IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 24, NO.

5, MAY 2009

1267

Digital Combination of Buck and Boost Converters to Control a Positive BuckBoost Converter and Improve the Output Transients
Young-Joo Lee, Student Member, IEEE, Alireza Khaligh, Member, IEEE, Arindam Chakraborty, and Ali Emadi, Senior Member, IEEE

AbstractA highly efcient and novel control strategy for improving the transients in the output voltage of a dcdc positive buckboost converter, required for low-power portable electronic applications, is presented in this paper. The proposed control technique can regulate the output voltage for variable input voltage, which is higher, lower, or equal to the output voltage. There are several existing solutions to these problems, and selecting the best approach involves a tradeoff among cost, efciency, and output noise or ripple. In the proposed method, instead of instantaneous transition from buck to boost mode, intermediate combination modes consisting of several buck modes followed by several boost modes are utilized to distribute the voltage transients. This is unique of its kind from the point of view of improving the efciency and ripple content in the output voltage. Theoretical considerations are presented. Simulation and experimental results are shown to prove the proposed theory. Index TermsControl, efciency, Li-ion battery, positive buck boost converter, transients.

I. INTRODUCTION

VERY common power-handling problem, especially for portable applications, powered by batteries such as cellular phones, personal digital assistants (PDAs), wireless and digital subscriber line (DSL) modems, and digital cameras, is the need to provide a regulated noninverting output voltage from a variable input battery voltage. The battery voltage, when charged or discharged, can be greater than, equal to, or less than the output voltage. But for such small-scale applications, it is very important to regulate the output voltage of the converter with high precision and performance. Thus, a tradeoff among cost, efciency, and output transients should be considered [1][5]. A common power-handling issue for space-restrained applications powered by batteries is the regulation of the output voltage in the midrange of a variable input battery voltage. Some of the common examples are 3.3 V output with a 34.2 V Li cell

Manuscript received August 2, 2007; revised May 29, 2008, and January 12, 2009. Current version published May 1, 2009. This paper was presented in part at the 2006 IEEE Power Electronics Specialists Conference. Recommended for publication by Associate Editor J. A. Cobos. The authors are with the Electric Power and Power Electronics Center, Electrical and Computer Engineering Department, Illinois Institute of Technology, Chicago, IL 60616 USA (e-mail: ylee35@iit.edu; khaligh@ece.iit.edu; chakari@iit.edu; emadi@iit.edu). Color versions of one or more of the gures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identier 10.1109/TPEL.2009.2014066

input, 5 V output with a 3.66 V four-cell alkaline input, or a 12 V output with an 815 V leadacid battery input [6][9]. With an input voltage range that is above and below the output voltage, the use of a buck or a boost converter can be ruled out unless cascaded. Cascaded combination of converters results in cascaded losses and costs; therefore, this approach is seldom used. In such a range of power demand, the transition of dc voltage from one level to another is generally accomplished by means of dc/dc power converter circuits, such as step-down (buck) or step-up (boost) converter circuits. There are various topologies such as inverting buckboost converters, single-ended primary inductance converters (SEPICs), Cuk converters, isolated buckboost converters, and cascaded buck and boost converters, which can be implemented to maintain a constant output voltage from a variable input voltage [10][13]. The important points of concern for such low-voltage-range power supplies are output ripple, efciency, space, and the cost. The aforementioned topologies are generally not implemented for such power supplies due to their lower efciency, higher size, and cost factors. The most difcult problem is the spikes in the output voltage, which causes the converter to lose efciency during the transition from buck mode to the boost mode. Cost, size, switching speed, efciency, and exibility all need to be considered in designing such power supplies. The advantage of having higher efciency is longer runtime at a given brightness level from the same set of batteries [14][20]. In this paper, a novel theory is presented in order to fulll the requirements of energy-efcient power supplies for batterypowered portable applications. This manuscript extends [1] and [2] by adding the experimental results, to justify the theoretical results. The two main important factors, the efciency and the voltage regulation, which are derived numerically from the experimental results, are shown, and their comparison with the conventional methods is tabulated. The proposed method improves the transition problem and tries to reduce the transients happening during the transition from the buck mode to the boost mode. The paper is organized as follows. Section II discusses the operation of the converter in different modes and issues associated with transition from buck to boost. Existing topologies and solutions to deal with this problem are addressed in Section III. The proposed new method of transient improvement is presented in Section IV. Section V addresses the simulation results applying the conventional and proposed methods of control. Analytical studies for each control technique are shown, and the method of control is elaborated. Experimental results

0885-8993/$25.00 2009 IEEE

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

1268

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 24, NO. 5, MAY 2009

Fig. 1.

Inputoutput curve for the power supply.

Fig. 3. Variations of duty cycle of buck, buckboost, and boost converters versus V in V o u t .

