(Detailed long text) Capacitor selection rules in power supply design
The power supply is often most neglected link in circuit design process. As a great design, design of power supply must be very important, it greatly affects performance and cost of entire system. The use of capacitors in power supply design is often most overlooked point in power supply design.
1. The principle of operation of capacitors in design of power supply
In power supply design applications, capacitors are primarily used for filtering and decoupling/bypass. Filtering is operation of filtering frequency of a specific band in a signal, and is an important measure for suppressing and preventing interference. Based on results of observing a random process, a probability theory and a method for estimating another related to a random process have been developed. The term filtering originated from communication theory, which is a method of separating useful signals from received signals containing interference. The "received signal" is equivalent to observed random process, and "useful signal" is equivalent to assumed random process.
Filtering mainly refers to filtering out external noise, while decoupling/bypassing (a kind of decoupling effect achieved as a bypass, which will be replaced by "decoupling" in future) is intended to reduce interference from external local circuit noise. Many people tend to confuse two concepts. Let's look at schema structure:
The power supply in figure is shown for A and B. The current passes through C1 and then passes through a section of PCB traces, splitting into two paths to power A and B, respectively. When A needs a large current at a certain moment, if there is no C2 and C3, then voltage at terminal A will become lower due to line inductance, and voltage at terminal B will also depend on voltage at terminal A, so local circuit The change in current A causes voltage local circuit B, thereby affecting signal of circuit B. Similarly, current change in B will also interfere with A. This is "co-channel interference".
After adding C2, when local circuit needs a short-term high current, capacitor C2 can temporarily provide current for A, even if there is inductance in common circuit, voltage at terminal A will not drop too much. . The impact on B will also be greatly reduced. Therefore, decoupling effect is played by current bypass.
General filtering mainly uses large capacitors that do not require very high speed, but require large capacitance values. If incomplete circuit A in figure refers to microcircuit, and capacitor is located as close as possible to power supply pin of microcircuit. And if "partial circuit A" refers to a functional module, then ceramic capacitors can be used. If capacitance is not enough, tantalum capacitors or aluminum electrolytic capacitors can also be used (provided that each microcircuit in functional module has a decoupling capacitor - a ceramic capacitor).
The capacity of filter capacitor can often be calculated from datasheet of power IC. If filter circuit uses electrolytic capacitors, tantalum capacitors, and ceramic capacitors at same time, place electrolytic capacitors closest to switching power supply to protect tantalum capacitors. The ceramic capacitors are located behind tantalum capacitors. Thus, best filtering effect can be obtained.
The decoupling capacitor must meet two requirements: capacitance requirements and ESR requirements. That is, decoupling effect of a 0.1uF capacitor may not be as good as two 0.01uF capacitors. In addition, 0.01uF capacitors have lower impedance in higher frequency ranges. If a 0.01uF capacitor can meet capacitance requirement in these frequency ranges, it will have a better decoupling effect than 0.1uF capacitors.
Many design guides for high-speed microcircuits with a large number of pins contain requirements for design of a power supply for decoupling capacitors. For example, a BGA package with more than 500 pins requires at least 30 ceramic capacitors for a 3.3V power supply. Somewhat large capacitors, total capacitance should be more than 200uF...
Secondly, correct selection of capacitors in various power supplies
As a basic component, capacitors play an important role in electronic circuits. In traditional applications, capacitors are primarily used for communication bypass, power filtering, DC blocking, and weak signal generation and delay. With development of electronic circuits, especially power electronic circuits, various special requirements are placed on capacitors for various applications. Let's start with design of capacitor. The simplest capacitor consists of plates at both ends and an insulating dielectric (including air) in middle . After electrification, plates are charged, forming a voltage (potential difference), but due to insulating material in middle, entire capacitor does not conduct current. However, this situation is possible provided that critical voltage (breakdown voltage) of capacitor is not exceeded. We know that any substance is relatively insulating. When voltage across a substance is increased to a certain level, substance can conduct electricity. We call this voltage breakdown voltage.
The capacitor is no exception. Once a capacitor has failed, it is not an insulator. However, at high school stage, no such voltage is observed in circuit, so they all operate below breakdown voltage and can be considered as an insulator. However, in an AC circuit, direction of current changes with time. The process of charging and discharging a capacitor has time, at which time a changing electric field is formed between plates, and this electric field is also a function of time change.
