Non-contact digitization of objects and surfaces with optical sensors based on fringe or pattern projection in combination with a CCD-camera allows a representation of surfaces with pointclouds equals x, y, z data points. To digitize the total surface of an object, it is necessary to combine the different measurement data obtained by the optical sensor from different views. Depending on the size of the object and the required accuracy of the measured data, different sensor set-ups with handling system or a combination of linear and rotation axes are described. Furthermore, strategies to match the overlapping pointclouds of a digitized object are introduced. This is very important especially for the digitization of large objects like 1:1 car models, etc. With different sensor sizes, it is possible to digitize small objects like teeth, crowns, inlays, etc. with an overall accuracy of 20 micrometer as well as large objects like car models, with a total accuracy of 0.5 mm. The various applications in the field of optical digitization are described.

The goal of optical 3D-measurements is the determination of Cartesian coordinates of surfaces. Using principles basing on fringe projection techniques coordinates are calculated of measured phases in fringes, image-coordinates of cameras and parameters of the system configuration. Mostly, three measured values unambiguous lead to one coordinate-tripel. A more comfortable way offers photogrammetric measurements where more than three values are measured to calculate coordinates and additional to parameters describing the system configuration. Applying these experiences to fringe projection techniques it is possible to combine the advantages of both techniques. That guarantees high number of object points, quick data acquirement and a simultaneous determination of coordinates and system parameters with photogrammetric bundle adjustment. Recent measurements show the capabilities of that principle. The reliability of the coordinates is given by a standard deviation of less than 10 microns while the object diameter was up to 280 mm.

Three-dimensional-metrology based on photogrammetric and topometric techniques is a powerful tool for the digitization and measurement of complex three-dimensional scenes and objects. Since several years advanced sensors and measurement systems are available for industrial applications. Especially the integration of topometric systems into measuring- and handling machines is supported by compact and light 3D- sensors. These sensors can be optimized for specific measuring tasks with respect to accuracy, field of view and further parameters. During the last two years one is going to describe both techniques by the same algorithms. Moreover, there are first approaches of 'topogrammetric' systems, that combine photogrammetric and topometric metrologies, especially by using calibration techniques that are well known in photogrammetry and which allow the on-line calibration of 3D- sensors. On the other hand the topometric projection of coded light provides a continuous indexing of the whole measuring scene where photogrammetric methods (without active illumination) are limited to a lower number of discrete index marks.

An approach to improve the quality of offline programming for laser beam welding is described. A CAD-dataset is combined with technological information using a feature model. A feature consists of the basic geometry, process parameters, a set of strategies in which way it can be processed, and rules to select the optimum strategy depending on the boundary conditions. The resulting welding task is represented by a list of features from which an NC-dataset is generated, containing all process information. The aim of the development is to design a feature based technology module which is integrated into a flexible, fault tolerant and process near planning tool.

Rapid prototyping technologies -- especially stereolithography and plastic laser sintering -- did substantially contribute in reducing product development times. But models that are produced by rapid prototyping technologies do not meet the requirements of functional prototypes, as neither the serial material nor the serial production processes are used. This concerns injection molding and pressure die-casting prototypes. Therefore rapid tooling techniques are necessary. The main emphasis of this presentation is on the significance of rapid tooling for integrated product and process development. This is discussed using the example of evolution tooling (tools that grow with the development status). Furthermore, a survey is presented of the most important rapid tooling processes that employ rapid prototyping technologies. A distinction is made here between the process chains for rapid prototyping and conversion processes and the direct manufacture of tool components. The IWB User Center (Anwenderzentrum) in Augsburg, Germany is the first European user of the indirect metal laser sintering process. In order to communicate what has been learned from initial experience with this procedure, sample tools that have been realized are described.

