The most attractive features of resonant scanners are high reliability and eternal life as well as extremely low wobble and jitter. Power consumption is also low, electronic drive is simple, and the device is capable of handling large beams. All of these features are delivered at a low cost in a small package. The resonant scanner's use in numerous high precision applications, however, has been limited because of the difficulty in controlling its phase and resonant frequency. This paper introduces the concept of tunable/controllable resonant scanners, discusses their features, and offers a number of tuning techniques. It describes two angular scanner designs and presents data on tunable range and life tests. It also reviews applications for these new tunable resonant scanners that preserve the desirable features of earlier models while removing the old problems with synchronization or time base flexibility. The three major types of raster scanning applications where the tunable resonant scanner may be of benefit are: 1. In systems with multiple time bases such as multiple scanner networks or with scanners keyed to a common clock (the line frequency or data source) or a machine with multiple resonant scanners. A typical application is image and text transmission, also a printer with a large data base where a buffer is uneconomical. 2. In systems sharing data processing or laser equipment for reasons of cost or capacity, typically multiple work station manufacturing processes or graphic processes. 3. In systems with extremely precise time bases where the frequency stability of conventional scanners cannot be relied upon.

A high precision controller using two synchronization photodiodes or phototransistors is used to compensate the errors introduced by the position sensor of the galvanometer and its electronic driving circuit. Errors of gain, offset, and phase of a periodic waveform are easily compensated using this technique. This paper will describe the approach used. A comparison with the conventional method of reading the signal directly from the position sensor will also be made. Experimental results will be presented.

A fast access control method of optical disk memory was developed. When this method is used, an optical spot can directly reach any destination track by driving a linear motor which carries the optical head in radial direction of a pregrooved disk, even though its track pitch is as narrow as only 1.6 μm. To realize this method, in the first, distorted track crossing signal by recorded pits, which was detected by a split-type photo detector in an optical head, was compensated by an appropriate signal processing method. Therefore, the total number of crossing tracks can be counted correctly and the traveling velocity of the head can be controlled accurately by utilizing the compensated crossing sensor signal. In the next, a state observer based on the advanced control theory was applied to make the velocity controlling performance improve high enough. As a result, the track crossing velocity of the optical spot has been able to be controlled accurately while seeking. Since it hardly needs to make reseeks and any track jump motion, the average access time of optical disk memory has been shortened no more than 70 msec including average rotational latency.

Minimizing obscuration in Cassegrain optical systems places limits on the secondary mirror diameter and width of the supporting spider vanes. Accompanying those limits is the requirement that the focus, tilt, and decentration drives fit within the obscuration limits. This paper presents a design solution to the problem of minimizing obscuration while providing adjustment capability for focus, tilt, and decentration using stepper motors with integral harmonic drives coupled to eccentrics and flexures to ultimately achieve resolutions of half-micron in focus, 5 micron translation and 70 microradians in the tilt mode. The resulting design eliminates friction, hysteresis and injurious thermal effects by means of flexural connecting members of super invar.

A chopping secondary mirror system is at present in use at ESO's 3.6m telescope for the cancellation of the sky- and telescope background emission and its variations. The chopper system achieves: - square wave and sawtooth chopping with amplitudes up to 10 arcmin and frequencies from 0-40 Hz, - stability of <1 arcsec ( <.2 on sky) in flat part of waveform, - risetime of about 6 msec, - flexure of <7.5 arcmin for any 30 degree tilt, - no vibrations induced in or transmitted to the rest of the mechanics (telescope). The chopper system is attached to a subsystem which enables the mirror to be rotated and focussed. The mechanics, optics and electronic control of the system are described and its performance is presented.

