Ladies and Gentlemen! It is a great pleasure to me to welcome you to the "16th International Congress on High-Speed Photography and Photonics" at Strassburg as German National Delegate.

This International Congress has been an example of co-operation and exchange of information between many Nations, indeed all and any Nations which chose to attend. The bond and common interest is the fascination of these techniques which not only provide data to solve problems but which also have artistic and "temporal" beauty. The progressive measurement of time into smaller and smaller instants is enough in itself to absorb thoughtful men.

We propose a photonic high speed analog data recording system that, when compared with conventional oscilloscope-based systems, offers economy, higher bandwidth, electrical isolation, and compactness. The basis of this approach is to record photon analog signals rather than analog electrical signals. A prototype High Speed Multi-Channel Data Recorder has been developed as a key component. Initial applications of this photonic system in pulsed power experiments have already shown its versatility and usefulness. The system, including photonic sensors, is described and its overall impact in diagnostic technology is discussed.

The evolution of ultra-fast cinematography from fast classical cameras to picosecond electro-optic and electronic cameras reaching picosecond temporal resolution allows the metrology of the shortest luminous events observed in ballistics, detonics, laser and plasma physics. This paper recalls some of the important steps concerning the development of framing cameras and streak cameras as well, which now allow the analysis of picosecond phenomena in almost all the spectral range between infrared and X-rays. Chronometers as well as oscilloscopes have been widely used in physics experiments, but they have been progressively replaced by electro-optic and electronic cameras which have the great advantage to give pictorial images of the physical event being analysed. The main improvements of cameras concern the spectral extension of photocathodes, the image intensification, the development of new electron-optics, the electronic circuitry and the image read-out systems. Besides the extension of the spectral range in the infrared with S1 photocathodes and toward X-rays with gold or CsI, in a sealed or open tube structure, we recall the principles, advantages and inconvenients of both framing and streak cameras.

A new high spatial resolution electrostatically focused streak tube is described. Previous streak tube design work emphasized shorter time resolutions with little regard for the amount of spatial information that could be obtained. The design goals of the initial version of this tube included a time resolution of only 100 ps, but 500 line pairs of spatial resolution across the 25-mm diameter of the photocathode (about twice the number of spatial resolution elements ,previously available). Computer programs were used in the tube design to simulate electron trajectories in the three-dimensional azimuthally symmetric electric imaging fields. Both tangential and sagittal trajectories were considered so that the tangential and sagittal focal surfaces could be designed to be in close proximity to each other for off-axis points, thus affording better spatial resolution. Test results obtained with the prototype tube are presented. The tube is limited in time and spatial resolution by the spread in the energy distribution of photoelectrons leaving the photocathode. The addition of an accelerator electrode close to the cathode is predicted by computer analysis to improve the time resolution to a few picoseconds and to double the useful diameter of the photocathode--doubling the number of spatial resolution elements. A modified version of the tube that includes an accelerator electrode is presently being fabricated.

The performance of the Picoframe I framing camera in a number of different experimental arrangements is described, and an exposure time capability of 130ps is demonstrated. The dynamic spatial resolution obtained in these schemes is shown to be in the range 42ℓp/mm to lOtp/mm (at the input photocathode). Further developments, including a method for obtaining multiple frames, are briefly outlined.

Better than 500fs(femtosecond) Temporal resolution for optical phenomena has been obtained using a newly developed subpicosecond streak camera. The temporal resolution has been measured with CPM ring dye laser system which can generate shorter than 300 fs laser pulse at 113MHz repetition rate. Another characteristics of the streak camera are also discussed.

Progress in laser fusion research has increased the need for detail and precision in the diagnosis of experiments. This has spawned the development and use of sophisticated sub-nanosecond resolution diavostic systems. These systems typically use ultrafast x-ray or optical streak caAleras in combination. with spatially imaging or spectrally dispersing elements. These instruments provide high resolution data essential for understanding the processes occurrilltg in the interaction. of high. intensity laser light with targets. Several of these types of instruments and their capabilities will be discussed. The utilization of these kinds of diagnostics systems on the nearly completed 100 kJ Nova laser facility will be described.

The P 700 is a new picosecond streak tube implementing a lamellar electron optics. Compared to the previous generation lamellar tube, the P 700 has an improved spatiotemporal resolution, a better photonic gain and an enhanced deflection sensitivity.

A magnetic focusing streak tube has been developed aiming at femtosecond time resolution. In focusing of photoelectron beam swept very fast, an unique focusing method has been used in order to improve the defocusing of the beam induced under such sweeping conditions. The limiting time resolution better than 500fs has been experimentally obtained.

The design features of a circular-scan streak tube which is intended as a primary receiver component in a spaceborne laser-ranging system are reviewed. Preliminary experimental data are also presented which indicate that a circular-scan Photochron streak camera has an instrumental function of less than 6ps in both single-shot and repetitive modes of streak operation.

The Development Of A Streak Camera Using Microchannel Plate Electron Multiplier As The Cathode Is Presented. Potential Applications To Y-Rays, Low Energy X-Rays And Charged Particles Are Described. Preliminary Test Shows That 150 Ps Resolution Can Be Obtained.

A universal streak camera which works in synchroscan and single-shot modes with 3 plug-ins (synchroscan, fast and slow) has been developed utilizing a microchannel plate-incorporated streak tube. The synchroscan plug-in features low jitter of less than 4 ps, and high-speed sine-wave deflection from 80 to 160 MHz, achieving a limiting temporal resolution of 5 ps. The fast plug-in offers temporal resolution of better than 2 ps with triggering jitter of less than t20 ps. The slow plug-in provides longer time windows from 10 ns to 1 ms/15 mm in 16 ranges. Shutter operation down to 100 ns in duration is available with a maximum repetition rate of 1 - 10 KHz for the 3 plug-ins. An on-off ratio of more than 1:106 has been achieved by using double gate operation at photocathode and built-in microchannel plate.

The number of streack cameras used in research laboratory has been continuously increased du-ring the past years. The increasing of this type of equipment is due to the development of various measurement techniques in the nanosecond and picosecond range. Among the many different applications, we would mention detonics chronometry measurement, measurement of the speed of matter by means of Doppler-laser interferometry, laser and plasma diagnostics associated with laser-matter interaction. The old range of cameras have been remodelled, in order to standardize and rationalize the production of ultrafast cinematography instruments, to produce a single camera known as TSN 506. Tne TSN 506 is composed of an electronic control unit, built around the image converter tube it can be fitted with a nanosecond sweep circuit covering the whole range from 1 ms to 200 ns or with a picosecond circuit providing streak durations from 1 to 100 ns. We shall describe the main electronic and opto-electronic performance of the TSN 506 operating in these two temporal fields.

kVe have improved the performance of the avalanche-transistor deflector-driver (sweep) circuitry used in the high-speed, electrooptic streak camera at Lawrence Livermore National Laboratory (LLNL). In the previous design for the sweep circuit, trigger-to-output delay time drifted on some cameras. This delay drift is a function of a somewhat randomly unstable breakdown voltage of some avalanche transistors. I.oth temperature and differences in manufacturing ethods for transistors affect this instability. However, a sig-nificant improvement in system performance is achieved by long-term burn in and by selection of only the most stable transistors for the sweep circuit. The peak-to-peak sweep voltage has been increased about 8O by increasing the number of avalanche transistors in the string and raising the Q of the resonant circuit. The result is an improvement in sweep uniformity by a factor of appoximately 2. Design equations for selecting components are given. Fast-recovery diodes are used to prevent undershoot and to keep the beam out of the intensifier field of view until after the intensifier is gated off. The sweep time range has been extended to over 100 ns.

Three kinds of x-ray streak cameras in wavelength regions ranging from several Å to 2000 It are developed. A time resolution in the shorter wavelength region is successfully improved to about 10 ps by adopting most desirable parameters, especially high electric field and by minimizing velocity distributions of electrons emitted from a photo-cathode. The design and the performance of these streak cameras are described. Plasma experiments are performed by a YAG laser with 30 ps of pulse-width and 100 mJ of output energies. Both time and spatial resolutions are investigated by using higher harmonics of a YAG laser and soft x-rays from laser-produced plasmas.

The SANDYL electron-photon coupled Monte Carlo code has been compared with previously published experimental bremsstrah-lung data at 20.9 MeV electron energy. The code was then used to calculate forward-directed spectra, angular distributions and dose-reduction factors for three practical configurations. These are: 20 MeV electrons incident on 1 mm of W + 59 mm of Be, 45 MeV electrons of I mm of W, and 45 MeV electrons on 1 mm of W + 147 mm of Be. The application of these results to flash radiography is discussed.

Flash radiographs have long been useful for interpreting dynamic events. Such interpretation ranges from merely looking at a radiograph to aid in understanding the event to fully deconvolving film densities to reconstruct actual masses and positions. Between these two extremes are the possibilities of using trained observers to select features of interest or of using densitometers to aid in such descriptions. Several examples of these latter methods will be illustrated to indicate that reasonable precision can be obtained in this way and that flash radiography can indeed.be a quantitative science. The examples will be from the field of shock-wave hydrodynamics.1 The radiographs were taken at the PHERMEX facility,2 a 25-MeV machine emitting about 30 R at a meter in a pulse of 100 ns for these examples. The first situation involves the selection of the position of a thin, low-density beryllium foil embedded in a moving high-density material, uranium. The second will involve the selection of the location or of the alignment direction of step mass discontinuities such as those found in shock waves or in interfaces between materials of different density. Such discontinuities may be slowly curving, and the problem is that of determining tangents at selected locations. Both spall in lead and multiple intersecting shocks in lead will be described. The third situation will involve the selection of the alignment of the head of a rarefaction wave--the problem of a density gradient with a discontinuous first derivative. The example will be oblique reflection of a detonation wave from an open edge in the explosive PBX-9404. The last situation will involve the problem of a material density gradient with continuous first derivative. The example will be low-density material ejected from a lead surface that has been shocked. In this last case the answer is not a single selection, but a variation as a function of location.

In explosively driven experiments, flash radiography can record a wealth of information about material densities and boundaries. Obtaining accurate quantitative data from these radiographs requires careful design of the experiment so that one can control and measure the scattered radiation background that is a part of any experiment. We have used collimators at the x-ray source to match the incident x-ray flux to the transmission of the object, thereby reducing the production of scattered radiation while still preserving a complete view of the object. Multi-hole collimators (at the film plane) with a length-to-diameter ratio of -20:1 have been used to measure the scatterad radiation field with several exposure geometries and with various shielding methods.

We present a paper giving a survey about Flash X-Ray Diffraction (FXD) and its progress during the last years. After a brief introduction and a historical survey we determine how to improve flash X-ray systems in order to use them for FXD. A first possibility for increasing the emitted radiation by flash X-ray tubes is to in-crease the voltage. Some authors were successful in applying this method to the recording of powder patterns using 300 kV, 10 ns, pulses. Another possibility for increasing the emitted intensity without increasing the voltage is to use a pulse generator with a very low self-inductance, for example a Blumlein line, which enables attainment of a high discharge current. Finally it should be noted that the intensity of the X-rays falling on the sample can be in-creased by an appropriate shape of the anode in order to keep the focus very small 1,5 mm in diameter). The sensitivity of the detector can be appreciably improved by using highly sensitive films together with fluorescent screens. Another method is to use an image intensifier or a microchannel plate. FXD has been applied to the analysis of the evolution of different kinds of patterns (Laue, Debye-Scherrer,...) of cristals submitted to high deformation rates. We give examples of such rapid changes in the crystalline state due to exploding foils, shock waves or shaped charge effects.

