This lecture is to call to mind this great pioneer of high-speed photography and cinematography, who would have been 90 years old a few weeks ago. In addition to his career, his various scientific contributions, his interests, and his far-sightedness in scientific cooperation, personal experiences also are mentioned.

Recent advancements in the Ultranac camera are described. The camera incorporates all solid state electronics which are used to drive the all metal/ceramic image converter tube. Individual exposure and inter-frame times are fully programmable via an external computer. A review is made of the applications in which Ultranac is being employed -- these include combustion studies, particle flow, ballistics and detonation phenomena.

This paper reports the results of a study that empirically determined the gating time of an MCPI tube must be about eight times more than the iris time to prevent spatial resolution degradation. During the study, MTF curves were taken under dc conditions, as well as dynamic gating conditions from 1.6 microsecond(s) to 1 ns gate widths on a tube that has an iris time of 1.2 ns. Under dc conditions, the 50% MTF was 10 lp/mm; under dynamic conditions, the 50% MTF was 7 lp/mm from 1.6 microsecond(s) to 20 ns FWHM. However, when the gate width was reduced to 10 ns, the resolution began to decrease and became 5 lp/mm at 1.0 ns. The results of the experiment and the MTF measurement apparatus are discussed.

A totally new approach has been taken to develop an image intensifier for use in front of high speed rotating prism and image converter cameras at framing rates up to 1 million pictures per second. Cinemax II is a high brilliance image intensifier which can also record UV radiation down to 180 nm, increasing the scope of the automotive engineer investigating the combustion process. Cinemax II is gated on at a frequency of 100 Hz to facilitate setting up and wide aperture critical focussing. Variable gain allows the image brightness to be adjusted to satisfy different subject lightning conditions. In operate mode the gate on period can be readily synchronized to the event and camera to ensure optimum results are achieved and safety thresholds are not exceeded. Cinemax II has successfully been used with the Imacon 792, Photec, and Hyspeed cine cameras for combustion and plasma research.

Common high-speed gated proximity focused multichannel plate image intensifiers allow for a typical gate width of 3 to 5 ns. We have studied an alternative way to accomplish sub- nanosecond time-resolved imaging by operating a gatable proximity focused intensifier as a radio-frequency phase-sensitive camera. In this operating mode, we apply a dc bias voltage between the photocathode and the microchannel plate input that brings the effective intensity transfer function to 50% relative to the `ON' state. Then we add a rf signal to the dc voltage which is phase-locked to the intensity modulation phase of the imaging light source. This phase-locking causes a steady-state image at the intensifier output screen which is recorded using a CCD camera, and stored in an image processor. At least two images are recorded at two different intensifier modulation phase settings relative to the light source modulation phase. Finally, the two images are subtracted, or divided, pixel-by-pixel to create the time- resolved image. The rf phase-sensitive camera has been applied to distance-selective image suppression. We have created black-and-white contrasts for target distance differences of 3.75 cm, i.e., for photon transit time differences of 0.25 ns. The camera also can be used to accomplish lifetime-selective fluorescence suppression.

The current demand for inexpensive streak camera manufacturing leads to the necessity in development of a variety of relatively simple and low cost image-converter tubes. One such tube, known as PIF-C, designed and manufactured in the Photoelectronics Department of the General Physics Institute (GPI), is now commercially available. Its experimentally measured time resolution in streak mode has approached one picosecond, and 3 ps in synchroscan mode at 82 MHz operation frequency. In single frame mode at 100 ns exposure time, the spatial resolution over 6 mm input area is within 15 lp/mm. Electron optical magnification of the tube is 1.5 ^x. PIF-C tubes may be supplied with one of the S1/S20/S25 photocathodes, fabricated either on borosilicate glass, UV-glass, or MgF₂ substrate. Its P11 phosphor screen is deposited onto the fiber optic window. EBI of the PIF-C/S1 tube is in the range of 5 (DOT) 10^-10 A/cm².

Obtaining detailed photographs of the early stages of the first hydrogen bomb explosion in 1952 posed a number of problems. First, it was necessary to invent a continuous-access camera which could solve the problem that existing million-picture-per-second cameras were blind most of the time. The solution here was to alter an existing camera design so that two modified cameras could be mounted around a single high-speed rotating mirror. A second problem, acquiring the necessary lenses of precisely specified focal lengths, was solved by obtaining a large number of production lenses from war surplus salvage. A third hurdle to be overcome was to test the new camera at an A-bomb explosion. Finally, it was necessary to solve the almost impossible difficulty of building a safe camera shelter close to a megaton explosion. This paper describes the way these problems were solved. Unfortunately the successful pictures that were taken are still classified.