Fig. 2.

Duty ratio variation for the buck and boost modes. Fig. 4. (a) General block diagram of error amplier and drivers. (b) Operation modes.

are presented in Section VI to verify the simulations that are carried out for different control techniques. Finally, conclusions are drawn based on each method in Section VII. II. TRANSITION PROBLEM An example for a battery-powered application is shown in Fig. 1. The input voltage of the battery when fully charged is 6 V and when discharged is 3.6 V [21]. This supply needs to continuously provide a steady output of 5 V. Thus, the converter needs to operate in the buck mode for the period TA , followed by the buckboost mode for TB , and nally in the boost mode for TC . The change in the duty ratio for different modes of operation based on different ranges of the input voltage is shown in Fig. 2. Referring to Fig. 1, for the period TA , the converter works in the buck mode, where the duty changes from the minimum value when the input is 6 V to the maximum when the input is approximately between 5.1 and 4.9 V, as shown in Fig. 2; for the time TB , the converter is in buckboost mode; and, nally, for the period TC , it is working in the boost mode. The points A and B in Fig. 2 are in reference to those in Fig. 1. A. Analytical Studies In buck topology, when Vin is equal to Vout , the duty cycle will approach to 1, Dbuck = Vout /Vin . In the boost operating topology, when Vin approaches Vout , the duty cycle moves to-

ward zero, Db o ost = 1 (Vin /Vout ). In the buckboost operating condition, since Dbuckb o ost = Vout /Vin + Vout , therefore, for Vin equal to Vout , duty cycle becomes 0.5. In other words, when Vin decreases toward Vout , the duty cycle should follow the pattern of 1 for buck to 0.5 for boost and then zero in the boost operating topology. This is demonstrated in Fig. 3. Instantaneous transition from buck to buckboost or boost incorporates a sudden change in the duty cycle from Dbuck,m ax = 1 to Dbuck-boost = 0.5 or Db o ost,m in = 0. Output voltage variations are associated with the sudden changes in the duty cycle change, due to the fact that output voltage is a nonlinear function of the duty cycle. This is shown in Section III. Transition between different modes of operation is a result of comparing the output of the error amplier (EAout ) and its level shifted value (VLS) with the sawtooth waveform, as shown in Fig. 4. Since during a transition from buck to buck boost the duty cycle changes from 1 to 0.5, the EA output signal should be level shifted by 0.5 Vpp . However, to prevent toggling between buck mode and buckboost mode, the VLS should be less than 0.5 Vpp . In buck mode, Dbuck = EAout . Vin approaches Vout , and Dbuck approaches 1. In buck boost mode, Dbuck-boost = EAout 1 + 0.5. When Vin is equal to Vout , Dbuck-boost becomes 0.5. In boost mode, Db o ost = EAout 1 + , where = (Vout,m ax Vin )/Vout,m ax from

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

LEE et al.: DIGITAL COMBINATION OF BUCK AND BOOST CONVERTERS TO CONTROL A POSITIVE BUCKBOOST CONVERTER

1269

buckboost mode. Therefore, in each mode, the sensitivity of the control parameter EAout with respect to time is (Dbuck /t) = (EAout /t), (Dbuckb o ost /t) = (EAout /t), and (Db o ost /t) = (EAout /t). Based on the relationship of the inputoutput voltage of the buck (Dbuck = Vout /Vin ), buckboost ([Vout /Vin ] = [Dbuckb o ost /1 Dbuckb o ost ]), and boost ([Vout /Vin ] = [1/1 Db o ost ]) operation topologies, we have (Vout /Vin ) Dbuck =1 Dbuck t (Vout /Vin ) Dbuck-boost 1 = Dbuck-boost t (1 Dbuck-boost )2 (Vout /Vin ) Db o ost 1 = . Db o ost t (1 Db o ost )2 (1) (2) (3)

Fig. 5.

Cascaded boost and buck converter.

Based on (1), output voltage of the converter in buck mode is a linear function of the duty cycle. In other words, for the case of switching frequencies higher than natural frequency of the system, the controller is able to regulate the transient in the output voltage. However, the output voltage of the converter in boost and buckboost modes is a nonlinear function of the duty cycle, based on (2) and (3). This means that the direct transition from buck to buckboost or boost will incorporate output voltage variations, which cannot be well regulated by conventional control techniques. III. EXISTING SOLUTIONS A. Demerits of Using an Ordinary BuckBoost Converter There are various issues about an ordinary buckboost converter, which prevents its use for such specic applications. The biggest problem associated with such a converter is that the output of such converter is inverted. Of course, it can be inverted, but it requires a transformer, which adds to the cost and space and sacrices the efciency of the converter. B. Demerits of Using SEPIC Converter A very popular buckboost topology that requires more components but produces a noninverting output is the single-ended primary inductance converter (SEPIC). SEPIC, a popular buck boost circuit has limited efciency (usually maximum about 85%) and requires either a transformer or two inductors. Thus, including a transformer or two inductors would occupy more space and, thereby, increase the size and the cost. The use of these components would add to the losses, and thereby degrade the efciency of the converter [10], [11]. C. Disadvantages of a Cascaded Buck and Boost Converter Such a topology applies two dcdc converters cascaded together (see Fig. 5); hence, the loss of the whole single converter is actually doubled in this case, which results in poor efciency. The number of external components such as inductors, decoupling capacitors, and the compensation networks needed for both controllers in this case is more. Due to more components, more space is occupied, which results in higher cost [12]. In