The surface accuracy of the primary mirror of a large telescope can be compromised by polishing errors, thermal distortions, material inhomogeneities, stress relaxation and errors in support forces. One method of correcting the mirror figure is to apply forces to the mirror to introduce deformations that cancel out these errors. The image quality goals of the National New Technology Telescope (NNTT) will require highly accurate surfaces on 8-meter diameter primary mirrors, and it is anticipated that an active force system will be necessary to maintain the required accuracy. An active support system has been built and tested on a 1.8-meter prototype of the primary mirrors for the NNTT. The existing astatic counterweight support system was modified to allow computer-controlled variation of the forces exerted . Tests of the system have been conducted in two different modes; it has been used to compensate measured optical surface distortions, as well as to correct the thermal distortion predicted by computer modeling based on measured mirror temperatures. Test results show that distortions corresponding to astigmatism, coma, spherical aberration, trefoil and quatrefoil can be effectively countered.

A method has been developed to least-mean square fit Zernike polynomials to surface deformations along the component centerline or lurface normal during finite element structural analysis. This procedure uses a preprocessor to MSC/NASTRAN and allows the polynomial fit to be calculated during dynamic response as well as static loading.

A tracking telescope designed, built and tested by Perkin-Elmer contains several unique mechanical design solutions that may be applied to other types of equipment with similar design requirements. Three design areas that required special approaches include an azimuth overtravel stop mechanism permitting +190 degrees of motion, a drive system with very low gear-induced errors and zero backlash and a cable wrap system.

The availability of microcomputer control has allowed a new design philosophy to be introduced which is of particular interest for remote or unattended instrumentation. Instead of attempting to create systems that will retain their attributes (e.g., mechanical dimensions) in the presence of all perturbations, it is now possible to use an arrangement in which errors are expected, but reduced to tolerable values by the exercise of machine intelligence. This involves sensors to monitor performance and effectors that actively reduce the error. With such a concept it is frequently possible to achieve better results with lighter and more compact hardware and to eliminate the need for human intervention to retain calibration. The presence of a microcomputer also makes it possible for a machine to perform functions that were previously left to human judgement. One of these, the resolution of seeming ambiguities, is the subject of this paper.

A detailed mechanical design of a large diameter (120-inch radius) centrifuge was completed. The overall g-stability of the system under all operating conditions is projected to be 5 ppm. The design includes an externally pressurized dual gas bearing main axis spindle and a two-degree of freedom platform. Significant design features and methods are discussed.

The present paper proposes a new technique for collecting and processing carried fringe image. Special softwares are designed for photoelastic, holographic, speckle carried fringe image. The procedure is demonstrated and some main subroutines are described in detail. The whole-field grades are determined automatically.

The blasting strain fields between the adjacent boreholes initiated simultaneously were investigated with the dynamic moire-photoelasticity and the strain gages. A multiple spark dynamic photoelastic apparatus (Mode WZDD-1) and the super-dynamic. measuring system were utilized in the dynamic recording. According to the dynamic moire-photoelastic fringe patterns, the superposition of blasting stress waves was analysed, and the distribution of az(in the direction of the line linking two adjacent borehole centers, which is simply called linking-center line in the following) and ar (in the direction which is normal to the linking-center line) in the linking-center line and ea. in the line which is vertical to the linking-center line through the middel point of the linking-center line were calculated quantitatively. The oscillations of strain vs. time in the linking-center line due to the superposition of blasting stress waves were recorded with the foil strain gages. The laws of superposition and attenuation of stress waves between boreholes derived from the measured and calculated results are in agreement with the results from the theoretical analyses.

This paper discusses and describes applications of a recently developed infrared scanning system (SPATE) sensitive enough to be used for the full-field experimental stress analysis of dynamically loaded structures. A review is presented of the progress of the thermo-elastic stress analysis research programme at the National Engineering Laboratory over the past four years. Stress analyses performed on several components and engineering structures are described, and developments in such areas as calibration, design evaluation, and random loading are discussed.