A new electronic control system provides intelligent (computer) control features for high speed film cameras. This system incorporates the most significant advances made since the advent of analog servo controls in high speed cameras a generation ago. All of the camera's former electronics are replaced with digital electronics operating under 32-bit microprocessor control which directs and monitors each camera function. Typical camera control functions include selection of: camera operating speed, phase-lock synchronization, shutter phasing delay in multiple-camera installation, film load, acceleration time, etc. Entries are made on a keypad, on the camera housing, or may be made by an external data link. Keypad inputs are displayed on a 16 character liquid crystal display (LCD) located adjacent to the keypad panel. A 15- character LED numeric display, at the edge of the film aperture, prints the IRIG-B time code on film, or real-time data supplied via an RS232 input from sensors associated with the test or process being filmed. This data can also be stored in a RAM memory so that it precisely matches the numerically recorded information on each frame of film. The stored data can then be downloaded to an external computer for analysis, thereby speeding the data reduction process and eliminating the likelihood of human error in reading frame-by-frame film data.

Exposure time reduction with rotating prism cameras is usually achieved with narrow angle shutters, or slits. Exposure times of 1 microsecond are possible with this low cost method, but there are disadvantages: (1) Vast amounts of continuous lighting are required to produce a correctly exposed film, even with the fastest materials available. (2) Total frame recording time remains constant for a given setting. `Focal plane' distortion occurs, particularly with extremely fast events, although subject movement is `frozen.'

With SST and with its variations in BiSST, OSST, or especially with BiOSST the characteristics of a shaped charge jet, such as tip velocity, length, particulation times and distances, diameters, transverse velocities, tumbling rates, and 3-D jet deviations, can be fully analyzed. All this can be done quantitatively with high accuracy. Observation of the jet surface under bright front illumination provides new insight into its structure under very high strain rates during extreme elongation.

By combining the attributes of electro-optic shuttering and pulsed laser illumination in a large format camera system, we have developed a multi-frame image converter camera with a laser illuminator that produces sequential photographs of fast phenomena with very high resolution. The combination of the large size image plane (75 mm), short exposure time per frame (minimum 12 - 15 ns), and monochromatic laser illumination provides clear, sharp, front-lit images of surfaces and shapes with no degradation by luminous air shocks or motion blur. The unique modular design of the camera and the laser allow for a variety of configurations and applications. The current camera system produces eight independent pictures or four stereo pairs. A Q-switched ruby laser with multiple pulse capability provides individual illumination for each frame. This system has photographed a variety of fast phenomena including the first stereo sequential photographs of the initial formation and early time history of high velocity shaped-charge jets.

High-speed photography at framing rates of up to 1 million per second has been employed to investigate the mechanisms of initiation of an explosion and its propagation in a small arms cartridge. Specially designed models have been used to simulate the conditions inside the cartridge and the photography has been carried out using reflected light. It is found that the impact of the striker pin on the percussion cap of the cartridge causes several reaction sites lying close to the impact axis and it is argued that these sites are located at intergranular frictional hot spots in the primary explosive composition. It is also shown that the reaction from the cap to the main propellant charge is transmitted by the streaming of the hot reaction products of the former into the grains of the latter.

A PV001 streak image tube supplied with an atmosphere resistant (AlMgCu)O_x photocathode is reported. Quantum efficiency of such a photocathode being irradiated by 5 ps light pulses at 354 nm wavelength is up to 10^-3. It is shown that this photocathode may be exposed for many cycles (> 20) of air admittance without losses in its quantum efficiency. It is believed that the PV001/oxide tube may be used in electron diffraction experiments.

Cameras and laser illuminators have been upgraded and combined into a coaxial streak and frame system. The argon laser illuminating the streak camera over the duration of the event is combined with the short pulsed dye laser for stop motion framing, using a polarization splitter. The combined beams are sent to the target, viewed by the same objective lens, then split by a second polarizing splitter and relayed to the appropriate cameras. A half-wave plate before the second splitter allows adjustable levels of light of each laser to be leaked to the opposite camera for fiducial purposes. A particular advantage of this arrangement is that both the laser beams are tailored and aimed separately. The dye laser is spatially filtered and focused to cover the frame. The argon laser is focused to a line on the target which is then imaged to the streak camera slit. The ability to see the focal line of the argon laser on the frame camera allows precise collocation of the fields of view of the two cameras.

