Chapter #2: Signals and Amplifiers from Microelectronic Circuits Text by Sedra and Smith Oxford Publishing Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Introduc;on IN THIS CHAPTER YOU WILL LEARN The terminal characterisHcs of the ideal op‐amp. How to analyze circuits containing op‐amps, resistors, and capacitors. How to use op‐amps to design amplifiers having precise characterisHcs. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Introduc;on IN THIS CHAPTER YOU WILL LEARN How to design more sophisHcated op‐amp circuits, including summing amplifiers, instrumentaHon amplifiers, integrators, and differenHators. Important non‐ideal characterisHcs of op‐amps and how these limit the performance of basic op‐amp circuits. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.1.1. The Op Amp Terminals terminal #1 inverHng input terminal #2 non‐inverHng input terminal #3 output terminal #4 posiHve supply VCC terminal #5 negaHve supply VEE Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.1.2. Func;on and Characteris;cs of Ideal Op Amp ideal gain is defined below !! = "$!" ! !# % ideal input characterisHc is infinite impedance ideal output characterisHc is zero impedance differenHal gain (A) is infinite bandwidth gain is constant from dc to high frequencies Q: But, is an amplifier with infinite gain of any use? Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.1.2. Func;on and Characteris;cs of Ideal Op Amp ideal gain: is defined below !! = "$!" ! !# % ideal input characteris;c: infinite impedance ideal output characteris;c: zero impedance Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.1.2. Func;on and Characteris;cs of Ideal Op‐Amp An amplifier’s input is composed of two components… differen;al input (vdfi) – is difference between inputs at inverHng and non‐inverHng terminals common‐mode input (vcmi) – is input present at both inverHng and non‐inverHng terminals *+,,+&-,+!$ "&./'01&$%# 2 !"##$%$&'"() "&./'01&!"# 2 6 474 8 678 &#' = (34 + 3) ! (34 ! 3) = (34 ! 34 ) + (3 + 3) Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.1.2. Func;on and Characteris;cs of Ideal Op‐Amp Similarly, two components of gain exist… differen;al gain (A) – gain applied to differenHal input ONLY common‐mode gain (Acm) – gain applied to common‐ mode input ONLY #$%%$&'%$() $4-54- !"#$ (*++),)&-*./ $4-54- )0102!! =!3 +! )0102!" =!3 !! 64 4744 8 64 4744 8 64 4744 8 6 474 8 = ('%& !3 + '!) ! ('%& !3 ! '!) = '%& (!3 ! !3 ) + ' (! + !) Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.1.2. Func;on and Characteris;cs of Ideal Op Amp Table 2.1: Characteris;cs of Ideal Op Amp infinite input impedance zero output impedance zero common‐mode gain (Acm = 0) complete common‐mode rejecHon infinite open‐loop gain (A = infinity) infinite bandwidth Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.1.3. Differen;al & Common‐Mode Signals Q: How is common‐mode input (vcmi) defined in terms of v1 and v2? !"#$%&!"'(!")*& 64 47448 %1 = %!"# ! %$# 3 2 { +,--,".-,/$(!")*& 64 4744 8 1 /!00 %!"# = 9%1 + %2 : ((4*&(567,888(( 2 %2 = %!"# + %$# 32 1442443 ",".!"#$%&!"'(!")*& Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.1.3. Differen;al & Common‐Mode !"##"$%#"&'()$*+, 644744 8 Signals 1 %!"# = 3%1 + %2 8 2 4+,(567"999 )$-'.,)$/()$*+, 64 47448 %1 = %!"# ! %$# :2 { &)00 %2 = %!"# + %$# :2 1442443 $"$%)$-'.,)$/()$*+, Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.2. The Inver;ng Configura;on Q: What are two basic closed‐loop op‐amp configuraHons which employ op‐amp and resistors alone? A: 1) inverHng and 2) non‐inverHng op amp Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.5: The inverHng closed‐loop configuraHon. 2.2. The Inver;ng Configura;on R2 facilitates “negaHve feedback” R1 regulates level of ques;on: what are two basic closed‐loop op amp configuraHons which this feedback employ op‐amp and resistors alone? answer: inverHng and non‐inverHng op amp note: here we examine the inverHng type source is applied to Oxford University Publishing inverHng input Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) non‐inverHng input is grounded 2.2.1. Closed‐Loop Gain Q: How does one analyze closed‐loop gain for inverHng configuraHon of an ideal op‐amp? step #1: Begin at the output terminal step #2: If vOut is finite, then differenHal input must equal 0 virtual short circuit btw v1 and v2 virtual ground exists at v1 Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) !"