Control principle of single-ended flyback switching power supply
Single-ended flyback switching power supply is a power supply circuit whose working principle is mainly based on the single-ended operation of the magnetic core. When the switch tube is turned on, the primary winding of the high-frequency transformer stores energy, and when the switch tube is turned off, the secondary winding releases the stored energy. This process allows the electric energy to be transferred from the primary winding to the load through the secondary winding and the rectifier diode.
Single-ended flyback switching power supply has the advantages of low cost, low power consumption, high efficiency and suitability for fixed loads. However, its output voltage ripple is large and it is not suitable for processing high-power electric energy. This power supply circuit is often used in occasions such as auxiliary power supply required by control systems.
Single-ended flyback switching power supply is similar to single-ended forward switching power supply in form, but the working conditions are different. When the switch tube is turned on, the rectifier diode is in the off state and the transformer stores energy; when the switch tube is turned off, the transformer releases energy to the load through the rectifier diode. Compared with the single-ended forward switching power supply, the transformer structure of the single-ended flyback switching power supply is more complex and larger in size, and is less practically used.
The working principle of the single-ended flyback switching power supply is mainly based on the single-ended operation of the magnetic core. When the switch tube is turned on, the primary winding of the high-frequency transformer stores energy, and when the switch tube is turned off, the secondary winding releases the stored energy. This process allows the electric energy to be transferred from the primary winding to the load through the secondary winding and the rectifier diode.
The single-ended flyback switching power supply has the advantages of low cost, low power consumption, high efficiency and suitability for fixed loads. However, its output voltage ripple is large and it is not suitable for processing high-power electric energy.
Considering the power of 10W and the small size, the circuit uses a single-ended flyback circuit. The characteristics of the single-ended flyback circuit are: simple circuit, small size and low cost. The single-ended flyback circuit consists of an input filter circuit, a pulse width modulation circuit, a power transfer circuit (composed of a switch tube and a transformer), an output rectifier filter circuit, an error detection circuit (composed of a chip TL431 and surrounding components) and a signal transmission circuit (composed of an isolation optocoupler and a resistor). This power supply is designed as a surface-mounted module power supply, and its specific parameter requirements are as follows:
Maximum output power: 10W
Input AC voltage: 85~265V
Output DC voltage/current: +5V, 500mA; +12V, 150mA; +24V, 100mA
Ripple voltage: ≤120mV
Control principle of single-ended flyback switching power supply
The so-called single-ended means that the TOPSwitch-II series device has only one pulse modulation signal power output terminal, the drain D. Flyback means that when the power MOSFET is turned on, the electric energy is stored in the primary winding of the high-frequency transformer, and only when the MOSFET is turned off, the electric energy is transmitted to the secondary. Since the switching frequency is as high as 100kHz, the high-frequency transformer can quickly store and release energy, and a DC continuous output can be obtained after high-frequency rectification and filtering. This is also the basic working principle of the flyback circuit. The feedback loop adjusts the duty cycle by controlling the current at the control end of the TOPSwitch device to achieve the purpose of voltage regulation.
Selection and introduction of TOPSwitch-Ⅱ series chips
The drain (D) of the TOPSwitch-Ⅱ series chip is connected to the internal power switch device MOSFET, and is connected to the main power supply through the load inductor. In the startup state, the internal switch-mode high-voltage power supply provides internal bias current and is equipped with current detection. The control electrode (C) is used for the input pin of the error amplifier and feedback current for duty cycle control. It is connected to the internal parallel regulator to provide internal bias current during normal operation. It is also the capacitor connection point for bypass, automatic restart and compensation functions. The source (S) is connected to the source of the MOSFET of the high-voltage power circuit and serves as the common point and reference point of the primary circuit. The duty cycle of the internal output MOSFET decreases linearly with the increase of the control pin current. The typical value of the control voltage is 5.7 V, the limit voltage is 9 V, and the maximum allowable current of the control terminal is 100 mA.
