Research on the development and intelligent application of power environmental protection platform based on big data

2.1 EMI circuit design and simulation

When the power of the driving power supply is higher than 25 W, the Electromagnetic Compatibility (EMC) standard mandates that the driving power supply must meet the requirements for electromagnetic compatibility in the relevant provisions of the standard. Conducted electromagnetic interference (EMI) in electromagnetic compatibility is one of them. Electromagnetic interference (EMI) is an undesired disturbance that affects an electrical circuit or path due to external sources, also known as radio frequency interference. EMI can result in poor performance, malfunction or complete failure of electronic devices. Both natural and human-made sources can cause EMI. When a metal wire is exposed to magnetic fields, it induces an electric current along the wire, which may interfere with network signals and potentially lead to network shutdown. An EMI filter is used in electronic circuits to reduce the amount of electromagnetic interference or noise that can affect the performance of the circuit. The filter is designed to block or attenuate unwanted noise by using a combination of inductors, capacitors, and resistors arranged in a specific configuration. It can be placed at the input or output of a device or circuit to improve the electromagnetic compatibility (EMC) of electronic devices. The EMI filter adopts a scheme of combining common mode (CM) inductors and differential mode (DM) inductors, and uses the leakage inductance of a common mode suppressing inductor to form a differential mode suppressing inductor, as shown in Figure 1. Common mode (CM) inductors and differential mode (DM) inductors are two types of inductors used in electronic circuits to filter out unwanted signals or noise. CM inductors are designed to block or attenuate noise that appears identically on both signal lines, while DM inductors attenuate noise that appears as a difference between the two signal lines. This structure saves the number of magnetic components, can meet the design requirements, and is cost-effective and easy to popularize. In Figure 1, L1a and L1b are the two coils of the common mode inductor, which are used to form a common mode filter, and L1a′ and L1b′ are the leakage inductances of the common mode inductor coil, which are used to form a differential mode filter. Differential mode filter and common mode filter are used in electronic circuits to filter out unwanted signals or noise. Differential mode filter attenuates noise that appears as a difference between two signal lines, while common mode filter attenuates noise that appears identically on both signal lines. The filters are designed using various components and configurations depending on the circuit’s specific application and requirements.

Figure 1:

Combined filter inductor of common mode and differential mode.

The slope of the inductor current waveform shown in Figure 2 is the current change rate di/dt. The change of the current will generate a changing magnetic field, which radiates outward. The larger the current change rate, the stronger the magnetic field and the more serious the electromagnetic radiation interference.

Figure 2:

PFC circuit and waveforms.

Figure 3 is the EMI conduction test result of EN55015 standard of Light-emitting diode (LED) power test without EMI Filter. From the test results, considering the 10 dB margin, the EMI filter needs to attenuate 22 dB at 200 KHz to meet the design requirements.

Figure 3:

EMI test results without EMI filter.

According to the table, the input impedance of the power supply can be calculated as:

The operating frequency of the PFC in the LED drive power supply is 70 KHz, the operating frequency of the resonant half-bridge is 35 KHz, and the corner frequency of the EMI Filter needs to be greater than 10 times the input AC frequency. According to the analysis of the above test results, 22 dB attenuation is required at 200 KHz.

Figure 4 is an effective circuit of a designed common mode filter.

Figure 4:

Equivalent circuit diagram of the second-order common mode filter.

The transfer function of this filter is:

Among them, ω is the angular frequency.

R L is the common mode noise load resistance.

The size of the ζ coefficient reflects the different gain of the second-order filter time domain response at the corner frequency. Figure 5 shows the relationship between them. The second order filter attenuates 12   d B per octave after the corner frequency.

Figure 5:

Second-order frequency response of different damping coefficients.

Therefore, the design goal of needing to attenuate 22   d B at 200   K H z requires that the corner frequency f c u t o f f of the filter be calculated from equation (3), which is 56.1   K H z .

