Soft switch design method of push-pull resonant converter
By designing soft switches in push-pull resonant converter, optimizing the resonant cavity and transformer parameters, the voltage ringing problems caused by low-frequency next resonant current and hard switch are solved, and zero voltage shutdown and turn-on are achieved, which improves the converter's Efficiency and safety.
Patent Information
- Application Number
- CN202410981505.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-08
AI Technical Summary
Traditional push-pull converters have subresonant current phenomena and voltage ringing problems caused by hard shutdown of switch tubes at low frequency, which increases switching losses and breakdown risks.
By designing a soft switching method for push-pull resonant converter, the resonant frequency, transformer parameters and switching frequency are determined, the zero voltage is turned off and on, the resonant cavity and transformer parameters are optimized, and the sub-resonant mode and hard switch are avoided.
The zero-voltage shutdown and turn-on of the switch tube is realized, which reduces switching losses, improves the efficiency of the converter, avoids voltage ringing caused by the resonant current and hard switches in the low-frequency operation, and improves the safety and efficiency of the equipment.
Smart Images

Figure CN118920815A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of resonant direct current converters, and in particular to a soft switch design method for a push-pull resonant converter. Background Art
[0002] The number of switch tubes in a push-pull converter is half that of a full-bridge converter, so the conduction loss and switching loss of the switch tube are also less. In addition, all switch tubes of a push-pull converter are grounded, and the drive circuit is simple and low-cost, so the push-pull converter is suitable for high-power and high-ratio DC boost converters. However, the primary switch tube of a traditional push-pull converter is hard-off and hard-on, with large switching losses and low efficiency. In addition, the inductance of the transformer and the parasitic capacitance of the switch tube produce very large voltage ringing at the end of the dead time of the switch tube, which may even break down the switch tube.
[0003] To avoid the above problems, push-pull resonant converters have been proposed and widely used. In particular, when the series resonant cavity is located on the high-voltage side of the transformer of the push-pull resonant converter, the resonant current flowing through the resonant cavity is small, which can reduce the conduction loss of the resonant cavity. However, there are also some problems with the push-pull resonant converter. For example, in practical applications, it is necessary to reasonably design the resonant cavity parameters and transformer parameters to ensure the soft shutdown and soft opening of the primary side switch tube of the transformer and reduce the shutdown ringing of the switch tube. The details are as follows:
[0004] On the one hand, when the switching frequency f s Below the resonant frequency f r If the quality factor Q of the secondary LC resonant cavity and the AC equivalent resistor is too large, the resonant cavity current will continue to conduct after crossing zero, and a reverse current will appear on the switch tube, causing the switch tube to be hard turned off, while increasing the turn-off loss, and then leading to an increase in the turn-off voltage ringing peak. Therefore, it is necessary to reasonably design the resonant cavity parameters to reduce the turn-off loss of the primary switch tube of the transformer.
[0005] On the other hand, if the switch tube is turned off when the current in the resonant cavity reaches zero, the switch tube enters the dead time. At this time, the parasitic capacitance of the switch tube in parallel in the off state is discharged by the excitation inductance current, and the excitation current simultaneously charges the parasitic capacitance of the switch tube of the other winding. If the primary inductance of the transformer (excitation inductance and leakage inductance) is greater than the ideal value, the inductance current is too small, so that the parasitic capacitance voltage of the switch tube cannot be discharged to zero (or fully charged to the shutdown voltage) within the dead time. At this time, triggering the switch tube will cause the parasitic capacitance of the switch tube to be quickly discharged to zero (or quickly charged to the shutdown voltage). Because the parasitic capacitance charging and discharging circuit is connected in series with the leakage inductance of the transformer, an LC oscillation circuit is formed, which causes the switch tube to generate a shutdown voltage ringing, and the voltage ringing peak voltage is as high as three times the input voltage. In this process, the switch tube is hard turned on, which increases the switching loss. Therefore, it is necessary to reasonably design the primary inductance parameters of the transformer to obtain the ideal inductance current, and use the primary inductance current of the transformer to achieve zero voltage shutdown and zero voltage turn-on of the primary switch tube, thereby eliminating the shutdown voltage ringing on the switch tube.
[0006] To this end, it is necessary to optimize the parameters of the push-pull resonant converter to avoid the sub-resonant current phenomenon in the resonant cavity under low-frequency operation, and to solve the turn-off voltage ringing caused by the hard switching of the switch tube during the dead time to avoid the risk of the switch tube being broken down. Summary of the invention
[0007] The present invention aims to optimize the parameters of a push-pull resonant converter, avoid the sub-resonant current phenomenon of the resonant cavity under low-frequency operation, solve the problem of turn-off voltage ringing caused by hard switching of the switch tube during the dead time, and avoid the risk of breakdown of the switch tube. A soft switch (soft switch, i.e., parameters of zero voltage shutdown and zero voltage start-up) design method for a push-pull resonant converter is provided, which reduces switching losses and improves the efficiency of the converter.
[0008] The technical solution adopted by the present invention is:
[0009] A soft switching design method for a push-pull resonant converter, the method comprising the following steps:
[0010] Step 1: Determine the resonant frequency, rated power, input and output rated voltage;
[0011] Step 2: Calculate the transformer turns ratio, load resistance and output current;
[0012] Step 3: The dead time of the push-pull resonant converter realizes zero voltage switching, that is, the duty cycle is set to 0.47~0.49, and the switching frequency f is calculated from this s Slightly below the resonant frequency f r , f s =0.94f r ~0.98f r ;
[0013] Step 4: Calculate the maximum quality factor Q required to avoid sub-resonant mode based on expression (24) and the normalized switching frequency expression (15): max ;
[0014] Step 5: Basis and Design equivalent resonant inductance and resonant capacitance to ensure Q < Q max , and contains a certain margin;
[0015] Step 6: Pass Calculate the peak value of the resonant current;
[0016] Step 7: Select the switch tube in the push-pull resonant converter and determine the number of switch tubes and the size of parasitic capacitance;
[0017] Step 8: Basis Calculate the magnetizing inductance and leakage inductance required for zero voltage switching.
