Boost Controller#

../_images/boost-converter-schematic.png

Averaged Inductor Current and Capacitor Voltage#

Volt-Seconds

\[\begin{align*} \langle v_{L}(t)\rangle = L\frac{d}{dt} \langle i(t)\rangle &= \bigg[\bigg(\langle{v_g}(t)\rangle-\langle i(t)\rangle R_{on} \bigg)d(t) + \bigg( \langle{v_g}(t)\rangle-\langle i(t)\rangle R_{on} - \langle v(t)\rangle \bigg)d'(t) \bigg] \\[0.5em] \langle v_{L}(t)\rangle = L\frac{d}{dt} \langle i(t)\rangle &= \bigg[\langle{v_g}(t)\rangle-\langle i(t)\rangle R_{on} - \langle v(t)\rangle d'(t) \bigg] \\[0.5em] \frac{d}{dt} \langle i(t)\rangle &= \begin{matrix} \dfrac{1}{L} \\[1em] \end{matrix} \bigg[\langle{v_g}(t)\rangle-\langle i(t)\rangle R_{on} - \langle v(t)\rangle d'(t) \bigg] \\[0.5em] \end{align*}\]

Charge Balance

\[\begin{align*} \langle i_C (t) \rangle = C\frac{d}{dt} \langle v(t)\rangle &= \bigg[\bigg(\dfrac{-\langle v(t)\rangle}{R}\bigg)d(t) +\bigg(\langle i(t)\rangle -\frac{\langle v(t)\rangle}{R} \bigg)d'(t)\bigg] \\[1em] \langle i_C (t) \rangle = C\frac{d}{dt} \langle v(t)\rangle &= \bigg[\dfrac{-\langle v(t)\rangle}{R} + \langle i(t)\rangle d'(t) \bigg] \\[1em] \frac{d}{dt} \langle v(t)\rangle &= \begin{matrix} \dfrac{1}{C} \\[1em] \end{matrix} \bigg[\dfrac{-\langle v(t)\rangle}{R} + \langle i(t)\rangle d'(t) \bigg] \\[1em] \end{align*}\]

Differential Solution

x =

(2)#\[\begin{bmatrix} \langle i(t)\rangle \\ \langle v(t)\rangle \end{bmatrix}\]

u =

(3)#\[\begin{bmatrix} d(t) \\ \langle v_g(t)\rangle \end{bmatrix}\]
\[\begin{align*} \dot{x} &= f(x(t),u(t)) = \dfrac{d}{dt} \begin{bmatrix} \langle i(t)\rangle \\[1em] \langle v(t)\rangle \end{bmatrix} = \begin{matrix} \dfrac{1}{L} \\[0.5em] \dfrac{1}{C} \end{matrix} \begin{bmatrix} \langle{v_g}(t)\rangle-\langle i(t)\rangle R_{on} - \langle v(t)\rangle d'(t) \\[1em] \dfrac{-\langle v(t)\rangle}{R} + \langle i(t)\rangle d'(t)s \end{bmatrix} \\[0.5em] \end{align*}\]

Linearized Small-Signal Model#

Derive a linearized small-signal model

\[\begin{align*} \dot{\hat{x}}(t) &\approx A \hat{x}(t) + B \hat{u}(t) \\[0.5em] \hat{y}(t) &= C \hat{x}(t) + E \hat{u}(t) \end{align*}\]

where

(4)#\[\begin{matrix} x(t) = \hat{x}(t) + X && X = \begin{bmatrix} I \\ V \end{bmatrix} \\[0.5em] u(t) = \hat{u}(t) + U && U = \begin{bmatrix} D \\ V_g \end{bmatrix} \\[0.5em] \end{matrix}\]

Volt-Seconds (Large Signal)

\[\begin{align*} \langle v_{L}(t)\rangle &= L\frac{d}{dt} \langle i(t)\rangle = \bigg[\langle{v_g}(t)\rangle-\langle i(t)\rangle R_{on} - \langle v(t)\rangle d'(t) \bigg] = 0\\[0.5em] 0 &= V_g - I R_{on} - VD' \\[0.5em] VD' &= V_g - I R_{on} \\[0.5em] V &= \dfrac{V_g - I R_{on}}{D'} \bigg|_{I = \dfrac{V}{D'R} } \\[0.5em] V &= \dfrac{V_g - \dfrac{V}{D'R} R_{on}}{D'} \\[0.5em] V &= \dfrac{V_g}{D'}\cdot \dfrac{1}{1+\dfrac{ R_{on}}{D'^2 R}} \\[0.5em] \end{align*}\]

Charge Balance (Large Signal)

\[\begin{align*} \langle i_C (t) \rangle &= C\frac{d}{dt} \langle v(t)\rangle = \bigg[\dfrac{-\langle v(t)\rangle}{R} + \langle i(t)\rangle d'(t) \bigg] = 0\\[1em] 0 &=\dfrac{-V}{R} + I D' \\[1em] I &= \dfrac{V}{D'R} \\[1em] I &= \dfrac{1}{D'R}\cdot \frac{V_g}{D'}\cdot \dfrac{1}{1+\dfrac{ R_{on}}{D'^2 R}} \\[1em] I &= \frac{V_g}{D'^2 R}\cdot \dfrac{1}{1+\dfrac{ R_{on}}{D'^2 R}} \\[1em] \end{align*}\]

