Magnetic springs are a fatigue-free alternative to mechanical springs that could enablecompliant actuation concepts in highly dynamic industrial applications. The goals of this article are:(1) to develop and validate a methodology for the optimal design of a magnetic spring and (2) tobenchmark the magnetic springs at the component level against conventional solutions, namely,mechanical springs and highly dynamic servo motors. We present an extensive exploration of themagnetic spring design space both with respect to topology and geometry sizing, using a 2D finiteelement magnetostatics software combined with a multi-objective genetic algorithm, as a part of aMagOpt design environment. The resulting Pareto-optima are used for benchmarking rotationalmagnetic springs back-to-back with classical industrial solutions. The design methodology has beenextensively validated using a combination of one physical prototype and multiple virtual designs.The findings show that magnetic springs possess an energy density 50% higher than that of stateof-the-art reported mechanical springs for the gigacycle regime and accordingly a torque densitysignificantly higher than that of state-of-the-practice permanently magnetic synchronous motors.

For high-speed bearingless disk drives, certain topologies seem advantageous. The authors have published works on a bearingless disk drive for high speeds that is characterized by a slotless stator and a toroid winding set. Several different variations of this setup are imaginable. The ones which this paper focuses on are the variation of the number of phases, of the number of coils, and of the applied coil connection. A comparison of the constructed and tested setup with the possible variations is presented in the course of this paper.

Optimization of total cost of ownership (TCO) is an important, and challenging design target for present day manufacturing machines. This paper is concerned specifically with production machines with fast reciprocating loads (> 1 Hz), e.g. weaving looms and plate punching machines. Subsequent acceleration and deceleration give rise to a reciprocating energy flow that can be handled either mechanically or electrically. The chosen solution will affect the total cost of ownership. In addition to the cost of the energy storage device itself, there are the energy bill, the size and cost of the electric drive and power supply to consider. Moreover, there are certain constraints to be met: lifetime, DC-bus voltage limits and total power factor. This paper presents a methodology that takes all these aspects into account. It applies it to a bar linkage mechanism, which is representative for the targeted applications. In the mechanical domain, springs are considered for energy storage. The structural design of the spring is included in the analysis in order to account for lifetime and inertia added by the spring. In the electric domain, three different topologies are compared: a purely passive front end, where energy is stored directly on the DC-bus, a passive front end combined with a DC/DC converter and a separate storage capacitor, and an active front end.

In several industrial applications, drivetrains impose highly dynamic oscillating motions to inertial loads. By introducing springs, the system's natural frequencies can be matched to the required operating frequencies, lowering energy consumption of the drivetrain. However, xed-stiness springs only have a positive eect in a limited range of frequencies. To solve this problem, variable stiness springs are proposed.A discussion of the eect of the series and parallel topology, as well as the possibility of adapting the spatio-temporal characteristics of the motion to the spring, is presented. Furthermore, a practical implementation of a variable stiness spring is proposed. Its eectiveness is validated in experiments.