This paper discusses a laboratory method based on generating a buoyant thermal cloud through explosively bursting an aluminum foil by a rapid electric discharge procedure. The required electric energy is stored in a bank of capacitors and is discharged into the foil through a trigger circuit on external command. The aluminum first vaporizes and becomes an aluminum gas plasma at high temperature (approximately 8000 K) which then mixes with the surrounding air and ignites. The cloud containing these hot combustion products rises up in an unstratified anechoic environment. As the cloud rises, it entrains the air from the surroundings due to turbulent mixing and it grows. To characterize this cloud rise, three different types of photographic techniques are used. They are: high-speed photography (6000 fps), low-speed photography (200 fps), and video photography (30 fps). These techniques cover various time scales in foil firing schedule beginning from early time (up to 10 msec) to late time (up to 4 secs). Images obtained by video photography technique have been processed into a digital format. In digitizing the video tape data, an optical video disk player/recorder was used together with pc-based frame grabber hardware. A simple software routine was developed to obtain cloud size/rise data based on an edge detection technique.

The work of two types of optical fiber gauges on the basis of polymeric light guides was developed and investigated for registration of x,t-diagrams of shock waves movement in solids, liquids, and free surfaces of moving bodies. End-type gauges are proposed for self- luminous phenomena registration. For investigating phenomena in the low-pressure range we used loop gauges. The results of investigations of the gauges' work in gas-dynamic experiments are presented.

An experimental technique for investigating the air flow passing a domestic electric fan by using flow visualization and high speed photography is reported in this paper. A rotary-prism type high speed camera and a motion analyzer were applied in the experiments to investigate the flow pattern closely neighboring the fan blades. The smoke emerged from dry-ice was used as the trace particles. The particle paths and some significant flow phenomena in the flow field were then clearly observed. On the basis of the results obtained by use of this technique, significant improvement over the existing ceiling fan blades has been made for some dozen manufacturers in both mainland China and Hong Kong, resulting in great benefits.

For the precise observation of high-speed impact phenomena, a compact high-speed streak camera recording system has been developed. The system consists of a high-pressure gas gun, a streak camera, and a long-pulse dye laser. The gas gun installed in our laboratory has a muzzle of 40 mm in diameter, and a launch tube of 2 m long. Projectile velocity is measured by the laser beam cut method. The gun is capable of accelerating a 27 g projectile up to 500 m/s, if helium gas is used as a driver. The system has been designed on the principal idea that the precise optical measurement methods developed in other areas of research can be applied to the gun study. The streak camera is 300 mm in diameter, with a rectangular rotating mirror which is driven by an air turbine spindle. The attainable streak velocity is 3 mm/microsecond(s) . The size of the camera is rather small aiming at the portability and economy. Therefore, the streak velocity is relatively slower than the fast cameras, but it is possible to use low-sensitivity but high-resolution film as a recording medium. We have also constructed a pulsed dye laser of 25 - 30 microsecond(s) in duration. The laser can be used as a light source of observation. The advantage for the use of the laser will be multi-fold, i.e., good directivity, almost single frequency, and so on. The feasibility of the system has been demonstrated by performing several experiments.

For investigations of fast nonstationary events, e.g., flows, injection jets, etc., the high speed cinematography is particularly suitable, but there are difficulties with 3-dimensional motions. First the paper deals with problems and objectives which arose during the development of a new drum camera that is also suitable for high speed holography. The result is a drum camera for up to 200,000 f/s, also for use with holograms. The camera set-up and first test results are described, and possibilities of further developments are shown.

A mirror-rotating type compact streak camera whose maximum streak rate was faster than 10 mm/microsecond(s) and a strong Xenon-flash lamp with a wide flash area were produced for impact- shock study of solids. The streak camera mainly consists of the high-frequency motor with an air bearing of 120,000 rpm in a maximum rotating rate, the square-shaped mirror, the lens system with a 15 micrometers -width slit, and the circular-shaped film mount case of 830 mm in inner diameter. The flash lamp consists of two parallel Xenon tubes of 40 mm in length and 10 mm in diameter, which were stimulated with two high voltage condensers of 100 (mu) F and 2 kV. The shock-wave measurement experiments of solids in several 10 s of GPa region were performed by the inclined-mirror method using the streak camera and flash lamp in combination with the keyed-powder gun.

Increasing frame rates and the sensitivity to heat of test objects has forced the development of improved lighting systems. In response to the need for better high speed lighting systems, several advances have been made over the last few years. The primary one is the conversion to the use of metal halide arc lamps instead of tungsten halogen incandescent types. This was followed by the development of electronic power supplies to drive these lamps. The EPS in conjunction with the lamps can provide flickerless light with the capability of accurate power control. This paper presents the key points in this development and describes some of the features of practical `boost lighting systems' presently in service at several test sites. It also describes some of the advantages that will be available in the future with the addition of computer control functions to these systems.

A high pressure arc lamp capable of producing over ten million lumens is presented. The arc's unique cooling allows this type of arc to operate at continuous power ten times greater than any other. Spectral data is presented that shows a white light continuum with strong near infrared spectral lines. Continuous operation and millisecond pulses are outlined. Examples include a 3 by 5 meter area illuminated to 400 klux and a 6 m diameter parabola that produces a 2 degree beam angle and a peak of 4 billion candle power.

