PI controller is based on the TI F28M35H52C1 microcontroller. It is a simple BUCK converter (a bunch of design consideration, and other constrains where taken into account to choose a simple BUCK at the end). The switching frequency is 100kHz.
I had really good results with the PS in the testing phase, with resistive load transients (up to 67kW for 6sec), but finally we got to a point, where we connected it to the system it should deliver power to. This system is a complex electrical equipment with various loads present.
There are two motors driven by an inverter in this system, one of them is a pump, the other one is a 3ph motor, both around 200-300W. The PS starts fine, powers the system on preload which is about 1kW. As soon as I turn on one of the motors, it just goes crazy, the PI controller acts really strange and the whole power output becomes very unstable (turn on of the 3ph ~300W motor):

I tried to adjust the PI controller parameters (I don't use the derivative gain yet, maybe I should?!), without any luck. Today I was experimenting with hardware problems, maybe some sort of ringing in the feedback or on the 24V DC power supply for the gate driver ant the controller, but I could not get any significant improvements.
What you can see on the above picture is the following:
- 300V DC output is stable on preload
- 3ph motor started, the inverter increases the frequency of the motor in 2sec intervals, this is pretty much visible on the PID and V OUT signals
- 3PH motor revs up, and on required speed, in this state the 300VDC is stable again
If I slow down the controller to a very concerning amount, which makes no sense at all, I can get a result like this at a certain point, slowing it further messes up everything again:

But this is extremely slow reaction time for this controller, imagine putting a 50kW load on this (I decreased the output voltage to 285V so high fluctuations of the output voltage don't ruin the subsystem)...
So I'm quite clueless at this point, if anybody has some advise, please let me know!
Cheers