TWI825773B - Flyback power converter having emulated demagnetized signal and primary-side control circuit and control method thereof - Google Patents

Abstract

A flyback power converter includes: a first transistor switching a transformer for generating a primary switching current and an output voltage; and a second transistor generating a circulated current to achieve ZVS (zero voltage switching) of the first transistor; wherein the flyback power converter actively forces at least one switching cycle to be operated in a DCM (discontinuous conduction mode) operation when the primary switching current is determined to have been operating in a non-DCM operation for a predetermined number of switching cycles. The flyback power converter generates a demagnetized signal which emulates the demagnetized time of the transformer for controlling the second transistor during the non-DCM operation. The flyback power converter calibrates the demagnetized signal according to the demagnetized time during the actively fored DCM operation.

Description

Flyback converter with simulated demagnetization signal and primary side control circuit and control Preparation method

The present invention relates to a flyback converter, and in particular to a flyback converter with a simulated demagnetization signal. The present invention also relates to a switching control circuit and a control method for controlling a flyback converter with a simulated demagnetization signal.

Please refer to Figure 1. Figure 1 shows an asymmetrical duty cycle flyback converter (Asymmetrical Duty Cycle Flyback Converter) in the prior art US Pat. No. 5,959,850. This prior art discloses an Asymmetrical Duty Cycle Flyback Converter with zero voltage switching (ZVS). Half-bridge flyback converter to achieve higher power efficiency. Zero-voltage switching can be defined as switching a transistor on when the voltage across the transistor (for example, the drain-to-source voltage) is zero or close to zero. However, a disadvantage of this prior art is that the power conversion efficiency of the power converter is low in light load conditions.

Another shortcoming of the above-mentioned prior art is that the output voltage of the power converter is non-variable. Specifically, if the above-mentioned prior art is to be changed to a zero-voltage switching flyback converter with a variable output voltage, it must be Detect the demagnetization period of its transformer and control the switching of the transformer.

Another prior art, US Patent No. 7,151,681, is a multiple-sampling circuit for measuring reflected voltage and discharge time of a transformer. This prior art discloses a method for detecting the output voltage and discharge time of a transformer. The method of degaussing period, however, this prior art cannot achieve zero-voltage switching of the power converter, which is used for discontinuous conduction mode (DCM) operation.

Figure 2 shows a waveform diagram of a prior art half-bridge flyback converter operating in discontinuous conduction mode in a light load state. The drive signal SH is used to drive the upper-bridge switch of the half-bridge flyback converter to excite the transformer. The drive signal SL is used to drive the lower-bridge switch of the half-bridge flyback converter. The signal waveform of the excitation current IM shows that the transformer operates in Discontinuous conduction mode. When the output power of the half-bridge flyback converter decreases, the pulse width PW of the drive signal SH decreases due to the feedback control of the half-bridge flyback converter, and the pulse width of the drive signal SL also decreases accordingly. Therefore, the half-bridge As the switching frequency of the flyback converter increases, the switching losses also increase. When the driving signal SH turns to a low level (off), the first pulse of the driving signal SL is enabled during the demagnetization period of the transformer. The second pulse wave of the driving signal SL is enabled to generate a circulating current, thereby achieving zero-voltage switching of the upper bridge switch.

The disadvantage of the above-mentioned prior art is that when operating in the discontinuous conduction mode, the driving signal SL needs to be switched on/off twice in one switching cycle. Therefore, the average switching frequency of the driving signal SL is greatly increased, resulting in a large amount of switching losses and leading to Energy loss in the lower-side switch.

Compared with the prior art US Patent No. 7,151,681, the present invention provides a resonant half-bridge flyback converter with omitted cycles to improve power efficiency in medium load and light load operating conditions.

Compared with the prior art US Patent No. 5,959,850, the present invention provides a method for generating a demagnetization signal and a switching control circuit, in which the period of the demagnetization signal is equal to the period of the transformer. During the demagnetization period, the present invention can be used in zero-voltage switching flyback converters with programmable output voltages, such as USB PD power converters.

Compared with the prior art in Figure 2, the present invention provides a control circuit for an asymmetrical half-bridge (AHB) flyback converter, using three transistors to improve the power in the operating states of medium load and light load. conversion efficiency.

From one of the aspects, the present invention provides a flyback converter, including: a first transistor for switching a transformer to generate a primary-side switching current and an output voltage; and a second transistor, Used to generate a circulating current to achieve zero voltage switching (ZVS) of the first transistor; wherein the flyback converter determines that there has been a preset number of switching cycles based on the primary side switching current. When operating in a non-discontinuous conduction mode (non-DCM), the flyback converter actively forces at least one switching cycle to operate in a discontinuous conduction mode (DCM).

In a preferred embodiment, the flyback converter is used to generate a demagnetization signal, wherein the demagnetization signal simulates a demagnetization period of the transformer to control the second transistor during the non-discontinuous conduction mode. , wherein the flyback converter corrects the degaussing signal according to the degaussing period during the actively forced discontinuous conduction mode.

In a preferred embodiment, in the actively forced discontinuous conduction mode, the flyback converter detects a reflected voltage from an auxiliary winding of the transformer to correct the first transistor during the off period. A degaussing signal, in which the length of time the reflected voltage occurs is related to the degaussing period of the transformer.

In a preferred embodiment, the first transistor and the second transistor form a half-bridge circuit, wherein in the actively forced discontinuous conduction mode, after the first transistor is turned off, the second transistor The crystal is controlled off, causing the half-bridge circuit to operate in an asynchronous mode.

In a preferred embodiment, the flyback converter further includes a cycle counter, the cycle counter is used to count the number of switching cycles in which the flyback converter operates in the non-discontinuous conduction mode, so that when the primary side switch When the current has been operating in the non-discontinuous conduction mode for a preset number of switching cycles, the flyback converter is enabled to actively force operation in the discontinuous conduction mode.

In a preferred embodiment, a pulse period of the demagnetization signal is used to control the minimum conduction time of the second transistor, thereby demagnetizing the transformer after the first transistor is turned off.

In a preferred embodiment, a pulse period of the degaussing signal is related to a level of an input voltage in an excitation period of the transformer, a level of the output voltage of the transformer and the first voltage. A conduction period of the crystal.

In a preferred embodiment, the flyback converter further includes a resistor, wherein a pulse period of the degaussing signal is determined based on the resistance value of the resistor; wherein the flyback converter is based on the transformer in the transformer. The resistance value of the resistor is adjusted during the degaussing period during the active forced discontinuous conduction mode to correct the degaussing signal.

In a preferred embodiment, after turning off the first transistor during the actively forced discontinuous conduction mode, a portion of a demagnetization current of the transformer flows through a body diode of the second transistor.

In a preferred embodiment, the flyback converter is a resonant flyback converter and further includes a resonant capacitor, and the resonant capacitor is coupled in series to the transformer.

In a preferred embodiment, a first pulse wave of a second driving signal after the discontinuous conduction mode is used to turn on the second transistor to provide an exciting current from the resonant capacitor to The transformer energizes the transformer, thereby generating a negative circulating current to achieve zero-voltage switching of the first transistor.

From another point of view, the present invention also provides a switching control circuit for controlling a flyback converter, wherein the flyback converter includes: a first transistor for switching a transformer to generate a The secondary side switching current and an output voltage; and a second transistor to generate a circulating current to achieve zero voltage switching (ZVS) of the first transistor, the switching control circuit includes: a control element , used to generate a first signal and a second signal, the first signal and the second signal respectively control the first transistor and the second transistor according to a feedback signal, wherein the feedback signal is related to the Output voltage; and a period counter for calculating the number of switching cycles in which the flyback converter operates in a non-discontinuous conduction mode based on the primary-side switching current; wherein when the control element determines based on the primary-side switching current When there has been a preset number of switching cycles operating in the non-discontinuous conduction mode, the control element actively forces at least one switching cycle to operate in a discontinuous conduction mode (DCM).

In a preferred embodiment, the switching control circuit further includes a demagnetization simulator, which is used to generate a demagnetization signal to simulate a demagnetization period of the transformer, and then during the non-discontinuous conduction The second transistor is controlled during the mode, wherein the degaussing simulator corrects the degaussing signal according to the degaussing period during the actively forced discontinuous conduction mode.

In a preferred embodiment, in the actively forced discontinuous conduction mode, the degaussing simulator detects a reflected voltage from an auxiliary winding of the transformer to correct the first transistor during the off period. A degaussing signal, in which the length of time the reflected voltage occurs is related to the degaussing period of the transformer.

In a preferred embodiment, the first transistor and the second transistor form a half-bridge circuit, wherein in the actively forced discontinuous conduction mode, after the first transistor is turned off, the second transistor The crystal is controlled off, causing the half-bridge circuit to operate in an asynchronous mode.

In a preferred embodiment, a pulse period of the demagnetization signal is used to control the minimum conduction time of the second transistor, thereby demagnetizing the transformer after the first transistor is turned off.

In a preferred embodiment, a pulse period of the degaussing signal is related to a level of an input voltage in an excitation period of the transformer, a level of the output voltage of the transformer and the first voltage. A conduction period of the crystal.

In a preferred embodiment, the switching control circuit further adjusts the resistance value of a resistor to correct the degaussing signal according to the degaussing period of the transformer during the active forced discontinuous conduction mode, wherein the resistor couple Connected to the flyback converter, a pulse period of the demagnetization signal is determined according to the resistance value of the resistor.

In a preferred embodiment, during the active forced discontinuous conduction mode, the control element controls the second transistor to turn off, so that after the first transistor turns off, a part of the demagnetization current of the transformer Flow through a body diode of the second transistor.

In a preferred embodiment, a first pulse of a second drive signal after the discontinuous conduction mode is used to turn on the second transistor to provide an excitation from a resonant capacitor of the flyback converter. Current flows to the transformer to excite the transformer, thereby generating a negative circulating current to achieve zero-voltage switching of the first transistor, wherein the resonant capacitor is coupled in series to the transformer.

In a preferred embodiment, the degaussing simulator includes: an integrating capacitor and a switch, wherein the switch is coupled to sample and hold. "Sample and hold is a special verb. You can explain it when you have time, or you can check it online first." Now, it is quite common to use a current sensing signal, wherein the current sensing signal is related to the primary side switch current; and a discharge circuit to generate a discharge current to discharge the integrating capacitor, wherein the discharge current is related to The output voltage; where the demagnetization signal root It is generated according to a voltage across the integrating capacitor, wherein the discharge current discharges the integrating capacitor from one end of the energized transformer.

From another point of view, the present invention also provides a control method for controlling a flyback converter, wherein the flyback converter includes: a first transistor for switching a transformer to generate a primary side switch current and an output voltage; and a second transistor for generating a circulating current to achieve zero voltage switching (ZVS) of the first transistor. The control method includes: generating a first signal and a second signal, the first signal and the second signal respectively control the first transistor and the second transistor according to a feedback signal, wherein the feedback signal is related to the output voltage; according to the primary side switch current to calculate the number of switching cycles in which the flyback converter operates in a non-discontinuous conduction mode; and when it is determined based on the primary side switch current that a predetermined number of switching cycles have been operated in the non-discontinuous conduction mode. , actively forcing at least one switching cycle to operate in a discontinuous conduction mode (DCM).

