In the automotive industry, the pursuit of Electromagnetic Compatibility (EMC) plays a pivotal role in ensuring the reliability and performance of vehicles. Unlike the CE marking commonly used in other industries, the automotive sector adopts E marking for its products. To gain approval for vehicle sales, automakers adhere to standards such as UNECE Regulation 10, which is currently in its 6th revision.

While UNECE R10 serves as the baseline, automakers often supplement these standards with their in-house vehicle-specific EMC standards. For example, Daimler AG has its own vehicle specification MBN 10284-1, which outlines vehicle-level EMC tests. Automakers frequently collaborate with electric control unit (ECU) manufacturers, often referred to as Tier-1 manufacturers, providing them with stringent component or subsystem EMC specifications. This ensures that when assembling vehicles, automakers have a higher likelihood of meeting vehicle-level EMC requirements.

This top-down approach is a common practice in manufacturing large and complex products/systems, including those in aerospace, space, and military applications.

Emissions

Emissions tests for components and subsystems are typically based on CISPR 25 standards. While each automaker may slightly adjust limits based on CISPR 25, they often closely align with the most stringent CLASS 5 of CISPR 25. Notable component-level EMC standards include Ford FMC1278 and FMC280, JLR-EMC-CS, and Volvo STD 515-0003. These standards require that the test set-up and procedures align with the specifications defined in CISPR 25.

As battery technology matures, the automotive industry is experiencing a profound transformation, with electric vehicles (EVs) rapidly replacing internal combustion engine (ICE) vehicles. Consequently, EMC standards must evolve to accommodate this technological shift. An example of this evolution is found in CISPR 25, which now includes high-voltage (HV) tests to address components like battery packs, DC-DC converters, on-board chargers (OBCs), and electric motors, all of which are now standard components in vehicles.

Immunity

In the automotive realm, immunity tests primarily reference ISO standards rather than IEC standards, except for specific components such as on-board chargers that must withstand IEC immunity tests (as this is the only component that is connected to the mains network). Key ISO standards include the ISO 7637-x series, ISO 11451, and ISO 11452 series.

As automotive technology evolves, some previously relevant tests lose their relevance. For instance, the load dump test, which simulates a sudden battery disconnection from the alternator, becomes obsolete as EVs no longer have alternators. Instead, EVs incorporate OBCs that require compliance with standards such as IEC 61000-3-X. Moreover, newly developed immunity tests address high-voltage transient events frequently encountered in HV systems. For instance, disconnecting a HV contact may trigger EFT/Burst-type transient events within the vehicle's HV system.

As the automotive industry continues to embrace electric and hybrid technology, EMC standards will continue to adapt to meet the demands of this ever-evolving landscape.

Which Ground? - An EV case study

Before the era of EVs, vehicles typically did not have a designated 'ground' connection because they were intended to travel on the ground, not be attached to it. However, with the advent of EVs, there is now a 'ground' connection when the vehicle is being charged, which poses an interesting challenge for EV EMC, particularly in the case of the OBC conducted emission test.

The OBC, which is the primary unit in operation during the charging process, is bonded to the vehicle chassis. When testing the OBC as a component, the test 'ground' point is referenced to the Line Impedance Stabilization Network (LISN) earthing point, which is directly bonded to the test ground, this test ground represents the chassis of a vehicle in this case. But when testing a vehicle for conducted emissions while it is plugged into the mains network, the test 'ground' point is referenced to LISN earthing point, which represents the mains earthing point. In this case, the vehicle chassis does not share the same 'ground' as the test ground (while in the component test, it does). This presents a significant challenge for the EV industry. When a supplier tests its OBC, it follows component-level EMC standards (such as CISPR 25), where the unit under test is bonded to the test ground plane, representing the vehicle chassis. But when a vehicle manufacturer tests the OBC, there are non-conductive elements like wheels between the vehicle chassis and the test ground plane.

This means that a pass in the conducted emission test for an OBC at the component level may result in a fail at the vehicle level. This discrepancy is commonly observed in test houses. Simply grounding the vehicle chassis to the test ground plane would show improvement, but it is not practical in real-world applications.

To address this issue, vehicle manufacturers often require OBC suppliers to achieve more than a 6 dB (better to be 10 dB) pass margin, ensuring they can pass vehicle-level tests.

Another interesting observation is that on the power input port, the conducted emissions sweep between 150kHz to 30 MHz, based on the IEC 61000 series requirements. However, on the power output port and signal port, the conducted emissions often sweep between 150kHz up to 110 MHz, aligning with CISPR 25 requirements. This is because the emissions from the output port and signal port could potentially affect ECUs in the vehicle, all of which are bonded to the chassis. Therefore, automotive standards apply.

Since the vehicle remains stationary during the charging process and connects to the national supply grid, the equipment used in this charging mode must comply with commercial testing standards such as Harmonics & Flickers, commercial CE, Surge, and EFT testing.

On the other hand, equipment that is active during the normal mode (non-charging mode) must adhere to automotive standards, limits, and test levels.