Anda di halaman 1dari 35
CHAPTER 12 CLASS E ZERO-VOLTAGE-SWITCHING RESONANT INVERTER 12.1, INTRODUCTION ‘There are two types of Class E DC-AC inverters: Class E zero-voltage-switching (ZVS) inverters, which are the subject of this chapter, and Class E zero-current- switching (ZCS) inverters, which we study in Chapter 13. In Class E inverters, the transistor operates.as a switch. Both Class EZVS and Class E ZCS inverters belong to the familly of soft-switching inverters, Class E ZVS inverters [1}-(65) are the most, efficient inverters known so far. The current and voltage waveforms of the switch are displaced with respect to time to produce a very high efficiency of the inverter. In particular, the switch turns on at zero voltage if the component values of the resonant circuit are properly chosen. Since the switch current and voltage waveforms cdo not overlap during the switching time intervals, switching losses are virtually zero, yielding high efficiency. ‘We start by presenting a simple qualitative description of the operation ofthe Class ZVS inverter. Though simple, this description provides considerable insight into the performance of the inverter as a basic power cell. Next, we quickly move to a ‘quantitative description of the inverter. Finally, we present matching resonant circuits and give a design procedure of the inverter. By the end of the chapter, the reader will, bbe able to perform rapid first-order analysis as well as design a single-stage Class E ZNS inverter. ‘Resonant Power Converters, Second Eaton, By Marin K. Kazmiercak and Dariusz Caskowski {Copyright ©2011 tn Wiley de Sons, le 334 {22PRINOIPLEOF OPERATION — 335, T” [We i + Fy % © FIGURE 12:1 Class E zero-voltage-switching inverter. (a) Circuit. b) Equivalent circuit for operation above resonance. 12.2 PRINCIPLE OF OPERATION 122.1 Circuit Description The basic circuit ofthe Class E ZVS inverter is shown in Fig. 12.1(@). Itconssts of a power MOSFET operating as a switch, aL-C-R;series-resonant circuit, a shunt capa- citor C1, and a choke inductor Ly. The switch turns on and off at the operating frequency f = «/(2n) determined by a driver. The transistor output capacitance, the choke parasitic capacitance, and stray capacitances are ineluded inthe shunt capaci- tance Cy. For high operating frequencies all of capacitance Cy can be supplied by the overall shunt parasitic capacitance, The resistor R; is an AC load. The choke induc- tance Ly is assumed to be high enough so thatthe AC ripple on the DC supply current {canbe neglected. Asmall inductance witha large current ripple is also possible [22], but the consideration ofthis ease is beyond the scope of tis text. When the switch is ‘on; the resonant circuit consists of L, C, and Rj because the capacitance C1 is short- circuited by the switch, However, when the switch is or, the resonant circuit consists ‘of C1, L,C, and R, connected in series. Because Cy and C are connected in series, the ‘equivalent capacitance Czy = CC; /(C + Ci) is lower than Cand C;. The load network is characterized by two resonant frequencies and two loaded quality factors. When the Switch is 08, for = 1/@nVEC) and Oxs = wi L/Ri = 1/(c9iCR,). When the switch is om fia = 1/{2n VLCC (C+ Cy) and Qr2 = wal /R = I/[wealCCi/(C + C1) Ie should be noted that foi /fo2 = Qu1/Qta = V/CIACi +). Figure 12.1(b) shows an 336 CLASS E ZERO-VOLTAGE SWITCHING RESONANT INVERTER ‘equivalent circuit of the inverter for operation above resonance. Ifthe operating fre- quency f is greater than the resonant frequency fo, the L-C-R; series-resonant circuit represents an inductive load at the operating frequency f. Therefore, the inductance L can be divided into two inductances, Ly and Lp, connected in series such that L=Lg+Ly and La resonates with C at the operating frequency f .e., 1 a 124 Vat vee The loaded quality factor defined atthe operating frequency is ol _ ola th) 1, why Oe me gare peck ah 12.2.2 Circuit Operation “Figure 12.2 shows current and voltage waveforms in the Class E ZVS inverter for three cases: dus(wt)/d(wt)=0, dus(wt)/d(wt) <0, and dvs(wt)/d(wt)>0 at wt =2x when the switch turns on, In all three cases, the voltage us across the switch and the shunt capacitance Cy is zero when the switch turns on. Therefore, the energy stored in the shunt capacitance C; is zero when the switch turns on, yielding 2er0 turn-on switching loss. Thus, the ZVS condition is expressed by vs(2n) = 0. 