In a preferred embodiment, the control method further includes: generating a demagnetization signal to simulate a demagnetization period of the transformer, thereby controlling the second transistor during the non-discontinuous conduction mode; and according to the active forcing The degaussing signal is corrected during the degaussing period during the discontinuous conduction mode.

In a preferred embodiment, the step of correcting the degaussing signal includes detecting a reflected voltage from an auxiliary winding of the transformer in the active forced discontinuous conduction mode to turn off the first transistor. The degaussing signal is corrected during a period, wherein the length of time during which the reflected voltage occurs is related to the degaussing period of the transformer.

In a preferred embodiment, the first transistor and the second transistor form a half-bridge circuit, wherein the step of correcting the demagnetization signal further includes: in the actively forced discontinuous conduction mode In the formula, after the first transistor is turned off, the second transistor is controlled to be turned off, so that the half-bridge circuit operates in an asynchronous mode.

In a preferred embodiment, the step of generating the first signal and the second signal includes: controlling the minimum conduction time of the second transistor according to a pulse period of the degaussing signal, thereby controlling the first After the transistor is turned off, the transformer is demagnetized.

In a preferred embodiment, a pulse period of the degaussing signal is related to a level of an input voltage in an excitation period of the transformer, a level of the output voltage of the transformer and the first voltage. A conduction period of the crystal.

In a preferred embodiment, the step of correcting the degaussing signal includes: adjusting a resistance value of a resistor to correct the degaussing signal according to the degaussing period during the active forced discontinuous conduction mode of the transformer, The resistor is coupled to the flyback converter, and a pulse period of the demagnetization signal is determined according to the resistance value of the resistor.

In a preferred embodiment, the control method further includes: controlling the second transistor to turn off during the active forced discontinuous conduction mode, so that after the first transistor turns off, a demagnetization of the transformer Part of the current flows through a body diode of the second transistor.

In a preferred embodiment, the control method further includes: using a second driving signal to conduct the second transistor in the first pulse wave after the discontinuous conduction mode to convert the flyback converter from A resonant capacitor provides an excitation current to the transformer to excite the transformer, thereby generating a negative circulating current to achieve zero-voltage switching of the first transistor, wherein the resonant capacitor is coupled in series to the transformer.

It will be easier to understand the purpose, technical content, characteristics and achieved effects of the present invention through detailed description of specific embodiments below.

10: Transformer

100: Secondary side controller

20: Resonant capacitor

200: Primary side controller

201: Primary side controller

205: Clock generator

208: Primary side controller

22: timer

230: Capacitor

231: switch

240:Control components

243:Control components

248:Control components

25: timer

250:Degaussing simulator

255: Resistor

260: Period counter

271,272: Transistor

280: Comparator

285:Logic circuit

30:The first transistor

300: Resonant half-bridge flyback converter

35:Body diode

40:Second transistor

45:Body diode

51,52,55,60: Resistor

70: Secondary side synchronous rectifier

75:Body Diode

90: Optocoupler

900: Resonant half-bridge flyback converter

C: capacitance value

CPO: comparator output

DCM: discontinuous conduction mode

ID: discharge current

IM: exciting current

IP: primary side switching current

IS: secondary side switching current

kn: knee point

Lr: leakage inductance

LX: switch node

M1: the first transistor

M2: Second transistor

M3: The third transistor

n,m: turns ratio

NA: auxiliary winding

NC: positive integer

NNP: coupling node

NP: primary winding

NS: secondary winding

PW: pulse width

PZV: zero voltage switching pulse

Rs: resistance value

Rt: resistance value

S1: first driving signal

S2: second drive signal

S3: The third driving signal

Sdmg: degaussing signal

SG: drive signal

SH: drive signal

SL: drive signal

SMP: sampling signal

t1-t9: time point

t3’: time point

ta-te: time point

TA: first period

ta’,tc’: time point

TB: Second period

TC: The third period

Tcyc1: switching cycle

Tcyc2: switching cycle

Td1: first non-conduction period

Td2: The second non-conduction period

TDS: degaussing period

TDSX: conduction period

TDSX’: conduction period

TRH: time period

TRL: time period

TSL: conduction period

TW: Excitation period

Tx: Omit cycle

TZ: The third non-conduction period

VAUX: auxiliary signal

VC: voltage across

Vcr: cross voltage

VCS: current sensing signal

VCSp: voltage level

VDP: voltage drop

VFB: feedback signal

Vg: voltage level

VHB: switching node voltage

VIN: input voltage

Vinx: voltage level

VNA: auxiliary winding signal

VO: output voltage

VPK: voltage surge

Vref: reference voltage

Vth: voltage threshold

VX: reflected voltage

Figure 1 shows a prior art asymmetric duty cycle flyback converter.

Figure 2 shows a waveform diagram of a prior art half-bridge flyback converter operating in discontinuous conduction mode in a light load state.

FIG. 3 shows a schematic diagram of an embodiment of the resonant half-bridge flyback converter of the present invention.

FIG. 4 shows an operating waveform diagram corresponding to the embodiment of FIG. 3 .

FIG. 5 shows an operation waveform diagram for reducing the switching frequency of the driving signal SH and the driving signal SL.

FIG. 6 shows an operating waveform diagram of an embodiment of the resonant half-bridge flyback converter with omitted periods of the present invention.

FIG. 7 shows a block diagram of an embodiment of a primary-side controller in the resonant half-bridge flyback converter of the present invention.

FIG. 8 shows a block diagram of an embodiment of a primary-side controller in the resonant half-bridge flyback converter of the present invention.

FIG. 9 shows the operation waveform diagram of the degaussing signal generated by the degaussing simulator of the present invention.

FIG. 10 shows a schematic diagram of a specific embodiment of the degaussing simulator generating the degaussing signal Sdmg according to the present invention.

FIG. 11 shows a schematic diagram of a preferred embodiment of the resonant half-bridge flyback converter of the present invention.

FIG. 12 shows an operation waveform diagram of a preferred embodiment of the primary-side controller 201 operating in the discontinuous conduction mode of the present invention.

Figure 13 shows a block diagram of a preferred embodiment of the primary side controller of the present invention.

The diagrams in the present invention are schematic and are mainly intended to represent the coupling relationship between circuits and the relationship between signal waveforms. The circuits, signal waveforms and frequencies are not drawn to scale.

FIG. 3 shows a schematic diagram of an embodiment of the resonant half-bridge flyback converter of the present invention. The resonant half-bridge flyback converter 300 includes a first transistor 30 and a second transistor 40 to form a half-bridge circuit. The transformer 10 and the resonant capacitor 20 are connected in series with each other and coupled to the switching node LX of the half-bridge circuit. The transformer 10 includes a primary winding NP, a secondary winding NS and an auxiliary winding NA, wherein the primary winding NP and the secondary winding NS have The turns ratio n, the secondary winding NS and the auxiliary winding NA have a turns ratio m. The primary side controller 200 generates the driving signal SH and the driving signal SL, which switch the transformer 10 through the half-bridge circuit to generate the output voltage VO on the secondary side of the transformer 10 . The driving signal SH drives the first transistor 30 to excite the transformer 10 . The driving signal SL turns on the second transistor 40 during the demagnetization and resonance period of the transformer 10 . The driving signal SL is also used to turn on the second transistor 40 to generate a circulating current flowing through the transformer 10 to achieve the operation of the first transistor 30 . Zero voltage switching. The resistor 60 generates the current sensing signal VCS by detecting the primary side switching current IP of the transformer 10 .

The driving signal SH and the driving signal SL are generated according to the feedback signal VFB, wherein the feedback signal VFB is generated according to the output power of the resonant half-bridge flyback converter 300 . The secondary side controller 100 is coupled to the output voltage VO to generate a feedback signal VFB. The feedback signal VFB is coupled to the primary side controller 200 through the optical coupler 90 . The secondary side controller 100 is also used to generate a driving signal SG to drive the secondary side synchronous rectifier 70 during the demagnetization period TDS of the transformer 10 . The auxiliary winding NA generates the auxiliary winding signal VNA when the transformer 10 switches. The resistors 51 and 52 are used to attenuate the auxiliary winding signal VNA to generate the auxiliary signal VAUX. The auxiliary signal VAUX is coupled to the primary side controller 200 . In one embodiment, the resistor 55 is coupled to the primary side controller 200 , and parameters are set through the resistor 55 to generate the demagnetization signal Sdmg.

FIG. 4 shows an operating waveform diagram corresponding to the embodiment of FIG. 3 . When the driving signal SH is turned on, the transformer 10 is excited and generates an exciting current IM. When the driving signal SH is not turned on, the transformer 10 is demagnetized. During the demagnetization period TDS, the transformer 10 generates a secondary side switching current IS, and the driving signal SL is related to the demagnetization period TDS of the transformer 10 . In one embodiment, the conduction period TSL (ie, the pulse width) of the driving signal SL is equal to or longer than the demagnetization period TDS of the transformer 10, thereby preventing the transformer 10 from operating in a continuous conduction mode (CCM). During the demagnetization period TDS of the transformer 10 , a reflected voltage VX is generated on the resonant capacitor 20 , where the relationship between the reflected voltage VX and the output voltage VO is: VX=n*VO.

When the driving signal SH is not conductive, the driving signal SL may be turned on, and when the driving signal SL is not conductive, the driving signal SH may be turned on. The period between the driving signal SH and the driving signal SL (that is, when neither the driving signal SH nor the driving signal SL is conductive) may include a dead time (such as a period TRH, a period TRL).

Details of the operations in different time periods in Figure 4 are detailed in the following description.

The period from time point t1 to time point t2 is the excitation transformer period. During this period, the first transistor 30 is turned on and the second transistor 40 is turned off. The primary side switch current IP flowing through the transformer 10 increases and the voltage of the resonant capacitor 20 also increases. , at this time, the transformer 10 is excited and the resonant capacitor 20 is charged. The secondary side synchronous rectifier 70 is turned off and its body diode 75 has a reverse bias. Therefore, no energy is converted to the secondary side at this time.

The period from time point t2 to time point t3 is the first circulating current period. During this period, the first transistor 30 and the second transistor 40 are both turned off. The circulating current of the transformer 10 forces the switching node voltage VHB of the half-bridge circuit to decrease until the first circulating current period. Until the body diode 45 of the two transistors 40 is turned on. The period from time point t2 to time point t3 is related to the quasi-resonant period to achieve zero-voltage switching of the second transistor 40. At this time, the primary side voltage of the transformer 10 is the same as the voltage of the resonant capacitor 20 at time point t3.