[0018] Preferably, the method for determining the resonant frequency, rated power, input and output rated voltages in step 1 is:
[0019] Due to the limitations of power, heat dissipation, output ripple, electromagnetic compatibility (EMC) and actual transformer volume, if the rated power is above 300W and the input voltage is low, in order to ensure actual requirements such as heat dissipation, the resonant frequency is usually designed to be 100kHz~200kHz;
[0020] The rated input voltage is determined by the voltage of the preceding power supply, such as the number of cells connected in series in a battery pack;
[0021] The rated power and output voltage are determined by specific requirements. For the DC converter in the front stage of the vehicle-mounted boost inverter, the output voltage is at least 380V and the rated power is above 1.5kW.
[0022] Preferably, the method for calculating the transformer turns ratio, load resistance and output current in step 2 is:
[0023] Transformer turns ratio
[0024] Output Current
[0025] Load resistance
[0026] Where V0 is the rated output voltage, V in is the rated input voltage, P0 is the rated power, I d is the rated output current.
[0027] Preferably, the switching frequency f is calculated in step 3 s Slightly below the resonant frequency f r , f s =0.94f r ~0.98f r The method is:
[0028] Since dead time is required to achieve zero voltage switching, a duty cycle D less than 0.5 needs to be designed. The duty cycle is the ratio of the conduction time of the switch tube on a certain bridge arm per unit cycle to the entire signal cycle time. The expression of the duty cycle D is as follows:
[0029]
[0030] Where T on is the on-time, T d is the dead time; since the realization of zero voltage switching requires the existence of dead time, it can be seen from formula (4) that the duty cycle D must be less than 0.5;
[0031]
[0032] where f s is the switching frequency, f on is the conduction frequency without taking into account the dead time; thus we get
[0033] f s =2Df on (6)
[0034] And because the DC value of the resonant inductor current in half a switching cycle is:
[0035]
[0036] in is the peak value of the secondary resonant current. Substituting equation (5) into equation (7), we get:
[0037]
[0038] From equation (8), we can see that the reduction of D will cause the resonant current peak The voltage and current stress of each component will increase accordingly, which will lead to an increase in cost. Therefore, the duty cycle D is selected to be 0.47-0.49, that is, f s =0.94f r ~0.98f r ;
[0039] At this time, the resonant peak current increases by 0.01 to 0.03 times compared with the resonant current peak value with a duty cycle of 0.5.
[0040] Preferably, in step 4, the maximum quality factor Q required to avoid the sub-resonance mode is calculated according to formula (24) and the normalized switching frequency formula (15): max The method is:
[0041] F is the normalized switching frequency,
[0042]
[0043] Preferably, in step 5, and Design equivalent resonant inductance and resonant capacitance to ensure Q < Q max , and the method that includes a certain margin is:
[0044] (a) Due to the symmetry of the resonant waveform, the analysis and derivation of half the switching cycle of the push-pull resonant converter yields:
[0045] The Q (quality factor) composed of the secondary LC resonant cavity and the AC equivalent resistance,
[0046] AC equivalent resistance
[0047] Equivalent resonant inductance L' r =N 2 L σ1 +L σ2 +L r (11)
[0048] Among them, R e is the AC equivalent resistance, R o is the load resistance; L' r is the equivalent resonant inductor, which is composed of the primary leakage inductance L σ1 , secondary leakage inductance L σ2 and resonant inductor L r Composition, C r is the resonant capacitor;
[0049] (b) According to the relationship between the capacitor voltage and current (12), the resonant capacitor voltage value at any time τ is obtained:
[0050]
[0051] in, is the initial value of the resonant capacitor voltage, and τ is any time;
[0052] (c) Assume that the initial value of the capacitor voltage is Then the resonant capacitor voltage reaches its maximum value after half a resonant cycle. Since the capacitor voltage is a sinusoidal symmetrical waveform, Substituting it into equation (12) yields the maximum value of the resonant capacitor voltage expression (13), where T r is the resonant cycle time, i.e. T r =1 / f r Since only half of the resonant current of the resonant cycle flows through the resonant capacitor in half a switching cycle, τ at any time in equation (12) is T r / 2. It should be noted that in order to ensure that the resonant current reaches zero when the switch is turned off, the switching cycle time needs to be greater than or equal to the resonant cycle time, that is, T r ≤T s ;
[0053]
[0054] (d) The DC value of the resonant inductor current in half a switching cycle is obtained from equation (7): d , Substituting equation (7) into equation (13) and simplifying it, we get equation (14); where T s is the unit switching cycle time, that is, T s =1 / f s , F is the normalized switching frequency taking into account the dead time, that is, formula (15), where f r is the resonant frequency, that is, equation (17), and f s is the switching frequency, V0 is the rated output voltage;
[0055] F' is the normalized switching frequency without taking into account the dead zone, that is, formula (16), where f s ' is the ideal switching frequency without dead time. At this time, the push-pull resonant converter waveform has T d Dead time, hence the switching frequency f s Less than the resonant frequency f r ; However, the on-time of the switch tube remains unchanged, so the actual gain is still unity gain; but for the traditional gain expression, since D < 0.5, the traditional theoretical gain is already less than unity gain, so it is necessary to introduce the ideal switching frequency f without dead time s 'and the normalized switching frequency F' excluding the dead zone;
[0056]
[0057] (e) Considering the existence of dead time, the series resonant gain M of this push-pull resonant converter is different from the traditional series gain. The series gain of this converter can be calculated as formula (22), where D is the duty cycle. It should be noted that D < 0.5;
[0058] According to the secondary side AC equivalent model, the voltage division expression can be obtained:
[0059]
[0060] where ω s is the switching angular frequency, defined as ω s =2πf s , j is the imaginary unit, Nv in (jω s ) is the square wave voltage output by the primary full bridge and equivalent to the secondary side, v o (jω s ) is the square wave voltage of the secondary full-bridge input. After simplifying expression (18), and combining expression (9) and expression (16), we get expression (19):
[0061]
[0062] Taking the modulus of expression (19) yields the expression for gain M:
[0063]
[0064] Considering the influence of dead time on the resonant circuit gain, the relationship between the normalized frequency F' without taking the dead time into account and the normalized frequency F taking the dead time into account is obtained as formula (21):
[0065] F'=F+1-2D (21)
[0066] Substituting equation (21) into gain expression (20) yields equation (22);
[0067]
[0068] (f) The critical condition for the push-pull resonant converter to generate sub-resonant current is given by equation (23);
[0069]
[0070] From equations (13) and (23), it can be seen that once the resonant inductor current is too large, the voltage on the resonant capacitor will exceed the sum of the input voltage on the secondary side of the transformer and the output voltage on the load side. This will cause the rectifier bridge to turn on again, and the resonant capacitor will discharge along the loop to the load side, generating a negative half-wave current and entering the sub-resonant mode.
