88 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 23, NO. 1, JANUARY 2008 Coupled Input-Series and Output-Parallel Dual Interleaved Flyback Converter for High Input Voltage Application Ting Qian and Brad Lehman, Member, IEEE Abstract—This paper proposes an integrated magnetic dc–dc converter suitable for high input voltage application. The converter is based on a coupled input-series and output-parallel dual interleaved Flyback converter concept. All the center and outer legs are gapped, and the transformers are integrated into one magnetic core with not so tight coupling. The gap is beneficial for suppressing current spike caused by the voltage mismatch between the windings. The two transformers are inversely coupled, and current ripple reduction can be achieved with suitable coupling design. A prototype with 350–450-V input and 24-V/4-A output is built. Experimental results verify the performance of the new topology. Index Terms—Flyback converter, integrated magnetic, interleaved. I. INTRODUCTION HERE is an increasing demand in modern power electronics for high density power converters. In most cases, the size of the magnetic components, including transformers and inductors, significantly influences the overall profiles of the converters. Integrated magnetic techniques [1]–[17] seem to be suitable solutions for high density application. The attraction is that transformers and inductors are combined in a single core, and therefore, cost and size of the converters may be reduced. Generally, there are two dominant types of isolated topologies using integrated magnetics: buck mode topologies, such as forward [3], [4], push–pull [6], [7], half-bridge [8], [9] and full-bridge [8], [9], and buck-boost mode topologies, such as dual Flyback [15], [16]. More recently, there are newly proposed integrated boost converters [10], [17] that are beginning to see applications in automotive power electronics. For large step-down power conversions, such as 48 to 1 V, buck mode isolated topologies have become popular solutions. As a result, integrated magnetic techniques [1]–[9] have been widely developed for buck mode topologies to minimize the number of magnetic components and improve the ripple cancellation. Historically, magnetic integration is utilized for forward converters [3], [4]. However, recent research pays more attention to full wave integrated magnetic dc–dc converters [6]–[9], T Manuscript received March 29, 2007; revised June 15, 2007. Recommended for publication by Associate Editor J. Pomilio. The authors are with Northeastern University, Department of Electrical and Computer Engineering, Boston, MA, 02115 USA (e-mail: tqian@ece.neu.edu; lehman@ece.neu.edu). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TPEL.2007.911867 which includes push–pull, half-bridge and full-bridge, etc. For example, magnetic integration is utilized for push–pull forward circuit with coupled-inductor current doubler circuit [6], [7]. A single EI or EE core is used for all the magnetic components, including the input inductor, the step-down transformer and the output filtering inductor. The transformer’s primary and secondary windings, as well as the inductor windings, are wound on the two outer legs. Only the center leg has an air gap. The flux ripple is cancelled in the center leg, resulting in a lower core loss in the center leg. As well as reducing the overall size of the converter, coupled output inductor greatly reduces the current ripple, and thus improves the power efficiency. Incorporating an independent inductor winding into the transformer is another example for full wave integrated magnetic topologies [8], [9]. The original full wave buck mode circuit operation is retained for this approach. A flexible output inductor winding is added in the center leg to optimize the design of output inductor. The inductor can be designed specifically to satisfy the current ripple restrictions and the magnetic constraints. This research mainly focuses on the discussion of integrated magnetic techniques related with buck-boost mode isolated topologies. In this definition, an integrated magnetic buck-boost mode topology means double or multiple Flyback circuits using a single magnetic core, since a single Flyback utilizes only one gapped magnetic core to store the energy in the power transformer. Flyback converter is a popular choice for low power application due to its simplicity and low cost [18], [19]. For example, it is suitable for ac–dc converters (a dc–dc converter stepping down a 400-V input to a relatively low voltage output, such as 24 V, is generally connected in cascade with a power factor correction circuit) at a power level of 150 W or