Network Functions aria ee INTRODUCTION ‘tis chapter, the concept of transform impedance and transform admittance is studied. inter, a function relating currents or voltages at different parts of the network, called a sosée function, is found to be mathematically similar to the transform impedance function. lee transform impedance functions are called network functions. 2 PORTS AND TERMINAL PAIRS I “ez 16. (a) is a representation of a one-port network. The pair of terminals is customarily il 4toan energy source, which is the driving force for the network. The pair of terminals tceore known as the driving points of the network. Figure 16.1 (b) is a two-port network. aa | tT I-02 | vo— }—o2' @ (o) Fig. 16.1. (a) One-port network (b) Two-port network (0) m-port network 723nnected to the driving force and is called the in, Put the output port. There may be n-port oe aa ork, ie are Sire The port 1 - 1’ is assumed to be co i is called ; Seay | ae n-terminal networks in some cases, which i ig. 16.1 . There may be ae Batiste Our emphasis will be on one-port and two-port network. The array of quantities enclosed by straight-line brackets @yy yg 413 -+- An Gq, pq M93 ++ Fan Gn Gnz ng +++ Ann is known as a determinant A of order n. The element aj, is in the ith row and jth colum, Second and third-order determinants have expansions, that are familiar from studies jy elementary algebra as A= ay M2 Cece fegcasat eacieeal Oi) 12" 18 and A=| 421 422 423 31 932 9g M20 pq Ong G2 9413 32 433 Pao. Gea, O32 O33 =U ay, +a ay +0" ay ul @gyfbgg ~ gy lbgg) ~ py (@yoftgg — AyyGtys) + Oy(A gpg ~ Bpyftyg) = where Aj, is the cofactor and can be written as Ay =(-Li** My M,, is the minor which can be obtained by deleting the ith row and kth column determinant A. A determinant of order n is equal to the sum of the product of the elements of a or column multiplied by their corresponding (n — 1) order cofactors. Applying this rule expansion of the third-order determinant along the first column, we get = M. A= ayyAyy + Og Ayy + Oy, A5, = 044M, — 2gMoy + Aart from be = 432 Gag, ay. a ay 413 =a, paren nia O38 | M1} d32 3g || ag9 agg] 21] a2 923 Now, by KVL % by KVL, the generalized mesh equations ean be written as a6! ZI=Vfar Fra Ang tnn Jn Ln By? mer’s rule, the solution, 7, ean be written as . T= DY wy Vjsia hn (16.2) im ser Ais the loop-basis system determinant, i.e., A = |Z| and admittance matrix Y having elements 4 eae. y= (16.3) 4, is the cofactor which can be obtained by deleting the jth row and ith column from the ‘topbasis determinant A and multiplying by (- 1). Inasimilar way, by KCL, the generalized node equation can be written as Ye=I (16.4) Mn I DIV - Ye Yon +» Yon || Vo |_| Lo Yn Ina» Inn |LVn) Un ‘i general, the solution can be written as pee i af bet © Vis o Ziply EEA yn! (16.5) “here 4 kel i naa the node-basis system determinant, i.e, 4’ = |Y| and A’,, is the cofactor of A’, matrix Z having elements 2, a8 Si 46.6) Rat Ae,has been defined as the ratio of the voltage own in Fig. 16.2, with zero initial condition, pt controlled sources. Thus, transfor, ‘The transform impedance at a port Sand wi With ny current transform of a network, as sh internal voltage or current sources exce] Vis) As) = Ts) ie Similarly, the transform admittance Y(s) is defined as the ratio Ks) 1 M9 = Ve) Ze) ig Fig. 16.2. Two-port network Note that the transform impedance and transform admittance must relate to the same port. The impedance or admittance found at a given port is called the driving-point impedance or admittance function. The two quantities are assigned one common name, immittane function, which may be either an impedance or an admittance. ‘The transfer function is used to describe networks which have at least two ports. In general