Developments in the design of image converter camera systems are reviewed. Designs of image converter tubes are discussed. Final image capture by photographic film and CCD camera and analysis systems are considered.

A high-speed imaging device has been built that is capable of recording several hundred images over a time span of 25 to 400 ns. The imager is based on a streak camera, which provides both spatial and temporal resolution. The system's current angular resolution is 16 X 16 pixels, with a time resolution of 250 ps. It was initially employed to provide 3-D images of objects, in conjunction with a short-pulse (approximately 100 ps) laser. For the 3-D (angle-angle-range) laser radar, the 250 ps time resolution corresponds to a range resolution of 4 cm. In the 3-D system, light from a short-pulse laser (a frequency-doubled, Q-switched, mode-locked Nd:YAG laser operating at a wavelength of 532 nm) flood-illuminates a target of linear dimension approximately 1 m. The returning light from the target is imaged, and the image is dissected by a 16 X 16 array of optical fibers. At the other end of the fiber optic image converter, the 256 fibers form a vertical line array, which is input to the slit of a streak camera. The streak camera sweeps the input line across the output phosphor screen so that horizontal position is directly proportional to time. The resulting 2-D image (fiber location vs. time) at the phosphor is read by an intensified (SIT) vidicon TV tube, and the image is digitized and stored. A computer subsequently decodes the image, unscrambling the linear pixels into an angle-angle image at each time or range bin. We are left with a series of snapshots, each one depicting the portion of target surface in a given range bin. The pictures can be combined to form a 3-D realization of the target. Continuous recording of many images over a short time span is of use in imaging other transient phenomena. These applications share a need for multiple images from a nonrepeatable transient event of time duration on the order of nanoseconds. Applications discussed for the imager include (1) pulsed laser beam diagnostics -- measuring laser beam spatial and temporal structure, (2) reflectivity monitoring during pulsed laser annealing of microelectronics, and (3) detonics or shock wave research, especially microscopic studies of shocks produced by laser pulses.

An experimental prototype of a femtosecond streak image converter camera was built around a specially designed femtosecond streak image tube having a cylindrical type electron focusing lens. Experimentally measured temporal resolution of the camera is better than 500 fs while its spatial resolution is not worse than 40 lp/mm across the slit direction.

A new model of a Scancross-type streak tube is proposed. It can be useful in studies of short light flashes aimed at elucidation of not so much spatial as temporal light flux energy distribution. The Scancross tubes differ from their predecessors in that photoelectrons escaping a 40 mm photocathode surface are gathered in an aberration-free crossover due to an electrostatic field formed by a special `quasispherical' electron-optical system. At the crossover output, the electron beam aperture is compressed, whereupon the beam is accelerated and focused onto the phosphor screen plane. The focusing system field also performs an anisotropic transformation of a point crossover, i.e., compression of the latter's image on the screen in the scanning direction with simultaneous crosswise extension. A rectangular crossover image in the form of a 0.035 mm X 3 mm dash was obtained with the screen size of about 40 mm in the scanning direction. Sensitivity of the high-frequency deflection system was 25 V/mm. A strong compression of the electron beam emanated from the photocathode enabled a 4 orders of magnitude brightness magnification. Employment of such a streak tube allows a much more simple oscilloscope optical system for collection of radiation to be used, since there is no longer need for a small-dimension focusing spot on the photocathode as when using conventional streak tubes. This also simplifies detection of electromagnetic radiation in a very wide spectral range. Calculations suggest the possibility of extending the detection dynamic range by at least two orders using special signal procession accounting for light distribution both in the scanning direction and crosswise.

As a part of the joint research projects between General Physics Institute and Yonsei University/V.TEK Company, an experimental prototype of an image converter camera is designed and manufactured. The camera operates both in single frame and single shot streak modes. Single frame exposures are varied in the 250 - 1000 ns range, while recording intervals in streak mode are adjusted within the 2 - 1000 ns range over a 25 mm-wide output screen area. Temporal resolution at maximum streak speed is better than 10 ps. Total camera gain is 5 (DOT) 10⁴. The camera is equipped with a specially designed PIF-V.1 image converter tube. Available are choices among S1, S20, or S25 photocathodes fabricated onto Molibden glass/UV glass, or MgF₂ substrate.