less. Fig. 1 shows a traditional full wave buck-boost (Flyback) power converter [15], [16] that utilizes one magnetic core to integrate two Flyback transformers. In this case, the two transformers are coupled with a low reluctance in the outer legs of the magnetic core. The relatively higher reluctance magnetic property of the inductor is integrated into the magnetic structure center leg. In other words, only the center leg has the gap. The or is turned ON, and transformers store the energy when and are turned release the energy to the load when both OFF. Ideally, and conduct simultaneously to deliver the energy to the load so that the two inductors are coupled to reduce the current ripple. Since the converter has full wave operation, the duty ratio of each Flyback is less than 50%. This paper proposes a new integrated magnetic dual Flyback converter that is more suitable for high input voltage applica- 0885-8993/$25.00 © 2007 IEEE QIAN AND LEHMAN: COUPLED INPUT-SERIES AND OUTPUT-PARALLEL DUAL INTERLEAVED FLYBACK CONVERTER Fig. 1. Traditional full wave buck-boost power converter [15], [16]. tions. Specifically, two interleaved Flyback converters are connected in series on the primary side and in parallel on the secondary side. Due to interleaved operation of the Flyback circuits, the current ripple is, thus, reduced as in typical interleaved Flyback circuits [20]–[22]. A benefit of sharing the output current is that each Flyback handles half of the load power. All the center and outer legs are gapped in the new converter, and the transformers are integrated into one magnetic core with not so tight coupling. This is beneficial for suppressing current spike caused by the voltage mismatch between the windings. The two transformers are inversely coupled so that significant current ripple reduction can still be fulfilled without tight coupling. It should be remarked that leakage inductance between the primary winding and the related secondary winding cannot be minimized. This problem also exists in other Flyback converters, since Flyback converters utilize gapped magnetic core for transformers to store and release energy. The details of current ripple reduction are explained in Section III. The following features are presented. • Connecting the primary side in series reduces the voltage stresses of the primary components. The primary switches are rated at half of the corresponding voltage values in a single Flyback or a traditional full wave buck-boost power converter [15], [16]. The on-resistance and cost of the primary switches are, thus, significantly reduced. Alternatively, this implies that for the same rated switches, the proposed converter can handle twice the input voltage compared with the converter in Fig. 1 [15], [16]. • Interleaved operation with paralleled connection on the secondary side keeps the benefits of ripple reduction as a traditional full wave buck-boost power converter [15], [16] does. • To satisfy a certain current ripple requirement, magnetizing inductance can be reduced by means of suitable inductor coupling design. Due to the mutual influence of the two coupled inductors that is combined with the transformers, further current ripple reduction can be achieved when compared with interleaved Flyback converters without coupling. Also, the amount of the magnetic components is reduced by integrating the two transformers 89 into one magnetic core, although the efficiency of the new approach may not be improved when compared with conventional discrete interleaved Flybacks. • Current spike caused by the voltage mismatch between the windings can be suppressed by gapping all the center and outer legs. Compared with tight coupling, such as the condition in a traditional full wave buck-boost power converter shown in Fig. 1, gapping all the legs weakens the coupling between the two Flybacks’ windings, and thus, reduces the effect of the voltage mismatch. Section II of the paper presents basic operation principles of the proposed converter, highlighting the differences between the proposed converter and known topologies [15], [16]. Then, Section III presents more technical detail of the coupled inductor interaction. Section IV gives further magnetic flux analysis. Experimental results are presented in Section V, followed by the conclusion. II. OPERATION PRINCIPLE Fig. 2 shows the proposed coupled input-series and outputparallel dual interleaved Flyback converter, and Fig. 3 represents the related waveforms. The primary switches are rated at half of the corresponding voltage values in a single Flyback or a traditional full wave buck-boost power converter [15], [16] due to series connection of the two Flybacks. Also, gap filling