3-2 BLOCK DIAGRAM REPRESENTATION OF FEEDBACK CONFIGURATIONS An op-amp that uses feedback is called a feedback amplifier. A feedback ampli- fier is sometimes refertred to as a closed-loop amplifier becmise the feedback forms a closed loop between the input and the output, A feedback amplifier es- sentially consists of two parts: an op-amp and a feedback circuit. The feedback cireuit can take any form whatsoever, depending on the intended application of the amplifier. This means that the feedback circuit may be made up of either pas- sive components, active components, or combinations of both. This chapter, in or- der to develop the basic feedback concepts, presents only purely resistive feed- back circuits. A closed-loop amplifier can be represented by using two blocks, one for an op-amp and another for a feedback circuit. There are four ways to connect these two blocks. These connections are classified according to whether the voltage or current is fed back to the input in series or in parallel, as follows: 1. Voltage-series feedback 2. Voltage-shunt feedback 3. Current-series feedback 4. Current-shunt feedback The four types of configurations are illustrated in Figure 3-1. In Figure 3- 1(a) and (b) the voltage across load resistor R, is the input voltage to the feed- back circuit. The feedback quantity (either voltage or current) is the output of the feedback circuit and is proportional to the output voltage. On the other hand, in the current-series and current-shunt feedback circuits of Figure 3—1(c) and (d), the load current i, flows into the feedback circuit. The output of the feedback cir- cuit (either voltage or current) is proportional to the load current i,. Note that in all four of these configurations the signal direction through the op- amp is from the input to the output. On the other hand, in the ideal case the signal direction through the feedback circuit is exactly opposite: from output to input. The voltage-series and voltage-shunt feedback configurations are important because they are most commonly used. An in-depth analysis of these two con- figurations is presented here, computing voltage gain, input resistance, output re- sistance, and bandwidth for each. The other two configurations are not discussed in this chapter. However, the interested reader may analyze the latter using the same procedures as illustrated for the first two configurations. 3-2 Block Diagram Representation of Feedback Configurations 89circuit ireuit ——. in Te opame n , x } Vn Ry A : Feedback ae chen a FIGURE 3-1 Feedback configurations. (a) Vollage-series. (b) Voltage-shunt. (c) Current series. (d) Currentshunt. Arrows indicate the signal flow directions. 3-3 VOLTAGE-SERIES FEEDBACK AMPLIFIER 90 The schematic diagram of the voltage-series feedback amplifier is shown in Figure 3-2. The op-amp is represented by its schematic symbol, including its large-signal voltage gain A, and the feedback circuit is composed of two resistors, R, and R,. The circuit shown in Figure 3—2 is commonly known as a noninverting ampli fier with feedback (or closed-loop noninverting amplifier) because it uses feedback, and the input signal is applied to the noninverting input terminal of the op-amp. Before proceeding, it is necessary to define some important terms for the volt- age-series feedback amplifier of Figure 3-2. Specifically, the voltage gain of the op-amp with and without feedback, and the gain of the feedback circuit are de- fined as follows: open-loop voltage gain (or gain without feedback) A = Via closed-loop voltage gain (or gain with feedback) Ap = 22 gain of the feedback circuit pat y, ‘An Op-Amp With Negative Feedbackt+—Feedback Le === 57 circuit FIGURE 3-2 Vollogeseries feedback amplifier (or noninverting amplifier with feedback). 