SLVSHI9 Data sheet

SLVSHI9A March 2025 – September 2025 TPS7H5020-SEP , TPS7H5020-SP

PRODMIX

8.2.2.11 Frequency Compensation Component Selection

The poles and zeros of the flyback converter can be determined with the following equations. Note that the flyback converter also has a right-half-plane zero.

Equation 72.

f_{Z_E S R} = \frac{1 + D_{M A X}}{2 π \times C_{O U T} \times R_{E S R}}

Equation 73.

f_{Z_E S R} = \frac{1 + 0.35}{2 π \times 470 μ F \times 4 m Ω} = 114.3 k H z

Equation 74.

f_{P} = \frac{1}{2 π \times C_{O U T} \times \frac{V_{O U T}}{I_{O U T}}}

Equation 75.

f_{P} = \frac{1}{2 π \times 470 µ F \times \frac{5 V}{4 A}} = 270.9 H z

Equation 76.

f_{R H P Z} = \frac{\frac{V_{O U T}}{I_{O U T}} \times {(1 - D_{M A X})}^{2}}{2 π \times \frac{L_{P R I}}{{N_{P S}}^{2}} \times D_{M A X}}

Equation 77.

f_{R H P Z} = \frac{\frac{5 V}{4 A} \times {(1 - 0.35)}^{2}}{2 π \times \frac{30 µ H}{2^{2}} \times 0.35} = 32.0 k H z

Type 2A compensation network can be utilized in order to properly place the pole and zero of the error amplifier to achieve stability. Note that this compensation technique is for a flyback operating in continuous conduction mode. The crossover frequency is typically targeted to be anywhere from one-fourth of the RHP zero frequency to a full decade below. A crossover frequency of 4kHz was targeted for this design. The error amplifier network gain is set to achieve the target crossover frequency, which is dependent on R_COMP. The equation for R_COMP, as derived from Equation 21, Equation 23, and Equation 24, is shown in Equation 78. The calculation of K_FB is as shown in Equation 25.

Equation 78.

R_{C O M P} = \frac{2 π \times f_{C} \times C_{O U T} \times A_{C S} \times R_{C S}}{(1 - D_{M A X}) \times N_{P S} \times K_{F B} \times g_{m}}

Equation 79.

R_{C O M P} = \frac{2 π \times 4 k H z \times 470 μ F \times 1 \times 0.1 Ω}{(1 - 0.35) \times 2 \times 0.12 \times 1750 \frac{μ A}{V}} = 4326.88 Ω

The error amplifer zero is set to one-tenth of the crossover frequency, which allows for selecting the value of C_COMP.

Equation 80.

C_{C O M P} = \frac{1}{2 π \times (0.1 \times f_{C}) \times R_{C O M P}}

Equation 81.

C_{C O M P} = \frac{1}{2 π \times (0.1 \times 4 k H z) \times 4.32 k Ω} = 91.96 n F

Lastly, the high frequency pole is set to the lower of the ESR zero and the RHP zero. In this specific case, the RHP zero is lower.

Equation 82.

C_{H F} = \frac{1}{2 π \times f_{R H P Z} \times R_{C O M P}}

Equation 83.

C_{H F} = \frac{1}{2 π \times 32 k H z \times 4.32 k Ω} = 1.15 n F

Using standard component values, the initial values of R_COMP, C_COMP, and C_HF selected were 4.32 kΩ, 100nF, and 1nF, respectively. It is important to note that these calculated values provide a starting point. The frequency compensation is often tuned during both simulation and testing to settle on the final compensation values of the design.