Fig. 6.

Cascaded boost converter and LDO regulator.

Fig. 7. Combined-method-based control logic for deciding modes of operation.

addition, subharmonic problem [22] is another issue, which prevents utilizing cascaded converters. D. Boost Converter Cascaded to a Low-Dropout (LDO) Voltage Regulator This topology is another cascaded topology composed of two different converter circuits. The rst one is a boost converter, followed by an LDO voltage regulator (see Fig. 6). Here, the varying voltage of the battery is stepped up using the boost converter, and then the output of the boost converter is regulated using the LDO to obtain a voltage in the middle range of the varying battery voltage. The biggest disadvantage of an LDO circuit is lower efciency, due to the fact that it is a cascaded network of two converters, which, in turn, double the number of loss components used in the proposed converter stage. Another concern here is the cost; thus, this topological control is not much desired [12]. IV. PROPOSED NEW METHOD The proposed method is to add interface modes, which are a combination of buck and boost operating topologies (Figs. 7 and 8). As shown in Fig. 8, when the input voltage is considerably higher than V1 , the converter operates in purely buck mode. However, during the time period, where the input voltage is between V1 and V2 , threshold voltage, the combination mode A comes into operation, followed by the buckboost mode for the voltage range V2 and V3 . In the voltage range V3 and V4 , the converter operates in the combination mode B. Finally, for the input voltages below V4 , the converter operates purely in the boost operating mode. By adding the combination modes A and B during the time periods T1 and T3 just before and after the stage, where vin vout , the transient at the output of the converter can

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

1270

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 24, NO. 5, MAY 2009

Fig. 10.

Circuit topology of a positive buckboost converter.

Fig. 8.

Voltage curves for combined-method-based control.

Fig. 9.

Block diagram of the theory of digital combination of power converters.

be improved signicantly. Operation of the converter in buck boost mode decreases the efciency of the converter. In order to improve its efciency, buckboost mode should be eliminated. This is another major contribution of this paper. In that case, time periods T2 will be eliminated and the operation mode will change from buck to the boost through intermediate combination modes. Each combination mode is a combination of several buck and boost operating topologies. In the proposed method, instead of a sudden transition from buck operating mode to a buckboost operating mode, a combination of buck and boost operating topologies is applied to distribute the voltage transient and, therefore, obtain smoother output waveform. This is the concept of digital combination of power converters (DCPCs), which is applied to a noninverting buckboost converter in this paper. In combination mode A, the number of buck operating topologies is usually higher than the number of boost operations. Similarly, in the combination mode B, the number of boost operating modes will be higher than buck operations. The concept of DCPC has been introduced in Section IV-A, and then Sections IV-B and IV-C present the analyses and operation of the positive buckboost converter. A. Digital Combination of Power Converters Control approach based on DCPC improves the dynamic response of the converter during transients by switching between different converter topologies to spread out the voltage spikes that are an inevitable part of the transients. Fig. 9 demonstrates the block diagram of the proposed approach. In this gure, converter topology control (CTC) controls the operating topologies