3-3-1 Negative Feedback Referring to the circuit of Figure 3-2, Kirchhoff’s voltage equation for the input loop is -y G-) input voltage feedback voltage via = difference input voltage Recall, however, that an op-amp always amplifies the difference input volt- age viz From Equation (3-1), this difference voltage is equal to the input voltage vig minus the feedback voltage v, In other words, the feedback voltage always op- Poses the input voltage (or is out of phase by 180° with respect to the input volt. age); hence the feedback is said to be negative. Returning now to the analysis of the voltage-series feedback amplifier, we should note that it will be performed by computing closed-loop voltage gain, in. put and output resistances, and the bandwidth. 3-3-2 Closed-Loop Voltage Gain As defined previously, the closed-loop voltage gain is However, by Equation (2-9), Up = A(vy — 02) 3-3 Voltage-Series Feedback Amplifier cAIReferring to Figure 3-2, we see that | mS Un wey, since R, > Ry Therefore, Ry, Rearranging. we get (Ry + Ryn R, + Rp + AR, Thus ACR: + Re) = exact) ou Are oR +R TAR ‘ wu Generally, A is very large (typically 10°). Therefore, AR, >> (Ri + Rp) and (R, + Re + AR) = AR, Thus Rr 5 ae 1+ R, (ideal) B-3) Equation (3-3) is important because it shows that the gain of the voltage-se. Ties feedback amplifier i is determined by the ratio of two resistors, R, and Re. For instance, if a gain of 11 is desired, we can then choose R, = 1 kM and Rr = 10k or R, = 1000 and Ry = 1 kQ. In other words, in setting the gain the ratio of R, and R; is important, and not the absolute values of these resistors. As a general mule, however, all external component values should be less than 1 MQ so that they do not adversely affect the internal circuitry of the op-amp. This is especially true for older-generation ICs such as the 709 and 741. Another interesting result can be obtained from Equation (3-3). As defined previously, the gain of the feedback circuit (B) is the ratio of v; and v,. Referring to Figure 3-2, this gain is Ar = 2 y 4 Be =e i g (3-4) gia R, + Rp Comparing uation’ ( (3-3) and (3-4), we can conclude that 1 Ap = (ideal) 3-5) B 92 An Op-Amp With Negative Feedback‘Summing junetion A FIGURE 3-3 Block diagram = fopresentation of noninverting amplifier with foodback shis means that the gai fe ‘| me euaaze gail ee the feedback circuit is the reciprocal of the closed- pone » for given R, and R; the values of A, and B are that, Equation (3-5) i J Isa ir i i simpler form makes it easier to remember. ae Finally, the closed-loop voltage gai Boalt op voltage gain Ay can be expressed i - joop gain A and feedback circuit gain B as follows. Reeeeering Eaution (222) we get = ay a2 =) Ap = > Rit Re AR, Ri + Re Ry + Rp Using Equation (3-4) yields A Ar Ty AB ey where Ar = closed-loop voltage gain ‘A = open-loop voltage gain B = gain of the feedback circuit AB = loop gain |4 one-line block diagram of Equation (3-6) is shown in Figure 3-3. This block diagram illustrates a Standard form for representing a system with feedback and oh indicates the relationship between the different variables of the system. The block-diagram approach helps to simplify the analysis of complex closed-loop networks, particularly if they-are composed of nonresistive feedback circuits. 3-3-3 Difference Input Voltage Ideally Zero Let us reconsider Equation (2-9), which can be rewritten as 2 Be via = y Since A is very large (ideally infinite), 0 G78) via = 93 3-3. Voltage-Series Feedback Amplifier5B yey fideal) a Bauation (3=70) says that the voltage al the noninverting input terminal of gq amp is approximately equal to that at the inverting input terminal provided iy. ig very lange. This concept is useful in the analysis of closed-loop op-anyy" cuits, For example, ideal closed-loop voltage gain [Equation O31 can Ci, tained! using the preceding results as follows, In the circuit of Figure 3-2, ‘That is, Oh, vy = Yin vty Rivy *R, + Re Substituting these values of v, and vy in Equation (3-7b), we get Rivy Vin = : + Rp That is, Ap= a oie z 3-3-4 Input Resistance with Feedback Figure 3~4 shows a voltage-series feedback amplifier with the op-amp equivalen “circuit. In this cireuit R; is the input resistance (open loop) of the op-amp, and R Vee FIGURE 3-4 Derivation of input resistance with feedback. ow" An Op-Amp With Negative Feedbackput resistance of the amplifier wi ick is defined ne Diifier with feedback. ‘The input resistance with Ry = Min Hin = oy vid, However, Yo Ma = and y,=—A_ / T+ ap ve Therefore, Vin "UdA R; Vin Au,/(. + AB) = RQ + AB) stance of the op-amp with feedback is (1 + AB) @B-8) This means that the input resi times that without feedback. 