Francesco Gambelli

One brief started from a blank sheet: a motor that met the required performance while staying compatible with materials and geometries already available on the market, without oversizing the component and paying for it in cost and mass.

Another arrived with the opposite constraint: envelope and geometry fixed. The challenge became reducing losses and raising specific performance by working only on the degrees of freedom left unconstrained in the design.

Designing a DC–AC converter, the first battle was component selection, active devices and their parasitics, followed by weighing every unwanted effect and disturbance against the converter's overall budget at its operating point, all under legal and homologation constraints, with high-efficiency operation as the north star.

Bringing a set of electric-powertrain components together meant accounting for both conducted and radiated emissions in the environment of interest: changing geometries and materials, creating controlled-impedance paths, and reaching the performance the assembly required while staying inside the limits set by the standards.

On several occasions the full propulsion system of an electric vehicle had to be simulated. The mathematical representation of each component was the first challenge: the model had to be fast and accurate. Introducing non-ideal behaviour added a further layer of detail and the simulation began to converge on reality; of particular interest was a battery model that reproduced the charge and discharge cycles of a real pack.

Adding the dynamics (aerodynamic drag, the suspension-and-tyre response, rolling friction, road-surface irregularity) completed the picture around the existing motor, inverter and battery, making it possible to replay very realistic drive cycles and evaluate the system's response in credible conditions.

Finite-element simulation of electromagnetic and electrostatic fields was demanding for a specific reason: finding the best trade-off between computation time and mesh quality while keeping the results valid. It was a self-taught process, many attempts checked against the literature, completed experiments and specialised colleagues.

At the test bench, the devices under test are only half the story; the boundary conditions of the working environment are the other half. The bench is the proof that nothing can be neglected: from the cables to the direct and parasitic couplings between components, through to writing control code or squeezing the best out of what is already installed. It demands a deep grasp of both electrical and mechanical behaviour. An excellent school for the mind.

Keeping pace with new trends means choosing what can improve everyday engineering work. I explore automations, scripts and AI to make repetitive tasks more efficient and leave more time for design and analysis.

This has led me to study these tools in greater depth, with a focus on repeatable simulation workflows, structured result databases and future AI integration.