is used for all the three legs. As shown in Fig. 2, the upper Fly, , , and two windback (Flyback 1) consists of ings in one outer leg. The lower Flyback (Flyback 2) consists of , , , and two windings in the other outer leg. Since gapping all the three legs weakens the coupling, the two Flybacks can actually operate independently (if desired). The current ringing caused by the voltage mismatch between the two Flybacks’ windings can be significantly suppressed due to the weakened coupling. Therefore, the duty ratio can exceed 50%. It should be emphasized that weakened coupling does not necessarily sacrifice ripple cancellation in this case. Detailed ripple calculation is available in Section III. However, synchronized operation with suitable coupling design is beneficial for ripple reduction. This paper mainly analyzes the interleaved condition and when duty ratio is less than 50% as shown in Fig. 3. are the main primary switches. and are the voltages across the primary windings as shown in Fig. 2. and are and are the the current through the primary windings. current through the secondary windings. An E magnetic core is used in this paper. Each transformer utilizes one outer leg for windings. The following description explains the detailed principle of the proposed topology. – is defined as one duty cycle . The length of – and – are defined as , which repand are the turns of the resents the duty ratio. Assume primary and secondary windings. The input to output transfer ratio is (1) Interval 1 – : The condition of this period is shown in is turned on. The voltage Fig. 4(a). During this interval, across the primary winding of the upper Flyback is equal to . The upper Flyback circuit delivers energy from 90 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 23, NO. 1, JANUARY 2008 Fig. 2. Proposed dual coupled Flyback converter. circuits switch the condition. is turned on. The voltage across the primary winding of the upper Flyback is equal to . The lower Flyback circuit delivers energy from to , the transformer on the primary side. The primary current which is shown in Fig. 2, increases linearly. At the same time, remains off. The secondary side current of the upper Flyback, , flows through and charges . The current which is is applied to value begins to decrease. The output voltage the secondary winding of the upper Flyback. Interval 4 – : Fig. 4(d) shows the condition of the last interval. This case is the same as interval 2. Both Flyback circuits release energy on the secondary sides. The output voltage is applied to the two secondary windings. III. RIPPLE CALCULATION FOR COUPLED CONDITION For two coupled inductors, the following typical relations can be obtained [6], [11], [14]: (2) (3) Fig. 3. Timing and waveforms of the proposed circuit. to the transformer on the primary side. The primary current , which is shown in Fig. 2, increases linearly. The secondary is reverse-biased. At the same time, is off. The diode secondary side current of the lower Flyback, which is , flows and charges . The current value begins to dethrough is applied to the seccrease. In this case, the output voltage ondary winding of the lower Flyback. – : As shown in Fig. 4(b), is turned off Interval 2 during this interval. Both Flyback circuits start to release the stored energy from the transformer to the output capacitor and goes through diode and goes load. Therefore, current through diode . The voltage across each secondary winding is equal to . – : The condition of this interval, which is Interval 3 shown in Fig. 4(c), is similar as interval 1. The two Flyback In the above case, let , and and are is the voltages applied on the two corresponding windings. the coupling inductance. For this proposed coupled Flyback, the following equations about the current ripple can be achieved based on typical equations and specific conditions. Using the derived equations, optimized design can be achieved for converters according to their specific requirement. When all the legs have gaps with approximately same distances, is approximately equal to 1/3. Assume is the coupling inductance between the two primary windings. and are the magnetizing inductance of the primary and are the input and output voltand secondary winding. is applied. ages. For each Flyback, : During this interval, currents circuInterval 1 late through the primary winding of Flyback 1 and the secondary winding of Flyback 2. These two windings are considis applied to the priered as two coupled inductors. Since is applied to the secondary mary winding of Flyback 1, and QIAN AND LEHMAN: COUPLED INPUT-SERIES AND OUTPUT-PARALLEL DUAL INTERLEAVED FLYBACK CONVERTER 91 Fig. 4. Detailed description of the