of the converter. In addition, transition control (TC) units control the transitions between various topologies. For instance, TCi controls the transition behavior of the converter from the ith converting topology to the (i + 1)th topology. In the transition from ith topology to the (i + 1)th topology, instead of instantaneous transition from ith topology to (i + 1)th topology, for i switching cycle, converter operates in ith topology and for i switching cycle, it operates in the (i + 1)th topology. Operation with i and i switching cycles in transition mode will be repeated for i times. For a specic case, where i = mi , i = 1 and i = mi 1, in the transition from the ith topology to the (i + 1)th topology, for the mi switching cycle, the converter operates in ith topology and for one switching cycle, it switches to the (i + 1)th topology. Then, in the next cycle, TCi tries to increase the number of cycles of operation in the (i + 1)th topology and decrease the number of cycles of operation in the ith topology. Then, in the next cycle, TCi tries to increase the number of cycles of operation in the (i + 1)th topology and decrease the number of cycles of operation in the ith topology. Finally, before the complete transition to the (i + 1)th topology, for one switching period, it operates in the ith topology and, for mi switching cycle, it switches to the (i + 1)th topology. DCPC imposes the fact that the proposed converter topology should have the capability of operating in various converter topologies with only controlling the switching components of the circuit. B. Operation of Positive BuckBoost Converter The circuit topology of a positive buckboost converter is shown in Fig. 10. In buckboost operating mode, always, two switches, Q1 and Q2 , and two diodes, D1 and D2 , are switching in the circuit. A positive buckboost converter can operate as a buck converter by controlling switch Q1 and diode D1 , when Q2 is OFF and D2 is conducting. It can also work as a boost converter by controlling switch Q2 and diode D2 , while Q1 is ON and D1 is not conducting. When the voltage of the battery is more than the output reference voltage, converter operates as a buck converter. As soon as the voltage of the battery drops to a value less than the output reference voltage, the converter should switch to boost mode. The added advantage of the converter is that the output of such a converter is always positive [16][20]. Table I presents the values of 1 , 1 , and 1 for the conventional method of transition from buck to buckboost and the values of 2 , 2 , and 2 from buckboost to the boost operating conditions.

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

LEE et al.: DIGITAL COMBINATION OF BUCK AND BOOST CONVERTERS TO CONTROL A POSITIVE BUCKBOOST CONVERTER

1271

TABLE I TRANSITION MODE CONTROL PARAMETERS (, , AND ) FOR THE CONVENTIONAL METHOD OF TRANSITION FROM A BUCK TO BOOST OPERATING TOPOLOGY

TABLE II PARAMETERS OF POSITIVE BUCKBOOST CONVERTER

Now, for the output voltage variation to be minimum or zero, s (/t) vout |CombA = 0 ; thus 1 (1db o ost )2 (vout /2L) db o ost T I0,b o ost + (1/R) vout = . 1 (1dbuck ) (vout /2L) T +I0,buck (1/R) vout (6) Similarly, for the combination mode B, the ratio of boost and buck pulses can be dened in terms of circuit parameters. By implementing the proposed combination modes, since transients get almost evenly distributed due to the application of these combined modes, the output voltage transients will signicantly improve. Fig. 12 presents the control signals for the combination modes A and B. V. SIMULATION RESULTS C. Analysis of the Combined Method In period combination mode A, combining 1 buck samples and 1 boost samples, the output voltage variation can be expressed as vout t 1 = C(1 + 1 ) 1 i L ,buck + 1 1 vout R 1 vout R (4) where i L ,buck = (vin vout /2L) dbuck T + I0,buck , i L ,b o ost = (vin /2L) db o ost T + I0,b o ost , and I0,buck and I0,b o ost represent the initial currents of the inductor in the buck and boost modes, respectively vout t CombA = 1 C(1 + 1 ) vin vout 1 dbuck T + I0,buck vout 1 2L R vin 1 db o ost T + I0,b o ost vout +1 (1db o ost ) 2L R The simulation results have been obtained for the converter based on the parameters shown in Table II. A. Conventional Method of Solving the Transition Problem Simulations are carried out on the positive buckboost converter using the conventional methods. Fig. 13 presents the output voltage waveform, buck and boost pulses for a direct transition from buck to boost mode. There is about 12%14% ripple in the output voltage during direct transition from buck to boost. Fig. 14 shows the output voltage, input voltage, and buck and boost pulses for a transition from buck to boost with an intermediate buckboost mode. In this method, the converter initially works in the buck mode, when the input voltage is greater than the output voltage, followed by the buckboost mode when the voltages are almost equal. Finally, the converter works in the boost mode when the input voltage is lower than the output voltage. The simulation results indicate that the presence of spikes in the output voltages during transitions through different modes is about 6%. B. Simulation Results of the Proposed Method (5) Applying the parameter values from Table II for the calculation of buck and boost samples and using (6), just before vin vout , the rounded ratio of 1 and 1 is 3:1. Similarly, just after vin vout , the ratio of 2 and 2 is found to be 1:2. Thus, we choose 1 = 3 or three buck cycles and 1 = 1 or one boost cycle for the time period in combination mode A and 2 = 1

Fig. 11.

Closed loop control strategy for the proposed method.

The overall system level closed loop control strategy of the proposed method is shown in Fig. 11. The control logic for deciding the modes of operation is based on the idea of Fig. 8. Here, both the input and the output voltages are sensed and the proper duty ratios are applied to the switches Q1 and Q2 based on proper duty setting and the desired mode of operation.

CombA

(1 db o ost ) i L ,b o ost

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

1272

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 24, NO. 5, MAY 2009

Fig. 12. Buck, boost, and buckboost pulses in (a) transition from buck mode to combination mode A, (b) combination mode A, (c) transition from combination mode A to buckboost mode, (d) transition from buckboost mode to combination mode B, (e) combination mode B, and (f) transition from combination mode B to boost mode.