GREC, an acronym for "Generateur de Radiographie Eclair" was designed for visualization, measurements and studies on dynamic behavior of materials under high explosive loading. Since the first operating time, 1975, this facility has been improved, taking advantage of progress in the techniques involved, especially focusing and transport of intense electron beams. Operating with a 2 m, 56.Q MITL,-dose levels of 250 + 50 R at a distance of 1 m from the source were routinely achieved. By employing a higher impedance MITL (free space impedance 150.E ), plasma erosion switches to eliminate prepulse and increase the input current rise time, shaped cathodes to improve pinch quality, and a low Z, thin foil over the high Z (tantalum) target, output dose levels of 550 + 50 R were routinely obtained. Using plasma erosion switches, an average increase by a factor of two in the input current rise time was observed. Spot size measurements with both 56 Ω and 1508 MITLs are compared. The direct areal masses measurement method by means of a multihole leaden collimator is described. The results obtained with the new MITL show that not only the dose but also the spectral hardness of the photon beam have been increased.

A facility for high-speed cineradiography at high energy designed to study detonic effets in framing or streak mode is operated since 1983 in VAWOURS (Commissariat a l'Energie Atomique). This facility includes two parts : a linear accelerator as a source of X photons and an opto-electronic recording system with 1 to 3 cameras depending on the observation mode (framing or streak). The linear accelerator, a travelling waves accelerator (mean energy : 45 MeV), built by CGR-MeV delivers 30 R at 1 meter for each of the three 50 ns short pulses in a 15/us period or 30 R//us during a 15/us pulse. The filling time of each section could allow, if necessary, 15 pulses of 30 to 100 fps. Three image-intensifier tubes ITT-F 4113 with variable luminous gain (50 to 500) or a streak camera CFTH-TSN 506 are used to record the images. The electronic cameras are in a shrapnel-proof mobile vessel designed by the C.E.A. This vessel can be moved along the fire table in order to be suited to different detonic experimentations.

Some peculiarities of X-ray pulse bursts production with the high pulse repetition rate have been considered. Two designs of pulse generators, based on the X-ray thermal emission tube and the X-ray tube with a burst-emission cathode, are described. The units can generate pulsgs with the energy up to 150-500 keV and the pulse repetition rate in a burst up to 54-10 Hz.

An X-ray-acoustic detector was realized which is sensitive for extremely soft X-ray flashes (5 eV - 10 keV, T < 1 μS). The total dose of the flash is measured by the X-ray acoustic effect. The absorbed energy induces an acoustical pulse which is registered with a microphone. The microphone signal depends on the absorbed energy only. It is independent of the wavelength of the X-rays. The dynamic range is limited only by the constance of thermal properties of the X-ray absorber and by the noise of the preamplifier. The influences on the signal are discussed for some properties of the absorber and for the X-ray flash time.

Flash radiographic bio-medical studies at sub-microsecond intervals were performed by using both a new type of flash X-ray(FX) apparatus with spectrum control functions and Fuji Computed Radiography(FCR). This single flasher tends to have a comparatively long exposure time and the electric pulse width of the FX wave form is about 0.3,usec. The maximum FX dose is about 50mR at 1m per pulse, and the effective focal spot varies according to condenser charging voltage, A-C distance, etc., ranging from 1.0 to 3.0mm in diameter, but in the low dose rate region it can be reduced to less than 1.0mm in diameter. The FX dose is determined by the condenser charging voltage and the A-C distance, while the FX spectrum is determined by the average voltage of the FX tube and filters. Various clear FX images were obtained by controlling the spectrum and dose. FCR is a new storage medium for medical radiography developed by the Fuji Photo Film Co., Ltd. and this apparatus has various image forming functions: low dose radiography, film density control, image contrast control, subtraction management and others. We have used this new apparatus in conjunction with our FX radiography and have obtained some new and interesting biomedical radiograms: the edge enhancement image, the instantaneous enlarged image, and the single exposure energy subtraction image using the FX spectrum distribution.

In 1826 Abel presented the method of obtaining the profile of a circularly symmetric 2-D object from its projection. It is shown that the extension of Abel inversion, the reconstruction of a 3-D axially symmetric object from a single radiograph, offers significant benefits as an image analysis tool. These benefits include improved delineation of material boundaries, enhanced. display of minor deviations from axial symmetry, as produced by defects, and accurate estimation of the linear attenuation coefficients of the materials. Frequently this technique allows observation of features in the object that are too subtle to be seen in the original radiograph.

This paper deals with the fundamental block diagram peculiarities of the system for investigation of high-speed processes in optic and X-ray ranges as well as in electron beams. The system comprises the high-voltage pulse generator based on the iron-free pulse resonance transformer with shock excitation and spark-gap switch in the secondary circuit. Different X-ray and electron beam tubes connected with the generator by a flexible cable are used in a load. The system design and its high-energy parameters make it possible to work with two and more loads simultaneously. The system possibilities become wider due to the computer processing of the images obtained.

Computorized tomography has been extensively used for investigation of the spatial distribution of refractive index resulting from optical probing /1/. The information about tomographic projections in such investigations is carried either by interferograms or shadow-grams. One first needs to process and decode the pictures to put the projection data into a computer. This procedure is rather time-conconsuming. Yet, there is quite a broad range of problems,such as nondestructive test of glass fiber manufacturing, express-analysis of plasma, aero- and hydrodynamic processes,which are characterized by a huge amount of experimental projection data and/or the requirement to obtain results in the real time scale. The authors have earlier proposed a one-step method for direct measurement of refractive index distribution patterns called tomo-graphic interferemetry /2/. The method essentially consists in combination of the probing step, i.e., obtaining projection data, with the following step of tomogram reconstruction which is done by analog processing in the optical systems of tomographic interferometers. Holographic interferometry technique is used to visualize the synthesized wave field that carries tomogram information.

The axisymmetric microscope, first discussed by Wolter1-2, provides high resolution and sensitivity for investigating the soft X-ray emission of laser-driven plasmas3-4. Such a device having a 10 X magnification has been constructed. Its resolution has been evaluated empirically using several grids backlighted by either an X-ray or a visible light source. Due to surface roughness and departure from theoretical contours, the X-ray resolution was found to be near 25 µm whereas the visible light image of the 12 tim grids was still good. We present a comparison between the images of laser-driven plasmas given by this microscope and by a 10 X pinhole camera. Until now these images were recorded on X-ray film. We have shown5 that film could be replaced by C.C.D. in a pinhole camera when the photon energy lies within the 1-10 keV range. Below 1 keV the quantum yield is too low so we have used an image converter tube made by RTC. It is a diode-inverter tube with a soft X-ray photocathode and a P20 phosphor deposited on an optic fiber plate. The electronic image appearing on the screen is read by a C.C.D. working in the visible spectral fields. An electronic image readout chain, which is identical to those associated with streak cameras6, then processes automatically and immediately the images given by the microscope.

The idea of using a laser for materials processing is more than 20 years old. Although the concept of a non-contact method for processing with a beam of light has been pursued with great interest and enthusiasm, the practical use of laser beam processing was slow to develop. The lasers available in the 1960's were fragile and of relatively low power. In the 1970's lasers in the multi-kilowatt range were developed but the problem of laser acceptance by the customer had to be overcome. Today, reliable Nd-Yag and CO2-lasers are available and laser processing is a fast growing market. An additional boost is expected with the development of the next generation of lasers and with increased knowledge of the physical phenomena that underlie laser material processing. This paper will review latest developments in laser technology and laser-workpiece interaction with special emphasis on the impact of high speed photography on the research work in these areas.

Recent experimental data are reported which show that an S20 Photochron IV streak camera has a temporal resolution of O.7ps at 617nm in the single-shot mode of operation. Results are also presented for a Photochron IV Synchroscan streak camera with which an instrumental function of l.7ps has been demonstrated. Additionally, an application involving a repetitively-operating demountable miniaturised "Photochron IV-M" is described from which it can be directly inferred that negative electron affinity photocathodes can be made to exhibit photoemissive response times'of less than lOps.

The temporal characteristics of the frequency doubled pulses of a continuous-wave ,Iode-locked or mode-locked and Q-switched Nd:YAG laser have been investigated by means of a streak camera operating in synchroscan (Fig. 1) or triggered mode (Fig. 3). The second harmonic of the cw mode-locked Nd:YAG laser using a KTP crystal, has resulted in 532 nm optical pulses of 60 ps duration with a conversion efficiency of 12 % (Fig. 2A). The frequency doubled output of the continuously pumped acoustooptically Q-switched and mode-locked Nd:YAG laser showed pulses of 80 ps width (Fig. 2B). Stable, reproducible and single pulses at 560 nm of 10 laJ energy and 30 ps duration have been obtained in a synchronously pumped and cavity dumped dye laser (Fig. 4). The cavity dumper consisted of a Pockels cell with avalanche transistor driver. Single shot (Figs. 5A,B and 6A,B) and 500-800 Hz repetition rate (Figs. 5C and 6C) measurements have been carried out by means of a triggered streak ca-mera. Moreover, interesting information of the pulse formation in the cavity has been obtained by triggering the dumper at the beginning (Fig. 5) and at the end (Fig. 6) of the pulse train : for the last condition it appeared short and symmetrical pulses (FWHM 25 ps).

A picosecond synchroscan photometer has been developed to obtain a picosecond temporal resolution and an improved dynamic range of more than 105 : 1, utilizing a newly developed synchroscan streak tube and a direct photomultiplier readout. Advantages include a potentially improved S/N ratio, a lower detection threshold and reduction in cost. A greatly improved high dynamic range of 2 x 105 : 1 with a temporal resolution of better than 47 ps has been demonstrated, by measuring picosecond pulses from a laser diode. The limiting time resolution of this system is estimated to be =10 ps. This technique and its limitations are discussed.

At present the ultrashort light pulses' (USP) shape and duration is measured by streak cameras with time resolution of the order of 1.10-" s /1/. Besides, USP measurement is frequently performed by nonlinear frequency converters (intensity correlators) (see, for instance, /2/) which utilize the second harmonic generation and provide information only on the pulse duration; this should be viewed as a shortcoming of such correlators. Although the intensity autocorrelation function G(2)(t) detected by this type of correlators and accompanied by the calculation of the phase of the pulse intensity Fourier spectrum makes it possible to recover in principle the USP shape as well /3/, such recovery is ambiguous since for both the direct and the time-reversed pulses the function G (2) (t) is the same. At the same time, it has been demonstrated in /4/ that the USP shape can be determined if one measures the third order intensity correlation function (ICF) G3 (t1 , t2 ). The authors of /5/ obtained this quantity by using three-photon fluorescence. The present contribution suggests a method for ICF measurement which is based on the cascade third harmonic generation of the initial pulse frequency. The advantages of the procedure in question with respect to the method of three-photon fluorescence resemble those of the conventional second harmonic generation as compared with two-photon fluorescence.