Photocathodes based on In_0.53Ga_0.47As/InP heterostructures (HS-photocathodes) with Schottky barriers for a spectral range of 0.9 - 1.6 micrometers were investigated. The maximum external quantum yield was 0.5% at (lambda) equals 1.5 micrometers and dark current was I_ed equals 3*10^-8 A/cm². It has been shown that in such photoemitters under reverse bias of about U equals 30 V, the electric field completely penetrates into the working layer of the photocathode. Since the dark current does not depend on the value of the reverse bias, HS-photocathodes may be used for time analyzing tubes to record picosecond pulses with milliwatt peak intensity. To increase the signal/noise ratio we suggest using InP/In_0.53Ga_0.47As superlattice (SL) for designing a SL-photocathode with internal amplification.

Low-jitter reliable short-duration spark sources have been developed for application in color schlieren methods. With an Xe-filling of a previously used nanosecond spark source (NANOSPARK 1400), the light quantity could be increased up to tenfold while the pulse width rose up to 500 ns depending on gas pressure. When filled with air, the new extended version NANOSPARK 4000, which is about three times the size of the NANOSPARK 1400, exhibits pulse widths of 50 - 80 ns (FWHM) and BCP five times as large as the smaller version. With an Xe-filling, the light quantity is again magnified by a factor of ten, but also the pulse width goes up to 900 ns depending on gas pressure. These sources have adequately exposed customary ASA 400 film in various color schlieren experiments. Moreover they have been used in a shearing interferometer and in several monochrome visualization setups.

The fundamental studies for the flash vacuum-ultraviolet (VUV) source utilizing a surface- discharge substrate are described. This flash VUV source consists of the following essential components: a high-voltage power supply, a polarity-inversion-type high-voltage pulser with a condenser capacity of 14.3 nF, an oil diffusion pump, and a flash VUV chamber with a glass body. The VUV chamber employed a surface-discharge ferrite substrate that's pattern was formed by means of the copper vacuum evaporation and was connected to an oil diffusion pump with a pressure of 1.3 X 10^-3 Pa. The combined ceramic condenser in the pulser was charged from 10 to 30 kV by a power supply, and the electric charges in the condenser were discharged to the radiation chamber after closing a gap switch. Then the flash VUV rays were generated. The maximum values of the cathode voltage and the tube current were about -21 kV and 1.7 kA, respectively. The VUV outputs were measured by a combination of a plastic scintillator and a photomultiplier. The pulse durations of the VUV rays were nearly equivalent to the durations of the damped oscillations of the voltage and current, and their values were about 10 microsecond(s) .

A simple technique for obtaining sinusoidally modulated laser radiation, of which the modulation period can be smoothly varied in the range from hundreds of femtosecond to tens of picosecond, is described. The idea of this method consists in the use of linear frequency modulated laser radiation when each axial mode has the same linear chirp.

This paper describes the EMI ruby laser stroboscope and two examples of its application, the study of behind the armor effects of a KE-projectile and the initiation of HE (high explosive) by a hypervelocity debris cloud. The flash x-ray technique is the common method to visualize terminal ballistic and impact phenomena which are mainly combined with light flashes, dust, and fragment clouds. The laser stroboscope is a useful completion to the other techniques, it allows the visualization of long duration (1 ms) events with high-time resolution and high accuracy. At a frequency of 200 kHz, for example, about 260 frames are taken with the camera at exposure times less than 20 ns.

An illumination system is developed for a microscope with a newly developed straight-type Xe flash tube as the light source. An optical fiber is used as the light transmitter to connect the microscope and the light source. The axis of the fiber is coincided to that of the plasma excited in the Xe flash tube. The efficiency of the light beam injected to the fiber is improved up to 5 times of that of the conventional type illumination system.

A novel powerful strobe for high-speed photography is described which can replace the high power cw light source, to save energy and synchroflash with the camera. In this strobe, three- phase transformerless direct rectifier, high current SCR switch and pre-ionization technique are used so that the energy consumption goes down greatly, and its total weight is less than 25 Kg. Its principal parameters are as follows: average power, 50 KW; light emitting pulse width, 1 - 100 ms; pulse rise time, less than 0.05 ms; pulse fall time, less than 0.1 ms.

Usefulness of holographic interferometric flow visualization applied to shock wave research is presented. Reflected shock transition from Mach to regular reflections over wedges is discussed in detail. Examples of shock wave reflection from curved wedges are presented in conjunction with the shock transition phenomena.

The imaging of high velocity (> 2000 m/s), 7 mm Cuboids impacting on various targets is discussed. The reasons why conventional H.S. Cine techniques, even framing at 40,000 pps, are inadequate to record the detail required are outlined. Four different methods of image capture are illustrated giving a direct comparison between state-of-the-art technologies.