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

LEE et al.: DIGITAL COMBINATION OF BUCK AND BOOST CONVERTERS TO CONTROL A POSITIVE BUCKBOOST CONVERTER

1273

Fig. 15.

Proposed number of cycles for the addressed application.

Fig. 13. Input and output voltages and buck and boost pulses for direct transition from buck to boost mode. Fig. 16. Input and output voltages for the transition from buck to boost through combination mode A, buckboost, and combination mode A.

Fig. 14. Input and output voltages and buck and boost pulses for transition from buck to buckboost and then to boost mode.

or one buck cycle and 2 = 2 or two boost cycles for the time period in combination mode B. This ratio is presented in terms of block diagram in Fig. 15 for solving the addressed problem. 1) Proposed Combination Method With BuckBoost Mode in the Middle: The simulations were carried out on the converter using the exact combination method along with the buckboost in the middle. This method improves the ripple content in the output voltage of the converter when the input voltage becomes almost equal to the output voltage and during other transition modes. The waveforms are shown in Fig. 16. It is seen that the peak transient happening during the transition is about 4%. 2) Without BuckBoost Mode in the Middle: The buck boost mode, in the middle, was neglected to save the efciency of the converter, since during this mode of operation, both the switches are operated simultaneously. By applying this combination method of control and simulating the converter, the results shown in Fig. 17 are obtained. The simulation results show that output voltage transients during transition from combination mode A to combination mode B are somehow similar to transients available in transition from combination mode A to buckboost mode. This voltage variation in this method is about 4%; however, canceling the buckboost operating mode in between

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

1274

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 24, NO. 5, MAY 2009

Fig. 19.

Experimental circuit conguration.

VI. EXPERIMENTAL RESULTS The hardware of a positive buckboost converter is designed based on the parameters listed in Table II. Then the converter operates at 100 kHz switching frequency. Two n-type MOSFET switches and two Schottky barrier diodes are used for positive buckboost converter conguration. The MOSFET switches and diodes are IRF540 and 1N5817, respectively. Controller has been implemented using a Texas Instrument digital signal processor (DSP) (320F2812). The output voltage reference is set to 5.0 V, and input voltage varies from 6.0 to 3.6 V. Two high-side gate drivers are designed for IRF540. The dedicated gating logic is applied to the converter for proper operation as buck, buckboost, and boost converters. The operating modes are dependent on mode selection signals, applied from DSP. The overall conguration of converter and controller is shown in Fig. 19. G1 and G2 are buck pulse and boost pulse, sequentially. MD_SEL0, MD_SEL1, and MD_SEL2 determine operation modes: buck, combination mode A; buckboost, combination mode B; and boost, in each control period. Fig. 20 presents the experimental setup that is composed of DSP controller, gate driver, gating logic, and converter. Fig. 21 demonstrates the output voltage of the converter, buck pulses, and boost pulses during a direct transition from buck mode to boost mode. Output voltage has a considerable variation during the transition from buck to boost. Figs. 21 and 22 present the output voltage of the converter, buck pulses, boost pulses, and buckboost pulses for an indirect transition from buck to boost. In this case, transition from buck to boost is carried out through a transition from buck to buck boost mode, followed by a transition from buckboost to boost mode. Fig. 23 presents the input and output voltages of the converter in a whole range of operation. Input voltage decreases from 6 to 3.6 V. Output voltage is tightly regulated at 5 V. In this experiment, transition from buck to boost is a combination of four different operating modes, as shown in Fig. 12. Fig. 12(a) demonstrates the output voltage variation as well as buck and boost pulses in transition from buck to combination mode A. Fig. 12(b) demonstrates the buck and boost pulses

Fig. 17. Input and output voltages and buck and boost pulses during transition from buck to combined mode A and then to combined mode B and boost without buckboost mode.

Fig. 18.

Output voltages for all four different transition methods.

signicantly improves the efciency of the converter. This is also proved through the measurements in Section VI. Fig. 18 presents the output voltages for all different transition methods discussed.

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

LEE et al.: DIGITAL COMBINATION OF BUCK AND BOOST CONVERTERS TO CONTROL A POSITIVE BUCKBOOST CONVERTER

1275

Fig. 20.

Experimental setup composed of (a) converter, (b) DSP controller, and (c) gate drivers.

Fig. 21.

(a) Output voltage, buck, and boost pulses in a direct transition from buck to boost and (b) zoomed view of the waveforms.

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

1276

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 24, NO. 5, MAY 2009

Fig. 22.

(a) Output voltage, buck, and buckboost pulses in transition from buck to buckboost and (b) zoomed view of the waveforms.

Fig. 23. Input and output voltages for the transition from buck to boost mode through proposed method.

