TLV713P-Q1: TIDA-01604: how to get D

Hello Zhonghui,

The D in Equation 8 of the TIDA-01604 Design Guide (https://www.ti.com/lit/pdf/tidue54) refers to the instantaneous duty cycle of each switching cycle of the PFC boost converter. As you mentioned, Vout in the equation D=1-Vin/Vout is a fixed value, but Vin in this equation is the instantaneous amplitude of the input voltage.  It is not an RMS value.  The value of Vin changes with each switching cycle as the sine wave rises and falls. 

Equation 12 can be rearranged to solve for L and appears to be independent of D.  Of course, L then depends on the value of Iripple, which is determined by Equation 13.

The reason Equation 12 appears to be independent of D, is because this is a special case where D=0.5.  D = 0.5 when Vin = Vout/2.  Equation 12 can be derived from Equation 10 by substituting 1/2 for D and Vout/2 for Vin.  In a boost converter, Iripple is maximum at the point where D= 0.5.  In TIDA-01604, the design target for Iripple is 10% of the peak current at high line (Equation 13). Note that the Vin_max here is the peak amplitude, not the RMS value. 

To be honest, I'm not sure if Equation 13 is entirely correct.  It uses Pout_max, which is the entire 6.6kW, but the L to be calculated is one of 3 phases, so it's value should be based on processing 1/3 of the full power.  Also the root-2 factor does not make sense to me. It would convert Vin_max into an rms value and I don't think Iripple should be based on an rms value. Maybe I am missing something. 

But in any case, the maximum peak-to-peak target value of Iripple used in Equation 12 should be low enough to avoid DCM until very light loads.  Once the boost PFC begins to operate in DCM, harmonic distortion will go up significantly.

(By the way, this D has nothing to do with UCC28700-Q1 either.) 

Regards,
Ulrich