On-Site EMC Assessment for Very Large Power Conversion Systems

Client Context

The client is a small to medium enterprise developing cutting-edge, very high-power power converters for industrial applications. One of the key challenges associated with such high-power systems is their physical size. In this case, the complete installation consists of an active front end (AFE), battery storage, an innovative energy storage device, and multiple variable speed motor drives. Combined, the system occupies a volume of approximately 5 m × 4 m × 3 m.

Challenges

Testing a system of this size in a conventional EMC chamber is both technically and commercially challenging. For radiated emissions, true far-field measurements—particularly at the lower end of the frequency range—would require a 10 m test distance. A 3 m chamber, while commonly used, does not strictly satisfy far-field conditions for such a large installation.

In addition, transporting and reinstalling a system of this scale into an EMC facility would be complex and costly. For many companies facing similar constraints, an on-site EMC assessment provides a practical alternative. Based on the assessment results, manufacturers can make informed design decisions and, where appropriate, proceed via the manufacturer self-declaration route to meet EU EMC Directive and UKCA requirements.

Scope of Work

The manufacturer engaged Mach One Design to carry out an on-site EMC assessment. Given the size and complexity of the system, a one-day assessment was planned, focusing on emissions only. The scope included:

• Conducted emissions

• Radiated emissions

• Harmonics

With the available test equipment, the objective was to perform an early-stage EMC evaluation that would provide a clear indication of whether the system was likely to pass or fail formal compliance testing in an EMC chamber or open area test site.

Test Methodology

Figure 2 shows the use of a three-phase LISN to measure conducted emissions from the three-phase system. The LISN was rated at 32 A per phase, which limited the maximum operating power during testing. A common question raised in such situations is whether conducted emissions measured at reduced power levels are representative of full-power operation.

In practice, experience shows that conducted emissions in high-power converters are often dominated by common-mode mechanisms rather than load current. Common-mode voltage, generated by fast switching transitions, couples through parasitic capacitances associated with the power devices, heatsinks, and enclosure. The resulting common-mode current follows:

I=C⋅dVdtI=C⋅dtdV​

Since the switching dV/dt is largely independent of load current, and the parasitic capacitance is fixed by the mechanical and electrical design, conducted emissions measured at low load are often a good indicator of worst-case behaviour.

As a rule of thumb, if a system demonstrates at least a 10 dB margin below the limit lines under low-power operation, there is generally good confidence that it will pass conducted emissions across its full operating range.

That said, this assumption does not apply universally. Some converters operate in different switching modes depending on load, and motor drives may change control strategies under varying mechanical conditions. These factors must always be considered during interpretation of the results.

Radiated Emissions Assessment

Although a 10 m measurement distance would normally be required for a system of this size, site constraints meant that radiated emissions were measured at an approximate distance of 3 m, using a single antenna location and fixed height. A biconical antenna was used, as the majority of emissions were expected to fall within the 30–300 MHz range.

Figure 4 shows the biconical antenna in a vertically polarised configuration. One of the main challenges when performing radiated measurements in industrial environments is ambient noise. Figure 4 also illustrates a typical factory or office noise floor. In this case, the ambient noise level was more than 10 dB below the relevant limit line, making it feasible to perform meaningful EMC assessments on site.

To further improve measurement confidence, the following steps were taken:

• Identification and exclusion of known external transmitters, such as FM and DAB radio signals

• Switching off non-essential equipment, particularly lighting and air-conditioning systems, which are common sources of broadband noise

• Adjustment of resolution bandwidth to reduce the noise floor and help distinguish emissions originating from the system under test

In addition to antenna measurements, RF current probes were used where appropriate. For systems of this nature, radiated emissions are often cable-driven, making current-based diagnostics particularly effective.

Outcome

The on-site EMC assessment allowed clear identification of key EMI risks and “red flags” within the system. Based on these findings, targeted recommendations were provided to guide the next stage of EMI mitigation and design improvement.

The resulting EMC assessment report also provides supporting technical evidence that the manufacturer can use as part of a self-declaration process for EMC compliance.