The period from time point t3 to time point t4 is the resonance period (positive current). During this period, in the zero-voltage switching state, the first transistor 30 is turned off and the second transistor 40 is turned on. At this time, the output voltage VO is equal to the resonant capacitance. The cross voltage Vcr of 20 is divided by the turns ratio n, the current starts to flow through the secondary side synchronous rectifier 70, and the energy stored in the transformer 10 is converted to the output terminal to generate the output voltage VO. Since the leakage inductance Lr of the transformer 10 and the resonant capacitor 20 (Cr) form an inductance-capacitance tank (LC tank), the secondary side current is in the form of a sinusoidal wave during the period determined by the resonant frequencies Lr and Cr. The primary side current of the transformer 10 is the sum of the exciting current IM and the secondary side switching current IS. The current flowing through the resonant tank (Lr, Cr) is still a positive current, which is mainly driven by the magnetizing inductance of the transformer 10 and flows through the resonant capacitor 20 .

The period from time point t4 to time point t5 is the resonance period (negative current). During this period, the first transistor 30 continues to be turned off and the second transistor 40 continues to be turned on. Energy continues to be converted to the secondary side, but the resonant tank current is blocked by the resonant capacitor. The voltage of 20 is reversely driven, and the energy of the resonant capacitor 20 is not only converted to the secondary side, but also used to pull the excitation current level of the transformer 10 when the second transistor 40 continues to conduct (for example, from time t4 to time t5). to negative values.

The period from time point t5 to time point t6 is the reverse excitation transformer period (negative current). In this period from the end of the demagnetization period TDS of the transformer 10 to when the second transistor 40 is turned off, the resonant capacitor 20 reversely excites the transformer 10, and generate negative current.

The period from time point t6 to time point t7 is the second circulating current period. During this period, the first transistor 30 and the second transistor 40 are both turned off. The negative current of the transformer 10 is induced from time point t5 to time point t6 to force The switching node voltage VHB on the switching node LX in the half-bridge circuit increases until it turns on the body diode 35 of the first transistor 30 .

After time point t7, another period similar to the period from time point t1 to time point t2 begins. The first transistor 30 is turned on and the second transistor 40 is turned off in a zero-voltage switching state. If the circulating current in the resonant tank of the transformer is still negative current, the excess energy in the resonant tank will be sent back to the input terminal (the node supplying the input voltage VIN).

In the light load state, when the output power decreases, the pulse widths of the drive signal SH and the drive signal SL will decrease accordingly. Therefore, the switching frequency of the drive signal SH and the drive signal SL increases in the light load state. Due to the core loss ( Power losses such as core loss and switching loss increase, thus causing the power conversion efficiency of the power converter to deteriorate.

FIG. 5 shows an operation waveform diagram for reducing the switching frequency of the driving signal SH and the driving signal SL. One way to improve power efficiency is to reduce the switching frequency by extending the time between the turning off of the driving signal SL (for example, time point t3) and the turning on of the driving signal SH (for example, time point t5). However, the turning off of the driving signal SL will A circulating current is generated, which in turn causes a voltage surge VPK of the switching node voltage VHB and a voltage drop VDP of the auxiliary signal VAUX. The voltage surge VPK and voltage drop VDP will cause power loss and noise.

It should be noted that the aforementioned turning on or off of the driving signal SH and the driving signal SL respectively corresponds to the turning on or off of the first transistor 30 and the second transistor 40 .

FIG. 6 shows an operating waveform diagram of an embodiment of the resonant half-bridge flyback converter with omitted periods of the present invention.

Referring to FIG. 6 , in one embodiment, the driving signal SH is turned on during the excitation period of the excitation transformer 10 (for example, from time point t1 to time point t2 ) to excite the transformer 10 . After the driving signal SH is turned off, the driving signal SL is turned on during the resonance period (for example, from time point t2 to time point t3) and has a resonance pulse wave (for example, from time point t2 to time point t3). An excitation period and a resonance period form a switching period ( For example, time point t1 to time point t3).

As shown in FIG. 6 , in one embodiment, the omitting period Tx starts from the non-conducting time point when the driving signal SH turns non-conductive (for example, time point t4), and when the omitting period Tx ends (for example, time point t6), the driving signal SH SL turns conductive. In one embodiment, when the output power is reduced due to power saving, the omission period Tx will be correspondingly increased (ie, the switching frequency is reduced).

Please continue to refer to Figure 6. Compared with the period without omitted periods, such as time point t1 to time point t3, the driving signal SL is not conductive and has no resonant pulse in the omitted period (such as Tx). For example, in the prior art, A pulse wave of the driving signal SL from time point t4 to time point t5, that is, the resonance pulse wave of the driving signal SL, has been omitted in this embodiment, as shown in Figure 6. Therefore, in the omitted period (time point t4 to time point t5), At time t6), no negative circulating current is generated. In the prior art, the voltage surge VPK generated by the switching node voltage VHB and the voltage drop VDP generated by the auxiliary signal VAUX have also been avoided in this embodiment. In one embodiment, as shown in FIG. 6 , the driving signal SH is also in a non-conducting state during the omitted period (eg, Tx).

In one embodiment, after the driving signal SH is turned off, part of the demagnetization current of the transformer 10 flows through the second part of the omitted cycle (for example, part of the time between time t4 and time t5). Body diode 45 of transistor 40 . In other words, in one embodiment, there are no double pulses in the driving signal SL. In one embodiment, there is no double pulse wave in the driving signal SH. From one point of view, after a single pulse wave of the driving signal SH, a single pulse wave of the driving signal SL is then generated, and after a single pulse wave of the driving signal SL, a single pulse wave of the driving signal SH is generated, even if the resonant half bridge The same goes for flyback converters operating with omitted periods. From another point of view, between two consecutive pulse waves of the driving signal SH, the driving signal SL includes at most one pulse wave. Between two consecutive pulse waves of the driving signal SL, the driving signal SH includes at most one pulse wave. .

In one embodiment, when the output power is lower than the preset threshold, the omitted period Tx is generated. In one embodiment, the omission period Tx increases correspondingly as the output power decreases. In one embodiment, even when the driving signal SL cannot achieve zero-voltage switching of the first transistor 30, the second driving signal does not include a second pulse wave between two consecutive pulse waves of the first driving signal. , so the second pulse wave is not used to achieve zero-voltage switching of the first transistor 30 .

Please continue to refer to FIG. 6 . In one embodiment, the zero-voltage switching pulse wave (for example, PZV) of the driving signal SL turns on the second transistor 40 after the omitted period passes, so as to achieve a zero-voltage switching period (for example, from time point t6 to time point t7 ).

As shown in FIG. 6 , in one embodiment, after the zero-voltage switching pulse wave PZV after the omitted period, at least one switching period (for example, time point t7 to time point t9 ) is generated.

Please continue to refer to FIG. 6 . In one embodiment, the resonance period may include a continuous zero-voltage switching period (for example, from time point t3′ to time point t3 ). The continuous zero-voltage switching period is used to achieve zero-voltage switching of the first transistor 30 . In other words, in this embodiment, the first part of the resonant pulse wave (for example, from time point t2 to time point t3') is used to achieve resonance between the transformer 10 and the resonant capacitor 20, and the second part of the resonant pulse wave (for example, from time point t3' to time point t3 ) is used to generate circulating current to achieve zero-voltage switching of the first transistor 30 .

FIG. 7 shows a block diagram of an embodiment of a primary-side controller in the resonant half-bridge flyback converter of the present invention. In one embodiment, the primary side controller 200 includes a timer 25 and a control component 240 . In one embodiment, the control element 240 is used to generate the driving signal SH and the driving signal SL according to the input voltage VIN (via the auxiliary signal VAUX) and the feedback signal VFB, and the timer 25 is used to generate the aforementioned omitted period Tx.

As shown in FIG. 7 , in one embodiment, the timer 25 determines whether the output power is lower than a preset threshold based on information related to the output power. When the output power is determined to be lower than the preset threshold, the timer 25 starts The omission period Tx is calculated, and the control element 240 is controlled to omit the pulse waves of the driving signal SH and the driving signal SL in the omission period Tx.

Please refer to Figure 4 again. When the resonant half-bridge flyback converter is in a medium load or light load state, the resonance period from time point t4 to time point t5 is short and cannot generate enough negative current (energy) to achieve zero-voltage switching. , therefore, the main part of the negative current is the current generated from time point t5 to time point t6.

However, a higher negative current will lead to a higher power loss. In order to control the negative current to achieve zero-voltage switching at an appropriate level, the demagnetization period must be accurately controlled. Therefore, a demagnetization signal Sdmg corresponding to the transformer 10 needs to be generated. Degaussing period TDS.

FIG. 8 shows a block diagram of an embodiment of a primary-side controller in the resonant half-bridge flyback converter of the present invention. In one embodiment, the primary side controller 208 includes a demagnetization simulator 250 and a control component 248 . In one embodiment, the control element 248 is used to generate the driving signal SH and the driving signal SL according to the input voltage VIN (via the auxiliary signal VAUX) and the feedback signal VFB, and the demagnetization simulator 250 is used to generate the driving signal SH according to the demagnetization related signal. The demagnetization signal Sdmg is generated to simulate the demagnetization period TDS, where the demagnetization related signal is such as the reflected voltage of the transformer 10 (via the auxiliary signal VAUX).

Please also refer to FIG. 9 , which shows an operation waveform diagram of the degaussing simulator of the present invention generating a degaussing signal.

During the switching cycle, the resonant half-bridge flyback converter periodically operates in the non-discontinuous conduction mode (for example, from time point ta to time point tc'), and the driving signal SH first turns on the first transistor 30 to excite the transformer 10 to generate The primary side switching current IP (for example, time point ta' to time point tb), after the first transistor 30 is turned off, the driving signal SL is used to conduct during the resonance period (time point tb to time point tc) (for example, time point tb to time point tc' ) the second transistor 40, and is used to generate a circulating current (for example, from time point tc to time point tc′) to achieve zero-voltage switching of the first transistor 30. In the switching cycle of the non-discontinuous conduction mode, the conduction period TSL of the driving signal SL (for example, time point tb to time point tc') is determined by the pulse width (for example, TDSX') of the demagnetization signal Sdmg, in which the demagnetization signal Sdmg is determined by the demagnetization signal Sdmg. The magnetic simulator 250 is generated based on the correction in the previously forced-in discontinuous conduction mode. In one embodiment, the conduction period TDSX′ of the demagnetization signal Sdmg is corrected in the previous actively forced discontinuous conduction mode period, and is used to cause the control element 248 to control the minimum conduction time of the second transistor 40, thereby in The transformer 10 is demagnetized during a non-discontinuous conduction mode after the transistor 30 is turned off. In one embodiment, as shown in FIG. 9 , the conduction period TSL of the driving signal SL (for example, time point tb to time point tc’) may be the conduction period of the demagnetization signal Sdmg. A delay time (such as time point tc to time point tc') is added to the pass-on period TDSX' to establish a negative circulating current on the primary side switch current IP after the demagnetization period to achieve zero-voltage switching of the first transistor 30 .

It should be noted that non-discontinuous conduction mode refers to an operation mode that is not a discontinuous conduction mode, such as continuous conduction mode (CCM) or quasi-resonant mode (QRM) operation. The resonant mode is also called the boundary conduction mode (BCM).

