Forward power supply--resonant demagnetization technology

The single-ended forward switching power supply uses only one power switch tube, and the overall circuit structure is relatively simple, and has been widely used in the occasion of small and medium power output. However, the characteristic of this topology is that the power transformer works in the first quadrant of the BH curve. The potential hazard of the core saturation of the transformer must be properly demagnetized. The magnetizing energy stored in the power transformer when the switch is turned on is cut off. Evacuate or consume during the period. Otherwise, after a plurality of switching cycles, due to the remanence, the operating point of the transformer gradually moves up, and it is easy to generate an approximate short-circuit state due to saturation of the core, resulting in a large current flowing through the power switch tube, which is burned out beyond its rated value. .

The classical demagnetization methods commonly used in engineering include adding demagnetization winding, active clamping, RCD clamping method, ZVT clamping method, etc. The common idea is: after the main power switch is cut off, through a certain way, the transformer is made. The remaining magnetization energy is vented or consumed on the passive power resistor.

In fact, since the current switching power supply generally adopts a MOSFET as a power switch, the demagnetization operation can be better accomplished by using only its distribution parameters, that is, demagnetization is performed by using a resonance technique. The basic principle of resonant demagnetization is: after the power switch is turned off, the self-inductance of the transformer and the distributed capacitance of the components in the circuit are used to resonate, and the magnetization energy of the transformer is transferred. In this way, the relatively complicated demagnetization design is omitted, and the circuit structure is simplified.

Before analyzing the working principle of transformer demagnetization using resonance technology, the following assumptions are made:

(1) The entire system is in a state of dynamic equilibrium.

(2) The output inductance LO and the output capacitance CO are approximately infinite compared to the distributed components participating in the resonance.

(3) The leakage inductance of the transformer is negligible.

(4) Both the switching transistor and the diode are ideal devices, that is, the on-resistance of the switching transistor and the forward voltage drop of the diode can be approximately considered to be zero.

(5) The transition time of the switching device is short compared to the switching period and the resonant operating time.

For a single-ended forward power supply, the basic circuit component distribution associated with the resonant demagnetization method is shown in Figure 1:

Where Lm is the equivalent inductance of the primary coil of the transformer; Ct is the equivalent capacitance of the primary winding of the power transformer, and is connected in parallel with Lm; Cs is the drain-source junction capacitance of the switching transistor Q1 and paralleled to improve its switching environment The sum of the external capacitors; C1 is the junction capacitance of the output rectifier diode.

Figure 2 is a schematic diagram of the equivalent of these components to the primary of the transformer. As can be seen from the figure, the junction capacitance C1 of Dr is equivalent to the capacitance C2 of the transformer primary:

And it is in parallel with Ct. At the same time, assuming that the input voltage source Vin is an ideal voltage source, its internal resistance is negligible, so in the AC resonant state, Cs is also in parallel with Ct.

During a complete switching cycle, the entire process of resonant demagnetization consists of the following phases:

The first stage: the T1 phase in Figure 3. Prior to this, Q1 is in an off state, the voltage on the drain-source is the input voltage Vin, Df is continuously turned on, and the magnetizing current flowing through the transformer core is a negative value I1 (the magnitude and direction are further explained later). Starting from t=0, Q1 is controlled to conduct, and the magnetizing current Imag of the core of the main power transformer changes linearly, gradually changing from a negative value to 0, and starts to increase in the forward direction. At this stage, due to the polarity relationship, Dr is turned on and Df is turned off. The terminal voltages of C1 and Cs are both approximately zero, and the energy is coupled from the input to the output load through the transformer. Assuming that the transformer primary magnetizing current is I1 at the beginning of this phase and I2 at the end, the relationship between I1 and I2 is:

The second stage: the T2 stage in Figure 4. At the beginning of this phase, Q1 is turned off by the action of the control signal, and its drain-source voltage Vds starts to rise rapidly. When Vds exceeds the input voltage Vin, the polarity of the secondary winding of the transformer is reversed, Dr is turned off, and Df is turned on. Due to the cutoff of Q1, the transformer primary inductance Lm forms a parallel resonant circuit with the equivalent capacitance Cr (the sum of C2, Ct, Cs) in the circuit, and starts the resonance operation. The demagnetizing current Imag starts to change in a sinusoidal shape and flows through the resonant circuit. . It can be known from the circuit theory that when an LC parallel circuit operates in a resonant mode, the current on the inductor and the voltage on the capacitor are both sinusoidal and are 90 degrees out of phase with each other. The energy stored in the inductor and capacitor participating in the resonance is mutually exchange. Since the terminal voltage of Cr in the previous stage is 0, there is no stored energy, and the energy in Lm reaches the maximum before the switch is turned off, so Lm and Cr generate energy exchange; the duration of this phase is T2, and T2 is one Half of the full resonant period.