R L is usually selected as 50   Ω . In order to not produce oscillation at the corner frequency, the minimum damping coefficient ζ is selected to be 0.707. According to the above parameters, the value of the component can be calculated as follows:

The maximum Y capacitance value specified by the E M C standard to meet the leakage current requirements is 0.05   μ F . For safety reasons, 6.8   n F is chosen for the Y capacitor in this design. Therefore, it is necessary to select the inductance value of the minimum value 1.2   m H to design the filter. The actual corner frequency of the computational filter is 55.7   K H z , which is close to and less than 56.1   K H z . The actual damping coefficient is ζ = 4.2 , and the actual damping coefficient greater than or equal to 1 is acceptable.

The designed differential mode filter is a Π -type balanced filter, and its equivalent circuit is shown in Figure 6.

Figure 6:

Equivalent circuit diagram of the third-order differential mode filter.

Therefore, the design target of 22 dB attenuation at 200 KHz requires that the corner frequency f c u t o f f d m of the filter is calculated by equation (4) to be 85.72 KHz.

The inductance value calculated according to equation (5) is 0.226   m H .

The capacitance value calculated according to equation (6) is 0.0153   μ F .

According to the calculation results of equations (5) and (6), it is reasonable to choose the value of DM Choke as 300   μ H and the value of capacitance as 0.2   μ F .

Combined with the design results of the previous two sections, the schematic diagram of the complete EMI filter circuit is obtained as shown in Figure 7.

Figure 7:

Complete EMI filter circuit diagram.

Among them, L1a and L1b are the two windings of the CMChoke, with a minimum inductance of 1.2 mH, L1a′ and L1b′ are the leakage inductance of the CMChoke, and the minimum inductance is 150 µH. CMChoke chooses ferrite EE20 core, double slot skeleton. The Y capacitors CY1 and CY2 are both 6.8 nF, and the X capacitors CX1 and CX2 are both 0.1 µF. The special feature of this circuit is that the leakage inductance of the CMChoke is used to provide the differential mode inductance required by the DMFilter, which reduces the use of magnetic materials and costs.

2.2 Design and calculation of CRMPFC circuit parameters

PFC output capacitors can be used in series with two 300 Vdc or 400 Vdc withstand voltages, and use two identical resistors in series to achieve equal voltages on each capacitor.

1. The realization of PFC voltage and overvoltage protection. The resistor divider network composed of PFB pin and R5, R6 is used to realize the PFC overvoltage alarm protection function. When the maximum PFC voltage is set to 430 Vdc, it is required to turn off the PFC drive signal when the PFC output voltage is higher than 430 V. The maximum PFC voltage of 430 Vdc can be estimated by checking the specifications of the PFC circuit or device being used. If the PFC output voltage exceeds this limit, it can result in damage to the PFC circuit or other system components, pose a safety hazard, and cause overvoltage stress leading to insulation breakdown and other failures. Hence, it is crucial to ensure that the PFC output voltage does not surpass the manufacturer’s maximum voltage rating or specified limit. This only requires selecting the resistance values of R5 and R6 according to equation (7).

According to equation (7) and the current driving capability of the PFB pin, the selected R5 is 5.4 MΩ and R6 is 31.5 KΩ to meet the requirements.

1. Design of PFCBULK capacitor. PFCBULK capacitors improve power factor, reduce harmonic distortion, enhance voltage regulation, and increase equipment lifespan in electrical systems. Their importance lies in improving system efficiency and reliability, resulting in cost savings and other benefits. The selection of the PFCBULK capacitor needs to meet the requirements of the ripple voltage and the withstand voltage of the BULK capacitor needs to be higher than the maximum BoostPFC voltage and half the ripple voltage. The ripple voltage in a circuit that employs a capacitor to filter out AC noise from a DC power source varies inversely with the capacitor’s value. Therefore, increasing the capacitance of the capacitor decreases the ripple voltage, while decreasing the capacitance increases the ripple voltage. This relationship is crucial when designing circuits that need steady DC power supplies. Typically, a 1 µF PFCBULK capacitor is chosen for 1WPFC output power. Therefore, it is reasonable to select two 400 V capacitors with a withstand voltage of 150 µF in series for this design of 60 W power supply, that is, the capacitance after the series is an equivalent capacitor of 75 µF, and then calculate the ripple voltage according to equation (8) as follows:

Among them, f is 50 Hz. The ripple calculated by equation (8) is the low-frequency ripple, because the ripple caused by the high-frequency switching frequency is negligible compared to the low-frequency ripple. According to equation (8), the ripple value can be obtained [5.9 V, 6.8 V], corresponding to the PFC output voltage [375 V, 430 V]. This value is the acceptable ripple range. Therefore, the choice of two 400 V, 150 µF capacitors in series can meet the design requirements.

