What are the crosstalk, jitter, tilt, and skin effects in PCB design?

In order to achieve efficient design, several factors need to be considered when designing a printed circuit board (PCB). This article discusses some general but key factors that need to be considered. These are the key considerations for predicting the final result of a designed printed circuit board (PCB).

crosstalk

• General introduction to crosstalk：

Crosstalk refers to the transmission of unwanted signals from one network to another, and it occurs between each pair of networks close to each other. Generally speaking, two single-ended signals in a printed circuit board (PCB) or other type of system are called crosstalk. Crosstalk usually only occurs between two wires, not necessarily on a printed circuit board (PCB).

• Detailed description of crosstalk：

We need to describe how the noise is coupled from one line to another. When defining crosstalk, we have two lines, one is the "attacker line“ and the other is the ”victim line". As the name suggests, the attacker will cause some noise on the victim. The situation may also be the opposite. The victim may be a drive signal, which may cause some noise in another signal. This means that the roles are interchangeable because they are reversible. Generally, these lines contain the driver, impedance, and load impedance in the circuit.

• The root cause of crosstalk：

It is very important to understand how the coupling between two parallel transmission lines occurs and the basic role of the edge field effects that cause capacitive and inductive coupling. These two fields, namely the electric field and the magnetic field, are not only limited to the return path, they are the diffusion of the field line and will also affect the surrounding conductors. If we move the network closer to the active network, more edge field effects will be observed. If we want to reduce the impact of the edge field, we need to increase the spacing between the transmission tracks running in parallel. The inference here is that the influence of the edge field is inversely proportional to the spacing between orbits running in parallel.

• Factors that determine them

When there is a signal on the interference line and the signal is being transmitted in the direction of the load, during the switching event, it will generate noise on the disturbed trace. Therefore, during the switching operation, we will see the change of dV/dt or signal voltage over time, and dI/dt, that is, the change of current over time, during this period of time. This change in the electric or magnetic field will lead to changes in the electric and magnetic fields, thereby inducing a certain amount of current in the adjacent conductors. Therefore, the edge field effect will produce small capacitance and inductance between two parallel orbits, which we call mutual capacitance and mutual inductance.

• Crosstalk noise categories of different transmission lines：

Suppose a setup has two transmission lines, and each transmission line has two transmitters and two receivers. There are attackers and victims between the transmission lines. The noise voltage measured in this setting has a very different mode for the noise across the victim network. Therefore, in order to distinguish the two ends, we uniquely mark them as the near end and the far end. The end close to the source is marked as the near end of the victim's network. Similarly, the end away from the source is marked as the far end of the victim's network. Here, the attacker and the victim are located between the two transmitters and the two receivers. The two transmitters are Tx1 and Tx2, and the two receivers are Rx1 and Rx2. In transmission line 1 (Tx1), the crosstalk observed in another transmission line (Tx2) is called near-end crosstalk. The noise suffered by the victim (Rx2) due to Tx1 is called remote crosstalk.

• Direction parameters：

The direction of signal propagation plays an important role here. The near end will be located in the backward direction of signal propagation, while the far end will be located in the forward direction.

Jitter

• Definition of jitter：

Jitter refers to the deviation of the signal period. If we take a pulse signal, if it is transmitted through a channel, ideally, when the receiver receives it, it should be very thick. However, sometimes, due to jitter problems, the conversion of the signal will be advanced or delayed.

• Types of jitter：

Generally speaking, jitter in printed circuit boards (PCBs) is divided into two categories.

• Random jitter
• Deterministic jitter

• Random jitter：

Random jitter may occur at any point in time and cannot be predicted by the designer. This is mainly due to the presence of white noise in the circuit board due to thermal problems.

• Deterministic jitter：

Deterministic jitter, as the name suggests, can be predicted in advance.

Signal offset

• Definition of signal deviation：

This is a very important parameter to be considered in high-speed design. We have several media, such as air, vacuum, metal, etc. In a printed circuit board design, there is a copper wire between the driver and the receiver. When a signal is transmitted from the transmitter at time t=0, the receiver will not receive the signal immediately. It takes some time for the signal to be transmitted from the driver side of the PCB to the receiver side, and this transmission delay (td) is proportional to the length of the line (L) between the driver and the receiver.

• How does signal deviation occur?

If we assume a synchronization interface, it means that we have a clock and a data signal. When the function interface appears, there will be a clock signal with rising and falling edges.

We sample the signal at specific time intervals at the receiving end. If we assume that there is a clock signal and a root data signal on the printed circuit board, we always make sure that the length of the clock signal is approximately equal to the length of the data signal.

Skin effect

Definition of skin effect：

Skin effect is the most common phenomenon, that is, when an AC signal (AC) is transmitted along a transmission line located directly above the main body, current will gather or gather around the edge of the trace. The main body here refers to the substrate layer. When alternating current is transmitted, it increases the amount of current in it and distributes it along the edge of the trace. As the signal frequency increases, the amount of current here will gradually decrease, and the closer it is to the edge of the trace, the closer it will be.

Occurrence of skin effect：

Here we assume that there is a ground plane below the trace of the PCB substrate and a transmission line above the substrate. When the signal passes through the substrate, a potential difference or voltage will be generated between the trace and the ground plane, resulting in a current. The electric and magnetic fields here are current and voltage.

pointer：

• Causes of skin effect

The skin effect only appears in alternating current, but not in direct current. The main reason is that there is no frequency component in direct current.

• Appearance

This effect is not everywhere, but it does occur in transmission lines.

• Simplify the skin effect

When current flows inside the conductor, the current density does not flow in the center of the conductor, but outside the conductor. This is called the “skin effect.” The “skin" here refers to the outer surface of the conductor.

• Common problems stemming from skin effects

Since the current flows less in the center part of the conductor than on the outer surface, the internal resistance of the center point of the conductor increases. As the resistance increases, the power loss will also increase, because the resistance is inversely proportional to the conductivity.

• Radio frequency problems caused by this effect

With the help of radio frequencies, FM (frequency modulation) and television signals can work. However, due to this skin effect, radio frequency is greatly affected. This is why it is necessary to reduce the skin effect.

• Factors affecting them
• The shape of the conductor depends to a large extent on
• Material type
• Conductor diameter

conclusion

Therefore, attention to these parameters must be mandatory to reduce any impact on the output.