Automotive capacitive sensing electronic system compatibility test

The compatibility of electronic systems in products is subject to electromagnetic compatibility (EMC) testing. Electronic products must pass a series of specific tests before they can be sold and used in certain regions. Automotive electronic systems have dedicated EMC testing because electronic subsystems need to function properly near other electrical devices that generate noise. The vibration and temperature ranges in the automotive environment also require additional certifications defined by independent chip-level certification processes such as AEC-Q100.

EMC is the ability of an electronic system to operate properly without adversely affecting the performance of other electronic systems. ISO and IEC compatibility specifications include both radiated (transmitted through the air) and conducted (through wiring harnesses) tests. Each test covers both the emissions (radiated outward from the system) and immunity (external sources that affect the subsystem under test) capabilities of an electronic subsystem.

Therefore, there are four different EMC tests, each corresponding to a different EMC test standard:

1. Conducted immunity test (ISO11452-4)
2. Radiated immunity test (ISO 11452-2)
3. Conducted emission test (CISPR25)
4. Radiated emission test (CISPR25)

In fact, there is a fifth test, called the electrostatic discharge (ESD) test, which is often discussed along with the EMC tests listed above. But this test is distinct from the others—it uses an electrostatic discharge gun to inject static charge into a subsystem to measure its immunity. This article will take a closer look at the first four EMC tests.

Below, we will discuss EMC testing in more detail using capacitive sensing applications as an example. In addition, this article will introduce some custom EMC tests specified by regional automotive manufacturers (such as Toyota, BMW, Volkswagen, etc.). We will also introduce some system-level aspects of designing electronic systems compatible with the automotive environment and the challenges encountered in the industry when conducting EMC testing.

Fault level overview

EMC testing is to evaluate whether the system under test and other surrounding systems can operate normally in a noisy environment. Some EMC failure symptoms in the automotive environment are as follows:
1. Non-critical electronic equipment failure when the vehicle is started (for example: car entertainment system, power window control, interior lighting, etc.)
2. Critical electronic equipment such as engine control unit (ECU) or braking system fails due to power surge during load dump
3. The sound output of the car's car entertainment system is like the sound of the engine or starter motor starting
4. Permanent damage to the circuit during the jump start of the car
5. Electronic system errors and unexpected restarts due to electrical interference

The severity of the above symptoms is confirmed by the automobile manufacturers through the fault classification. A typical example of fault classification is shown in Table 1.

It is important to note that severity does not represent the importance of the fault. All systems, critical or non-critical, can experience any of the above categories of faults. However, critical systems such as ECU, brake control, airbag control, powertrain, etc. have different levels of fault acceptance. These systems must not have any faults, not even level 5 faults. Non-critical systems such as multimedia, in-car entertainment systems, power windows, etc. tend to have a lower level of acceptance. Some level 4 and level 5 faults are usually acceptable, depending on the overall impact of the fault on the system. The level of acceptance is determined by the manufacturer based on the possible consequences of the fault on each subsystem of the car and the cost of resolving non-critical issues.

Certification of automotive integrated circuits

All electronic components, especially semiconductor integrated circuits used in automotive embedded systems, must be able to operate properly in the harsh automotive environment. Equipment standards such as AEC-Q100 are developed by the Automotive Electronics Council and serve as guidance documents for semiconductor manufacturers to certify their equipment for automotive use.

The following lists some of the main parameters of integrated circuits tested in AEC-Q100 certification:

1. Parts operating temperature test (normally up to 125°C)
2. Accelerated life test
3. Stress test
4. Wire and solder ball shear test
5. Human body model (HBM) ESD test and latch-up test
6. Non-volatile memory endurance test
7. Short circuit reliability evaluation

Since automotive systems have strict requirements for the reliability of electronic components, system designers must select integrated circuits that are AEC-Q100 qualified. Given the complexity involved, it is recommended to contact the manufacturer to obtain specific information about AEC-Q100 qualified parts that are not covered in this article. Manufacturers also provide detailed documentation and application support for products to pass automotive EMC testing.