[0071] (h) Substituting equation (14) and equation (22) into equation (23), we can simplify equation (24);
[0072]
[0073] Since parasitic parameters such as parasitic inductance in the actual circuit will affect the Q value of the resonant cavity, there is a certain deviation from the theoretical Q value. Therefore, a margin is left in the actual design. For example, when a double margin is reserved, the Q value should be less than Q max / 2 to ensure that the DC converter does not enter sub-resonant mode;
[0074] Combining equations (9) and (17) to obtain the equivalent resonant inductance L' r and resonant capacitor C r The expression is:
[0075]
[0076] Preferably, in step six, The method for calculating the peak value of the resonant current using formula (28) is:
[0077] Substituting the design parameter duty cycle D and the normalized frequency F including the dead time into equation (24), the critical Q value of the sub-resonance mode can be obtained. The circuit can be prevented from entering the sub-resonance mode by designing the equivalent resonant inductor, resonant capacitor and load resistor.
[0078] The resonant current expression (27) is obtained through the complex frequency domain model of the circuit,
[0079]
[0080] Formula (27) is used to calculate the resonant current i Lr (t) is the peak current; where I L (0 - ) is the initial value of the inductor current, V c (0 - ) is the initial value of the capacitor voltage, ω r is the resonant angular frequency;
[0081] Since the resonant current of the push-pull resonant converter needs to be designed as a complete half-wave to achieve the soft switching characteristics, the resonant current always reaches zero within half a switching cycle, that is, I L (0 - )=0; According to the resonant sine waveform, when When , the resonant current is the peak current, so equation (27) is simplified to equation (28);
[0082] Peak value of resonant current
[0083] Preferably, in step seven, the method for selecting the switch tubes in the push-pull resonant converter and determining the number of switch tubes and the size of parasitic capacitance is:
[0084] The peak value of the resonant current flowing through the secondary resonant cavity is obtained by equation (28). The peak value of the resonant current on the secondary side is equivalent to the sum of the value of the primary side and the transformer inductance current, that is, Ni Lr_max +i M_P , as the basis for selecting the switch tube and determining the number of switch tubes n;
[0085] Because the switch tube has a limit current that it can withstand, once the current flowing through the switch tube exceeds the limit current, the switch tube is at risk of burning out; for the boost resonant circuit, the primary current stress flowing through the switch tube is very large, and a single switch tube may not be able to withstand the secondary side equivalent to the primary side resonant peak current, so it is necessary to use n switch tubes in parallel to reduce the current stress and conduction loss flowing through each switch tube;
[0086] The parasitic capacitance C of a single switch tube is obtained from the switch tube selection data sheet. oss .
[0087] Preferably, in step eight, The method to calculate the magnetizing inductance and leakage inductance required for zero voltage switching is:
[0088] Through the inductor volt-second balance, we can get:
[0089]
[0090] From formula (29), we can get the transformer inductor current peak value i after the switch is turned off: M_P ,in is the time when the primary switch is turned on, T d is the dead time in one cycle, L m is the transformer’s magnetizing inductance;
[0091] Since the voltage of the switch tube when it is turned off is 2V in , so the total amount of charge stored in the parasitic capacitance of the switch tube is q total As shown in formula (30):
[0092]
[0093] Substituting equation (29) into the charge equation (30) and simplifying it, we can obtain equation (31);
[0094]
[0095] In order to ensure that the dead time can achieve zero voltage switching characteristics, it is known from equation (29) that the sum of the primary excitation inductance and leakage inductance needs to be less than or equal to the theoretical value to obtain sufficient i M_P , then equation (31) evolves into inequality (32);
[0096]
[0097] Once the restriction of inequality (32) is met, the push-pull resonant converter can utilize the non-mutational characteristic of the excitation inductor current during the preset dead time to charge the parasitic capacitance in the switch tube that is about to be turned off through the freewheeling process of the excitation current, and discharge the parasitic capacitance in the switch tube that is about to be turned on, so as to achieve zero voltage switching characteristics and realize zero voltage shutdown and zero voltage turn-on.
[0098] Beneficial effects: The present invention realizes the zero voltage turn-off and zero voltage turn-on characteristics of the switch tube of the push-pull resonant converter by designing the maximum value of the equivalent inductance (the sum of the excitation inductance and the leakage inductance) of the primary transformer and the length of the dead time, and determines the range of the Q value of the resonant cavity to ensure that the resonant cavity does not generate sub-resonant current and avoid the switch tube from suffering hard switching, thereby improving the efficiency of the converter. The influence of the zero voltage switching process on the voltage gain is analyzed and calculated (different from the reverse resonant current method of the traditional series resonance frequency reduction voltage regulation, the present invention uses the equivalent inductance current of the primary transformer in the dead zone to achieve zero voltage switching).
