WO2011119102A1

WO2011119102A1 - High frequency switch mode power supply capable of gate charge recovery

Info

Publication number: WO2011119102A1
Application number: PCT/SG2010/000110
Authority: WO
Inventors: Sajith Yasanga Goonesekera
Original assignee: Individual
Current assignee: Individual
Priority date: 2010-03-22
Filing date: 2010-03-22
Publication date: 2011-09-29
Anticipated expiration: 2012-09-22

Abstract

A method of operating switch mode power supply, capable of gate charge recovery comprising an oscillator; a power switch device; a gate driver; and a transformer unit, wherein gate driver configured to carry out activation and deactivation of the power switch device and in order to deactivate and recover gate charge by discharging gate charge of said power switch device to gate charge recovery circuit, said gate driver configured to have a gate charge recovery path to discharge gate charge of the power switch device through a winding of said transformer unit and wherein said gate driver, controlled by gate drive signal, configured to, provide capability to operate said power switch device in order to magnetize the magnetic core of said primary transformer when the gate drive signal having frequency around resonance frequency of the magnetic core of said primary transformer.

Description

HIGH FREQUENCY SWITCH MODE POWER SUPPLY

CAPABLE OF GATE CHARGE RECOVERY

Technical Field

[1] This invention relates to switch mode power supply, and more particularly, a

improved method of operating a switch mode power supply, in which operating in high frequency and capable of recovering gate turn-off loss of the power switch device.

Background Art

[2] A switched-mode power supply is an electronic power supply unit that incorporates a switching regulator in order to provide the required output voltage. A switch mode power supply is a power converter that transmits power from a source (e.g. a battery or the electrical power grid) to a load (e.g. a personal computer, notebook computer, mobile phone). The function of the converter is to provide a reliable output voltage often at a different level than the input voltage.

[3] Essentially switched mode power supplies act as DC to DC converters by first rectifying an AC input voltage (110V / 240V) to DC and depending upon the design considerations chosen, chops this DC in a 'chopper' and converts it to a higher and/or lower level of DC.

[4] Almost all electrical devices used today requires some sort of a power supply in order to convert available electrical power source to required electrical power level needed by the electronic or electrical device. Currently this is accomplished by switch mode power supplies which operate at low frequencies (e.g. ~60Khz, ~150Khz, ~250Khz) in which gate drive losses are negligible.

Disclosure of Invention

Technical Problem

[5] Today number of mobile computer and mobile phone users are increases every day. And the mobile equipment like portable computers and mobile phones are getting smaller. But the size of the power supply used to supply power and charge the battery did not get significantly smaller. If the size of the power supply of the mobile electronic devices can be reduced, traveling with mobile equipments can be made more convenient.

Technical Solution

[6] One of the ways to reduce the size of the switch mode power supply is by reducing the size of the components used in it. In switch mode power supply three largest components are the filtering inductors, transformer and smoothing capacitors. By designing to operate switch mode power supply in high frequencies, the sizes of the smoothing capacitors used to minimize ripples in the output stage and filtering inductors can be reduced. And by using resonance frequency of the magnetic core of the transformer as illustrated below which in most cases may be in- high frequencies, the amount of energy transferred from primary side to secondary side can be increased by using magnetic materials like ferrite or other ferritic materials as magnetic core.

[7] As illustrated in FIG. 1, output wave 4 will have pulse 7 (circled in dotted line) , when a transformer 6 having a magnetic core 8 is supplied with a square wave input signal 9, if the input signal 9 has high slew rate rising edge 5, specifically having enough slew rate necessary to transfer magnetic core 8 resonance frequency. This pulse 7 which has a width approximately half a cycle width of the resonance frequency of the magnetic core 8 is the preferred pulse width of this invention. The amplitude of the output pulse 7 depends on type of the magnetic core 8, ratio of primary 2 and secondary 3 turns, slew rate of the raising edge 5 of the input signal and the amplitude of the input signal 9. This is the primary principle incorporated in this invention in order to reduce the size of the magnetic core of the transformer. Resonance frequency of the transformer is the preferred frequency.

[8] Solution proposed by this invention is to operate the switch mode power supply, at resonance frequency of its magnetic cores. When switch mode power supply operates in that frequency, size of the transformer need to transfer power, size of the filtering inductors and the capacitors need to reduce ripples from the output will be smaller. Most of the magnetic cores are capable of transferring higher power when they operate in resonance frequency usually falls in to megahertz range.