1. Filter Capacitor After rectifying AC (mains frequency or high frequency), capacitor filtering is required to smooth output voltage, which requires a large capacitance of capacitor, and aluminum is generally used electrolytic capacitor. The main problem when using aluminum electrolytic capacitors isThere is a relationship between temperature and service life, which basically follows 50°C rule. Therefore, in many cases where high temperature and high reliability are required, electrolytic capacitors with a long service life (for example, more than 5000 hours or even 105°C, 5000 hours) should be selected. As a rule, small volume electrolytic capacitors have a relatively short life.
It is used for input filter capacitor of DC/DC switching power supply, because switching converter consumes power from power supply in pulses, so a large high-frequency current flows through filter capacitor, when equivalent series resistance (ESR) of electrolytic capacitor is large, there will be large losses, which will lead to heating of electrolytic capacitor. A low ESR electrolytic capacitor can greatly reduce heat generated by ripple (especially high frequency ripple) current.
An electrolytic capacitor used to rectify output of a switching regulated power supply requires that its impedance frequency response does not show an upward trend at 300 kHz or even 500 kHz. However, conventional electrolytic capacitors begin to show an upward trend after 100 kHz, and effect of rectification and filtering on output of switching power supplies is relatively weak. The author through experiment found that electrolytic capacitor 4700uF, 16V in ordinary type CDII, ripple and peaks used for output filter of switching power supply is not lower than that of high-frequency electrolytic capacitor type CD03HF 4700uF, 16V, and increase The temperature of a conventional electrolytic capacitor is relatively large. Under heavy load, transient response of conventional electrolytic capacitors is much worse than that of high frequency electrolytic capacitors.
Since aluminum electrolytic capacitors cannot play a good role in high frequency range, they should be supplemented with ceramic or non-inductive film capacitors with good high frequency performance. Main advantages: good high frequency performance, low ESR, such as 1uF MMK5 type capacitor, resonant frequency is higher than 2MHz, equivalent resistance is less than 0.02Ω, which is much lower than electrolytic capacitors, and smaller capacitance, higher resonant frequency (up to 50MHz). or more), so output frequency response or power supply dynamics will be very good.
In filter capacitor section, we will focus on how to choose a filter capacitor in a switching power supply
How to choose a filter capacitor for a switching power supply
Filter capacitors play a very important role in switching power supplies. How tobut choice of filter capacitors, especially choice of output filter capacitors, is a matter of great concern for every engineer.
Conventional electrolytic capacitors used in 50 Hz circuits have a ripple voltage frequency of only 100 Hz, and charging and discharging times are in order of milliseconds. To obtain a smaller ripple coefficient, required capacitance reaches hundreds of thousands of microfarads. Therefore, purpose of conventional low-frequency aluminum electrolytic capacitors is to increase capacitance. The main parameters of its advantages and disadvantages. The electrolytic capacitor of output filter in switching power supply has a sawtooth voltage frequency reaching tens of thousands of hertz, or even tens of megahertz. At present, capacitance is not main indicator, and standard for measuring quality of high-frequency aluminum electrolytic capacitors is "impedance-frequency" characteristic. It is required to have a smaller equivalent impedance within operating frequency of switching power supply, and at same time have a good filtering effect of high-frequency peak signals generated by operation of a semiconductor device.
Many electronics designers know role of a filter capacitor in a power supply, but filter capacitor used at output of a switching power supply is different from filter capacitor used in a power frequency circuit. The frequency of ripple voltage on it is only 100 Hz, and charge and discharge time on order of milliseconds. To obtain a small ripple factor, capacitance required reaches hundreds of thousands of microfarads. Therefore, conventional aluminum electrolytic capacitors are usually used for low frequencies. The goal is to increase capacitance Capacitors cons.
An electrolytic capacitor is used as an output filter in a switching regulated power supply, frequency of sawtooth voltage on it reaches tens of kilohertz or even tens of megahertz. Its requirements are different from those of low-frequency applications, and capacitance is not main indicator. measures its quality, is its impedance-frequency response, which requires it to have a low and equal impedance in operating frequency band of a switching regulated power supply. can also have a good filtering effect. Generally, a common electrolytic capacitor for low frequency is about 10 kilohertz, and its impedance starts to seem inductive, which cannot meet requirements of switching power supply.