[0099] The present invention designs a transformer to meet the requirement of zero voltage switching. The primary inductance of the transformer is 1000mA. In the 70ns dead time, the drain-source voltage of the switch tube has dropped to 0 before the turn-on signal, achieving the zero voltage turn-on characteristic. The drain-source voltage of the switch tube of the other bridge arm rises smoothly, and the voltage ringing disappears (such as Figure 5 and Figure 6 In addition, since the resonant current will not have a negative half-wave current within the designed Q value range, the switch tube can be turned off after the resonant current reaches zero (as shown in Figure 6 The peak efficiency of the optimized push-pull resonant converter is as high as 98.3%, which avoids the sub-resonant current phenomenon of the resonant cavity under low-frequency operation, solves the turn-off voltage ringing caused by the hard switching of the switch tube in the dead time, and provides zero voltage shutdown and zero voltage start of the push-pull resonant converter.
[0100] The present invention solves the problem that the ringing phenomenon generated when the switch tube is turned off increases the probability of the switch tube being broken down, and solves the problem that when the switching frequency is lower than the resonant frequency, additional turn-on loss is generated, and the negative half-wave current of the resonant cavity causes additional turn-off loss of the primary switch. The present invention is also applicable to isolated DC boost converters. BRIEF DESCRIPTION OF THE DRAWINGS
[0101] Figure 1 is a flow chart of the soft switching design method of the push-pull resonant converter;
[0102] Figure 2 It is a structural schematic diagram of a zero voltage switching circuit of a push-pull resonant converter;
[0103] Figure 3 It is the working waveform of the push-pull resonant converter;
[0104] Figure 4 It is the modal equivalent circuit diagram of the push-pull resonant converter;
[0105] Figure 5 This is the experimental waveform diagram of the transformer primary switch not being able to turn on at zero voltage;
[0106] Figure 6 It is the experimental waveform of the transformer primary switch being turned on with zero voltage;
[0107] Figure 7 This is a schematic diagram of the waveform of the resonant current to zero of the primary switch tube of the transformer when the Q value is small;
[0108] Figure 8 This is a schematic diagram of the negative half-wave current turn-off waveform of the primary switch tube of the transformer when the Q value is large;
[0109] Fig. 9 This is the effect of the dead time of zero voltage switching on the gain;
[0110] Fig.10 It is the equivalent circuit diagram of the primary side to the secondary side;
[0111] Fig.11 It is a schematic diagram of the complex frequency domain model;
[0112] Fig.12 It is a schematic diagram of the secondary side AC equivalent model. DETAILED DESCRIPTION
[0113] Specific implementation method 1. Reference Figures 1 to 12 Specifically describe this embodiment, the soft switch design method of the push-pull resonant converter described in this embodiment is a control method for the push-pull resonant converter, and the method includes the following steps: The soft switch design method of the push-pull resonant converter includes the following steps:
[0114] Step 1: Determine the resonant frequency, rated power, input and output rated voltage;
[0115] Step 2: Calculate the transformer turns ratio, load resistance and output current;
[0116] Step 3: The dead time of the push-pull resonant converter realizes zero voltage switching, that is, the duty cycle is set to 0.47~0.49, and the switching frequency f is calculated from this s Slightly below the resonant frequency f r , f s =0.94f r ~0.98f r;
[0117] Step 4: Calculate the maximum quality factor Q required to avoid sub-resonant mode based on expression (24) and the normalized switching frequency expression (15): max ;
[0118] Step 5: Basis and Design equivalent resonant inductance and resonant capacitance to ensure Q < Q max , and contains a certain margin;
[0119] Step 6: Pass Calculate the peak value of the resonant current;
[0120] Step 7: Select the switch tube in the push-pull resonant converter and determine the number of switch tubes and the size of parasitic capacitance;
[0121] Step 8: Basis Calculate the magnetizing inductance and leakage inductance required for zero voltage switching.
[0122] Specific implementation method 2: This implementation method is a further description of the soft switch design method of the push-pull resonant converter described in implementation method 1. In this implementation method,
[0123] In the soft switching design method of the push-pull resonant converter, the method for determining the resonant frequency, the rated power, and the input and output rated voltages in step 1 is:
[0124] Due to the limitations of power, heat dissipation, output ripple, electromagnetic compatibility (EMC) and actual transformer volume, if the rated power is above 300W and the input voltage is low, in order to ensure actual requirements such as heat dissipation, the resonant frequency is usually designed to be 100kHz~200kHz;
[0125] The rated input voltage is determined by the voltage of the preceding power supply, such as the number of cells connected in series in a battery pack;
[0126] The rated power and output voltage are determined by specific requirements. For the DC converter in the front stage of the vehicle-mounted boost inverter, the output voltage is at least 380V and the rated power is above 1.5kW.
[0127] Specific implementation method three, this implementation method is a further description of the soft switch design method of the push-pull resonant converter described in implementation method one. In this implementation method, the method for calculating the transformer turns ratio, load resistance and output current in step two is:
[0128] Transformer turns ratio
[0129] Output Current
[0130] Load resistance
[0131] Where V0 is the rated output voltage, V in is the rated input voltage, P0 is the rated power, I d is the rated output current.
[0132] Specific implementation mode 4: This implementation mode is a further description of the soft switching design method of the push-pull resonant converter described in the implementation mode 1. In this implementation mode, the switching frequency f is calculated in the step 3. s Slightly below the resonant frequency f r , f s =0.94f r ~0.98f r The method is:
[0133] Since dead time is required to achieve zero voltage switching, a duty cycle D less than 0.5 needs to be designed. The duty cycle is the ratio of the conduction time of the switch tube on a certain bridge arm per unit cycle to the entire signal cycle time. The expression of the duty cycle D is as follows:
[0134]
[0135] Where T on is the on-time, T d is the dead time; since the realization of zero voltage switching requires the existence of dead time, it can be seen from formula (4) that the duty cycle D must be less than 0.5;
[0136]
[0137] where f s is the switching frequency, f on is the conduction frequency without taking into account the dead time; thus we get
[0138] f s =2Df on (6)
[0139] And because the DC value of the resonant inductor current in half a switching cycle is:
[0140]
[0141] Among them Lr_max is the peak value of the secondary resonant current. Substituting equation (5) into equation (7), we get:
[0142]
[0143] From equation (8), we can see that the reduction of D will cause the resonant current peak value i Lr_maxThe voltage and current stress of each component will increase accordingly, which will lead to an increase in cost. Therefore, the duty cycle D is selected to be 0.47-0.49, that is, f s =0.94f r ~0.98f r ;
[0144] At this time, the resonant peak current increases by 0.01 to 0.03 times compared with the resonant current peak value with a duty cycle of 0.5.
