F: Elektrotechnik und MedientechnikHochschulschrift
Elmar Pitschke
Techniques for the Production of High Quality Lenses by the Utilisation of a Knowledge Based System
2011
Abstract anzeigen
The rapid increase in the employment of optics-based technologies in modern life has spurred the search for lenses that exhibit exceptionally high optical accuracies, and yet are simple and cheap to manufacture. Currently, most complex lenses are based on multiple lens elements of spherical form, and it is common for a number of these spherical elements to be combined into a single optical objective. However, a characteristic of spherical lenses, spherical aberration, prevents the spherical lens from achieving perfect refraction and leads to distortion of the image at the focus point. It is for this reason that multiple spherical elements are generally necessary. Aspherical, sometimes referred to as free-form, lenses enable the incorporation of compensation for spherical aberration. Such aspherical lenses may exhibit a perfect focus, will be lighter in weight and will be smaller in size than equivalent multi-spherical objectives. However, the production of asphericallenses is complex, a consequence of their intricate shape. Research was undertaken to analyse the methods for producing lenses. In particular magnetorheological finishing was studied extensively. This process utilises a fluid that is stiffened under the influence of a magnetic fluid as the polishing agent. It is an extremely capable and adaptable process, but yet is subject to many complex factors that influence the manner in which it may be applied to the optical finishing process. The magnetorheological finishing process was investigated, and procedures for synthesising the machine characteristic, referred to as the influence function, were proposed. Employment of the technique that was developed will reduce the quantity of optical material used for testing and machine characterisation, so reducing scrap, while saving the manufacturing time that must normally be devoted to the characterisation operation. In addition, a knowledge-based system was established, which may provide guidance and direction as to the most appropriate lens-processing conditions to adopt. The results of this work have generic applications, in diverse processes, wherever surface preparation and surface finishing are required - See more at: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.572873#sthash.SQKYg44i.dpuf
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenZeitschriftenartikel
Markus Schinhärl, R. Stamp, Elmar Pitschke, Rolf Rascher, L. Smith, G. Smith, Andreas Geiss, Peter Sperber
Advanced techniques for computer-controlled polishing
Current Developments in Lens Design and Optical Engineering IX, vol. 7060, no. 70600Q ff.
2008
DOI: 10.1117/12.808036
Abstract anzeigen
Computer-controlled polishing has introduced determinism into the finishing of high-quality surfaces, for example those used as optical interfaces. Computer-controlled polishing may overcome many of the disadvantages of traditional polishing techniques. The polishing procedure is computed in terms of the surface error-profile and the material removal characteristic of the polishing tool, the influence function. Determinism and predictability not only enable more economical manufacture but also facilitate considerably increased processing accuracy. However, there are several disadvantages that serve to limit the capabilities of computer-controlled polishing, many of these are considered to be issues associated with determination of the influence function. Magnetorheological finishing has been investigated and various new techniques and approaches that dramatically enhance the potential as well as the economics of computer-controlled polishing have been developed and verified experimentally. Recent developments and advancements in computer-controlled polishing are discussed. The generic results of this research may be used in a wide variety of alternative applications in which controlled material removal is employed to achieve a desired surface specification, ranging from surface treatment processes in technical disciplines, to manipulation of biological surface textures in medical technologies.
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenZeitschriftenartikel
Markus Schinhärl, Rolf Rascher, R. Stamp, L. Smith, G. Smith, Peter Sperber, Elmar Pitschke
Utilisation of time-variant influence functions in the computer-controlled polishing
Precision Engineering, vol. 32, no. 1, pp. 47-54
2008
DOI: 10.1016/j.precisioneng.2007.04.005
Abstract anzeigen
In the computer controlled polishing, a polishing tool moves in a well-defined manner across the workpiece surface in order to individually remove the surface error-profile. The commonly used technique to calculate the moving of the polishing tool is the dwell time method. Based on a constant (time-invariant) removal characteristic of the polishing tool (influence function) the amount of material to be removed is controlled via the dwell time. The longer the polishing tool is in contact with a particular area of the workpiece, the more material is removed at this position.
Mathematical basics to calculate dwell time-profiles are shown, and a new approach considering time-variant influence functions for the computer controlled polishing is introduced. The results point out that time-variant influence functions may contribute to further decrease the process time, and thus to make a computer controlled polishing process more efficient. The reduction of the process time was observed to approximately 35% using a combination of the dwell time method with time-variant influence functions.