In one embodiment, when the primary side switch current IP has a preset number (for example, a positive integer NC) of switching periods (for example, time point ta to time point t1), the operation is in a non-discontinuous conduction mode (for example, quasi-resonant mode). , at least one switching cycle is actively forced to operate in the discontinuous conduction mode (for example, from time point t1 to time point t3). Therefore, the degaussing simulator 250 is used in the forced insertion discontinuous conduction mode to correct the conduction period TDSX of the degaussing signal Sdmg according to the degaussing period TDS of the transformer 10 .

As shown in FIG. 9 , in the forced insertion discontinuous conduction mode, the degaussing period TDS of the transformer 10 starts from the rising edge of the auxiliary signal VAUX, and ends between the auxiliary signal VAUX (for example, from time t2 to time t3). The falling edge (i.e. knee point kn) ends. Specifically, in this embodiment, the reflected voltage can be detected by sensing the auxiliary signal VAUX, which comes from the auxiliary winding NA of the transformer 10 during the off period of the first transistor 30 . The length of time during which the reflected voltage appears, that is, the pulse width of the auxiliary signal VAUX from the rising edge to the knee point kn, is related to the demagnetization period TDS of the transformer 10 .

In one embodiment, the primary side controller 208 further includes a cycle counter 260. The cycle counter 260 is used to calculate the number of switching cycles operating in the non-discontinuous conduction mode according to the primary side switch current IP, and when the primary side switch current IP is When it is determined that the preset number of switching cycles have not been operated in the discontinuous conduction mode, the period counter 260 is used to control the control element 248 to actively force the operation in the discontinuous conduction mode. In one embodiment, the period counter 260 can sense the primary-side switch current IP through the current sensing signal VCS, thereby determining that the operation is in the non-discontinuous conduction mode.

In one embodiment, as shown in FIG. 9 , during the forced discontinuous conduction mode switching period, the driving signal SL continuously controls the second transistor 40 to be non-conductive, so that the half-bridge circuit not only operates in the discontinuous conduction mode, but also operates in the discontinuous conduction mode. In the asynchronous switching mode, in which during the forced discontinuous conduction mode switching period, a portion of the demagnetization current of the transformer 10 (eg, IP from time point t2 to time point t2 ′) flows through the body diode 45 of the second transistor 40 .

Please continue to refer to FIG. 9. After the discontinuous conduction mode DCM (for example, from time point t4 to time point t5), the first pulse wave of the driving signal SL conducts the second transistor 40 to excite the transformer 10 from the resonant capacitor 20 to the transformer 10, and then A negative circulating current (IP from time point t4 to time point t5) is generated to achieve zero-voltage switching of the first transistor 30 .

FIG. 10 shows a schematic diagram of a specific embodiment of a demagnetization simulator for generating a demagnetization signal Sdmg according to the present invention. In one embodiment, the degaussing simulator 250 includes a timing generator 205, a comparator 280, and a logic circuit 285.

In one embodiment, the timing generator 205 includes an integrator, which is composed of a switch 231 and a capacitor 230. The switch 231 is controlled by the sampling signal SMP, and the sampling signal SMP is related to the driving signal SH to sample the current sensing signal VCS. The discharge current ID is related to n*VO and is used to discharge the cross-voltage VC of the capacitor 230 . The cross voltage VC is compared with the reference voltage Vref by the comparator 280 . The logic circuit 285 generates the degaussing signal Sdmg according to the comparator output CPO and the sampling signal SMP related to the driving signal SH. In one embodiment, the reference voltage Vref is 0 volts, and when the primary side switch current IP is 0, the current sensing voltage VCS is 0.

In one embodiment, the time length of the demagnetization signal Sdmg is related to the voltage level (Vinx) of the input voltage of the transformer 10 , that is, as shown in FIG. 3 , the coupling node NNP of the primary winding NP and the resonant capacitor 20 The voltage and the time length of the demagnetization signal Sdmg are also related to the voltage level of the output voltage of the transformer 10 (for example, n*VO) and the excitation of the transformer 10 when the first transistor 30 is turned on. Time period (TW). It should be noted that the voltage level Vinx of the input voltage of the transformer 10 is equal to the input voltage VIN minus the voltage Vcr across the resonant capacitor 20 .

According to the fact that the magnetic flux demagnetized by the transformer 10 is equal to the magnetic flux excited by the transformer 10, the following equation 1 can be listed: Vinx*TW=n*VO*TDS (Equation 1)

TW is the occurrence time of the voltage level Vinx of the input voltage of the transformer 10 during the excitation period of the transformer 10 ; n*VO is the voltage of the transformer 10 during the demagnetization period TDS of the transformer 10 . n is the turns ratio of the primary winding NP and the secondary winding NS, and VO is the voltage of the secondary winding NS (ie, the output voltage).

After the transformer 10 is excited, the level VCSp of the current sensing signal VCS is related to the peak value of the primary-side switching current IP at the end of the exciting process, and is generated on the resistor 60 shown in Figure 3, which can be expressed by the following equation 2: VCSp =(Vinx/L)*TW*Rs (Formula 2)

Wherein, L is the inductance of the primary winding NP of the transformer 10, Rs is the resistance value of the resistor 60, and VCSp is the voltage level of the transformer 10 at the end of the excitation process.

Assume ID=n*VO/Rt, where Rt is the resistance value of resistor 55.

The pulse width TDSX of the demagnetization signal Sdmg can be expressed as: TDSX=(C*VCSp)/ID, where C is the capacitance value of the capacitor 230.

TDSX=(Rt*C*VCSp)/(n*VO)

TDSX=(Rt*C/(n*VO))*(Rs/L)*Vinx*TW

Let Rt=L/(Rs*C) (Formula 3)

TDSX=(Vinx*TW)/(n*VO) (Formula 4)

When the condition of Equation 3 is satisfied, the conduction period TDSX of the demagnetization signal Sdmg shown in Equation 4 is equal to the demagnetization period TDS of the transformer 10 .

Please continue to refer to Figure 10. The switch 231 is turned on to sample the current sensing signal VCS to the capacitor 230, and when the switch 231 is turned off (that is, when the excitation ends), the level VCSp of the current sensing signal VCS is maintained at the capacitor 230. The switch 231 is controlled by the sampling signal SMP. When the switch 231 is turned off, the demagnetization signal Sdmg is enabled (for example, through the logic circuit 285). In other words, when the demagnetization signal Sdmg starts to be enabled, the voltage VC across the capacitor 230 is the peak value of the current sensing signal VCS. After the switch 231 is turned off, the discharge current ID begins to discharge the capacitor 230. When the capacitor 230 is completely discharged through the discharge current ID (ID=n*VO/Rt) (VC=0V), the demagnetization signal Sdmg is disabled. The resistor 55 shown in FIG. 10 and FIG. 3 is used to set the preset pulse width of the demagnetization signal Sdmg.

In one embodiment, in the discontinuous conduction mode switching period of forced insertion, the pulse width TDSX of the degaussing signal Sdmg can be compared with the degaussing period TDS represented by the pulse width of the auxiliary signal VAUX by the degaussing simulator 250 Comparison is made, so the pulse width TDSX of the demagnetization signal Sdmg can be corrected and used for the next non-discontinuous conduction mode switching cycle. In one embodiment, the degaussing simulator 250 is further used to adjust the resistance value of the resistor 255 according to the degaussing period TDS detected in the discontinuous conduction mode to correct the pulse period TDSX of the degaussing signal Sdmg.

In other embodiments, in addition to adjusting the resistance value of the resistor 255, the degaussing simulator 250 can also correct the pulse period TDSX of the degaussing signal Sdmg in the following manner: adjusting the voltage threshold Vth to determine the end of the degaussing signal Sdmg. , or adjust the capacitance value of the capacitor 230, or adjust, for example, the ratio of the current mirror composed of the transistor 271 and the transistor 272 in FIG. 10 .

FIG. 11 shows a schematic diagram of a preferred embodiment of the resonant half-bridge flyback converter of the present invention. The resonant half-bridge flyback converter 900 is similar to the resonant half-bridge flyback converter 300 of FIG. 3 . In this embodiment, the resonant half-bridge flyback converter 900 includes a first transistor M1, a second transistor M2, and a third transistor M3. The first transistor M1, the second transistor M2, and the third transistor are M3 is used to form a half-bridge circuit. From one point of view, the first transistor M1 and the third transistor M3 are matched The second transistor M2 is configured as a lower bridge transistor of the resonant half-bridge flyback converter 900 , and the second transistor M2 is configured as an upper bridge transistor of the resonant half-bridge flyback converter 900 .

According to the feedback signal VFB and the input voltage VIN, the primary side controller 201 is used to generate the first driving signal S1, the second driving signal S2 and the third driving signal S3. The first driving signal S1, the second driving signal S2 and the third driving signal S3 are generated. The drive signal S3 is coupled to switch the transformer 10 through the half-bridge circuit, thereby generating an output voltage VO on the secondary side of the transformer 10 . The second driving signal S2 drives the second transistor M2 to excite the transformer 10. The third driving signal S3 turns on the third transistor M3 during the demagnetization and resonance period of the transformer 10. The third driving signal S3 is also used to turn on the third transistor M3. The transistor M3 generates a circulating current flowing through the transformer 10 and achieves zero-voltage switching of the second transistor M2 in a heavy load state. In other words, the second transistor M2 is a primary-side high-bridge switch of the resonant half-bridge flyback converter 900 and can correspond to the first transistor 30 of FIG. 3 , and the third transistor M3 is a resonant half-bridge flyback converter. The primary side lower bridge switch 900 may correspond to the second transistor 40 of FIG. 3 . From one point of view, the first transistor M1 is connected in parallel to the third transistor M3 and serves as an auxiliary primary-side low-bridge switch and has an independent control signal S1.

In one embodiment, after the transformer 10 is excited by turning on the second transistor M2 at a light load state and operating in the discontinuous conduction mode, the third transistor M3 is controlled during the demagnetization and resonance period of the transformer 10 for conduction. After demagnetization, when the third transistor M3 continues to be turned off, the first driving signal S1 is used to turn on the first transistor M1 to generate a circulating current flowing through the transformer 10 to achieve zero-voltage switching of the second transistor M2 ,. Therefore, the third transistor M3 can avoid switching twice in one switching period of the discontinuous conduction mode.

Since the first transistor M1 is only used to generate circulating current to achieve zero voltage switching, in one embodiment, the actual size (such as aspect ratio) of the first transistor M1 can be configured to be much smaller than the actual size of the third transistor M3 size. Therefore, the driving capability of the first transistor M1 and the parasitic capacitance (eg For example, the gate capacitance) is lower than the parasitic capacitance of the third transistor M3, so the switching loss of the first transistor M1 is also lower than the switching loss of the third transistor M3.

For example, the gate switching loss Pg of the transistor can be expressed as: Pg=0.5*Ciss*Vg*Vg*Freq

Where Ciss is the input capacitance of the transistor, Vg is the voltage level of the gate drive signal, and Freq is the switching frequency of the gate drive signal.

As shown in the above switching power loss equation, the first transistor M1 with a smaller actual size is used exclusively to achieve zero-voltage switching of the second transistor M2 in the discontinuous conduction mode. Therefore, the gate switching loss of the first transistor M1 is lower than The third transistor M3 is actually larger in size.