operation principle of the proposed circuit. (a) Interval 1 (t –t ); (b) Interval 2 (t –t ); (c) Interval 3 (t –t ); (d) Interval 4 ( t –t ) . winding of Flyback 2, the following formulas related with current ripple are obtained according to (2) and (3): (4) Since , we also have . The capacitor is in a discharge state. is increasing approxiis positive. mately linearly because – : During this interval, currents circulate Interval 2 through the secondary winding of both Flyback circuits. These two windings are considered as two coupled inductors. Since is applied to the secondary winding of each Flyback, the following formulas related with current ripple can be achieved according to (2) and (3): , we also have . Since Similar to Interval 1, the capacitor is in a discharge state and is increasing approximately linearly. – : During this interval, currents circulate Interval 4 through the secondary winding of both Flyback circuits. These two windings are considered as two coupled inductors. Since is applied to the secondary winding of each Flyback, the following formulas related with current ripple are obtained: (7) Similar to Interval II, . The capacitor is in a charge state. The two coupled secondary inductors are both releasing their energy into the load. Further Discussions on the Proposed Circuit: As shown in Fig. 3, the current ripple is determined by (5) (8) . The capacitor is Also, . The two coupled in a charge state since secondary inductors are both releasing their energy into the load. – : During this interval, currents circulate Interval 3 through the second winding of Flyback 1 and the primary winding of Flyback 2. These two windings are considered as is applied to the secondary two coupled inductors. Since winding of Flyback 1, and is applied to the primary winding of Flyback 2, the following formulas related with current ripple are obtained: (6) When compared with no coupling condition (suppose using same inductance value , and the current ripple is determined .), by should be satisfied in order to achieve a reduced current ripple. In other words, smaller inductor can be used to satisfy identical current ripple requirement when . In the prototype for experimental verification, is approximately equal to 1/3 (all the legs have gaps with approximately same distances). ( , , , and ) is achieved. Therefore, smaller inductance is required when compared with two discrete interleaved Flybacks. 92 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 23, NO. 1, JANUARY 2008 Fig. 5. Magnetic reluctance circuit of the proposed topology. Specifically, the following differences are remarked when compared with a traditional converter [15], [16] in Fig. 1. 1) The new approach connects the input of the two Flybacks in series, while the traditional method connects the input in parallel. 2) The proposed converter has three gaps in all legs. The traditional converter only has one gap in the center leg. 3) The secondary windings are inversely coupled in the new converter, and directly coupled in the traditional converter. From the first difference, it can be concluded that the proposed method has half of the voltage stress when compared with the traditional method. According to the second and third differences, the secondary windings of the traditional converter are directly coupled with tight coupling. However, for practical circuits, especially for this high voltage application, switches cannot operate exactly simultaneously. There is a voltage mismatch between the two secondary windings. The voltage difference is applied to the leakage inductance between the two windings. Because the two outer legs have low reluctances, and the center leg has a high reluctance, the two secondary windings have good coupling and small leakage inductance. Minor mismatch can lead to high current spikes and resonance. On the other hand, the secondary windings of the new converter are inversely coupled with not so tight coupling since all three legs are gapped. By gapping all the three legs, a big leakage inductance between the windings of the two transformers is obtained, since the flux generated by each winding in the outer legs can go through all three legs. ( is approximately equal to 1/3 when all the legs have gaps with approximately same distances.) The large inductance value can avoid high current spike when a voltage mismatch is applied. Also, current ripple reduction can still be achieved with suitable coupling design. Detailed calculation for current ripple is given in Section III. We also remark that the weakened coupling between the two transformers allows the duty ratio to be greater than 50% in the propose circuit. Fig. 6. Flux path in the core of