23) The choke inductor Ly forces current. To achieve zero-voltage switching turn-on of the switch, the operating frequency f = o/(2x) should be greater than the resonant frequency foi =1/(2x-VLC), i.e., f > for However, the operating frequency f is usualy lower than for = 1/(2x EC), its f 2.5), the shape of the waveform of current is approximately sinusoidal. If Qislow, the shape ‘of the waveform of current i becomes close to an exponential function {19}, (22)."The ‘combination of the choke inductor and the L-C-R; series-resonant circuit acts as @ ‘current source whose currents ;— i. When the switch is ox, the currently —i flows through the switch. When the switch is or, the current 1; ~ i flows through capacitor Cy, producing the voltage across shunt capacitor C; and the switch. Therefore, shunt ‘capacitor C; shapes the voltage across the switch. 12.2.3. Optimum Operation Figure 12.2(a) shows current and voltage waveforms for optimum operation. In case, both the switch voltage vs and its derivative dus /dt are zero when the switch turns on. The second condition is given by dvs(oo d(o) az) "22PRINOPLEOF OPERATION 337 gon, sore gon, srg $y, srg fama eee) a fb a 4 es Je foo teal lant 2 iter | | Weg ‘ ir, tee, re is So D. ; pe IN ORT ' iyo \ ii ' nist 1 a \ i ! Lr 1 Nal : Wa ht a aN A fl I AM fa pc a al eg ot te Aa rs Lhe I Lk 1 Aa INTs pos vs all vay! wi vp LIT MO LAN LN VW vea\ / f ° ut ou bs ° ths @ (o) © FIGURE 12.2 Waveforms in Class Ezero-voltage-switching tion. (b) For suboptimum operation with dos (wt)/d( ux) <0 at a ‘operation with dvs(.wt)/d(at)>O at a eres. (a) For optimum oper- >. () For suboptimury ‘The first condition is called zero-voltage switching (ZVS), and the second con- dition is called zero-derivative switching (ZDS) or zero-slope switching. The two conditions are the “nominal” conditions. The optimum conditions are the operating conditions at which the maximum drain efficiency is achieved. For real compo- nents, the optimum conditions are off-nominal conditions. The smaller the parasitic components, the closer are the nominal and optimum operating conditions. Because the derivative of us is zero at the time the switch turns on, the switch ‘current ig increases gradually from zero after the switch is closed. It should be noted that both the switch voltage and the switch current ae positive for optimum operation. ‘Therefore, there is no need to add any diode to the s 338 CLASSE ZEROVOLTAGE.SWITCHING RESONANT INVERTER, In an optimum design yielding the maximum drain efficiency, the switch voltage us at the switch turn-on time is usually 10% to 50% of the peak switch voltage, which is a nonzero-voltage-switching condition. In addition, the switch voltage derivative atthe switch turn-on time is zero or slightly positive or negative. Close relationships among Ci, Ls, Rif, and D must be satisfied to achieve 1um operation (22]. Therefore, optimum operation can be achieved only at an um load resistance Ri =Ropr. If Rj > Roy, the amplitude fp of the current i through the L-C-R; series-resonant circuit is lower than that for optimum operation, the voltage drop across the shunt capacitor C; decreases, and the switch voltage vs is greater than zero at turn-on. On the other hand, if Ri < Rog, the amplitude Jy i higher than that for optimum operation, the voltage drop across the shunt capacitor CC increases, and the switch voltage vs is less than zero at turn-on. In both cases, assuming a linear capacitance C, the energy stored in C, just before turn-on of the switch is