In designing high-speed time-resolving image converter tubes (ICT) it is necessary to estimate beforehand the time resolution of the image converter system (ICS) to be employed. To obtain a correct estimate one must take into account the spreading of converged electron beam in the ICS due, first, to the scatter of the initial longitudinal velocity of electrons /I, 2/, second, to the scatter of the initial transverse velocity of electrons and the finite dimensions of the photocathode image (i.e., dut to different electron path deflections from the axis) /3/ and, third, to the electrostatic (Coulomb) interactions between electrons in the converged beam /4-6/. In /4, 5/ the focusing effect, which en-hances the interactions of electrons was not taken into acount; in /6/ the calculation were obtained with the help of computer.

A new, small-size image-converter tube type PIF-01 has been developed for recording of various high speed phenomena with subpicosecond time resolution. The designing goal of the new image-converter tube having four electrodes geometry (photocathode, acceleration mesh, focusing electrode, anode diaphragm) and incorporating with a modern deflection system, was to simultaneously improve temporal resolution and dynamic range.Preliminary experimental results confirm the efficiency of the software developed for picofemtosecond streak tube computation and optimization.

Several fast-gated imaging systems to measure ultra-fast single-transient data have been developed for time-resolved imaging of pulsed radiation sources. These systems were designed to achieve image recording times of 1-3 ms and dynamic ranges of >200:1 to produce large two-dimensional images ( ≥10⁴ spatial points) of 1-2 ns exposure and small two-dimensional images (>200 spatial points) of .0.5 ns exposure. Both MCP intensified solid-state two-dimensional framing cameras and streak camera/solid-state camera systems were used; the framing camera system provides snap shots with high spatial resolution whereas the streak camera system provides for limited spatial points each with high temporal resolution. Applications of these systems include electron-beam, x-ray, gamma-ray, and neutron diagnostics. This report reviews the characteristics of the major components of fast-gated imaging systems developed at Lawrence Livermore National Laboratory. System performances are described in view of major experiments, and the diagnostic requirements of new experiments in atomic physics (x-ray lasers) and nuclear physics (fusion) are indicated.

The control voltage sensitivity is one of the most important characteristics of pulse elec tam-optical image converter tubes (lC ). Losv-voltage broad-band shutter and image scanning rsystems make it possible to use semiconductors and sntall-size efficient electron tubes in voltage control circuits. This, in its turn, notably lowers the size, weight and power supply of the electron-optical devices.

Proximity-focus image intensifiers of the diode type or with microchannel plate may be used for high speed photography even in the subnanosecond range. For high speed gating the conductivity of the photocathode has to be greatly increased by a conductive layer. Additionally the connection of the image intensifier with the gating-impulse generator requires special measures. The behaviour of the gated image intensifier can be simulated using a simplified transmission line equation. Thereby all the parameters influencing the gating behaviour can be evaluated. All measures which are increasing the gating capability, that are increase of the cathode conductivity, reduction of the optically active cathode area, and increase of the electrode distance, also affect other parameters in a disadvantageous way. Thus a compromise between the different requirements is necessary. Exposure times even in the sub-nanosecond range are possible. The experimental evaluation of the exposure time generally can be done by measurements of the temporal behaviour of the luminous screen. In the subnanosecond range, however, this may require other methods due to lack of extremely fast photo-multipliers.

The sweep speed and dynamic range performance of an electronic streak camera is for a part defined by its photon gain. RTC-HYPERELEC have developed a wafer type MCP image inten-sifier with a 40 mm useful diameter which is particularly well suited to the amplification of images from large output format streak tubes. The design and technology parameters are discussed. This tube - fitted with F.O. input/output windows - has a minimum gain of 500 W/W at 460 nm.

We have developed a high-speed, electro-optic camera that features a large format; eight frames; and an accompanying eight-pulse, ruby laser illuminator. This system offers greater resolution and versatility than has been previously available. The eight 75-mm frames provide high dynamic spatial resolution (15 line pairs/mm) and time resolution (as short as 10-ns exposure time) to record fine physical details of rapid events. Four independent two-frame (tube) cameras complete with objective optics comprise the eight-frame array. Finally, pulsed ruby laser illumination with spectral filtering allows photography in intense ambient light.

A laboratory version of electron-optical camera permitting 9-frame photography of radiation in the wavelength range of from 0.4 to 1.2 has been designed and tested.

The S1 photocathode was the first cathode available for practical applications ; in spite of this its mechanism of photoemission has remained enigmatic. S1 semitransparent photocathode is the only one that can be used to study the 1.06 μm neodynium laser pulses of less than 10 ps duration. This recent application and the difficulties to manufacture stable and sensitive S1 photocathode at this wavenlength gave rise to new researches which aim is to have a better knowledge of this structure. We first review the recent results obtained at the Paris Observatory (research sponsored by the CEA) and report on the lifetime in the 1-μm range of the photocathodes processed four years ago. In a second part we will try to analyse the researches which have been investigated during these last ten years in different laboratories to determine the role of the main constituants (silver particles, Co oxydes) and their contributions to photoemission in order to improve the sensitivity and the stability of S1 photocathode.

Streak camera tubes have been developed with alkali antimonide photocathodes. At the be-ginning it was mainly monoalkali type S 9 or S 11 (caesium antimonide). These cathodes have a good sensitivity in blue and are currently used in photomultipliers for nuclear detec-tion. From theoretical and experimental results obtained in laboratory (Laboratoires d'Electronique et de Physique Appliquee - LEP), development studies on multialkali antimonide thin films have been recently carried out on industrial equipments at the Hyperelec factory. It is now possible to produce with a good reproducibility thin films with high red sensitivity. Particularly the use of transfer technique allows to obtaina very high sensitivity in the bandwidth of 800 - 850 nm. This characteristic is very important for calibration of streak camera tubes because it is made by using LED or ultra fast laser diodes emitting in the red. It is also important for increasing the detection sensitivity when streak tubes are used in connection with long fiber optics which transmit preferably red signal. The paper will describe the main parameters on which it is possible to act and the photo-cathode results expected on streak camera tubes processed by transfer technique.

The S-1 semi-transparent photocathode is the only one that can be used to study the 1.06 μm neodynium laser pulses of less than 10 ps duration. We first reviewed the recent results obtained at the Paris Observatory (research sponsored by the CEA), and then we tried to determine the role of the main constituents and their contributions in photoemission.

Coulomb corrections to temporal resolution of electron-optical image converter systems are estimated. A general treatment of the self-consistent field Coulomb problem for an electron bunch shows the temporal broadening of the electronic signal at small acceleration dis-tances to be mainly due to momentum space relaxation which is difficult to suppress.

Application of the 16 mm high-speed camera synchronized with a digital storage oszillograph for the process analysis of stochastic events on the sector of welding technique, camera recording technique and evaluation with film examples.

The combustion phenomena of stratified charge combustion in a divided combustion bomb simulator under the initial conditions of room temperature and atmospheric pressure was investigated by using combustion bomb pressure record and high speed schlieren photography in comparison with that of homogeneous charge combustion in the same combustion bomb. For the demonstrative illustration of high speed schlieren photography to show the characteristics of stratified charge combustion and homogeneous charge combustion, four different cases of fuel/air mixture charging combinations were chosen. The results show that stratified charge combustion can reduce its total burning time comparing with the total burning time of homogeneous charge combustion and pressure rise-up of stratified charge combustion is rather faster and higher than that of homogeneous charge combustion. It is also indicated that stratified charge combustion can enhance its leaner inflammability limit than homogeneous charge combustion. The detailed high speed schlieren photography visualize small wrinkled or turbulent flamelets and discrete vorticity associated with flame jet expanding into the main chamber through the connecting orifice between the pre-chamber and the main chamber.

Photography techniques past and present, are briefly reviewed. The advantages and limitations of side valve and over-head valve engines for combustion research are discussed. The paper presents a method by which high speed photographs of the combustion process are analysed to obtain flame speeds and rates of combustion. Results obtained using a side valve engine are presented.

The principle of the new approach taken consists in viewing the fuel injection and combustion process through specially designed endoscopes. Compared to conventional film techniques, the endoscope as photography system requires very small windows fitted into the cylinder head of the engine, thus causing minimal interference with the combustion process. The windows are made of fused silica to withstand the high gas temperature and pressure inside the combustion chamber of a diesel engine. The outcoming image from the endoscope is transmitted to the camera through several rigid optical relay systems, interconnected with rotating optical joints. A flash bulb located behind a second window is used to illuminate the non-luminous process of fuel spray formation during the ignition delay period.

To investigate the penetration of liquid jets in a gas, the methods of high speed cine-matography and short time interferometry are particularly qualified. Therefore a set up containing a laser, an interferometer and a rotating mirror camera was designed, built and tested. The results are demonstrated by special examples.

This paper describes analysis and design of a digital video image processing system the authors have developed. Emphasis is laid on improvement of the throughput including the interactive operation in image processing of a film taken with a high speed camera by using advanced digital video techniques. An example of the application of this system to the measurement of temperature and soot concentration of flame in a diesel engine has also been presented.

Araidite models of two types of notched bar, namely three point bend and double notched bar, were subjected to an impact from a falling weight pendulum The resultant dynamic photoelastic patterns were captured using a 100 nano second duration spark source, a lens polariscope and a still camera. The spark was fired at different times, to an accuracy of 1 microsecond, for each reproducible impact and hence a sequential series of photographs built up. Dynamic stress intensity factors at the notch tips were measured from the photo-elastic fringes at the given times.

Some of the mechanisms involved in the reduction of wood chips to papermaking pulp in a commercial disc refiner have been determined by high speed photography. Flow patterns of pulp through the refiner, including an unexpected recirculation pattern, have been recorded. Cine-photography was also employed to show how wood chips are transported by a ribbon screw feeder into the refiner. Some aspects of photographing in a hostile environment are described. The following salient observations have been made during these studies. Chips and dilution water fall to the base of the feeder housing and are fed along it to the refiner eye, where the chips are reduced to coarse pulp. This coarse pulp proceeds through the breaker bars into the refining zone. Some pulp in the inner part of the refining zone flows back to the breaker bars along grooves of the stationary plates, giving rise to considerable recirculation. Pulp in the outer part of the refining zone moves radially outwards. For a short fraction of its passage through the refiner, most of the fibrous material is constrained to move in the direction of rotation of the moving plates. Some of this material is stapled momentarily in a tangential orientation across the bars of both sets of plates. The immobilized fibres are then subjected to the refining action between the relatively moving bars before being disgorged into the adjacent grooves.

The protection of people and property must be assured by every situation around an industrial power plant. That is why the FRENCH Commissariat a l'Energie Atomique has defined the size of the confinement of Super Phenix to withstand the worst highly hypothetical accident. The study of the strength of the confinement has been carried out by two complementary means : - Calculation (Display poster # 491 188), - Experiment : reactor mock-up. The latter is presented in the film. The solution which have been adopted for the problems encountered are emphasied ; the work with high speed camera is presented. The film is illustrated with some fast movie sequences.

A measuring system combining subnanosecond laser-induced fluorescence with microscopic signal detection was installed and used for diverse projects in the biomedical and environmental field. These projects are ranging from tumor diagnosis and enzymatic analysis to measurements of the activity of methanogenic bacteria which effect biogas production and waste water cleaning. The advantages of this method and its practical applicability are discussed.

The use of cinephotography for evaluation of disturbed knee joint function was compared in three groups of patients. While a sampling rate of 50 images per second was adequate for patients with neuromuscular disorders, a higher frequency of around 300 i.p.s. is necessary in osteoarthritis and ligamentous knee joint injuries, but the task of digitizing is prohibitive unless automated.