#$%&"'! '()'()+()(," 644 7448 %"#$ %* " %- = =. ! { ! 2.2.1. Closed‐Loop Gain step #3: Define current in to inverHng input (i1). step #4: Determine where this current flows? refer to following slide… !"#$%&' (#)%*+ } -#!" . ! -#, . #!" ! / #!" $, = = = %, %, %, Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.2. The Inver;ng Figure 2.5: The inverHng closed‐loop configuraHon. Configura;on i1 ques;on: what are two basic closed‐loop op amp configuraHons which employ op‐amp and resistors alone? answer: inverHng and non‐inverHng op‐amp 1 note: here we examine the inverHng type i Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) i = 0 2.2.1. Closed‐Loop Gain step #5: Define vOut in terms of current flowing across R2. step #6: SubsHtute vin / R1 for i1. !"#$%&' (#)%*+ &"#$ } = /&- 0 ! /'-(, 0 = !'-(, &"#$ (, = ! &!% (.)'%$")* note: this expression is one of the fundamentals of electronics Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.6: Analysis of the inverHng configuraHon. The circled 2.2.1. numbers indicate the order of the analysis steps. Closed‐Loop Gain ques;on: how will we… step #4: define vOut in terms of current flowing across R2 step #5: subsHtute vin / R1 for i1. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) closed‐loop gain G = ‐R2/R1 2.2.1. Effect of Finite Open‐Loop Gain Q: How does the gain expression change if open loop gain (A) is not assumed to be infinite? A: One must employ analysis similar to the previous, result is presented below… "! =! } (#$% ")1 6 )2 )1 "!<! = = (" (&' )2 # 2 + 7)1 6 )2 8 $ 2+% & ! ) ' 1442443 non‐ideal gain !"#!=! #$%&'#$%&#()&*!+,./!'#&0()&--!+'#!-#3!&45&5 Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) ideal gain 2.2.1. Effect of Finite Open‐Loop Gain Q: Under what condiHon can G = ‐R2 / R1 be employed over the more complex expression? A: If 1 + (R2/R1) << A, then simpler expression may be used. "! "! #$%%" + << ! %%&'()%%#!=! = " %%(*-(%%#!<! = "" "" ideal gain Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) ""! + "" # " + ,"! + "" . $ "+% & ! ' ( non‐ideal gain Example 2.1: Simple InverHng Amplifier Problem Statement: Consider an inverHng configuraHon with R1 = 1kOhm and R2 = 100kOhm. Q(a): Find the closed‐loop gain (G) for the cases below. In each case, determine the percentage error in the magnitude of G relaHve to the ideal value. cases are A = 103, 104, 105… Q(b): What is the voltage v1 that appears at the inverHng input terminal when vIn = 0.1V. Q(c): If the open loop gain (A) changes from 100k to 50k, what is percentage change in gain (G)? Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.2.3. Input and Output Resistances Q: What is input resistance for inverHng op‐amp? How is it defined mathemaHcally? A: R1 (refer to math below) Q: What does this say? A: That, for the combinaHon of ideal op‐amp and external resistors, input resistance will be finite… !"#$%&" #$%&'$() !"#$%&"*#$%&'$() this assumes that 64 4744 8 678 $ $!" $ ideal op‐amp and %# = !" = = !" = %! external resistors are #{ 6$!" ! ${! 78 %! $!" 8 %! !" considered “one #+%, -$./0+' +#*#! 1.2034 Oxford University Publishing unit” =5 Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Example 2.2: Another InverHng Op‐Amp Problem Statement: Consider the circuit below... Q(a): Derive an expression for the closed‐loop gain vOut/vIn of this circuit. Q(b): Use this circuit to design an inverHng amplifier with gain of 100 and input resistance of 1Mohm. Assume that one cannot use any resistor with resistance larger than 1Mohm. Q(c): Compare your design with that based on tradiHonal inverHng Oxford University Publishing configuraHon. Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.8: Circuit for Example 2.2. The circled numbers indicate the sequence of the steps in the analysis. Example 2.2: Figure 2.9: A current amplifier based on the circuit of Fig. 2.8. The amplifier delivers its output current to R Another InverHng 4. It has a current gain of (1 + R2 /R3), a zero input resistance, and an infinite output resistance. Op‐Amp The load (R4), however, must be floaHng (i.e., neither of its two PART B: Use this circuit to design an inverHng amplifier with gain of 100 and input resistance of 1Mohm. Assume that one cannot use any resistor with resistance larger than 1Mohm. terminals can be connected to ground). The largest resistor on may choose is 1Mohm Q: Where does one begin (in choosing the resistor values)? Which resistor would you define to be 1Mohm? A: The input resistance (R1) should be set as high as possible, therefore 1Mohm Q: What other resistor values should be defined? A: R2 = 1Mohm, R4 = 1Mohm, R3 = 10.2kohm Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.2.4. An Important Applica;on – The Weighted Summer weighted summer ‐ is a closed‐loop amplifier configuraHon which provides an output voltage which is weighted sum of the inputs. Figure 2.10: A weighted summer. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.2.4. An Important vOut = ‐[ (R f./RIn1)vIn1 + (Rf./RIn2)vIn2 + (Rf./RIn3)vIn3 + … ] Applica;on – The Weighted Summer weighted summer ‐ is a closed‐loop amplifier configuraHon which provides an output voltage which is weighted sum of the inputs. vIn1 vIn2 vIn3 RIn1 RIn2 RIn3 Rf Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) vOut Figure 2.10: A weighted summer. 2.3. The Non‐Inver;ng Configura;on non‐inver;ng op‐amp configura;on – is one which uHlizes external resistances (like the previous) to effect voltage gain. However, the polarity / phase of the output is same as input. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.12: The non‐inverHng configuraHon. 2.3. The Non‐Inver;ng Configura;on R1 and R2 act as voltage divider, regulaHng negaHve feedback to the inverHng input inverHng input is grounded through R1 node #1 node #2 source is applied to non‐inverHng input Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Characteris;cs of Non‐Inver;ng Op‐Amp Configura;on '!"# '$% } " (! # ( " + $(! % (" & !"#$%&'$!(# $ & >> " + % )##### )&=! = " + ! (" ' (" " + " + $(! % (" & & & " + $(! % (" & (*( + !"#$%&'$!( ) ###################( ) = {&<! " + $(! % (" & '!"# "+ & '$% " + $(! % (" & " + $(! % (" & ,#-.#(/&'$!(&#--*- ) ###############*+, = "'' & + " + $(! % (" & " + " + $(! % (" & & " (" # " + $(! % (" & !(0#-/!('&!(,1/&,*/#(/!$% ) ###('" = '!"# $ % Oxford University Publishing & (" + (! ' " + " + $(! % (" & Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) & Configura;on and Characteris;cs of Buffer / Voltage‐Follower Op‐Amp Configura;on Figure 2.14: (a) The unity‐gain buffer or follower amplifier. (b) Its equivalent circuit model. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Configura;on and Characteris;cs of Buffer / Main point? For the buffer amp, output voltage is equal Voltage‐Follower Op‐Amp Configura;on (in both magnitude and phase) to the input source. However, any current supplied to the load is drawn from amplifier supplies (VCC, VEE) and not the input source (vI). Figure 2.14: (a) The unity‐gain buffer or follower amplifier. (b) Its equivalent circuit model. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4. Difference Amplifiers difference amplifier – is a closed‐loop configuraHon which responds to the difference between two signals applied at its input and ideally rejects signals that are common to the two. Ideally, the amp will amplify only the differenHal signal (vdfi) and reject completely the common‐ mode input signal (vcmi). However, a pracHcal circuit will behave as below… )!"# = *)$%& + *'()'(& Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4. Difference Amplifiers common‐mode input common‐mode gain differenHal input differenHal gain )!"# = *)$%& + *'()'(& Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4. Difference Amplifiers common‐mode rejec;on ra;o (CMRR) – is the degree to which a differenHal amplifier “rejects” the common‐mode input. Ideally, CMRR = infinity… !$%% = #"$!"#!" Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) # #!" 2.4. Difference Figure 2.15: RepresenHng the input signals to a differenHal amplifier in terms of their differenHal and common‐mode components. Amplifiers #&&' = #"$!"#!" Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) %!" %#$ 2.4. Difference Amplifiers Q: The op amp itself is differenHal in nature, why cannot it be used by itself? A: It has an infinite gain, and therefore cannot be used by itself. One must devise a closed‐loop configuraHon which facilitates this operaHon. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4. Difference Figure 2.16: A difference amplifier. Amplifiers Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.1. A Single Op‐Amp Difference Amp Q: What are the characterisHcs of the difference amplifier? A: Refer to following equaHons… '!"# *&! + &" ,&# &! = '$%! ! '$%" *&# + &$ ,&" &" "&" = &$ # &! %&'()+(((( $ % ((((('-./((((('!"# = ('$%! ! '$%" ) &" &&! = &# ' Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) A Shie in Nota;on Before this point… The parameter A is used to represent open‐loop gain of an op amp. The parameter G is used to represent ideal / non‐ideal closed‐loop gain of an op amp. Aoer this point… The parameter A is used to represent ideal gain of an op amp in a given closed‐loop configuraHon. The parameter G is not used. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.2. The Instrumenta;on Amplifier Q: What is one problem associated with the difference amplifier? A: Low input impedance. Q: And, what does this mean pracHcally? A: That source impedance will have an effect on gain. Q: What is the soluHon? A: Placement of two buffers at the input terminals, amplifiers which transmit the voltage level but draw minimal current. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.2. The Instrumenta;on Amplifier Q: However, can one get “more” from these amps than simply impedance matching? A: Yes, maybe addiHonal voltage gain??? Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.2. The Instrumenta;on Figure 2.20: A popular circuit for an instrumentaHon amplifier. Amplifier stage #1 stage #2 quesHon: however, can we get “more” from these amps than simply impedance matching? answer: yes, maybe addiHonal voltage gain??? non‐inverHng op amp (A1) vOut = (1 + R2/R1)vIn non‐inverHng op amp (A2) Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) difference op amp (A3) vOut = (R4/R3)vdfi 2.4.2. The Instrumenta;on Amplifier Q: However, can one get “more” from these amps than simply impedance matching? A: Yes, maybe addiHonal voltage gain??? !"#$%&'"(&)$*!+,$(&," +$%!")-'$!#!+,$(#-./+&+'"(,&(&+0)"'(25265 644474448 %1 ! %2 " +&'$ = # 4 + $ +()* %3 % %4 & 14243 ! !"#$ 7% 8 Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) addiHonal voltage gain 2.4.2. The Instrumenta;on Amplifier advantages of instrumentaHon amp very high input resistance high differenHal gain symmetric gain (assuming that A1 and A2 are matched) disadvantages of instrumentaHon amp ADi and ACm are equal in first stage – meaning that the common‐mode and differenHal inputs are amplified with equal gain… Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) What is problem with ACm = A? vIn1 vIn1 A = 10 A = 25 A = 10 x 25 vIn2 vIn2 differenHal gain >> common‐mode gain differenHal gain = common‐mode gain Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) differen;al gain >> common‐mode gain vIn1 = 10.03V A = 10 x 25 vIn2 = 10.02V Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) vOut= 250 x (10.03‐10.02)V vOut = 2.5V no problem!!! differen;al gain = common‐mode gain vIn1 = 10.03V A = 10 vOut1= 10 x 10.03 = 15V saturaHon A = 25 vOut= 25 x (15‐15)V vOut = 0V problem!!! vIn2 = 10.02V vOut2= 10 x 10.02 = 15V saturaHon Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.2. The Instrumenta;on Amplifier advantages of instrumentaHon amp very high input resistance high differenHal gain symmetric gain (assuming that A1 and A2 are matched) disadvantages of instrumentaHon amp ADi and ACm are equal in first stage – meaning that the common‐mode and differenHal inputs are amplified with equal gain… need for matching – if two op amps which comprise stage #1 are not perfectly matched, one will see unintended Oxford University Publishing effects Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.2. The Instrumenta;on Amplifier Q: How can one fix this (alleviate these disadvantages)? A: Disconnect the two resistors (R1) connected to node X from ground, making the configuraHon “floaHng” in nature… A: Refer to following slide… Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.20: A popular circuit for an instrumentaHon amplifier. (b) 2.4.2. The The circuit in (a) with the connecHon between node X and ground Instrumenta;on removed and the two resistors R 1 and R1 lumped together. This Amplifier simple wiring change dramaHcally improves performance. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.2. The Instrumenta;on Amplifier Q: How can one analyze this circuit? Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.2. The Instrumenta;on Amplifier step #1: note that virtual short circuit exists across terminals of op amp A1 and A2 step #2: define current flow across the resistor 2R1 step #3: define output of A1 and A2 Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) %&'()&*+(&,(-.,(%! (-/0(%" 64748 &# + $ ! &# ! $ = : (((((999*+1'1%&'1 (((((((((((((((&#+ $ = &#!$ !"#$ 6 4&74 8 &!"" ! &!"! $'! = "'! )12-341(/&(23''1/*(5677(%7&5 6/*&(601-7(&,(-.,8(-77(&%($'! (5677( %7&5(-2'&44('" 6447448 &()* ! = &!"! ! $'!'" &()* " = &!"" + $'!'" 2.4.2. The Instrumenta;on Amplifier short‐ckt vOut1 iR1 vOut2 Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.2. The Instrumenta;on 64444444744444448 Amplifier " $ ' !' % # !"#$%&#$%&'()&$*'+,-$&./01(,)2$03,4&5$16& %(''&+&)1(07$()8/1$'!"# ! !'!"# " $1,$2109&$:! step #4: Define output of A1 and A2 in terms of input alone '!"# ! ! '!"# " = &'$%! + ( $%! $%" ) &! ' ! K !&" + * 2444 ,1444 3 '!"# ! ='$% ! + (& "&! " $ '$%! ! '$%" % # $$$$$$$$$$$$$$$$$$$$$$$$$$$$$ K ! &'$%" ! ( ) &! ' !&" + * 2444 ,1444 3 '!"# " ='$% " !(& "&! !"#$%&#$;,-3()&$1&+-2 64444444 744444448 $%)( $6 4'74 8% ( '$%! ! '$%" ) '!"# ! ! '!"# " = *'$%! ! '$%" 5 + ! ( ) &! 1424 3 ( !&" ) '$%)( * + $ !& % '!"# ! ! '!"# " = ( " + ! ) '$%)( * !&" + Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.4.2. The Instrumenta;on Amplifier step #5: Define output of A3 . step #6: Define gain of revised instrumentaHon amplifier. !"#$%&*+,-./0-+/0 )-12%+&.+$!"# $%&' )! 64748 = 3$%&' " ! $%&' $ 4 )# $%&' )! " ")" # = $$ + % $!"# )# & ")$ ' $%&' )! " ")" # = *(# = $ $ + % $!"# )# & ")$ ' %&'()/&0 Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.5. Integrators and Differen;ators integrator / differen;ator amplifier – is one which outputs an integral or derivaHve of the input signal. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.5.1. The Inver;ng Configura;on with General Impedances Q: Does the transfer funcHon for the inverHng op amp change if the feedback and input impedances are not purely resisHve? A: No, not in form… Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Example 2.4: Other Op‐Amp ConfiguraHons Consider the circuit on next slide page. Q(a): Derive an expression for the transfer funcHon vOut / vIn. Q(b): Show that the transfer funcHon is of a low‐pass STC circuit. Q(c): By expressing the transfer funcHon in standard form of Table 1.2, find the dc‐gain and 3dB frequency. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.23: Circuit for Example 2.4. Example 2.4: Other Op‐Amp ConfiguraHons 2.5.2. The Inver;ng Integrator Q: How can inverHng op‐amp be adapted to perform integraHon? A: UHlizaHon of capacitor as feedback impedance. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.5.2. The Inver;ng Figure 2.24: (a) The miller or inverHng integrator. (b) Frequency response of the integrator. Integrator !"!#!$% &'#('# )&%#$*+ ! " 678 - ! #1$"2!+"#.3+241!(#!&"./3405......8'"#! /! 0 = # $ ( '$% /! 0(! % # '"#! /!, 0 )-*& & ! =, ' '"#! 2#+$3672#$#+.3+241!(#!&"./$405. =# '$% +)-*& Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.5.2. The Inver;ng Integrator Q: What is the problem with this configuraHon (related to dc gain)? A: At dc frequency (ω = 0), gain is infinite Gain = 1 / (ω.R1CF) Q: SoluHon? A: By placing a very large resistor in parallel with the capacitor, negaHve feedback is employed to make dc gain “finite.” Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.25: The Miller integrator with a large resistance RF 2.5.2. The Inver;ng Integrator connected in parallel with C in order to provide negaHve feedback and hence finite gain at dc. +/-&')$&+"#$'0/)%+)(&"1#023"""""""#$%$&#'"(&")&%*+"'),&-.666 '!"