the transfer function may have the following possible forms: (@) The ratio of one voltage to another voltage, and is called voltage transfer functioo- (@) The ratio of one current to another current, and is called current transfer functo (iii) The ratio of one current to another voltage or one voltage to another current andis called transfer admittance or transfer impedance function. In case of a two-port network, the output quantities are V,(s), 1y(s) and the i? ‘Quantities are V,(s), 1,(s)There may be four transfer functions, as mentioned below . inction Gy(s) = Va(s) asa) V\(s) 10) function (6) = a as 1s, A 1) funetion ¥,(0) = 220 (16 Vi(s) 16 Note that, for « one-pg (9) Sad Zig# Wy Gy # Voryy hetwork, Z(s) = 1/Y(s), but for a two-port networkpoe eS ‘Asan illustration, the driving-point impedance function for tho sorion RLO eireult in fu 183 @is 1 LCs" + RC; Zs) = Re Ls + a (16.18) ‘The degree of numerator and denominator polynomials are two and one, respectively, CWO R Ls A z = Ves = i = {te oo! mt it (a) (b) 18) IAs) =0 1,(6) 1h(s) =0 R Ls cs i p fad 7 @ (@ i) wes Is) =0 : va(s) nage we “ «Fig. 16.3. The network for whieh driving-point function has to be obtained _ The driving-point impedance function of Fig. 16.3 (b) is yin 1 at 2 wee RE (16.14) het Pie a a = Cs a) On RLatTG (16.15)(aii) The voltage tranator function of the network in ig 16.5 (0) can jy mi Mii, 1 Vy Ww Vylw Cu! WhO So “ “ On" VO "Tey Lp wy 0 RE Gr} (My under no-load condition, ie 1,68) # 0. ‘The driving-point admittance funetion under no-load condition in Its) 4 4 a a Yy,@) © Via) Ri at ‘ai (1949 {i) The voltage transfor function of the network in Fig, 16.3 (d) can be writh, y Written ag i Ci LC Gy) = tae . Ca eS . ; 1 (ue Z,] 4 cee i ‘under no-load condition, ie., [,(s) = 0. {v) The voltage transfer function of the network in Fig. 16.3 (¢), under no-load conditg, ie, I,(s) = 0, can be written as G,(6) = Vals) _ 1,(8) V0) Fang (16.19) (a+ aeuR ii 1 ae Cae, (1630 Now, the transfer function of this network G(s) = (162 st Ry+ Ry RRC = We oF nin automatic control systems and is known as lead compensating etwas et nee rk tons (@riving-point immittance fneoot ee : /e computed so far are the ratio of polynomials is s having the se form as Beat toe tt 18 on by. 8) Bg + by8 Th By 8+ Oy __ which is @ rational function of and m, n are integers. ‘The degree of the nume' i rator poly™ the denominator polynomial is m,=] a ay pa 1% | t oe eo LS Fig. 16.4. Ladder network ‘The simple ladder network is shown in Fig, 16.4, The Zand Y' i ee he Z and Y's are the impedance and tance of the elements of the ladder network. The driving-point immittance function will ke eee ee ey Z3+ 1 Y, att: T (16.23) pe Ye This is known as continued fraction. Write the driving-point impedance of the ladder network in Fig. 16.5. eeu 1H Z=s8 vy Altace Fig, 16.5. Network of Example 1 fn psh + Bal 41 (16.24) Zyse+ if 8° +28 at a+= cs V, (8) a; (16.25) ea a= TG) aor +D Vals) 1 Su» Vy) +087 +1 (16.26)The driving-point 4s+1 16s+8 8 a + Bee + el088 +802s° + 752s + 604s° + 242s” + 78s +16 + 1885" +3585" +6265° +710s° + 7585" + 470s° + 330s" + 785+! 1 where, Bap = ads 1 _s?+2s+1 Bite i a ast 2s +2 8 impedance function of the ladder network in Fig. 16.