The investigation of certain areas of nuclear reactor safety involves the study of high speed phenomena with timescales ranging from microseconds to a few hundreds of milliseconds. Examples which have been extensively studied at Winfrith are firstly, the thermal interaction of molten fuel and reactor coolant which can generate high pressures on the 100 msec timescale, and which involves phenomena such as vapour film collapse which takes place on the microsecond timescale. Secondly, there is the response of reactor structures to such pressures, and finally there is the response of structural materials such as metals and concrete to the impulsive loading arising from the impact of heavy, high velocity missiles. A wide range of experimental techniques is used in these studies, many of which have been developed specially for this type of work which ranges from small laboratory scale to large field scale experiments. There are two important features which characterise many of these experiments:- i) a long period of meticulous preparation of very heavily instrumented, short duration experiments and; ii) the destructive nature of the experiments. Various forms of High Speed photography are included in the inventory of experimental techniques. These include the use of single and double exposure, short duration, spark photography; the use of an Image Convertor Camera (IMACON 790); and a number of rotating prism cine cameras. High Speed Photography is used both in a primary experimental role in the studies, and in a supportive role for other instrumentation. Because of the sometimes violent nature of these experiments, cameras are often heavily protected and operated remotely; lighting systems are sometimes destroyed. This has led to the development of unconventional techniques for camera operation and subject lighting. This paper will describe some of the experiments and the way in which High Speed Photography has been applied as an essential experimental tool. It will be illustrated with cine film taken during the experiments.

Considering the difficulty of achieving the goal of this session due to the large number of participants involved in the discussion (workshop, tuesday 28, 18 : 30), an ad hoc committee of concerned users was appointed to review the definitions and parameters relevant for specifying streak camera system characteristics.

In /1, 2/ were described high-speed optomechanical instruments for recording fast proceses in a broad spectrum of 0.2 to 10.6 μm, viz., the photographic framing camera ZSFK-2 and the streak camera (chronometer) ZSKh-1. Both cameras employ optical reflector systems. Presenting considerable improvement over optomechanical means of recording fast processes that employ refractive lenses, the ZSFK-2 and ZSKh 1 cameras still need refinement of their design and improvements of basic performance parameters such as photographic resolution and frame size. It is also desirable to supplement these instruments with spectral adapter units for wide-range spectral investigation of fast processes.

A Scanning record of high speed photoraphy can present only an one-dimentional information; moreover, for rotating Mirror streak cameras, there is a severe degradation of the temporal resolution. In this paper, we have theoretically explained that it is possible to approxmitaly get scanning records any spatial direction as desired from one or two framing records so long as given that the spectral band width of a transient event, the adaptable photographic rate depending on the maximitm velocity of the event, and te adequate number.

The present paper counders the purpose of high-speed VK-12 camera. Its specific features are discussed. It is pointed out that the camera is provided with a set of accessory input objectives and plug-in units thus allowing the camera to be used in framing and steak operation modes. The removable relay lenses plug-in units with 1-2-4 row arrngement make it possible to provide a wide range of framing speeds ranging from 50 x 10) to 4,8 x 10--) frames per second. The optical trains of the camera in framing and streak operations are given and the camera operation in both modes are discussed. The circuit diagram for camera control is discussed briefly. The technical characteristics of the camera are given.

To enhance the time resolution of high-speed cine equipment during the investigation of rapidly flowing processes, a light source to illumi late an object under test is represented b7 solid-state laser exposure devices operating in the mode of Q-factor flodulation. With a high-speed eine cafiera being run in the continuous scanning mode, these devices will permit a sequence of fra Mlles to be obtained within a short exposure time of 150 ns to 200 nanoseconds. At scanning speeds of up to 250 m/s this will ensure satisfactory image quality from the slear viewpoint. In the case of faster continuous scanuin speeds and of shorter exposure times, it becomes necessary to run the high-speed cauera in the fl ode of frame-by-frame cinematography.

Overwhelming majority of high-speed cameras with optical-mechanical commutation (i.e., rotating mirror framing cameras) has a cylindrical film track' and it results, to a considerable degree, in their enlarged overall dimensions and weight. A rare exception is the high-speed camera discribed in report² when they managed to design a flat film track by adding a flat mirror to each commutating lens (i.e., framing lens) but the construction became more complex. Thus 48 flat mirrors should be inserted into the optical path at 48 frames being recorded. Nevertheless camerae is rather compact and provided with cassettes of "dry" developing that considerably raises photoresearch work rate. The aim of the paper is to introduce two other systems of mirror scanning, which also allow to get the flat film.

The experiments I want to talk about were published in steps over the years 1 . Improved diagnostics with subnanosecond time resolution now provide more detailed information about pulse shapes, liglatrise and time jitters. This has stimulated new interests for early spark channel development. The search for controlled highpower diffuse plasma presently is being organized in the puls-ower .prograrno Fast and low jitter switching appears a key application. Wanted is a plasma without a space charge controlled statistical streamer (Ii., .Haether), which determines spark breakdowns in many environments. The initial channel of the narrow gap "Na.noliten, however., appears to grow diffuse from both electrodes without streamer formation and no instabilities, as is demonstrated in Fig. 2°

A high frequency stroboscope for flash rates up to the 101-1z region will be described. The flashes are generated by light emitting diodes which emit - especially in the pulse mode - high light currents in the visible region. The pulse width of the flashes is controlled automatically in such a way that the mean light. current stays constant for flash rates between 100 Hz up to 1 MHz. The ratio between flash duration and period length is adjustable over a wide range, hereby the height of the light pulses is regulated internally for h.,-lean light current output. Some applications of the stroboscope in Mach-Zehnder-, 'N'Tirau- and Differential-Interferometry and an example for the observation of a desintegrating water-jet will be given.

A method of infrared photography on silver halide light sensitive materials is discussed. The idea of the method consists in the modulation of sensitivity to actinic light by the IR exposure. An attempt is made to explain the effect theoretically from the basic principles, and some experimental results of recording IR radiation with the wavelength λ = 10. μm are described.

A method of infrared photography with the help of silver halide films was described in /1-2/. The essence of the method is that the sen-sitivity of silver halide films to actinic light may be varied by additional IR radiation effected either immediately prior to or during the exposure. The best results were obtained when radiation with it = 10.6 μm was recorded on the Kodak High Copy 5069 film /3/; in that case the threshold sensitivity was ~102 cm2/J; dynamic range, 20; resolution, 100 mm-1; and maximum difference of the darkening optical density, ~0.4.

The characterisation of reactions occuring during the thermally induced explosion of an energetic material is hampered by severe experimental difficulties. Dynamic techniques are essential in the study of such fast events but, unlike in the case of, for example, flash photolysis, synchronisation is difficult to arrange. We suggest here an approach in which data is obtained at a high rate and stored continuously on a video recorder so that the need for synchronisation is eliminated. Its use is illustrated here for a mass spectrometry study of the explosion of mercuric nitrotetrazole.

Many of the limitations associated with conventional (incoherent) stroboscopic sources for high-speed photography can be overcome by the use of the copper laser for illumination. Properties of this short pulse, high-repetition-rate visible source are discussed and its use with a high-speed video camera described. Many of the limitations associated with conventional (incoherent) light sources for high-speed film or video recording can be overcome by the use of the copper vapor laser (CVL) for illumination. These limitations include: 1) mechanical shutter-speed constraints 2) limited repetition rate of strobes 3) relatively long pulse widths (4 microseconds) 4) proximity requirements of light source to subject 5) substantial target heat-loading These problems are reduced or eliminated by the use of the copper laser for illumination. The CVL is a pulsed, visible laser which has recently gained wide commercial acceptance as an efficient, rugged laser source for a variety of scientific and commercial applications. The laser produces pulses of 2-8 mJ in energy at a repetition rate of 4-6 kHz. The resulting total average power is 10-40 W at 510 nm and 578 nm. The laser's repetition rate exceeds by a factor of three to ten that of the fastest conventional strobe sources, allowing blur-free multiple images within a reasonable illumination field size. For example, a projectile traveling at 1500 m/s moves a distance of less than 50 microns during CVL illumination and only about 25 cm between pulses. Thus the copper laser can be used to measure not only projectile velocity but acceleration, spin rate and spin acceleration as well.

This paper describes a steel rotating mirror. The mirror has been applied in the high speed photographic camera of the compensation type as the optical compensator. The article discusses the compensation principle of the optical unit and equation of parameters calculation. We use it as the optical compensator which has given optical compensation for a moving film up to 150m/s. The camera rate is raised to 10⁵ fps.

During the tests in wind tunnel, flow visualization techniques offer to the aerodynamicists general views of the whole or part of flows they are studying. Even when the image observation only provides qualitative informations, the high density of these informations leads to a synthetic interpretation of the very complex physical phenomena which take place in the course of the interaction between a flow and a solid in relative motion. The variety of these techniques arises from the problem diversity according to the fluid relative velocity and to the solid wall proximity. On the wall, informations are obtained by examination of the flows of more or less viscous, colored or not, liquid spread on the surface, or still thin tufts motion. Out of the wall, the techniques to be used are different according that the external flow velocity is more or less fast as that of the sound in this medium. In subsonic flow, the air is seeded with tracers in the shape of small solid or liquid particles. Observed individualy or in groups, the particle motion informs on this of the surrounding air. Images are obtained thanks to the particle scattering properties regarding the illumination light. Changes in the illuminating conditions (total illumination or by a light sheet) allow to show particular features such as vortices, separation and recirculating zones, ..In the supersonic domain, new physical phenomena appear, providing rough or progressive changes of the gas density (particularly, shock wave apparition). These density variations involve propagation light perturbations (phase changes in a light wave or light ray de-viations in a beam). It is the domain where techniques such as interferometry, schlieren and shadowgraphy are developed. They provide not only qualitative visualizations, but, in certain cases, such as two-dimensional or axisymetrical flows, they allow density distribution determinations. Some phenomena are very brief (detonation wave propagation) or evolve fast (turbulent structures). In order to allow a detailed analysis, it is then necessary to fix temporarily some aspects of these phenomena. Separate pictures are obtained thanks to fast shutters, pulse light illuminations (spark lamps or lasers), or image converter cameras. Sequences with high repetition rate (from 25 frames per second to beyond a million of frame per second) necessitate the use of special materials: high speed cameras (ordinary or with compensating prism), rotating drum or mirror cameras, image converters,..To all these recording means on photographic film, is now added the video, recording on magnetic band. Although it does not yet reach the same number of pixels in an image, the video presents some advantages for its easy use, fast restitution and long play recording. in the field of high speed video, frame rates of 200 and 2000 frames per second, in full image, are obtained. More, the video development now allows to take advantage of the image observation informative richness to facilitate the interpretation of large data sets acquired either by measuring means or at the computer output. The aerodynamicists are particularly interested with it for their result presentation and comparison between experiment and calculation.

The paper contains the description of high-speed photographic systems applied on the ballistic ranges of A.F.Ioffe Physical-Technical Institute of the USSR Academy of Sciences. The specific features of these (shadow, interferometric, holographic and spectral) systems application for physical-gasdynamic research are considered.

Semiconductor devices are sensitive to X-rays. Array sensors can be constructed with integrated circuits as CCD's. Their application -to X-ray imaging is discussed especially to get a high spatial resolution. This is shown to be possible with some 10 μm resolution using a CCD-sensor covered with a converter screen. Furthermore such a screen protects the CCD against radiation damage. It could be demonstrated that soft X-radiation produces most of the damage. The life time of a sensor can be enhanced by a factor of more than 100 if a converter screen absorbs most of the soft radiation.