[0145] Specific implementation method 5. This implementation method is a further description of the soft switching design method of the push-pull resonant converter described in implementation method 1. In step 4, the maximum quality factor Q required to avoid the sub-resonance mode is calculated according to formula (24) and the normalized switching frequency formula (15). max The method is:
[0146] F is the normalized switching frequency,
[0147]
[0148] Specific implementation method 6: This implementation method is a further description of the soft switch design method of the push-pull resonant converter described in implementation method 1. and Design equivalent resonant inductance and resonant capacitance to ensure Q < Q max , and the method that includes a certain margin is:
[0149] (a) Due to the symmetry of the resonant waveform, the analysis and derivation of half the switching cycle of the push-pull resonant converter yields:
[0150] The Q (quality factor) composed of the secondary LC resonant cavity and the AC equivalent resistance,
[0151] AC equivalent resistance
[0152] Equivalent resonant inductance L' r =N 2 L σ1 +L σ2 +L r (11)
[0153] Among them, R e is the AC equivalent resistance, R o is the load resistance; L' r is the equivalent resonant inductor, which is composed of the primary leakage inductance L σ1 , secondary leakage inductance L σ2 and resonant inductor L r Composition, C ris the resonant capacitor;
[0154] (b) According to the relationship between the capacitor voltage and current (12), the resonant capacitor voltage value at any time τ is obtained:
[0155] in, is the initial value of the resonant capacitor voltage, and τ is any time;
[0156] (c) Assume that the initial value of the capacitor voltage is Then the resonant capacitor voltage reaches its maximum value after half a resonant cycle. Since the capacitor voltage is a sinusoidal symmetrical waveform, Substituting it into equation (12) yields the maximum value of the resonant capacitor voltage expression (13), where T r is the resonant cycle time, i.e. T r =1 / f r Since only half of the resonant current of the resonant cycle flows through the resonant capacitor in half a switching cycle, τ at any time in equation (12) is T r / 2. It should be noted that in order to ensure that the resonant current reaches zero when the switch is turned off, the switching cycle time needs to be greater than or equal to the resonant cycle time, that is, T r ≤T s ;
[0157]
[0158] (d) The DC value of the resonant inductor current in half a switching cycle is obtained from equation (7): d , Substituting equation (7) into equation (13) and simplifying it, we get equation (14); where T s is the unit switching cycle time, that is, T s =1 / f s , F is the normalized switching frequency taking into account the dead time, that is, formula (15), where f r is the resonant frequency, that is, equation (17), and f s is the switching frequency, V0 is the rated output voltage;
[0159] F' is the normalized switching frequency without taking into account the dead zone, that is, formula (16), where f s ' is the ideal switching frequency without dead time. At this time, the push-pull resonant converter waveform has T d Dead time, hence the switching frequency f s Less than the resonant frequency f r; However, the on-time of the switch tube remains unchanged, so the actual gain is still unity gain; but for the traditional gain expression, since D < 0.5, the traditional theoretical gain is already less than unity gain, so it is necessary to introduce the ideal switching frequency f without dead time s 'and the normalized switching frequency F' excluding the dead zone;
[0160]
[0161]
[0162] (e) Considering the existence of dead time, the series resonant gain M of this push-pull resonant converter is different from the traditional series gain. The series gain of this converter can be calculated as formula (22), where D is the duty cycle. It should be noted that D < 0.5;
[0163] According to the secondary side AC equivalent model, the voltage division expression can be obtained:
[0164]
[0165] where ω s is the switching angular frequency, defined as ω s =2πf s , j is the imaginary unit, Nv in (jω s ) is the square wave voltage output by the primary full bridge and equivalent to the secondary side, v o (jω s ) is the square wave voltage of the secondary full-bridge input. After simplifying expression (18), and combining expression (9) and expression (16), we get expression (19):
[0166]
[0167] Taking the modulus of expression (19) yields the expression for gain M:
[0168]
[0169] Considering the influence of dead time on the resonant circuit gain, the relationship between the normalized frequency F' without taking the dead time into account and the normalized frequency F taking the dead time into account is obtained as formula (21):
[0170] F'=F+1-2D (21)
[0171] Substituting equation (21) into gain expression (20) yields equation (22);
[0172]
[0173] (f) The critical condition for the push-pull resonant converter to generate sub-resonant current is given by equation (23);
[0174]
[0175] From equations (13) and (23), it can be seen that once the resonant inductor current is too large, the voltage on the resonant capacitor will exceed the sum of the input voltage on the secondary side of the transformer and the output voltage on the load side. This will cause the rectifier bridge to turn on again, and the resonant capacitor will discharge along the loop to the load side, generating a negative half-wave current and entering the sub-resonant mode.
[0176] (h) Substituting equation (14) and equation (22) into equation (23), we can simplify equation (24);
[0177]
[0178] Since parasitic parameters such as parasitic inductance in the actual circuit will affect the Q value of the resonant cavity, there is a certain deviation from the theoretical Q value. Therefore, a margin is left in the actual design. For example, when a double margin is reserved, the Q value should be less than Q max / 2 to ensure that the DC converter does not enter sub-resonant mode;
[0179] Combining equations (9) and (17) to obtain the equivalent resonant inductance L' r and resonant capacitor C r The expression is:
[0180]
[0181] Specific implementation method 7: This implementation method is a further description of the soft switch design method of the push-pull resonant converter described in implementation method 1. In step 6, The method for calculating the peak value of the resonant current using formula (28) is:
[0182] Substituting the design parameter duty cycle D and the normalized frequency F including the dead time into equation (24), the critical Q value of the sub-resonance mode can be obtained. The circuit can be prevented from entering the sub-resonance mode by designing the equivalent resonant inductor, resonant capacitor and load resistor.