In addition, in one embodiment, the amplitude of the voltage level (ie, Vg) of the first driving signal S1 is lower than the amplitude of the voltage level of the third driving signal S3, so the switching loss of the first transistor M1 can be further reduced. And in one embodiment, the gate maximum rating (eg gate-source voltage) of the first transistor M1 may also be lower than the gate maximum rating of the third transistor M3.

The resistor 60 generates the current sensing signal VCS by detecting the primary-side switching current IP of the transformer 10. The primary-side controller 201 is used to generate the first driving signal S1 according to the input voltage VIN, and to output the first driving signal S1 according to the input voltage VIN. voltage VO to generate the third driving signal S3. The primary side controller 201 is further used to generate the second driving signal S2 according to the feedback signal VFB.

FIG. 12 shows an operation waveform diagram of a preferred embodiment of the primary-side controller 201 operating in the discontinuous conduction mode of the present invention. In the discontinuous conduction mode operation, the primary side controller 201 operates in the first switching period Tcyc1 and controls the first driving signal S1 to conduct the first transistor M1 in the first period TA, thereby generating a circulating current to achieve the second Zero voltage switching when transistor M2 is on. After the first period TA, the first driving signal S1, the second driving signal S2 and the third driving signal S3 are used to turn off the first transistor M1 and the first non-conducting period Td1 (ie, the dead period). The second transistor M2 and the third transistor M3. In one embodiment, the first non-conduction period Td1 is related to To achieve a quasi-resonant period of zero-voltage switching of the second transistor M2. After the first non-conduction period Td1, the second driving signal S2 turns on the second transistor M2 in the second period TB, and the conduction of the second transistor M2 is used to excite the transformer 10. After the second period TB, the first driving signal S1, the second driving signal S2 and the third driving signal S3 are used to turn off the first transistor M1, the second transistor M2 and the third transistor M1 during the dead period. Tritransistor M3. In one embodiment, the second non-conduction period Td2 is related to another quasi-resonant period for achieving zero-voltage switching of the third transistor M3. After the second non-conduction period Td2, the third driving signal S3 turns on the third transistor M3 during the third period TC, and the third transistor M3 turns on during the demagnetization period of the transformer 10. After the third period TC, the first driving signal S1, the second driving signal S2 and the third driving signal S3 are used to turn off the first transistor M1, the second transistor M2 and the third non-conduction period TZ. In the three-transistor M3, the exciting current IM is maintained at zero in the third non-conduction period TZ (ie, the discontinuous conduction mode). After the third non-conduction period TZ, another switching period Tcyc2 starts.

Figure 13 shows a block diagram of a preferred embodiment of the primary side controller of the present invention. In one embodiment, the primary side controller 213 includes a timer 22 and a control component 243 . The control element 243 is used to generate the first driving signal S1, the second driving signal S2 and the third driving signal S3 according to the input voltage VIN (via VAUX) and the feedback signal VFB.

The timer 22 is used for timing to generate the third non-conduction period TZ. The third non-conduction period TZ starts from the end of the pulse wave of the third driving signal S3 (for example, the falling edge). In one embodiment, when the output power of the resonant half-bridge flyback converter decreases, the third non-conduction period TZ increases correspondingly. Therefore, the switching frequency of the resonant half-bridge flyback converter can also be affected by the resonant half-bridge flyback converter. The output power of the Chi converter is reduced accordingly, thereby improving performance in light load operating conditions.

FIG. 3 shows a schematic diagram of an embodiment of the resonant half-bridge flyback converter of the present invention. The resonant half-bridge flyback converter 300 includes: a first transistor 30 and a second transistor 40 for Constitute a half-bridge circuit. The transformer 10 and the resonant capacitor 20 are connected in series with each other and coupled to the switching node LX of the half-bridge circuit. The transformer 10 includes a primary winding NP, a secondary winding NS and an auxiliary winding NA, wherein the primary winding NP and the secondary winding NS have The turns ratio n, the secondary winding NS and the auxiliary winding NA have a turns ratio m. The primary side controller 200 generates the driving signal SH and the driving signal SL, which switch the transformer 10 through the half-bridge circuit to generate the output voltage VO on the secondary side of the transformer 10 . The driving signal SH drives the first transistor 30 to excite the transformer 10 . The driving signal SL turns on the second transistor 40 during the demagnetization and resonance period of the transformer 10 . The driving signal SL is also used to turn on the second transistor 40 to generate a circulating current flowing through the transformer 10 to achieve the operation of the first transistor 30 . Zero voltage switching. The resistor 60 generates the current sensing signal VCS by detecting the primary side switching current IP of the transformer 10 .

The driving signal SH and the driving signal SL are generated according to the feedback signal VFB, wherein the feedback signal VFB is generated according to the output power of the resonant half-bridge flyback converter 300 . The secondary side controller 100 is coupled to the output voltage VO to generate a feedback signal VFB. The feedback signal VFB is coupled to the primary side controller 200 through the optical coupler 90 . The secondary side controller 100 is also used to generate a driving signal SG to drive the secondary side synchronous rectifier 70 during the demagnetization period TDS of the transformer 10 . The auxiliary winding NA generates the auxiliary winding signal VNA when the transformer 10 switches. The resistors 51 and 52 are used to attenuate the auxiliary winding signal VNA to generate the auxiliary signal VAUX. The auxiliary signal VAUX is coupled to the primary side controller 200 . In one embodiment, the resistor 55 is coupled to the primary side controller 200 , and parameters are set through the resistor 55 to generate the demagnetization signal Sdmg.

FIG. 4 shows an operating waveform diagram corresponding to the embodiment of FIG. 3 . When the driving signal SH is turned on, the transformer 10 is excited and generates an exciting current IM. When the driving signal SH is not turned on, the transformer 10 is demagnetized. During the demagnetization period TDS, the transformer 10 generates a secondary side switching current IS, and the driving signal SL is related to the demagnetization period TDS of the transformer 10 . In one embodiment, the conduction period TSL (ie, the pulse width) of the driving signal SL is equal to or longer than the demagnetization period TDS of the transformer 10, thereby preventing the transformer from 10Operates in continuous conduction mode (CCM). During the demagnetization period TDS of the transformer 10 , a reflected voltage VX is generated on the resonant capacitor 20 , where the relationship between the reflected voltage VX and the output voltage VO is: VX=n*VO.

When the driving signal SH is not conductive, the driving signal SL may be turned on, and when the driving signal SL is not conductive, the driving signal SH may be turned on. The period between the driving signal SH and the driving signal SL (that is, when neither the driving signal SH nor the driving signal SL is conductive) may include a dead time (such as a period TRH, a period TRL).

Details of the operations in different time periods in Figure 4 are detailed in the following description.

The period from time point t1 to time point t2 is the excitation transformer period. During this period, the first transistor 30 is turned on and the second transistor 40 is turned off. The primary side switch current IP flowing through the transformer 10 increases and the voltage of the resonant capacitor 20 also increases. , at this time, the transformer 10 is excited and the resonant capacitor 20 is charged. The secondary side synchronous rectifier 70 is turned off and its body diode 75 has a reverse bias. Therefore, no energy is converted to the secondary side at this time.

The period from time point t2 to time point t3 is the first circulating current period. During this period, the first transistor 30 and the second transistor 40 are both turned off. The circulating current of the transformer 10 forces the switching node voltage VHB of the half-bridge circuit to decrease until the first circulating current period. Until the body diode 45 of the two transistors 40 is turned on. The period from time point t2 to time point t3 is related to the quasi-resonant period to achieve zero-voltage switching of the second transistor 40. At this time, the primary side voltage of the transformer 10 is the same as the voltage of the resonant capacitor 20 at time point t3.

The period from time point t3 to time point t4 is the resonance period (positive current). During this period, in the zero-voltage switching state, the first transistor 30 is turned off and the second transistor 40 is turned on. At this time, the output voltage VO is equal to the resonant capacitance. The cross voltage Vcr of 20 is divided by the turns ratio n, the current starts to flow through the secondary side synchronous rectifier 70, and the energy stored in the transformer 10 is converted to the output terminal to generate the output voltage VO. Since the leakage inductance Lr of the transformer 10 and the resonant capacitance 20 (Cr) form an inductance-capacitance tank (LC tank), the secondary side The current is in the form of a sine wave during the period determined by the resonant frequencies Lr and Cr. The primary side current of the transformer 10 is the sum of the exciting current IM and the secondary side switching current IS. The current flowing through the resonant tank (Lr, Cr) is still a positive current, which is mainly driven by the magnetizing inductance of the transformer 10 and flows through the resonant capacitor 20 .

The period from time point t4 to time point t5 is the resonance period (negative current). During this period, the first transistor 30 continues to be turned off and the second transistor 40 continues to be turned on. Energy continues to be converted to the secondary side, but the resonant tank current is blocked by the resonant capacitor. The voltage of 20 is reversely driven, and the energy of the resonant capacitor 20 is not only converted to the secondary side, but also used to pull the excitation current level of the transformer 10 when the second transistor 40 continues to conduct (for example, from time t4 to time t5). to negative values.

The period from time point t5 to time point t6 is the reverse excitation transformer period (negative current). In this period from the end of the demagnetization period TDS of the transformer 10 to when the second transistor 40 is turned off, the resonant capacitor 20 reversely excites the transformer 10, and generate negative current.

The period from time point t6 to time point t7 is the second circulating current period. During this period, the first transistor 30 and the second transistor 40 are both turned off. The negative current of the transformer 10 is induced from time point t5 to time point t6 to force The switching node voltage VHB on the switching node LX in the half-bridge circuit increases until it turns on the body diode 35 of the first transistor 30 .

After time point t7, another period similar to the period from time point t1 to time point t2 begins. The first transistor 30 is turned on and the second transistor 40 is turned off in a zero-voltage switching state. If the circulating current in the resonant tank of the transformer is still negative current, the excess energy in the resonant tank will be sent back to the input terminal (the node supplying the input voltage VIN).

In the light load state, when the output power decreases, the pulse widths of the drive signal SH and the drive signal SL will decrease accordingly. Therefore, the switching frequency of the drive signal SH and the drive signal SL increases in the light load state. Due to the core loss ( Power losses such as core loss and switching loss increase, thus causing the power conversion efficiency of the power converter to deteriorate.

FIG. 5 shows an operation waveform diagram for reducing the switching frequency of the driving signal SH and the driving signal SL. One way to improve power efficiency is to reduce the switching frequency by extending the time between the turning off of the driving signal SL (for example, time point t3) and the turning on of the driving signal SH (for example, time point t5). However, the turning off of the driving signal SL will A circulating current is generated, which in turn causes a voltage surge VPK of the switching node voltage VHB and a voltage drop VDP of the auxiliary signal VAUX. The voltage surge VPK and voltage drop VDP will cause power loss and noise.

It should be noted that the aforementioned turning on or off of the driving signal SH and the driving signal SL respectively corresponds to the turning on or off of the first transistor 30 and the second transistor 40 .

FIG. 6 shows an operating waveform diagram of an embodiment of the resonant half-bridge flyback converter with omitted periods of the present invention.