X-ray imaging adapted to the laser-matter interaction experiments consits in recording plasma images from its X-ray emission ; those phenomena have between 100 ps and some nanoseconds duration. When we only need spatial information on 1-10 keV X-ray emission, the most simple imaging device is the pinhole camera ; the two dimension image of the plasma is temporally integrated by an X-ray sensitive detector. Until now, X-ray film was used. Its operation and processing were long and tedious, so we replaced it by a television camera built around a Charge Coupled Device (C.C.D.). This camera is directly integrated in the pinhole camera. The X-ray detection is made by the silicon substrat of a C.C.D. without input window working in the vacuum of the expe-riment chamber ; a compact camera head (40 mm diameter, 120 mm length) located near the C.C.D. (1 to 2 cm) makes the charge/voltage conversion and the signal amplification. The immediate operation of images is done by an image acquisition and processing unit after digitizing the video signal on 8 bits. From measurements made on a continuous X-ray source (5,4 keV) we could point out the fact that a THOMSON-CSF THX 31135 CCD is 10 times more sensitive than the X-ray SB2 KODAK film that we use in pinhole cameras. The dynamic range measured in these conditions was about 300. The first experimental results obtained on a pulsed X-ray source are presented.

We describe a method for designing micro-electrophotonic systems to be integrated at a high level. This process does'nt require a mask and we can control the shape of the synthe--sized profiles with a high precision. We have also developed an algorithm to optimize the tomographic reconstruction of 3-D images. This method was applied to optic fibers.

The installation and operation of high power lasers, requires to know both the spatial repartition of energy in the beams and their location. Infrared films are less and less frequently used because they require a long and tedious processing. They have been progressively replaced by television systems built around tubes (vidicon with silicon or PbSe targets). Those systems simplify the laser alignment operations, nevertheless their limited dynamic range is an important drawback. In order to replace them by a new generation of cameras, we have studied several CCD sensors and we present the last result obtained when illuminating them with C.W. laser light at X = 1.06 μm and 0.35 We have also tested the behavior of CCD illuminated by a 50 ps laser pulse at X = 1,06 μm. These measurements show that the THOMSON TH 7861 CDA FO sensor has the best features for our applications. The associated television camera is also described.

Beeing a powerful way to catch high speed luminous events snapshot Video finds a lot of applications in Instrumentation. For covering a large range of short time exposures (from continuous to 2,5 ns) three basic equipements have been developped and described among their various applications.

Whether your plant manufactures, assembles or packages computer components, chain saws, circuit breakers or other products, a common denominator for success is maximum productivity with minimum waste, rejects and returns.

High-speed video equipment can be used to optically image up to 2,000 full frames per second or 12,000 partial frames per second. X-ray image intensifiers have historically been used to image radiographic images at 30 frames per second. By combining these two types of equipment, it is possible to perform dynamic x-ray imaging of up to 2,000 full frames per second. The technique has been demonstrated using conventional, industrial x-ray sources such as 150 Kv and 300 Kv constant potential x-ray generators, 2.5 MeV Van de Graaffs, and linear accelerators. A crude form of this high-speed radiographic imaging has been shown to be possible with a cobalt 60 source. Use of a maximum aperture lens makes best use of the available light output from the image intensifier. The x-ray image intensifier input and output fluors decay rapidly enough to allow the high frame rate imaging. Data are presented on the maximum possible video frame rates versus x-ray penetration of various thicknesses of aluminum and steel. Photographs illustrate typical radiographic setups using the high speed imaging method. Video recordings show several demonstrations of this technique with the played-back x-ray images slowed down up to 100 times as compared to the actual event speed. Typical applications include boiling type action of liquids in metal containers, compressor operation with visualization of crankshaft, connecting rod and piston movement and thermal battery operation. An interesting aspect of this technique combines both the optical and x-ray capabilities to observe an object or event with both external and internal details with one camera in a visual mode and the other camera in an x-ray mode. This allows both kinds of video images to appear side by side in a synchronized presentation.

In the field of investigations and research on materials, as we shall see, the highspeed video observation technique is especially well suited to the gathering of visual information. Investigations of such structural materials as concrete and steel are generally based on the results of tests that are most often destructive. Since the phenomena preceding failure, and the failure itself, are rapid, little visual information is available other than that yielded by a preliminary examination of the materials and a check of their appearance after failure. Since the phenomena cannot be observed directly, we must have recourse to observation by means of images taken at4high speed. Moreover, with tests of this type it is not possible to determine the instant of failure, in advance, to within ten seconds or so, and this uncertainty must be overcome by a sufficiently long observation time. An SP 2000 system was used for the high-speed video recording of the tests shown, These tests consisted of compressing a series of cylindrical concrete test specimens at their edges and observing the formation and propagation of the cracks induced by the applied force. These observations were made possible by the speed and capacity of the SP 2000 system, which also offers a new and very real convenience of use, since the high-speed video technique provides real-time access to the information, both during the preparation of the test, in the form of a display on the monitor, and during the processing and interpretation of the data taken from the recorded images. Moreover, since this type of laboratory test is easily repeated, it becomes possible to conduct a parametric investigation of the phenomenon by crushing test specimens having dif-ferent consistencies in terms of type of cement and aggregate site and by varying, say, the loading rate, with the advantage of being able to alter a test according to the results of the last one, if necessary, thanks to the real-time availability of the information.

Photogrammetry, the science of measurement from images may be either two or three dimensional in nature utilising single images or pairs of stereoscopic images. The technique was developed originally at RR APD for the measurement of images from high energy x-radiography or neutron radiography. Recently, a programme of research has been carried out using a double flash argon spark source to arrest the motion of bubbles in a flowing stream of lubricating oil - two phase flow. A pair of stereoscopic images with single or double exposure have been obtained providing material for subsequent analysis in a stereocomparator It has been possible to derive information on local flow velocities from the double exposures and to measure the volume of gas in the bubbles flowing in the liquid - the so called void fraction. From a system calibration it has been possible to obtain local two phase pressure drop non-invasively using photogrammetry. A high speed video system, the spin physics SP2000 has been used to take mono and stereoimages of bubbles at 2000 pictures per second. Using the freeze frame mode it has been possible to obtain three dimensional images and obtain a better understanding of the dynamics of the various flow regimes. The SP2000 has now also been used to analyse a number of other engineering problems including flow from a reheat burner ring, birdstrike problems and the laser drilling of holes. The SP2000 has been found to be fast in its acquisition and display of images and the measurement system has provided quantitative information a short time after the experimental work. It is intended to pursue further the largely untouched field of high speed video - photogrammetry using the SP2000.

In packaging,vision of making process is difficult due to high speed of machines. The setting of box erecting is tedious. The origin of filling hazards tedious is often unknown. Some performances on line are not reproducible. To visualisate, decompose and analyse the fast moving phenomenons which are causing packaging troubles, the French Paper and Board Research Institute has obtained video NAC eouinment (200 pictures/sec.). An articulate stand able to move on a carriage is designed to facilitate shooting on machine. An elaborate image analysis is undertaken to find the characteristics of the hoards which are necessary to optimize making the folding box erection on the packaging lines. Actual image analysis with computer systems are large, expensive and necessitate a specific program for each problem. Important equipment are actually used exclusively for special fields. For packaging field where machines and products are diversified we have developed an easy electronical technique for picture analysis. This technique is suitable for all kinds of processes or defects, visualised by high speed video from video shooting on machine. The processes and the manufacture incidents are analysed, controled with a cell on video screen. The light intensity variations are detected and writed on self-recording apparatus. Materials move, their forming modifications and the moves of machine elements are expressed like "signature". All changes on "signature" show hazardous or reproducible variations according to defects of manufacture processes. A short time event, visible on few images only is located with normal or slower speed of the magnetic tape according to the importance of variation. This technique is used to measure in real time the packaging deformations on line. Ease for use, speed of setting and quantity of data are operating qualities of efficient image analysis.

The pursuit of quality is essential in the development and production of drugs. The pursuit of excellence is relentless, a never ending search. In the pharmaceutical industry, we all know and apply wide-ranging techniques to assure quality production. We all know that in reality none of these techniques are perfect for all situations. We have all experienced, the damaged foil, blister or tube, the missing leaflet, the 'hard to read' batch code. We are all aware of the need to supplement the traditional techniques of fault finding. This paper shows how high speed video systems can be applied to fully automated filling and packaging operations as a tool to aid the company's drive for high quality and productivity. The range of products involved totals some 350 in approximately 3,000 pack variants, encompassing creams, ointments, lotions, capsules, tablets, parenteral and sterile antibiotics. Pharmaceutical production demands diligence at all stages, with optimum use of the techniques offered by the latest technology. Figure 1 shows typical stages of pharmaceutical production in which quality must be assured, and highlights those stages where the use of high speed video systems have proved of value to date. The use of high speed video systems begins with the very first use of machine and materials: commissioning and validation, (the term used for determining that a process is capable of consistently producing the requisite quality) and continues to support inprocess monitoring, throughout the life of the plant. The activity of validation in the packaging environment is particularly in need of a tool to see the nature of high speed faults, no matter how infrequently they occur, so that informed changes can be made precisely and rapidly. The prime use of this tool is to ensure that machines are less sensitive to minor variations in component characteristics.

1. Introduction - Present situation of manufacturing industries and its targets In recent years, the trend to automate production lines has gathered increasing momentum across the whole range of manufacturing industries. Very high levels of automation are now commonplace in the manufacture of such diverse products as Automobiles to Video Recorders and minute Intergrated Circuits. Other industries manufacturing such products as foods, cosmetics, and pharmaceuticals have rapidly invested in automated methods, particularly for the filling and packaging processes.

For over half a century, High Speed Motion Picture Photography has provided a unique opportunity to analyse fast motion in industrial machinery, as well as providing special effects in Film/TV drama and documentary work.

It is the purpose of this paper to describe an opto-electronic method which allows non-contacting and non-reactive measurements to be performed of the space position of a rod-shaped body in free-flight such as a projectile. The beam of light from a laser is widened to form a thin lighting row oriented preferably normal to the direction of flight. As the subject under investigation passes through the plane of measurement, a light spot is observed on the surface of the test object, which drifts more or less strongly with the angle of attack and angle of precession. This motion is imaged on two uniaxial position-sensitive photodiodes. The photocurrent generated is recorded by means of a four-channel transient recorder which, in turn, is connected to an electronic computer. Here the data measured are analyzed, the velocity, angle of attack and angle of precession of the test object thereby being determined from the discrete absolute values of the photocurrents and their functions of time. The first measurements achieved show the angular resolution to be 0.1°.

Beschrieben wird eine optoelektronische Messmethode, die zur berUhrungslosen und rUck-wirkungsfreien Messung der raumlichen Lage eines freifliegenden, stabformigen Korpers, z.B. eines Geschosses, eingesetzt werden kann. Ein Laserstrahl wird zu einem &Innen Lichtband aufgeweitet, das vorzugsweise senkrecht zur Flugrichtung anzuordnen ist. Durchquert der zu vermessende KOrper diese Messebene, so kann an seiner Oberflache ein Lichtfleck beobachtet werden, der je nach Anstell- und Prazessionswinke1mehr oder weniger stark auswandert. Diese Bewegung wird auf zwei einachsige positionsempfindliche Photodioden abgebildet. Die hier erzeugten Photostrome werden mittels eines vierkanaligen Transientenspeichers registriert, der seinerseits mit einem elektronischen Rechner verbunden ist. Dort erfolgt die Auswertung der Messdaten, wobei sich die Geschwindigkeit, der Anstellwinkel und der Prazessionswinkel des Flugkorpers aus den diskreten Absolutwerten der Photostrome und deren Zeitfunktionen ermitteln lassen. Erste Messergebnisse liegen bereits vor und bestatigen eine Winkelauflo-sung von 0.1°.