[9] The power switch device used in most of the switch mode power supply currently being used are metal-oxide-semiconductor field-effect transistors. These power switch devices activate and deactivate by charging and discharging gate capacitance. During traditional deactivation, power stored in the gate is dissipated as heat. When the switch mode power supply operates in high frequency, gate turn-off loss of the power switch device use as chopper to deliver resonance frequency, become considerably higher.

[10] Gate charge losses: Pgate

[11] The average current required to drive the gate capacitor of the Mosfet:

[12] Igate_avg=fsw.Qg_t0,

[13] Pgate = Igate_avg.Vdr

[14] Wherein;

[15] fsw = Switching frequency

[16] Qg_to, = Total gate charge

[17] Igate_avg= Average current utilized to charge the gate.

[18] As shown in above formula when frequency (fsw) increases, gate charge loss

(Pgate) also increases. In order to minimize the gate charge loss, gate charge recovery circuit has to be incorporated in to the switch mode power supply. Proposed gate charge recovery method is, to transfer the gate charge to a winding of a transformer in order to utilize through a tapping of same winding or another winding of the same transformer, during the deactivation period of the power switch device by the gate driver. This method can be applied to any power switch device which is activated and deactivated by charging and discharging gate capacitance.

[19] Slew rate of a sinusoidal waveform can be illustrated using following formula:

[20] SR = 2pi.fr. Vpk

[21] Wherein fr is the frequency (Resonance frequency of the magnetic core used in the transformer), and Vpk is the peak amplitude of the waveform. Slew rate is usually expressed in units of ν/μ≤. In a sinusoidal waveform maximum slew rate occur at the zero cross point. Maximum slew rate of the resonance frequency is the slew rate at the zero crossing point of the resonance frequency at the same peak voltage.

Advantageous Effects

[22] The main advantage of designing a switch mode power supply having gate charge recovery circuit and magnetic core capable of transferring few times more power when it operate in its resonance frequency is that switch mode power supply can be design in a smaller foot print. By the same principle, sizes of the some of the large components (e.g. capacitors, inductors, transformers) will be smaller and switch mode power supply can be designed in smaller housing. There for switch mode power supply can be made more portable.

Description of Drawings

[23] The accompanying drawings, which are included to provide a further understanding of the invention, illustrate embodiments of the invention and together with the description serve to explain the principle of the invention.

[24] FIG. 1 illustrates the principle behind the resonance frequency of the magnetic core.

[25] FIG. 2 shows simplified block diagram of the switch mode power supply.

[26] FIG. 3 shows first embodiment of the gate charge recovery method.

[27] FIG. 4 shows second embodiment of the gate charge recovery method.

[28] FIG. 5 shows third embodiment of the gate charge recovery method.

[29] FIG. 6 shows fourth embodiment of the gate charge recovery method.

[30] FIG. 7 shows fifth embodiment of the gate charge recovery method.

[31] FIG. 8 shows more detailed block diagram of the switch mode power supply.

[32] FIG. 9 shows embodiment of switch mode power supply wherein the gate driver, the power switch device and transformers contribute to the function the oscillator.

Best Mode

[33] The objective of this invention is to further develop a switch mode power supply in order to reduce in size. This can be achieved by operating switch mode power supply in high frequencies and recovering gate charge losses of power switch device which used to deliver high frequency. There are two main design aspects in the proposed switch mode power supply. The first one is frequency it operates to deliver maximum power and the second one is the way it recovers the power used to activate power switch device, which is gate charge. The optimum operating frequency of the switch mode power supply is the resonance frequency of the magnetic core of the transformer Used in it. If if is design to operates in other frequencies, but if intend to capture some benefits of operating in resonance frequency , minimum slew rate of rising edge of the pulse provided to the primary winding should be at least maximum slew rate of the resonance frequency of the magnetic core at the corresponding voltage level.

[34] Proposed improved switch mode power supply, optimum pulse width to activate the power switch device is around half a cycle width of the resonance frequency of the magnetic core used in the transformer which is the preferred pulse width. While the switch mode power supply operates its maximum capacity, deactivation pulse width is preferred to be the almost same as the preferred pulse width, i.e. power switch device is operating at approximately at same frequency as preferred frequency.