A relatively simple method is explained, which makes it possible to determine the detonation velocity of an infinitely large charge, using a very small size sample. However, this requires additional instrumentation and a simple trick in the design of the sample. A prerequisite for this is that the detonation velocity of a reference high explosive must be known accurately and that this velocity is larger than that of the high explosive to be examined. By observing the front face, the true detonation velocity can then be determined, even with a very small sample size. The theoretical background, test set-up, and execution of the experiments are described below. Beside this special application, sandwiched explosive charges are of general interest in connection with shaped charge design, and generally in the effort to increase the efficiency of explosive energy transmission.

The Engineering Department at Oxford University has a history of designing, building, and operating its own devices. Highspeed photography has been no exception. This paper looks at the progression from novel home built rotating mirror cameras through the building of a 9 spark Cranz-Schardin dynamic photoelastic bench, including a typical test and a sequence of photographs using this equipment, to results from an in depth study into the dynamic behavior of a new material, super-plastically formed sandwich panels using the Cordin 377 camera.

Experiments are described where normal impacts of caliber 7.62 mm projectiles on ceramic faced steel plates were observed by high speed photography. Three kinds of ceramic tiles in size were used, and they were bonded to a 10 mm thick backing of strong steel plates. When the target was a steel plate alone, a core of M2 AP projectile has easily passed through the target and the core has been almost intact. The ceramic facing has had two effects on the penetrator effectiveness in this case, one due to blunting of the core tip and the second due to removal of the core mass in erosion. Because of these two effects, the defeated core has been far less effective at penetrating steel plate than that of the intact core. We have obtained minimum safe facing composition with each adhesive.

A new procedure is proposed of generating converging or of colliding shock waves in solids. The method is based on the refraction phenomena of a plane shock front at a shaped material interface, due to the difference in shock velocity of the materials of each side. A high-pressure gas gun is used to produce plane shock waves in a composite target assembly. The assembly is composed of aluminum and of polyethylene. The conically converging shock wave is generated in a lower-impedance material, i.e., polyethylene. In this case, polyethylene material is machined to the shape of a cone, and is inserted and glued to the aluminum plate having just the same inner surface. The processes of shock convergence are observed by a compact high-speed streak camera together with a pulsed dye laser as a light source. We have performed a series of experiments by varying several parameters. The realized converging angle of shock waves is found to be about 55 - 60 degrees in polyethylene medium. It is shown that the converging wave front looks almost continuously curved, and it is not easy to discriminate the boundary of the Mach stem. In other words, the growth of the Mach region seems faster than expected. This result is attributed to the combined effects of wave convergence and Mach reflection.

Multiflash photography is a simple, economic, and efficient technique for the research of desert physics. The study of the micro-characteristics of sand blown by the wind in the desert is one of the basic subjects in desert research. The research on this gas-solid two phase flow was conducted using high speed multiflash photography both in the environment sand wind tunnel and in the natural desert environment for studies of the flight trajectories of moving by jump of sand grains and the structure of wind-blown sand. The experimental results indicate the moving trajectories of sands by wind presents the form of opposite parabola. The taking- off angles have linear relation with the descending angles. The distribution function of the sand content of the wind-blown sand is verified through the statistical analysis.

A configuration for sequential holographic recording based upon a new system for obtaining a train of spatially separated light pulses at frequencies ranging from 6 MHz to 35 MHz has been developed. The multipulse system uses a high energy Q-switched Nd:YAG laser as a light source,and incorporates a phase-front preserving optical delay line and a specially graded beamsplitter to produce up to ten spatially separated light pulses of nearly equal energy. The temporal spacing between successive output pulses may be varied discretely in increments of 28.3 ns from 28.3 ns to 169.8 ns. The system is currently used for high-speed time-resolved holography of dynamic events with lifetimes between 100 ns and 1.5 microsecond(s) . Other applications include using the system either as a single point or `phased-array' source for laser generation of ultrasound.

The interaction of laser radiation with materials can lead to the generation of thermal and acoustic transients. If the laser pulse is of short duration and high power as in the case of a Q switched Nd YAG laser, then shock waves may be formed with fast rising pressure fronts. The optical diagnostics of laser generated sound waves therefore requires high spatial, approximately 10 micrometers , and high temporal, approximately 10 ns, resolution, in order to make quantitative measurements. A nitrogen laser pumped dye laser with a pulse length of 0.4 ns has been used as the light source for a Mach-Zehnder interferometer. The interferometer was modified to enable dynamic photoelastic interferograms to be taken. Recording was by means of a CCD video camera and computer controlled digitizing frame grabber. This system has been employed to study laser generated waves at water-polymer boundaries and the diffraction of ultrasound by defects in the solid material. The isochromatic fringe patterns, together with an approximate Abel inversion procedure, allow the radial pressure distribution in the polycarbonate test samples to be calculated. The ultimate aim is to obtain the near field diffraction coefficients for use in NDT modeling.

