Access Official Technical PDF

Share Summary:LinkedInEmail

Choosing a DC link voltage for electric aircraft propulsion is a system-level decision, not simply a way to reduce current in a cable. Voltage affects transmission losses and conductor sizing, but it also changes the requirements for motor insulation and inverter components. A coupled design space exploration of the CENTRELINE fuselage fan drive illustrates how these interactions can shift the overall balance of mass and efficiency.

Why DC link voltage matters in electric propulsion

Electric propulsion gives aircraft designers more options for arranging power generation and distribution than conventional engine architectures. That flexibility also creates more interactions between components: choices made for the electrical network can affect the motor, power electronics, cabling and ultimately the propulsion system’s weight and efficiency.

Higher voltage is attractive because, for a given power, it allows lower current. Lower current can reduce resistive losses and the conductor cross-section needed to transmit power. However, a voltage increase does not benefit every component equally. Insulation requirements grow, and inverter design depends on the voltage ratings of available semiconductor devices and the topology used.

Aircraft electrical systems also face operating constraints that differ from many ground-based installations. The study notes that conventional aircraft systems have been limited to ±270 V DC to avoid corona at typical flight altitudes, while higher power demands have motivated investigation of higher-voltage distribution. Selecting a higher level therefore requires considering both electrical performance and equipment design.

What the CENTRELINE drive-unit study examined

The analysis focused on the fuselage fan drive unit in the CENTRELINE turbo-electric aircraft concept. Its propulsion architecture uses a fuselage fan at the rear of the aircraft, driven by an electric motor. The assessed subsystem included the motor, DC/AC inverters and DC transmission cables; other aircraft and propulsion components were outside the model’s scope.

The drive unit required 8 MW of shaft power. Power was carried through eight parallel DC lanes, with each lane serving a corresponding inverter and motor winding system. The cable route was 86 metres long. The design exploration varied DC link voltage from 1,500 V to 3,000 V and compared component designs using system mass and efficiency as key performance indicators.

Rather than treating each component as an independent estimate, the approach linked analytical models for electrical, thermal and structural considerations. This matters because a voltage choice can influence several design steps at once. For example, motor insulation affects slot space and heat transfer; the motor’s electrical characteristics inform inverter sizing; and the DC voltage directly affects transmission current and cable dimensions.

How voltage changes motor, inverter and cable design

Motor insulation creates a trade-off

In the modeled permanent-magnet synchronous motor, higher DC link voltage increased the thickness of slot, phase and turn insulation. The resulting increase in required slot area made the motor heavier. Thicker insulation also impedes heat transfer, affecting cooling-channel requirements.

The motor’s efficiency changed only slightly across the voltage levels considered. The study attributed that small reduction mainly to increased core losses as the iron mass rose. In other words, reducing current in the wider system did not automatically make the motor lighter or more efficient.

Inverter results depend on topology and device ratings

The inverter design was evaluated across several two-, three- and five-level topologies. Its mass and efficiency did not vary smoothly with voltage. Semiconductor modules have discrete blocking-voltage ratings, so a small change in the selected DC link voltage can require a different module. That change can produce a noticeable step in inverter performance.

Within a range where the selected modules remain unchanged, higher voltage can reduce current-related demands, including some capacitor requirements. But moving to a device with a higher blocking-voltage rating can increase mass. Consequently, inverter topology and available semiconductor ratings need to be assessed alongside the voltage target—not after it has been fixed.

Cable mass falls as transmission voltage rises

The cable showed the clearest benefit from increasing voltage. Lower current reduced conductor losses and allowed a smaller conductor cross-section under the modeled thermal constraints. Aluminum was selected for the transmission cables because its conductivity-to-weight relationship made it preferable to copper for this application. The study also found that insulation became a larger share of total cable mass at higher voltage, suggesting that the cable benefit may not continue indefinitely as voltage rises.

System-level results reveal the dominant effect

When the motor, inverters and cables were assessed together, higher voltage generally enabled lower total system mass and higher efficiency across the explored range. The cable’s mass reduction outweighed the modest increase in motor mass and the voltage-dependent changes in inverter mass. For the aircraft configuration studied, cable mass was a substantial share of the drive unit’s total mass, making transmission design a major system-level driver.

The comparison also showed why voltage levels should not be ranked on cable performance alone. Around a semiconductor selection boundary, the inverter’s mass and efficiency could change enough to affect the combined result. A voltage that appears favorable for cabling may therefore be less attractive once the inverter and motor are included.

Practical implications for engineering teams

For engineers and system integrators evaluating electric power architectures, the study points to several useful practices:

  • Model component interactions early. Include motor insulation, thermal behavior, inverter characteristics and cable sizing in the same assessment.
  • Check semiconductor rating boundaries. Discrete device options can create performance changes that a smooth, voltage-only estimate will miss.
  • Compare system KPIs. Evaluate total mass and efficiency together, rather than optimizing one component in isolation.
  • Make model boundaries explicit. The assessed system omitted other components, including thermal-management equipment, and identified further model refinement as necessary. Results should be interpreted within that scope.

Potential ways to reduce the influence of transmission voltage on the motor and inverter include placing DC/DC converters between the rectifier and inverter, or investigating superconducting technologies. These are possible architectural directions, not conclusions that were validated by the sizing exercise.

Conclusion

DC link voltage selection involves balancing transmission advantages against the insulation, thermal and power-electronics consequences of operating at higher voltage. In the CENTRELINE drive-unit analysis, lighter DC cabling was the dominant factor behind improved system-level results, even though motor and inverter designs did not always improve. The practical lesson is to assess voltage within a coupled propulsion-system model, while accounting for component constraints and the limits of the analysis. As models expand to include additional equipment and aircraft-level integration, the resulting comparisons can better inform architecture decisions.

Access Official Technical PDF

About The Author