The maximum voltage that can be reached by the voltage on Cr is:

The Q1 drain-source voltage Vds also reaches a maximum when Cr reaches a maximum value:

Thus, at the end of this phase, the exciting current Imag reaches the maximum value in the negative direction. Since the system is in a stable dynamic equilibrium state and can be completely demagnetized, its value is equal to -I2. At this time, the Q1 drain-source voltage Vds is equal to the input voltage Vin.

The equivalent capacitance Cr at this stage is:

The resonant frequency is:

From the initial conditions, the changes in magnetizing current and equivalent capacitor voltage can be obtained as follows:

In the above two stages, the change in the magnetic field strength H in the transformer coincides with the change in the magnetizing current Imag: in the T1 phase, H increases in the positive direction; and in the next T2 phase, H changes in the opposite direction due to the resonance. In this way, the excitation energy of the transformer is transferred by resonance, and finally the reverse flow of the magnetization current is realized, thereby achieving the purpose of demagnetization.

The third stage: the T3 stage in Figure 5. During this period of time, Q1 remains in the off state. Since the voltage on Cr in the previous stage changes resonantly to 0, the voltage across Q1 is Vin. When the voltage on Cr attempts to continue to resonate and further decreases, Dr is turned on. Therefore, at the beginning of the time period, the terminal voltages of Np and Ns are both 0, the terminal voltage of Cr is clamped to 0, and the resonance ends. At this time, there is no alternating AC voltage across Cs in parallel with Q1, only stable DC. Voltage Vin. Both Dr and Df can be considered to be in a "on" state. The negative magnetization current can continue to flow because only one channel such as Df-Dr-Ns, and the magnetizing current I1 maintains a constant negative value I1 at this stage. This mode of operation continues until the next switching cycle. . When the system is in a stable working state and the demagnetization is ensured for each switching cycle, the magnetizing current I1 is also equal to I1 at the beginning of the next switching cycle, namely:

If the resonant frequency of the circuit is exactly equal to the time the switch is turned off, the duration of Ts is zero. However, if the resonance period is greater than Tr, there may be cases where I1 and I2 are not equal. In this case, the first half of the resonance period before the start of the next switching cycle does not end, so the drain-source voltage on the main power switch exceeds Vin at the beginning of each switching cycle; thus, the switching loss is increased. At the same time, it is impossible to effectively achieve complete demagnetization of the transformer.

1 Reduces the 50% duty cycle requirement for the control circuit. The single-ended forward switching power supply is usually demagnetized by adding a third winding to the main transformer in actual engineering. Due to the limitation of the withstand voltage of the switching tube, the ratio of the demagnetization winding to the primary winding is usually set to 1:1. In this way, the maximum duty cycle can only reach 50%. Simultaneously. In order to reduce the voltage spike when the switch is turned off, the reset winding and the primary winding require close coupling in the process, so the design and processing process of the transformer is complicated. Resonance demagnetization only requires that at least half of the resonant period work be completed during the cut-off period of the switching tube. This can be fully ensured by the selection of the resonant frequency and the adjustment of the parameters of the resonant element. Thus, the duty cycle is no longer required by 50% and the power supply can operate over a wide range of input voltages. It also makes sense for simplifying the circuit structure.

2 Compared with the conventional demagnetization winding method and resonant demagnetization, it can be seen that in the conventional demagnetization winding method, the magnetizing current can always be regarded as a non-negative value, linearly increasing when the switch is turned on, and linearly decreasing when the switch is turned off. Therefore, its B-H characteristic is the first quadrant; and the magnetization current of the resonance demagnetization has a negative value for each period of time, and therefore belongs to the bidirectional magnetization current change. When selecting a large magnetic induction swing (ΔB) for power transformer design, resonant demagnetization has more advantages in preventing core saturation.

3 From the theoretical analysis and the actual waveform of the subsequent desktop circuit test, it can be seen that when the resonance demagnetizes, the drain-source voltage waveform of the switch tube is a relatively smooth half sine wave, and the demagnetization winding method is a pulse with a steep edge of the waveform. Square wave, the former undoubtedly has smaller harmonic components than the latter. Therefore, the EMI problem of the switching power supply has also been improved.

4 Resonance demagnetization, determination of resonant component parameters

In the case of resonant demagnetization, in order to ensure that the demagnetization process can be completed for half a resonance period during the switching off period, the parameters of the resonant element need to be carefully determined. Therefore, on the basis of theoretical analysis, it is necessary to carefully observe the drain-source waveform of the switching tube under various working conditions in the test to determine a suitable resonant frequency.

