How to solve the high-frequency effects generated by electromagnetic compatibility measurement

Electromagnetic compatibility problems usually occur at high frequencies, with the exception of individual problems (voltage drops, transient interruptions, etc.). Therefore, in electromagnetic compatibility design, it is necessary to have high-frequency thinking. In short, it is important to pay attention to the characteristics of equipment and circuits under high-frequency conditions, which are different from traditional frequency states. If you still judge and analyze according to common engineering thinking, you will enter into a design misunderstanding.

1、 Capacitance

In medium to low frequency or DC situations, a capacitor is an energy storage component that only exhibits the characteristics of a capacitor. However, in high frequency situations, it is not only a capacitor, but also has ideal capacitor characteristics, leakage current (R on the high-frequency equivalent circuit), wire inductance, and ESR (equivalent series resistance), resulting in voltage pulse fluctuations, as shown in the figure:

Analyzing from this diagram can help our designers get many beneficial design ideas.

Firstly, according to the conventional thinking, Z=1/(2 π FC), Z is the capacitive reactance of the capacitor. The higher the frequency, the smaller the capacitive reactance, and the better the filtering effect. That is, the higher the frequency, the easier the clutter is to leak, but this is not the case. Due to the existence of lead inductance, a capacitor has the best filtering effect when its overall impedance is the smallest when its 1/2 π FC=2 π FL equation is established. If the frequency is high or low, the filtering effect will decrease. Therefore, we can analyze why two capacitors, one electrolytic capacitor and one ceramic capacitor, are added to the ICVC terminal, with a capacitance difference of more than 100 times, to increase the filter bandwidth of the capacitor.

resolvent:

Replace multiple ordinary decoupling capacitors with BDL filters.

BDL is a new generation of capacitors. In the high-frequency state, BDL's patented internal structure greatly reduces high-frequency parasitic effects. At high frequencies, due to the balance of its internal structure, it has better filtering and decoupling effects.

To understand the equilibrium characteristics of BDL, it is first necessary to understand its physical structure, as can be seen from the figure:

The standard bypass capacitor consists of alternating parallel electrode plates connected to the poles A and B.

BDL adds two parallel reference electrodes G1. G2, which are printed in parallel between A. B electrodes to form a Faraday shielding cage or coaxial structure to achieve balance and reduce ESL.

What this patented structure brings us is that BDL has a wider filtering frequency band than ordinary capacitors. Currently, only one BDL filter is needed to solve the filtering frequency band that originally required multiple ordinary capacitors, and compared to similar through-center capacitors, it is not subject to flow restrictions.

2、 Inductance

Inductance and capacitor are a somewhat opposite device, one with low resistance; One is low resistance. If the distribution of capacitors is temporarily ignored, the inductive impedance mainly consists of two parts: one is the resistance component (R), and the other is the inductive reactance component (FL), namely:

Z=2πfl

The resistance component comes from the resistance of the winding inductance and the loss of the magnetic core. As an inductor for electromagnetic interference suppression, I hope that the larger the resistance component, the better. Because resistance can convert interference energy into heat consumption, inductive resistance simply reflects interference energy back to the signal source.

Although the inductive impedance formally increases with increasing frequency, its properties are completely different within different frequency ranges.

Low frequency: The magnetic core has high permeability, large inductance, small inductance and resistance components, and the impedance is mainly inductive reactance. It is an inductor with low loss and high Q value characteristics.

When the frequency is high: As the frequency increases, the magnetic permeability of the magnetic core decreases, resulting in a decrease in the inductance of the inductive component of the inductor. However, when the core loss increases, the resistance component increases, mainly the resistance component. Therefore, when high-frequency signals pass through the ferrite, electromagnetic energy dissipates in the form of heat.

Terms of settlement:

The core material is the key point of selection, whether it is a common mode inductor or a differential mode inductor. The saturation characteristic of the magnetic core material is the most important when using a differential mode inductor magnetic core. As a common mode inductive core, more attention is often paid to the magnetic permeability of the core material.

Generally, there are two materials used as differential mode inductor cores: one is iron powder core, and the other is iron nickel molybdenum core. The price of the iron core is relatively low, but when used at 400 Hz current conditions, overheating may occur. The biggest advantage of these two materials is that they are not easily saturated. However, the magnetic permeability is low.