¢ can be Wittai o pridged T-network in Fig. 16.7 (a), eval forte nittance Yo. valuate the driving-point admittance Y,, ‘ra a the Joop-basis system determinant A can be written as e rales, re : os 8 Ate Pal odggo id 8 (16.27) a hae} Rr peg ost, 8 s 19 Ge , 1 1F He ha Ou on uJ afte 19 z 10 10 10 (a) (6) Fig. 16.7. (a) Bridged T-network of Example 3 (b) Ladder form of network when output is shorted Expanding the appropriate cofactor A,,, we get loa Ay=Cyrs Be (16.28) a B 2 ce a ‘i 8 definition, the driving-point admittance Y,, can be written as “aes Au _ a +58+1 (16.29) Te A 8? +58+2 , transfer admittance : (-0?t! rn i gs? +25+2 ) 4, 8 -_—— (16.30) eee, 2.4/2 Note Tye ‘om to ladder form, as ‘heap the output port is shorted, the network reduces& om ‘The other three 2er08 are ats = », A dissipationless (lossless) network h™ oe * hich is a ratio of two polynomials of ‘sean, We have shown that the network function, rewritten as pls) _ 208" $ays" 1 tot dq 18+ dq ———————— Ms) = as) &s” $59" be Hb 8+ py (16.3)) where a, b’s are the coefficients of real positive value. Let p(s) = 0 has n roots as z : and g(s) = 0 has m roots @8 Py» Pay» Pe Then N(s) can be rewritten as My ya (e=2;X6- 2g)». (8 — 2n) No) =H Tp Ne= Pa)» 8 — Pm) (16.39) elare = scale factor. For s = z,; 1 = 1, 2,...,n, the networ function vanishes. Such complex frequencies are known as the zeros of the network function Fors =p, ;1= 1, 2, . m, the network fanction becomes finite. Such complex frequencies are known as the poles of the network function. Poles and zeros are useful in describing the network function. The network function is completely specified by its poles, zeros and the scale factor. The poles and zeros are distinct when z; # Z; and p; #P; for all possible i, j. ros have the same value, the pole or zero is said to have repeated When r finite poles or ze multiplicity r. If finite poles or zeros are not repeated, it is said to be simple or distinct. ‘When n > m, then (n — m) zeros are at s = , and for m > n, (m —n) poles are at s =~. If, for any rational network function, poles and zeros at zero and infinity are taken into account in addition to finite poles and zeros, then the total number of zeros is equal to the total number of poles. The symbol x is for pole and o for zero. Poles and zeros are critical frequencies because, at poles, the network function becomes infinite, while at zeros the network function becomes zero, At any frequency other than poles and zeros, the network function has a finite non-zero value. As an example, the network function ees (s+ I(s+2+ js +2-jD ad te 46g Sd thes finite poles ~2~j, as shown in Fig. 16.8, Note that complex poles and zeros are in con) A network function Ma) = 7 i (s +2) (6+ 1+ jXo+1—jXe+1+ j2Xs+2-J2 has one finite zero at » = ~2 and four finite poles at #=-14/,-1-j,-2+2,-2-j2. ag/by is a constant known as the Ms) = gen i228 agate. of ps jyimos” - Poles and zeros. 3 only A network function havi : Ga polen bob nagative, Ce nee et and complex poles and zeros i stable if ., the poles are lying in the left-half of the s-plan’ the rea! po3 polos 2 zor08 | Fig. 16.8. Pole-zero configuration in s-plano Vout (8), Vin (8) atc, (16.33) . Veui(8) = Gai(s) Vi,(8) = > j=l Pj tal out s d vt)=L7V,(s)= )) kexp(pjt)+ D) kexp,t) (16.34) jo ist tizeu is the number of poles of G,,(s), v is the number of poles of V;,(s), fs, and h,’s are mses, s are the natural complex frequencies corresponding to free oscillation and s, are ple frequencies of the driving force corresponding to forced oscillations. ‘num and Non-Minimum Phase Function “evil discuss the minimum and non-minimum phase functions. Stable network functions Uilllst-halt plane zeros are classified as minimum-phase functions; those with any zero fonathalf plane are non-minimum phase functions. From the name it is clear that it is se tative nature. If the zeros of a network function are all reflected about the jo axis, will be no change in magnitude of the network function. pel only difference arising will be in phase shift characteristics of two functions are ig itean be seen that the net phase shift over the frequency range from zero to infinity # network function having all its zeros in the left-hand side of the s-plane. the network function G,(s) and G(s) as +z