Conventional low-frequency electrolytic capacitors begin to show inductance at about 10,000 Hz, which cannot meet requirements of switching power supplies. High frequency aluminum electrolytic capacitor designed for switching power supplypower supply, has four terminals. The two ends of positive aluminum sheet are respectively extended as positive electrode of capacitor, and two ends of negative aluminum sheet are also respectively extended. as a negative electrode. Current flows from one positive end of four-terminal capacitor, passes through capacitor, and then flows from other positive terminal to load; current returning from load also flows from one negative terminal of capacitor, and then flows from other negative terminal to negative terminal of power supply.
The high-frequency aluminum electrolytic capacitor designed for switching power supply has four terminals. The two ends of positive aluminum sheet are respectively drawn out as positive electrode of capacitor, and two ends of negative aluminum sheet are also drawn out as negative electrode. Regulated power supply current is drawn from one positive terminal of a four-terminal capacitor, passes through interior of capacitor, and then flows from other positive terminal to load; current returned from load also flows from one negative end of capacitor and then flows from other negative end to negative terminal of power supply. Because four-terminal capacitor has good high-frequency characteristics, it provides an extremely favorable means for reducing ripple component of output voltage and suppressing switching noise.
The switching regulated power supply has multifunctional comprehensive protection: in addition to most basic stable voltage function, voltage regulator must also have overvoltage protection (more than +10% output voltage), undervoltage protection (lower than -10% output voltage %), phase loss protection, short circuit protection and overload protection are most basic protection functions. Surge Suppression (Optional): Sometimes power grid will experience sharp, high amplitude, narrow pulse width pulses that will destroy electronic components with a low withstand voltage. The anti-surge components of a regulated power supply can suppress such surges very well.
High-frequency aluminum electrolytic capacitors also have a stranded form that divides aluminum foil into several short lengths and connects them in parallel with several leads to reduce resistive component in capacitive reactance. - resistive materials are used together. A screw rod is used as a terminal to enhance capacitor's ability to withstand large currents.
Layered capacitors are also called non-inductive capacitors. Generally, cores of electrolytic capacitors are rolled into cylindrical shapes and equivalent series inductance is relatively large; setoff, it is biased, which reduces value of inductance and has better high-frequency performance. This type of capacitor is usually square-shaped, which is easy to correct, and can also reduce volume of machine accordingly.
Drawing of step-down capacitor power circuit
2. Absorbing and switching capacitor
As power rating of gate controlled semiconductor devices increases, switching speed becomes faster and rated voltage becomes higher and higher, snubber circuit capacitors only require sufficient withstand voltage, capacitance and excellent high-frequency performance is not enough.
In power electronic circuits, since switching speed of IGBT is less than 1 µs, snubber capacitor voltage change rate is usually required to be dv/dt> V/µs, and some require V/µs or even V/µs.
For ordinary capacitors, especially dv/dt<100V/µs of ordinary metalized capacitors, dv/dt≤200V/µs of special metalized capacitors, and dv/dt of low capacitance (less than 10nF) of special double metalized capacitors ≤1500V/µs , and a larger capacitance (less than 0.1uF) 600V / μs, which is difficult to withstand under influence of such a huge peak current with a high repetition rate. Phenomena that damage power electronic circuits.
At present, special capacitors for absorbing circuits, that is, metal foil electrodes, can withstand large peak currents and RMS current impacts, such as: small capacitance (below 10nF) can withstand voltage change rate of 100,000 V / μs ~ 455,000 V/µs, 3700 A peak current and 9 A effective current (e.g. CDV30FH822J03); large capacitance (more than 10nF, less than 0.47uF) or larger size can withstand more than 3400V/µs and 1000A peak current.
It can be seen that although same non-inductive capacitors, metallized capacitors and metal foil capacitors will have different characteristics in absorption circuit, similar shapes, but different characteristics are absolutely not interchangeable here. The size of capacitor will affect dv/dt and peak capacitance current of capacitor. Generally speaking, longer length, smaller dv/dt and peak current.
