In the frame of an international R&D project, a team from Japan and Germany developed a replica process to manufacture a CFRP honeycomb mirror for visible application. The main target was to generate the optical surface with minimized effort to achieve a microroughness of better (1 – 2) nm RMS. This technology would enable the team to manufacture cost-effective mirrors for larger amounts, like for constellations. In this paper, we will present the recent results of this development and the finally achieved performance of a 300 mm light CFRP mirror. We will also present lessons learned for next steps of development to achieve such mirrors with an optical performance not only for microroughness but also for surface accuracy.
Standard interferometers are not suitable for measuring highly aspherical lenses due to the obligatory null configuration. In a tilted wave interferometer (TWI), however, the surface under test is illuminated from multiple directions by a point light source array 1,2. Thus, there is at least one source for each location on the test object, which illuminates the specimen perpendicular so an evaluable measuring signal is generated. In combination with the stitching approach, even large convex apertures can be evaluated. However, the system must be calibrated before measuring with a TWI. During this calibration, a known sphere is measured in several positions. The obtained information describes the entire imaging system with a finite number of Zernike polynomials. High-frequency structures in the system components are not covered and can lead to calibration errors. For this reason, they must be reduced to a minimum. In order to control high-frequency aberrations, the wavefront of each individual lens and the entire system must be determined. In classical methods with a return mirror, many error effects are encountered, such as air turbulences at large distances. The proposed model for the mathematical determination of wavefront aberration uses single measurements of the front and back surface. This reduces measurement uncertainties and simplifies the measuring effort. In addition, the costs for otherwise necessary return mirrors are also omitted. However, the lenses used must be verified additional for impurities or streaks3. Different measurement configurations are compared with the proposed method.
A challenge of coaxial - measurement cavity based - interferometer is to realize an interference contrast in the vicinity of one and to realize a complete elimination of the parasitic reflections. Another challenge, which also exists in non-coaxial setups, is the phase transfer function of extended measurement cavities. Ideally, the surface under test (SUT) and the reference surface (REF) are both exactly imaged onto the detector plane. In practice, SUT and REF have to be placed within the depth of field (DOF), which refers to the object space. The term depth of focus refers to the image space. To avoid confusion, the depth of field might be referred to as DOOF (depth of object field) and the depth of focus might be referred to as DOIF (depth of image field).
However, in many measurement situations, the REF is not placed within the DOOF, which is the small z-range, which is imaged onto the detector plane. Furthermore, the phase transfer function (PTF) of the REF and the image distortion of the REF are both dependent on the focal plane used to image the SUT onto the detector plane. Effects as phase deformation, image distortion and image blurring have to be taken into account when using extended measurement cavities. This can be done by using a look up table (LUT), which contains simulated and/or calibrated data. Thus, the related system error can be subtracted. A remaining challenge is an unknown object under test (OUT), which is measured by using a double path arrangement. The measured wave front depends on the two surfaces of the OUT and the position of the return mirror. For simplicity, a homogeneous substrate and a perfect return mirror might be presumed. The simulation of waves propagating within extended measurement cavities, as well as measurement results, will be discussed. In addition, the influence on the power spectral density (PSD) will be described. This is important for high end correction techniques as e.g. magneto rheological figuring (MRF) and ion beam figuring (IBF).
Measuring large surfaces interferometrically is a straight forward established technology, as long as they are concave and spherical. The situation chnages completely if aspheres and freeforms have to be measured. The application of a Tilted Wave Interferometer opens up possibilities to measure large concave surfaces of any shape without compensation optics. For the investigation of large convex aspheres, it is necessary to make use of stitching methods. Due to the freeform capability of the Tilted Wave Interefrometer, it is possible to acquire larger subapertures compared to null interferometers. Therefore measurement and computation time are reduced.
In the past, steadily increasing demands on the imaging properties of optics have led more and more precise spherical apertures. For a long time, these optical components have been produced in a satisfying quality using classic polishing methods such as pitch polishing. The advance of computer-controlled subaperture (SA) polishing techniques improved the accuracy of spheres. However, this new machine technology also made it possible to produce new lens geometries, such as aspheres.
In contrast to classic polishing methods, the high determinism of SA polishing allows a very specific correction of the surface defect. The methods of magneto-rheological finishing (MRF) [1], [2] and ion beam figuring (IBF) [3], [4] stand out in particular because of the achievable shape accuracy. However, this leads to the fact that a principle of manufacturing "As exact as possible, as precise as necessary" [5] is often ignored. The optical surfaces often produced with unnecessary precision, result at least in increased processing times.
The increasing interconnection of the production machines and the linking with databases already enables a consistent database to be established. It is possible to store measurements, process characteristics or tolerances for the individual production steps in a structured way. The difficulty, however, lies in the reasonable evaluation of the measurement data.