transformer. Fig. 7. Equivalent magnetic reluctance circuit during each interval. (a) Interval 1 (t –t ); (b) Interval 2 (t –t ); (c) Interval 3 (t –t ); (d) Interval 4 (t –t ). metrical. Also, Fig. 6 shows the flux path of in the core of transformer. Because of the air gap, the center leg is no longer a low magnetic reluctance path for the fluxes. The flux generated by each winding in the outer legs can go through all three legs. The flux interaction between the windings affects the current ripple and peak values of the flux densities. According to Fig. 5, the following fundamental equations is, thus, obtained [6], [11], [14]: (9) IV. FLUX DISTRIBUTIONS ANALYSIS For the convenience of magnetic design, it is necessary to provide the formulas for flux distributions. To prevent the magnetic core from saturating, the peak values of the flux densities in all three legs should be kept below a specified maximum value. Therefore, this section will discuss the detailed calculation for both the ac fluxes and dc fluxes in all three legs. As a result, the peak values of the fluxes are the sum of the dc fluxes and half of the ac fluxes. Fig. 5 shows the magnetic reluctance circuit of the proposed topology, assuming the conditions of the two outer legs are sym- (10) Furthermore, Fig. 7 shows the detailed equivalent magnetic reluctance circuit during each interval. The effect of the corresponding winding is removed when the current in the winding is zero. A. AC Fluxes According to Farad’s law, the ac fluxes in the two outer legs, and , are determined by the time integral of voltages QIAN AND LEHMAN: COUPLED INPUT-SERIES AND OUTPUT-PARALLEL DUAL INTERLEAVED FLYBACK CONVERTER across the windings. Therefore, the magnitude of the peak-topeak ac fluxes in the two outer legs can be derived as follows: 93 and Since the averaged value of is , according to (9), (10), and Fig. 7, the dc fluxes for all the outer and center legs are derived as (11) (21) On the other hand, the flux rate of the center leg is derived by summing the flux rates of both the outer legs (22) (12) As a result, the peak values of the fluxes are the sum of the dc fluxes and half of the ac fluxes. During Interval 1 – , and are applied to the primary winding of Flyback 1 and the secondary winding of Flyback 2. The flux rates of the two outer legs are and can be derived as (23) . Similarly, the flux rate during all intervals – (13) – (14) – (15) – (16) (24) To avoid the saturation of the magnetic core, the peak value of the flux densities of all three legs should be computed. Assume is the crossing area of the center leg, the peak values are derived as follows: (25) As a result, the peak-to-peak ac flux in the center leg is achieved according to (13)–(16) (17) (26) By using the calculation of the current ripple (8) and the peak flux densities (25), (26), suitable magnetic core size and the turn numbers of the windings can be chosen to satisfy the current ripple restrictions and the magnetic constraints. B. DC Fluxes V. EXPERIMENTAL RESULTS Another important factor in integrated magnetic design is the dc flux bias in the core. The detailed flux distribution is shown in Fig. 6. The fluxes , , and in the three core legs can be determined from Fig. 5 and Fig. 7 as (18) (19) (20) To verify the principle of the proposed scheme, a prototype with 350–450-V input and 24-V/4-A output is built by using the proposed topology switched at 200-kHz hard switched. FDD6N50TF (500-V, 6-A, DPAK) is used for the primary switches. (It should be remarked that MOSFETs rated at 800–1000 V are needed in this specification for a single Flyback or a full wave Flyback [15], [16] shown in Fig. 1.) and ) use 12CWQ10FN (100-V, Secondary diodes ( 12-A, DPAK). The two coupled transformers are integrated into one E22 magnetic core. The cross-sectional area of the magnetic core is 78.5 mm . The gaps are the same in each leg. Therefore, is approximately equal to 1/3. Each power transformer has 60 turns for the primary winding and 13 turns for the secondary winding. Using (25) and (26), the designed peak values of the flux densities of all three legs are below 2600 Gauss during the whole input voltage variation range at full load. In this prototype, the drain-source of each primary switch is paralleled with a 330-pF capacitor. On the secondary side, a 94 IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 23, NO. 1, JANUARY 2008 Fig. 8. V (Channel 1, 250 V/div), V (Channel 2, 250 V/div), primary current (Channel 3, 1 A/div) and voltage across the secondary diode (Channel 4, 50 V/div). (a) Primary current of the upper Flyback is shown; (b) primary current of the lower Flyback is shown. Fig. 9. V (Channel 1, 250 V/div), V (Channel 2, 250 V/div), secondary current (Channel 3, 5 A/div) and voltage across the secondary diode (Channel 4, 50 V/div). (a) Secondary current of the upper Flyback is shown; (b) secondary current of the lower Flyback is shown. 