A noncontact shock (or acoustic) pulse detection system has been developed using optical probing of the refractive-index change associated with the pulse. The pulse profiles are monitored by the transient photorefractive deflection of collimated He-Ne probe laser beams. An increase in sensitivity of detection is achieved by an angular magnification of the deflection by refracting the deflected laser beam through a peripheral section of a converging lens. The transient deflection is observed by using a knife-edge to block half of the probe beam and detecting the deflection signal with a fast photodiode. By using two or more probe beams at different displacements from the pulse source and detecting their deflection at different times, the pulse velocity and attenuation in the medium can be determined. A demonstration of the capability of this system is made by the velocity measurements for shocks in a tube and for spark generated sound pulses in various media such as gas, liquids, and solid. Results agree well with those obtained by conventional technique. The proposed technique is superior to the conventional methods in that it is noncontact and easily applied to any hostile environment.

For a given flow field obtained in a shock tube several visualization techniques are applied to reveal both the weak and the strong points of each individual method. Particular emphasis is put on two newly developed schlieren techniques, which prove to be most suitable for a detailed optical investigation of a large number of problems in fluid mechanics.

A color schlieren system, which accurately indicates density gradient directions, has been developed. This system differs from the classical schlieren system by replacing the dark/bright regions on a uniform background with colored regions, each color representing a gradient direction. A component description of the system is given, with the formation of the color schlieren image being explained. The new system is compared to other color schlieren techniques, and photographs of schlieren tests are included.

In this paper shearing interferometry as used at the Prins Maurits Laboratory for studying shock wave diffraction around structures is reviewed. The performance of the shearing interferometer is illustrated by a number of applications. It is shown that, for not too complicated flow geometries, accurate density distributions of the flow field can be obtained from an analysis of the recorded interferograms. A comparison of results obtained with a numerical code, based on the flux-corrected transport (FCT) method, and experimental data also are presented.

We briefly describe the origin, history, and present status of pico/femtosecond image converter photography in Russia to which the late Professor Boris M. Stepanov devoted his talent, knowledge, and life. Taking into consideration the scientific background toward the end of the forties to the beginning of the fifties, we show why this field of technology has attracted people and resources, who was involved in this activity, where R&D work and manufacturing facilities were established, and when and how this area was developed. The present level of pico/femtosecond image converter photography is described in general and future trends are outlined.

The use of a high-speed (up to 100 MHz) programmable pattern generator and special clock driver/translator circuits for clocking solid-state multiple output imagers is discussed. A specific example of clocking a developmental 256 X 512 two-port CCD is illustrated. Reference to a prior report of clocking an eight-port CCD is included. Future use in clocking a CID imager is discussed.

A new optically deflected streaking camera with performance of nanosecond-range resolution, superior imaging quality, high signal detectability, and large format recording has been conceived and developed. Its construction is composed of an optomechanical deflector that deflects the line-shape image of spatial-distributed time-varying signals across the sensing surface of a cooled scientific two-dimensional CCD array with slow readout driving electronics, a lens assembly, and a desk-top computer for prompt digital data acquisition and processing. Its development utilizes the synergism of modern technologies in sensor, optical deflector, optics and microcomputer. With laser light as signal carrier, the deflecting optics produces near diffraction-limited streak images resolving to a single pixel size of <EQ 25 micrometers ². A 1kx1k-pixel array can thus provide a vast record of 1,000 digital data points along each spatial or temporal axis. High signal detectability is expected from the combined advantage of sharp focusing, 75% QE 514 nm of a thinned backside-illuminated CCD, absence of any undesirable signal conversion involving diffusing emission or inefficient coupling, and low noise of CCD readout electronics. Since only one photon-to-electron conversion exists in the entire signal recording path, the camera responses linearly to the incident light over a wide dynamic range in excess of 10⁴:1.

The characterization of a 512 by 512 pixel, eight-output full frame CCD manufactured by English Electric Valve under part number CCD13 is discussed. This device is a high- resolution Silicon-based array designed for visible imaging applications at readout periods as low as two milliseconds. The characterization of the device includes mean-variance analysis to determine read noise and dynamic range, as well as charge transfer efficiency, MTF, and quantum efficiency measurements. Dark current and non-uniformity issues on a pixel-to-pixel basis and between individual outputs are also examined. The characterization of the device is restricted by hardware limitations to a one MHz pixel rate, corresponding to a 40 ms readout time. However, subsections of the device have been operated at up to an equivalent 100 frames per second. To maximize the frame rate, the CCD is illuminated by a synchronized strobe flash in between frame readouts. The effects of the strobe illumination on the imagery obtained from the device is discussed.

