Manufacturing of a toroidal mold for the production of polymeric lenses

Optical elements are used in various fields, in which it is required to measure spherical, aspherical, and free-form surface shapes. We use a nanoprofiler previously developed by our research group to measure a cylindrical surface shape. Unlike spherical or aspherical shapes, the cylindrical surface shape of free-form surface is rotationally asymmetrical and thus difficult to determine the measured position. Therefore, we processed fiducial marks on the measured sample and specified the measurement position by fiducial marks. Then we performed comparative measurements with a Fizeau interferometer using a computer generated hologram (CGH) as a reference plane. The nanoprofiler and interferometer measured to accuracies of 43.7 nm peak-to-valley (PV) and 60.9 nm PV, respectively; the difference between the two was 17.2 nm PV. Furthermore, a spherical shape and an aspherical shape were similarly measured by both devices. The spherical shape had a difference of 8.4 nm PV and that of the aspherical shape was 9.7 nm PV; thus the difference between the measurements of the cylindrical surface shape was larger. Considering that the guaranteed accuracy of CGH for measuring a cylindrical surface shape is ∼30 nm PV, we concluded that this is a reasonable result and fiducial marks are effective to determine the position.

Conformal optical dome can induce the drag of the missile. The conformal dome test is an important way to make the dome perfect for imaging. This paper shows a interference test of elliptical conformal dome's inside surface. First, the test system is designed by zero position method. The test system is composed by two parts: the focal lens and the spherical mirror. The interferometer used is operating in 632.8nm. The ideal test system's resolution is 0.0065waves in PV(peak to valley) value and 0.004 waves in RMS(root mean square) value. The next step is tolerance analysis. According to the optical and mechanical manufacture level, the tolerance analysis is done by optical software ZEMAX. The results are 1.219 waves in PV value and 0.249 in RMS value. The working wavelength is 4.2μm for conformal dome. After calculation, the PV value is 0.179 waves and the RMS value is 0.0366 waves. Finally, the bolster is designed, and for the key structure the topology optimizing is simulated

Straight bevel gears are usually manufactured with various machines and systems, and the tooth profiles are produced by grinding or machining by means of a tool with many cutting edges. In recent years, the straight bevel gear has been manufactured by an end mill using a CNC milling machine because the use of the auxiliary apparatus, special cutters, and special machine tools are not needed. Using this method, the gear manufacturing with high accuracy is important problem. In this paper, the coordinate measurement of the straight bevel gear manufactured in this method is performed and the gear accuracy is evaluated. The tooth profiles of a straight bevel gear generated by a quasicomplementary crown gear instead of a conventional complementary crown gear is introduced. For this study, first the tooth profiles of the straight bevel gear were modeled using a 3D-CAD system and the gear was manufactured by an end mill using a CNC milling machine based on a CAM process. Afterward, the coordinates of many points on the gear tooth surfaces were measured at random using a coordinate measuring machine. This coordinate measurement provides the information about the factors relation to the gear accuracy such as pressure angle, tooth angle error, root cone angle, and apex to back and so on. Therefore, the values of the above factors were estimated and were compared with the theoretical ones, respectively.

ABCO Tool & Die, Inc. is a mold manufacturer specializing in the manufacturing of plastic injection molds for molded optical parts. The purpose of this presentation is to explain the concepts and procedures required to build a mold that produces precision optical parts. Optical molds can produce a variety of molded parts ranging from safety eyewear to sophisticated military lens parts, which must meet precise optical specifications. The manufacturing of these molds begins with the design engineering of precision optical components. The mold design and the related optical inserts are determined based upon the specific optical criteria and optical surface geometry. The mold manufacturing techniques will be based upon the optical surface geometry requirements and specific details. Manufacturing processes used will be specific to prescribed geometrical surface requirements of the molded part. The combined efforts result in a robust optical mold which can produce molded parts that meet the most precise optical specifications.