75- resistor and a 1000-pF capacitor (connected in series) are connected in parallel with each secondary winding. (Channel 1, 250 V/div), Fig. 8 shows the waveforms of (Channel 2, 250 V/div), primary current (Channel 3, 5 A/div) and voltage across the secondary diode (Channel 4, 50 V/div) when the input voltage is 400 V. Fig. 9 shows the (Channel 1, 250 V/div), (Channel waveforms of 2, 250 V/div), secondary current (Channel 3, 5 A/div) and voltage across the secondary diode (Channel 4, 50 V/div) when the input voltage is 400 V. The two Flyback circuits are interleaved. The waveforms of the secondary side are in phase with the primary side. The change rate of the secondary current of one Flyback differs as expected when the other Flyback operates at different modes. This phenomenon occurs due to the mutual effect between the coupled windings. By comparing the waveforms from the upper and lower Flyback circuits, it can be clearly seen that the two parts are self-balanced. Each part shares half of the input voltage. As seen in Fig. 8, the voltage across the primary winding of the transformer is equal to half of the input voltage when the main primary switch is drops to zero. Since is approximately turned on and equal to 1/3 when all the legs have gaps with approximately same distances, the expected current ripple can be derived according to (8). The magnetizing inductance of the primary . The calculated primary windings in the prototype is 460 peak-to-peak current ripple is, thus, approximately 0.71 A, which is close to the peak-to-peak values of about 0.75 A in channel 3 of Fig. 8(a) and (b). Also, there are some oscillations on the primary currents. The ringing is caused due to the effect of the leakage inductance between the primary winding and the related secondary winding. To keep more safety margin, clamp circuit is recommended to reduce the voltage ringing across the switches. An alternate method to achieve more safety margin is to use higher voltage rated MOSFETs. It is possible that the power conversion efficiency of the converter might decrease 1%–2%. Also, improving the coupling between the primary winding and the related secondary winding is beneficial for ringing suppression. By applying the proposed coupled input-series and output-parallel dual interleaved Flyback converter, an efficiency of 89.1% at 400-V input and 24-V/4-A output is obtained. Fig. 10 shows the efficiency curve during the whole input voltage variation range. The experimental results verify the principle and performance of the scheme. QIAN AND LEHMAN: COUPLED INPUT-SERIES AND OUTPUT-PARALLEL DUAL INTERLEAVED FLYBACK CONVERTER Fig. 10. Efficiency of the built prototype. VI. CONCLUSION A coupled input-series and output-parallel dual interleaved Flyback converter for high input voltage application is proposed. Connecting the primary side in series reduces the voltage stresses of the primary components. All the center and outer legs are gapped, and the transformers are integrated into one magnetic core with not so tight coupling. The current ringing caused by the voltage mismatch between the two Flybacks’ windings can be suppressed due to the weakened coupling. The two transformers are inversely coupled so that significant current ripple reduction can still be fulfilled with not so tight coupling. Experimental results verify the performance of the new topology. REFERENCES [1] W. Chen, G. Hua, D. Sable, and F. Lee, “Design of high efficiency, low profile, low voltage converter with integrated magnetics,” in Proc. Applied Power Electronics Conf. Expo., 1997, vol. 2, pp. 911–917. [2] E. Bloom, “Core selection for & design aspects of an integrated-magnetic forward converter,” in Proc. Applied Power Electronics Conf. Expo., 1986, pp. 141–150. [3] E. Bloom, “New integrated-magnetic dc–dc power converter circuits & systems,” in Proc. Applied Power Electronics Conf. Expo., 1987, pp. 57–66. [4] R. Severns and E. Bloom, Modern DC/DC Switchmode Power ConverterCircuits. New York: Van Nostrand, Dec. 1985. [5] L.-P. Wong, Y.-S. Lee, K.