snes STE and G, = stp arp G(s) = bee patterns of G(s) and G,(s) are shown in Figs. 16.9 (a) and (®) aes eat G(s) and G(s) have the same magnitude shareegristie _ ee but their phase characteristics are obviously different,ar ate (@ Gie)= ON aes Fig. 16.9. (a) and (b) Pole-zero patterns ‘The phase shift of G,( jw) at any frequency © = ©, can be written as 2G,(jo,) = 0, - 8, = Z tan“ (w,/z) — Z tan“! («,/p) which is an acute angle for > 0 and clearly ranges over less than 90° as is apparent from the figure. ‘The phase shift of G,( jo) at any frequency w = «, can be written as £G,{ jo,) = 0,* — 8 = (m— Z tan“ («,/z)) — Z tan“ (w/p) which is an obtuse angle for w > 0. Strictly speaking, the phase shift of G,(jw) is acute and that of G,{ja) obtuse. An all pass network is realized with symmetrical lattice network as in Fig. 16.10 (a) ‘ith all component value as unity. Derivation of the transfer function, that is Yet) a flo: in) = Vous(s) _ Vy(s) ~ Vo(s) i ‘ Vi, (8) Vin (8) hie Vic (aoe, Vis) vu so Ato aut sD +1 — KCRDe+1jeal pattern of poles — 0.5 J) /O75 and zeros am r 0.8 8) 07K j - in Fig. 16.10 (6). Again the magni 8 *J/0c76 is obtained for thi in ig itude remains co: < is gotheyio, whereas the phase characteristic varios from 0° to sient ie frequency as a 16 phat & = 180°~ 6, and O*, = 180°— 9%, The pha 0 860°. Tein evident Ron 8 ae jsanon-minimum phase symmetrical lattice are ma G0). = 200; + 64,), "sn telephone communication networks, ork and has wide applications o @ S 1 o =180°} ----------- : ® inn o 3 c) ai PEIEIO (9) A pase network (0) pote-zr0 pom (0) trequency response of Magnitude cure, and (d) frequency response of phase curve X, Secu for a network function to be a driving-point function with common Ln Polynomial p(s) and denominator polynomial q(s) cancelled are listed ts in the polynomials p(s) and q(s) of N(s) = p(sv¥q(e) must be real and * aoe ae es ah @ginary poles and zeros must be conjugate.5 must not be positive. Fa 3. (a) The real part of all poles and zero: (6) If the real part is zero, then that pole or zero must be simple. 4. The polynomials p(s) and qs) lowest degree, unless all even 5. The degree of p(s) and g(s) may differ 6. The terms of lowest degree in p(s) and g(s) may di must not have missing terms between the hi or all odd terms are missing, by either zero or one only. hes any iffer in degree by one at the moss tions for a network function to be a transfer function with common factors Necessary condi pis) and g(s) cancelled, are listed below: 1. The coefficients in the polynomials p(s) and q(s) of N(s) = p(s)/q(s) must be real, ang those for g(s) must be positive. 2. Poles and zeros must be conjugate if imaginary or complex. 3. (a) The real part of poles must be negative or zero. (®) If the real part is zero, then that pole must be simple. This includes the origin. 4. The polynomial q(s) may not have any missing term between that of the highest and lowest degree, unless all even or all odd terms are missing. 5. The polynomial p(s) may have terms missing between the terms of lowest and highest degree; and some of the coefficients may be negative. 6. The degree of p(s) may be as small as zero, independent of the degree of q(s). 7. (a) For G,, and o,,, the maximum degree of p(s) is the degree of q(s). (6) For Z,, and Y,», the maximum degree of p(s) is the degree of q(s) plus one. To find conditions which are both necessary and sufficient would require that we understand how to find a network from the given network function, a topic which is included in the study of network synthesis, However, a hint has been given in the following section ‘obtaining the network from the given network function, equations are suitably used in the design or synthesis of different ty?" passive and active networks, Find the network functioy The first step in the aerate for the bridged P-network shown in Fig. 161° s-domain, as shown in Fig. 16,11 (b) ah express the admittance of each element” respect to ground ( : In such a circuit, the voltages at nodes 1, 2 and 3: node), are designated V\(s), V.