1. Design and selection of PFCMOSFET and diode.

A PFCMOSFET (Power Factor Correction Metal-Oxide-Semiconductor Field-Effect Transistor) is an MOSFET specifically designed for use in power factor correction circuits. Design considerations include high voltage rating, low on-resistance, fast switching speed, and low gate charge. Selection factors include voltage and current ratings, switching frequency, and on-resistance among others. Selecting the appropriate PFCMOSFET is critical to achieve efficient and reliable power factor correction. The ultra-fast recovery diode MURS360 times of 600V3A can be selected to meet the design requirements.

1. Design of PFC inductor

The peak current of the PFC inductor at the minimum input voltage is calculated as 2.35 A according to equation (10), and the mean current can be calculated as 0.96   A according to equation (11).

The design calculation of the PFC inductor is calculated according to the specification parameters in the table and equation (12). Among them, the minimum switching frequency is selected as 30   K H z , and it can be calculated that the inductance value is less than 1200   μ H at the minimum AC input. At the maximum AC input voltage, and when the maximum PFC output voltage is 400   V , the maximum value of the inductance is 1050   μ H .

In order to make the PFC controller work in the verified frequency range, it is more reasonable to choose the inductance as 650 μH–700 μH. To protect a PFC controller operates in a verified frequency range, it is crucial to select the right inductance value for the PFC inductor. An inductance value in the range of 650 μH–700 μH is often the most reasonable choice since it provides a resonant frequency that matches the switching frequency of many PFC controllers, reduces ripple current, and minimizes the inductor size. However, the selection of the specific inductance value should depend on the application requirements, input voltage, output power, and component selection. It is advisable to refer to the PFC controller datasheet or consult an experienced engineer for the optimal inductance value. According to the selected maximum and minimum inductance and equation (13) to calculate the switching frequency operating range, the operating frequency range of the PFC controller can be obtained between the minimum 45   K H z and the maximum 54 K H z .

In order to provide the auxiliary voltage between all the way 15   V d c − 18   V d c , it is more appropriate to choose the auxiliary coil turns ratio of 10:1. In order to maintain the principle that the material selection of the whole design is as consistent as possible, PQ-2020 ferrite core is selected as the core of the PFC inductor and the resonant half-bridge isolation transformer. Based on the 0.83   A input RMS current, a three-strand AWG#30 field wire was chosen for the main winding coil to reduce AC losses. Three-strand AWG#30 field wire is a type of stranded wire consisting of three thin wires twisted together. This wire offers increased flexibility, better durability, and reduced electrical resistance compared to a solid wire of the same gauge, making it ideal for small electrical and electronic projects. In terms of reducing AC losses, stranded wire has an advantage over solid wire since it decreases the effects of skin and proximity effects. In a solid wire, the AC current concentrates near the surface of the conductor, leading to increased resistance and power loss. However, in a stranded wire, the current flows through each strand, increasing the surface area of the conductor and reducing the skin effect. Its diameter is 0.61   m m . The inner winding width of the skeleton of the magnetic core is about 12   m m . This means that 18 turns per layer, for a total of 4 or 5 layers, can yield 75 cores. Based on the above analysis, it is reasonable to choose the PQ-2020 core. The required number of turns can be calculated according to equation (14).

In the known conditions, B max is 3500 G, A _e is 0.6, I _pk is 2.35 A, L is 650 μH. It can be calculated that N is a minimum of 72 turns, and 75 turns are selected.

To prevent core saturation, the core must be air-gapped according to equation (15). The calculated air gap length is 0.063   c m .

