Recent technological developments have caused a renewed interest in the brushless excitation system. With the application of wireless communication, the conventional diode bridge has been replaced with fully controllable thyristors on the shaft. It offers the same dynamic performance as the conventional static excitation system. The thyristor bridge of the conventional three-phase exciter needs to be controlled with a high firing angle in normal operation in order to fulfill a requirement of both a high ceiling voltage and a high ceiling current. A high firing angle causes high torque ripple to be absorbed by the exciter stator and a low power factor results in a low utilization of the designed exciter. In this contribution, we present a strategy that solves this problem by looking into combinations of thyristor configurations of a double-star six-phase connection of the exciter. Experimental results are used to verify the circuit models implemented for this investigation. A hybrid-mode 12-pulse thyristor bridge configuration seems to be a good solution for implementations in commercial apparatus. An additional switch interconnects two separate thyristor bridges from parallel- to series connection at the rectifier output, and utilizes the advantages of both topologies.

his paper investigates the performance of different power electronic interfaces for a rotating brushless permanent magnet exciter, designed for a synchronous generator test setup. A passive rotating diode bridge is commonly used as the rotating interface on conventional brushless excitation systems. Those systems are known to be slow dynamically, since they cannot control the generator field voltage directly. Including active switching components on the rotating shaft, like thyristors or transistors, brushless excitation systems can be comparable to static excitation systems. Brushless excitation systems has the benefit of less regular maintenance. With permanent magnets on the stator of the designed exciter, the excitation system improves its field forcing capability. Results show that modern power electronic interfaces utilize the exciter machine optimally, increase the power factor, reduce the torque pulsations, maintain the available field winding ceiling voltage and improve the field winding controllability.

The static exciter is dominating among large grid-connected generators due to the weak dynamic performance of conventional brushless exciters. In this paper, a six-phase outer pole permanent magnet rotating brushless exciter is evaluated with different active rectification topologies. Both thyristor-based and chopper-based topologies are considered. A fast-response brushless excitation system is obtained by replacing the conventional rotating diode bridge rectifier with the proposed active rectification topologies on the shaft. The given two-stage system generates its own excitation power directly from the shaft, contrary to static exciters. The selection of an appropriate rectification topology could minimize the rotor armature phase currents for a given generator field current. The objective is a high power factor and a high utilization of the exciter machine. An optimal rectification topology makes higher ceiling currents possible, improving the transient behavior of the synchronous generator. In this paper we show that six-phase topologies add complexity, but improve exciter redundancy, increase the available ceiling voltage and reduce the steady state torque ripple. Experimental results are given for validating the models implemented for the analysis.

We calculated analytically and with the finite-element method the force between two identical coaxial cylindrical magnets for nine different magnetic materials at separations of 1, 3, 5, 7, and 9 mm and for height-to-diameter ratios between 0.1 and 2. For the analytical calculations, we assumed homogeneous magnetization. For the nonhomogeneous case, we used the finite-element method taking into consideration the magnetic flux density vs. field (B-H) curves of the materials. The results show that, similarly for cuboid magnets, the maximum force for a given volume of permanent-magnet material was achieved for a height-to-diameter ratio around 0.4 for homogeneously magnetized (ideal) permanent magnets. We also show that this is not always true for the nonhomogeneous cases. The resulting forces are dependent on the B-H curve of the magnets. As the B-H curves of the materials deviate from their ideal counterparts, the aspect ratio that yields maximum force increases.

Permanent magnets are an attractive material to be utilized in thrust bearings as they offer relatively low losses. If utilized properly, they have a long service lifetime and are virtually maintenance free. In this contribution, we communicate the results of the tests performed on a permanent magnet thrust bearing that was custom built and installed in a hydropower synchronous generator test-rig. Tridimensional finite element simulations were performed and compared with measurements of axial force. Spin down times and axial force ripple have also been measured. We found good correspondence between the measurements and the simulations.