This is where this publication comes in. The smart evaluation of the measurement data with the widespread Zernike polynomials should result in a classification, depending on the required manufacturing tolerance. In combination with the so-called ABC analysis, all surface defects can be categorized. In this way, an analytic breakdown of a - initially confusing - overall problem is made. With the aid of cost functions [6] an evaluation and consequently a deduction of actions is made possible. Thus, for example, the isolated processing of rotationally symmetrical errors in spiral mode, setup times and machining times can be reduced while avoiding mid spatial frequency errors (MSFE) at the same time.
The trend in the optic industry shows, that it is increasingly important to be able to manufacture complex lens geometries on a high level of precision. From a certain limit on the required shape accuracy of optical workpieces, the processing is changed from the two-dimensional to point-shaped processing. It is very important that the process is as stable as possible during the in point-shaped processing. To ensure stability, usually only one process parameter is varied during processing. It is common that this parameter is the feed rate, which corresponds to the dwell time.
In the research project ArenA-FOi (Application-oriented analysis of resource-saving and energy-efficient design of industrial facilities for the optical industry), a touching procedure is used in the point-attack, and in this case a close look is made as to whether a change of several process parameters is meaningful during a processing. The ADAPT tool in size R20 from Satisloh AG is used, which is also available for purchase. The behavior of the tool is tested under constant conditions in the MCP 250 CNC by OptoTech GmbH. A series of experiments should enable the TIF (tool influence function) to be determined using three variable parameters. Furthermore, the maximum error frequency that can be processed is calculated as an example for one parameter set and serves as an outlook for further investigations. The test results serve as the basic for the later removal simulation, which must be able to deal with a variable TIF. This topic has already been successfully implemented in another research project of the Institute for Precision Manufacturing and High-Frequency Technology (IPH) and thus this algorithm can be used.
The next step is the useful implementation of the collected knowledge. The TIF must be selected on the basis of the measured data. It is important to know the error frequencies to select the optimal TIF. Thus, it is possible to compare the simulated results with real measurement data and to carry out a revision. From this point onwards, it is possible to evaluate the potential of this approach, and in the ideal case it will be further researched and later found in the production.
The current demand of even more powerful and efficient microprocessors, for e.g. deep learning, has led to an ongoing trend of reducing the feature size of the integrated circuits. These processors are patterned with EUV-lithography which enables 7 nm chips [1]. To produce mirrors which satisfy the needed requirements is a challenging task.
Not only increasing requirements on the imaging properties, but also new lens shapes, such as aspheres or lenses with free-form surfaces, require innovative production processes. However, these lenses need new deterministic sub-aperture polishing methods that have been established in the past few years. These polishing methods are characterized, by an empirically determined TIF and local stock removal.
Such a deterministic polishing method is ion-beam-figuring (IBF). The beam profile of an ion beam is adjusted to a nearly ideal Gaussian shape by various parameters. With the known removal function, a dwell time profile can be generated for each measured error profile. Such a profile is always generated pixel-accurately to the predetermined error profile, with the aim always of minimizing the existing surface structures up to the cut-off frequency of the tool used [2].
The processing success of a correction-polishing run depends decisively on the accuracy of the previously computed dwell-time profile. So the used algorithm to calculate the dwell time has to accurately reflect the reality. But furthermore the machine operator should have no influence on the dwell-time calculation. Conclusively there mustn’t be any parameters which have an influence on the calculation result. And lastly it should take a minimum of machining time to get a minimum of remaining error structures. Unfortunately current dwell time algorithm calculations are divergent, user-dependent, tending to create high processing times and need several parameters to bet set.
This paper describes an, realistic, convergent and user independent dwell time algorithm. The typical processing times are reduced to about 80 % up to 50 % compared to conventional algorithms (Lucy-Richardson, Van-Cittert …) as used in established machines. To verify its effectiveness a plane surface was machined on an IBF.
Precision optical components are usually machined by grinding and polishing in several steps with increasing accuracy. Spherical surfaces will be finished in a last step with large tools to smooth the surface. The requested surface accuracy of non-spherical surfaces only can be achieved with tools in point contact to the surface. So called mid-frequency errors (MSFE) can accumulate with zonal processes. This work is on the formation of surface errors from grinding to polishing by conducting an analysis of the surfaces in their machining steps by non-contact interferometric methods. The errors on the surface can be distinguished as described in DIN 4760 whereby 2nd to 3rd order errors are the so-called MSFE. By appropriate filtering of the measured data frequencies of errors can be suppressed in a manner that only defined spatial frequencies will be shown in the surface plot. It can be observed that some frequencies already may be formed in the early machining steps like grinding and main-polishing. Additionally it is known that MSFE can be produced by the process itself and other side effects. Beside a description of surface errors based on the limits of measurement technologies, different formation mechanisms for selected spatial frequencies are presented. A correction may be only possible by tools that have a lateral size below the wavelength of the error structure. The presented considerations may be used to develop proposals to handle surface errors.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.