3-3-5 Output Resistance with Feedback Output resistance is the resistance determined looking back into the feedback am- plifier from the output terminal as shown in Figure 3-5. This resistance can be Vee \eRee\ V FIGURE 3-5 Derivation of output resistance with feedback. 3-3 Voltage-Series Feedback Amplifier 9596 pendent sources, Specifically, to 4 rendent source thy {0 7eF0, ani "i App y he sin's theorem for de Hing current é,. In short, the p aback yy. reduce indey obtained by using Théver on calculate the rest sistance with fee output re) age tip, and th an extemal is defined as follows Rye oe 7 (31 ¢ Sny fy nat output node NV, we get Writing Kirchhot?’s current equatio' ig + iy i since [(Rp + By) |] Ri] 2 Ro and fg 2 dy» Therefore, lo = ba d by writing Kirchhoff’s voltage equation for the ou, The current #, can be foun put loop: Up — Roig — Avia = 9 Avia However. Via = 11 Therefore, i, a Equation (3—9a), we get Vo Roe = ape al Re (w, + ABv,/Re B-9) eRe 1+ AB tance of the voltage-series feedback ampli- tance R, of the op-amp. That is, the out- k is much smaller than the output re- Substituting the value of i, in the output resist s the output resis! dbac] This result shows that fier is 1/(1 + AB) times put resistance of the op-amp with fee sistance without feedback. 3-3-6 Bandwidth with Feedback The bandwidth of an amplifier is defined as the band (range) of frequencies for which the gain remains constant. Manufacturers generally specify either the gain- bandwidth product or supply open-loop gain versus frequency curve for the op- amp. For the 741 op-amp the latter is typical. pe ‘An Op-Amp With Negative Feedback G108 a Van isv Tan 26 7 298 FIGURE 3- ; a E 3-6 Openloop gain fo Frequency (Hz) wae frequency curve of the ; ee a ae open-loop gain versus frequency curve of the 741C Soe eor the Bait be vg fcr 4 gain of 200,000, the bandwidth is approximately 5 Hz; : gi ra andwidth | Product is (200,000 x 5 Hz) = 1 MHz. On the ater ex reme: he bandwidth is approximately 1 MHz when the gain is 1. Thus. he gain-l product is constant, However, this holds tru for op- n C e only for op- amps like the 741 that have just one break frequency below ‘unity ‘gain bandwidth. For the 741, 5 Hz is the break fre : i the gain A is 3 dB down from its value at 0 His We sil ene fo ae other hand, the frequency at which the gain equals 1 is known as the unity gain bandwidth (UGB). The relationship between the break frequency f,, oper loo voltage gain A, bandwidth with feedback f,, and the closed-loop gain Ar can be established as follows. Since for an op-amp with a single break frequency f. the gain-bandwidth product is constant, and equal to the unity gain-bandwidth (UGB), we can write, UGB = (A)(f,) (10a) where A = open-loop voltage gain f, = break frequency of an op-amp or, alternatively, only for a single break frequency op-amp, UGB = (A;)(fr) (3-10) closed-loop voltage gain bandwidth with feedback where A, tr Therefore, equating Equations (3-10a) and (3—10b), (A)(f.) = (Ar) fr) 3-3 Voltage-Series Feedback Amplifier 978 or (A) fe A ie However, for the noninverting amplifier with feedback, i A Are TAB of A, in Equation (3-10c), we get (AVG) WC + AB) ‘Therefore, substituting the value tt te = fo + ABY Oty ] uation (3—-10d) indicates that the bandwidth of the noninverting ampjig, with feedback, fr, is equal to its bandwidth without feedback, f,, times (1 + AR In other words, if negative feedback is used as shown in Figure 3-2, gain 4 te W/L + AB); consequently, the open-loop bandwidth f, (the break jn, quency) should increase to f,(1 + AB). For instance, let us assume that the Tic ig used