Referring to FIG. 6 , in one embodiment, the driving signal SH is turned on during the excitation period of the excitation transformer 10 (for example, from time point t1 to time point t2 ) to excite the transformer 10 . After the driving signal SH is turned off, the driving signal SL is turned on during the resonance period (for example, from time point t2 to time point t3) and has a resonance pulse wave (for example, from time point t2 to time point t3). An excitation period and a resonance period form a switching period ( For example, time point t1 to time point t3).

As shown in FIG. 6 , in one embodiment, the omitting period Tx starts from the non-conducting time point when the driving signal SH turns non-conductive (for example, time point t4), and when the omitting period Tx ends (for example, time point t6), the driving signal SH SL turns conductive. In one embodiment, when the output power is reduced due to power saving, the omission period Tx will be correspondingly increased (ie, the switching frequency is reduced).

Please continue to refer to Figure 6. Compared with the period without omitted periods, such as time point t1 to time point t3, the driving signal SL is not conductive and has no resonant pulse in the omitted period (such as Tx). For example, in the prior art, A pulse wave of the driving signal SL from time point t4 to time point t5, that is, the resonance pulse wave of the driving signal SL, has been omitted in this embodiment, as shown in Figure 6. Therefore, in the omitted period (time point t4 to time point t5), At time t6), no negative circulating current is generated. In the prior art, the switching node voltage VHB The voltage surge VPK generated and the voltage drop VDP generated by the auxiliary signal VAUX have also been avoided in this embodiment. In one embodiment, as shown in FIG. 6 , the driving signal SH is also in a non-conducting state during the omitted period (eg, Tx).

In one embodiment, after the driving signal SH is turned off, part of the demagnetization current of the transformer 10 flows through the second part of the omitted cycle (for example, part of the time between time t4 and time t5). Body diode 45 of transistor 40 . In other words, in one embodiment, there are no double pulses in the driving signal SL. In one embodiment, there is no double pulse wave in the driving signal SH. From one point of view, after a single pulse wave of the driving signal SH, a single pulse wave of the driving signal SL is then generated, and after a single pulse wave of the driving signal SL, a single pulse wave of the driving signal SH is generated, even if the resonant half bridge The same goes for flyback converters operating with omitted periods. From another point of view, between two consecutive pulse waves of the driving signal SH, the driving signal SL includes at most one pulse wave. Between two consecutive pulse waves of the driving signal SL, the driving signal SH includes at most one pulse wave. .

In one embodiment, when the output power is lower than the preset threshold, the omitted period Tx is generated. In one embodiment, the omission period Tx increases correspondingly as the output power decreases. In one embodiment, even when the driving signal SL cannot achieve zero-voltage switching of the first transistor 30, the second driving signal does not include a second pulse wave between two consecutive pulse waves of the first driving signal. , so the second pulse wave is not used to achieve zero-voltage switching of the first transistor 30 .

Please continue to refer to FIG. 6 . In one embodiment, the zero-voltage switching pulse wave (for example, PZV) of the driving signal SL turns on the second transistor 40 after the omitted period passes, so as to achieve a zero-voltage switching period (for example, from time point t6 to time point t7 ).

As shown in FIG. 6 , in one embodiment, after the zero-voltage switching pulse wave PZV after the omitted period, at least one switching period (for example, time point t7 to time point t9 ) is generated.

Please continue to refer to FIG. 6. In one embodiment, the resonance period may include a continuation of a zero-voltage switching period (for example, from time point t3' to time point t3). The continuation of the zero-voltage switching period is used to achieve the first transistor Zero voltage switching of body 30. In other words, in this embodiment, the first part of the resonant pulse wave (for example, from time point t2 to time point t3') is used to achieve resonance between the transformer 10 and the resonant capacitor 20, and the second part of the resonant pulse wave (for example, from time point t3' to time point t3 ) is used to generate circulating current to achieve zero-voltage switching of the first transistor 30 .

FIG. 7 shows a block diagram of an embodiment of a primary-side controller in the resonant half-bridge flyback converter of the present invention. In one embodiment, the primary side controller 200 includes a timer 25 and a control component 240 . In one embodiment, the control element 240 is used to generate the driving signal SH and the driving signal SL according to the input voltage VIN (via the auxiliary signal VAUX) and the feedback signal VFB, and the timer 25 is used to generate the aforementioned omitted period Tx.

As shown in FIG. 7 , in one embodiment, the timer 25 determines whether the output power is lower than a preset threshold based on information related to the output power. When the output power is determined to be lower than the preset threshold, the timer 25 starts The omission period Tx is calculated, and the control element 240 is controlled to omit the pulse waves of the driving signal SH and the driving signal SL in the omission period Tx.

Please refer to Figure 4 again. When the resonant half-bridge flyback converter is in a medium load or light load state, the resonance period from time point t4 to time point t5 is short and cannot generate enough negative current (energy) to achieve zero-voltage switching. , therefore, the main part of the negative current is the current generated from time point t5 to time point t6.

However, a higher negative current will lead to a higher power loss. In order to control the negative current to achieve zero-voltage switching at an appropriate level, the demagnetization period must be accurately controlled. Therefore, a demagnetization signal Sdmg corresponding to the transformer 10 needs to be generated. Degaussing period TDS.

FIG. 8 shows a block diagram of an embodiment of a primary-side controller in the resonant half-bridge flyback converter of the present invention. In one embodiment, the primary side controller 208 includes a demagnetization simulator 250 and a control component 248 . In one embodiment, the control element 248 is used to generate the driving signal SH and the driving signal SL according to the input voltage VIN (via the auxiliary signal VAUX) and the feedback signal VFB, to simulate demagnetization. The device 250 is used to generate the demagnetization signal Sdmg according to the demagnetization related signal, such as the reflected voltage of the transformer 10 (via the auxiliary signal VAUX), to simulate the demagnetization period TDS.

Please also refer to FIG. 9 , which shows an operation waveform diagram of the degaussing simulator of the present invention generating a degaussing signal.

During the switching cycle, the resonant half-bridge flyback converter periodically operates in the non-discontinuous conduction mode (for example, from time point ta to time point tc'), and the driving signal SH first turns on the first transistor 30 to excite the transformer 10 to generate The primary side switching current IP (for example, time point ta' to time point tb), after the first transistor 30 is turned off, the driving signal SL is used to conduct during the resonance period (time point tb to time point tc) (for example, time point tb to time point tc' ) the second transistor 40, and is used to generate a circulating current (for example, from time point tc to time point tc′) to achieve zero-voltage switching of the first transistor 30. In the switching cycle of the non-discontinuous conduction mode, the conduction period TSL of the driving signal SL (for example, time point tb to time point tc') is determined by the pulse width (for example, TDSX') of the demagnetization signal Sdmg, in which the demagnetization signal Sdmg is determined by the demagnetization signal Sdmg. The magnetic simulator 250 is generated based on the correction in the previously forced-in discontinuous conduction mode. In one embodiment, the conduction period TDSX′ of the demagnetization signal Sdmg is corrected in the previous actively forced discontinuous conduction mode period, and is used to cause the control element 248 to control the minimum conduction time of the second transistor 40, thereby in The transformer 10 is demagnetized during a non-discontinuous conduction mode after the transistor 30 is turned off. In one embodiment, as shown in FIG. 9 , the conduction period TSL of the driving signal SL (for example, from time point tb to time point tc') may be the conduction period TDSX' of the demagnetization signal Sdmg plus a delay time (for example, from time point tc to time point). tc'), so as to establish a negative circulating current on the primary side switch current IP after the demagnetization period, so as to achieve zero-voltage switching of the first transistor 30 .

It should be noted that non-discontinuous conduction mode refers to an operation mode that is not a discontinuous conduction mode, such as continuous conduction mode (CCM) or quasi-resonant mode (QRM) operation. The resonant mode is also called the boundary conduction mode (BCM).

In one embodiment, when the primary side switch current IP has a preset number (for example, a positive integer NC) of switching periods (for example, time point ta to time point t1), the operation is in a non-discontinuous conduction mode (for example, quasi-resonant mode). , at least one switching cycle is actively forced to operate in the discontinuous conduction mode (for example, from time point t1 to time point t3). Therefore, the degaussing simulator 250 is used in the forced insertion discontinuous conduction mode to correct the conduction period TDSX of the degaussing signal Sdmg according to the degaussing period TDS of the transformer 10 .

As shown in FIG. 9 , in the forced insertion discontinuous conduction mode, the degaussing period TDS of the transformer 10 starts from the rising edge of the auxiliary signal VAUX, and ends between the auxiliary signal VAUX (for example, from time t2 to time t3). The falling edge (i.e. knee point kn) ends. Specifically, in this embodiment, the reflected voltage can be detected by sensing the auxiliary signal VAUX, which comes from the auxiliary winding NA of the transformer 10 during the off period of the first transistor 30 . The length of time during which the reflected voltage appears, that is, the pulse width of the auxiliary signal VAUX from the rising edge to the knee point kn, is related to the demagnetization period TDS of the transformer 10 .

In one embodiment, the primary side controller 208 further includes a cycle counter 260. The cycle counter 260 is used to calculate the number of switching cycles operating in the non-discontinuous conduction mode according to the primary side switch current IP, and when the primary side switch current IP is When it is determined that the preset number of switching cycles have not been operated in the discontinuous conduction mode, the period counter 260 is used to control the control element 248 to actively force the operation in the discontinuous conduction mode. In one embodiment, the period counter 260 can sense the primary-side switch current IP through the current sensing signal VCS, thereby determining that the operation is in the non-discontinuous conduction mode.

In one embodiment, as shown in FIG. 9 , during the forced discontinuous conduction mode switching period, the driving signal SL continuously controls the second transistor 40 to be non-conductive, so that the half-bridge circuit not only operates in the discontinuous conduction mode, but also operates in the discontinuous conduction mode. In the asynchronous switching mode, in which during the forced discontinuous conduction mode switching period, a portion of the demagnetization current of the transformer 10 (eg, IP from time point t2 to time point t2 ′) flows through the body diode 45 of the second transistor 40 .

Please continue to refer to FIG. 9. After the discontinuous conduction mode DCM (for example, from time point t4 to time point t5), the first pulse wave of the driving signal SL conducts the second transistor 40 to excite the transformer 10 from the resonant capacitor 20 to the transformer 10, and then A negative circulating current (IP from time point t4 to time point t5) is generated to achieve zero-voltage switching of the first transistor 30 .

FIG. 10 shows a schematic diagram of a specific embodiment of a demagnetization simulator for generating a demagnetization signal Sdmg according to the present invention. In one embodiment, the degaussing simulator 250 includes a timing generator 205, a comparator 280, and a logic circuit 285.

In one embodiment, the timing generator 205 includes an integrator, which is composed of a switch 231 and a capacitor 230. The switch 231 is controlled by the sampling signal SMP, and the sampling signal SMP is related to the driving signal SH to sample the current sensing signal VCS. The discharge current ID is related to n*VO and is used to discharge the cross-voltage VC of the capacitor 230 . The cross voltage VC is compared with the reference voltage Vref by the comparator 280 . The logic circuit 285 generates the degaussing signal Sdmg according to the comparator output CPO and the sampling signal SMP related to the driving signal SH. In one embodiment, the reference voltage Vref is 0 volts, and when the primary side switch current IP is 0, the current sensing voltage VCS is 0.

