Peak current control of DAB on RT-Box

RT Box
1

Hello,

I have a problem with the waveforms of the DAB converter generated by the RTbox 1. In my HIL simulation, I receive control PWM (100 kHz) signals from the LAUNCHXL-F280049C development kit using the PWM capture block. I feed the measured signals to the ADC pins of the kit via the Analog out block. In the case of control using single-phase shift, the control worked. When I outputted the current waveform through the inductor to the analog output and displayed it on the oscilloscope, I found that the current through the inductor of the DAB converter is delayed by 2.76 microseconds (see the oscilloscope screenshot). For this reason, I could not test the control using the peak current technique, since the PWM signals of the secondary full bridge depend on the current waveform through the inductor. I want to ask if this is caused by a bad simulation/coder setting or is the problem elsewhere? Thank you for your answer.

schematic
schematic1920×1007 168 KB

DS0003
DS0003800×480 12.7 KB

2

Hi matusskorvaga,

From the schematic, the circuit is divided into two segments, each solved by a separate solver:

  • Nanostep solver

    • This solver runs on the FPGA and handles the DAB circuit at 7.5 ns on the RTBox1.
    • The voltage at the terminals of the DAB nanostep component is sampled and fed into the equations associated with the DAB.
    • The calculated currents are then fed back into the nodes.

  • CPU solver

    • While the internals of the DAB are solved by the nanostep solver, the rest of the circuit outside the DAB is solved by the CPU solver, which runs at the discretization step size defined in the Coder Options > Scheduling.
    • In your simulation setup, the analog outputs are updated at the CPU rate.
    • In addition to the slower CPU update rate, the analog outputs also have an I/O latency of approximately three times the CPU update interval. These additional delays are typically acceptable for applications using average current mode control or voltage mode control.
    • However, for peak current mode control, these delays can significantly impact controller performance and are not well suited. This limitation is one reason the RT Box4 was introduced, which features six fast DACs enabling users to test control algorithms like peak current mode controllers. There is still some latency in the fast DACs, but it is less than 200 ns.

Hope this helps!

3

Hi Ahmed,

thank you for your answer. At least now I know how to defend this issue in my thesis. :slight_smile:

Kind regards,
Matúš Škorvaga

1 Like