[35] The main aspect of the proposed gate charge recovery method is to transfer gate charge to a winding of a transformer in order to utilize through another winding or tapping of same winding of the same transformer, during the deactivation period of the power switch device. This can be achieved in several different methods. FIG. 2 shows a simplified block diagram of a switch mode power supply. Gate charge recovery system 500 is isolated by dotted line. FIG. 3 to FIG. 7 show different embodiments of the gate driver designs configured to recover the gate charge by the said method which can be incorporated in gate charge recovery system 500 in FIG. 2. The power switch device used in those embodiments are preferably a metal oxide semiconductor field effect (MOSFET) transistor or an insulated gate bipolar transistor (IGBT) or any variation of those two. In all figures which contain capacitor "Ciss", is drawn for illustration purposes only. "Ciss" is the gate capacitor of the power switch device next to it.

[36] FIG. 3 shows first embodiment of the gate driver design, which is capable of gate charge recovery. Output 926 of gate driver 928 is connected to gate 922 of power switch device 920. In this embodiment, gate charge is transferred to transformer 924 in order to extract from secondary side. The power switch device 920 is activated by closing switch 934 and opening switch 936 in gate driver 928 in order to provide current to charge the capacitor of the gate 922. The power switch device 920 is deactivated by opening the switch 934 and closing the switch 936 in the gate driver 928 to recover gate charge by discharging the gate charge along gate discharge path 932 through winding 930 of the transformer 954. The unique difference is gate charge is transferred by the winding 930 to secondary side of the transformer 924 during the deactivation of the power switch device 920.

[37] FIG. 4 shows second embodiment of the gate driver design, which is capable of gate charge recovery. Output 956 of gate driver 958 is connected to gate 952 of power switch device 950. In this embodiment, gate charge is transferred to transformer 954 in order to extract from tapping 963 and/or any secondary winding. The power switch device 950 is activated by closing the switch 964 and opening switch 966 in the gate driver 958 in order to provide current to charge the capacitor of the gate 952. Power switch device 950 is deactivated by opening switch 964 and closing switch 966 in the gate driver 958 to recover gate charge by discharging the gate charge along gate discharge path 962 through tapping 961 of winding 960 of the transformer 954.

Winding 960 having a tapping 963 of the transformer 954, with diode 968 and capacitor 969 can be configured to supply power to primary side of the components, for example gate driver, oscillator etc.

FIG. 5 shows third embodiment of the gate driver design, which is capable of gate charge recovery. Output 976 of gate driver 978 is connected to gate 972 of power switch device 970. In this embodiment, gate charge is transferred to transformer 974 in order to extract from winding 985 and/or any secondary winding. The power switch device 970 is activated by closing switch 984 and opening switch 986 in the gate driver 978 in order to provide current to charge the capacitor of the gate 972. The power switch device 970 is deactivated by opening the switch 984 and closing the switch 986 in the gate driver 978 to recover gate charge by discharging the gate charge along gate discharge path 982 through the winding 980 of the transformer 974. The transformer 974 comprising winding 985 which is connected to diode 988 and capacitor 989 can be configured to supply power to primary side of the components, for example gate driver, oscillator etc.

The transformer used for gate charge recovery in FIG. 3 to FIG. 5 can be shared. But transformer used for gate charge recovery in FIG. 6 and FIG. 7 is dedicated for gate charge recovery. The transformer unit is defined as one or more transformer.

FIG. 6 shows fourth embodiment of the gate driver design, which is capable of gate charge recovery. Output 856 of gate driver 858 is connected to gate 852 of power switch device 850. In this embodiment, gate charge is transferred to transformer 854 in order to extract from tapping 863. The power switch device 850 is activated by closing the switch 864 and opening switch 866 in the gate driver 858 in order to provide current to charge the capacitor of the gate 852. Power switch device 850 is deactivated by opening switch 864 and closing switch 866 in the gate driver 858 to recover gate charge by discharging the gate charge along gate discharge path 862 through tapping 861 of winding 860 of the transformer 854. Winding 860 having a tapping 863 of the transformer 854, with diode 868 and capacitor 869 can be configured to supply power to primary side of the components, for example gate driver, oscillator etc. or secondary side.

