EMI troubleshooting & filter design for an automotive DC-DC converter

Client Context

A UK-based electronics design company developed a 2.4 kW DC-DC converter for electric vehicle applications. The company’s background was primarily in medical and industrial electronics, where EMC requirements are typically less stringent than those imposed by the automotive sector.

When transitioning into automotive design, meeting CISPR 25 requirements proved challenging. The first hardware revision failed CISPR 25 conducted & radiated emissions testing significantly, performing well above Class 1 limits. For automotive DC-DC converters of this power level, CISPR 25 Class 3 is generally considered the minimum acceptable compliance target.

The client lacked in-house expertise in both automotive EMC testing and mitigation strategies and therefore engaged us to diagnose the root causes and implement robust, production-ready solutions.

Scope of Work

• Perform bench-top conducted emissions testing at the client’s premises

• Use near-field and RF current probe techniques to predict far-field radiated emissions

• Identify dominant EMI sources across the system

• Design and implement EMC fixes, including filters, PCB layout improvements, and enclosure-level considerations

• Achieve CISPR 25 Class 3 compliance with confidence ahead of accredited lab testing

Technical Approach

While conducted emissions testing is often associated with screened rooms or accredited laboratories, this is not strictly necessary for accurate diagnostics. Using our calibrated bench-top EMC setup, we were able to achieve conducted emissions measurement accuracy comparable to that of an accredited EMC facility. This approach allowed rapid iteration and real-time troubleshooting directly at the client’s site.

For radiated emissions, although a full anechoic chamber was not available, RF current probe measurements were used to estimate far-field radiation levels. This method provides a reliable indication of compliance risk and is particularly effective during early troubleshooting stages.

Initial conducted and radiated emissions measurements were captured and documented to establish a clear baseline for comparison as design fixes were introduced.

Filter design was initiated using circuit-level simulation in SIMetrix and subsequently validated with VNA measurements—an approach that forms part of our in-house EMC design methodology.

Troubleshooting & Mitigation

We applied a systematic investigation using a combination of near-field probes and RF current probes to localise EMI sources across the four PCBs within the product.

As expected, the dominant noise originated from the main power conversion stage. The converter employs a dual-active-bridge (DAB) topology using SiC devices, which offers high efficiency but presents significant EMC challenges due to fast switching edges and high dv/dt.

Transformer behaviour proved to be a critical factor in the overall EMI performance. Our mitigation strategy included the following key actions:

Key Actions

• Transformer characterisation - The parasitic capacitance between primary and secondary windings was measured and found to be excessive, leading to strong common-mode noise coupling. A revised transformer design incorporating an electrostatic shield was developed to significantly reduce this coupling.

• Main power stage optimisation - The SiC switching PCB layout was refined to minimise loop areas, and the snubber network was tuned to control voltage overshoot and ringing.

• Input and output filter design - Dedicated EMC filter stages were designed and implemented to provide sufficient attenuation across the CISPR 25 frequency range.

• Control board grounding fixes - A grounding issue on the control PCB—originating from the layout of an on-board buck-boost converter—was identified and corrected.

• Spread-spectrum modulation

  
  

Each modification was validated incrementally using the same measurement setup, allowing fast confirmation of effectiveness before proceeding to the next step.

Results & Validation

Following implementation of the design improvements, the device was retested using the same bench-top EMC configuration. The results showed that:

• Conducted emissions were reduced to well within CISPR 25 Class 3 limits

• Radiated emissions were significantly lowered, providing high confidence of passing accredited laboratory testing

• A comprehensive technical report was delivered, documenting measurement data, root causes, design changes, and EMC rationale

• This documentation now serves as a reference guide for future product revisions and additional automotive programmes.

Conclusion

This case study demonstrates how targeted EMC expertise and in-situ troubleshooting can transform a non-compliant automotive power converter into a robust, standards-ready product. By addressing EMI at the source—through transformer design, PCB layout, filtering, and grounding—we enabled the client to reduce certification risk, avoid costly redesign cycles, and move confidently toward market release under stringent automotive EMC requirements.