In summary, the PFC inductor design can be summed up as follows:

The 63.5WCRMPFC inductor core type under 70 KHz is ferrite PQ-2020, the material is PC40, the inductance is 650 μH–700 μH, and the skeleton type is P Q − 202014 -pin P C material. The main winding coil is 1.3 feet, 75 turns, 3 strands of AWG#30 excitation wire are paralleled, and the auxiliary winding coil is 4.6 feet, 8 turns, single-strand AWG#30 wire is wound in a single layer. The wound inductor needs to be evacuated and immersed in water, and a safety withstand voltage test of more than 1000   V d c is required.

In the LLC resonant converter, the resonant tank element inductor L _R and capacitor C _R are connected in series with the primary side of the transformer, while the resonant inductor L _M is connected in parallel with the primary side of the transformer, as shown in Figure 8a. The LLC resonant converter is a type of resonant converter used in power electronics, often with a transformer and a magnetizing inductance (L _m ). The converter operates by switching the input voltage across a resonant circuit of L _s and C _s . During the on-time of the switch, the magnetizing inductance (L _m ) limits the rate of change of current and during the off-time, it limits the rate of change of voltage. The L _m acts as a buffer between the primary and secondary sides of the transformer, enabling efficient power transfer and controlling the current and voltage stresses. The LLC resonant converter uses the transformer magnetizing inductance to generate another resonant frequency, which is much lower than the main resonant frequency generated by the resonant tank elements L _R and C _R . LLC resonant converters are designed to operate at switching frequencies much higher than the resonant frequency of the resonant tank composed of L _R and C _R . The LLC converter is a DC/DC converter with soft-switching capabilities that is based on a resonant circuit. With zero-voltage switching, the LLC resonant circuit lowers switching loss (ZVS). Unlike the SLR converter, the LLC converter can maintain a constant output voltage under minimal load conditions.

Figure 8:

Schematic diagram of LC resonant converter.

The advantage of the LLC resonant converter is that it can achieve a narrow operating frequency range under light load conditions and ZVS capability under no-load conditions. Its DC characteristic curve, which is different from SRC and PRC, is shown in Figure 8b. This particular DC characteristic curve shows that LLC resonant converter is an excellent choice for DC–DC converter application in this project. The two resonant frequencies are given by formulas (16) and (17). The first resonant frequency is determined by L _R and C _R , while the other is determined by L _R , C _R and L _M .

2.3 Design of LLC resonant half-bridge circuit parameters

LLC resonant half-bridge circuit (LLCHBR) contains resonant components C _R , L _R and L _M , power MOSFETs, isolation transformers, output rectifier diodes, output capacitors, etc. A common practice is to use an isolation transformer to provide the inductance required by the two resonant inductors.

1. Determination of the resonance element. LLC resonant converters operate at a fixed frequency, thus mitigating EMI issues and improving efficiency. In order for the resonant converter to operate in this mode, the switching frequency needs to match the resonant frequency of the resonant cavity. The resonant frequency is shown in equation (18).

1. Design of resonant half-bridge isolation transformer. Because the resonant half-bridge isolated converter operates at a fixed frequency with a symmetrical duty cycle, the design of the transformer is relatively simple. Because the resonant capacitor is divided into two parts to form a voltage divider network, half of the HBR input voltage is added to the primary side of the transformer. A voltage divider network is typically made using a pair of resistors in series across a voltage source. However, instead of resistors, a voltage divider network can also be created using resonant capacitors. This is done by connecting two capacitors in series across the voltage source, with their resonant frequency equal to the input signal frequency. At resonance, the capacitors behave like resistors, and their impedance becomes purely resistive. The output voltage is obtained from the junction between the two capacitors, and its value depends on the capacitance ratio. Resonant capacitors can be advantageous in high-frequency circuits, reducing voltage drops and power losses. The maximum PFC output voltage was previously chosen to be 400 Vdc, and produces a maximum output voltage of 40 Vdc. Isolation transformers are transformers that have separate primary (input) and secondary (output) windings. These transformers are designed to transfer electrical power from an AC power source to a device or equipment while keeping the powered device electrically isolated from the power source. This is done for safety reasons and to minimize transients and harmonics. The primary and secondary windings are separated from each other by a dielectric insulation barrier, effectively creating an electrical separation between the input and output power. Therefore, the turns ratio of the isolation transformer is:

