Simplify your Ethernet design, Part 1: Ethernet PHY basics and selection process

Is it 100BASE-T1, 1000BASE-T, 100BASE-TX, 10BASE-T or 10BASE-Te? For those who are not well versed in Ethernet physical layer (PHY) terminology, it is very difficult to evaluate various types of terminology. What do these numbers, symbols and abbreviations refer to? What is the media independent interface (MII)? What is the difference between the automotive physical layer and the industrial physical layer? How to choose the physical layer for network protocol cameras, car networking control units, and programmable logic controllers? Do all physical layers meet various fieldbus requirements?

Is it 100BASE-T1, 1000BASE-T, 100BASE-TX, 10BASE-T or 10BASE-Te? For those who are not well versed in Ethernet physical layer (PHY) terminology, it is very difficult to evaluate various types of terminology. What do these numbers, symbols and abbreviations refer to? What is the media independent interface (MII)? What is the difference between the automotive physical layer and the industrial physical layer? How to choose the physical layer for network protocol cameras, car networking control units, and programmable logic controllers? Do all physical layers meet various fieldbus requirements?

In the first part of the technical article series “Simplify your Ethernet design”, we will introduce the basics of the Ethernet physical layer to help you choose the appropriate physical layer for the end application. We will also provide TI physical layer selection flowchart to help you simplify the physical layer selection process.

What is the Ethernet physical layer?

In fact, the basic Ethernet physical layer is very simple: as shown in Figure 1, it is a physical layer transceiver (transmitter and receiver) that can physically connect one device to another. This physical connection can be copper wire (such as CAT5 cable-a blue patch cable used in households) or fiber optic cable.

Simplify your Ethernet design, Part 1: Ethernet PHY basics and selection process

Figure 1: Block diagram of the Ethernet system

The initial concept of the Internet was a network that could exchange data from one university to another quickly, reliably, and securely, which gave rise to the birth of Ethernet. Subsequently, electrical and Electronic engineers (IEEE) expanded on the basis of Ethernet, using new speeds (data rates), physical media (cable materials) and physical layer functions, making the expansion of Ethernet far beyond computer networks.

What are the functions of the Ethernet physical layer?

The Ethernet physical layer has two main functions.

First, the physical layer (PHY) has a digital domain that is directly connected to the device’s media access controller (MAC), such as a field programmable gate array (FPGA), microcontroller (MCU), or central processing unit (CPU). The PHY will have an MII, 4-bit wide data bus to varying degrees, with control lines and clock lines in the sending and receiving directions. MII has various forms, depending on the speed of MAC and PHY, and there will be different pin counts. Table 1 shows the most common MIIs and provides a summary of the pros and cons to be considered in the selection.

interface

Pin (pin count)

Speed ​​support

(Mbps)

profit

Disadvantages

MII

RX_D[3:0], RX_CLK, RX_DV, CRS, COL TX_D[3:0], TX_CLK, TX_EN

(14)

10, 100

Common pin assignment, low speed, easy wiring, lowest latency

No 1-Gbps support, high pin count

MII reduction (RMII)

RX_D[1:0], CRS_DV, TX_D[1:0], TX_EN

(6)

10, 100

Pin count reduction

Low deterministic latency (due to first in, first out), no 1-Gbps support

Gigabit MII (GMII)

RX_D[7:0], GRX_CLK, RX_CTRL, TX_D[7:0], GTX_CLK, TX_CTRL

(20)

10, 100, 1000

1-Gbps support, low latency

High pin count, generally not supported

Gigabit MII reduction (RGMII)

RX_D[3:0], RX_CLK, RX_CTRL, TX_D[3:0], TX_CLK, TX_CTRL

(12)

10, 100, 1000

1-Gbps support, common pin assignment

Difficult wiring, poor electromagnetic compatibility (EMC)

Serial Gigabit MII (SGMII)

SO_P, SO_M, SI_P, SI_M

(4)

10, 100, 1000

1-Gbps support, common pin assignment, excellent electromagnetic compatibility, easy wiring

Integrated circuits are more expensive

Table 1: List the common MII according to the number of pins and speed support

Secondly, PHY has a media independent interface (MDI), which connects a device (similarly, an FPGA, MCU, or CPU) to another device through a physical medium. This is often referred to as the analog domain of the physical layer because it is a continuous time-varying signal.