-W. D. Cheng, and M. H. L. Chow, “Two-phase forward converter using an integrated magnetic component,” IEEE Trans. Aerosp. Electron. Syst., vol. 40, no. 4, pp. 1294–1310, Oct. 2004. [6] P. Xu, M. Ye, and F. C. Lee, “Single magnetic push-pull forward converter featuring built-in input filter and coupled-inductor current doubler for 48 V VRM,” in Proc. Applied Power Electronics Conf. Expo., 2002, vol. 2, pp. 843–849. [7] P. Xu and F. C. Lee, “Design of high-input voltage regulator modules with a novel integrated magnetics,” in Proc. Applied Power Electronics Conf. Expo., 2001, vol. 1, pp. 262–267. [8] L. Yan and B. Lehman, “Integrated magnetic full wave converter with flexible output inductor,” IEEE Trans. Power Electron., vol. 18, no. 2, pp. 670–678, Mar. 2003. [9] J. Sun, K. F. Webb, and V. Mehrotra, “Integrated magnetics for currentdoubler rectifiers,” IEEE Trans. Power Electron., vol. 19, no. 3, pp. 582–590, May 2004. [10] L. Yan and B. Lehman, “An integrated magnetic isolated two-inductor boost converter: Analysis, design and experimentation,” IEEE Trans. Power Electron., vol. 20, no. 2, pp. 332–342, Mar. 2005. [11] P.-L. Wong, P. Xu, P. Yang, and F. C. Lee, “Performance improvements of interleaving VRMs with coupling inductors,” IEEE Trans. Power Electron., vol. 16, no. 4, pp. 499–507, Jul. 2001. 95 [12] M. Xu, J. Zhou, K. Yao, and F. C. Lee, “Small signal modeling of a high bandwidth voltage regulator using coupled inductors,” IEEE Trans. Power Electron., vol. 22, no. 2, pp. 399–406, Mar. 2007. [13] D. K. Cheng, L. Wong, and Y. S. Lee, “Design, modeling, and analysis of integrated magnetics for power converters,” in Proc. Power Electronics Specialists Conf. 2000, pp. 320–325. [14] P.-L. Wong, W. Wu, P. Xu, B. Yang, and F. C. Lee, “Investigating coupling inductors in the interleaving QSW VRM,” in Applied Power Electronics Conf. Expo., 2000, vol. 2, pp. 973–978. [15] F. O. Barthold, “Full wave buck-boost power converter with buck power converter properties,” U.S. Patent 5 436 818, Jul. 25, 1995. [16] G. Q. Morris, “Magnetically Integrated Full Wave dc to dc Converter,” U.S. Patent 5 555 494, Sep. 10, 1996. [17] Y. Jang and M. M. Jovanovic, “New two-inductor boost converter with auxiliary transformer,” IEEE Trans. Power Electron., vol. 19, no. 1, pp. 169–175, Jan. 2004. [18] B.-R. Lin and F.-Y. Hsieh, “Soft-switching zeta-flyback converter with a buck-boost type of active clamp,” IEEE Trans. Ind. Electron., vol. 54, no. 5, pp. 2813–2822, Oct. 2007. [19] H. S.-H. Chung, S. Y. R. Hui, and W.-H. Wang, “A zero-current-switching PWM flyback converter with a simple auxiliary switch,” IEEE Trans. Power Electron., vol. 14, no. 2, pp. 329–342, Mar. 1999. [20] H. Martin, “Topology for miniature power supply with low voltage and low ripple requirements,” U.S. Patent 4 618 919. [21] H. Wang, C. Gong, H. Ma, and Y. Yan, “Research on a novel interleaved flyback DC/DC converter,” in Proc. IEEE Conf. Industrial Electronics Applications, 2006, pp. 1–5. [22] C. A. Gallo, A. S. Morais, F. L. Tofoli, L. C. de Freitas, V. J. Farias, E. A. A. Coelho, J. B. Vieira, Jr., and J. A. C. Pinto, “Association of a high power factor power supply using an interleaved boost-flyback converter and a soft-switching full bridge topology,” in Proc. Applied Power Electronics Conf. Expo., 2004, vol. 2, pp. 717–722. Ting Qian received the B.S and M.S. degrees in electrical engineering from Zhejiang University, Hangzhou, Zhejiang, China, in 1999 and 2002, respectively, and the Ph.D. degree in electrical engineering from Northeastern University, Boston, MA, in 2007. He is currently with Texas Instruments, Warwick, RI. His research interests include high-density dc–dc power converters, active power filters, and digital control on power electronics. Brad Lehman (M’92) received the B.E.E. degree from the Georgia Institute of Technology, Atlanta, the M.S.E.E. from University of Illinois at Champaign-Urbana, and the Ph.D. E.E. from the Georgia Institute of Technology in 1987, 1988, and 1992, respectively. Previously, he was a Hearin Hess Distinguished Assistant Professor at Mississippi State University, Mississippi State, MS, an NSF Presidential Faculty Fellow, a recipient of an Alcoa Science Foundation Fellowship, a Visiting Scientist at the Massachusetts Institute of Technology (MIT), and the head swimming and diving coach at the Georgia Institute of Technology. He is presently a Professor in the Department of Electrical and Computer Engineering at Northeastern University, Boston, MA. He conducts esearch in the areas of power electronics, with a primary focus of his research is in the modeling, design and control of dc–dc converters with applications to high brightness LEDs, battery chargers, telecommunication power supplies, and VRMs. Prof. Lehman serves as an Associate Editor of the IEEE TRANSACTIONS ON POWER ELECTRONICS, and from 1993–1997, he served as an Associate Editor for the IEEE TRANSACTIONS ON AUTOMATIC CONTROL. In 1999, he served as a science advisor to the to Commonwealth of Massachusetts, Science and Technology Committee (State Senate) for the Y2K issue in the Power Industry.