To support an expert planning lasercells to advise prospective buyers, the Institute for Machine Tools and Industrial Management (IWB) developed a lasercell-planning-system (LaPlaS), which includes tools for configuration and evaluation. Configuration is supported by a database module. Therefore a framework was developed which allows the modeling of lasercells for cutting and welding. Moreover methods to represent knowledge of former plannings were introduced. LaPlaS supports the expert by using this knowledge based data to select matching lasercells or laser components. On the other hand the expert has the possibility to employ his own experience by selecting lasercells or laser components with user defined search criteria. Evaluation is supported by a simulation module. A three dimensional model is used to represent the selected lasercell and to examine the working motion. Data gained from simulation lead to statements about cycle time, collision and motion behavior as well as other evaluation criteria without the need of a real cell. Moreover the simulation module may be used for automatic layout optimization. Therefore functions were developed which generate the optimal lasercell layout based on user defined optimization criteria as minimal cycle time for example.

The demand of machine users towards the complete processing of a workpiece in one setting requires the use of different processing technologies in one machine. In the past years lasers have found new applications in production engineering as a tool for surface modification, cutting, welding and marking. By combination of conventional metal cutting technologies with laser processes in one machine, the complete processing of a workpiece with different technologies in one setting can be realized. The advantages are the processing in one setting and the reduction of material flow between the production machines. The possibility of integrating lasers for material processing in a cutting machine are examined. Processing technology as well as different constructive solutions depending on the type of machine used and integration method are evaluated. Parallel to this approach, examinations of the economy of these systems are carried out.

Increasing laser beam power for industrial material processing causes increasing demands on the beam guiding system and the optical components used. Especially the thermally induced deformation of the optics influences the beam propagation at higher power levels. This in particular is true for beam guiding systems which often include a large number of mirrors and/or long beam paths, where the diverging effect upon the beam accumulates with every single mirror. The thermally induced deformation of reflective optics is strongly related to the power and diameter of the incident laser beam. Therefore, the propagation of a beam through a system which is subject to thermally induced deformation is of nonlinear nature. The calculation of the beam propagation as well as the design of beam guiding systems for high laser powers has to take this effect into account. In order to include the diverging effect of a deformed reflective mirror in the calculation of the propagation, it is necessary to develop an analytical description of the relation of the beam parameters and the optical effects upon the beam and the beam-mirror interaction, respectively. Our approach for an analytical description is based on a finite-element analysis of laser mirrors being deformed by various beams with different beam powers, beam diameters and energy distributions. The calculated surface deformations were analyzed using a new method, which allows the calculation of the resulting focal length of a deformed surface in relation to the diameter and energy distribution of the incident beam. The numerical results of this analysis are compared with measurements. For minimizing the effect of the thermally induced deformation, a new cooling concept for metal mirrors has been investigated numerically and experimentally. The results of this optimization are discussed in detail.

To increase the flexibility of the generative LAPS-J (laser aided power solidification -- powder jet) process, a special focusing device has been developed. The beam of a Nd:YAG laser is delivered into the production machine via a step-index glass fiber. To utilize the resulting top-hat intensity distribution, the end facet of the fiber is imaged on the workpiece by a special optical system, consisting of four lenses. It allows the variation of the scale of imaging between 4:1 and 1:1. By the computer controlled movement of two motors it is possible to change the width of the generated tracks in this range during laser processing. Integrated in a turning center, the LAPS-J proces allows new and complex applications, e.g. in the fields of cladding, rapid prototyping or repairing of metal parts. With an additional process control, the quality and accuracy of generated metallic parts can be considerably increased.

Complex three-dimensional parts can be manufactured directly from CAD data using rapid prototyping processes. SLS selective laser sintering is a rapid prototyping process developed at the University of Texas at Austin and commercialized by DTM Corporation. SLS parts are constructed layer by layer from powdered materials using laser energy to melt CAD specified cross sections. Polymer, metal, and ceramic powders are all potential candidate materials for this process. In this paper the fabrication of complex metal parts rapidly using the investment, die and sand casting technologies in conjunction with the selective laser sintering process are being explained and discussed. TrueForm and polycarbonate were used for investment casting, while RapidSteel metal mould inserts were used for the die casting trials. Two different SandForm materials, zircon and silica sand, are currently available for the direct production of sand moulds and cores. The flexible and versatile selective laser sintering process all these materials on one single sinterstation. Material can be changed fast and easily between two different builds.