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenZeitschriftenartikel
Elmar Pitschke, Markus Schinhärl, Rolf Rascher, Peter Sperber, L. Smith, R. Stamp, M. Smith
Simulation of a complex optical polishing process using a neural network
Robotics and Computer-Integrated Manufacturing, vol. 24, no. 1, pp. 32-37
2008
DOI: 10.1016/j.rcim.2006.07.003
Abstract anzeigen
Most modern manufacturing processes change their set of parameters during machining in order to work at the optimum state. But in some cases, like computer-controlled polishing, it is not possible to change these parameters during the machining. Then usually a standard set of parameters is chosen which is not adjusted to the specific conditions. To gather the optimum set of parameters anyway simulation of the process prior to manufacturing is a possibility. This research illustrates the successful implementation of a neural network to accomplish such a simulation. The characteristic of this neural network is described along with the decision of the used inputs and outputs. Results are shown and the further usage of the neural network within an automation framework is discussed. The ability to simulate these advanced manufacturing processes is an important contribution to extend automation further and thus increase cost effectiveness.
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenZeitschriftenartikel
Markus Schinhärl, G. Smith, R. Stamp, Rolf Rascher, L. Smith, Elmar Pitschke, Peter Sperber, Andreas Geiss
Mathematical modelling of influence functions in computer-controlled polishing. Part II
Applied Mathematical Modelling, vol. 32, no. 12, pp. 2907-2924
2008
Abstract anzeigen
Computer-controlled polishing (CCP) is commonly used to finish high-quality surfaces, such as optical lenses. Based on magnetorheological finishing (MRF), a mathematical model to calculate the polishing tool characteristic (influence function) was developed and verified experimentally. The second part of this paper describes the calculation of the distribution of material removal within the size of an influence function and is based on Preston’s fundamental polishing equation. The complete influence function model was implemented using MATLAB. The result is a user-friendly and easy-to-use software tool that enables fast computation of MRF influence functions without the current cumbersome determination procedure, and thus gives improved and more economical production of high-quality surfaces.
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenZeitschriftenartikel
Markus Schinhärl, G. Smith, R. Stamp, Rolf Rascher, L. Smith, Elmar Pitschke, Peter Sperber, Andreas Geiss
Mathematical modelling of influence functions in computer-controlled polishing. Part I
Applied Mathematical Modelling, vol. 32, no. 12, pp. 2888-2906
2008
Abstract anzeigen
Computer-controlled polishing (CCP) is commonly used to finish high-quality surfaces, such as optical lenses. Based on magnetorheological finishing (MRF), a mathematical model to calculate the polishing tool characteristic (influence function) was developed and verified experimentally. The first part of this paper introduces the model to predict the size and shape of an influence function. The second part of this paper describes the calculation of the distribution of material removal within the size of an influence function. The model supersedes the current cumbersome procedure for determining an influence function and thus results in considerably improved and more economical manufacture. Furthermore, the model enables the quality of the final surface to be enhanced when polishing complex, for example aspherical or free-form, workpiece geometries and provides the first step in the application of time-variant influence functions.
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenVortrag
Markus Schinhärl, G. Smith, Andreas Geiss, L. Smith, Rolf Rascher, Peter Sperber, Elmar Pitschke, R. Stamp
Calculation of MRF influence functions
Optical Manufacturing and Testing VII, SPIE, San Diego, CA, USA
2007
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenZeitschriftenartikel
Markus Schinhärl, G. Smith, Andreas Geiss, L. Smith, Rolf Rascher, Peter Sperber, Elmar Pitschke, R. Stamp
Calculation of MRF influence functions
Optical Manufacturing and Testing VII, vol. 6671
2007
Abstract anzeigen
Magnetorheological finishing (MRF) is a commonly used computer-controlled polishing (CCP) technique for high precision optical surfaces. The process is based on a magnetorheological abrasive fluid, which stiffens in a magnetic field and may be employed as a sub-aperture polishing tool. Dependent upon the surface error-profile of the workpiece and the polishing tool characteristic (influence function) an individual polishing procedure is calculated prior to processing. However, determination of the influence function remains a time consuming and laborious task. A user friendly and easy to use software tool has been developed, which enables rapid computation of MRF influence functions dependent on the MRF specific parameters, such as, magnetic field strength or fluid viscosity. The software supersedes the current cumbersome and time consuming determination procedure and thus results in considerably improved and more economical manufacture. In comparison with the conventional time period of typically 20 minutes to ascertain an influence function, it may now be calculated in a few seconds. An average quality improvement of 57% relating to the peak-valley (PV) value, and approximately 66% relating to the root-mean-square (RMS) of the surface error-profiles was observed during employment of the artificial computed influence functions for polishing.