in the circuit of Figure 3-2 and that the desired voltage gain A, is 19 then the closed-loop bandwidth fy [using Equation (3—10c) or (3-104)] will bg approximately 100 kHz. ~~ The closed-loop bandwidth can also be determined from the open-loop gain versus frequency plot. To do this we locate the closed-loop voltage gain value on “the gain axis and draw a line through this value parallel to the frequency axis. Then we project the point of intersection of the line with the curve on the frequency axis and read the value of the closed-loop bandwith. Using this procedure in Figure 3-6, the bandwidth is approximately 100 kHz for a closed-loop gain of 10. 3-3-7 Total Output Offset Voltage with Feedback In an op-amp when the input is.zero, the output is also expected to be zero. How- ever, because of the effect of input offset voltage and current, the output is sig- nificantly larger, a result in large part of very high open-loop gain. That is, the high gain aggravates the effect of input offset voltage and current at the output We call this enhanced output voltage the total output offset voltage Voor. In an open-loop op-amp the total output offset voltage is equal to either the positive or negative saturation voltage. The saturation voltages are specified on the data sheets as output voliage swing. Since with feedback the gain of the noninverting amplifier changes from Ato A/C + AB) [Equation (3-6)], the total output offset voltage with feedback must also be 1/(1 + AB) times the voltage without feedback. That is, total output offset) __ total output offset voltage without feedback voltage with feedback } ~ 1+ AB ‘An Op-Amp With Negative FeedbackeS o a1 ‘ Veer * any Tear AB) is always less than Land + 1 ges the output of an op. Remember that in an ope inpot of an op-amp can cause decause of its very high volt tages. the max amp can reach ‘n-loop configur the output to re: ration even a very small voltage at the ch maximum value (+ Vor ~V.) £ain. Therefore, according to Equation (3-11). for a given op-amp circuit the V_., Viz can be either positive or negath Negative feedback can also be is either positive or negative voltage because ve. used to reduce significantly the effect of noise, ges in temperature on the output voltage fer the value of (1 + AB), the smaller Ply voltages and changes in tempera- 'g amplifier. joninverting amplifier with feedback ex- ‘ect voltage amplifier. That is, it has very high inpat pesiifance. very low output resistance, stable voltage gain, large bandwidth, and very litle (ideally zero) output offset voltage. 3-3-8 Voltage Follower ‘The lowest gain that can be obtained from a non is 1. When the noninverting amplifier is configured for unity gain. it is called a voltage follower because the output voltage is put. In other words, in the voltage follower th: Although it is similar to the preferred because it has much hig! is exactly equal to the input. To obtain the voltage follower from the noninverting amplifier of Figure 3-2, simply open R, and short Ry. The resulting circuit is shown in Figure 3-7. In this, figure all the output voltage is fed back into the inverting terminal of th consequently, the gain of the feedback circuit is 1 (B = Ap = variations in sup tage of a noninventin; ture on the output vol From this analysis it is clear that the m hibits the characteristics of the pe1 ie Output follows the input. screte emitter follower, the voltage follower is her input resistance, and the output amplitude FIGURE 3-7 Voltage follower. 