CAN bus chokes from EPCOS prevent EMC problems in automotive networks

The FlexRay consortium has designated the SIMDAD 1812 as reference type for conformity tests. The proportion of electronics used in automobiles has risen significantly in recent years.

This trend is continuing with the integration of additional functions designed to increase safety, efficiency, reliability and convenience as well as lower emissions. The same trend is associated with the increasing demands made on the bus systems to ensure reliable communications between the most diverse control units. In particular, safety-critical applications controlled via bus systems such as controller area networks (CAN) or FlexRay systems must satisfy the highest EMC requirements. Common mode chokes (CMC) in the data lines offer improved protection from malfunctions caused by EMC problems.

Complex control functions in today’s motor vehicles are characterized by ever-faster data rates, deterministic behavior and fault tolerance. Whereas cost-critical or multimedia functions are connected via LIN (local interconnect network) or MOST (media oriented systems transport) buses, CAN or FlexRay bus systems are used for safety-critical functions such as engine management, ABS, airbag, etc. CAN and FlexRay use twisted pairs, are deterministic and support fast data rates – up to 1 Mbit/s for the CAN bus and up to 10 Mbit/s for FlexRay. For these two bus systems, physical layer and data transfer protocols are optimized to assure high reliability. In view of the increasing complexity of modern vehicles, however, these measures alone cannot completely prevent malfunctions due to EMC problems.

Automotive bus systems must satisfy high EMC requirements in terms of high immunity from transients, electrostatic discharge (ESD) and electromagnetic interference (EMI). But the bus systems must not interfere with other electronic components, which means that interference emissions must be minimized. As a result of the increasing proportion of electronics in vehicles, however, EMC cannot be tested in advance under all conditions, which implies the risk of malfunctions or even damage to the control units. Data line chokes from EPCOS help to make sure this doesn’t occur.

Generally, one must distinguish between differential mode and common mode interference. Differential mode interference is superposed on the data signal. Common mode interference on the other hand is related to ground and and is produced by asymmetries and and parasitic effects. To minimize common mode interference, special attention must be paid to the layout of the bus signal leads, the termination filters, connectors and to the circuit board itself. Parasitic capacitances and inductances of the through-contacts or connector contacts as well as the layout of the bus signal leads on the circuit board can lead to asymmetries and thus to common mode interference.

A common method for the EMC measurement of the RF influences is the Te direct power insertion (DPI) method: a signal from a signal generator increasing in power up to 36 dBm is coupled into the bus leads and the signal outputs are observed. If a fault occurs the signal level of the signal coupled in can then be recorded. This process is repeated in steps for each corresponding frequency in the relevant range. The RF interference emissions for a defined bus configuration are determined by measuring the common mode voltage at the bus and at all inputs and outputs with an EMC test receiver.

The RF immunity and interference emissions were measured on test boards for the DPI measurement (Fig.1) with and without EPCOS chokes. Figs. 2 and 3 show the corresponding results for the CAN bus. It is obvious that the RF immunity is greatly improved by the use of B82789 common mode chokes. At the same time, the implementation of the chokes in data lines significantly reduces interference emissions.

Data line chokes can significantly increase the reliability of CAN bus systems. That is why European automobile manufacturers require their use by suppliers. Although chokes of higher inductance produce better EMC results due to their higher attenuation, additional aspects such as scatter inductance, signal integrity, ground shift, ruggedness of the wiring, etc. must also be considered when selecting an ideal choke. The B82789 series from EPCOS offers inductances in the range from 11 to 100 µH.

EMC and EMI for FlexRay

FlexRay is a serial, deterministic and fault-tolerant bus system that was defined by the FlexRay consortium founded in 2000 by BMW, DaimlerChrysler, Motorola (today Freescale) and Philips (today NXP); Bosch, General Motors, Volkswagen and others joined later. FlexRay is designed to satisfy the tougher requirements of future automotive networks, especially for fast data rates, real-time capability and failsafe behavior. However, current focus is on faster data rates.

The physical layer application note specifies the general requirements on data line chokes for FlexRay networks such as lead resistance (≤ 1 Ω), inductance (> 50 µH) and scatter inductance (< 1 µH). The measurements for electromagnetic radiation and EMI were carried out as for the CAN bus. However, in view of the higher transmission rate of 10 Mbit/s, the signal integrity was examined more closely. Fig. 4 shows the corresponding evaluation results of the DPI measurement for chokes with various inductances in comparison with a test circuit without chokes. Fig. 5 shows the results of the EMI evaluation obtained with another test board, again for chokes of different inductance. Here too, it is clear that use of data line chokes improves EMC and reduces interference emissions. This effect becomes increasingly significant at higher inductances. However, higher inductance values have a negative impact on signal integrity.

An aperture diagram is used to investigate the signal integrity for FlexRay. If the curve of the data signal is within the aperture, then secure data transmission is guaranteed. These requirements are fulfilled when using the B82789C0104 choke with an inductance of 100 µH (Fig. 6). In view of the results of the EMC measurements and the negligible influences on the data signal, in other words the guaranteed signal integrity, the EPCOS B82789C0104 N002 (bifilar winding) choke was defined as the reference for the physical layer conformity test for FlexRay.

Model libraries facilitate development

EPCOS provides suitable simulation models for efficient designing on the basis of the new chokes. To these ends their electrical characteristics were determined for the entire relevant frequency range and incorporated in the respective models. Because terminal capacitances and inductances and connecting resistances are all dependent on customer designs they are not included in the models. This allows customers to evaluate the function of a choke without the time-consuming design of an entire system. Simplified models that allow fast and accurate simulation are available for a first design estimation. Advanced models supply more exact simulation results, but require a longer simulation time.

For more information on the full range of EPCOS products available from Anglia please call +44 (0)1945 474747 or email info@anglia.com

Alternatively, click on this link to go to the main EPCOS section where you can view other news articles and product data.

back to product news