Based on MDI, select the appropriate Ethernet physical layer for your system

Now that we have introduced the functions of the physical layer, let us apply this knowledge to find the appropriate physical layer for your system. Most integrated circuit manufacturers stipulate that its physical layer has the following specifications and characteristics:

・ Data rate (10 Mbps, 100 Mbps, 1 Gbps)

・ Interface support (MII, RMII, GMII, RGMII, SGMII)

・ Media support (BASE-T, BASE-Te, BASE-TX, BASE-T1)

With this information, you can study this list starting from the data rate and match it to the data rate required by the end application. Next, determine the standards commonly used by the application. For example, since 2015, automotive Ethernet has been greatly expanded and is now usually provided by semiconductor manufacturers. Therefore, media standards are an important consideration because BASE-T1 is completely different from BASE-T.

To give another example, consumer electronics and most industrial applications use 10BASE-Te, 100BASE-TX, and 1000BASE-T because PCs support these standards. If your application is automated, then the physical layer supporting BASE-T1 is the most suitable solution. The exception to this rule is the automotive on-board diagnostic (OBD) port, which usually uses a BASE-T or BASE-TX interface to (again) support a PC connection. Table 2 summarizes the common MDI and its common systems.

MDI

IEEE specification (data rate)

Typical system

medium

profit

Disadvantages

10BASE-T/Te

IEEE802.3u

(10 Mbps)

Industrial lighting

CAT5

General support

Long distance

Low standby power

Low speed

10BASE-T1L

IEEE802.3cg

(10 Mbps)

Field transmitter; switch; heating, ventilation and air-conditioning controller; escalator

Unshielded twisted pair (UTP), shielded twisted pair (STP)

Ultra-long distance, single pair bidirectional, data power coupling

Low speed

100BASE-TX

IEEE802.3u

(100 Mbps)

PLC, IP camera, OBD port

CAT5

Universal support, used by fieldbus

High emission, external components

100BASE-T1

IEEE802.3bu

(100 Mbps)

Display cluster, audio host, gateway, infotainment, avionics communication, robotics, machine vision

UTP, STP

Low emission, high immunity, single-pair bidirectional cable

Uncommon (no PC connection support), short cable length

1000BASE-T

IEEE802.3ab

(1 Gbps)

IP camera, test and measurement

CAT6

1-Gbps speed

Cable is expensive

1000BASE-T1

IEEE802.3bp

(1 Gbps)

IoV control unit, gateway, avionics communication, robotics, machine vision

UTP, STP

1-Gbps speed, single pair bidirectional

Uncommon (no PC connection support), short cable length

Table 2: Comparison table of commonly used MDI

Most commercial and industrial physical layers support multiple data rates. These physical layers include a mechanism called auto-negotiation, which is a way for the physical layer to exchange information about function support so that they can be connected as fast as possible.

TI Ethernet physical layer selection flowchart

If you are ready to put your knowledge of the Ethernet physical layer into practice, Figure 2 is a simple physical layer selection flow chart that can help you determine the TI device suitable for your design. To learn more about the equipment in this flowchart, including the DP83826E low-latency Ethernet physical layer for supporting Industry 4.0 applications and the DP83TC811S-Q1100BASE-T1 Ethernet physical layer for space-constrained automotive applications, please visit us An overview of the Ethernet physical layer.

Simplify your Ethernet design, Part 1: Ethernet PHY basics and selection process

Figure 2: TI Ethernet physical layer selection flow chart

Stay tuned for the second part of our physical layer selection series. We will explore the best practices of physical layer schematic capture and layout to minimize noise, emission and signal loss.

The Links:   LP150X05-A2M2 NL10276BC28-11G

Related Posts