Document

advertisement

Filters

Passive Filters

 A filter is a circuit that is designed to pass signals with desired frequencies and reject or attenuate others.

 A filter is a passive filter if it consists of only passive elements R , L , and C .

 A lowpass filter passes low frequencies and stops high frequencies.

 A highpass filter passes high frequencies and rejects low .

 A bandpass filter passes frequencies within a frequency band and blocks or attenuates frequencies outside the band.

 A bandstop (band reject) filter passes frequencies outside a frequency band and blocks or attenuates frequencies within the band.

1

Filter Responses

Ideal Filter Classification

Low pass f filter

High pass f filter

Band pass f filter

Band reject f filter

2

Passive (1 st order) Lowpass Filters

RL circuit

(  

R

 

 j

R L

  R L

1

1

Plot |H(j  )| versus  :

Ideal

Actual

H j  

1

2

(  ) max

R

L

H

LPF

( j  

1

1 j c

3

Passive (1 st order) Lowpass Filters

RC circuit

(  

R

1 ( 

 

)

1 ( j C ) 1

Plot |H(j  )| versus  :

1

Ideal

Actual

H j  

1

2

(  ) max

1

RC

H

LPF

( j  

1

1 j c

4

Passive (1 st order) Highpass Filters

(  

R j L

1

 c

R

L

1 j   c j c

 c

1

RC

(  

R

R

 

1 ( j C ) 1

1 j   c j c

5

Example

Effects of Loading on Filter Performance

H

)

 sL

R sL

Ks

/

R

, K

L

 1

H

H

)

Z eq , Z eq

 ||

L eq

)

 s

R s

 sLR

L sL R

L

 sLR

L sL R

L

R

L

R

L

R

L

L

L

 sLR

L

( 

L

)  sLR

L

Ks

 

,

K 

R

L

R R

L

 1;

) )

 

)

6

Example cont`d

Let R=R

L

= 500  and L =5.3mH

Then f c

(unloaded) = 15kHz and f c

(loaded) = f c`

= 7.5 kHz

Effects of Loading on Filter Performance

 Changes the cutoff frequency

 Changes the gain in the passband

For R

L

= R

7

Passive 2 nd order Lowpass Filter

1 sC

  1 sC

1 

(1 LC ) s 2  ( )  (1 LC )

 2 o s 2 s 2 o

 H

LPF

1

1 LC

8

Passive Bandpass Filters

Series RLC circuit

H ( s ) 

R  sL

R

 1 s 2 sC

 ( R

( R

L )

L s

)

 s

( 1

RsC

LC )

RsC s 2 LC s 2

 1

 s

 s   o

2

 H

BPF

( s )

B

1 LC

R L

9

Passive Bandpass Filters

Parallel RLC circuit

Z eq , Z eq

 sL ||1 sC 

L C sL  1 sC eq

R sL

L C

 1 sC

 L C sL  1 sC

 s s 2   s   2 o

L C

( 1 sC ) L C s 2

(1 )

 (1 )  (1 LC )

 H

BPF

( ) 1

B    1

LC

RC

10

Passive Bandpass Filters

Frequency response characteristics:

Only 2 of the 5 characteristics of a bandpass filter can be specified independently. The others are related as follows:

    c 2

B c

Q c 1

11

Example

Design a bandpass filter with a center frequency of 2 kHz and a bandwidth of 500 Hz. Use a

250  resistor.

B 

1

RC

 C 

1

 R

1

2 500Hz 250  o

1

LC

 L

 1.27 F

1

2 C

 4.97 mH

Note: Q

 3

 4

500

The Q is so low in order to achieve the specified bandwidth.

1

12

Passive (Bandstop) Bandreject Filters

Series RLC circuit

 sL 

R sL

1

 sC

1 sC

 1

 1

 s 2  ( s 2  (1 LC )

)  (1 LC )

 s 2 s 2   2 o

    2 o

 H

BRF

( )

1 LC

B    R L

13

Passive (Bandstop) Bandreject Filters

Parallel RLC circuit

R

, Z eq

 sL ||1 sC 

L C sL  1 sC eq

R 

R

L C sL  1 sC

 s 2   2 o s 2 s 2 o

(

(

 1

 1 sC

 H

BRF

( )

) sC

)

L C

 s 2 s 2  (1 LC )

 (1 )  (1 LC )

1

B    1

LC

RC

14

Example 1

Determine what type of filter is shown. Calculate the corner or cutoff frequency.

Take R = 2 k  , L = 2 H, and C = 2 μ F.