In order to realize testing of a large aperture convex sphere, large aperture standard transmission spheres are designed and manufactured. The system focal length of 4400 mm, 350 mm aperture are required. The design and performance comparison, surface quality and radius of curvature testing, large aperture SiC convex sphere surface testing are complished. The simulation result analysis of optical design software shows that the wavefront performance of the system design is 0.0001 [peak to valley (PV), =632.8 nm], and it can offer a high-quality spherical wave and the reference spherical surface quality is up to 0.088 (PV, =632.8 nm), 0.006 [root mean square (RMS), =632.8 nm]. For a SiC convex sphere of 320 mm aperture and 4092 mm radius of curvature the manufactured results of the surface quality are up to 0.102 (PV, =632.8 nm), 0.011 (RMS, =632.8 nm). The results indicates that the large aperture standard transmission spheres are helpful for the testing large aperture convex sphere surface with large radius of curvature, the highly precision testing is complished only by standard sphere and non-null testing method based on digital mask, and the problems of large-aperture convex asphere or freeform mirror testing are solved.

Typically, optical molds have been made from silicon carbide (SiC) or tungsten carbide (WC). Magnetorheological Finishing (MRF) polishing results of SiC and WC molds will be reviewed. Impressive figure corrections have been demonstrated on both types of materials. The roughness performance of CVD-SiC, WC and binderless WC will be compared. However, the hardness and polycrystalline nature of these materials make them difficult to manufacture. In this paper we report positive initial results using an alternate mold material, glassy carbon. Test samples have been ground, pre-polished and finish polished to a 38 nm surface figure peak-to-valley (PV) and a 6 a rms surface roughness, with improved cycle times versus SiC and WC. Glassy carbon is a promising mold material candidate as an amorphous material of lower hardness. The lower hardness leads to more effective diamond grinding process and results in a better surface rms roughness following MRF. After reviewing key material properties of glassy carbon material, this paper will describe some collaborative activities between Toshiba Machine Co., Ltd. and QED Technologies (QED) to manufacture representative examples of glassy carbon. Details of the grinding, pre-polishing and final polishing process will be provided along with the resultant metrology results after key steps. Molding experiments based on these developments will also be presented.

This article describes the difference between coordinate measurement results obtained on a coordinate measuring machine and on a CNC milling machine. It also includes the raw materials used for measurement, the geometry of the workpiece used, the tools used, the principles used to evaluate the measurement and the measurement results. In terms of measurement results, machine measurement has been shown to be more accurate. For flatness, the deviation between the two measurements averaged 4 µm, for circularity 4 µm, for single-interpolation machining with one axis 3 µm, their parallelism 5 µm, for machining two ten-slots together 2 µm, their parallelism 5 µm. The difference between the errors proves that milling and measuring machine measurement do not differ to such an extent as to require investment in a measuring machine, if circumstances do not require it.