The Edwards-Duromedics (ED) mechanical heart valve prosthesis is of a bileaflet design, incorporating unique design features that distinguish its performance with respect to other mechanical valves of similar type. Leaflet motion of mechanical heart valves, particularly during closure, is related to valve durability, valve sounds and the efficiency of the cardiac output. Modifications to the ED valve have resulted in significant improvements with respect to leaflet motion. In this study a high-speed video system was used to monitor the leaflet motion of the valve, and to compare the performance of the Modified Specification to that of the Original Specification using a St. Jude Medical as a control valve.

Immediately after the launching of VCR in VHS Format by JVC for consumer use in 1976, NAC has also started to work on development of VCR in VHS Format for the first time under the licensing arrangement with JVC, and has succeeded in the development of a special type VCR in VHS Format in 1979 which is durable in the special severe environment required for the airborne use. NAC has worked on development of High Speed Video System by utilizing the technologies acquired from the development of VCR in VHS Format. NAC has then introduced in 1981 HSV—200 Sytem for the first time in Color as High Speed Video System. This HSV—200 SystemU is capable of recording at 200 fields per second with scanning lines of 262.5 lines per field based on the VHS (NTSC) Video Tape Standard. Since this has allowed to use the standard VHS video cassette, this System has provided the various merits; it has lowered its operation cost and the video cassette recorded by HSV—200 System has become possible to playback on the standard VCR of VHS Format for consumer use. This basic concept has been carried on all the subsequent models NAC developed to this date. Based on HSV—200 System, NAC has announced a serie of different models; FHS—200 specially designed for field use in 1982, HVRB—2002) designed for airborne purpose in 1983, and MHS—200 for use with a microscope in 1986. The recording speed of these models are all 200 fields per second and these models have been conveniently used for various applications in the new and different sectors of High Speed Video Instrumentation market. In 1987, a new feature of 400 fields per sçcond in half size was incorporated into HSV—200 System and a new model, HSV—4003) with field memory and various kinds of playback modes was announced. In 1990, HSV—l0004) System was introduced. This model is capable of recording at 500 fields per second in full size and at 1000 fields per second in half size. The VCR of HSV—l000 System has been adopted with new feature to handle S—VHS Format in addition to the current VHS Format and it has realized much higher resolution. The camera has also been improved by changing the imaging device from the current image pickup tube to the solidstate MOS Image Device and it has successfully reduced the lag practically to Zero of which resulted in a great improvement of Dynamic Resoluti on. In 1991, HSV—5005) System in S—VHS Format was developed, as the successor of HSV—400 System, based on the technologies cultivated at HSV—1000 System. HSV—500 System is capable of recording at speed of 250 fields per second in full size and 500 fields per second in half size in S—VHS/VHS Formats. HSV—500 and HSV—l000 Systems now have become the main product lines of NAC in the recent years. Followings are the major steps taken for the technological development of new features and functions of NAC High Speed Video Systems

High-speed, high x-ray energy imaging of implosions is a key diagnostic technique in the glass laser implosion program in the USA. With the correct x-ray energy, time-gated images can measure the symmetry and mix of the imploding shell into the stagnated fuel if the spatial resolution of 10 or 5 micrometers matched by a temporal resolution of 100 or 50 psec. Several 100 psec microchannelplate (MCP) x-ray pin hole cameras have bee installed and run on large laser systems with improving reliability. To increase the sensitivity of the imaging system, a ring aperture microscope has been coupled to a gated MCP detector allowing imaging at up to 8 keV.

High resolution x-ray streak cameras using tubes with bilamellar electron optics have been developed for a long time in CEL-V, in collaboration with Philips Components (now Philips Photonics) for the tubes. The last improvements on their electronics, implemented on the commercial cameras developed with ARP, have allowed us to get the best results from the tubes, especially for the quality of their focusing. We have already been able to verify that the performance of the C850X camera in the UV range is as good as expected from calculations. New experiments made in the x-ray range show that it is also true in the x-ray range. We describe the results of these experiments and we also show preliminary results obtained with the C750X camera (which is identical to C850X except that its deflection structure is a meander travelling wave line instead of conventional plates) which show that the sweep speed reached gives us a 1 ps temporal resolution in the x-ray range.

The heating and the sounding pulse are generated by the same laser. The laser output radiation is split into two independent channels, one of them is converted into an ultraviolet picosecond pulse by means of backward stimulated scattering. When a 3 - 10 ns pulse is produced by a Q- switched laser, the temporal compression of the sounding pulse is performed with a cascade Brillouin and Raman backward stimulated scattering. When a 0.1 - 1 ns pulse is produced by a mode-locked laser, the sounding pulse compression is achieved with a backward stimulated Raman scattering. The beam is then frequency-doubled into 315 nm in a BBO crystal. The ultraviolet pulsewidth is about 10 - 20 ps and its energy is larger than 0.2 mj.