3-3 Voltage-Series Feedback Amplifier 99Y Since the voltage follower is a special case of De noninverting AMDILifigy the formulas developed for th are indeed applicable to the formes ent Saal erhetapnlia " that the gain of the feedbac 1 (B= 1). The applicable formula I . Rye = AR, Reo or = Se = Mo + Veat Voor = ve since (1 + A) =A. : The voltage follower is also called a noninverting buffer because, when Placeg between two networks, it removes the loading on the first network. SOLUTION vn aee Pa Let us first calculate the value of B, Then the closed-loop parameters Ap, Rip, Ron Wand V,,7 can be obtained by using Equations (3-6), (3-8), (3—-9b), (310d), and . (3-11), respectively. Ry a. 1k. wale Rit Rp 1kKQ+10KQ 1 V+ AB = 1 + 200000 ye 1908 100 An Op-Amp With Negative Feedback— 200,000 18,182.8, Rp = 2 MQ(18,182.8) = 36.4 GO. es 18,182.8, Se = (5 Hz)(18,182.8) = 90.9 kHz, +13 Voor = ————— = T= Te rgzg 7 20.715 mv bs = 10.99 ‘oF = 4.12m0 Note that the (==) sign indicates that Vier could be of either polarity In the example above, the voltage gain calculated using the exact equation (Equation (3-2)] is 10.99. The gain would have been 11 if we had used the ideal voltage-gain equation [Equation (3-3)]. Thus the difference error is very small (0.09%) and can be ignored. That is, for all practical purposes we may use the ideal voltage-gain equation, provided that A>> Ay. Remember that as Ap ap- proaches A, the difference error also increases. ? For the voltage follower, B = 1; therefore, 1 + AB = 200,000. To compute the closed-loop parameters, we merely substitute the known values into Equations (3-6), (3-8), (3-9b), (310d), and (3-11). Ap=1 : Rip = 2MQ(200,000) = 400 GA 759 eS = 0375 ma oF = 399,000 = 9375 ™ fr = (5 Hz)(200,000) = 1 MHz +13V = = +65 nV Voor = 390,000 a ‘Thus the input and output resistances of the voltage follower approach ideal val- ues, and the bandwidth is equal to the maximum operating frequency of the op- amp. In addition, since (1 + AB) = A, the smallest possible value for Vir is Possible. 3-3 Voltage-Series Feedback Amplitier 101Feedback circuit FIGURE 3-8 Voltage-shunt feedback amplifier (or inverting amplifier with feedback), 3-4 VOLTAGE-SHUNT FEEDBACK AMPLIFIER Figure 3-8 shows the voltage-shunt feedback amplifier using an op-amp. The in. put voltage drives the inverting terminal, and the amplified as well as inverted out. put signal is also applied to the inverting input via the feedback resistor R. This arrangement forms a negative feedback because any increase in the output signal results in a feedback signal into the inverting input, causing a decrease in the out. put signal. e Note that the noninverting terminal is grounded, and the feedback circuit has only one resistor Rr. However, an extra resistor R, is connected in series with the input signal source vj. First we derive the formulas for the voltage gain, input and output resistances, bandwidth, and the total output offset voltage. Then we study the special cases of this configuration. 3-4-1 Closed-Loop Voltage Gain The closed-loop voltage gain A, of the voltage-shunt feedback amplifier can be obtained by writing Kirchhoff’s current equation at the input node v, (see Figu 3-8) as follows: 3-121) Sipt ly in Since R; is very large, the input bias current /, is negligibly small. For instance, R, = 2MO and Ip = 0.5 yA for the 741C. Therefore, fin = bp That is, ee ee (3-120) 102 An Op-Amp With Negative FeedbackHowever from Equation (2-9), u% 1-6, —% Since 1) = OV, oy v; fe A Substituting this value of vin Equation (3-126) and rearranging, we get int tolA _ (ul) ~ vy RX Ry (G-13) _ ARy : in Ry + Rp FAR, at negative sign in Equation (3-13) indicates that the input and output si os ae snvettion thew : ee of ppposite polarities). In fact, because of th s sion, tion in Fi 8 i inverti Pi ae toon iguration in Figure 3-8 is commonly called an inverting Since the internal AR, >> Ry + Rp. This Ag vu, (exact) gain A of the op-amp is very large (i infini o rge (ideally infinity), means that Equation (3-13) can be rewritten as Ar= Snr =F (ideal) G14) This equation shows that the gain of the inverting amplifier is set by selecting a ratio of feedback resistance Rr to the input resistance Rj. In fact, the ratio R,/R; can be set to any value whatsoever, even to less than 1. Because of this property of the gain equation, the inverting amplifier configuration with feedback lends it- self to a majority of applications as against those of the noninverting amplifier. Let us now rewrite Equation (3-13) in the