In one embodiment, the time length of the demagnetization signal Sdmg is related to the voltage level (Vinx) of the input voltage of the transformer 10 , that is, as shown in FIG. 3 , the coupling node NNP of the primary winding NP and the resonant capacitor 20 The voltage and the time length of the demagnetization signal Sdmg are also related to the voltage level of the output voltage of the transformer 10 (for example, n*VO) and the excitation period (TW) of the transformer 10 when the first transistor 30 is turned on. It should be noted that the voltage level Vinx of the input voltage of the transformer 10 is equal to the input voltage VIN minus the voltage Vcr across the resonant capacitor 20 .

According to the fact that the magnetic flux demagnetized by the transformer 10 is equal to the magnetic flux excited by the transformer 10, the following equation 1 can be listed: Vinx*TW=n*VO*TDS (Equation 1)

TW is the occurrence time of the voltage level Vinx of the input voltage of the transformer 10 during the excitation period of the transformer 10 ; n*VO is the voltage of the transformer 10 during the demagnetization period TDS of the transformer 10 . n is the turns ratio of the primary winding NP and the secondary winding NS, and VO is the voltage of the secondary winding NS (ie, the output voltage).

After the transformer 10 is excited, the level VCSp of the current sensing signal VCS is related to the peak value of the primary-side switching current IP at the end of the exciting process, and is generated on the resistor 60 shown in Figure 3, which can be expressed by the following equation 2: VCSp =(Vinx/L)*TW*Rs (Formula 2)

Wherein, L is the inductance of the primary winding NP of the transformer 10, Rs is the resistance value of the resistor 60, and VCSp is the voltage level of the transformer 10 at the end of the excitation process.

Assume ID=n*VO/Rt, where Rt is the resistance value of resistor 55.

The pulse width TDSX of the demagnetization signal Sdmg can be expressed as: TDSX=(C*VCSp)/ID, where C is the capacitance value of the capacitor 230.

TDSX=(Rt*C*VCSp)/(n*VO)

TDSX=(Rt*C/(n*VO))*(Rs/L)*Vinx*TW

Let Rt=L/(Rs*C) (Formula 3)

TDSX=(Vinx*TW)/(n*VO) (Formula 4)

When the condition of Equation 3 is satisfied, the conduction period TDSX of the demagnetization signal Sdmg shown in Equation 4 is equal to the demagnetization period TDS of the transformer 10 .

Please continue to refer to Figure 10. The switch 231 is turned on to sample the current sensing signal VCS to the capacitor 230, and when the switch 231 is turned off (that is, when the excitation ends), the level VCSp of the current sensing signal VCS is maintained at the capacitor 230. The switch 231 is controlled by the sampling signal SMP. When the switch 231 is turned off, the demagnetization signal Sdmg is enabled (for example, through the logic circuit 285). In other words, when the demagnetization signal Sdmg starts to be enabled, the voltage VC across the capacitor 230 is the peak value of the current sensing signal VCS. Turn off switch 231 Afterwards, the discharge current ID begins to discharge the capacitor 230. When the capacitor 230 is completely discharged through the discharge current ID (ID=n*VO/Rt) (VC=0V), the demagnetization signal Sdmg is disabled. The resistor 55 shown in FIG. 10 and FIG. 3 is used to set the preset pulse width of the demagnetization signal Sdmg.

In one embodiment, in the discontinuous conduction mode switching period of forced insertion, the pulse width TDSX of the degaussing signal Sdmg can be compared with the degaussing period TDS represented by the pulse width of the auxiliary signal VAUX by the degaussing simulator 250 Comparison is made, so the pulse width TDSX of the demagnetization signal Sdmg can be corrected and used for the next non-discontinuous conduction mode switching cycle. In one embodiment, the degaussing simulator 250 is further used to adjust the resistance value of the resistor 255 according to the degaussing period TDS detected in the discontinuous conduction mode to correct the pulse period TDSX of the degaussing signal Sdmg.

In other embodiments, in addition to adjusting the resistance value of the resistor 255, the degaussing simulator 250 can also correct the pulse period TDSX of the degaussing signal Sdmg in the following manner: adjusting the voltage threshold Vth to determine the end of the degaussing signal Sdmg. , or adjust the capacitance value of the capacitor 230, or adjust, for example, the ratio of the current mirror composed of the transistor 271 and the transistor 272 in FIG. 10 .

FIG. 11 shows a schematic diagram of a preferred embodiment of the resonant half-bridge flyback converter of the present invention. The resonant half-bridge flyback converter 900 is similar to the resonant half-bridge flyback converter 300 of FIG. 3 . In this embodiment, the resonant half-bridge flyback converter 900 includes a first transistor M1, a second transistor M2, and a third transistor M3. The first transistor M1, the second transistor M2, and the third transistor are M3 is used to form a half-bridge circuit. From one perspective, the first transistor M1 and the third transistor M3 are configured as lower bridge transistors of the resonant half-bridge flyback converter 900 , and the second transistor M2 is configured as a resonant half-bridge flyback converter. 900 upper bridge transistor.

According to the feedback signal VFB and the input voltage VIN, the primary side controller 201 is used to generate the first driving signal S1, the second driving signal S2 and the third driving signal S3. The first driving signal S1, the second driving signal S2 and the third driving signal S3 are generated. The drive signal S3 is coupled to switch the transformer 10 through the half-bridge circuit, thereby generating an output voltage VO on the secondary side of the transformer 10 . The second driving signal S2 drives the second transistor The body M2 excites the transformer 10, and the third driving signal S3 turns on the third transistor M3 during the demagnetization and resonance period of the transformer 10. The third driving signal S3 is also used to turn on the third transistor M3 to generate a voltage flowing through the transformer 10. Circulating current enables zero-voltage switching of the second transistor M2 in a heavy load state. In other words, the second transistor M2 is a primary-side high-bridge switch of the resonant half-bridge flyback converter 900 and can correspond to the first transistor 30 of FIG. 3 , and the third transistor M3 is a resonant half-bridge flyback converter. The primary side lower bridge switch 900 may correspond to the second transistor 40 of FIG. 3 . From one point of view, the first transistor M1 is connected in parallel to the third transistor M3 and serves as an auxiliary primary-side low-bridge switch and has an independent control signal S1.

In one embodiment, after the transformer 10 is excited by turning on the second transistor M2 at a light load state and operating in the discontinuous conduction mode, the third transistor M3 is controlled during the demagnetization and resonance period of the transformer 10 for conduction. After demagnetization, when the third transistor M3 continues to be turned off, the first driving signal S1 is used to turn on the first transistor M1 to generate a circulating current flowing through the transformer 10 to achieve zero-voltage switching of the second transistor M2 ,. Therefore, the third transistor M3 can avoid switching twice in one switching period of the discontinuous conduction mode.

Since the first transistor M1 is only used to generate circulating current to achieve zero voltage switching, in one embodiment, the actual size (such as aspect ratio) of the first transistor M1 can be configured to be much smaller than the actual size of the third transistor M3 size. Therefore, the driving capability and parasitic capacitance (such as gate capacitance) of the first transistor M1 are lower than the parasitic capacitance of the third transistor M3, and the switching loss of the first transistor M1 is also lower than the switching loss of the third transistor M3. .

For example, the gate switching loss Pg of the transistor can be expressed as: Pg=0.5*Ciss*Vg*Vg*Freq

Where Ciss is the input capacitance of the transistor, Vg is the voltage level of the gate drive signal, and Freq is the switching frequency of the gate drive signal.

As shown in the above switching power loss equation, the first transistor M1 with a smaller actual size is used exclusively to achieve zero-voltage switching of the second transistor M2 in the discontinuous conduction mode. Therefore, the gate switching loss of the first transistor M1 is lower than The third transistor M3 is actually larger in size.

In addition, in one embodiment, the amplitude of the voltage level (ie, Vg) of the first driving signal S1 is lower than the amplitude of the voltage level of the third driving signal S3, so the switching loss of the first transistor M1 can be further reduced. And in one embodiment, the gate maximum rating (eg gate-source voltage) of the first transistor M1 may also be lower than the gate maximum rating of the third transistor M3.

The resistor 60 generates the current sensing signal VCS by detecting the primary-side switching current IP of the transformer 10. The primary-side controller 201 is used to generate the first driving signal S1 according to the input voltage VIN, and to output the first driving signal S1 according to the input voltage VIN. voltage VO to generate the third driving signal S3. The primary side controller 201 is further used to generate the second driving signal S2 according to the feedback signal VFB.

FIG. 12 shows an operation waveform diagram of a preferred embodiment of the primary-side controller 201 operating in the discontinuous conduction mode of the present invention. In the discontinuous conduction mode operation, the primary side controller 201 operates in the first switching period Tcyc1 and controls the first driving signal S1 to conduct the first transistor M1 in the first period TA, thereby generating a circulating current to achieve the second Zero voltage switching when transistor M2 is on. After the first period TA, the first driving signal S1, the second driving signal S2 and the third driving signal S3 are used to turn off the first transistor M1 and the first non-conducting period Td1 (ie, the dead period). The second transistor M2 and the third transistor M3. In one embodiment, the first non-conduction period Td1 is related to the quasi-resonant period used to achieve zero-voltage switching of the second transistor M2. After the first non-conduction period Td1, the second driving signal S2 turns on the second transistor M2 in the second period TB, and the conduction of the second transistor M2 is used to excite the transformer 10. After the second period TB, the first driving signal S1, the second driving signal S2 and the third driving signal S3 are used to turn off the first transistor M1, the second transistor M2 and the third transistor M1 during the dead period. Tritransistor M3. In one embodiment, the second non-conduction period Td2 is related to another quasi-resonant period for achieving zero-voltage switching of the third transistor M3. After the second After the non-conduction period Td2, the third driving signal S3 turns on the third transistor M3 during the third period TC, and the third transistor M3 turns on during the demagnetization period of the transformer 10. After the third period TC, the first driving signal S1, the second driving signal S2 and the third driving signal S3 are used to turn off the first transistor M1, the second transistor M2 and the third non-conduction period TZ. In the three-transistor M3, the exciting current IM is maintained at zero in the third non-conduction period TZ (ie, the discontinuous conduction mode). After the third non-conduction period TZ, another switching period Tcyc2 starts.

Figure 13 shows a block diagram of a preferred embodiment of the primary side controller of the present invention. In one embodiment, the primary side controller 213 includes a timer 22 and a control component 243 . The control element 243 is used to generate the first driving signal S1, the second driving signal S2 and the third driving signal S3 according to the input voltage VIN (via VAUX) and the feedback signal VFB.

The timer 22 is used for timing to generate the third non-conduction period TZ. The third non-conduction period TZ starts from the end of the pulse wave of the third driving signal S3 (for example, the falling edge). In one embodiment, when the output power of the resonant half-bridge flyback converter decreases, the third non-conduction period TZ increases correspondingly. Therefore, the switching frequency of the resonant half-bridge flyback converter can also be affected by the resonant half-bridge flyback converter. The output power of the Chi converter is reduced accordingly, thereby improving performance in light load operating conditions.