본 논문에서는 37채널을 갖는 적응광학계용 SiC(Silicon Carbide) 변형거울의 파면 보상 성능 검증에 관한 내용을 다룬다. 컴퓨터 시뮬레이션을 통해 SiC 변형거울의 파면 보상 성능을 예측하였고, 실제 closed-loop 적응광학계를 구성하여 파면 보상 성능을 확인 하였다. Closed-loop 적응광학계는 광원, 위상판, SiC 변형거울, 고속 샥-하트만 센서 그리고 제어용 컴퓨터로 구성되어있다. 회전하는 위상판에 의해 왜곡된 파면을 샥-하트만 센서로 측정하고, SiC 변형거울을 이용하여 왜곡된 파면을 보상해주는 시스템이다. 결과적으로 closed-loop 적응광학계에서 500 Hz의 속도로 PV(Peak-to-Valley) <TEX>$0.3{\mu}m{\sim}0.9{\mu}m$</TEX>, RMS(Root-Mean-Square) <TEX>$0.06{\mu}m{\sim}0.25{\mu}m$</TEX>의 왜곡된 파면을 PV <TEX>$0.1{\mu}m$</TEX>, RMS <TEX>$0.03{\mu}m$</TEX>이하로 보상시킬 수 있었다. In this paper, we deal with the wavefront compensation capability of a silicon carbide (SiC) deformable mirror (DM) with 37 actuators for adaptive optics. The wavefront compensation capability of the SiC DM is predicted by computer simulation and examined by actual experiments with a closed-loop adaptive optics system consistsing of a light source, a phase plate, a SiC DM, a high speed Shack-Hartmann sensor, and a control computer. Distortion of wavefront is caused by the phase plate in the closed-loop adaptive optics system. The distorted wavefront has a peak-to-valley (PV) wavefront error of <TEX>$0.3{\mu}m{\sim}0.9{\mu}m$</TEX> and root-mean-square (RMS) error of <TEX>$0.06{\mu}m{\sim}0.25{\mu}m$</TEX>. The high-speed Shack-Hartmann sensor measures the wavefront error of the distortion caused by the phase plate, and the SiC DM compensates for the distorted wavefront. The compensated wavefront has residual errors lower than <TEX>$0.1{\mu}m$</TEX> PV and <TEX>$0.03{\mu}m$</TEX> RMS. Consequently, we conclude that we can compensate for the distorted wavefront using the SiC DM in the closed-loop adaptive optics system with an operating frequency speed of 500 Hz.

Non-rotational symmetric aspheric surface has many significant advantages, but it still can not be widely used because the limiting that there is no method can tests it precisely. At present, the coordinate contour measuring machine is the main testing method for the aspheric surface with non-rotational symmetric, but the measurement accuracy of this method is not high. In this paper, the method of diffraction compensator (computed graphic holograph) has been adopted to test the combined aspheric surface, which can compensate the phase caused by tested lens. The sample surface is the combined aspheric surface with diameter of 33.84mm, and the process from optical software simulation design, the fabrication of the computed graphic holograph (CGH) to experimental platform built is given in detail after testing via the CGH technology. The simulation results show that the root mean square (RMS) of remnant wave-front error is 0.004 λ, and the peak to valley (PV) is 0.0245 λ. The free-from surface has been tested by Zygo interferometer, and the experimental results show that the RMS is 0.49 λ, the PV is 4.69 λ. The accuracy of the result is higher than that of coordinate contour measuring machine. The system error caused by optical elements analysed is 0.1149λ. The accurate result means that the CGH technology for testing the combined aspheric surface is realized.

Due to the great demand of modern optical systems, many researches on magnetorheological finishing (MRF) technology have been carried out. In this paper, the common MRF was discussed. To reduce the figure error of optical surfaces, dual-rotation magnetorheological finishing (DRMRF) technology was studied. A DRMRF setup was designed and built. Customised magnetorheological polishing fluid was developed and rheological properties were tested. Experiments of removal functions and micro-topography of common MRF and DRMRF were conducted. An experiment of figure error correction was conducted on a Fused Silica workpiece by DRMRF. The figure error was corrected from initial 110.549 nm PV (peak to valley) and 18.201 nm RMS (root mean square) of the whole surface (φ100 mm) to 21.474 nm PV and 2.513 nm RMS within optical aperture (φ96 mm). Results indicate that dual-rotation magnetorheological finishing is an effective method of the figure error correction for ultra-precision optical surfaces.