The aims of this work were to demonstrate the feasibility of the visualization of the ignition phenomena in a translucent case, simulating a 20 mm ammunition case and to study the instrumentation, including the data processing, suitable for this kind of experiments. We observed many conventional and experimental 20 mm ignition systems, in order to test the maximum of possible configurations, with two cameras, a classical HITACHI and an electronic HADLAND camera. Pressure was measured with a KISTLER piezoelectric gauge, and a few tubes were fitted with strain gauges to check case deformation. These tests allowed the setting of the experimental methods and to consider several types of data processing, keeping in mind a number of ideas for further experimentations.

A technique is developed that permits gas temperature to be determined in highly pressurized combustion chambers by means of specially designed emission gauges. Two different designs are developed and tested, and sample results of temperature measurements are presented.

This paper presents measurement and analysis of impact and penetration effects upon aluminum alloy targets by the caliber 7.62 mm and the caliber 12.7 mm armor piercing bullet. The impact velocity ranged from 250 m/s to 850 m/s, and the impact angles were o° and 30°. A number of aluminum alloy plates were used as the targets, with different materials and thickness. Using the measurement values of ballistic limit, an empirical relation between the initial incident energy of the core of the bullet and the quasistatic piercing energy is obtained.

In studying the behaviour of different projectiles when penetrating a target, in this case gelatin, the development of an integrated flash X-ray, high speed photography and ballistic screen instrumentation system has proven useful for gaining qualitative and quantitative data of the projectile motion before, during and after penetration. The system discussed permits a three dimensional analysis of projectile behaviour inside the target, thereby providing measurements of projectile velocity, yaw and tumble, and enabling the study of the cavity formed by the projectile at various instants during penetration. Methodologies are compared and the uncertainties of the measurement techniques are discussed.

The techniques described in this paper were developed during a study of high velocity liquid/solid impact. This form of impact is relevant to a number of technological problems such as the rain erosion of aircraft, the erosion of steam turbine blades, cavitation damage and the cleaning, cutting or fragmentation of materials using high velocity liquid jets. There are advantages in creating two-dimensional impact geometries since these enable processes occurring inside the impacted liquid mass and the onset of jetting to be studied in detail. Liquid wedge and drop geometries are produced by first casting water/gelatine sheets and then cutting out the required shapes. The impacting solid is a plate fired from a rectangular bore gas gun. The impact is viewed at high magnification using high speed photography at microsecond framing rates with an Imacon camera, and the shocks are visulised using Schlieren photography. The results give information on the shock structures produced, the "water hammer" and "edge" pressures and the critical conditions for jetting. The two-dimensional techniques have also been applied to studies of cavity collapse. In this case, a single cavity, or an array of cavities, is formed in the gelatine layer. An advantage of the technique is that the cavity size and spacing can be accurately controlled. It is well known that when a shock passes over a cavity, a jet is formed in the direction of the shock, by involution of the cavity surface. The impact of this jet onto a solid surface is an important feature of cavitation damage. In our experiments, we have also studied shocks passing over an array of cavities. Cavity collapse is also important as an ignition mechanism with explosives and as a source of "hot-spots" which aid the build-up to, and propagation of, fast reaction in explosives. However, there is still controversy about the relative importance of adiabatic heating of the gas in the cavity and shock heating when the jet impacts the far side of the cavity. We have investigated this problem by photographing the collapse of two-dimensional cavities containing small quantities of primary explosives.

To investigate blockage phenomena and its effects on the loading of a model, in the rect-angular test section of a shock tube, for several blockage ratios, realized by simple models namely closed box-like structures, shadowgraphs were taken from the flow and wave processes, initiated by a non-decaying shock wave. Besides, pressure-time and impulse-time histories registrated at measuring points at the front, top and rear of the model are discussed. Re-sulting from these experiments recommendations can be given about the admissible blockage ratio of a simulator.

During the process of shock wave reflection at plane and curved walls (wedges and cylinders) the front reflection pressure was measured in shock tube runs using pressure transducers and indirectly by evaluating schlieren pictures. Shock incidences were especially chosen in those angular regions where transition from regular reflection to irregular reflection occurs. A basic difference in the trend of the reflection pressure is recognizable at the various shaped obstacles. In the case of the plane wall the pressure increases with increasing angles of incidence until the termination of the phase of regular reflection and decreases during the subsequent phase of irregular reflection. This enhanced front reflection pressure can be higher than that resulting from head-on reflection. At the curved wall, the pressure decreases monotonically throughout the entire phase of regular reflection. The first case, being pseudosteady, is predicted by the von Neumann reflection theories, not so the latter case, which is truly a non-stationary process. However, not only the geometric shape of the body surface influences the reflection phases but also its quality. By a series of shadow-schlieren pictures it is qualitatively shown that the surface roughness delays the development of regular reflection and Mach reflection remarkably and that at plane obstacles as well as at curved ones. So it must be taken into account that the trend of pressure also depends on realistic parameters.

A simple method for obtaining moire' patterns sequence of aluminium plate at high strain rate and large flexivity is presented. At different instant full field flexivity and three dimensional coordinates of all interested points on the plate are obtained by the moire' fringes. Thus displacement and strain components may be deduced. Possible errors are also estimated.

Freshly cleaved {100} faces of potassium chloride crystals were impacted at normal incidence with glass and tungsten carbide spheres of diameter 0.4 and 0.7 mm, respectively, and of velocities of up to 200 m s-1. This caused the plastic flow and a system of fractures at the impact site; one such system is generated by the interactions of screw dislocations on {100}45 planes inclined to the impacted plane at 45°. The extent of these cracks has been shown to be almost the same as that of the boundary of the plastically deformed region. Therefore the movement of this system of cracks directly gives us information about the velocity of the dislocation front, thus providing us with a new method of monitoring the dislocation velocities. High-speed framing microphotography using a Beckman and Whitley model 189 rotating mirror camera working at a rate of - 2 million frames per second has been successfully used to follow the propagation of these cracks. The maximum velocity of the dislocations was found to be 700 m s-1 corresponding to a shear stress of - 8 MPa. The dislocations only propagated during the loading part of the impact, and their velocity was linearly dependent on the shear stress.

The conical convergent shock wave in a cylindrical specimen can bring forth Mach reflection, e.g. the shock wave overcomes the singularity on the axis by performing a triple point between the incoming front, the head wave and the reflected shock wave. The conical convergent shock wave is produced by the explosion of a high explosive which forms the outer coating of the probe. If the detonation is made running parallel to the axis of the probe the head wave - Mach disk -runs in the same direction with the detonation velocity of the outer explosive. In water all the three types of shock waves were detected as well from X-ray flash photographs as by measuring the electrical conductivity of the water under shock load. X-ray flash photographs of PMMA (Polymethylmetacrylat) specimens in contrast to the experiments with water show no triple point configuration. On top of the incoming shock wave cone a shock wave with steady curvature is observed. The centre part of this shock wave is nearly plane and stands perpendicular to the axis of the probe. There is no evidence for a reflected shock front. Optical streak photographs of the PMMA probes were made using several collapsing argon gaps. The results obtained by this method are very similar to those known from literature. A distance of about three times the probe diameter is necessary to get a stable shock configuration. The interpretation of the streak records is made difficult by the great range of intensity within one streak curve. The centre zone shows very little light due to the rapid closure of the gap due to the nearly parallel striking of the shock wave. The results were confirmed by time resolved pressure measurements with composite carbon resistors which have a great piezoresistive coefficient. From these experiments the shock contours were reconstructed with 10 ns time resolution in accordance with the optical and X-ray measurements. The pressure records too show no second pressure jump of a reflected shock front in PMMA.

Optical fiber gauges are commonly used in hydrodynamic experiments to measure the arrival time of shock waves or free surfaces at different locations. The light pulse emitted by the shock loaded gauges is recorded by means of a streak opto-electronical camera (this technique has been described by J.P. GAUTHIER at the 13th International High Speed Photography and Photonics meeting in TOKYO 1978). The aim of the work presented in this paper is to characterize the response of such gauges under shock loading between 10 GPa and 430 GPa. Optical density versus time is analysed in order to know if optical fiber gauges can be used as pressure transducers. There is effectively an important variation of the emitted light versus pressure. Assuming the first optical density peak is related to the maximum air gap pressure of gauges, we study the variation of this peak versus shock pressure. An important dispersion, due to the gauges as well as to the transmission line and the recording camera, can be observed on the results. Nevertheless, it is possible to get an optical density versus pressure relation and also to correlate the optical variation density versus time with the shock propagation inside the gauge. Thus we notice that the rise time varies from 10 ns for 428 GPa to 100 ns for 16 GPa.

The improvement of shaped charge performance depends to a large extent on an improved precision of all components and on the improved homogeneity of the high explosive charge and of the liner 1. The deviation of the individual jet elements from the axis must be less than 0.1 % if these elements are to transform their kinetic energy into an increase in depth at the crater bottom 4.

Fibre optic probes easily can be placed on high explosive charges. Thus the detonation front can be observed while breaking through the surface of the charge. By inser ting those probes in a high explosive charge, this technique can be extended to a diagnostic tool for observation detonation phenomena inside the charge. Arrival time at the tip of the probes can be measured simultaniously to the break-through of the detonation front at the outer surface of the charge when a streak camera is used Each of the light fibres can transduce one light pulse only If there is need for observation in several planes of the high explosive charge the number of probes has to be increased to an amount which however causes severe disturbance of the detonation wave propagation. The detonation wave can be observed in a multi-planar way by means of a new probetechnique avoiding the disadvantage mentioned above. The light fibre probe now has several airgaps placed appropriately along the probe, each of them giving an output of light at the end of the probe when the detonation wave propagates down the probe. Thus by multiplanar diagnostic of the detonation wave inside a high explosive charge it is possible to measure the detonation front with comparatively small disturbances and little effort. Practicable solutions for the set-up of the light fibre probes are discussed and the advantage of this technique is demonstrated on examples.

Warheads in conventional munitions contain three explosive components - a detonator, an augmenting charge known as a booster and a main filling. High-speed photographic experiments have indicated that a significant improvement in the initiating ability of boosters can be obtained if minor modifications are made to their containers, or housings. It is shown that in the situation where an aluminium alloy booster housing is inserted into a recess in the main filling, enhanced radial initiation can be achieved. This is especially welcome where insensitive plastic bonded high explosives are used as main charge fillings.

Shock wave propagation in plexiglas, generated by an oblique detonation is monitored using a shadowgraph framing camera technique. The transparent material loses its transparency upon arrival of the shock wave and then regains it behind the shock front where the pressure drops. The experimental dataare compared to a semianalytic calculation and to a two dimenstional numerical simulation. Good agreement is obtained with both calculations.

