Advancing automotive networks: robust ESD protection for Ethernet applications

Ethernet technology has emerged as a robust and reliable solution for data communication in various industrial and computing applications. While its adoption in the automotive sector has been relatively limited, automotive Ethernet is gaining momentum due to its ability to deliver fast and resilient data transfer. With its high bandwidth, flexibility in bus topologies, and support for multiple electronic control units (ECUs), Ethernet technologies are strong contenders for advancing automotive networks from domain-based to zonal architectures. This article delves into the requirements and properties of modern semiconductor electrostatic discharge (ESD) protection devices, specifically in connection with the 100BASE-T1 and 1000BASE-T1 standards, and explores how these devices, when synergistically integrated with the rest of the circuitry, contribute to the creation of a robust system that defends against destructive ESD events and electromagnetic compatibility (EMC) issues.

Standards and testing
Ethernet solutions have a long-standing presence in industrial and computing applications. However, the automotive industry has traditionally been less inclined to embrace this technology. Nevertheless, recognizing the immense potential of automotive Ethernet, the One Pair Ethernet Network (OPEN) Alliance committees, along with the Institute of Electrical and Electronic Engineers (IEEE), have collaborated to develop standards specifically tailored to meet the unique demands of the automotive sector. In 2016, the 100BASE-T1 and 1000BASE-T1 standards were drafted, and in more recent years, the OPEN Alliance has been actively working on developing two additional standards, namely 10BASE-T1s and MGB-T1. The OPEN Alliance, comprising multiple technical committees, strives to standardize Ethernet-based technologies in the automotive market. At the same time, the IEEE has taken charge of the 100BASE-T1 and 1000BASE-T1 standards, addressing automotive requirements, particularly those related to EMC. This collaborative effort ensures that Ethernet technology in the automotive domain aligns with industry-wide standards and undergoes rigorous testing to ensure compliance and performance.

ESD protection requirements for 100BASE-T1 and 1000BASE-T1
Automotive applications greatly benefit from the flexibility offered by Ethernet connections. Star topologies can employ these connections, where a switch serves as a central point, interconnecting multiple domains such as Advanced Driver Assistance Systems (ADAS), infotainment, and more. Alternatively, Ethernet can be utilized in a bus topology, similar to the traditional CAN and FlexRay applications (Figure 1).

Figure 1: Typical configuration of Ethernet nodes in a modern vehicle

Notably, the standardization of 100BASE-T1 and 1000BASE-T1 employs unshielded twisted pair (UTP) cables (Figure 2), commonly used in the automotive industry due to their widespread availability and cost-effectiveness. However, cables present challenges related to EMC behavior, particularly regarding electromagnetic interference (EMI) in bundled cable configurations.

Figure 2: Two Ethernet nodes are connected using an unshielded twisted pair (UTP)

In a modern vehicle, numerous electrical units are interconnected by hundreds of meters of cables, ranging from simple climate control units to powerful generators. These cables are often bundled together, increasing the risk of electromagnetic interference (EMI) between them. Extensive investigations have revealed that, in worst-case scenarios, this EMI can induce peak voltage amplitudes of up to 100 V in unshielded twisted pair (UTP) cables. This poses a challenge as stable data transfer is required during normal operation, highlighting the need for robust Ethernet circuitry to withstand these EMC issues.

The circuitry of each node in the Ethernet system, standardized by the OPEN Alliance, includes a common mode choke (CMC) to filter out unwanted common mode noise that couples in the UTP. In conjunction with common mode termination, the CMC addresses these EMC concerns. The properties of the CMC for 100BASE-T1 and 1000BASE-T1 are defined in the CMC Test Specifications specific to these standards. Apart from its filtering and EMC characteristics, the CMC also plays a crucial role in electrostatic discharge (ESD) protection, which will be discussed in the following section.

Figure 3: Circuitry of the 100BASE-T1 and 1000BASE-T and the ESD performance of the ESD device

From the perspective of ESD protection devices, there are several important considerations. Firstly, considering the potential electromagnetic noise on the UTP, the ESD device should remain inactive within a voltage range of up to 100 V. In terms of ESD device parameters, the triggering threshold for the ESD device is set above 100 V, as shown in Figure 3 of the Transient Voltage Suppressor (TVS) graph. While this voltage threshold may seem high, it is important to note that the specific circuitry configuration, including the CMC, provides robust protection for the physical layer (PHY) components, such as high-performance cameras and displays.

Secondly, the ESD protection device is subjected to rigorous testing to ensure its robustness, with requirements of withstanding 15kV ESD discharges for a minimum of 1000 cycles. This stringent requirement underscores the criticality of reliable operation for Ethernet-based applications in the demanding automotive environment. Combined with the 24V operating voltage, similar to that of CAN applications, these specific requirements create a unique set of criteria, as shown in Table 1.

Additional ESD testing for OPEN Alliance ESD protections
In addition to the requirements outlined in Table 1, more tests need to be performed to ensure the effectiveness of OPEN Alliance ESD protections. ESD vendors typically conduct these tests, and their results can be valuable for hardware design engineers.

Table 2: Additional ESD testing for OPEN Alliance standards

When it comes to 100BASE-T1 and 1000BASE-T1, the tests are quite similar, but they have different pass criteria. The first two tests focus on the impact of the ESD protection device on signal integrity (SI) in automotive Ethernet applications. It is important to assess how the device affects SI by measuring insertion loss (IL), return loss (RL), and common mode rejection ratio (CMMR), as shown in Figure 4. To ensure compliance, specific limits for these parameters are provided in the “Specification of the ESD protection for 100BASE-T1 and 1000BASE-T1” document.