Temperature compensation measures are also taken for the threshold voltage during design to eliminate the change of drain current caused by the change of drain-source on-resistance with temperature. When the junction temperature of the chip is greater than 135℃, the overheat protection circuit outputs a high level and turns off the output pole. At this time, the control voltage Vc enters the hysteresis regulation mode, and the waveform at the Vc end also becomes a sawtooth wave with an amplitude of 4.7V to 5.7V. To restart the circuit, you need to turn off the power and then turn on the circuit switch, or reduce Vc to below 3.3V, and then use the power-on reset circuit to reset the internal trigger to zero, so that the MOSFET can resume normal operation.
When using the TOPSwitch-Ⅱ series to design a single-chip switching power supply, fewer external components are required, and the device is much less sensitive to the circuit board layout and input bus transients, so the design is very convenient, the performance is stable, and the cost performance is higher.
For the selection of chips, the input voltage and power are mainly considered. From the design requirements, it can be seen that the input voltage is a wide range input and the output power is not more than 10W, so TOP222G is selected.
Circuit design
The schematic diagram of this switching power supply is shown in Figure 1.
The main circuit of the power supply is flyback type. C1, L1, and C2 are connected to the AC power supply input end to filter out grid interference. C5 is connected between the high voltage and the ground to filter out the common mode interference generated by the primary and secondary of the high-frequency transformer and the capacitor. It is called "Y capacitor" in international standards. C1 and C5 are both called safety capacitors, but C1 specifically filters out the series mode interference between the grid lines and is called "X capacitor".
In order to withstand the electric shock that may come from the grid line, a varistor VSR with a nominal voltage u1mA of 275V can be connected in parallel at the AC end.
In view of the fact that at the moment when the power MOSFET is turned off, the leakage inductance of the high-frequency transformer generates a peak voltage UL. In addition, an induced reverse electromotive force UOR is generated on the primary side, and the two are superimposed on the DC input voltage. In typical cases, after the AC input voltage is rectified by the rectifier bridge, its maximum voltage UImax=380V, UL≈165V, UOR=135V, and UOR+UL+UOR≈680V. This requires that the power MOSFET can withstand at least 700V of high voltage. At the same time, a clamping circuit must be added to the drain to absorb the peak voltage and protect the power MOSFET in TOP222G. The clamping circuit of this power supply consists of D2 and D3. D2 is a transient voltage suppressor (TVS) P6KE200, and D3 is an ultra-fast recovery diode UF4005. When the MOSFET is turned on, the upper end of the primary voltage is positive and the lower end is negative, so that D3 is cut off and the clamping circuit does not work. At the moment when the MOSFET is turned off, the primary voltage becomes positive at the lower end and negative at the upper end. At this time, D1 is turned on and the voltage is limited to about 200V.
Output link design
Take the +5V output link as an example. The high-frequency voltage on the secondary coil passes through the UF5401 type 100V/3A ultra-fast recovery diode D7. Since the +5V output power is relatively large, a post-stage LC filter is added to reduce the output ripple voltage. The filter inductor L2 uses a 3.3μH through-hole inductor called a "magnetic bead" to filter out the switching noise generated by D7 during the reverse recovery process.
For the other two outputs, only filter capacitors need to be added to the output ends. R3 and R4 are the dummy loads of the outputs, which can reduce the no-load and light-load voltages of their respective output ends.
Feedback link design
The feedback path is mainly composed of PC817 and TL431 and several capacitors and resistors. U2 is TL431, which is an adjustable precision shunt regulator. The reference voltage value is obtained by dividing the voltage with resistors R5 and R6. The output voltage regulation value can be adjusted by adjusting the values of R5 and R6. C8 is the frequency compensation capacitor of TL431, which can improve the transient frequency response of TL43l. C7 is a soft start capacitor. When C7=22μF, the soft start time can be increased by 4ms. When the TOP222G itself has a 10ms soft start time, the total is 14ms.