Fig. 24. Output voltage, buck pulses, and boost pulses in transition from combination mode A to combination mode B.

in combination mode A, where each three buck pulses are associated with one boost pulse. In Fig. 12(c), buck, boost, and buckboost pulses in transition from combination mode A to buckboost mode are demonstrated. Similarly, Fig. 12(d) shows

the buck, boost, and buckboost pulses in transition from buck boost mode to combination mode B. In addition, Fig. 12(e) illustrates the buck and boost pulses in combination mode B before transition to boost mode is complete. In combination

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

LEE et al.: DIGITAL COMBINATION OF BUCK AND BOOST CONVERTERS TO CONTROL A POSITIVE BUCKBOOST CONVERTER

1277

Fig. 25.

(a) Output voltage, buckboost, and boost pulses in transition from buckboost to boost and (b) zoomed view of the waveforms. TABLE III SUMMARY OF THE OUTPUT VOLTAGE TRANSIENTS DURING TRANSITION FROM DIFFERENT MODES OF OPERATION

mode B, each buck pulse is associated with two boost pulses. Finally, Fig. 12(d) shows the output voltage for a transition from combination mode B to boost mode. In order to improve the efciency of the converter in addition to improving the output voltage transients, dynamic behavior of the system for a direct transition from combination mode A to combination mode B is investigated. Fig. 24 presents the output voltage, buck pulses, and boost pulses in a direct transition from combination mode A to combination mode B. The ripple during direct transition from buck to boost mode is about 12% (Fig. 20). The voltage ripple is dened as the peak peak ripple as a percentage of the output voltage of the converter. This ripple can be descended by adding the interface buckboost mode. Addition of the buckboost mode improves the voltage transient to about 4% in transition from buck to buckboost (Fig. 22) and decreases the ripple transient from buckboost to boost to about 6% (Fig. 25); however, staying in buckboost mode for a long time sacrices the efciency of the converter. Adding combination modes A and B between buck and boost modes signicantly improves the voltage transients, as shown in Figs. 12, 23, and 25. Table III summarizes the output voltage ripple during transients from different modes of operation. Fig. 26 presents the efciency of the converter versus the input voltage variations for various transition modes for different loads. Efciency is a function of load prole. Increasing the load power decreases the efciency because of increasing the ohmic and switching losses. Fig. 26(a) presents the efciency of

the converter in the transition from buck to buckboost and then boost mode. Efciency of the converter in the region where input voltage is approximately equal to output voltage plus diode voltage drop descends to 73% due to the operation of the converter in buckboost mode. Similarly, transition from buck to combination mode A and then buckboost mode sacrices the efciency when input voltage is about 5.4 V. Eliminating the buckboost mode in Fig. 26(c) improves the efciency by about 17% and 20%, respectively, in the cases of R = 5.1 and R = 19.08 . Comparing the voltage regulation plots of Figs. 12 and 2325 and the efciency plots of Fig. 26 demonstrates that the proposed methodology improves the efciency and output voltage regulation during transition from one operating mode