[0183] The resonant current expression (27) is obtained through the complex frequency domain model of the circuit,
[0184]
[0185] Formula (27) is used to calculate the resonant current The peak current size; where I L (0 - ) is the initial value of the inductor current, V c (0 -) is the initial value of the capacitor voltage, ω r is the resonant angular frequency;
[0186] Since the resonant current of the push-pull resonant converter needs to be designed as a complete half-wave to achieve the soft switching characteristics, the resonant current always reaches zero within half a switching cycle, that is, I L (0 - )=0; According to the resonant sine waveform, when When , the resonant current is the peak current, so equation (27) is simplified to equation (28);
[0187] Peak value of resonant current
[0188] Specific implementation method 8: This implementation method is a further description of the soft switch design method of the push-pull resonant converter described in implementation method 1. The method for selecting the switch tube in the push-pull resonant converter and determining the number of switch tubes and the size of parasitic capacitance in step 7 is:
[0189] The peak value of the resonant current flowing through the secondary resonant cavity is obtained by equation (28). The peak value of the resonant current on the secondary side is equivalent to the sum of the value of the primary side and the transformer inductance current, that is, Ni L_max +i M_P , as the basis for selecting the switch tube and determining the number of switch tubes n;
[0190] Because the switch tube has a limit current that it can withstand, once the current flowing through the switch tube exceeds the limit current, the switch tube is at risk of burning out; for the boost resonant circuit, the primary current stress flowing through the switch tube is very large, and a single switch tube may not be able to withstand the secondary side equivalent to the primary side resonant peak current, so it is necessary to use n switch tubes in parallel to reduce the current stress and conduction loss flowing through each switch tube;
[0191] The parasitic capacitance C of a single switch tube is obtained from the switch tube selection data sheet. oss .
[0192] Specific implementation method 9: This implementation method is a further description of the soft switch design method of the push-pull resonant converter described in implementation method 1. In step 8, according to The method to calculate the magnetizing inductance and leakage inductance required for zero voltage switching is:
[0193] Through the inductor volt-second balance, we can get:
[0194]
[0195] From formula (29), we can get the transformer inductor current peak value i after the switch is turned off: M_P ,in is the time when the primary switch is turned on, T d is the dead time in one cycle, L m is the transformer’s magnetizing inductance;
[0196] Since the voltage of the switch tube when it is turned off is 2V in , so the total amount of charge stored in the parasitic capacitance of the switch tube is q total As shown in formula (30):
[0197]
[0198] Substituting equation (29) into the charge equation (30) and simplifying it, we can obtain equation (31);
[0199]
[0200] In order to ensure that the dead time can achieve zero voltage switching characteristics, it is known from equation (29) that the sum of the primary excitation inductance and leakage inductance needs to be less than or equal to the theoretical value to obtain sufficient i M_P , then equation (31) evolves into inequality (32);
[0201]
[0202] Once the restriction of inequality (32) is met, the push-pull resonant converter can utilize the non-mutational characteristic of the excitation inductor current during the preset dead time to charge the parasitic capacitance in the switch tube that is about to be turned off through the freewheeling process of the excitation current, and discharge the parasitic capacitance in the switch tube that is about to be turned on, so as to achieve zero voltage switching characteristics and realize zero voltage shutdown and zero voltage turn-on.
[0203] The soft switch design method of a push-pull resonant converter relates to the field of resonant DC converters. Firstly, by determining the maximum value of the quality factor Q of the resonant cavity, it is ensured that the resonant current reaches zero before the switch tube is turned off, and then the resonance is stopped, and no reverse current is generated in the switch tube. At this time, the switch tube will not generate additional turn-off loss, and increase the turn-off voltage ringing peak; secondly, by designing the maximum value of the equivalent inductance (the sum of the excitation inductance and the leakage inductance) of the primary transformer, the minimum excitation current required to achieve zero voltage switching is obtained, and the charge in the parasitic capacitance of the switch tube is transferred to the parasitic capacitance in the switch tube of the other winding of the transformer through the excitation current, so that the switch tube obtains zero voltage switching and zero voltage switching characteristics within the dead time. At this time, the transformer inductance is equivalent to a current source, so as to achieve the effect of eliminating the turn-off voltage ringing on the switch tube. The peak efficiency of the optimized push-pull resonant converter is as high as 98.3%, avoiding the sub-resonant current phenomenon of the resonant cavity under low-frequency operation, and solving the turn-off voltage ringing caused by the hard switching of the switch tube within the dead time, providing zero voltage switching and zero voltage switching of the push-pull resonant converter.
[0204] Figure 1 is a flow chart of the method described in the present invention; Figure 2 1 is a schematic diagram of the structure of the circuit of the present invention, from which the structure included in the circuit can be seen; Figure 3 This is the working waveform of the push-pull resonant converter. It can be seen that the converter is turned on with zero voltage and turned off with zero voltage. Figure 4 This is the modal equivalent circuit diagram of the push-pull resonant converter, from which four modes can be seen; Figure 5 This is the experimental waveform diagram of the transformer primary switch not having zero voltage turn-on, in which the terminal voltage Vds of the switch tube produces a large voltage ringing, and does not achieve zero voltage turn-on, resulting in large losses; Figure 6 This is the experimental waveform of the transformer primary switch with zero voltage turn-on. It can be seen from the figure that after the zero voltage switch design method, the ringing disappears and the zero voltage turn-on is achieved, which reduces the ringing phenomenon generated when the switch tube is turned off, thereby reducing the probability of the switch tube being broken down; the zero voltage switch solves the problem of additional turn-off loss when the switching frequency is lower than the resonant frequency; Figure 7 It is the resonant current to zero turn-off waveform of the primary switch tube of the transformer under a smaller Q value; Figure 8 It is the negative half-wave current turn-off waveform of the primary switch tube of the transformer under a larger Q value; Fig. 9 This is the effect of the dead time of zero voltage switching on the gain; Fig.10 It is the equivalent circuit diagram of the primary side to the secondary side; Figure 10 The first is a schematic diagram of the complex frequency domain model. Fig.12 It is a schematic diagram of the secondary side AC equivalent model.