In the automotive domain, there is a new factor to consider — the ESD discharge current, which quantifies the current flowing into the PHY during an ESD event. It is crucial to evaluate this current to ensure proper protection. Additionally, RF clamping is employed to simulate the noise on the UTP, taking into account the > 100 V requirement.

By conducting these additional tests, engineers can gain a deeper understanding of how the ESD protection devices perform in terms of signal integrity and current flow, ensuring the robustness and reliability of the Ethernet system in automotive applications.

Figure 4: S-parameter results for an ESD protection device for 1000BASE-T1, including the limits in yellow

Optimal placement, routing, and layout techniques
Let’s dive into the crucial aspects of placing, routing, and laying out the ESD protection devices in an Ethernet design. It’s not just about the performance of the ESD device itself; its implementation on the PCB plays a vital role. Figure 3 gives us a visual representation, highlighting the importance of placing the ESD device at the connector. This strategic placement ensures that any ESD pulse is immediately clamped down to ground right at the connector, safeguarding the entire circuitry, including the CMC, CMT, and PHY. For a tangible demonstration of the significance of placement, refer to Figure 5.

Figure 5: Field scan of the Ethernet circuitry during an ESD event.

In Figure 5, a field scan of the Ethernet circuitry during an ESD event reveals areas of high current density, indicated by red. The OPEN Alliance (OA) approach demonstrates that having the ESD device close to the connector results in the lowest current density at the PHY location. Consequently, the entire circuitry exhibits optimal ESD performance for the system.

It’s important to grasp the role of the CMC in reducing ESD stress for the PHY. To understand this, let’s examine the behavior of the CMC under pulsed conditions, as depicted in Figure 6. We utilize the transmission line pulse (TLP) method to illustrate the blocking and saturation phases of the CMC. When a transient pulse, such as an ESD pulse, heads towards the CMC, it initially blocks the current for a specific period of time during the blocking phase (phase II), with the peak (in phase I) being a measurement artifact. The duration of the blocking phase depends on the voltage level of the pulse — the higher the voltage, the shorter the blocking phase. After the blocking phase, the CMC enters the saturation phase (phase III), acting as an inductor driven into saturation by the pulse. Once saturation occurs, the CMC starts conducting the current, leading to a drop in voltage across it.

Figure 6: Current and voltage response of a typical CMC for 1000BT1 applications based on TLP measurements.

This intriguing finding demonstrates that when an ESD pulse approaches the 100BASE-T1 or 1000BASE-T1 circuit, the CMC effectively blocks the current for the first few nanoseconds. Simultaneously, the voltage across the ESD protection device increases. When the voltage reaches the trigger level, typically around 140 V (Figure 3), the ESD device clamps the ESD pulse to ground. As a result, the voltage across the entire circuit drops to the clamping voltage of the ESD device, typically ranging from 30 to 40 V (see TLP plot in Figure 3). This showcases the synergistic effect of combining a high-trigger ESD protection device with a CMC during ESD events. It’s worth noting that a CMC with an inductance in the range of approximately 100µH exhibits sufficient blocking behavior, which is already covered by the CMC specification.

ESD protection devices are typically available in various packages. One widely used package is the SOT23, a common and established choice in automotive applications. Another option is the DFN1006BD (SOD882BD), a leadless package. Figure 7 illustrates different options for routing the differential lines to and from the package, and their rankings are provided in Table 3.

Table 3: Ranking of the different routing options with regard to ESD and SI

When routing the ESD packages, it’s generally advisable to keep the routes straight, avoiding any stubs or bends. Specifically, for ESD considerations, the traces of the differential lines should pass over the pad of the ESD device, as depicted in options [A] and [C] for SOT23, as well as for DFN1006BD in option [A]. It’s crucial to avoid stubs and ensure that the impedance of the differential lines remains at 100Ω to maintain signal integrity (SI). This can be achieved by keeping the lines separated. For SOT23, options [B] and [C] are preferable, while for DFN1006BD, option [A] is the recommended choice.

Figure 7: Routing options for SOT23 and DFN1006BD (SOD882BD)

As a general rule, try to minimize unnecessary layer changes to achieve optimal signal integrity. Layer changes will inevitably impact SI and electromagnetic compatibility (EMC). If you must change layers, route the signal over the pad of the ESD device, as shown in Figure 8 (left and right). Avoid routing signals via stubs, as depicted in Figure 8 (middle).

By following these guidelines for placement, routing, and layout, you can effectively enhance the performance, reliability, and protection of ESD in your Ethernet design.

Final thoughts
The article emphasizes the unique circuitry and ESD protection requirements in 1000BASE-T1 and 100BASE-T1 applications. It highlights the strong synergy between ESD protection devices and the blocking capability of the CMC, resulting in a highly robust Ethernet system that effectively combats EMC noise and ESD. The significance of positioning the ESD protection directly at the connector is underscored through an EMI scanner. Additionally, it should be noted that discussions within the OPEN Alliance committees are underway for an additional standard, 10BASE-T1S, which is expected to have similar requirements regarding the high trigger voltage, given the similarity in topology with 1000BASE-T1 and 100BASE-T1. Further research and development in this area will contribute to the continuous improvement of ESD protection solutions and their role in supporting automotive Ethernet applications.