Skip to main content

Why switched capacitors; design for 100–400 rad/s peak

Short answer

On-chip resistors consume large area and have poor absolute tolerance; switched capacitors give time constants set by accurate capacitor ratios and clock. From the Bode plot, T(s)=(s+100)(s+400)/(s+200)². With fc=10 kHz, C=100 pF: section1 CR1=1 pF, CR2=2 pF; section2 CR1=4 pF, CR2=2 pF.

Why are resistors replaced by switched capacitors in modern IC technology? Design a switched capacitor filter to realize the magnitude response given by the plot below: [Figure: |T| vs ω: 0 dB below ω = 100, rising at 20 dB/decade to a 6 dB peak at ω = 200, then falling at -20 dB/decade back to 0 dB at ω = 400]

Why resistors are replaced by switched capacitors

  • A large resistor (hundreds of kΩ or more) takes a very large chip area, whereas a small capacitor plus two MOS switches takes very little area.
  • The absolute values of IC resistors and capacitors vary by ±20 % or more, so an RC time constant is inaccurate. In an SC circuit the time constant is τ=C/(CRfc)\tau = C/(C_Rf_c), where CC is the integrating capacitor. It depends only on a capacitor ratio (accurate to about 0.1 %) and on the clock frequency (crystal-controlled, so very accurate).
  • Capacitor ratios track well with temperature and ageing.
  • The response can be tuned by changing the clock frequency.
  • MOS technology makes good capacitors, switches and op-amps on one chip together with digital circuits.
         phi1        phi2
  V1 o---/ ----+---- / ---o V2
               |
              === C
               |
              GND

Two non-overlapping clock phases ϕ1\phi_1 and ϕ2\phi_2 at frequency fc=1/Tf_c = 1/T:

  • During ϕ1\phi_1 the capacitor connects to V1V_1: charge q1=CV1q_1 = CV_1.
  • During ϕ2\phi_2 it connects to V2V_2: charge q2=CV2q_2 = CV_2.

Charge moved from V1V_1 to V2V_2 in one clock period: Δq=C(V1−V2)\Delta q = C(V_1 - V_2). Average current:

I=ΔqT=Cfc(V1−V2)⇒Req=V1−V2I=1Cfc=TCI = \frac{\Delta q}{T} = C f_c (V_1 - V_2) \quad\Rightarrow\quad R_{eq} = \frac{V_1 - V_2}{I} = \frac{1}{C f_c} = \frac{T}{C}

This holds when fcf_c is much higher than the signal frequencies.

Transfer function from the Bode plot

Reading the straight-line plot: the slope changes from 0 to +20 dB/dec at ω = 100 (a zero), from +20 to −20 dB/dec at ω = 200 (a double pole), and from −20 back to 0 at ω = 400 (a zero). Check: the rise 20log⁡(200/100)20\log(200/100) = 6 dB matches the 6 dB peak.

T(s)=(s+100)(s+400)(s+200)2=s+100s+200⋅s+400s+200T(s) = \frac{(s+100)(s+400)}{(s+200)^2} = \frac{s+100}{s+200}\cdot\frac{s+400}{s+200}

DC gain =100×400/2002=1= 100 \times 400/200^2 = 1 (0 dB). (The exact peak of this function is about 1.9 dB; the 6 dB is the asymptotic value.)

Building block

A first-order section with an op-amp, input admittance Y1=sC1+1/R1Y_1 = sC_1 + 1/R_1 and feedback admittance Y2=sC2+1/R2Y_2 = sC_2 + 1/R_2:

T(s)=−Y1Y2=−C1C2⋅s+1/(R1C1)s+1/(R2C2)T(s) = -\frac{Y_1}{Y_2} = -\frac{C_1}{C_2}\cdot\frac{s + 1/(R_1C_1)}{s + 1/(R_2C_2)}

With C1=C2=CC_1 = C_2 = C the zero is at a=1/(R1C)a = 1/(R_1C) and the pole at b=1/(R2C)b = 1/(R_2C). Each resistor is then replaced by a switched capacitor CR=1/(Rfc)C_R = 1/(Rf_c), i.e.

CR1=aCfc,CR2=bCfcC_{R1} = \frac{aC}{f_c}, \qquad C_{R2} = \frac{bC}{f_c}
            +-------| C2 |-------+
            +----[SC: CR2]-------+
            |                    |
 Vi o--+--| C1 |--+              |
       |          |              |
       +-[SC: CR1]+-----(-)      |
                          A -----+----o Vo
                 GND-----(+)

Choices: fcf_c = 10 kHz, CC = 100 pF.

SectionZero aaPole bbC1=C2=CC_1 = C_2 = CCR1=aC/fcC_{R1} = aC/f_cCR2=bC/fcC_{R2} = bC/f_c
T1T_1100200100 pF1 pF2 pF
T2T_2400200100 pF4 pF2 pF

Answer: fcf_c = 10 kHz; section 1: C = 100 pF, CR1C_{R1} = 1 pF, CR2C_{R2} = 2 pF; section 2: C = 100 pF, CR1C_{R1} = 4 pF, CR2C_{R2} = 2 pF (two inverting stages give a positive overall gain).

✓ Answer checked by Claude Sonnet 5.5

All Filter Design chapter 9 questions →