feedback form of Equation (3-6), for a couple of reasons. First, it facilitates analysis of the inverting amplifier with feedback. Second, it helps compare and contrast inverting and noninverting am- plifier configurations, as we shall soon see. However, to express Equation (3-13) in the form of Equation (3-6), we must represent the current-summing junction at the input terminals of an amplifier as a voltage-summing junction (see Figures 3-8 and 3-9). , ; To begin with, we divide both numerator and denominator of Equation (3-13) by (R; + Rr): : _ ARAR, + Re + AR R, + Rp Ar = 1 (3-15) AK & ~~ T+ AB 3-4 Voltage-Shunt Feedback Amplifier 103RE 104 Summing junction = 8 gram of inverting amplifier with feedback using a voltage. FIGURE 3-9 Block cia Ssdel for current summing. summing junction as o ™ Rr 4 voltage attenuation factor R, + Re Ry gain of the feedback circuit R, + Rr where K = ne ; ion (3-15) with the feedback Equation (3-6) indicates 8 comparison of Equation (315) with sign), the closed-loop gain of the in, fat. in addition to the phase inversion ( loop gi . ving amplifier is K times the closed-loop gain of the noninverting amplifier, . ee ie block diagram of the inverting amplifier with feedback is shown in Figure 3-9. The reason for the block diagram is twofold: (1) to facilitate the analysis of the inverting amplifier, and (2) to express the performance equations in the same form as those for the noninverting amplifier. The block diagram in Figure 3-3 for the noninverting amplifier and the block diagram in Figure 3-9 for the inverting amplifier are identical, except for the K block. However, the major difference is that in Figure 3-9a voltage-summing junction is being used as a model for what is actually current summing. To derive the ideal closed-loop gain, we can use Equation (3—15) as follows. If AB >> 1, then (1 + AB) = AB and GB-16) 3-4-2 Inverting Input Terminal at Virtual Ground Refer again to the inverting amplifier of Figure 3-8. In this figure, the nonin- verting terminal is grounded, and the input signal is applied to the inverting ter- minal via resistor R,. However, as discussed in Section 3-3-3, the difference in- Put voltage is ideally zero; that is, the voltage at the inverting terminal (v,) is approximately equal to that at the noninverting terminal (v,). In other words, the inverting terminal voltage v, is approximately at ground potential. Therefore, the inverting terminal is said to be at virtual ground. This concept is extremely use- ‘An Op-Amp With Negative Feedbackjul in the analysi cjosed-loop gain pi as follows: In the circuit of Figure 3-8, of closed-loop invertin tation — 14 i amplifier circuits. For example, ideal )] can be obtained using the virtual-ground con- Q-17) That is. However, Therefore, or This is the same result obtained in Equation (3-14), 3-4-3 Input Resistance with Feedback The easiest method of finding the input resistance is to Millerize the feedback re- sistor Ry; that is, split Rr into its.two Miller components, as shown in Figure 3-10. Ry FIGURE 3-10 Inverting amplifier with Millerized feedback resistor. 3-4 Voltage-Shunt Feedback Amplifier 105In the circuit of Figure 3 10, the inj R, ye (exact) na Rea RT ya Since R, and A are very large Rr Jason +A Hence (ideal) Ri + Ri tance with Feedback he resistant t resistance of the m 3-4-4 Output Resis! The output resistance with feedback Ror ist a nal of the feedback amplifier. The uit ing Thévenin’s theorem, plifier was obtained by using aan everting amplifier. Thévenin's equivalent circuit lifer is shown in Figure 3~11. Note that @ Ptly the same as that for the noninverting ampli put resistance Ror of the inverting amplifier mus reninverting amplifier [Equation (3-96)]- Specifically, 106 An Op-Amp With Negative Feedback. put resistance with feedback Rip ig th nh Outs, Guy, ce measured at the outpy, oninverting am. ‘and we can do the same for t for Rg of the inverting am. his Thévenin’s equivalent circuit is ex. fier (Figure 3-5) because the out- t be identical to that