When selecting the resonant frequency, it is necessary to comprehensively consider the contradiction between the rated voltage and the demagnetization effect of the switching tube. At present, in medium and small power applications, the primary inductance of a single-ended forward power transformer is usually tens to hundreds of microhenries, and the junction capacitance of a switching transistor is usually several hundred to several thousand pF, so that only the primary is utilized. When the inductor and the device junction capacitance are resonantly demagnetized, the resonant frequency can generally reach several hundred kHz or higher. In order to reduce the voltage stress Vds of the main switch tube at the resonance, it is sometimes necessary to connect a certain amount of capacitance in parallel with the switch tube Q1 or the diode Dr to appropriately lower the resonance frequency. However, the capacitance of the capacitor should not be too large, otherwise the problem of resonance demagnetization could not be completely achieved.

Figure 6 is a typical voltage waveform of the drain-source of the switching transistor when different resonant parameters are selected under the same duty cycle conditions. In the picture:

(1) is the ideal voltage waveform after selecting a suitable resonant frequency, and its shape is consistent with the theoretical analysis of the previous section;

(2) is a waveform when the resonance frequency is selected too high. This waveform is the case when the equivalent capacitance Cr is small under the conditions in which the various parameters of the transformer are determined. It can be seen from the figure that although the basic shape is exactly the same as (1), the demagnetization process can be completed quickly, but since the equivalent capacitance is small, the resonance frequency is high, and the same transformer primary excitation energy causes equivalent capacitance. The magnitude of the resonant voltage V2 on Cr far exceeds V1. This requires the main switch tube to have a higher withstand voltage and increase the cost.

(3) The capacitance connected in parallel between the switch tube and the output diode is too large, resulting in the resonance equivalent capacitance Cr being too large, so the resonance frequency is low, and it is not even enough to completely satisfy the resonance period within the cut-off period of the switch tube. work. According to the previous analysis, when the system is dynamically balanced, the complete demagnetization condition is that the magnetizing current at the beginning of each switch's conduction period should be the same as the magnetizing current at the end of the last switching period. Obviously, the demagnetization process in (c) is not complete. At the end, this is especially true when the input voltage is low and the switch is turned on for a long time. In circuit design and testing, this waveform should be avoided as much as possible. It can also be seen from these figures that in engineering design, in order to reduce the loss of the power switch tube, the capacitance is connected in parallel at both ends, which has an effect on the effect of the resonance demagnetization, and therefore needs to be comprehensively considered.

Four design examples:

Under the guidance of the above theory, the desktop circuit test (12V/20W) of the actual single-ended forward power supply demagnetized by the resonance technique was carried out; and on this basis, the design of the initial sample of a certain type of product was completed. The basic principle is shown in Figure 7.

In this circuit, the control device selects UC1843 (LCC20 package); the switching frequency is set to nearly 300kHz, the maximum duty ratio is selected to be about 60%; the switching transistor Q1 is 2N6798 (IRF230), its Coss is 250pF; the rectifier diode Dr is 15CLQ100. The transformer core is RM6 core of MAGNETICS, the primary coil is 8 turns and the secondary coil is 9 turns. The inductance of the primary coil of the core is measured to be about 160μH, the secondary rectifier diode is not connected in parallel, and the primary MOSFET is connected in parallel with 510p capacitor; the input voltage range is 23V-33V. The measured Q1 drain-source waveform is at the lowest input voltage and the highest input voltage, as shown in Figure 7 (the abscissa in the two figures is the time amount 1μs/div; the ordinate is the voltage amount 20V/div) :

The operation of this core reset method can be seen by measuring the drain-source voltage waveform of the actual circuit power MOSFET. It can be roughly estimated from the figure that the resonance frequency at the time of demagnetization is about 300 kHz. The actual circuit parameter calculation is also roughly in this range. The power supply products of the actual design were subjected to long-term continuous power-on test under high and low temperature conditions respectively, and their working performance was stable, which proved the technical effectiveness of the method.

KNL5-63 Residual Current Circuit Breaker

KNL5-63 Moulded Case Circuit Breaker is MCCB , How to select good Molded Case Circuit Breaker suppliers? Korlen electric is your first choice. All moulded Case Circuit Breakers pass the CE.CB.SEMKO.SIRIM etc. Certificates.

Moulded Case Circuit Breaker /MCCB can be used to distribute electric power and protect power equipment against overload and short-current, and can change the circuit and start motor infrequently. The application of Moulded Case Circuit Breaker /MCCB is industrial.
Korlen electric also provide Miniature Circuit Breaker /MCB. Residual Current Circuit Breaker /RCCB. RCBO. Led light and so on .

KNL5-63 Molded Case Circuit Breaker,Small Size Molded Case Circuit Breaker,Electrical Molded Case Circuit Breaker,Automatic Molded Case Circuit Breaker

Wenzhou Korlen Electric Appliances Co., Ltd. , https://www.zjmotorstarter.com