Frequency Response; Filters

Frequency response: Passive Filters Let’s consider again the RC filter shown on Figure 1 R + 1 jC ω Vs Vc - Figure 1 When the output is taken across the capacitor the magnitude of the transfer function is H (ω ) = By letting ω0 = 1 1 + (ω RC ) 2 (1.1) 1 the transfer function becomes RC H (ω ) = 1 ⎛ω ⎞ 1+ ⎜ ⎟ ⎝ ω0 ⎠ 2 (1.2) The overall characteristics of the transfer function may be determined by considering what happens at ω = 0 and at ω → ∞ . ω = 0, ω → ∞, H (ω ) = 1 H (ω ) → 0 It is also interesting to look at the value of the transfer function at the frequency ω0 For ω = ω0 the magnitude of the transfer function becomes 1 H (ω ) ω =ω = 0 2 (1.3) 1 The frequency ω0 is called the corner, cutoff, or the ½ power frequency. Also, by considering the definition of the dB we have H (ω ) dB = 20 log ( H (ω ) ) (1.4) H (ω ) dB = −3dB (1.5) Which at ω = ω0 gives And so the frequency ω0 is also called the 3dB frequency. For our example RC circuit with R=10kΩ and C=47nF the Bode plot of the transfer function is shown on Figure 2. In this case the corner frequency equals 2,127 rad/sec and it is indicated on Figure 2. Figure 2 If the output is taken across the resistor, the magnitude of the transfer function becomes H (ω ) = ω ω0 ⎛ω ⎞ 1+ ⎜ ⎟ ⎝ ω0 ⎠ 2 (1.6) In this case the limits are ω = 0, ω → ∞, H (ω ) = 0 H (ω ) → 1 2 The plot of this transfer function is shown on Figure 3 Figure 3 Filtering and Filters By investigating Figure 2 and Figure 3 we see that the magnitude of the output signal is a very strong function of frequency. The attenuation of the signal amplitude with frequency is also called filtering and the circuits that perform this operation are called filters. In general we say that filters are circuits which allow a specific range of frequencies to be passed (or rejected) as they are transmitted from an ac source to a load. Schematically the system is shown on Figure 4. AC source Filter Load Figure 4 Filters in general fall into one of the following categories: • • • • Low Pass: passes low frequencies (that is signals with low frequencies) and attenuates high frequencies High Pass: passes high frequencies (that is signals with high frequencies) and attenuates low frequencies Band Pass: passes frequencies in a certain range and attenuates frequencies outside this range Band Stop: attenuates frequencies within a certain range and passes frequencies outside this range. 3 For the RC circuit, when the output is taken across the capacitor we obtain a Low Pass filter. By contrast when the output is taken across the resistor we have a High Pass filter. The corresponding plots are shown on Figure 5. Low Pass Filter High Pass Filter Figure 5 The transition frequency which indicates that range of frequencies that are allowed and those that are rejected is given by the cutoff frequency ω0 . In practical situations the design of a High pass or Low pass filter is guided by the value of the cutoff or corner frequency ω0 . For our example RC circuit, with R=10kΩ and C=47nF, the cutoff frequency is 338 Hz. We may obtain a band pass filter by combining a low pas and a high pass filter. Consider the arrangement shown on Figure 6. a R Vs C C R Vo b Figure 6 The transfer function may be calculated very easily if we first consider the equivalent circuit to the left of a-b as shown on Figure 7 ZC a ZTh VTh R Vo b Figure 7 4 The voltage VTh is VTh = Vs ZC ZC + R (1.7) And R ZC R + ZC (1.8) Vo ZC R = Vs ( R + ZC ) 2 + R ZC (1.9) ZTh = The transfer function now becomes H (ω ) = And upon simplification the magnitude becomes R H (ω ) = 1 ⎞ ⎛ (3R ) + ⎜ R 2ωC − ωC ⎟⎠ ⎝ By looking at low and high values for ω we have 2 (1.10) 2 ω = 0, ω → ∞, H (ω ) = 0 H (ω ) → 0 1 1 the magnitude becomes