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

1278

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 24, NO. 5, MAY 2009

to the other. Fig. 27 demonstrates the control signals in combination mode A and combination mode B. VII. CONCLUSION A highly efcient control strategy to control a pulsewidth modulation (PWM) dc/dc positive buckboost switching converter has been illustrated in this paper. The proposed control scheme can regulate the output voltage for an input voltage, which changes based on the charge status of the battery supply. The technique introduced in this paper is unique of its kind in improving the efciency and the ripple content of the output voltage for a positive buckboost converter whenever smooth transition is needed from the buck mode to the boost mode. In addition, the concept of DCPC is introduced, which improves the transition ripple by distributing the voltage transients. In this method, the capability of skipping over higher loss interface stages such as buckboost mode in the case of a positive buckboost converter signicantly improves the efciency of the circuit topology. The presented simulations and experimental results validate the proposed theory and its merits. In this manuscript, the proposed theory has been utilized to improve the output voltage transients in transition from buck to boost mode. This is an enabling technology to improve voltage transients in any applications that require transition between different converter topologies. REFERENCES
[1] A. Chakraborty, A. Khaligh, A. Emadi, and A. Pfaelzer, Digital combination of buck and boost converters to control a positive buckboost converter, in Proc. IEEE Power Electron. Spec. Conf., Jun. 2006, vol. 1, pp. 16. [2] A. Chakraborty, A. Khaligh, and A. Emadi, Combination of buck and boost modes to minimize transients in the output of a positive buckboost converter, in Proc. IEEE 32nd Ind. Electron. Annu. Conf., Paris, France, Nov. 2006, pp. 23722377. [3] R. W. Erickson, Fundamentals of Power Electronics, 4th ed. Norwell, MA: Kluwer, 1999. [4] Y. Konishi and Y. F. Huang, Soft-switching buck boost converter using pulse current regenerative resonant snubber, Electron. Lett., vol. 43, pp. 127128, Jan. 2007. [5] L. S. Yang, T. J. Liang, and J. F. Chen, Analysis and design of a novel three-phase ACDC buckboost converter, IEEE Trans. Power Electron., vol. 23, no. 2, pp. 707714, Mar. 2008. [6] B. Sahu and G. A. Rincon-Mora, A low voltage, dynamic, noninverting, synchronous buckboost converter for portable applications, IEEE Trans. Power Electron., vol. 19, no. 2, pp. 443452, Mar. 2004. [7] B. Bryant and M. K. Kazimierezuk, Derivation of the buckboost PWM DCDC converter circuit topology, in Proc. Int. Symp. Circuits Syst., May 2002, vol. 5, pp. 841844. [8] Y. Zhang and P. C. Sen, A new soft-switching technique for buck, boost, and buckboost converters, IEEE Trans. Ind. Appl., vol. 39, no. 6, pp. 17751782, Nov./Dec. 2003. [9] C. Jingquan, D. Maksimovic, and R. Erickson, Buckboost PWM converters having two independently controlled switches, in Proc. IEEE 32nd Power Electron. Spec. Conf., Jun. 2001, vol. 2, pp. 736741. [10] D. Adar, G. Rahav, and S. Ben-Yaakov, A unied behavioral average model of SEPIC converters with coupled inductors, in Proc. IEEE Power Electron. Spec. Conf., Jun. 1997, vol. 1, pp. 441446. [11] W. M. Moussa, Modeling and performance evaluation of a dc/dc SEPIC converter, in Proc. IEEE Appl. Power Electron. Conf., Mar. 1995, vol. 2, pp. 702706. [12] M. Hengchun and V. J. Thottuvelil, Switching controller for a buck+boost converter and method of operation thereof, U.S. Patent 6 037 755, Mar. 14, 2000. [13] X. Ren, Z. Tang, X. Ruan, J. Wei, and G. Hua, Four switch buckboost converter for telecom DCDC power supply applications, in Proc. IEEE 23rd Appl. Power Electron. Conf. Expo., Feb. 2008, pp. 15271530.

Fig. 26. Efciency plot versus the input voltage variations for (a) transition from buck to buckboost and then boost mode, (b) transition from buck to boost through buckboost and combination modes, and (c) transition from buck to boost without buckboost mode.

Fig. 27. Control signals in (a) combination mode A and (b) combination mode B.

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.

LEE et al.: DIGITAL COMBINATION OF BUCK AND BOOST CONVERTERS TO CONTROL A POSITIVE BUCKBOOST CONVERTER

1279

[14] R. S. Weissbach and K. M. Torres, A noninverting buckboost converter with reduced components uses a microcontroller, in Proc. IEEE Southeast Conf., Apr. 2001, vol. 2, pp. 7984. [15] M. Gaboriault and A. Notman, A high efciency, noninverting, buck boost DCDC converter, in Proc. IEEE Appl. Power Electron. Conf., 2004, vol. 3, pp. 14111415. [16] D. M. Dwelley and T. W. Barcelo, Control circuit and method for maintaining high efciency in a buckboost switching regulator, U.S. Patent 6 166 527. [17] A. Syed, E. Ahmed, and D. Maksimovic, Digital PWM controller with feed-forward compensation, in Proc. IEEE Appl. Power Electron. Conf. Expo., 2004, vol. 1, pp. 6066. [18] V. D. Yurkevich, Design of controller for buckboost converter, in Proc. Sci. Technol., 2005, pp. 741745. [19] B. W. Dishner, Bi-directional buck/boost dc/dc converter, U.S. Patent 4 736 151. [20] J. Paulkovich and G. E. Rodriguez, Buck/boost regulator, U.S. Patent 4 245 286. [21] Z. Shameem, Discrete circuit protects Li-ion battery charger, in Proc. Power Electron. Technol. Mag., Mar. 2007, pp. 5051. [22] A. M. Rahimi, A. Khaligh, and A. Emadi, Sub-harmonic production in multi-converter systems, in Proc. IEEE Veh. Power Propulsion Conf., Windsor, U.K., Sep. 2006, pp. 15.

Arindam Chakraborty received the M.S. degree in electrical engineering from Jadavpur University, Kolkata, India, and the Ph.D. degree in electrical engineering from the Illinois Institute of Technology (IIT), Chicago. He was a Research Fellow at the Grainger Power Electronics and Motor Drives Laboratory, IIT, Chicago, for three years, during which he published several papers in micro-, nano- and power electronics and digital control of power electronic converters. He is a Principal Design Engineer in the Electronics and Controls (R&D) of OSI, a Siemens company, for the past two years in Boston, MA. He holds one U.S. patent issued and three pending to his credit.