of the G-20)output resistance of the op-amp n-loop voltage gain of RAIN Of the of n of the feedback circuit 4-4-5 Bandwidth with Feedback enrioned previo C pai As ote conte tte gain bandwidth product of a single break frequency Lye atways less than the pain eit tha the gain of the amplifier with feed. n010.0 0 Oooo 3-4 Voltage-Shunt Feedback Amplifier 111As mentioned earlicr, current shunt and current-series feedback configura selelom find use in practice: therefore, they will not be discussed here, He interested readers may analyze them following the same procedure as wet’, voltage-series and voltage-shunt feedback configurations, ed f, 3-5 DIFFERENTIAL AMPLIFIERS We will now study the closed-loop differential amplifier configurations. Spa, cally, we will evaluate (wo different arrangements of the differential amplifier 4," ative feedback. We classify these arrangements according to the number of op amps used. That is, 1. Differential amplifier with one op-amp 2. Differential amplifier with two op-amps Generally, the differential amplifiers are used in instrumentation and indus. trial applications to amplify differences between two input signals, such as the outputs of the Wheatstone bridge circuits. Differential amplifiers are preferred in these applications because they are better able to reject common-mode (noise) voltages ‘than single-input circuits such as inverting and noninverting amplifiers, They also present a balanced input impedance. 3-5-1 Differential Amplifier with One Op-Amp Figure 3—14 shows the differential amplifier with one op-amp. We will analyze this circuit by deriving voltage gain and input resistanee. A close examination of Figure 3-14 reveals that a differential amplifier is a combination of inverting and noninverting amplifiers. That is, when v,-is reduced to zero the circuit is a non- inverting amplifier, whereas the circuit is an inverting amplifier when input v, i reduced to zero. Re Ry ve FIGURE 3-14 Differential amplifier with one op-amp. Ry = Rp and Rr = Rs. 112 An Op-Amp With Negative Feedback3-5-1(2) Voltage gain, = grout in Figure 3-14 has tw ‘osition theorem in order When v, = 0 V, the con «put due tov, only is © inputs, vy, and v,: i to ests Uy We will, therefore, use tt figuraiioe the relationship between inputs hy Pecomes an inverting amplifier: hence vy, = — Rely) ag jess similarly. when v= OV, the confi 1 Eo W ieee Jivider network composed Gf Re ey 4 Roninverting amplifi fier havi fore. 2 and R; at the noninverting input, There, », = Rey Ry + Ry That is. vy = —R_ (Rit Re R+R\ RK, )% Since Ry = R, and Rr = R;, Rr(vy) ae loy R, (3-24b) Thus, from Equations (3—24a) and (3—24b), the net output voltage is Uy = lo = Vor + Ugy es . Uo R (uz = vy) = or the voltage gain R Ad 7 in @-25) q Note that the gain of the differential amplifier is the same as that of the inverting amplifier. 3-5-1(b) Input resistance. The input resistance Rip of the differential amplifier is the resistance determined looking into either one of the two input terminals with the other grounded. 3-5 Differential Amplifiers 113Therefore, with v, = 0 V. the circuit in Figure 3-14 is an inverting amplifie, input resistance of which is 8-260) Similarly, with v, = 0 V, the differential amplifier of Figure 3-14 becomes a non. inverting amplifier whose input resistance can then be written as Riry = (Ra + Rs) (3-26) ) and (3-26b), it is obvious that the input resistances seen by the signal sources v, and 2, are not the same. This inequality can be corrected and both input resistances can be made equal if we modify the basic differential amplifier of Figure 3-14 as ‘shown in Section 3-5-3. However, for the differen- tial amplifier of Figure 3-14 to perform properly, both Ri and (Ry + Rs) can be made much larger than the source resistances so that the loading of the signal sources does not occur. If we need a variable gain, we can use the differential amplifier of Figure 3.15. In this circuit R, = Ro, Rr = Rs, and the potentiometer Rp = Ry. Therefore, depending on the