Structural dynamics for integrity assessment and condition monitoring of electrical power station plant can be approached using a variety of methods combining theoretical modelling with experimental measurements. In recent years experimental approaches have broadened to include non-contacting, full-field response measurement techniques using laser holography and more recently, stress pattern analysis from thermal emission (SPATE) to obtain the dynamic stresses. This paper presents the use of thermoelastic techniques for modal evaluation. The modal shapes of a welded 'T' section laboratory specimen obtained from the SPATE 8000 camera system are compared with results of finite element modelling, frequency response function and pulsed holographic techniques. In addition a compensating bellows unit taken from the gas circuit of a nuclear power station was dynamically tested in the laboratory using an electromagnetic shaker to excite resonances in the frequency range 300Hz to 400Hz. The dynamic stress data as recorded by the SPATE 8000 camera was compared with finite element model predictions. A further example describes a similar modal response investigation performed during a routine overhaul on an Advanced Gas Cooled Reactor gas circulator inner casing ring which is designed to isolate the stress concentrations present.

The development of thermoelastic stress analysis has provided a unique non-contact method of investigating the stress distribution over the surface of structures and components subject to cyclic loading. The technique has been extensively applied on a qualitative basis to identify areas of high or non-uniform stress in a wide range of structures, including those whose complexity would preclude investigation by more conventional means. However, particularly when applied to fibre composite materials, the derivation of accurate quantitative data requires further development. The present Paper describes the development of a rigorous theoretical basis for the application of thermoelastic stress analysis to a generalised anisotropic material. Some practical aspects of the application of the method to fibre composite materials are then discussed.

In an earlier paper, a revised theory of the thermoelastic effect was presented which offers an explanation of the mean stress dependence of the thermoelastic parameter. Further experimental results are presented here to validate this theory, and to demonstrate that the predicted higher harmonic thermal response of a body under a single frequency excitation may be observed and measured. The presence of this higher harmonic response is due to the weak quadratic nature of the stress temperature relationship. It is suggested that the ability to accurately measure this component may be the key to making practical residual stress measurements using this theory.

The field of "Opto-Mechanical Instrument Design" is becoming an identifiable discipline. From the time the first optical instruments were designed, fabricated and assembled, optical and mechanical engineering design have been coupled together. Now with the high accuracy and stability required for many optical systems, the mechanical engineering aspects of the main support structure have become of paramount importance (at times equaling the importance of the optical system design itself). Today the world of accurate measurements and alignment is based on the quality and performance of the opto-mechanical instruments available. This paper will highlight the topics that must be considered during the design, development, fabrication, assembly and test of an opto-mechanical instrument: materials of construction, fabrication, jointing, stress analysis, thermal analysis, vibration analysis, transmission (power), rotary motion, linear motion, adjustments and alignment. For additional information on optomechanical design, the following references will be useful: "Introduction to Opto-Mechanical Design" by D. Vukobratovich; "Opto-Mechanical System Design" by P. Yoder, Jr.; "Fine Adjustments for Optical Alignment" by R.S. Reiss. Telephone: (415) 965-3888.

A large solar furnace that has a parabolic mirror with a diameter of 10m, a focal length of 3.2m and a heliostat mirror with an area of 15x15m was made by the authors at T6hoku University in 1962, and subsequently a small solar furnace having a parabolic mirror with a diameter of 1.5m, a focal -length of 0.5m and a heliostat mirror with an area of 2x2m was constructed at T6hoku Gakuin University in 1986. The large solar furnace could melt tungsten with a melting point of 3400°C, and the small solar furnace drove a stirling engine made in West Germany that had a rated power of 400W. The parabolic mirror of the segment type at TohokU University was made by a grinding apparatus that used a cam mechanism, and the parabolic mirror at T6hoku Gakuin University was made by an apparatus (hand-made by students)which employed a link mechanism to draw the parabolic curve. In this paper, the grinding apparatus used for making the segmental parabolic mirror with a diameter of 2-3m and a focal length of 0.5-1.0 m is reported. This mirror was used in a middle-size solar heat engine. The heat engine in this system was a Stirling engine with a rated power of 1-3KW, and the grinding apparatus (the precision parts moved in a linear track ) employed a compact link mechanism.