Thermally assisted finishing of fused deposition modelling build part using a novel CNC tool

Although most of the current surfboards are manufactured industrially, the use of rigorous surface metrology is scarce to control the surface finish of CNC-milled expanded polystyrene (EPS) blanks. The present study aimed to evaluate the surface finish of EPS blanks machined by a CNC milling machine with a diamond disc tool, designed for surfboard manufacturing. Four EPS blanks of different brands and densities were characterized, machined at two feed rates: (a) 10 m min−1 and (b) 5 m min−1, and the quality of the surface finish was evaluated. Scanning electron microscopy as well as a confocal laser scanning microscope were used to evaluate the surface morphology. A range of topographical parameters was proposed and assessed. The structural characteristics of the EPS blanks demonstrated that the machined material is easily mouldable, not causing damage to the tool, thus requiring little maintenance for tool periodic replacement. The use of confocal microscopy proved to be efficient for the evaluation of the quality of the surface finish of EPS blanks and a set of topographical parameters was chosen. The higher density blanks showed better surface finish after machining at the feed rate of 5 m min−1, while the lower density blank had a better surface finish when machined at the feed rate of 10 m min−1. The parameters Sq, Sz, Sdq, Spk and Svk could not differentiate well the quality of the machined EPS surfboards under different conditions due to the large deviations of the measured values, probably resulting from the occurrence of through-bead cutting and bead removal. The parameters Ssk and Sku could detect problems in the surface morphology resulting from machining EPS blanks, enabling to differentiate between the through-bead cutting and bead removal mechanisms. When bead removal occurred significantly, Ssk values became more negative and Sku values were lower than 3.

Specifications of phase error, peak to valley (PV), and root mean square (rms) are not able to represent the properties of a wavefront reasonably because of their irresponsibility for spatial frequencies. Power spectral density is a parameter that is especially effective to indicate the frequency regime. However, it seems not convenient for opticians to implement. Parameters of phase gradient, PV gradient, and rms gradient are most correlated with a point-spread function of an imaging system, and they can provide clear instruction of manufacture. The algorithms of gradient parameters have been modified in order to represent the image quality better. In order to demonstrate the analyses, an experimental spherical mirror has been worked out. It is clear that imaging performances can be maintained while manufacture difficulties are decreased when a reasonable trade-off between specifications of phase error and phase gradient is made.

A Lap-MRF process is proposed for large aperture mirrors. In Lap-MRF, a lap is used to expand the polishing area, which improves the material removal rate. Moreover, the MR fluid can be renewed continuously to ensure the stability of the material removal rate, which improves the convergence efficiency of the surface profile error. In this paper, the figuring ability of the Lap-MRF removal function is analyzed. The Lap-MRF process, which is based on surface profile filtering, is presented. Finally, a series of figuring experiments on a Φ350 mm K9 mirror are carried out. For the Lap-MRF removal function, the volume removal rate is up to 0.48 mm3 /min and the cut-off frequency is about 0.03 mm-1 . After four times figuring using Lap-MRF, the surface profile error throughout the whole surface is improved to 4.84 λ (λ= 632.8 nm) PV (Peak-to-Valley), 0.69 λ RMS (Root Mean Square) from 25.24 λ PV, 4.31 λ RMS and its total convergence ratio of the RMS error is up to 6.32. These results verify the validity of the proposed method for large aperture mirrors.

The present article describes a method for enhanced accuracy in machining free-form surfaces produced on CNC milling machines. In this method, surface patch machining programs are generated based on their nominal CAD model. After the pretreatment, coordinate control measurements are carried out. The obtained results of the measurements contain information on the values and distribution of observed machining deviations. These data, after appropriate processing, are used to build a corrected CAD model of the surface produced. This model, made using reverse engineering techniques, compensates for the observed machining deviations. After regeneration of machining programs, the object processing and control measurements are repeated. As a result of the conducted procedure, the accuracy of the manufacture of the surface object is increased. This article also proposes the introduction of a simple procedure for the filtration of measurement data. Its purpose is to minimise the effect of random phenomena on the final machining error correction. The final part of the article presents the effects of the proposed method of increasing the accuracy of manufacturing on ‘raw’ and filtered measurement data. In both cases, a significant improvement in the accuracy of the machining process was achieved, with better final results obtained from the filtered measurement data. The method proposed in the article has been verified for three-axis machining with a ball-end cutter.