This PLECS demo model shows a dual active bridge (DAB) converter. The thermal behavior of the Wolfspeed C3M0030090K and C3M0065090D Silicon Carbide MOSFETs is included for this topology using the PLECS Thermal domain.
A dual active bridge is a bidirectional DC-DC converter with identical primary and secondary side full-bridges, a high frequency transformer, an energy transfer inductor and DC-link capacitors. Energy transfer inductance in the model refers to the leakage inductance of the transformer plus any necessary external energy transfer inductance.
The two legs of both full-bridges are driven with complimentary square-wave pulses. Power flow in the dual active bridge can be directed by phase-shifting the pulses of one bridge with respect to the other using phase shift modulation. The control directs power between the two DC buses such that the leading bridge delivers power to the lagging bridge. The applied square waves to the bridges create a voltage differential across the energy transfer inductance and direct its stored energy.
In ideal cases with dual active bridge converters, zero voltage switching (ZVS) can be realized when the voltage transfer ratio (M) across the transformer is equal to 1:
M = Vout / (n*Vin)
where, n is the transformer turn ratio, Vout is the output voltage and Vin is the input voltage.
In non-ideal cases, ZVS depends on the resonant relationship between the output capacitance across each device and the equivalent inductance of the circuit during different switching intervals. During switching events, the current through one of the complimentary devices is interrupted, but due to the energy transfer inductance, current is supplied through the output capacitor and forced through the anti-parallel diode of the device.
Each switch is on for 50% of its respective switching period. The switch pairs in the two bridges all have the same switching period but are operated such that between each bridge a phase shift is introduced that varies based on the modulation derived from feedback measurements. An output voltage error signal is generated based on a set point value and this is fed through a digital PI regulator to generate the phase shift ratio for the PWM modulator.
The transformer model can be configured as an ideal transformer for speeding up the simulation or as a more detailed transformer model that includes saturation behavior. For the detailed version, Payton Planar Magnetics’ T250-4-16 transformer is modeled using the PLECS magnetic domain. The parameters of the magnetic model are directly related to the geometry and material characteristics of the core, which in most cases can be obtained from datasheets. The transformer uses industry standard E43/28 core and TP4A ferrite material from TDG.
This model is available in the PLECS Demo Model library provided in both versions of PLECS.