Moire analogy is advantageously used to explain aspects of interferometry, in particular, holography and derived techniques. Transposition of concepts from moire methods to fringe pattern processing proved to be of merit and capable of producing development of new features of practical interest. It is proposed the exploration of the concepts of moire reference change, differential moire, integrative moire and moire derivation. Examples are given to present envisaged advantages particularly for application on the study of dynamical events where pulsed illumination with synchronized video recording is proposed for extended range analysis.

Experiment shows /1/ that in an image-converter tube (ICT) the central electrons reach the deflection plane before the peripheral electrons emitted from photocathode edges come there. This results in bending the isochrone lines on the ICT screen which confuses the processing of the image (which would be still more complicated as the sweep rate becomes higher). Such a bending can be remedied by using a concave shape of photocathode. But minor errors in designing the concave shape of photocathode may result in the opposite sign non-isochronism of the ICT, i.e., to isochrone bending in the opposite direction.

The problem of recording fast processes in IR spectrum remains an urgent one up till now. The authors of /1-3/ proposed a principally new device which they called ionization-type semiconductor system (ITSS). It features high sensitivity and rapid response in a broad spectral range. The ITSS is a promising means for recording and measuring the space-time parameters of IR flash sources.

In high magnification systems laser speckle limits the resolution of laser illuminated targets. We have developed a broad bandwidth laser illuminator which substantially reduces the speckle. This laser has been incorporated with intensified camera technology and suitable lensing to provide a microscope system with good resolution over a 75-mm diameter image. The microscope currently has a 2-ns exposure onto film with magnification adjustable between 5 X and 50 X .*

A new type of picosecond streak image tube has been developed. The electron optical system of this tube consists of a curved photocathode, a curved mesh, a spherecal focous electrode and an anode cone. The deflector with ultroshort parallel plate construction is adopted. S20 photocathode is made on the Cu-Ni mesh substrate having surface resistivity Qf 300*Cland with transmissivity of 30% for white light. The measured photosensitivity is up to,140,11A/1m. The curved mesh of 50 lines/mm is fabricated by using special technique. P11 phosphor screen is deposited on a fibre faceplate having working area of P50 mm. The obtained static optimum spatial resolution is 40 1μm. The deflection sensitivity of the deflector is 4.2 cm/kv. The magnification of the tube is 2 and it is spatial distortion free. The dynamic measurements show that at present the temporal resolution reaches 2 ps and the useful dynamic range is greater than 256 for the temporal resolution of 10 ps. The slit streak image curvature phenomena has been eliminated.

This paper describes the principles and methods of dynamic test. Some results of the dynamic performances of a new picosecond streak tube will be presented. When the input pho-tocathode is illuminated by laser pulses ( 0.53,x,(, m, 5ps ) from a passively modelocked Nd;glass laser, the temporal resolution of the streak tube is better than 3ps, and the dynamic range of the streak tube is about 32.

In investigating the rapidly flowing processes using high-speed and super-speed eine cameras, a problem arises of concurrent recording on one photographic medium of the pro cess dynamics in the shape of a eine film fraglent and of the oscillograms which characterise the changes in certain process parameters whereof the frequency range is from several fractions of one hertz to hundreds of kilohertz.

Measuring techniques for characterizing electronic streak tubes including dynamic range and resolution are described. A particular tube with a CCD imaging device within the envelope is described.

Streak cameras are now currently used as a basic tool in the diagnostics instrumentation or laser-plasma interaction studies. We have recently shown at the 15th congress which was held in San Diego the properties of the the P 650 soft X-ray cameral having both optimized spatial and temporal resolutions (16-20 1p.mm-1, 3 to 5 ps). During 1983-1984 we have continued our efforts to improve and extend the performances of these diagnostics. New developments include : - the used of a cylindrical mirror with the P 650 X-ray streak camera, in order to obtain a streaked soft X-ray imaging device which will be describe with some more details in the following paper. - study of a new X-ray streak camera using a modified image converter tube P 750 X with a best lamellar optics. - enhancement of photocathode sensitivity by testing new materials. - development of new photocathode-tube window arrangement in order to record both laser and X-ray signals. - another important improvement in the same field is the C.C.D. readout which is presented in other papers.

Understanding the laser-matter interaction experiments require a dynamic recording of the phenomena as well as a good knowledge of the laser pulse occuring during the irradiation of the target ; those measurements are made with streak cameras the increasing number of which leads to processing problems when the results are recorded on films. Furthermore, since physicists wish to have those temporal information immediatly, we unfolded automatic image readout devices fitted specially to streak cameras. The first one used an ISOCON tube operating with a slow sweep (0.5 s frame)1-2. The sensitivity of the tube was very good but its dynamic range was too limited when seeing pulsed images at low light level. So we developped two electronic readout chains with solid state devices which behave better for that kind of light. The first one was designed to get the most information along the temporal axis of the camera sweep (1024 points) in one or two spatial channels ; this device operates a linear 1024 photodiodes array the signal of which is digitized on 12 bits. We have obtained a temporal resolution better than 15 ps and a dynamic range over 500 ; this system is mainly useful to study laser pulses (rise time, temporal profile...)3. For applications requiring two dimension images, we studied and realized a device operating a CCD array and a fiber optics reducer (40-18 mm) adapting the image of the streak camera screen to the dimensions of the input fiber optics of the CCD. A comparison has been made on different CCD cameras on a test setup which simulates the experimental conditions, in order to choose the CCD which would fit the best for that purpose ; we present these results here, as well as those of the associated readout chains.

A CCD-type readout system combined with 3-stage electrostatically-focused image intensifier is described. The readout system is based on K580UK80-type microprocessor, 64K of 16 bit words frame memory and 10/12 bit A/D converter.

This paper deals with a digital streak camera readout device. The device consists in a low light level television camera made of a solid state C.C.D. array coupled to an image intensifier associated to a video-digitizer coupled to a micro-computer system. The streak camera images are picked-up as a video signal, digitized and stored. This system allows the fast recording and the automatic processing of the data provided by the streak tube. Starting from the output screen of the streak camera, the constitutive elements are : - A fiber optic taper (A.O. Scientific Instruments) set in contact with the fiber optic output window of the streak tube achieves the image demagnification ; - A double proximity focused image intensifier (RTC - XX1410 SP) achieves the bright-ness amplification without any distortion ; - A second fiber optic taper achieves the dimensional matching between intensifier output and C.C.D. sensitive area ;

A camera system has been assembled employing an image intensifier and a CCD array TV camera. Exposures from 100ns to 20us are obtained by gating the intensifier, and the video signal is digitally stored and displayed on a monitor. Randomly timed events may be captured as there is no synchronism between the gating and the TV scanning.

An electronic camera capable of delivering one image frame of video information in 1 ms has been designed around the TH 7882 solid-state area array image sensor (a single-field, frame transfer organization charge-coupled device. The prototype uses an image format set to 100 x 100 pixels and a 100 /Is integration time. The stroboscopic type of illumination, called for by the image sensor's single-field configuration, is compatible with the camera's applications for analyzing rapidly occurring phenomena. In this operating mode, the TH 7882 has a CTF (contrast transfer function) of 80 % at the Nyquist frequency and a sensitivity of 9.6

The modern high-speed electron-optical streak cameras employ image-converter tubes (ICT) with wide-band image display in form of a plane-parallel line /1-3/ in which the travelling-wave electric field is used for deflection of electron beam. In this report we investigate in the relativistic approximation the effect of magnetic field of the travelling wave on electron motion. We have determined the deflector system sensitivity and sweep distortions with allowance for the travelling wave magnetic field for differently applied deflection voltage. The edge effects are ignored.

The principle of "light-in-flight recording by holography" has been developed for studies of ultrafast phenomena. It produces a three-dimensional continuous motion picture with a resolution in the picosecond range and a time span of e.g. some nanoseconds. As the main advantage of this technique is in the study of objects with a velocity close to that of light, compensation for relativistic effects have to be made. This is true when the studied object is e.g. a wavefront of light itself, but it is also true for other high-velocity objects. It has been found that the "holo-diagram" (originally designed for holographic interferometry) is a practical device also for the evalution of relativistic effects. The relativistic rotation of wavefronts and aberration of lightrays, the Lorentz contraction and the transvers Dopplershift all can be explained by the use of this diagram. Thus the conceptual step becomes small from ordinary optics to the relativistic optics needed for the study of ultrahighspeed phenomena. Good agreement has been found between accepted relativistic equations and results from a slightly modified holo-diagram.

The term of "detonics" is generally applied to all the physical phenomena that occur in connection with the detonation of conventional high explosives. The reaction and/or detonation velocities are of the order of 1000 m/s to 9000 m/s, at pressures of several hundreds of kbar. In order to obtain records or pictures of the detonative event with little motion blur and high time resolution, exposure times of less than 10 ns and frame rates higher than 106/s are desirable. About one or two decades ago, new discoveries and improved knowledge in the field of detonics were heavily dependent upon developments in high speed photography. Nowadays, however, special cameras are available for diagnosing detonation phenomena, and to events caused by high pressure and shock waves, and their commercial development has produced reliable and easy-to-use instruments. Thus, the scientist working on detonics problems can now concentrate on developing measuring techniques instead of developing measuring instrumentation. A short overview will be given of the different most used and important optical equipments and cameras in the field of detonics diagnostics. Further some sophisticated, and even "tricky", configurations for optical high-speed measurements in the field of detonics will be shown in various examples which have recently been in refinded detonation diagnostics.

For many years opticists have tried to achieve 3D-cinematography with the help of holography. The term "cineholography" has been introduced in 1965. At that time the method consisted in superimposing on the same photographic plate various holograms recorded at different times. The image separation was achieved by rotating either the plate in its plane or the reference beam during both the recording and reconstructing processes. The number of views was limited by the principle itself, but a high repetition rate (100 kHz) has been obtained with a pulsed ruby laser. Then other experiments have been conducted by different authors, mainly in the Soviet Unions and in the United States of America, by using the principle of classical cinematography in which the image separation is obtained by translating the film. With this method we have recorded the first French holographic movies on 35 mm Agfa films with the help of a pulsed YAG laser built in our laboratory. This frequency doubled laser (X = 0.532 pm) delivers pulses of 20 ns with an energy of 30 HO at a repetition rate of 24 Hz. The experimental arrangements are described and some images of diffuse moving objects are presented. The volume of the recording scene is greater than one cubic meter. The coherence length of the laser is higher than one meter and remains steady during the recording process. Results are discussed and an outlook on the future is given with special respect to high repetition rate techniques.

A high speed holographic movie camera system has been developed in our laboratories at the Third Physical Institute of the University of Gdttingen. As a light source for holography a high power multiply cavity-dumped argonion laser is used to record very long hologram series with framing rates up to 300 kHz. For separating successively recorded holograms two spatial multiplexing techniques are applied simultaneously: rotating of the holographic plate or film and acousto-optic beam deflection. With the combination of these two techniques we achieve up to 4000 single holograms in one series.

A new approach in modeling and analysing propagation of light is presented. This approach is based on wavefront analysis and not on rays analysis. Traditional mathematical modeling can be reexamined in this new frame work. It turns out that it is well adapted to high speed problem analysis; particularly light in flight recording by holographic interferometry. This formalism can also be used conveniently for design of non conventional optical elements and systems: highly aspherical refractive or reflexive as well as holographic and hybrids.

A system allowing the recording of holograms at a long distance is described. Its luminosity is such that a 1 m3 scene could be recorded at 10 meters with the help of a low-energy pulse laser. A resolution of about 100 microns was observed in the object. An application to ballistics is presented.