15

Example 1 cont’d

R = 2 k  , L = 2 H, C = 2 μ F

Note:

Q p

 R

1

LC

C

L

1

 6

 500 rad/s

3

2 10  6

2

 2

16

Example 2

For the circuit shown, obtain the transfer function V o( ω )/ V i( ω ) . Identify the type of filter the circuit represents and determine the corner frequency. Take R

1

= 100 

= R

2

, L = 2 mH.

This is a highpass filter.

17

Example 2 cont’d

18

Example 3

If this bandstop filter is to reject a 200-Hz sinusoid while passing other frequencies, calculate the values of L and C .

Take R = 150  and the bandwidth as 100 Hz.

19

Example 4

Design this bandpass filter with a lower cutoff frequency of 20.1 kHz and an upper cutoff frequency of 20.3 kHz. Take R = 20 k  . Calculate

L , C , and Q .

 Too large !!

20

Take R = 20  .

Example 4 cont’d

Q

L

B o  101

R

B

20 

C 

1

 o

QR

 

 15.916 mH

1

  

 3.9 nF

21

Example 5

Use a 500 nF capacitor to design this band-reject filter. The filter has a center frequency of 4 kHz and a quality factor of 50.

1

LC

Q   o

RC

 L 

1

 2

0

C

 R 

Q

 o

C

1

   3

 2

 

50

      9

 9

 3.166 mH

 3.98 k 

22

Active Filters

Passive Filter Disadvantages:

Maximum gain in the pass band is 1 – no amplification in the pass band!

Adding a load to a passive filter changes the filter’s characteristics, like the cutoff frequency.

Passive band pass and band reject filters require the use of inductors, which are large, costly, and can introduce stray electromagnetic field effects.

No obvious method to make the filter characteristic more ideal.

23

Active Filters

Active filters

 Based on circuits with op amps, which are active devices, since they require an external power supply (  V

CC

).

 Eliminate all of the disadvantages of passive filters:

 Can create a pass band gain > 1;

 Can add loads without changing the filter characteristics;

 All four types of filters can be created using op amps, resistors, and capacitors – no inductors;

 Can cascade simple (first- and second-order) filters to create higher order filters that are more nearly ideal.

24

First-Order Active Lowpass Filter

Find the transfer function:

  o i

 

Z f  

Z i

 

R

1

1 

2

1

  sC

R

1

|| R

2

R

2

R R R R sC

 

R R s 

1

1



1 R C

R C

 s

K  c

  c

1

K  

R

2

R

1

25

Example

Design an active low pass filter with a gain of -5 and a cutoff frequency of 1 kHz. Use a 10 nF capacitor.

1

R

2

1

 9

  15.916 k 

R

1

R

5

2 

15,916

5

 3.184 k 

26

First-Order Active Highpass Filter

Find the transfer function:

H ( s )

V o

( s )

V i

( s )

 

R

1

R

2 sC sC  1

Z

Z i

 f

 s

 

R

2

1

R

2

R

1

R

1

 s

1

R

1

C sC

For K = -1

1

K  

R

2

R

1

27

Example 1

Design an active high pass filter with a gain of -10 and a cutoff frequency of 500 rad/s. Use a 0.1  F capacitor.

1

 500 rad/s

R

1

1

R

2

 10 R

1

 200k 

 20k 

28

Example 2

Design a first-order highpass op amp filter whose transfer function is given by

    4000

Use capacitances of any value. But you only have 10 k  resistors available

- no inductors!

The transfer function of an inverting op amp is

Thus,

 

1 s

 

 3 

 3

  s

 

 

3

 3

1

 9 s

 3 

 

 3

Therefore, f

( ) R 10k  i

( )  

1 sC where R  10k  and C  25nF s i f

7

 

3

 3

1

 8 s

29

First-Order Active Bandpass Filter

+ + +

=

30

First-Order Active Bandpass Filter

Gives  c2

Gives  c1

Gives K

31

First-Order Active Bandpass Filter

H

BPF

( )  H

LPF

( )

 s

 c 2

  c 2

 s

 s

  c 1

 R

R

 ( R f

R i

)  c 2 s i f s

2

HPF

( ) 

Inv

( )

( c 1 c 2

) s    c 2

 s

2 s

2

0 if    c 2

So this cascading technique only produces broadband filters.

32

Example

Design an active band pass filter with a passband gain of 2 and cutoff frequencies of 500 Hz and 50 kHz. Use a 200 nF capacitor.

Lowpass filter:

R

L

1

 15.9

Highpass filter:

R

H

1

 1.592 k 

Gain:

R f

R i

 500 

33

First-Order Active Bandstop Filter

Gives  c1 Gives K

Gives  c2

LPF

Summing

Amp.