EMC test methods at the system level

EMC tests can quantitatively test whether the system under test is suitable for use in a car. The system must also be "quiet" enough during operation so as not to interfere with other surrounding devices. In robustness or immunity testing, external noise is injected through the wiring harness (conducted immunity) or the RF antenna (radiated immunity). During the noise injection, the operation of the system is observed. Noise of different frequencies is injected, and the intensity of the noise is controlled according to the provisions in the standard specifications.

To quantitatively test the conducted immunity of the system under test, the noise voltage spectrum on the harness is plotted as a graph according to the spectral intensity. In the radiated immunity test, the noise spectrum is obtained through the corresponding RF antenna located at the distance specified by the standard specification. In either case, the spectrum must be below the qualified limit specified in the specification. Table 2 lists some common automotive EMC specifications.

In addition, automakers often develop their own specifications, most of which are more stringent than the general standards. Table 3 lists some of the automakers' self-developed automotive EMC standards.

Conducted immunity test

The conducted immunity test for the automotive environment is the most stringent of all EMC tests, and therefore the most challenging to pass. In the conducted immunity test, noise is injected into the system through the wiring harness. The system must be able to operate normally in the noisy environment.

Figure 1 is a capacitive sensing user interface for controlling a passenger car HVAC control unit - Cypress PSoC system. In the ISO11452-4 EMC standard test, the RF noise generator injects noise into the power harness of the system under test through the RF coupling clamp. The frequency range of the noise is 1MHz to 400MHz, and the intensity is the 200mA loop current in the RF coupling clamp. See Figure 2.

For example, capacitive sensing interfaces such as the Cypress CapSense interface have replaced mechanical buttons in many automotive user interfaces. The following faults can be used to determine whether a capacitive touch sensing system is functioning properly:

1. No false touches, i.e. the system does not detect touch input when a person is not touching the input sensor.
2. No drop in touch sensitivity, i.e. the user does not need to touch the input sensor hard for the system to sense touch input.
3. No stuck, i.e. the system should not
get stuck in the state of detecting touch after the user stops the touch input. 4. The system should not restart intermittently when EMC noise is injected.

To pass the conducted immunity test, care must be taken in designing the system, including the design of the voltage regulator circuit and the use of circuit protection components such as transient voltage suppression (TVS) diodes. In particular, the design of a low-ripple automotive power supply that provides a stable voltage to the subsystems from the 12-volt automotive battery is critical to product compliance. Figure 3 shows the output voltage ripple of the power rail in the ISO11452-4 test, with the blue color representing the non-automotive power supply circuit Rev1 board and the red color representing the carefully designed automotive power supply Rev2 board.

The ripple graph is plotted according to the noise frequency injected during the conducted immunity test. As shown in the figure, if the power supply circuit is well designed, the ripple in the 100MHz area will be greatly reduced. The object of this test is the Cypress PSoC4 capacitive sensing system designed for the automotive HVAC user interface control unit.

Another immunity test, ISO7637-2 transient immunity test, is a test specific to automotive systems. This test simulates the typical electrical noise generated on the wiring harness during in-vehicle events such as jump start, load dump, ignition start, reverse battery connection, etc. Using appropriate protection devices such as TVS diodes and series inductors in the power path can help reduce the impact of transients on the system under test.

Radiated emission test

In an automotive environment, radiated emissions testing is particularly important for systems that switch frequencies, such as touchscreen in-car entertainment systems and RF transceivers including in-car Bluetooth and key systems. Radiated emissions testing uses an RF antenna to measure the emission spectrum of the system under test. A spectrum analyzer is used to measure the spectrum intensity (FFT amplitude). Figure 4 shows the laboratory setup for measuring radiated emissions in an anechoic chamber.

The test frequency and its specified amplitude limits are based on the respective EMC standards. As the latest automotive systems generate higher amplitudes and frequencies, the highest frequency in today's EMC standards has reached 5GHz. In most cases, it is not possible to measure emissions over a wide range of frequencies using only one RF antenna, so multiple antennas are required when measuring multiple frequency bands. In this process, the antenna impedance of all antennas must be equal so that the entire frequency range can be compared. Currently, vehicle manufacturers are using broadband butterfly antennas that cover the entire frequency range. This is shown in Figure 5.