W(2x—) = }Civ}(2x-). This energy is dissipated in the transistor as heat after the switch is tured on, resulting in a turn-on switching loss. To obtain ZVS ‘operation at a wider load range, an antiparallel or a series diode can be added to the transistor. This improvement ensures thatthe switch automatically turns on at zero voltage for Ri < Roy For operation with zero-derivative switching (ZDS), Miller's effect is reduced to zero, Figure 12.3 depicts waveforms of gate-to-source voltage ugs and drain- to-source voltage ups in the Class E zero-voltage-switching inverter. In this case, the change in the drain-o-souree voltage Aups=0. The corresponding change in the gate-to-source voltage is Avgs. The instantaneous voltage gain is o t FIGURE 12.3 Waveforms of gute-to-source voltage ves and (Chass E zero-voltage-switching inverter. 40-source vollage ts in 122 PRINCIPLE OF OPERATION 339 Ay= Avps/Avps =0. Therefore, the gate-to-drain capacitance C,q reflected to the ‘gate-to-source terminals, called Miller's capacitance, is Cy =(I—Ay)Cga = Cya ‘The MOSFET input capacitance is C;=Ciss = Cyr + Cya- Since Miller's effect is cleminated and the MOSFET input capacitance is low, the gate-to-source voltage gs increases much faster than in the inverters with hard switching, in which the slope of the transistor voltage is large when the transistor turns on. In many applications, the load resistance varies over a certain range. The tur-on of the switch at zero voltage can be achieved for suboptimum operation for O 0. Figure 12.2(b) shows current and voltage waveforms forthe case when vs(2x)=0 and dvs(.)/d(or) <0 at ot =2x. Power MOSFETS are bidirectional switches because their curent can flow in both directions, but their voltage canbe only greater than ~0.7 V. When the switch voltage reaches ~0.7V, the antiparallel diode tums on and therefore the switch automatically turns on. The diode accelerates the ime at which the switch turns on. This time i no longer determined by the gate-o-source voltage. Since the switch turns on at zero voltage, the turn-on switching loss is zero, yielding high efficiency. Such an operation can be achieved for 0:0 at or =2r. It should be noted thatthe switch current is always postive, butthe switch voltage vs has positive and negative values. Therefore, 4 unidirectional switch for curent and bidirectional for voltage is needed. Such « switch canbe obtained by adding a diode in series with a MOSFET. When the switch voltage vs is negative the diode is or and supports the switch voltage, regardless ofthe state of the MOSFET: The MOSFETT is tumed on during the time interval when the switch voltage is negative. Once the switch voltage reaches 0.7 V with a positive derivative, the diode turns on, taming the switch on. The series diode delays, the time at which the switch tars on, The ange of Dy is D =D, = Daz, a8 shown in Fig. 12.2) by the shaded area, One disadvantage of the switch with a series jode is a higher on-voltage and a higher conduction loss. Another disadvantage is associated with the transistor output capacitance. When te switch voltage increases, the transistor output capacitance is charged vi a series diode to the peak value of the switch voltage and then remains at this voltage until the transistor turns on because the diode is or. At this time, the transistor output capacitance is discharged through the MOSFET on-tesstance, dissipating the stored energy. 340 CLASS E ZERO-VOLIAGE SWITCHING RESONANT INVERTER 12.3 ANALYSIS 12.3.1 Assumptions ‘The analysis of the Class E ZVS inverter of Fig. 12.1(@) is carried out under the following assumptions: 1. The transistor and diode form an ideal switch whose on-resistance is zero, off-resistance is infinity, and switching times are zero. 2. The choke inductance is high enough so that its se component is much lower than the DC component of the input current. 