[0205] The present invention is described by several specific embodiments, and it should be understood by those skilled in the art that various changes and equivalent substitutions may be made to the present invention without departing from the scope of the present invention. In addition, various modifications may be made to the present invention for specific situations or specific circumstances without departing from the scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed, but should include all implementation methods falling within the scope of the claims of the present invention.
Claims
1. A soft switching design method for a push-pull resonant converter, characterized in that: The method comprises the following steps: Step 1: Determine the resonant frequency, rated power, input and output rated voltage; Step 2: Calculate the transformer turns ratio, load resistance and output current; Step 3: The dead time of the push-pull resonant converter realizes zero voltage switching, that is, the duty cycle is set to 0.47~0.49, and the switching frequency f is calculated from this s Slightly below the resonant frequency f r , f s =0.94f r ~0.98f r ; Step 4: Calculate the maximum quality factor Q required to avoid sub-resonant mode based on expression (24) and the normalized switching frequency expression (15): max ; Step 5: Basis and Design equivalent resonant inductance and resonant capacitance to ensure Q < Q max , and contains a certain margin; Step 6: Pass Calculate the peak value of the resonant current; Step 7: Select the switch tube in the push-pull resonant converter and determine the number of switch tubes and the size of parasitic capacitance; Step 8: Basis Calculate the magnetizing inductance and leakage inductance required for zero voltage switching.
2. The soft switching design method of a push-pull resonant converter according to claim 1, characterized in that: The method for determining the resonant frequency, rated power, input and output rated voltage in step 1 is: Due to the limitations of power, heat dissipation, output ripple, electromagnetic compatibility (EMC) and actual transformer volume, if the rated power is above 300W and the input voltage is low, in order to ensure actual requirements such as heat dissipation, the resonant frequency is usually designed to be 100kHz~200kHz; The rated input voltage is determined by the voltage of the preceding power supply, such as the number of cells connected in series in a battery pack; The rated power and output voltage are determined by specific requirements. For the DC converter in the front stage of the vehicle-mounted boost inverter, the output voltage is at least 380V and the rated power is above 1.5kW.
3. The soft switching design method of a push-pull resonant converter according to claim 1, characterized in that: The method for calculating the transformer turns ratio, load resistance and output current in step 2 is: Transformer turns ratio Output Current Load resistance Where V0 is the rated output voltage, V in is the rated input voltage, P0 is the rated power, I d is the rated output current.
4. The soft switching design method of a push-pull resonant converter according to claim 1, characterized in that: The switching frequency f is calculated in step 3. s Slightly below the resonant frequency f r , f s =0.94f r ~0.98f r The method is: Since dead time is required to achieve zero voltage switching, a duty cycle D less than 0.5 needs to be designed. The duty cycle is the ratio of the conduction time of the switch tube on a certain bridge arm per unit cycle to the entire signal cycle time. The expression of the duty cycle D is as follows: Where T on is the on-time, T d is the dead time; since the realization of zero voltage switching requires the existence of dead time, it can be seen from formula (4) that the duty cycle D must be less than 0.5; where f s is the switching frequency, f on is the conduction frequency without taking into account the dead time; thus we get f s =2Df on (6) And because the DC value of the resonant inductor current in half a switching cycle is: in is the peak value of the secondary resonant current. Substituting equation (5) into equation (7), we get: From equation (8), we can see that the reduction of D will cause the resonant current peak The voltage and current stress of each component will increase accordingly, which will lead to an increase in cost. Therefore, the duty cycle D is selected to be 0.47-0.49, that is, f s =0.94f r ~0.98f r ; At this time, the resonant peak current increases by 0.01 to 0.03 times compared with the resonant current peak value with a duty cycle of 0.
5.
5. The soft switching design method of a push-pull resonant converter according to claim 1, characterized in that: In step 4, the maximum quality factor Q required to avoid sub-resonance mode is calculated according to equation (24) and the normalized switching frequency equation (15): max The method is: F is the normalized switching frequency, 6. The soft switching design method of a push-pull resonant converter according to claim 1, characterized in that: In step five, and Design equivalent resonant inductance and resonant capacitance to ensure Q < Q max , and the method that includes a certain margin is: (a) Due to the symmetry of the resonant waveform, the analysis and derivation of half the switching cycle of the push-pull resonant converter yields: The Q (quality factor) composed of the secondary LC resonant cavity and the AC equivalent resistance, AC equivalent resistance Equivalent resonant inductance L' r =N 2 L σ1 +L σ2 +L r (11) Among them, R e is the AC equivalent resistance, R o is the load resistance; L' r is the equivalent resonant inductor, which is composed of the primary leakage inductance L σ1 , secondary leakage inductance L σ2 and resonant inductor L r Composition, C r is the resonant capacitor; (b) According to the relationship between the capacitor voltage and current (12), the resonant capacitor voltage value u at any time τ is obtained: Cr (t), Among them, is the initial value of the resonant capacitor voltage, and τ is any time; (c) Assume that the initial value of the capacitor voltage is Then the resonant capacitor voltage reaches its maximum value after half a resonant cycle. Since the capacitor voltage is a sinusoidal symmetrical waveform, Substituting it into equation (12) yields the maximum value of the resonant capacitor voltage expression (13), where T r is the resonant cycle time, that is, T r =1 / f r Since only half of the resonant current of the resonant cycle flows through the resonant capacitor in half a switching cycle, τ at any time in equation (12) is T r / 2. It should be noted that in order to ensure that the resonant current reaches zero when the switch is turned off, the switching cycle time needs to be greater than or equal to the resonant cycle time, that is, T r ≤T s ; (d) The DC value of the resonant inductor current in half a switching cycle is obtained from equation (7): d , Substituting equation (7) into equation (13) and simplifying it, we get equation (14); where T s is the unit switching cycle time, that is, T s =1 / f s , F is the normalized switching frequency taking into account the dead time, that is, formula (15), where f r is the resonant frequency, that is, equation (17), and f s is the switching frequency, V0 is the rated output voltage; F' is the normalized switching frequency without taking into account the dead zone, that is, formula (16), where f s ' is the ideal switching frequency