(s) and V,(s) respectively:Fig. 16.11. (a) Bridged T-network of Example 4 . (b) Transformed network ‘Thenode equations obtained by using Kirchhoff’s current law at nodes 1, 2 and 3 are as. is: Node 1: 1 ; RH Vi Va) + 8, (Vy Va) =F ‘ 1 ot i... [Free] mg MO =h 1 (W,-V,) + Bey Va— Vs) + 80,Va= 0 Ju-g%-0 a Wy, -V,)+sC\Vs-V)=0 (16.35) (16.36) (16.37) (16.38)i at <,| =— +sC,| — +—-+s8C, RR, li Ry fe aeze-«)-a)) 28 eae +B. G so) RR, Ry RR, RR 24 6{—1 ‘ge eeoe aa aa) aE os a 0) { 2 2 RR, RRC, :. . is 18 It is observed that the nodal determinant is symmetrical about the aoe characteristic property of the nodal (and mesh) determinants of RLC networks. Sul ealize the function: Vois) aS testd (6 Vil) “ns? + as a ible to re & d)n and a are Positive numbers. It should be possible tained “ment values of the desired circuit can be Fig. 16.11 (a), The # ofs in Eqns, (16,39) and (16.40), a8 : 2oinnship ae cd (16.41) sbecapacitor C, is given by n_a Ce a (16.42) (ne choice of resistors is R, = R,, which leads to 2 R,=R,=7. (16.43) Substituting, the remaining unknown C, is 2 = (16.44) ste re é ‘Thus we see that if the coefficients, as in Eqns. (16.42) to (16.44), satisfy the relationship rg the network function of Eqn. (16.40) can be realized by the circuit in Fig. 16.11 (a) values as 2 2 C= 2 C, epee Ne ieeaaure cd n n ; lagul® ove discussion indicates that the results of anlysis can be used for the synthesis of functions. in Fig. 16.12, the poles s, and s, are 4 tr & itierens array of poles in the s-plane, as shown Fi tadpole and the response the "2% conjugate i and, ; Poles pole pairs. Further, 8, a Eo cacsieae ‘si *, and s, converge monotonically. 8, and 8, MAY correspond to the q & % + Bos +42. : : eet (16.45) Sy 84 =~ 80, # 0, 9-151. Fig. 16.12. Array of poles ‘Te contribution in the total response, due to poles s, and 84, is | K, exp (s,t) + K, exp (s,t) ‘The contribution in response due to the pole s, is predominant com ‘pared to that of, as | s, | >> | s, |. In this case, s, is dominant compared to Sq: The pole s, is the dominant ig. ‘amongst these two real poles. The response due to pole at s jg dies down faster compared to tha, ofs,. (16.45) Now, let us discuss the complex conjugate poles s, and $,* which belong to the quadrats " 8+ 280,54+0,2;5<1 (1647) ‘The roots are $0 $4" = — 80, + jo, 1-8? (1648 plex conjugate poles 84) 8, is 1H eH Bayt) x exp (—jo,( fT 5) 80,0 gives the mo fut notonically decreasing function, whereas th? sWained sinusoidal oscillation, ‘The resultant will sve 8 shown in Fig, 16.13, The rate of decay of the sinusoidal “Omplex poles (8a, ), wl tos, “omplex conjugate pole pair. The response dU® in njugate pole pai com aes , < and 8% are also the f shat due to 5, and » Hence, s,, and s¥ are the domin" a Plex pole pair s, and s,4.lt Ils) = Yi) Vis) = plsvals) = H S&—8— (= 8,Xs~5, Fatiesale factor. The time-domain response can be obtained by taking the Laplace inverse transform = te partial fraction expansion, as K, + Ke, Mh , Ke, Ke, Ka 82s, s=s, s~s, 8-s, s—s, 8-84 i= £419) = 24 : | (16.50) “eve rsduesK, and K,* are complex conjugate, as also K, and K.*. ay resid K, can be obtained as K,=Hx (s~s))...(s~8,) 5@-8)) (16.51) (s—8,)...(8—8,)...(8- 8p, This equation « "tion ation is composed of factors of the general form (s, ~s,), where both s, and s, : Plex .