Ferrite materials are mainly used as common mode inductor cores, commonly used in manganese zinc ferrite and nickel zinc ferrite. Although the DC magnetic permeability of manganese zinc ferrite is high, it decreases rapidly with the increase of frequency; In addition, due to the good conductivity of the magnet, there will be a large distributed capacitance between the winding and the magnetic core, so it is only suitable for low-frequency applications. The DC magnetic permeability of nickel zinc ferrite is low, but it can maintain a high frequency. In addition, the magnet has a large resistance and is suitable for occasions with high frequencies.

3、 Cable and PCB wiring.

The high-frequency equivalent characteristics of PCB wiring (as shown in the figure), regardless of high or low frequencies, are objectively present in wiring resistance, but for wiring inductance, it can only be displayed at high frequencies. In addition, there is a distributed capacitor, but when there is no conductor near the wire, this distributed capacitor is useless, just like a woman cannot have children without a man. This requires two conductors to function. Therefore, we should pay attention to common mode noise caused by cable or PCB wiring.

Common mode radiation is caused by unwanted voltage drops in a circuit. This voltage drop places some parts of the system at a high potential. When an external cable is connected to the system, the external cable is excited by a common mode voltage, forming an antenna that radiates an electric field.

Therefore, the filter of external cables and the layout of internal harnesses should be considered in equipment design.

resolvent:

1. Reduce the common mode voltage.

The purpose of reducing the common mode voltage is to reduce the common mode current. When the impedance of a common mode circuit is constant, reducing the common mode voltage can reduce the common mode current.

(1) Clean cable interface settings: Clean and noiseless voltage. Connecting the cable to the ground wire can effectively reduce the common mode voltage. Typically connected to a metal chassis to further reduce the common mode voltage.

(2) Shielded internal cable: When the internal cable is longer, it is easier to sense higher common mode voltages. At this point, the internal cables can be shielded, and the shielding layer needs to be connected to the metal chassis.

2. Control cable length.

On the premise of meeting the use requirements, try to use short cables. However, the cable length is often limited by the connection distance between devices and cannot be arbitrarily shortened. In addition, when the cable length cannot be reduced to 1/2 of the wavelength, the effect of reducing the cable length is not significant.

The spacing between cables should not be too close, otherwise crosstalk between signal cables can be caused by the presence of distributed capacitance in the wires. Of course, it is best for the signal line to be closer to the coupling of the ground wire, so that the fluctuating interference on the signal line can be easily discharged onto the ground wire.

3. Increase the common mode current loop impedance.

(1) Disconnect the connection between the circuit board and the chassis (only the low frequency band is valid);

(2) Connecting a common mode choke in series at the cable or interface end: A common mode choke can form a large impedance to the common mode current without affecting the differential mode signal, so it is very simple to use without considering signal distortion. The common mode choke does not need to be grounded and can be directly added to the cable.

4. Common mode filtering.

Another effective method to solve cable radiation is to perform common mode filtering on the cable. The principle of common mode filtering is to use high-frequency common mode current components on low-pass filtering cables, which is the main reason for cable radiation problems.

5. Shielded cable.

The shielding layer directly blocks the differential mode radiation of the differential mode signal circuit in the cable; Provide a path for common mode current to return to the common mode noise source, reducing the loop area of the common mode current.

The key to controlling common mode radiation with shielded cables is to provide a low impedance channel for common mode current, allowing it to flow back through the shielding layer to the common mode voltage source. The impedance of the common mode current channel provided by the cable shield layer consists of two parts: one is the impedance of the shield layer itself; The other part is the overlapping impedance between the cable shield and the metal chassis.

Therefore, in electromagnetic compatibility design, in order to form a low impedance channel, it is not only required that the shielding layer of the cable itself be of good quality (low RF impedance), but also that the overlapping impedance between the cable shielding layer and the metal chassis should be as low as possible. The way to ensure a low impedance overlap between the cable shield and the chassis is to connect the chassis within 360 degrees. In other words, the cable shield layer forms a complete shield with the metal chassis, regardless of whether the chassis is grounded.