Lately, more and more electrical bicycles with firmly inserted engines have appearde on the market. The Power Wheel, which will be developed by us, is based on another concept. The idea is to integrate the electrical motor into the rear wheel. With this system the owner will be able to decide between electrical and normal drive. The change occurs by exchanging the rear wheel with the Power Wheel.

The aim of the diploma thesis was to clarify whether the Power Wheel, built with a switched reluctance motor according to the transverse magnetic flux principle, can be implemented and if possible to develop a model. Our diploma thesis bases on the project thesis II/99 " SRM Power Wheel ". With the results of the following clarifications, we were able to deduce a mathematical model, which we used to simulate the behavior of the Power Wheel, and to check the requirements.

Whilst working on our project we discovered that product requirement specifications cannot be fulfilled with an SR-motor. With these simulations we found out that the engine should have two rows of motorpoles in order to even partially fulfill the requirements. The main problem lies in the build-up time of the current at high velocities. The reason is the high lever inductance. Additionally, the rotational speed of the motor is relatively small. Thus, huge forces are needed per pole. With these restrictions the required force support of 100 % is given only up to a velocity of 20 km/h, instead of the required 35 km/h. The weight increases because of the additional row of poles, up to 12.5 kg.

Compared with the electrical bicycles available on the market, the Power Wheel does not reach the same level, and therefore the Power Wheel will never be produced with an SR-motor.

Background: Mobile concrete pumps have a tubing system, by which the concrete is pumped to its destination. The convweyor line is carried by a pylon with a vertical swivelling axis and individual swivelling arms. These arms are connected by joints, so that by changing the individual angles of joint and turning of the whole distribution pylon the end of the tubing system can be positioned at will.

Aim: The aim of this diplomaproject was, to develop a control, which permits the end of the tube to be moved on any desired course without each joint being moved individually. The angles of the joints are to be given by the control, while at the control unit, only the desired direction and rate are chosen. The control can be demonstrated by a computer simulation of the pylon.

Function of the control: The control determines the new rated position of the point of the pylon from the actual angles of the joints and the control instructions. Based on this, the angles which belong to the rated position, are determined by kinematic calculations and geometrical correlations. The angles are now supplied to the automatic controller as rated value and the virtual model is adjusted to the rated position. The virtual distribution pylon reports the actual values of the angles to the control. Additionally in the calculation of the rated values of the angles, it had to be guaranteed that position specifications are intercepted and corrected accordingly, if they are not feasible with the mechanical construction. The complete control system was implemented in Matlab/Simulink and the virtual distribution pylon has been modelled in ADAMS. Thus the simulation runs in Simulink, whereas for the presentation and animation of the model, ADAMS is called automatically.

Rieter AG manufactures so-called drawroll for producing thread. The new thread runs over the hot, fast rotating cylinder. Because the temperature of the cylinder is strictly controlled and the temperature profile at the surface of the cylinder is flat the thread is always of the same quality. Inside the cylinder there is an inductor which produces a magnetic flow . This magnetic flow produce an eddy current in the cylinder. The eddy current warms up the cylinder.

Through earlier corporation between the companies NM GmbH and Rieter AG there is a rough numerical finite-element-model, which simulates the behaviour of the drawroll. The Software SESES was used to produce the model.

The goal of this diploma thesis was the further development of the old numerical model. With that model the development of drawroll at Rieter AG support and sped up.

The main task was to optimise the numerical model by using results of measuring produced by Rieter AG, so that the calculation of the temperature in the cylinder corresponds in the data from measurement. When the influence from the different parameters for the temperature on the cylinder was known, it was possible to adjust the parameters. It could be achieve that the model corresponded with the measured data.

A next task was to expend the inductor, which is normally a cylinder, to the form of a bone. This could be reached. The shape of the inductor is now adjustable with different parameters.

In the inset there is a wet thread which runs over the surface of the drawroll. The water of the wet thread evaporates, which make an additional load. The task to expend the model with that load could be performed.

The present model can now be used for the development of new drawrolls. With additional measuring the current model could be optimized.

The additional tasks control of the temperature and sensitivity referring to data of material and geometry could not be done, because the other tasks needs too much time.

The firm Genge & Thoma AG in Lengnau/Biel develops and produces different Joysticks. They want to improve their products with the help of magnetic circuits. One of the most important advantages of using a magnetic circuit is to avoid the abrasion of the electrical contacts. The development and simulation of methods of detecting the right angle of the joystick has been done in an earlier work.

The new idea of the company is to use this magnetic circuit to detect the zero position of the stick. As a result the aim of this project is to simulate the induction field below the iron ball in the centre of the Joystick and to increase the gradient of this field by small angles.

To achieve this, I used the software SESES developed by the NM GmbH. With this program it is possible to develop a three dimensional model of the iron ball, the contacts and the air around it. This model consists of a lot of small meshes so that the method of finite elements can be used to simulate the system. The first result of the simulation is that the induction below the iron ball is three times smaller than that one between the ball and the contacts. The gradient is unfortunately to small too be detected by the sensor used in detecting the zero position.

The next step is to develop possibilities to increase the change of the field. I also investigated the consequences of a change in the width of the airgap on the induction field. The most promising solution is to add an iron stick and to equalise the potential. In this case the sensor can detect the change of the induction field.

These suggestions are based solely on the simulated model. In practice, fixing the sensor at the surface of the top of the iron stick and achieving equalisation of the potential in the real system are still open problems.