FIG. 7 shows fifth embodiment of the gate driver design, which is capable of gate charge recovery. Output 876 of gate driver 878 is connected to gate 872 of power switch device 870. In this embodiment, gate charge is transferred to transformer 874 in order to extract from winding 885. The power switch device 870 is activated by closing switch 884 and opening switch 886 in the gate driver 878 in order to provide current to charge the capacitor of the gate 872. The power switch device 870 is deactivated by opening the switch 884 and closing the switch 886 in the gate driver 878 to recover gate charge by discharging the gate charge along gate discharge path 882 through the winding 880 of the transformer 874. The transformer 874 comprising winding 885 which is connected to diode 888 and capacitor 989 can be configured to supply power to primary side of the components, for example gate driver, oscillator etc. or secondary side.

Fourth (FIG. 6) and fifth (FIG. 7) embodiments can be used to recover gate charge in Buck, Boost, Buck-Boost, Split-Pi (or Boost-Buck) power regulators or any electronic equipment Uses power switch device preferably operate at high frequencies. In above embodiments FIG. 3 to FIG.7 a discharge path may go through a diode in order to control the direction of the current.

FIG. 8 shows more detailed schematic diagrams of a switch mode power supply 20 comprises:

an oscillator 240 having output 242 which provide gate drive signal and may have control input 244;

a gate driver 220 having input 224 and output 222 and may have control input 245; a transformer unit 210;

a rectifying and smoothing apparatus 238 having output 239;

a power source 211;

a feedback controller 230 having input 232 and/or input 235 and/or input 231 and output 234 and;

may have a signal controller 236 having input 247 and output 241 and control input 243.

a power switch device 200 (example metal oxide semiconductor field effect transistor or insulated gate bipolar transistor), which is capable of switching at the resonance frequency of the ferrite core of the transformer. Capacitor 208 ("Ciss") is to illustrate the gate capacitance of the power switch device.

The transformer unit may have one or more transformer. Transformer unit in embodiments shown in FIG. 3, FIG. 4 and FIG. 5 require only one transformer.

Transformer unit in embodiments shown in FIG. 6 and FIG. 7 require two

transformers.

The oscillator 240 is configured to provide preferably square wave gate drive signal 9 as shown in FIG. 1. If the gate driver 220 or any stage, after oscillator 240, is not designed to recondition gate drive signal to deliver enough slew rate, during the rising edge of the activation part of the gate drive signal to power switch device 200, the oscillator 240 may be required to generate gate drive signal with enough slew rate which is the slew rate of the resonance frequency of the magnetic core 218 at corresponding voltage levels. If there are any stages in between the oscillator 240 and gate driver 220, those stages should not distort the slew rate of the gate drive signal beyond the minimum required levels. In order to optimize the performance of the switch mode power supply 20, the first half tl of the signal 9, which is gate activation pulse, should preferably have width of approximately half a cycle width of the resonance frequency of the magnetic core 218 of the transformer 210. The control input 244 may be configured to activate and deactivate the output 242 or change the length of the second part t2 of signal 9 which is deactivation part of the signal 9. To operate the switch mode power supply 20 at its maximum power, the second part t2 of the signal 9 may required to have minimum of approximately half a cycle width of the resonance frequency of the magnetic core 218 of the transformer 210. In other words switch mode power supply 20 should operate approximately resonance frequency of the magnetic core 218 of the transformer 210. The key is the power switch device 200 should be able deliver resonance frequency of the magnetic core 210 to the first winding with minimum distortion for maximum power transfer to secondary side.

The function of the Signal Controller is to control the incoming signal from input 247 to output 241 by controlling input 243. The control input 243 is capable of blocking or unblocking the incoming signal from input 247 to output 241.

In order to regulate the output signal, the feedback controller 230 may have one or more monitoring inputs as listed below.

an input 232 to monitor output of the rectifying and smoothing circuit 238.

an input 235 to monitor winding 233.

an input 231 to monitor winding 214.

If input 235 is not used by the feedback controller 230, winding 233 may not be necessary. Control output 234 of the feedback controller 230 may control the oscillator 240 and/or the signal controller 236 and/or the gate driver 220 in order to regulate the output power. If the feedback controller is not configured to control the signal controller, may not be necessary to have a signal controller and in that case output 242 of the oscillator 240 may be directly connected to input 224 of the gate driver 220 bypassing signal controller 236.