The present invention has been described above with reference to the preferred embodiments. However, the above description is only to make it easy for those familiar with the art to understand the content of the present invention, and is not intended to limit the scope of rights of the present invention. The various embodiments described are not limited to single application, but can also be used in combination. For example, two or more embodiments can be used in combination, and part of the components in one embodiment can also be used to replace those in another embodiment. Corresponding components. In addition, under the same spirit of the present invention, those skilled in the art can think of various equivalent changes and various combinations. For example, the present invention refers to "processing or computing according to a certain signal or generating a certain output result", which is not limited to Depending on the signal itself, it also includes performing voltage-to-current conversion, current-to-voltage conversion, and/or ratio conversion on the signal when necessary, and then processing or calculating the converted signal to produce an output result. From this it can be It is understood that under the same spirit of the present invention, those skilled in the art can think of various equivalent changes and various combinations. There are many combinations, and they are not listed here. Accordingly, the scope of the present invention is intended to cover the above and all other equivalent changes.

DCM: discontinuous conduction mode

IP: primary side switching current

kn: knee point

NC: positive integer

Sdmg: degaussing signal

SH: drive signal

SL: drive signal

t1-t5: time point

ta-te: time point

ta’,tc’,t2’: time point

TDS: degaussing period

TDSX: conduction period

TDSX’: conduction period

TSL: conduction period

TW: Excitation period

VAUX: auxiliary signal

Claims (30)

A flyback converter includes: a first transistor for switching a transformer to generate a primary-side switching current and an output voltage; and a second transistor for generating a circulating current to achieve the first Zero voltage switching (ZVS) of a transistor; wherein the flyback converter is operating in a non-discontinuous conduction mode when it is determined based on the primary-side switch current that there has been a preset number of switching cycles ( non-discontinuous conduction mode (non-DCM), the flyback converter actively forces at least one switching cycle to operate in a discontinuous conduction mode (DCM). The flyback converter of claim 1, wherein the flyback converter is used to generate a demagnetization signal, wherein the demagnetization signal simulates a demagnetization period of the transformer during the non-discontinuous conduction mode. The second transistor is controlled, wherein the flyback converter corrects the degaussing signal based on the degaussing period during the actively forced discontinuous conduction mode. The flyback converter as claimed in claim 2, wherein in the actively forced discontinuous conduction mode, the flyback converter detects a reflected voltage from an auxiliary winding of the transformer to activate the first voltage The demagnetization signal is corrected during the crystal off period, wherein the length of time the reflected voltage occurs is related to the demagnetization period of the transformer. The flyback converter of claim 3, wherein the first transistor and the second transistor form a half-bridge circuit, wherein in the actively forced discontinuous conduction mode, when the first transistor is turned off Finally, the second transistor is controlled to be turned off, so that the half-bridge circuit operates in an asynchronous mode. The flyback converter of claim 1 further includes a cycle counter, the cycle counter is used to calculate the number of switching cycles in which the flyback converter operates in the non-discontinuous conduction mode, so that when the primary side switching current The flyback converter is enabled to actively force the flyback converter to operate in the discontinuous conduction mode when a predetermined number of switching cycles have been operated in the non-discontinuous conduction mode. The flyback converter of claim 2, wherein a pulse period of the demagnetization signal is used to control the minimum conduction time of the second transistor, thereby demagnetizing the first transistor after it is turned off. the transformer. The flyback converter of claim 2, wherein a pulse period of the demagnetization signal is related to a level of an input voltage in an excitation period of the transformer and a level of the output voltage of the transformer. and a conduction period of the first transistor. The flyback converter as described in claim 2 further includes a resistor, wherein a pulse period of the demagnetization signal is determined according to the resistance value of the resistor; wherein the flyback converter is based on the active voltage of the transformer. The resistance value of the resistor is adjusted during the degaussing period during the forced discontinuous conduction mode to correct the degaussing signal. The flyback converter of claim 1, wherein after turning off the first transistor during the active forced discontinuous conduction mode, a part of a demagnetization current of the transformer flows through the second transistor. A body diode. The flyback converter of claim 1, wherein the flyback converter is a resonant flyback converter and further includes a resonant capacitor, and the resonant capacitor is coupled in series to the transformer. The flyback converter of claim 10, wherein a second driving signal is used to conduct the second transistor in the first pulse wave after the discontinuous conduction mode to switch from the resonant circuit. An excitation current is provided to the transformer to excite the transformer, thereby generating a negative circulating current to achieve zero-voltage switching of the first transistor. A primary-side control circuit for controlling a flyback converter, wherein the flyback converter includes: a first transistor for switching a transformer to generate a primary-side switching current and an output voltage; and A second transistor is used to generate a circulating current to achieve zero voltage switching (ZVS) of the first transistor. The primary side control circuit includes: a control element used to generate a first signal and a second signal, the first signal and the second signal respectively control the first transistor and the second transistor according to a feedback signal, wherein the feedback signal is related to the output voltage; and a period counter for The number of switching cycles in which the flyback converter operates in a non-discontinuous conduction mode is calculated based on the primary-side switching current; wherein the control element determines that there has been a preset number of switching cycles based on the primary-side switching current. When operating in the non-discontinuous conduction mode, the control element actively forces at least one switching cycle to operate in a discontinuous conduction mode (DCM). The primary side control circuit of claim 12 further includes a degaussing simulator, the degaussing simulator is used to generate a degaussing signal to simulate a degaussing period of the transformer, and then in the non-discontinuous conduction The second transistor is controlled during the mode, wherein the degaussing simulator corrects the degaussing signal according to the degaussing period during the actively forced discontinuous conduction mode. The primary side control circuit of claim 13, wherein in the actively forced discontinuous conduction mode, the degaussing simulator detects a reflected voltage from an auxiliary winding of the transformer to apply the voltage to the first transistor. The degaussing signal is corrected during the off period, wherein the length of time the reflected voltage occurs is related to the degaussing period of the transformer. A control circuit as claimed in claim 14, wherein the first transistor and the second transistor form a half-bridge circuit, wherein in the actively forced discontinuous conduction mode, in the After the first transistor is turned off, the second transistor is controlled to be turned off, so that the half-bridge circuit operates in an asynchronous mode. The primary side control circuit as described in claim 13, wherein a pulse period of the demagnetization signal is used to control the minimum conduction time of the second transistor, thereby demagnetizing the first transistor after it is turned off. transformer. The primary side control circuit of claim 13, wherein a pulse period of the demagnetization signal is related to a level of an input voltage in an excitation period of the transformer and a level of the output voltage of the transformer and a conduction period of the first transistor. The primary side control circuit of claim 13, wherein the primary side control circuit further adjusts the resistance value of a resistor to correct the demagnetization according to the demagnetization period of the transformer during the active forced discontinuous conduction mode. A signal, wherein the resistor is coupled to the flyback converter, and wherein a pulse period of the demagnetization signal is determined according to the resistance value of the resistor. The primary side control circuit of claim 12, wherein during the active forced discontinuous conduction mode, the control element controls the second transistor to turn off, so that after the first transistor turns off, a portion of the transformer A portion of the demagnetizing current flows through a body diode of the second transistor. The primary side control circuit of claim 12, wherein a second driving signal is used to turn on the second transistor in the first pulse wave after the discontinuous conduction mode to convert from a The resonant capacitor provides an exciting current to the transformer to excite the transformer, thereby generating a negative circulating current to achieve zero-voltage switching of the first transistor, wherein the resonant capacitor is coupled in series to the transformer. The primary side control circuit as claimed in claim 14, wherein the degaussing simulator includes: an integrating capacitor and a switch, wherein the switch is coupled to sample and maintain a current sensing signal, wherein the current sensing signal is related to the primary side switch current; and a discharge circuit for generating a discharge current to integrate the integrating The capacitor is discharged, wherein the discharge current is related to the output voltage; wherein the demagnetization signal is generated according to a voltage across the integrating capacitor, and wherein the discharge current discharges the integrating capacitor from one end of the energized transformer. A control method for controlling a flyback converter, wherein the flyback converter includes: a first transistor for switching a transformer to generate a primary-side switching current and an output voltage; and a first transistor. Two transistors are used to generate a circulating current to achieve zero voltage switching (ZVS) of the first transistor. The control method includes: generating a first signal and a second signal, the first signal and The second signal controls the first transistor and the second transistor respectively according to a feedback signal, wherein the feedback signal is related to the output voltage; the flyback converter operation is calculated based on the primary side switch current. a number of switching cycles in a non-discontinuous conduction mode; and when it is determined based on the primary side switch current that a predetermined number of switching cycles have been operated in the non-discontinuous conduction mode, actively forcing at least one switching cycle to operate in a non-discontinuous conduction mode Continuous conduction mode (DCM). The control method of claim 22, further comprising: generating a demagnetization signal to simulate a demagnetization period of the transformer, thereby controlling the second transistor during the non-discontinuous conduction mode; and according to the active forcing The degaussing signal is corrected during the degaussing period during the discontinuous conduction mode. The control method of claim 23, wherein the step of correcting the demagnetization signal includes: detecting a reflected voltage from an auxiliary winding of the transformer in the active forced discontinuous conduction mode to generate a signal in the first voltage. The demagnetization signal is corrected during the crystal off period, wherein the length of time the reflected voltage occurs is related to the demagnetization period of the transformer. The control method of claim 24, wherein the first transistor and the second transistor form a half-bridge circuit, and the step of correcting the demagnetization signal further includes: in the actively forced discontinuous conduction mode, in After the first transistor is turned off, the second transistor is controlled to be turned off, so that the half-bridge circuit operates in an asynchronous mode. The control method as described in claim 23, wherein the step of generating the first signal and the second signal includes: controlling the minimum conduction time of the second transistor according to a pulse period of the demagnetization signal, thereby After the first transistor is turned off, the transformer is demagnetized. The control method as claimed in claim 23, wherein a pulse period of the demagnetization signal is related to a level of an input voltage in an excitation period of the transformer, a level of the output voltage of the transformer and the level of the transformer. A conduction period of the first transistor. The control method of claim 23, wherein the step of correcting the demagnetization signal includes: adjusting a resistance value of a resistor to correct the demagnetization period according to the demagnetization period of the transformer during the active forced discontinuous conduction mode. Magnetic signal, wherein the resistor is coupled to the flyback converter, wherein a pulse period of the demagnetization signal is determined according to the resistance value of the resistor. The control method as described in request 22 further includes: During the active forced discontinuous conduction mode, the second transistor is controlled to be turned off, so that after the first transistor is turned off, a part of a demagnetizing current of the transformer flows through a body 2 of the second transistor. polar body. The control method as described in claim 22, further comprising: using a second driving signal to conduct the second transistor in the first pulse wave after the discontinuous conduction mode, so as to convert a pulse from the flyback converter to The resonant capacitor provides an exciting current to the transformer to excite the transformer, thereby generating a negative circulating current to achieve zero-voltage switching of the first transistor, wherein the resonant capacitor is coupled in series to the transformer.

Images