(s, ~5,) is also a complex number expressed in polar coordinates, =M,, exp (j4,,) (16,52) Henge, M,M,, ..M, K,= ici i( ee SH x MM, Bi %P Un +4 FG, Gy ~ oo) 1658) § ete ead en converge, all'poles of the network function must lie in the left half of 4 Sea (the imasn’”€T Occur in the right half of the s-plane. Poles and zeros can be on the ‘Ne tranaton? axis) subject to the limitation that such poles and zeros are simple. Pair with double pole locations at +,jo,, as in Fig. 16.14 (a), and we getResponse envolope xt” (a) (b) Fig. 16.14. The inverse Laplace transform of the function having poles and zeros of (a) shown in (6). ‘The response is unbounded, as shown in Fig. 16.14 (b). Response for multiple lesa seros at other locations in the let-half of the s-plane are all bounded since such terme a ‘to terms of the form ¢” e~, and in the limit jt me =0 Hence, the response is bounded. “The closed-loop transfer function C(s/R(s) can be written from Fig. 16.15 as OO a, Ris) 5? +280,8+ 02 (16.55) (16.56) (aes? 6sy ing ratio and is dimensionless oe j= damping undamped natural frequency, i.e, frequency of oscillation when damping ratio is zero (5 = 0) <0, (1-8? = damped frequency of oscillation The dynamic behaviour of second-order system can be described in terms of two 0, * eters, 5 and ©, yo<8<1, the closed-loop poles are complex conjugate and lie in the left-half of the s- ‘The system is then called underdamped, and the transient response is oscillatory. If 6 «), the system is called critically damped. Overdamped systems correspond to 5 > 1. The transient response of critically damped and overdamped systems do not oscillate. If 5 = 0, the tansient response does not die out, gives sustained oscillation. ‘The rots of the characteristic equation vary widely for different values of the damping ratio. For 8<1;5,,s,=-Sa, jo, J1- 5? > Re [s,s] <0 (complex conjugate roots in underdamped case) 8=1;5,,8, =— 50, (negative real repetitive roots in critically damped case) 8>1;5s,,s,=— 80, +, J1- 3? (real negative roots in overdamped case) 3=0; s,s, =+/o, (sustained oscillation) 6<0;8,,s,=—80, +jo, f1- 8" ¥, (positive roots in negatively damped case and response diverges out) ity all now determine the response of the system shown in Fig. 16.15, to a unit step @ wasiteramped case (0 <6 < 1) “ase, C(s/R(s) can be written as 2 Gis) | cc} (16.59) = Ris) (s+ 80, + jog N(s +80, ~ Ja) a Nag, ®W1=, The frequency «, is called the damped frequency of oscillation, For a "Pt Cle) can be written as Cs) = on ; (16.60) PE oa : (8? + 250,.8 + 048 Ne inoue A r(t) = U(t)=1 RG) = £ r(t)= VsG(s) can be obtained easily if C(,) ; The inverse Laplace transform of O18 ‘ 0 wit ga form, OD C8) = 5 574 280,8 + OF mm ne ire 1 Oy = 5 Gtb0,)2+0% (s+50,)? +04 etG ia eres 5 ork Oy Sede «, et) = 6 5) +80," +0, fine +80) 757 isis S sin @yt eft) = 1- exp (— bui,¢) | 008 gt + FF sin toy (16.61) where, @,= 0,1- 3° ‘The response is damped sinusoid with overshoot and undershoot, as shown in cune (a) of Fig. 16.16. (Gi) Critically damped case (6 = 1) c(t) = £7) = £1 an s(s+o,)” a: _ -=1-(1+0,1) exp —0,t); 20) on _ +e response is increasing from zero value, monotonically towards the steady sil ‘value unity, as shown in curve (b) of Fig. 16.16. ns Bie septerth weinse9 nse Paw he? 2281 Pt 4ee =1+ Kye" + Kye™ re : . (16.63) residues and s,, , are the real negative roots for 5 > 1, 8,97 — 5a, £0, 87-1 5551 (16.64) i increasing from zero value towards the steady sta a te {c) of Fig. 16.16. The response is slower than the citcally pier oa curve a - peau diffrent ranges of values of 5, the characteristics of the system is given in al Table 16.1 Critically damped Underdamped Roots are negative real Roots are complex, and repetitive having negative real parts Not oscillatory Oscillatory Asymptotically stable ‘Asymptotically stable Rise time lesser een the roots of a characteristic equation