Discontinuous silicon carbide reinforced aluminum metal matrix composites (MMC) are a unique class of advanced engineered materials which have been developed and recently qualified for use in aerospace structures, inertial guidance systems, and lightweight optical assemblies. Such materials are as light as aluminum but exhibit significantly greater strengths and specific stiffness. Certain grades of these MMC's are isotropic and are more resistant to compressive microcreep than beryllium; and they can be tailored to match the coefficients of thermal expansion (CTE) of other materials, including beryllium, stainless steel, and electroless nickel. Since these composites can be easily forged, superplastically formed, and precision machined into complex shapes, they are ideal for use in the economical production of stable optical systems. This paper describes some of the enhanced properties of engineered MMC's, discusses some of the design considerations that have led to the specification of these materials for building an ultra-lightweight telescope, and presents some interesting results obtained from prototype testing.

Thermal cycling hysteresis affects surface figure in low-expansivity mirror substrates. Zerodur, ULE, and Cer-Vit 8-in.-diameter mirrors and dilatometer samples were thermally cycled at uniform rates of 6 K/hr and 60 K/hr, and somewhat faster for nonuniform heating. Figure distortions as large as λ/10 were observed following nonuniform heating of standard Zerodur, which was the only material exhibiting thermal hysteresis. A new experimental Zerodur appears to be free of this problem.

Fibre optic sensors are used to measure sample displacements caused by a mechanical device capable of 1-D motion. Test samples, composed of multilayered materials, are submitted to sinusoidal excitations by means of an electromagnetic driver. The driving platform, as well as several test points in the sample, are instrumented with fibre optic bundles that are both, conveying a controled amount of light, and receiving the reflected light from the driver and sample points, in order to establish their instantaneous position with time. The device can oscillate up to several kHz, and the position of test surfaces evaluated with a precision of 1.5 microns.

Industrial robot is mechanical manipulator with repeatability as its one of the most important characteristics. To measure the repeatability of a manipulator moving in space revolving diameter 2 to 5 meters, it is required to design a new test system with high accuracy. The spacial repeatability is tested by two sets of experimental setups arranged at right angle. Each setup is composed of the following items: 1). A He-Ne laser generater, a Fresnel zone plate, a corner cube prism and a quadrant cell which constitutes a optical two-dimensional sensing system; 2). A precision horizontal slide and a vertical mechanical slide both actuated by stepped motor which guide an optoelectronic device to trace automatically the laser beam reflected from the corner cube prism which is mounted on the hand of the manipulator; 3). Data processing circuits and movement control circuits of stepped motors controlled by a microcomputer; 4). Two inductosyn devices and related numerical display meter used to measure the two-dimensional displacement. This paper deals with the application of automatic tracing technique by means of the corner cube prism and laser on micrometrology as well as the setups of this system. The experimental results by using of this system are presented. It is possible to achieve an accuracy of 0.02 mm or 7.8*10-finch.

The ARACOR Model 4100XPL semiconductor test instrument delivers a train of Nd:YAG Q-Switched laser pulses at 10 Hz whose energies are selectable over 5 orders of magnitude by user data entry. Each pulse has the same energy within 5%. To achieve this, a feedback controlled beam attenuator module was designed with a bandwidth of about 1 Hertz. The beam is thereby also stabilized against thermal drift and performance degradation of the laser. A microprocessor monitors the beam energy via a Brewster window pick-off, photodiode and related electronics. This feedback is then used to position the beam attenuator module. Excellent control is achieved without adding noticeably to the normal pulse to pulse variations of the laser via a control law which avoids excessive servoing about the desired pulse energy.

Simple low cost methods are discussed for obtaining 96-mirror arrays using closely packed square mirrors and commercially available linear actuators. Advantages and disadvantages of orthogonal-kinematic and gimbaled-pivot-flexure designs that have been employed in the Aurora Project are addressed.

The program automatically designs and draws all parts and tools that are needed for manufacturing spherical lenses. The input is the lens production drawing; the output consists of production drawings of all the tools. Two blocking methods (pitch bonding and recessing) are available, each defining a complete set of tools. The program is integrated with the Euclid CAD system, and can be operated either in an interactive mode (on a graphic CAD screen), or in a batch mode.