Complex three dimensional parts can be manufactured directly from CAD data using the SLS^R selective laser sintering process. In addition to the currently used nylon materials, two new materials have been developed for the production of functional models. From an elastomeric polymer, commercially available as Somos 201, highly flexible, rubber-like parts can be produced directly in the sinterstation system. Also, a new product for producing stiff functional prototypes will be introduced in July, commercially available as VeriForm polymer. Its surface quality and detail definition is superior to the currently used nylon materials while the mechanical properties could be kept at the same level. Comparative data of VeriForm and the nylon materials are presented. For the production of real technical prototypes injection moulding inserts can be produced by the RapidTool process that can be employed for the production of up to 80,000 parts depending on the injected material and part geometry. By changes in the chemical composition and by new processing parameters, the average cycle time could be reduced from 4 to 5 days down to 2 to 3 days.

In this paper two processes of microfabrication technology are investigated. In the first part, laser generating process which is largely affected by the beam interaction time is described. In the second part, selective laser sintering of metallic powder shall be discussed. The results show that the two metal liquid phase sintering is most effective in the selective laser sintering of the metallic powder comprising of high and low melting point. Whereas in the generating process the beam interaction time greatly affect the structural development of the product with respect to its strength and quality. A few examples are demonstrated briefly.

Laser-assisted machining operations, i.e. the use of laser beam energy to enhance the cutting properties via local plastification of the material during machining operations, can be successfully used to cut materials which are difficult to machine. The following article shows the technological potential of this process highlighted by the results obtained through research laser-assisted turning of advanced ceramics. These studies indicate that laser-assisted hot machining of materials which are normally regarded as 'difficult to machine' decreases cutting forces and tool wear, thus increasing material removal rates.

The principle of laser cladding involves the use of high power carbon-dioxide lasers and powder deposition technology to provide wear and corrosion resistant surface coatings to engineering components. By injecting metal powder into a laser generated melt pool on a moving substrate a solidified metal track can be produced. Deposition of successive tracks produces a multi-layer build. Laser direct casting (LDC) utilizes a coaxial nozzle enabling consistent omnidirectional deposition to produce 3D components from a selection of metal powders. The influence of the principal process parameters over the process features namely, powder catchment efficiency, beam shape and build rates are presented with several successfully generated 3D components. Nickel, stainless steel and satellite powders were deposited at laser powders of 0.4 to 1.4 kW and speeds of 500 to 1000 mm/min achieving build rates of 3 to 9 mm³/s. Fully dense metallurgical structures have been produced with no cracking or porosity and powder catchment efficiencies up to 85% have been achieved.

The use of air bearings as friction free devices is state of the art in measurement machines. However in many applications air bearings cannot be used because of their insufficient dynamic characteristics. This disadvantage occurs in common air bearings because of pockets and chambers that are worked into the surface in order to improve their static behavior. Air bearings with optimized static and dynamic characteristics have been developed at the Lehrstuhl fuer Feingeraetebau der TU Muenchen in 1986. However the reliable production of these air bearings was very difficult because of high sophisticated working steps. In 1995 a new technique was developed and patented to manufacture dynamically optimized air bearings. This new technique allows a mainly automated production of the bearings without loosing any performance. The key is to bring a large number of nozzles with aid of a laser beam. The combination of this new manufacturing tool and the knowledge of more than 40 years of research has lead to air bearings with best dynamic behavior at a lower production price.

High speed 3D profiling is a key issue for many quality control systems in the manufacturing industry. Three- dimensional profiling in speeds that match the production demands of today's industries is very hard to achieve using conventional sensor technology. The solution presented in the paper is a sheet-of-light range imager based on a smart vision system. The 3D profiling system is based on a programmable smart vision sensor. The sensor consists of an image sensor, AD-converter, and RISC-processor integrated on the same silicon chip. The processor, being a line parallel bit-serial SIMD machine, handles both binary and gray scale information efficiently and has proven to be very suitable for sheet-of- light range imaging.