# ($ 7 (! '+$-#45'+-+$"#$'0/)%+)(&"1-023" =! '%& ! + )($ *$ Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Example 2.5: Miller Integrator Consider the Miller integrator… Q(a): Find response of a Miller Integrator to input pulse of 1V height and 1ms width. R1 = 10kOhm, CF = 10nF Q(b): If the integrator capacitor is shunted by a 1MOhm resistor, how will the response be modified? note: the op amp will saturate at +/‐ 13V Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.5.3. The Op‐Amp Differen;ator Q: How can one adapt integrator to perform differenHaHon? A: Interchange locaHons of resistors and capacitors. Figure 2.27: A differenHator. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.5.3. The Op‐Amp Differen;ator '! !" "% # %&'()*+(%$,+)-&*.%*/($",-#0$$$$$$$$!#$% "% # = !(& )! '% "#$% "*# )%+',12)%'%+$,+)-&*.%*/($"'-#0$$ = ! *(& )! "!" "*# Figure 2.27: A differenHator. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.5.3. The Op‐Amp Differen;ator filtering characterisHc is high pass filter magnitude of transfer funcHon is |VOut / VIn| = ωRFC1 phase of transfer funcHon is φ = ‐90O differenHator Hme‐constant is frequency at which unity gain occurs and defined as ω = 1 / RFC1 Q: What is the problem with differenHator? A: DifferenHator acts as noise amplifier, exhibiHng large changes in output from small (but fast) changes in input. As such, it is rarely used in pracHce. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.6. DC Imperfec;ons Q: What will be discussed moving on? A: When can one NOT consider an op amp to be ideal, and what effect will that have on operaHon? Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.6.1. Offset Voltage Q: What is input offset voltage (VOS)? A: An imaginary voltage source in series with the user‐ supplied input, which effects an op amp output even when idfi = 0. What will happen when short is Oxford University Publishing applied? Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.28: circuit model for an op amp with input offset voltage VOS. 2.6.1. Offset Voltage Q: What causes VOS? A: Unavoidable mismatches in the differenHal stage of the op amp. It is impossible to perfectly match all transistors. Q: Range of magnitude? A: 1mV to 5mV Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) !""#$%&'( !)%*)% !""#$% +!,%-.$ } } ! ( " )"#$%& = )$' # / + ! $ (/ & % This relaHonship between offset voltage (VOS) and offset dc output (V OsOut) applies to both inverHng and non‐inverHng op amp. 2.6.1. Offset Voltage However, only if one assumes that VOS is present at non‐inverHng input. Q: What causes VOS? A: Unavoidable mismatches in the differenHal stage of the op amp. It is impossible to perfectly match all transistors. Q: Range of magnitude? A: 1mV to 5mV Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) !""#$%&'( !)%*)% !""#$% +!,%-.$ } } ! ( " )"#$%& = )$' # / + ! $ (/ & % 2.6.1. Offset Voltage Q: How can this offset be reduced? A: offset nulling terminals – A variable resistor (if properly set) may be used to reduce the asymmetry present and, in turn, reduce offset. A: capaci;ve coupling – A series capacitor placed between the source and op amp may be used to reduce offset, although it will also filter out dc signals. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.30: The output dc offset voltage of an op‐amp can be trimmed to zero by connecHng a potenHometer to the two offset‐nulling terminals. The wiper of the potenHometer is connected to the negaHve supply of the op amp. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.31: (a) A capaci;vely‐coupled inverHng amplifier. (b) The equivalent circuit for determining its dc output offset voltage VO. dc signals cannot pass! Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.6.2. Input Bias and Offset Currents input bias current ‐ is the dc current which must be supplied to the op‐amp inputs for proper operaHon. Ideally, this current is zero… input offset current ‐ the difference between bias current at both terminals Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.32: The op‐amp input bias currents represented by two current sources IB1 and IB2. 2.6.2: Input Bias and Offset Currents Figure 2.32: The op‐amp input !"