3. Theloaded quality factor Q, ofthe L-C-R; series-resonant circuits high enough so that the current é through the resonant circuit is sinusoidal 12.3.2 Current and Voltage Waveforms ‘The current through the series-resonant circuit is sinusoidal and given by In sin(wot + 6) 12s) where Jn is the amplitude and ¢ is the initial phase of current i. According 10 Fig. 12.1@), istic = hh T= In sin(wt + 8). (12.6) For the time interval O) fou _ 2x( = D)sin@2D +9) 4 1 orp < 0.2848. 1220 Tr” cosh = cosa +4) * wy Differentiating the switch voltage waveform in (12.13) gives the value ofc at which the peak value of the switch voltage occurs 02.22) ae [= = cos(2nD + ®) atl —D) Substituting this into (12.18) yields the switch peak value Vsu¢/V; in numerical form, Figure 12.6 depicts plots of cin, fom, Ie /I1, and Vay /Vp versus D. 344 CLASS E ZERO-VOLIAGE:SWITCHING RESONANT INVERTER 360 220 200 in to 4 0-2 ogee Omega OO pee > ® 0 ° | 7] 5 3 s| 36 SF vawi¥h foul 2 1 g 02 cy 38 C3 1 . © FIGURE 12.6 Peak values of the switch current and voltage. (8) cfg and ton versus D. (@)Normalized switch peak curent /y /; andnormalized switch peak voltage Vx /V; versus D. Neglecting power losses, the AC output power Pp; is equal to the DC input power Vil. Hence, using [sw /l1 and Vs /V1, one obiains the power-output capability oe LM *TsuVom — Tsu Vsn" A plot of ¢p versus D is displayed in Fig. 12.7, A «a2.23) T2SANAWYSIS — 945 o1 0.09 0.08 0.07] 0.08 S 0095 0.08 0.03 0.02 001 °o Oz on 08 os 1 > FIGURE 12.7 Power-output capability ¢y versus D. ine on vm Ry Sum, of the series-resonant circuit above resonance at the FIGURE 128 — Equivalent cic operating frequency f 12.3.4 Input Impedance of the Resonant Circuit ‘The current through the series-resonant circuit is sinusoidal. Consequently, higher harmonies ofthe input power are zero, Therefor, itis sufficient to consider the input impedance of the series-esonant circuit at the operating frequency f. Figure 12.8 shows an equivalent circuit of the series-resonant circuit above resonance at the operating frequency fA phasor diagram for voltages at fundamental frequency is shown in Fig. 12.9. The fundamental component of the input voltage of the seres- resonant circuit atthe operating frequency is ver = Uni + 01s = Vain Sn( ot + 6) + Vein COS(at + 6) = Vim Sin (or + 6+ ¥). 12.24) ‘where Vani is the amplitude ofthe fundamental component ofthe switch voltage vs and tan Y= Viin/Vain- Using (12.13) and the Fourier formula, the amplitude ofthe voltage across the load resistance R is obtained as J sin xD sin(x 1p” ug intent + 6)d(aa = — 28RD SiNAD + Oy, a don mL ») oe 346 CLASS E ZERO-VOLTAGE-SWITCHING RESONANT INVERTER Vism FIGURE 129 Phasor diagram for voltages at fundemental frequency f. 2 18 16 1 12 1 08 os o4 02 o. Vim! Vises VV» iin! 0 02 oa 08 8 1 > FIGURE 12.10 Normalized amplitudes of the fundamental components Vain/¥; and Vein Vs versus D. Substituting (12.13) into the Fourier formula and using (12.16), the amplitude of the fundamental component of the voltage across the input reactance of the series- resonant circuit (equal to the reactance of the inductance Z) is obtained as 1p Vim = olan = = f ap COMO + BAC!) 1 = 2(1 — Dx? — 2e0s Gcos(2xD + 6) +008 2(nD + )[cos2xD — x(1 — D) sin2xD] 2 = Dyxcos(aD + OU = D)ncos D+ sin xD] (1226) Figure 12.10 shows Vrim/Vi and Viin/Vi as functions of the duty eycle D. WSANALYSIS 347 18) 16 14 12| PaslPyIVP o8| os! oa 0! ° oz oF 08 On 1 o FIGURE 12.11 Normalized output power PiRi/V? a function ofthe duty eye D. 123.5 Output Power From (12.25), one obtains the output power Vey _2sin? xDsin? (nD + OV? PR= Re wT = DPR: 227 Figure 12.11 shows a plot ofthe normalized output power PriRi/V} as a function of the duty cycle D. 12.3.6 Component Values Combining (12.10), (12.16), and (12.25), 2sin xD cos(xD + 6) sin(xD + ){(1 — D)x cos xD + sin xD] 321 = D) eR; 42.28) Similarly, using (12.10), (12.16), and (12.26), 2(1 — Dx? — 1 + 2e0scost2nD + 6) oly c0s2(nD + $)[c082xD — x1 — D)sin 2x] “Ri ~ Fsin xD cos (xD + 6) sin(xD + $)[ — D)xcos xD + sin xD] (12.29) any = ‘The product of (12.28) and (12.29) yields 2(1 — D)?x? ~ 1+ 20s cos(2nD + 6) = cos2(nD + 6)[c0s 2D — x(I = D) sin2xD] oe 2aX(I —D) (42.30) 348 CLASSE ZERO-VOLTAGE SWITCHING RESONANT INVERTER Figures 12.12-12.14 show plots of @C:Ri, «oLy/R;, and wL4Ci as functions ofthe Xc3. If the optimum operation occurs at Ri=Xc3, then Remax =Xc3/2 and the inverter operates under ZNS conditions at any load resistance R; [28]. This is because Ry < Romar = Xc3/2 at any values of R T2EMATCHING RESONANT CIRCUS 357 R= Kea R » FIGURE 12.18 Series equivalent resistance R, and reactance Xe, as functions of load resistance R; inthe circuit xa () Ry versus Ry. (6) Xex versus Ri 12.6.3 Resonant Circult x2a ‘The values of Ry, Xc1, and X¢ for the resonant circuit shown in Fig. 12.16(b) can be calculated for optimum operation at D=0.5 from (12.75), (12.71), and (12.73), respectively. The reactances of Ly and Lp are: Xz = oly = (LL) = (Qr- OR (o.- Baal, (12.86) Xin = ola 2.87) Es ‘The range of resistances that can be matched by the circuit shown in Fig. 12.16(b) Ry < Ry (12.88) ‘Suboptimum operation takes place for 0. Rasw) < Re 2.89) ‘and consequently for Ri < Rysady < 00. (12.90) 358 CLASSE ZERO-VOLIAGE-SWITCHING RESONANT INVERTER, ‘The relationship between resistances Ry and Ri is Re —_ (1291) 1+(#) ‘Thus, as R is increased from 0 to Xz Re increases from 0 10 Ranae = X:2/?, and as R; is increased from Xia to 00, Ry decreases from Xi2/2 to 0. Its clear that the Ri-L2 circuit behaves as an impedance inverter for Ri> Xia. If the optimum ‘operation occurs at Ri= Xia, then R,=Xr2/2 and the ZVS operation occurs at any load resistance [25], (30) In this case, Ry < Rimax =X12/2- 12.6.4 Resonant Circuit x1b ‘The values of Ry, Xc1, and X, for the circuit shown in Fig. 12.16(c) can be calculated for optimum operation at D=0.5 from (12.75), (12.71), and (12.72), respectively. ‘The values of Xc2 and Xc3 can be found from [R(Qx — 1.1525? + 1) po ca 2 1, — 1.1525 — / EAR RN Ri Gast < Ry, < RK. (12.94) Rigsuoy < Rie (12.96) 12.6.5 Resonant Circuit x4a ‘The values of Ry, Xct and Xe forthe circuit shown in Fig. 12.16(d) can be calculated for optimum operation at D=0.5 from (12.75), (12.71), and (12.72), respectively. ‘The reactances of Ly and Ly are RQ} +1) X= ola = Rif az97) 127 DESIGN EXAMPLE 359 Xu = oly = EY (1298) ‘This circuit can match resistances an < Ry < Ri. 1299) Suboptimum operation takes place for Raut) > Rs (412.100) and therefore for Rijsub < Rie (12.101) 127 DESIGN EXAMPLE EXAMPLE 12.1 Desgn he Cass ES inveneof Fig, 121). sats the folowing pein: Vi =100V, Primer =80W, and f . Solution: It is sufficient to design the inverter for the full power. Using (12.48), the full-load resistance is 2 8 VE 9.5168 x 0% —n.19 (22.102) B= PYAR R 30 From (12.41), the DC resistance of the inverter is 44 Ry = 1.7337 x 72.1 Roc 252. (12.103) ‘The amplitude of the output voltage is computed from (12.46) Vein = 2 V1 = 1.074 x 100 = 107.4. (12.104) ‘The maximum voltage across the switch and the shunt capacitor can be calculated from (12.43) as Vsue = 3.562V) = 3.562 x 100 = 356.2V. (12.105) 260 CLASSE ZEROWOLTAGE SWITCHING RESONANT INVERTER From (12.41, the DC input curentis 8M Me aR, 100 5168 x 7 = 08A. (12.106) ‘The maximum switch current obtained using (12.42) is (2 4.)u 2.862 x 0. Iu 229A. (12.107) ‘The amplitude ofthe current through the resonant circuit computed from (12.45) is lea 2 Assuming Qz =7 and using (12.31), (12.49), and (12.52), the component values of the load network are: I = 1.8621 x 0.8 = 149A. (32.108) OR 7x71 1 OR Oo (12.109) = FE (12.110 1 GP Dam, ~ TRG FD KAD AOR TET SPF C12 and 1 © mis — SRD] BTR TOE x TANTS) = 315 pF. azn It follows from (12.33) that in order to keep the current ripple in the choke inductor below 10% of the full-lond DC input current J), the value of the choke inductance ‘must be greater than ej) 7x7 yn2(Far)e TARE = 4205uH 2.12) “The peak voltages across resonant capacitor Cand inductor L are en 149 =e4v a2) 2 x 108 x 315 x 10-7 and Vim = Old, = 