Young-Joo Lee (S07) received the B.S. degree in electrical engineering from Korea University of Technology and Education, Cheonan, Korea, in 1996, and the M.S. degree from Gwang-Woon University, Seoul, Korea, in 2003. Currently, he is working toward the Ph.D. degree at the Illinois Institute of Technology, Chicago. His Ph.D. research has been focused on integrated bidirectional converter for plug-in hybrid electric vehicles. In 1995, he joined SunStar R&C, Incheon, Korea, which is highly specialized in industrial sewing machines and controllers, and motors and controllers for industrial sewing machines. Then he joined Genoray Company, Ltd., Seongnam, Korea, which manufactures X-ray uoroscopy equipment for medical surgery. He has ten years or more of experience in the industrial eld and has developed many commercial system controllers for sewing machines and medical X-ray uoroscopy equipment that require control over brushless dc (BLDC) motors, permanent magnet synchronous motors (PMSMs), induction motors, stepper motors, high-frequency full-bridge converters, X-ray electron tubes, and other electricpneumatic actuators.

Alireza Khaligh (S04M06) received the B.S. and the M.S. degrees (with highest distinction) in electrical engineering from the Sharif University of Technology, Tehran, Iran, and the Ph.D. degree from the Illinois Institute of Technology (IIT), Chicago. He was a Postdoctoral Research Associate in the Electrical and Computer Engineering Department, University of Illinois (UIUC), Urbana-Champaign. He is currently an Assistant Professor of Electrical and Computer Engineering at IIT, where he has established courses, curriculum, and research in energy harvesting and renewable energy systems. His major research interests include modeling, analyses, design, and control of power electronics converters and motor drives, energy scavenging, biomechanical energy conversion, electric and hybrid electric vehicles, and power management for portable electronics. Dr. Khaligh is an Associate Editor of the IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY. He is a Guest Editor for a special section of IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY on vehicular energy storage systems. He is also a Guest Editor for a special section of IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS on energy harvesting. He is the recipient of the Distinguished Undergraduate Student Award, presented jointly by the Minister of Science, Research, and Technology and the President of Sharif University, and Exceptional Talents Fellowship from the Sharif University of Technology. He is also the recipient of the Highest Standards of Academic Achievement Award at IIT.

Ali Emadi (S98M00SM03) received the B.S. and the M.S. degrees in electrical engineering with highest distinction from the Sharif University of Technology, Tehran, Iran, and the Ph.D. degree in electrical engineering from the Texas A&M University, College Station, TX. He is a Professor of electrical engineering and the Director of the Electric Power and Power Electronics Center (EPPEC), Illinois Institute of Technology (IIT), Chicago, where he has established research and teaching facilities as well as courses in power electronics, motor drives, and vehicular power systems. He is also the Founder, Director, and Chairman of the Board of the Industry/Multiuniversity Consortium on Advanced Automotive Systems (IMCAAS). Prof. Emadi was the General Chair of the 2005 IEEE Vehicle Power and Propulsion and SAE Future Transportation Technology Joint Conference. He is the author/coauthor of over 180 journal and conference papers as well as several books, including Vehicular Electric Power Systems (New York, NY: Marcel Dekker, 2003), Energy Efcient Electric Motors (New York, NY: Marcel Dekker, 2004), Uninterruptible Power Supplies and Active Filters (Boca Raton, FL: CRC Press, 2004), and Modern Electric, Hybrid Electric, and Fuel Cell Vehicles (Boca Raton, FL: CRC Press, 2004). He is an Associate Editor of the IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, and IEEE TRANSACTIONS ON POWER ELECTRONICS. He is also the Editor of the Handbook of Automotive Power Electronics and Motor Drives (New York, NY: Marcel Dekker, 2005). He is the recipient of numerous awards and recognitions. He has been named the Eta Kappa Nu Outstanding Young Electrical Engineer of the Year 2003 by virtue of his outstanding contributions to hybrid electric vehicle conversion. He also received the 2005 Richard M. Bass Outstanding Young Power Electronics Engineer Award from the IEEE Power Electronics Society. In 2005, he was selected as the Best Professor of the Year by the students at IIT. He is the recipient of the 2002 University Excellence in Teaching Award from IIT as well as the 2004 Sigma Xi/IIT Award for Excellence in University Research. He directed a team of students to design and build a novel motor drive, which won the First Place Overall Award of the 2003 IEEE/DOE/DOD International Future Energy Challenge for Motor Competition.

Authorized licensed use limited to: GOVERNMENT COLLEGE OF TECHNOLOGY. Downloaded on August 5, 2009 at 23:42 from IEEE Xplore. Restrictions apply.