Skip to main content

Sensitivity of -7; RLC lowpass ωc and Q

Short answer

S^x_y = -7 indicates a 1% rise in y produces roughly a 7% fall in x. For RLC lowpass: S^(omega_o)_L = S^(omega_o)_C = -1/2, S^(omega_o)_R = 0; S^Q_R = -1, S^Q_L = 1/2, S^Q_C = -1/2.

What do you understand when someone tells you that sensitivity of x with respect to y is -7? Compute the sensitivity expression for cutoff frequency and quality factor of RLC lowpass filter.

Meaning of Syx=−7S^x_y = -7

Δxx≈−7Δyy\frac{\Delta x}{x} \approx -7\frac{\Delta y}{y}

A 1% increase in yy causes about a 7% decrease in xx (opposite direction, seven times larger). xx is highly sensitive to yy, so yy must be held to a very tight tolerance.

Sensitivity of RLC low-pass filter

Take the series RLC low-pass circuit with the output across the capacitor:

Vin o--R--L--+--o Vout
             |
            === C
             |
GND o--------+--o
T(s)=1/(sC)R+sL+1/(sC)=1LCs2+RLs+1LCT(s) = \frac{1/(sC)}{R + sL + 1/(sC)} = \frac{\dfrac{1}{LC}}{s^2 + \dfrac{R}{L}s + \dfrac{1}{LC}}

Comparing with ωo2s2+(ωo/Q)s+ωo2\dfrac{\omega_o^2}{s^2 + (\omega_o/Q)s + \omega_o^2}:

ωo=1LC=L−1/2C−1/2,Q=ωoLR=1RLC=R−1L1/2C−1/2\omega_o = \frac{1}{\sqrt{LC}} = L^{-1/2}C^{-1/2}, \qquad Q = \frac{\omega_o L}{R} = \frac{1}{R}\sqrt{\frac{L}{C}} = R^{-1}L^{1/2}C^{-1/2}

Here ωo\omega_o is the cutoff (pole) frequency; for the Butterworth case Q=0.707Q = 0.707 it is also the half-power frequency.

Cutoff frequency:

SLωo=Lωo∂ωo∂L=Lωo(−ωo2L)=−12SCωo=−12,SRωo=0\begin{aligned} S^{\omega_o}_L &= \frac{L}{\omega_o}\frac{\partial\omega_o}{\partial L} = \frac{L}{\omega_o}\left(-\frac{\omega_o}{2L}\right) = -\frac12 \\ S^{\omega_o}_C &= -\frac12, \qquad S^{\omega_o}_R = 0 \end{aligned}

Quality factor:

SRQ=RQ∂Q∂R=RQ(−QR)=−1SLQ=+12,SCQ=−12\begin{aligned} S^Q_R &= \frac{R}{Q}\frac{\partial Q}{\partial R} = \frac{R}{Q}\left(-\frac{Q}{R}\right) = -1 \\ S^Q_L &= +\frac12, \qquad S^Q_C = -\frac12 \end{aligned}
ElementSωoS^{\omega_o}SQS^Q
R0−1
L−1/21/2
C−1/2−1/2

So a 1% rise in L or C lowers the cutoff by 0.5%, while R affects only Q: a 1% rise in R lowers Q by 1%.

✓ Answer checked by Claude Sonnet 5.5

All Filter Design chapter 7 questions →