One way to reduce emissions at higher frequencies is to control the rise time of the switching signals that cause the emissions. Another approach is to use spread spectrum techniques to sweep the resulting frequencies so that all the energy is evenly distributed over a wider frequency band. Sometimes manufacturers integrate these methods with specific parts. For example, to effectively meet radiated emission standards, Cypress offers both technologies in its capacitive sensing technology.

Radiated immunity test

Radiated immunity testing also uses the same antennas as the emission test. In the test, RF noise is injected into the system under test through the radiating medium (air). The passing criteria are similar to the conducted immunity test mentioned above. Methods such as shielding and conductive chassis are used to reduce the impact of radiation on the system under test.

Conducted emission test

In conducted emissions testing, the system injects noise into the power lines (wiring harness) that often serve as a common rail for multiple subsystem buses. In most cases, the automotive electronic subsystems are not the main source of electrical noise on the wiring harness. If they generate significant conducted noise, the results can be improved by designing the system with appropriate bypass capacitors at the power entry point.

Challenges of EMC Compatibility Standards

As electronic subsystems perform critical functions in modern cars, compatibility standards have become increasingly stringent. Many OEMs have mandated that even non-critical electronic subsystems must pass EMC immunity testing and must not exhibit any failures, including the Level 5 failures mentioned above. In addition, automotive EMC testing has become more and more involved, with each subsystem function being fully tested at a variety of noise frequencies. Non-critical features, such as the subsystem's user response time and low-power modes, also need to be fully tested and must not exhibit any deviations, even those that would not be visible to the user without EMC testing.

Since current EMC testing takes hours, an automated EMC testing process is needed. For example, a complete EMC test of a capacitive sensing subsystem can take more than 3 hours for conducted immunity testing alone, and more than 20 hours for the entire suite of EMC tests. Since mobile and Internet of Things (IoT) applications are often located in close proximity to the system under test (e.g., a user brings a mobile phone or device that connects to a WiFi wireless network into a car), the frequency bands in EMC radiated immunity and emission testing have expanded beyond 5GHz.
Most EMC is performed in an anechoic chamber, where the walls are made of absorbing materials that prevent radiated emissions from reflecting. This is an ideal environment that is completely different from the actual environment in which the automotive subsystem is located. Therefore, some vehicle manufacturers also use metal chambers or tuned reverberation chambers to deliberately create reflections toward the system under test. In this way, they can study the impact of these reflections on system functionality and immunity.

It is well known that the design difficulty of automotive electronic subsystems is increasing, especially in terms of compatibility. In addition, controlling costs and minimizing the increasing number of radiation sources make the design more complicated. Sometimes, the end user may bring incompatible or non-automotive grade devices into the car, which will affect the function of the system. This is difficult to detect through standard EMC testing. A common example is the car entertainment system making a strange sound when an incompatible mobile phone in the car receives an incoming call.

EMC Infrastructure Challenges

The equipment used in EMC testing is very expensive. Usually, ordinary automotive Tier 1 suppliers cannot purchase equipment for their own use, and the usual practice is to rent space and equipment from certified EMC laboratories. But the hourly rental cost is also very high. In addition, if the EMC test fails, the design engineer needs to modify the system (hardware, firmware, etc.) and incur additional costs to test it again in the laboratory. Due to the limited number of EMC laboratories and the large number of Tier 1 suppliers waiting for the laboratory's "schedule", it takes a long time from the designer's application to actual use, which further affects the overall time to market.

One way to reduce the cost of using external labs and increase efficiency is for companies to tune systems at their internal design sites using pre-certified equipment. Pre-certified equipment cannot measure intensity precisely, but it can provide an approximate noise spectrum that designers can use to fine-tune the system.

Product compatibility and electromagnetic compatibility of automotive electronic subsystems have become an increasing challenge. Electronic products must focus on compatibility from the design stage. Testing is difficult due to limited in-house RF expertise and test equipment. For this reason, simply using compliant chips is not enough for companies to ensure that they can pass automotive subsystem EMC testing. Chip manufacturers need to actively support their chip products. Although application notes can describe the best way to achieve EMC compliance, each subsystem is unique. Therefore, you must have experienced application engineers to help you review the hardware schematics and PCB layout, and tune firmware parameters.