HPF

34

First-Order Active Bandstop Filter

Find the transfer function:

H

BRF

( s )  H

Inv

( s )[ H

LPF

( s )  H

HPF

( s )]

 R f

R i

 s

 c 1

 c 1

 s

 s

  c 2

( R f s 2 

R i

) [ s 2

(  c 1

 2

 c 2

 c 1 s

) s 

  c 1

 c 1

 c 2 c 2

]

K s 2

(

 s 2

 s

 

2

0

)

2

0 if  c 1

  c 2

So again, this technique only produces broadband filters.

35

Example

Design a notch filter between 20 krad/s and 100 krad/s with a gain of

K = 5. You have only 10 k  resistors available.

Lowpass filter:

C

L

1

20krad/s 10k 

 5nF

Highpass filter:

C

H

1

100krad/s 10k 

 1nF

Gain:

R f

R i

 5  R i

 10 k 

R f

5 10 k 

36

Example cont’d

Note: Due to the fact that the lowpass and highpass filters used in this example go down with only 20 dB/dec, attenuation in the stopband might be limited.

37

v i

Second-Order Op-Amp Filters

Bandpass Filter

H v o j 

 

1  j  2

R

1

 R

3

R

1

 R

3

C

C  ( j  ) 2

R

1

 R

3

C 2

Exercise: Try to derive this transfer function.

0

1 R

1

 R

3 , Q 

2

1

1

2

0

, B 

0 

2

Q R C

, H   K 

R

2

2 R

1

Advantage: The bandwidth, thus Q factor, is entirely controlled by R

2

.

38

Example 2 nd -order Bandpass Filter

Design for 3kHz, K =2, Q =10:

39

v i

H

Bandstop Filter

H

 k

1 ( j  ) 2 R C 2

1  j  2(2  )  ( j  ) 2 R C 2 v o 1 2

H

 k

1   2 R C 2

 1   2 R C 2

 2

   2(2  )

 2

H 

 

1

RC

 0    dB

40

Scaling

There are two ways of scaling a circuit: magnitude or impedance scaling , and frequency scaling .

Both are useful in scaling responses and circuit elements to values within the practical ranges.

Magnitude scaling leaves the frequency response of a circuit unaltered.

Frequency scaling shifts the frequency response up or down the frequency spectrum.

41

Magnitude Scaling

 Magnitude scaling is the process of increasing all impedances in a network by a factor, the frequency response remaining unchanged .

Z

Z

Z

R

L

C

 

 

  k Z k Z k Z

R

L

C

1

R   k R

 

 k m

1

R   k R

L   k L

C  

C k m

 The primed variables are the new values and the unprimed variables are the old values. Consider the series or parallel RLC circuit. We now have

 o

 

1

1

 m

 m

1

LC

  o

42

Frequency Scaling

 Frequency scaling is the process of shifting the frequency response of a network up or down the frequency axis while leaving the impedance the same .

   k f

R   R

 

L    j k f

1

 

C 

1

R   R

L   /

C   / f f

 Again, if we consider the series or parallel RLC circuit, for the resonant frequency

 o

 

1

/ f

1



/ f

 k f

LC

 k f

 o

B   k B

43

Magnitude and Frequency Scaling

 If a circuit is scaled in magnitude and frequency at the same time, then

R   k R

L   k m k f

L

C  

C

   k f

 B   k B

44

Example 1

Redesign the active low pass filter to use a 200 pF capacitor.

Use magnitude scaling:

C  

C k m k m

C

C 

10nF

200pF

 50

R

1

 

R

2

 

50(3183.8 ) 159.19 k 

45

Example 2

Redesign the active low pass filter, which was designed for 1 kHz, so that the cutoff frequency is 7.5 kHz, using a 250 pF capacitor.

Use frequency scaling first, then magnitude scaling:

 c

C  

 k

C

 

C   250 pF; k f k m

  c 

 c

C

7500

1000

 7.5

10nF

(7.5)(250pF)

 5.3

R

1

 R

2

 5 R

1

 84.9k

46

Example 3

The fourth-order Butterworth lowpass filter is designed such that the cutoff frequency is ω c

10-k  resistors.

= 1 rad/s. Scale the circuit for a cutoff frequency of 50 kHz using

R   k R L   k m k f

L

C  

C

   k f

Frequency scaling: k f

' c 

  

1

3

Magnitude scaling: k m

R

R

 3

1

 10 4

10 5

47

mH

48

Example 4

The third-order Butterworth filter is normalized to ω c

= 1. Scale the circuit to a cutoff frequency of 10 kHz. Use 15 nF capacitors.

k f

C ' 

' c 

 

1

4

C

   4

 k m

C f

1

15 10

 9    4

10

4

3 

R '  k R

L '  k m k f

L

10 4

3 

1   1.061k

10 4 2H

   4

 33.77 mH

C

49

Download