The performance of a capacitor in a sink circuit is that duty cycle of high peak current is small and RMS current is not very large. Similar to this circuit is switching capacitor of a thyristor inverter, although dv required for this capacitor /dt is less than an absorbing capacitor, but peak current and RMS current are larger, and conventional capacitor cannot meet current requirement.
In some special applications, capacitor needs to be repeatedly and quickly discharged to store energy, and discharge circuit resistance is extremely low and parasitic inductance is small. In this case, absorbing capacitor can only be used in parallel to ensure reliability in long-term use.
3. Resonant CapacitorFir-trees, such as resonant switching power supplies and resonant capacitors in resonant circuits of thyristor intermediate frequency power supplies, often pass a lot of energy during high current operation. Another example is that if characteristics of resonant capacitor of electronic ballast are not properly selected, voltage across capacitor may not reach breakdown voltage, but will be damaged due to large resonant current flowing.
In a circuit containing capacitance and inductance, if capacitance and inductance are connected in parallel, there may be a short period of time: capacitance voltage gradually increases, and current gradually decreases; at same time, inductor current decreases, gradually increases, but coil voltage gradually decreases. And after a short period of time: voltage of capacitor gradually decreases, and current gradually increases; while inductor current gradually decreases, and inductor voltage gradually increases. An increase in voltage can reach a positive maximum value, and a decrease in voltage can also reach a negative maximum value. The direction of current will also change in positive and negative direction during this process. At this time, we call this electrical oscillation of circuit.
The phenomenon of circuit oscillation may gradually disappear or remain unchanged. When oscillation continues, we call it constant amplitude oscillation, also known as resonance. Resonance time The time during which two forging voltages of a capacitor or inductor change in one cycle is called resonant cycle, and reciprocal of resonant cycle is called resonant frequency. In this way, so-called resonant frequency is determined.
Summarizing, today's power technologies require capacitors with different characteristics for different applications, and they should not be mixed, abused or misused to minimize damage that should not occur and ensure product quality. performance.
Third, an example of design of a capacitor buck power supply
A common way to convert AC power to low voltage DC power is to use a transformer to step down voltage and then rectify and filter it. capacitor step down power supply.
1. The principle of operation of a capacitor step-down power supply
The basic circuit of a capacitor buck simple power supply is shown in Figure 1, C1 is a step-down capacitor, D2 is a half-wave rectifier diode, D1 provides a discharge circuit for C1 during negative half-cycle. network, D3 - zener diode, R1 - charge-discharge resistance C1 after power off. The scheme shown in fig. 2, often used in practical applicationsI. When it is necessary to provide more current to load, bridge rectifier circuit shown in Figure 3 can be used. The unregulated DC voltage after rectification is usually higher than 30 V and will fluctuate greatly as load current changes. This is due to large internal resistance of this type of power supply, so it not suitable for high current power. delivery use cases.
2. Device selection principle for RC buck circuit
(1) When designing a circuit, first determine exact value of load current, and then refer to example to select capacity of step-down capacitor. Excess current will flow through Zener tube, if maximum allowable current Idmax of Zener tube is less than Ic-Io, it will easily cause Zener tube to burn out.
(2) To ensure reliable operation of C1, its withstand voltage must be more than twice voltage of power supply. (3) The choice of discharge resistor R1 must ensure that charge on C1 is discharged within required time.
3. Design Examples
In fig. 2, given that capacitor C1 is 0.33uF and AC input voltage is 220V/50Hz, find maximum current that circuit can supply to load.
The capacitance Xc of capacitor C1 in circuit is: Xc=1/(2πf C)=1/(2*3.14*50*0.33*10-6)=9.65K Charging current flowing through capacitor C1 (Ic) is: Ic = U / Xc = 220 / 9.65 = 22 mA.
Typically, relationship between capacitance C of step-down capacitor C1 and load current Io can be approximated as: C=14.5 I, where unit of capacitance C is µF and unit of Io is A.
- What is drowning in gold? Why Shen Jin?
- This is a metaphor for EMI/EMS/EMC that can be understood at a glance.
- How many types of pads have you seen in PCB design?
- Summary of Common PCB Repair Techniques
- What is three anti-paint? How to use it correctly?
- Knowing these rules, you will not get confused looking at circuit diagram.
- How to make anti-interference PCB design?
- Can diodes do this?