U3 is a PC817 linear optocoupler with a current transfer ratio (CTR) range of 80% to 160%, which can better meet the design requirements of the feedback loop. However, the 4N25 and 4N26 commonly used in China are nonlinear optocouplers and should not be used. The voltage generated on the feedback winding is rectified and filtered by D4 and C9 to obtain a non-isolated +12V output to power the collector of the PC817 receiving tube. Since the output current of the feedback winding is small, the secondary uses the D4 silicon high-speed switch tube 1N4148. The optocoupler PC817 can isolate the +5V output from the power grid, and its emitter current is sent to the control end of TOP222G to adjust the duty cycle.
C3 is a control-end bypass capacitor, which can compensate the control loop and set the automatic restart frequency. When C3=47μF, the automatic restart frequency is 1.2Hz, that is, it detects whether the out-of-control fault has been eliminated every 0.83s. If it is confirmed that it has been eliminated, the switching power supply will automatically restart to resume normal operation.
R2 is the external current limiting resistor of the LED in PC817. In fact, in addition to the current limiting protection function, it also has an important influence on the gain of the control loop. When R2 changes, it will affect the following parameter values in turn: IF→IC→D→UO, which is equivalent to changing the current amplification factor of the control loop.
The following briefly analyzes the working principle of the feedback loop to achieve voltage regulation. When the output voltage UO fluctuates and the change is UO, after the sampling resistors R5 and R6 divide the voltage, the output voltage UK of TL431 will also change accordingly, thereby changing the working current IF of the LED in PC817. Finally, the duty cycle D is adjusted by the change in the control terminal current IC, so that UO changes in the opposite direction, thereby offsetting the fluctuation of UO. The above voltage regulation process can be summarized as:
UO ↑→UK ↓→IF ↑→IC ↑→D ↓→UO↓→Finally, UO remains unchanged.
The other outputs are not fed back, and the output voltage is determined by the number of turns of the high-frequency transformer.
Transformer design
The design of the transformer is the key to the entire power supply design, and its quality directly affects the power supply performance.
Determination of the magnetic core and skeleton
Since this article uses enameled wire winding, and the EE type magnetic core is cheap, has low magnetic loss and strong adaptability, EE22 is selected, and its magnetic core length A=22mm. From the magnetic core product manual provided by the manufacturer, it can be found that the effective cross-sectional area of the magnetic core SJ=0.41cm2, the effective magnetic path length 1=3.96cm, the magnetic core equivalent inductance AL=2.4μH/turn2, and the skeleton width b=8.43mm.
Determine the maximum duty cycle Dmax
According to the formula:
Where, UOR=135V, the minimum DC input voltage UImin=90V, the drain-source on-state voltage UDS(ON) of the MOSFET=10V, substitute into the above formula to get: Dmax=64.3%, close to the typical value of 67%. Dmax decreases with the increase of input voltage.
Calculate the current in the primary coil
The average value of the input current IAVG is:
The primary peak current IP is:
Wherein, KRP is the ratio of the primary ripple current IR to the primary peak current IP. When the voltage is input in a wide range, it can be taken as 0.9. Substituting Dmax=64.3% into IP=0.518A.
Determine the primary winding inductance LP
Wherein, the loss distribution coefficient Z=0.5, IP=0.518A, KRP=0.4, PO=10W, substituting into: LP≈1265μH.
Determine the winding method
And calculate the number of turns of each winding
The number of turns NP of the primary winding can be calculated by the following formula:
Among them, the core cross-sectional area SJ=0.41cm2, the core maximum flux density BM=60, IP=0.518A, LP≈1265μH, and NP=26.6 can be obtained by substitution, and 30 turns are actually taken.