ial distance from the roots to the i of the time real axis, i.e., pee (16.65)This signifies that ifroots are changed duo to the variation of ¢ (1) then the roots will lie on the Tine 8 * co# 0. "h Figure 16.18 shows the constant @,-loci, the constant loci, the @ and loci, It may be noted that 0 < 6 < © corresponds to the negative fonatant 4 (mg, -e< eam to tho positive half of the «plano, Por 0< 6 « w, ie Of the », lage In settle constant coda value because of the nogative Nerhine top report Negative damping wie aint @ response that grows without: hound ie xp ( lor damping corresponds to sustained oscillation. The effect of roots of the cha he ror val ", PMA ACLEry on “ODI tony rn Hi ‘on the damping ratio of'a second ord 5 ler aystom is shown in Figs, 16,19 and 16,20, Me oan Positive damping & » 0 rstl)o O>b>-1 7 (e) + cit) (f) Fig. 16.20. Effect of damping on roots and response of second order system (a) Negative real distinct roots (b) Negative real repetitive roots (¢) Complex roots having negative real parts (qd) Imaginary roots (@) Complex roots having positive real part (f) Positive real roots In a mechanical system, os aid of a 2N force as step input to the system, Determine thi dea with the parameter values of the given mechanical system. ca sn i ic! i fem lrepeee a the paramter values of M, D, K of the mechan yi ae 16s shown in Fig, 16.21 (a), the response is as show" in Pe a in the el values of L{M}, CLK), RUDI, ilx| in the series RLC ‘ oy The at the given mechanical system from /-v analogy. function of the mechanical system is X(s) a i 1 where, Xs) = Lxit), X9) Mo" + Ds+K s?+280,8+0! nc} Xi0) = £3 (0,3 = damping ratio, @, = undamped natural fea?"2 ~ (Ms? + Ds + K)s the final value theorem, the steady state value wie) = Lt, aXis) = UK = 0.1mfi ft yo network in defined a6 the ratio of the Se ore a: ileum of input, with all initial OmMBiNions v4. transform impedance and admittance function are the special case of network tunes For the sinusoidal case # = jo, the network function becomes # complex Hupp gives a direct relationship between the input and output phasors in the shondy frequency «, In case of linear lumped time-invariant circuit, the network function 19 4 function with real coefficients, In addition, for a passive network, all the coefficients are network function is the Laplace transform of the impulse response, Vth y 1, Ving on Her hig, Hale, of 5 ms Witine, 4 The poles and zeros are intimately related to the frequency TERDONKE 48 Well ag thy impulse response. Depending on the position of the poles and zeros, the response of the them, can be predicted. Poles lying in the negative half of the s-plane will ‘The system will be stable. On the other hand, a system having the s-plane will be unstable. The system response will be oscillatory for the damping ratio, Tange 0 < §< 1. The system is underdam will be critically damped for 6 = 1 and critically and overdamped cases, Kive a converging resyope, any pole in the positive hall f having values in the ped. The roots will be complex conjugate, The response overdamped when 6 > 1, The roots will be real for toh 1. Van Valkanburg, M.E., Network Analysis, Prentice-Hall of India Pvt. Ltd, New Delhi, 1974. 2. Kuo, BC., Linear Networks and Systems, McGraw-Hill, New York, 1967. 3. aa G., Principles of Active Network Synthesis, John Wiley & Sons, New York 4. Balabanian, N. and B: New York, 1969, L Show that the voltage transfor the network shown in aout ets Fig. P. 16.1 can be Vals) _ Beare be Fy Vi(s) ~ ags* + ays” +a, +20 where, a, =b, = R,R,RyC:CCs es = RaIR,CAC, + C,) + (Ry + RCO? % = RalR,CA(C, + C,) 1 +R REE ickart, T.A., Electric Network Theory, John Wiley & Sons b, = RAR, + R,)C,C, 4% = RC, +C,) +R,C, + (Cy + CR b= RC, +C,) y= by =1