The function of the gate driver is to control the power switch device 200 based on the input signal from the input 224 and the control input 245. The input signal generated by the oscillator 240 is received by the input 224 of the gate driver 220. During the tl period of the signal 9 the gate driver closes switch 226 and opens switch 227 while completes the circuit 221 through the gate 204 and the tapping 215 of winding 214. Power switch device 200 is deactivated to flow the gate charge along gate discharge path 223 through tapping 215 of winding 217 of the transformer 210 by opening switch 226 and closing switch 227 in the gate driver 220. The gate 204 and tapping 215 of winding 214 are connected in parallel. For optimal performance the gate driver have to activate the power switch device with enough slew rate, which is slew rate of the resonance frequency of the magnetic core at corresponding voltage levels during the rising edge of the gate drive signal. Gate driver 220 may be configured to condition the gate driver signal generated by the oscillator to said slow rate for the optimal performance.

[61] Once the startup power is given to start to operate the switch mode power supply 20, diode 246 and capacitor 248 which is connected to tapping 217 can be used to supply power, at least, to oscillator 240 and/or signal controller 236 and/or gate driver 220 and/or feedback controller 230. Tapping 215 and tapping 217 of the winding 214 may be interchanged if it is required for the optimal design.

The function of the power switch device 200 is to supply and cutoff (oscillate) power 225 to primary winding 212 of the transformer 210, based on the preferred input signals described before; Power switch device 200 should be capable of operating at the maximum frequency which is resonance frequency of the magnetic core 218.

Power switch device 200 preferably having low on resistance in order to reduce the conduction loss and the gate capacitance preferably as low as possible.

The transformers 210 magnetic core 218 play a major role in this design. In order transfer power from primary side to secondary side, preferably magnetic core having high magnetizing capability while operating in its resonance frequency. The winding 216 is the secondary winding which deliver the power to input 237 of the rectifying and smoothing circuitry 238 to convert to direct current (DC) and provide to output terminal 239. Winding 214, having tapping 215, works as a primary which absorb gate charge. The tapping 217 works as secondary which is configured to provide power and may provide feedback signal. Optional winding 233 may provide feedback signal also.

Power source 211 may be replaced by or connected to power supply 225 with proper power regulation. Gate charge recovery area of the schematic diagram of switch mode power supply 20 shown in FIG. 8 is marked in dotted line 300.Gate charge recovery embodiment 300 can be replaced by any other embodiment shown in FIG. 3 to FIG. 7.

As shown in FIG. 9 switch mode power supply 40 comprising a power switch device 400, a gate driver 420, a feedback controller 430, a transformer 410 and a oscillator 440, wherein the gate driver 420 and/or the feedback controller 430 and/or the transformer 410 contribute to the function of the oscillator. Gate charge embodiments shown if FIG. 3 to FIG.7 can be incorporated to switch mode power supply 40.

Mode for Invention

Various embodiments have been described in the best mode for carrying out the invention.

Industrial Applicability

This invention is applicable to any device using switch mode power supply to regulate and provide power with desired voltage levels.

Claims

A method of operating switch mode power supply, capable of gate charge recovery comprising:

an oscillator;

a power switch device;

a gate driver; and

a transformer unit,

wherein gate driver configured to carry out activation and deactivation of said power switch device and in order to deactivate and recover gate charge by discharging gate charge of said power switch device to gate charge recovery circuit, said gate driver configured to have a gate discharge path to recover gate charge of the power switch device through a winding of said transformer unit.

A method of operating switch mode power supply, capable of gate charge recovery as claim 1 where the said transformer unit comprises a primary transformer

wherein power switch device configured to activate and deactivate power to the first winding of said primary transformer and characterized by configuration of gate driver to carry out said activation, and in order to deactivate and recover gate charge by discharging gate charge of said power switch device to gate charge recovery circuit, said gate driver configured to have a gate charge recovery path to discharge gate charge of the power switch device through second winding or tapping of the second winding of primary transformer which connects to a circuit configured use recovered gate charge through said transformer's second winding or tapping of second winding or third winding.