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenZeitschriftenartikel
Elmar Pitschke, Peter Sperber, Rolf Rascher, R. Stamp, M. Smith, L. Smith, Markus Schinhärl
Lens production enhancement by adoption of artificial influence functions and a knowledge-based system in a magnetorheological finishing process
Optical Manufacturing and Testing VII, vol. 6671, no. September
2007
DOI: 10.1117/12.761356
Abstract anzeigen
High quality optical lenses are usually finished by magnetorheological finishing (MRF). In this process an abrasive fluid, with the ability to stiffen in a magnetic field, is used as the polishing tool in a computer-controlled machine tool. Although the machine is automated it is necessary for a skilled operator to set the machine and make judgments with regard to its operation. An investigation has been under way to examine the detailed operation of the MRF process, and the information that is necessary to establish best practice. This has resulted in the incorporation of a knowledge based system (KBS) into the machine control regime, and a methodology for the creation of artificial polishing tool characteristics, the machine influence function. The incorporation of the these elements has been instrumental in the operation of an enhanced MRF machine. This has been subject to extensive test procedures, and it has been demonstrated that the production process may be enhanced significantly and consistently. Batch production time may be significantly reduced, a figure in excess of a 50% reduction was met consistently during prolonged operation. Furthermore the incorporation of the KBS is instrumental in increasing the automation of the MRF process, reducing the levels of manual input necessary to manage machine operation.
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenZeitschriftenartikel
Markus Schinhärl, Rolf Rascher, R. Stamp, G. Smith, L. Smith, Elmar Pitschke, Peter Sperber
Filter algorithm for influence functions in the computer-controlled polishing of high-quality optical lenses
International Journal of Machine Tools and Manufacture, vol. 47, no. 1, pp. 107-111
2007
Abstract anzeigen
Computer controlled polishing (CCP) is widely used in the production of high-quality optical lenses. CCP enables surface error-profile-dependent calculation of polishing sequences prior to processing, and facilitates the cost-effective manufacture of high-quality optical surfaces. Calculation of an individual polishing sequence requires knowledge of the surface error-profile in addition to knowledge of the material removal characteristic (influence function) of the polishing tool. Measurement errors during both determination of the surface error-profile, and the influence function, may lead to an incorrect polishing sequence calculation, which in turn may result in an inadequate product quality. A new method has been developed which minimises the effects of measurement errors on the influence function. The resulting algorithm renders an influence function symmetrical and filters noisy data. Practical polishing tests with magnetorheological finishing have been performed to verify this new technique. The improvement of the peak-valley (PV) value of the surfaces polished with the symmetrical rendered influence function was observed to average 14% greater than that which related to the PV value improvement of those surfaces which were polished with the unmodified influence function. The algorithm developed is based on software and is easily implemented. Thus, artificial enhancement of an influence function is a straightforward technique to improve the result of the polishing process.
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenVortrag
Markus Schinhärl, Andreas Geiss, Rolf Rascher, Peter Sperber, R. Stamp, L. Smith, G. Smith, Elmar Pitschke
Coherences between influence function size, polishing quality and process time in the magnetorheological finishing
Current Developments in Lens Design and Optical Engineering VII, San Diego, CA, USA
2006
F: Angewandte Naturwissenschaften und WirtschaftsingenieurwesenF: Elektrotechnik und MedientechnikF: Maschinenbau und MechatronikI: Hochschulleitung und -einrichtungenZeitschriftenartikel
Elmar Pitschke, Markus Schinhärl, Rolf Rascher, Peter Sperber, L. Smith, R. Stamp, M. Smith
Simulation of a complex optical polishing process using a neural network
Robotics and Computer-Integrated Manufacturing, vol. 24, no. 1, pp. 32-37
2006
DOI: 10.1016/j.rcim.2006.07.003
Abstract anzeigen
Most modern manufacturing processes change their set of parameters during machining in order to work at the optimum state. But in some cases, like computer-controlled polishing, it is not possible to change these parameters during the machining. Then usually a standard set of parameters is chosen which is not adjusted to the specific conditions. To gather the optimum set of parameters anyway simulation of the process prior to manufacturing is a possibility. This research illustrates the successful implementation of a neural network to accomplish such a simulation. The characteristic of this neural network is described along with the decision of the used inputs and outputs. Results are shown and the further usage of the neural network within an automation framework is discussed. The ability to simulate these advanced manufacturing processes is an important contribution to extend automation further and thus increase cost effectiveness.
Version: 2.1.2-qf2