#$%&'(()*+ bias currents represented by #+%+)(,"*#-$ ./%#*0%.1 two current sources I B1 and IB2. 678 '!/ + '!1 "*5'+%!"#$%&'(()*+6%%%%%:%%%%%%%%%%'! = input bias current ‐ is the 1 dc current which must be 0"22)()*&) supplied to the op‐amp !)+3))*%!"#$4 inputs for proper 6 474 8 operaHon. "*5'+%722$)+%&'(()*+6%%%%%%%%%%%%%'"# = '!/ ! '!1 Ideally, this current is zero… input offset current ‐ the 64 7 4 8 difference between bias ()$'-+"*8%7'+5'+%97-+#8)6%%%%% (!"$% = '!/)& current at both terminals Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.6.2. Input Bias and Offset Currents Q: How can this bias be reduced? A: Placement of R3 as addiHonal resistor between non‐inverHng input and ground. Q: How is R3 defined? A: Parallel connecHon of RF and R1. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) "#$%$&'"()*+,#-(.#&/##0 0'01%02#"&%03(%0)4& +0-(3"'40-(5"! 6($7'4*#84+*()+"+**#*(,'00#,&%'0 '9(%02#"&%03(%0)4& "#$%$&+0,#(+0-(9##-.+,; 64748 ":"! "! = ": + "! 2.7.1. Frequency Dependence of the Open‐Loop Gain The differenHal open‐ loop gain of an op‐amp is not infinite. It is finite and decreases with frequency. It is high at dc, but falls off quickly starHng from 10Hz. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.39: Open‐loop gain of a typical general‐purpose internally compensated op amp. 2.7.1. Frequency Dependence of the Open‐Loop Gain internal compensa;on – is the presence of internal passive components (caps) which cause op‐amp to demonstrate STC low‐ pass response. frequency compensa;on – is the process of modifying the open‐ loop gain. The goal is to increase stability… Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.39: Open‐loop gain of a typical general‐ purpose internally compensated op amp. 2.7.1: Frequency The gain of an internally compensated op‐amp may be Dependence expressed as shown below… of the Open‐Loop Gain "4%',#)&%!)+,.4"5,!",!6'78'.&!95:'",;!!!!!!!!!!!0#1 = 2 + # 3 !! "4%',#)&%!)+,.4"5,!",!)%&*+&,./!95:'",;!!!!!!0 $! 1 = 2 + $! 3 !! "-!! 4%',#)&%!)+,.4"5,!)5%!<"=<!)%&*+&,."&#;!!!!!! !0 $! 1 " $!3 14 4244 !! !"#!$%&'(!)%&*+&,./ "-!! !% :'=,"4+9&!='",!)5%!<"=<!)%&*+&,."&#;!!!!!!!!! !0 $! 1 " " $! ! +,"4/!='",!5..+%#!'4!!% ;!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!% = "-!! Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.7.2. Frequency Response of Closed‐Loop Amplifiers (!"# ")1 9 )2 = ($% 2 + :2 + )1 9 )2 ;9 :{ '; !"#$ %!!" &'($ Q: How can we create a more accurate descripHon of closed loop gain for an inverHng‐type op‐amp? step #1: Define closed‐loop gain of an inverHng amplifier with finite open‐loop gain (A) step #2: Insert frequency‐ dependent descripHon of A from last slide step #3: Assume A0 >> 1 + R2/R1 Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) (!"# ")1 9 )2 ")1 9 )2 = = ($% 2 + 2 + )1 9 )2 # 2 + )1 9 )2 $ 2+% & :2 + * 9 !& ; # '4 $ ' 4 '144424443 ( % & 2 + * 9 !4 ( !"#$%&3)/"%(-)-5#/#)-#+,/ '14 243 ' )*+!,)-.! /%(0#/)6'78 (!"# = ($% ")1 9 )2 # 2 + )1 9 )2 $ * # 2 + )1 9 )2 $ 2+ % & + % & ' ! ' 4 & ' 4 '14243 ( 14 4244 3( !"#$%&3)+#"%'7#).(-5)4 6#7'=/#)'4 >>2+)1 9 )2 (!"# = ($% ")1 9 )2 * (2 + )1 9 )2 ) 2+ !# /!%=-(!$ !"#$%&3)+#"%'7#).(-5<<< 2.7.2. Frequency Response of Closed‐Loop Amplifiers Q: How can we create a more accurate descripHon of closed loop gain for an both inverHng and non‐ inverHng type op‐amps? #$%&'(#$)*+,*-., $+$/#$%&'(#$)*+,*-., 6444 474444 8 6444 474444 8 &!"# "'! 0 '" &!"# " + '! 0 '" = ********** = ( (" + '! 0 '" ) ( (" + '! 0 '" ) &$% &$% "+ "+ !# !# Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.7.2. Frequency Response of Closed‐Loop Amplifiers 3dB frequency – is the frequency at which the amplifier gain is axenuated 3dB from maximum (aka. dc ) value. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) !!"# !! = # + $" $ $# 2.8. Large‐Signal Opera;on of Op‐ Amps 2.8.1. Output Voltage Satura;on If supply is +/‐ 15V, then vOut will saturate around +/‐ 13V. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.8.2. Output Current Limits iOut current of op‐amp, including that which facilitates feedback, cannot exceed X. The book approximates X at 20mA. 2.8.3. Slew Rate slew rate – is maximum rate of change of an op‐ amp (V/us) Q: How can this be problemaHc? A: If slew rate is less than rate of change of input. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) !"#$%&'(#%)*+, 64 4744 8 $%!"# &' = $# -'. 2.8.3. Slew Rate Q: Why does slewing occur? A: In short, the bandwidth of the op‐ amp is limited – so the output at very high frequencies is axenuated… Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) 2.8.4. Full‐Power Bandwidth Op‐amp slewing will cause nonlinear distorHon of sinusoidal waveforms… sine wave rate of change Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) #!" = $!" !"# (!% ) &#!" = ! $!" $%! (!% ) &% 2.8.4. Full‐Power Bandwidth full‐power bandwidth (fM) – the maximum frequency at which amplitude of a sinusoidal input and output are equal maximum output voltage (VOutMax) – is equal to (A*vIn) note: an inverse relaHonship exists between fM and VOutMax note: beyond ωM, output may be defined in terms of ω Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) !"#$% &'#('# ,- )&*#"+$ /"0%1 = ).*!" } 678 +, = !# -$%&#'( +, .# = <" -$%&#'( 144244 3 2'**3(&4$!5/"0%46%#7 #7685)"*'$ 9"00&#5/$ +!$"#$! #7"05&0$ } #! $ -$%& = -$%&#'( % # & ' !3( 144 42444 !$*"#6&0876(5/$#4$$0 "9#'"*5&'#('#5"0%5:";6:':5 Conclusion The IC op‐amp is a versaHle circuit building block. It is easy to apply, and the performance of op‐amp circuits closely matches theoreHcal predicHons. The op‐amp terminals are the inverHng terminal (1), the non‐inverHng input terminal (2), the output terminal (3), the posiHve‐supply terminal (4) to be connected to the posiHve power supply (VCC), and the negaHve‐supply terminal (5) to be connected to the negaHve supply (‐ VEE). Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Conclusion (2) The ideal op‐amp responds only to the difference input signal, that is (v2 ‐ v1). It yields an output between terminals 3 and ground of A(v2 ‐ v1). The open‐loop gain (A) is assumed to be infinite. The input resistance (Rin) is infinite. The output resistance (Rout) is assumed to be zero. NegaHve feedback is applied to an op‐amp by connecHng a passive component between its output terminal and its inverHng (aka. negaHve) input terminal. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Conclusion (3) NegaHve feedback causes the voltage between the two input terminals to become very small, and ideally zero. Correspondingly, a virtual short is said to exist between the two input terminals. If the posiHve input terminal is connected to ground, a virtual ground appears on the negaHve terminal. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Conclusion (4) The two most important assumpHons in the analysis of op‐amp circuits, assuming negaHve feedback exists, are: the two input terminals of the op‐amp are at the same voltage potenHal. zero current flows into the op‐amp input terminals. With negaHve feedback applied and the loop closed, the gain is almost enHrely determined by external components: Vo/Vi = ‐R2/R1 or 1+R2/R1. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Conclusion (5) The non‐inverHng closed‐loop configuraHon features a very high input resistance. A special case is the unity‐ gain follower, frequently employed as a buffer amplifier to connect a high‐ resistance source to a low‐resistance load. The difference amplifier of Figure 2.16 is designed with R4/R3 = R2/R1, resulHng in vo = (R2/R1)(vI2 ‐ vI1). Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.16 Conclusion (6) The instrumentaHon amplifier of Figure 2.20(b) is a very popular circuit. It provides vo = (1+R2/R1)(R4/R3)(vI2 ‐ vI1). It is usually designed with R3 = R4 and R1 and R2 selected to provide the required gain. If an adjustable gain is needed, part of R1 can be made variable. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.20(b) Conclusion (7) The inverHng Miller Amplifier of Figure 2.24 is a popular circuit, frequently employed in analog signal‐processing funcHons such as filters (Chapter 16) and oscillators (Chapter 17). The input offset voltage (VOS) is the magnitude of dc voltage that when applied between the op‐amp input terminals, with appropriate polarity, reduces the dc offset at the output. Oxford University Publishing Microelectronic Circuits by Adel S. Sedra and Kenneth C. Smith (0195323033) Figure 2.24