A method of operating switch mode power supply, capable of gate charge recovery as claim 1 where the said transformer unit comprises a primary and secondary transformer

wherein power switch device configured to activate and deactivate power to the first winding of the primary transformer and characterized by configuration of gate driver to carry out said activation, and in order to deactivate by discharging gate charge of said power switch device to gate charge recovery circuit, said gate driver configured to have a gate charge recovery path to discharge gate charge of the power switch device through first winding or tapping of first winding of the secondary transformer which connects to a circuit configured use recovered gate charge through secondary transformer's first winding or tapping of first winding or second winding.

A method of operating switch mode power supply capable of gate charge. recovery as claimed above,

wherein said gate driver, controlled by gate drive signal, configured to operate, said power switch device in order to magnetize up to saturation level and demagnetize magnetic core through said first winding of said primary transformer when the gate drive signal providing frequency around resonance frequency of the magnetic core of said primary transformer.

A method of operating switch mode power supply capable of gate charge recovery circuit as in claimed in claim 1, 2 and 3,

wherein said gate driver, controlled by gate drive signal, configured to operate, said power switch device in order to magnetize and demagnetize the magnetic core through said first winding of said primary transformer when the gate drive signal providing frequency around resonance frequency of the magnetic core of said primary transformer.

wherein, during power activation to said first winding of said primary transformer, rising edge of activation having slew rate equal or more than around maximum slew rate of the resonance frequency of the magnetic core at corresponding voltage levels.

wherein, during power activation to said first winding of said primary transformer, rising edge of activation having slew rate around equal or more than slew rate at the zero crossing point of the resonance frequency of the magnetic core at corresponding voltage levels.

wherein, during power activation to said first winding of said primary transformer, rising edge of activation having slew rate more than 80% of slew rate at the zero crossing point of the resonance frequency of the magnetic core at corresponding voltage levels.

A method of operating switch mode power supply capable of gate charge recovery as in claim 4, 5, 6, 7 and 8,

wherein said gate driver signal generated by said oscillator, comprises activation and deactivation parts to operate said power switch device and starting edge of activation part of the gate drive signal having or gate driver configured to condition the gate drive signal to have slew rate equal or more than around maximum slew rate of said resonance frequency of the magnetic core of the primary transformer at corresponding voltage levels.

wherein the minimum width of said activation part of said gate drive signal having around half a cycle width of the resonance frequency of the magnetic core of said primary transformer.

A method of operating switch mode power supply capable of gate charge recovery as in claim 9 and 10 further comprising a feedback controller

wherein the feedback controller stabilizes the output voltage of the switch mode power supply by

enabling or disabling said gate drive signal to said power switch device from the gate driver or any stage before the gate driver and/or changing width of said deactivation part of said gate drive signal. A method of operating switch mode power supply capable of gate charge recovery as claimed in above,

wherein one or more of the gate driver, the power switch device and transformers contribute to function the oscillator.

A method of operating switch mode power supply capable of gate charge recovery as claimed in above,

wherein gate charge recovery path comprises a diode in order to control the direction of the current.

wherein the transformer core is made of ferritic material.

wherein power switch device is field effect transistor.

wherein power switch device is metal oxide semiconductor field effect transistor or insulated gate bipolar transistor or any variation of it.

an oscillator;

a power switch device;

a gate driver; and

a transformer unit,

wherein characterized by configuration of gate driver to carry out activation and deactivation of the power switch device and in order to deactivate and recover gate charge by discharging gate charge of said power switch device to gate charge recovery circuit, said gate driver configured to have a gate charge recovery path to discharge gate charge of the power switch device through a winding of said transformer unit and

wherein, during power activation to said first winding of said primary transformer, rising edge of activation having slew rate around equal or more than slew rate at the zero crossing point of the resonance frequency of the magnetic core at same peak voltage levels.

an oscillator;

a power switch device;

a gate driver; and

a transformer unit,

wherein said gate driver, controlled by gate drive signal, configured to operate, said power switch device in order to magnetize and demagnetize the magnetic core of said primary transformer when the gate drive signal providing frequency around resonance frequency of the magnetic core of said primary transformer and

A portable computer system comprising:

a central processing unit;

a random access memory sub system;

a disk sub system; and

a switch mode power supply capable of gate charge recovery as claimed in above.

A mobile phone system comprising:

a transmitting unit;

a receiving unit; and

a switch mode power supply capable of gate charge recovery as claimed claim 1 to claim 18.