Find A, B, C, D

\[\begin{align*} A = \dfrac{d}{d\hat{x}(t)} f(x(t),u(t))\bigg|_{x=X,u=U} &= \begin{bmatrix} \ \ \dfrac{\partial f_1}{\partial x_1} & \dfrac{\partial f_1}{\partial x_2} \ \ \\[1em] \ \ \dfrac{\partial f_2}{\partial x_1} & \dfrac{\partial f_2}{\partial x_2} \ \ \end{bmatrix}\bigg|_{x,u} = \begin{bmatrix} \dfrac{\partial f_1}{\partial \langle \hat{i}(t)\rangle} & \dfrac{\partial f_1}{\partial \langle \hat{v}(t)\rangle} \\[1em] \dfrac{\partial f_2}{\partial \langle \hat{i}(t)\rangle} & \dfrac{\partial f_2}{\partial \langle \hat{v}(t)\rangle} \end{bmatrix}\bigg|_{x,u} \\[1em] A &= \begin{bmatrix} \dfrac{-R_{on}}{L} & \dfrac{-D'}{L} \\[1em] \dfrac{D'}{C} & \dfrac{-1}{RC} \end{bmatrix} \\[1em] \end{align*}\]
\[\begin{align*} B = \dfrac{d}{d\hat{u}(t)} f(x(t),u(t))\bigg|_{x=X,u=U} &= \begin{bmatrix} \dfrac{\partial f_1}{\partial u_1} & \dfrac{\partial f_1}{\partial u_2} \\[1em] \dfrac{\partial f_2}{\partial u_1} & \dfrac{\partial f_2}{\partial u_2} \end{bmatrix}\bigg|_{x,u} = \begin{bmatrix} \dfrac{\partial f_1}{\partial \hat{d'}(t)} & \dfrac{\partial f_1}{\partial \langle \hat{v_g}(t)\rangle} \\[1em] \dfrac{\partial f_2}{\partial \hat{d'}(t)} & \dfrac{\partial f_2}{\partial \langle \hat{v_g}(t)\rangle} \end{bmatrix}\bigg|_{x,u} \\[1em] B &= \begin{bmatrix} \dfrac{V}{L} & \dfrac{1}{L} \\[1em] \dfrac{-I}{C} & 0 \end{bmatrix} = \begin{bmatrix} \dfrac{V_g}{D'L}\cdot \dfrac{1}{1+\dfrac{ R_{on}}{D'^2 R}} & \dfrac{1}{L} \\[1.5em] \dfrac{-V_g}{D'^2 RC}\cdot \dfrac{1}{1+\dfrac{ R_{on}}{D'^2 R}} & 0 \end{bmatrix} \\[1em] \end{align*}\]

C =

(5)#\[\begin{bmatrix} 0 & 1 \\ 1 & 0 \end{bmatrix}\]

E =

(6)#\[\begin{bmatrix} 0 & 0 \\ 0 & 0 \end{bmatrix}\]

System Outputs in Frequency Domain#

\[\begin{align*} \dot{\hat{x}}(t) &= A \hat{x}(t) + B \hat{u}(t) \\[0.5em] \dot{\hat{y}}(t) &= C \hat{x}(t) + D \hat{u}(t) \\[0.5em] \text{Laplace Transform } &\\[0.5em] s\hat{x}(s) &= A \hat{x}(s) + B \hat{u}(s) \\[0.5em] sI\hat{x}(s) - A \hat{x}(s) &= B \hat{u}(s) \\[0.5em] \hat{x}(s) &= \left(sI-A\right)^{-1}B \hat{u}(s) \\[0.5em] \hat{y}(s) &= \bigg(C(sI-A)^{-1}B +E \bigg)\hat{u}(s) = G(s) \hat{u}(s) \end{align*}\]

Solver

G(s) =

(7)#\[\begin{bmatrix} G_{id}(s) & G_{ig}(s) \\[1em] G_{vd}(s) & G_{vg}(s) \end{bmatrix}\]
(8)#\[\begin{bmatrix} \dfrac{D'RV_g}{D'^2R + R_{on}}\dfrac{(CRs + 2)}{(LCRs^2 + RR_{on}Cs + Ls + D'^2R +R_{on})} & \dfrac{RCs + 1}{LCRs^2 + RR_{on}Cs + Ls + D'^2R +R_{on}} \\[1em] \dfrac{RV_g}{D'^2R + R_{on}}\dfrac{(D'^2R - R_{on}-Ls)}{(LCRs^2 + RR_{on}Cs + Ls + D'^2R +R_{on})} & \dfrac{D'R}{LCRs^2 + RR_{on}Cs + Ls + D'^2R +R_{on}} \end{bmatrix}\]

Transfer Function

Description

\(G_{id}(s)\)

Small signal output current to duty ratio

\(G_{ig}(s)\)

Small signal output current to supply voltage

\(G_{vd}(s)\)

Small signal output voltage to duty ratio

\(G_{vg}(s)\)

Small signal output voltage to supply voltage