Common termination methods are: series termination, simple parallel termination, Devining termination, RC network termination, and diode termination. The following will be analyzed separately for these types of termination methods.
(1) Series termination
Series termination means that a resistor RS is connected in series as close as possible to the source to match the impedance of the signal source, so that the reflection coefficient at the source is zero, thereby suppressing the signal reflected from the load and then reflecting back from the source to the load. The output impedance ZS of the PS plus drive source should be equal to the transmission line impedance Zo, ie
The value of the series resistor is usually chosen to be 15 to 75 Ω, and the more choice is 33 Ω.
The advantage of series termination is that;
· Only one termination resistor is required for each line, no need to connect the DC power supply, so it does not consume too much power;
· Current limiting is provided when driving high capacitive loads, which can help reduce ground bounce noise.
The disadvantages are:
· Due to the voltage division of the series resistor, the voltage is only half of the source voltage in the middle of the trace path, so the distributed load cannot be driven;
· Since the resistor is connected in series with the signal path, the RC time constant is increased, which slows down the rise time of the load signal and is therefore not suitable for high-frequency signal paths (such as high-speed clocks). It should be noted that the series resistor must be as close as possible to the output of the source driver, and it is best not to use vias on the PCB because of the capacitance and inductance of the vias.
(2) Parallel termination
Parallel termination is also called DC parallel termination. This method is achieved by connecting a terminating resistor Pp (Ap=Zo) to the ground or pulling up to the DC power supply at the input of the receiver (ie, the end of the wiring network). The reflection is eliminated at the load end as shown in Figure (c). The advantage of this termination method is that it is simple and easy to design. The disadvantage is that it consumes DC power and cannot be used in portable devices that require low power consumption. In addition, this pull-up to the power supply can increase the drive capability of the driver, but it will raise the low level of the signal; while pulling down to ground can increase the current absorption capacity, but will pull the signal high.
(3) Dai Weining termination
The Thevenin termination is a voltage divider type termination that uses a pull-up resistor Rpl and a pull-down resistor Rp; forms a termination resistor that absorbs reflections through Rp1 and Rp2.
This impedance must be equal to the transmission line characteristic impedance Zo to achieve the best match.
The selection of the resistance of the termination resistors Rpl and Rp2 should be considered to avoid setting an inappropriate load voltage reference level for high and low logic transition points. The Rp1/Rp2 ratio determines the relative ratio of logic high and low drive current. When Rp1=Rp2, the drive requirements for high and low logic are the same; when Rpl
The advantages of Thevenin Termination are:
· Can be used with distributed loads across the network;
· Can completely absorb the transmitted wave and eliminate the reflection;
· Set the line voltage when there is no signal to drive the line;
· Especially suitable for bus use.
Its disadvantages are:
There is always a DC current from the power supply Vcc to ground, resulting in DC power dissipation in the matching resistor, reducing the noise margin unless the driver can provide a large current.
The Thevenin termination method is ideal for high-speed backplane designs, long transmission lines, and large-load applications where the voltage level of the load is maintained near the optimal switching point by two parallel resistors, and the driver can drive the bus with less power. .
(4) RC network termination
The RC network termination, also known as the AC load termination, uses a series RC network as the termination impedance to eliminate network end reflections. The termination resistor Rp is equal to the transmission line impedance Zo. The capacitor Cp should be chosen such that the time constant of the RC network should be greater than twice the propagation delay, ie "RpCp>2TD", typically using a 0.1μF multilayer ceramic capacitor. For specific designs, the capacitance is determined by simulation.
The advantage of AC termination is that the capacitor blocks the DC path without generating additional DC power, while allowing high-frequency energy to pass through and acting as a low-pass filter. The disadvantage is that the time constant of the RC network reduces the signal rate. In addition, additional resistors and capacitors occupy board space and increase cost.
(5) Diode termination
In the diode termination mode, one diode is connected in series between the end of the transmission line and the power supply Vcc, and the other diode is connected in series between the end of the transmission line and the ground. Schottky diodes are commonly used because Schottky diodes have low turn-on voltages.
Unlike other termination methods, the diode termination is not intended to match the characteristic impedance of the transmission line to eliminate reflections. When the receiver voltage overshoots, the diode begins to operate to stabilize the voltage. Although it can prevent overshoot, there are two disadvantages: the reflection still exists in the system; the reflection of high speed signals is slower. In order to get the advantages of this technology, you can use it together with the previous methods.