2 x 1.2 x 108 x 66.9 x 10° x 1.49=751.6V. (12.114) ‘Assume that the DC ESR of the choke Ly is r1y =0.15 2. Hence, from (12.54) the power loss in rip is Prag = rigl? = 0.15 x 0.8? = 0.096 W. 12.115) TR7DESIGNEXAUPLE 361 From (12.55), the rms value of the switch current is Iams = A ox 1538512310, (12116) If the MTPSN40 MOSFET is used whose on-resistance ros =19, the transistor conduction power loss is Prong = roel = 1x 1.231 515 W. (2.117) Using (12.57), one obtains the rms current through the shunt capacitor 4 4 18 x 0.6087 Totem = 1485 A. (12.118) Assuming the ESR of C; to be rci =76m®, one arrives atthe conduction power loss in rei Pict = rile ms = 0.076 x 0.485? = 0.018 W.. (2.119) ‘Assume the ESRs of the resonant inductor and capacitor to be 7, = 0.59 and rc =50 m8? atf = 1.2MHz, Hence, the power losses in the resonant components are iB _ 05x 1.49? Pa = Ta = 42.120) 2 «14st Pag = Tele = SOS XMAPT _ ose. (22 2 2 ‘The total conduction loss is Pr = Pros + Pag + Prei+ Pat Pre = 1.515 +0.096 + 0.018 + 0.555 + 0.056 = .24W. (12.122) ‘The inverter efficiency associated with the conduction loss is Pri 80 Pye P, ~ 8042.24 97.28%. (12.123) To estimate the turn-off switching loss, assume ty ~0.057 =005/f =0.05/ (1.2 109) = 41.7 1s, Hence, oy = 2 % 1.2 « 108 x 41.7 x 10? = 0.314 rad. Thus, from (12.64) the turn-off switching loss is 314? x 80 2 0.657 W. (42.124) 362 CLASSE ZERO-VOLTAGE SWITCHING RESONANT INVERTER For the MTPSN40 power MOSFET, Q, =27nC. Hence, assuming Vosm=8Y, ‘one obtains the gate-iive power PG =fVesmOz = 1.2 x 106 x 8 x 27 x 10- = 0.259 W. 42.125) ‘Thus, the power los is Pus = Pr+ Py + Po =2.24+0.657+0.259 =3.156W. (12.126) ‘The efficiency of the inverter becomes Pu _ «80 em 5s = ae aia 7 90®. 2.127) “The power gain ofthe inverter is Pa __80 ope eS 12. 9 = Feo asp (12128) ‘The equivalent capacitance when the switch is off is Ceg=CC1/(C + C1: and the resonant frequencies are foi =1/(2xVEC)=1.096MHz and fia=1/ (Ox ECaq)=1.524MHz. Notice that the operating frequency f is between the resonant frequencies fy and fa. 63 pF, 12.8 PUSH-PULL CLASS E ZVS INVERTER Fig. 12.19 shows a circuit of a push-pull Class E ZVS inverter. It consists of two transistors, two shunt capacitors C), two LC series-resonant circuits, and two RF chokes Ly. u coe FIGURE 12.19 Push-pull Class EZVS inverter, 12A0REFERENCES 963 12.9 SUMMARY + The transistor output capacitance, the choke parasitic capacitance, and the stray ‘capacitance are absorbed into shunt capacitance Cy in the Class E ZS inverter. + The turn-on switching loss is zero in the Class E ZVS inverter. Miller's effect is reduced to zero in the Class E ZVS inverter. + The operating frequency f is greater than the resonant frequency f= 1/ QnVLC) of the series-resonant circuit. This results in an inductive load for the switch when it is on. ‘+ The antiparallel diode of the switch turns off at low di/dt and zero voltage, reducing reverse-recovery effects. Therefore, the MOSFET body diode can be used and there is no need fora fast diode. + The zero-voltage-switching operation can be accomplished in the basic topology for load resistances ranging from zero to Rioyr. Matching circuits can be used to ‘match any impedance to the desired load resistance. + The peak voltage across the transistor is about four times higher than the input DC voltage. Therefore, the circuit is suitable for low input voltage applications. + The drive circuit is easy to build because the gate-to-source voltage of the transistor is referenced to ground, + The circuit is very efficient and can be operated at high frequencies, + The large choke inductance with a low current ripple can be replaced by a low inductance with a large current ripple. In this case, the equations describing the inverter operation will change [24]. ‘The loaded quality factor of the resonant circuit can be small.In the extreme case, the resonant capacitor becomes a large DC-blocking capacitor. The mathematical

Anda mungkin juga menyukai