Therefore, the turns ratio of the primary and secondary turns of the isolation transformer is 5:1. The secondary side of the isolation transformer uses two equal windings with a center tap. Because the magnetic core is based on the power level and the overall planning of the project, the PQ-2020 magnetic core has been selected. After the turns ratio is determined, the minimum number of turns on the primary side needs to be determined to avoid core saturation. Core saturation can cause energy efficiency loss and damage to a transformer due to excessive magnetization caused by high input voltage or current. To prevent core saturation, the primary side of the transformer needs to have a minimum number of turns. Increasing the number of turns on the primary side increases the inductance of the transformer, which helps to keep the magnetic flux density within safe limits. However, this also increases resistance and reduces efficiency. Therefore, determining the minimum number of turns required to prevent core saturation while maintaining high efficiency depends on the transformer’s design requirements and operating conditions. The required minimum number of turns can be calculated from equation (20).

Among them, N _p is the minimum number of turns of the primary side required by the isolation transformer; V is the maximum voltage of the primary side of the isolation transformer; F is the switching frequency of the LLC resonant half-bridge; BM is the maximum magnetic flux density of the ferrite; Ae is the cross-sectional area of the magnetic core. It is more appropriate to choose 96 turns.

The average current on the primary side of the isolation transformer is calculated according to equation (21).

According to the size of the current, we choose the wire diameter of the primary winding of the transformer, and choose the AWG#28 excitation wire, which can be wound in three layers, with 32 turns per layer, for 96 turns. The number of turns on the secondary side is 96/5 = 19.2 turns, so we can choose 19 turns.

2.4 Design of constant voltage and constant current feedback compensation circuit

The constant current feedback compensation circuit (CVCC) part in Figure 9 is the constant voltage and constant current feedback control loop. The output voltage of the resonant half-bridge is detected through R7 and R8, and the change of the output voltage is fed back to the operational amplifier in CVCC for conditioning and compensation. Then, it is fed back to the PFC and the resonant half-bridge controller to adjust the PFC output voltage, so that the change of the resonant half-bridge output voltage follows the change of the PFC output voltage.

Figure 9:

Constant voltage and constant current feedback control circuit.

Its transfer function is as follows equation (22).

By extracting the factor s / ω z and reconfiguring equation (22), the following equation (23) can be obtained.

From equation (23), the zero point, the zero-dB original pole and the IF gain can be obtained as follows:

As the frequency increases, C10 acts as a short circuit and the gain decreases to the ratio of the resistors, as shown in equation (27).

The phase angle is calculated according to equation (23). Because of the presence of the negative sign in front of G 0 , in the reverse configuration of the op amp, that is, operating in the second quadrant, the phase angle must be added by 180°, and the result is equation (28).

When ω tends to infinity, the term on the right is zero and the phase angle is reversed.

The AC signal from one of the two op amps interacts with the transfer network consisting of the optocoupler and its associated parts. Its transfer function is as follows:

Therefore, the transfer functions of voltage and current are given by equations (30) and (31), respectively.

Among them,

The high-frequency pole ω _p appears in both the voltage and current transfer functions. Therefore, when the high-frequency pole is fixed in one compensation loop, it is necessary to adjust the zero in the other loop to meet the required phase margin.

In order to limit the loss, the loss on the current sense resistor is generally limited to 100 mW. Therefore, the sampling resistance can be calculated according to equation (37):

Since the voltage drop of the sampling resistor calculated according to equation (37) is too small, it is reasonable to use a sampling resistor of 0.1   Ω . In the R9 and R10 networks, if R9 is assumed to be 3.3 K, R10 can be calculated as follows:

In equation (38), the expression of R10 is obtained as equation (39), and this equation is used to set the output current through the resistance value of R10 in the table.

According to the above calculation process, the value of R _out is 0.1 Ω, the value of R9 is 3.3 K, and the value of R10 is 50 K, that is, the maximum output current is set to be 1.65 A.

The setting of the output voltage is set by the value of the output voltage detection resistor divider network R7 and R8, and the maximum output voltage V _out is determined by equation (40).