The secondary winding adopts stacking winding, which is also a method often used by transformer manufacturers. Its characteristic is that the 5V winding provides part of the turns for the 12V winding, while the 24V winding includes the 5V, 12V windings and the newly added turns. The stacking winding technology is advanced, which can not only save wires and reduce the coil volume, but also increase the mutual inductance between windings and strengthen the coupling degree. Taking this power supply as an example, when the 5V output is fully loaded and the 12V and 24V outputs are lightly loaded, since the 5V winding also serves as part of the 12V and 24V windings, the leakage inductance of these windings can be reduced, which can prevent the filter capacitors in the 12V and 24V output circuits from being charged to the peak value by the peak voltage due to leakage inductance, that is, the so-called peak charging effect, which causes the output voltage to be unstable. Here, the 5V winding is used as the starting point of the secondary.
For multi-output high-frequency transformers, the number of turns of each output winding can be the same number of turns per volt. The number of turns per volt nO can be determined by the following formula:
The unit is turns/VO. Substituting NS into 5 turns, UO1=5V, UF1=0.4V (Schottky rectifier conduction voltage drop) into the above formula, we get nO=0.925 turns/V.
For 24V output, if UO2=24V and UF2=0.4V are known, the number of turns of the output winding is NS2=0.925 turns/V×(24V+0.4V)=22.57 turns, and 22 turns are actually taken.
For 12V output, if UO3=12V and UF2=0.4V are known, the number of turns of the output winding is NS2=0.925 turns/V×(12V+0.4V)=11.47 turns, and 11 turns are actually taken.
For the feedback winding, if UF=12V and UF3=0.7V (the conduction voltage drop of the silicon fast recovery rectifier diode), the number of turns of the output winding is NS2=0.925 turns/V×(12V+0.4V)=11.47 turns, and 11 turns are actually taken.
Determine the inner diameter of the primary/secondary conductor
First, according to the number of primary layers d, the skeleton width b and the safety margin M, use the following formula to calculate the effective skeleton width bE (in mm):
bE=d(b-2M) (7)
Substituting d=2, b=8.43mm, and M=0 into the above formula, we get bE=16.86mm.
Use the following formula to calculate the outer diameter (with insulation layer) DPM of the primary conductor:
DPM=bE/NP (8)
Substituting bE=16.86mm and NP=78 turns, we get DPM=0.31mm. After deducting the paint thickness, the inner diameter of the bare conductor DPM=0.26mm. The metric wire gauge close to the diameter of 0.26mm is 0.28mm, which is slightly thicker than 0.26mm and can fully meet the requirements, while the metric wire gauge of 0.25mm is slightly thinner and should not be used. The secondary winding uses the same wire as the primary winding, and is wound in parallel with multiple strands according to the current.
Test data
The input characteristic data of the switching power supply is shown in Table 1. When the voltage regulation rate of the main output UO1=5V (load is 65Ω) is SV=±0.2% in the wide range of u=85～245V, and the maximum output ripple voltage is about 67mV; the auxiliary output UO2=24V (load is 250Ω), the maximum output ripple voltage is about 98mV; the auxiliary output UO3=12V (load is 100Ω), the maximum output ripple voltage is about 84mV.
FlyBack Converter is also called single-ended flyback converter, also known as flyback converter, because its output end obtains energy when the primary winding is disconnected from the power supply, hence the name. Electronic equipment all need power supply, and switching power supplies are widely used. The most widely used topology for small and medium power supplies is the flyback structure. Let's take some practical application examples, such as laptop adapters, mobile phone chargers, etc.
Advantages:
1. Simple circuit, low cost, high reliability, can provide multiple DC outputs;
2. When the input and output voltages fluctuate greatly, it can still output stably and can realize AC input;
3. The transformer turns ratio is small;
4. High conversion efficiency and low loss;
Disadvantages:
1. The output voltage ripple is large and the load adjustment accuracy is not high, so the output power is limited;
2. Working in CCM mode, there is a large DC component, which easily leads to transformer core saturation, so an air gap must be added in this circuit to cause the transformer;
3. The converter has a DC component and can work in two different modes, CCM/DCM, at the same time, which makes the converter design and loop compensation design more difficult;
Share our interesting knowledge and stories on social media