We developed a simple method to calculate the axial force between concentric thin walled solenoids. To achieve this, the force between them was mapped as a function of their geometrical relations based on separation-to-diameter ratios. This resulted in an equation and a set of data. We used them together to calculate axial forces between two coaxial thin walled solenoids. With this method, direct evaluation of elliptical integrals was circumvented, and the forces were obtained with a simple expression. The results were validated against solutions obtained with an existing semi-analytical method and force measurements between high coercivity permanent magnets.

Synchronous motors are reliable and efficient, but it is relatively difficult to start them. In some cases, a variable frequency drive is utilized. In some other, asynchronous start is achieved by virtue of induced currents in a solid rotor, or a rotor damper cage installed for this purpose. In this contribution, a method to start a synchronous machine without a damper cage is presented. The starting was achieved by inverting the polarity of the rotor field winding in a timely manner with respect to the rotating stator field. The technique was verified with experiments performed on a 200 kVA experimental test rig and also simulated on a 20 MVA machine.

Unbalanced magnetic pull (UMP) is typically caused by rotor or stator shape defects, electrical short circuits, eccentric rotor/stator bores, as well as unreasonable pole-slot combinations. It leads to vibration and increases noise and energy losses of the machine. By actively controlling the magnetic fields and forces that arise between the rotor and stator by regulating the rotor field current of separated pole groups, it is possible to cancel it. In this paper, we measure and calculate the currents induced in the damper bars for a synchronous machine test rig under 20% static eccentricity with and without active compensation of UMP. This is done to validate our finite-element calculations. Afterward, we perform loss calculations for a 74-MVA synchronous generator with and without stator parallel circuits. We find that, with active compensation of UMP for an eccentric machine, the damper bar currents and stator parallel circuit circulating currents can be eliminated and the electromagnetic efficiency of the machine that has a static eccentricity fault increases.

Generally, PM machines are used as PMG pre-exciters in 3-stage brushless excitations systems. This paperpresents the design, characterization and prototyping of a rotatingbrushless PM exciter used in a proposed 2-stage excitation systemfor a synchronous generator. The proposed design reduces thenumber of components compared with conventional systems.A comparison with the state-of-the-art conventional excitationsystems is given. The design of a fast-response, or high initialresponse, brushless exciter requires active rectification on therotating frame, replacing the non-controllable diode bridge. Theobjective was to construct an exciter with the capability of a50 Aoutput field current as well as a high value of the available ceilingvoltage and ceiling current. The final exciter was constructed to befitted into an in-house synchronous generator test setup. A finiteelement model of the exciter was validated with experimentalmeasurements. The exciter prototype is also compared with analternative armature design with non-overlapping single-layerconcentrated windings but with the same main dimensions.The paper includes a general design procedure suitable foroptimization of PM brushless exciters that fulfill the requirementsof their synchronous generators and the grid.

Two types of passive magnetic lift bearings were evaluated in terms of thrust force and eddy current losses. The first type of bearings were based on two sets of segmented Halbach arrays mounted in repulsive mode, and the second type was based on ring-magnets. The eddy-currents studied arose in the bearing due to manufacturing variations of magnetic remanence, and due to non-radial magnetization. Both a 3D time-dependent and a quasi-stationary Finite-Element Method (FEM) formulation were used, and the simulated results were compared with lift-force measurements from experiment. The losses were found (by FEM) to be in the order of 25 W at a rotational speed of 30000 rpm while lifting a 45 kg rotor with a stiffness of 359 N/mm.

This paper deals with the characterization and construction of a rotating brushless PM exciter intended for synchronous generator excitation purposes. Traditionally, PM exciters are used as pre-exciters in synchronous generator excitations systems. In order to reduce the number of components and to increase the step time response of the system, a PM exciter is designed as an outer pole PM machine, with permanent magnets on the stator and armature windings on the rotor. The exciter was constructed electrically and mechanically to be fitted into an in-house synchronous generator test setup. A finite element model of the exciter was validated with no-load measurements of voltages and magnetic flux densities. The exciter was then characterized with unsaturated and saturated parameters.