without dead time. At this time, the push-pull resonant converter waveform has T d Dead time, hence the switching frequency f s Less than the resonant frequency f r ; However, the on-time of the switch tube remains unchanged, so the actual gain is still unity gain; but for the traditional gain expression, since D < 0.5, the traditional theoretical gain is already less than unity gain, so it is necessary to introduce the ideal switching frequency f without dead time s 'and the normalized switching frequency F' excluding the dead zone; (e) Considering the existence of dead time, the series resonant gain M of this push-pull resonant converter is different from the traditional series gain. The series gain of this converter can be calculated as formula (22), where D is the duty cycle. It should be noted that D < 0.5; According to the secondary side AC equivalent model, the voltage division expression can be obtained: where ω s is the switching angular frequency, defined as ω s =2πf s , j is the imaginary unit, Nv in (jω s ) is the square wave voltage output by the primary full bridge and equivalent to the secondary side, v o (jω s ) is the square wave voltage of the secondary full-bridge input. After simplifying expression (18), and combining expression (9) and expression (16), we get expression (19): Taking the modulus of expression (19) yields the expression for gain M: Considering the influence of dead time on the resonant circuit gain, the relationship between the normalized frequency F' without taking the dead time into account and the normalized frequency F taking the dead time into account is obtained as formula (21): F'=F+1-2D (21) Substituting equation (21) into gain expression (20) yields equation (22); (f) The critical condition for the push-pull resonant converter to generate sub-resonant current is given by equation (23); From equations (13) and (23), it can be seen that once the resonant inductor current is too large, the voltage on the resonant capacitor will exceed the sum of the input voltage on the secondary side of the transformer and the output voltage on the load side. This will cause the rectifier bridge to turn on again, and the resonant capacitor will discharge along the loop to the load side, generating a negative half-wave current and entering the sub-resonant mode. (h) Substituting equation (14) and equation (22) into equation (23), we can simplify equation (24); Since parasitic parameters such as parasitic inductance in the actual circuit will affect the Q value of the resonant cavity, there is a certain deviation from the theoretical Q value. Therefore, a margin is left in the actual design. For example, when a double margin is reserved, the Q value should be less than Q max / 2 to ensure that the DC converter does not enter sub-resonant mode; Combining equations (9) and (17) to obtain the equivalent resonant inductance L' r and resonant capacitor C r The expression is:
7. The soft switching design method of a push-pull resonant converter according to claim 1, characterized in that: In step six, The method to calculate the peak value of the resonant current is: Substituting the design parameter duty cycle D and the normalized frequency F including the dead time into equation (24), the critical Q value of the sub-resonance mode can be obtained. The circuit can be prevented from entering the sub-resonance mode by designing the equivalent resonant inductor, resonant capacitor and load resistor. The resonant current expression (27) is obtained through the complex frequency domain model of the circuit, Formula (27) is used to calculate the resonant current i Lr (t) is the peak current; where I L (0 - ) is the initial value of the inductor current, V c (0 - ) is the initial value of the capacitor voltage, ω r is the resonant angular frequency; Since the resonant current of the push-pull resonant converter needs to be designed as a complete half-wave to achieve the soft switching characteristics, the resonant current always reaches zero within half a switching cycle, that is, I L (0 - )=0; According to the resonant sine waveform, when When , the resonant current is the peak current, so equation (27) is simplified to equation (28); Peak value of resonant current 8. The zero voltage switching design method of a push-pull resonant converter according to claim 1, characterized in that: The method for selecting the switch tubes in the push-pull resonant converter and determining the number of switch tubes and the size of parasitic capacitance in step seven is: The peak value of the resonant current flowing through the secondary resonant cavity is obtained by equation (28). The peak value of the resonant current on the secondary side is equivalent to the sum of the value of the primary side and the transformer inductance current, that is, As the basis for selecting the switch tube and determining the number n of switch tubes; Because the switch tube has a limit current that it can withstand, once the current flowing through the switch tube exceeds the limit current, the switch tube is at risk of burning out; For the boost resonant circuit, the primary current stress flowing through the switch tube is very large, and a single switch tube may not be able to withstand the secondary side equivalent to the primary side resonant peak current. Therefore, it is necessary to use n switch tubes in parallel to reduce the current stress and conduction loss flowing through each switch tube. The parasitic capacitance C of a single switch tube is obtained from the switch tube selection data sheet. oss .
9. The soft switching design method of a push-pull resonant converter according to claim 1, characterized in that: In step eight, The method of calculating the magnetizing inductance and leakage inductance required for zero voltage switching using equation (32) is: Through the inductor volt-second balance, we can get: From formula (29), we can get the transformer inductor current peak value i after the switch is turned off: M_P ,in is the time when the primary switch is turned on, T d is the dead time in one cycle, L m is the transformer’s magnetizing inductance; Since the voltage of the switch tube when it is turned off is 2V in , so the total amount of charge stored in the parasitic capacitance of the switch tube is q total As shown in formula (30): Substituting equation (29) into the charge equation (30) and simplifying it, we can obtain equation (31); In order to ensure that the dead time can achieve zero voltage switching characteristics, it is known from equation (29) that the sum of the primary excitation inductance and leakage inductance needs to be less than or equal to the theoretical value to obtain sufficient i M_P , then equation (31) evolves into inequality (32); Once the restriction of inequality (32) is met, the push-pull resonant converter can utilize the non-mutational characteristic of the excitation inductor current during the preset dead time to charge the parasitic capacitance in the switch tube that is about to be turned off through the freewheeling process of the excitation current, and discharge the parasitic capacitance in the switch tube that is about to be turned on, so as to achieve zero voltage switching characteristics and realize zero voltage shutdown and zero voltage turn-on.
Citation Information
Cited By
Excitation inductor design method for guaranteeing soft switching of CLLC resonant converter
CN120090453A
Direct current (DC) -DC converter, audio device, vehicle and control method
WO2026174543A1