position of the wiper in Rp, voltage gain can be varied from the closed-loop gain of —2R;/R, ‘to the open-loop gain of A. From Equations (3-262 An Op-Amp With Negative Feedbacks nd Rip 'n Equations (3-25) and (3-26), we obtain Ay 10kQ. Ap = — 1kQ 10 Rie = 1kQ, R, Fy = 1kKO + 10KQ = 11k, b, Rearranging Equation (3-25) yields Vp = —Apy, ae a = ~(10)(2.7 ~ 3) =3y Pp sine wave at 100 Hz 3-5-2 Differential Amplifier with Two Op-Amps Recall that the gain expression for the differential amplifier of Figure 3-14 is the amplifier of Figure 3-8 [see Equations (3-14) and ial amplifier and also increase © op-amps. Such a circuit is shown in Figure z : this amplifier are identical to those of the non. inverting amplifier. 3-5-2(a) Voltage gain. A close examination of the circuit of Figure 3-16 shows that it is composed of two stages: (1) the noninverting amplifier, and (2) the differential amplifier with FIGURE 3-16 Differential amplifier with wo op-omps. 3-5 Differential Amplifiers 115Obtain the ove, unequal gains, By finding the gain of these (wo stages, we can “ gain of the circuit as follows: The output v, of the first stage is n) : =(1+5) =(14R)™ By applying the superposition theorem to the second stage, we can obtain the it voltage: x 2 Rd 4 (14+ Bo, a 8 = ~~ Re R ‘Substituting the value of v, from Equation (3-272), we get Re Rr By, + +(1 + Be), ao -G0+R) Since R, = R; and Rp = Ro, MRE Uy = ( + Be — wy) Therefore, where vp, =u, — Uy 3-5-2(b) Input resistance. The input resistance Rj, of the differential amplifier is the resistance determine looking into either one of the two noninverting input terminals with the othe grounded (see Figure 3-16). Note, however, that the first stage (A,) is a non verting amplifier; therefore [from Equation (3-8)], its input resistance is Riry = Ri + AB) (e284 where R; = open-loop input resistance of the op-amp ei Ry * i R, + Ry, Similarly, with v, shorted to ground (v, = 0 V), the second stage (A) also b comes a noninverting amplifier whose i input resistance can then be written as Ryy = Ri(1 + AB) am = ope feen input resistance of the op-amp R Ry + ReHowever, since Ry = Rs and Rp = Ro, the Riry # Rpg. Dading of the input sources v and 1, may oceu, icine ene Kiy & Rica the roading te smaller in amplitude than expected. This possible reduclion it eee a atthe output signal isthe drawback ofthe differential ampli on inthe am plitude Caventheless, with proper selection of components, both 2 ene ae isde much larger than the source resistances so that the loading of Ber can pu sources does not occur. SOLUTION a. From Equation (3-28), 6:8kO 680.0 Ap = 1+ ‘The input resistance can be calculated using Equations (3-29a) and (3-29b). 3 (2)0°)(6.8 “a e Ray 5 OMO) [: + eins oon | 2 S8 hentia (2)(10°) (680) a a Rip, = (2MQ) [: ‘oe sea "a0 ho 36.4GQ b. Output voltage can be calculated by rearranging Equation (3-28): R w=(1+ ay = (11)(-1.5 + 2) = 5.5 V pp sine wave at 1 kHz 3-5-3 Output Resistance and Bandwidth of Di intial Amplifiers with Feedback Resa that the closed-loop op-amp configurations—noninverting, inverting, and ifferential amplifier—employ negative feedback and use the same type of out- Pol goanections, Therefore, the output resistance of the two configurations must ¢ identical. In other words, the output resistance of the differential amplifier 3-5 Differential Amplifiers 17averting amplifier, excepy 4, should be the same as that of the noninverting oF B = WAp. That is, (339 ) Ror = 4 AAD | amplifier ain of the differential of the op-amp of the op-amp f the differential amplifier configuratic,, ifiers, the bandwidth of dit i gain of the amplifier ang is where Ap = closed-loop g R,, = output resistance = open-loop voltage gain 0 for each Oo! rting OF inverting ampl ds on the closed-loop Remember that Ap is different ‘As in the case of noninve ferential amplifiers also depen given by _ unity gain bandwidth F ~ “closed-loop gain Ap (3~31a) or, pw DD fp = Ap (3-31p) where f, is the open-loop break frequency of the op-amp. 3-6 PSPICE SIMULATION