H (ω ) = RC 3 The plot of Equation (1.10) is shown on Figure 8. This has the form of a band pass filter although the attenuation at the frequency 1/ RC is not desirable. We will next look at ways to improve this type of filter by considering the RLC circuit. Also we notice that for ω = Figure 8 5 Now let’s continue by exploring the frequency response of RLC circuits. L R Vs + Vc - C The magnitude of the transfer function when the output is taken across the capacitor is Vc = H (ω ) = Vs 1 (1 − ω LC ) + (ω RC ) 2 2 2 (1.11) Here again let’s look at the behavior of the transfer function, H (ω ) , for low and high frequencies. ω = 0, H (ω ) = 1 (1.12) ω → ∞, H (ω ) → 0 There is another frequency that has a significant effect on the behavior of H (ω ) . This is the frequency ω0 at which 1 = ω0 2 LC ⇒ ω0 = 1 LC (1.13) At this frequency the magnitude of the transfer function becomes H (ω0 ) = LC RC (1.14) From the scaling given by Equation (1.12) we see that this circuit corresponds to a low pass filter. Indeed Figure 9 shows the plot for H (ω ) for R=2kΩ, L=47mH and C=47nF (the values we also used in the laboratory). 6 Figure 9 The cutoff frequency in this case is given by the frequency ω0 = 1 = 21, 276 rad/sec . LC This is also indicated on the plot of Figure 9. The magnitude H (ω ) at ω = ω0 is inversely proportional to the resistor R . So let’s now investigate the behavior of the transfer function with R. Figure 10 shows the R dependence of the transfer function. We have plotted Figure 10 7 The peak observed at the frequency ω0 is called the resonance peak and the frequency ω0 is also referred to as the resonance frequency. A ω0 the transfer function becomes H (ω ) = Vc 1 = = Vs j RC LC 1 ( RC ) LC 2 e − π2 (1.15) Which shows that there is 90 degree phase difference between Vc and Vs. The current flowing through the capacitor is I = jω CVc (1.16) And thus the phase difference between the current I and the source voltage Vs is zero. Resonance if defined as the condition at which the voltage and the current at the input of a circuit is in phase. 8 If we take the output across the inductor the magnitude of the transfer function is VL = H (ω ) = Vs ω 2 LC (1 − ω 2 LC ) + (ω RC ) 2 2 (1.17) In this case, consideration of the frequency limits gives ω = 0, H (ω ) = 0 (1.18) ω → ∞, H (ω ) → 1 And it corresponds to a high pass filter. Figure 11 shows the plot of H (ω ) for various values of R. Here again we observe the resonance phenomenon at ω0 = 1 . LC Figure 11 9 If we take the voltage across the resistor the transfer function becomes VR = H (ω ) = Vs ω RC (1 − ω LC ) + (ω RC ) 2 2 2 (1.19) In this case, consideration of the frequency limits gives ω = 0, H (ω ) = 0 (1.20) ω → ∞, H (ω ) → 0 And it corresponds to a band pass filter. Figure 12 shows the plot H (ω ) for various values of R. Here again we observe the peak at ω0 = 1 . In this case however, the magnitude of the transfer function does not LC exceed 1. Figure 12 Figure 12 also shows that as the resistance increases, the peak becomes broader. This is a direct consequence of the increases damping provided by the resistor. We will analyze this phenomenon in detail next class. 10 Now we consider the voltage across the capacitor and inductor combination. In this case the magnitude of the transfer function is 1 − ω 2 LC VR = H (ω ) = Vs (1 − ω LC ) + (ω RC ) 2 2 2 (1.21) In this case, consideration of the frequency limits gives ω = 0, H (ω ) = 1 (1.22) ω → ∞, H (ω ) → 1 And it corresponds to a band reject filter. Figure 13 shows the plot H (ω ) for various values of R. Here again we observe the peak 1 . Here again we see the effect that R has on the details of the filter. We will LC investigate this phenomenon next class. at ω0 = Figure 13 11