The use of optical fibers in holographic set-up allows the recording of holograms of objects which are not directly accessible to conventional optical technics. The problems for recording holograms through optical fibers are of the same nature as for producing optical fiber sensors. But the optimal use of the two technics corresponds to opposite conditions: sensors require a great sensitivity to phase variations while the hologram quality is disturbed by a very low variation. There are two possibilities of freezing the phases during the recording of the hologram: 1) in c.w.-laser holography it is necessary to avoid any variation of physical parame-ters (temperature, pressure, electromagnetic field, vibrations, twisting, mechanical strain, air turbulence, etc...) influencing the phase or the polarisation of the light transmitted by the fiber; 2) use of a pulse laser with a very short time of emission. Unfortunately in this case non-linearity phenomena may occur in the fiber: reduction of the coherence length, light frequency variation. The focusing of the light beam may also ionize the air and damage the input face of the fiber. At last, the very high powers may destroy the fiber. In spite of those drawbacks holograms of small objects have been recorded through multi-mode and monomode fibers with the help of ruby and YAG lasers. A) Ruby laser: with a ruby laser giving a pulse of 20 mJ in 20 ns at 0.6943 pm the following set-ups have been tested: a) The object beam passes through a 1 mm-multimode step-index fiber; no fiber for the reference beam. b) Same arrangement for the object beam, the reference passing through a monomode fiber. The beamsplitter is located between the oscillator and the amplifier. B) YAG-laser: with a YAG-laser giving a pulse of 20 mJ in 20 ns at 0.53 pm the following set-ups have been tested: a) A 1 mm-multimode step-index fiber in the object beam and a 133 pm-multimode step-index fiber in the reference. b) Same arrangement for the object beam and a monomode fiber in the reference beam. In each case single- and double-exposure holograms have been recorded. Some typical results are presented and discussed.

The fleight target painted with moire pattern to be tracked by a high speed camera through a transparent grid give us a very high precision method for the measurement of target displacement. The film unflatness, air fluctuation and the vibration of the high speed camera are also effected on the shape change of the moire. The small amount change of the pitch, yaw, rotation and the direction of the fleight target can be easily measured by this method.

In contrast to the shape of the detonation wave in the normal shaped charge, a plane wave generator is used to transform a detonation wave from concentric circles to a flat wave. The construction of a plane wave generator consists of the combination of two high explosives of different V.O.D.'s, which when detonated in a certain manner will cause the detonation wave to reach all points at the bottom of the charge at the same time. In order to test the performance of the plane wave generator it is therefore necessary to monitor the final shape of the detonation wave. This can be done by the use of multi streak photography at very fast writing speeds, through an open iris. The streak writing speeds used in these experiments were 11Ons/mm and 56,8ns/mm. The ideal air gap between the column of explosives and plexiglass, which is used to mask the detonation, has also been researched at length and will be discussed in this paper.

Shock tube experiments using differential interferometry have been carried out to investigate compressible subsonic flows around and behind flat plates and cylinders. Pictures taken with an IMACON image converter camera with framing speeds of up to 2.10⁶ frames per second and streak records show that the trailing edge vortex separation induces pressure waves moving upstream. Each vortex separation is related to the formation of a pressure wave. The flow around the bodies is therefore influenced by the shedding frequency and is possibly coupled with the wake. The interaction of the wake with the flow around the bodies is a phenomenon of fundamental importance which has been neglected in gas dynamics up to now. After a symmetrical onset, the process of vortex separation and pressure wave generation develops into an asymmetrical wake with alternating separation. The experiments have shown that the dynamics of vortices close to the trailing edge is practically inviscid and incompressible. In contrast to this, the pressure waves moving upstream must be treated as a problem of purely compressible nature.

At the ONERA hydrodynamic visualization laboratory, high-speed photography and cinematography are used for analysing flow-phenomena around fixed or mobile models in the test section of three vertical water tunnels, operating by gravity draining. These studies in water are based on the hydraulic analogy of aerodynamic incompressible flows. Flow visualization is archieved by liquid tracers (dye emissions) or gaseous tracers (fine air bubbles in suspension in water). In many cases, the pictures at normal speed or long exposure time are insufficient, for they do not permit to distinguish all the details of the phenomena, due to an averaging or motion effect. Furthermore they must be completed with high speed pictures. This is illustrated by a few visua-lization examples recently obtained on following themes - two dimensional flow around a fixed cylinder, first at the start of the flow (symmetrical vortex), then in steady regime (periodic vortex street) ; - laminar-turbulent transition in a boundary layer along a cylindrical body at zero angle of attack ; - flow separation around a sphere and wake in steady regime at small and high Reynolds numbers; - flow separation around a profile, first with fixed incidence, then with harmonic oscillations in pitch ; - core structure of a longitudinal vortex issued from a wing first organized, then disintegrated under the effect of a lengthwise pressure gradient (vortex breakdown) ; - mixing zone around a turbulent axisymmetric jet, characterized by the formation of large vortex struc-tures ; - hovering tests of an helicopter rotor, first at the start of the rotation, then in established regime, finally in cruise flight ; - case of a complete helicopter model in cruise-flight, with air-intake simulation, gas exhaust and tail rotor ; - flow around a complete delta-wing aircraft model at mean or high angle of attack, first in steady regime, then with harmonic oscillations in yaw or pitch. These results illustrate the contribution of the high speed view recording and reveal the wide variety Nlpf their applications in both fundamental fluid mechanics and aerospace applied research.

By making suitable design changes in the Schlieren apparatus, the technique can be made to yield very attractive photographs which open up a wide field of new applications in commercial art, advertising, special effects, etc.

Irradiation of dense matter by pulsed laser beams with the ultimate goal of achieving thermonuclear fusion conditions provides a challenge to high-speed diagnostics. The concentrated deposition of energy by laser pulses of typically nanosecond duration induces laser-plasma interaction phenomena and hydrodynamic motion which have to be diagnosed on a picosecond temporal and micrometre spatial scale. The high-speed diagnostics developed for this purpose, such as shadowgraphy and interferometry by probe laser beams in the visible, x-ray backlighting, time-resolved spectroscopy in the soft x-ray region, and x-ray framing, will be described and illustrated by experimental results.

This paper describes the application of computer image enhancement and analysis to the interpretation of noisy low contrast flash X-ray photographs of weakly detonating mining safety explosives. It shows how the shapes and positions of detonation reaction zones can be highlighted even though the original X-ray photographs may be somewhat ill-defined.

Objects can be controlled by different methods, such as radiographic records for example. In this case, the image processing techniques allow to really measure the defects, and not only estimate them. This increasing precision of the results has been obtained by defining a set of correcting parameters in order to take into account experimental constraints, characteristic of the X-Ray generator and linked reception system, and of the use of a self-calibrating operating process. The measurement of the defect(s) can be seen as acted in three parts : a) Preliminary measurements so as to initiate the correcting parameters, activation of the corrections if necessary and if possible, measurements of the calibrating areas. b) Extracting the defective zone from the total image. The classical optimal separator based on bimodal gaussian distribution does not work well, and leads to important errors : the implicit hypothesis of stationarity cannot be held for the defective zone, but the background, or the whole zone, can be regarded as stationary, or varying so slowly to allow correcting it in order to obtain stationarity without modifying the boundary of the defective zone. Our aim is to find out the only pixels belonging to the background, for they constitute a set S, the stati,,tical parameters of which can be known. Pixels, the values of which are lower than a tnreshold Vo, belong to S. Pixels, the values of which are greater than Vo can belong to S, on the condition that a d-distance is lower than a value Vk determined by the number k of pixels already detected as belonging to S in the neighbourhood of the current pixel. The d-distance is based on the first and second statical moments of the S-population. The computation can be iterated and convergence is rapidly obtained. The defective pixels are immediatly obtained as the complementary set S. c) This set S leads to a binary space acting as a mask to be applied on the original image and identifying the pixels belonging to the defective domain. We can, now, measure the characteristics of the defects : maximum height, width, thickness at every location, volume, mass (or lack of mass) ... As an illustration to our algorithm, we give examples obtained from test-samples defined so as to measure the precision one can obtain, and verify the convergence, and results obtained from real defects.

Image degradation caused by relative motion between the object and the imaging system (like a camera with its platform) is detrimental to precision measurements with the image. Principal modes of relative motion are identified. The discussion, however, concentrates on the systematic motions, translatory and rotatory. Various known analogical approaches of compensating for the image motion are cited. An analytical-computational approach is presented. This considers the relationship of 3-dimensional transformation between the image and the object, known as the collinearity condition. The standard forms of collinearity equations are presented. Augmentation of these equations for both translatory and rotatory motions are expounded. With the increasing use of high speed computers (as well as analytical plotters in the realm of photogrammetry), this approach appears to be more cost-effective and finally seems to yield precision better than other approaches that con-centrate on analogical corrections to the image.

The last technological evolutions of imaging techniques and the appearance of new kinds of TV cameras lead us to think of new image acquisition and processing devices. Several kinds of cameras are operated in our laboratories with different video output signals (200 ms frame, CCIR and RS 170 standards, 40 ms image mode, sampled CCD output), so no existing acquisition device was easy to transfer between different experiments. Our new image memory is then designed to be linked to any TV camera. It can be connected to our computers too, through its IEEE 488 interface, although it may work on its own. Its three main purposes are - signal acquisition, managed by a 8086 microprocessor, which can be externally triggered ; - memorizing of four 512 x 512 x 8 bits images on which some pre-programmed processes can be performed ; - bi-directionnal communication of data with a computer or an image processing device, through three kinds of links. We tested that image memory on a laboratory setup, on which we stored images delivered by a CCD camera. We present here a comparison between those results and those obtained on the same setup with a Pericolor 1000, and we emphasize on the automatic conversion threshold adaptation, and on the possibilities of adjusting the conversion scale, allowed by this device.

An image processing software was developed for analyzing two dimensional streak images. The software is a part of a picosecond electron-optical measuring system. The software was adopted for general purpose microprocessor based image processing unit Pericolor-1000 produced by NUMELEC company. The main advantage of using Pericolor-1000 is the possibility of increasing software for preparing a special set of streak image processing functions. The system provides time and intensity scales mapping, image distortion and sweep rate nonlinearity correction, computer estimation of the streak image parameters, fast fourier transform, signal restoration and some other functions.

The more and more intensive use of Fabry-Perot velocimeters to study impact phenomena is leading to more and more films to be processed. Thus, engineers having to deal with this increasing plenty were asking for automatic tools in order to analyse optical records. A software package named ARIDEL has been written so as to fulfil this requirement. When looking at the DLI films, one can notice several points : a) Traces are very bright and large at the beginning of the film, and become less and less bright and therefore less and less large when time is going on. b) Traces vanish, most of the time on one side of the fringes and sometimes on both sides ; they may appear again later on. c) There are artifacts from place to place, leading to false detections. d) Optical density of the background may not be constant, especially at the beginning of the film. e) Velocity variations can make a new fringe appear or a previous one disappear. f) At the very beginning of the dynamic part of the record, it is impossible to know the exact jump of fringes without any outcoming information. Analysing computational errors due to discrete data (round-off noise), one can find out a sampling criterion allowing to get the maximum information from the film at the lowest computer memory allocation. After the film has been digitized and the image stored in the computer memory, ARIDEL software computes a new but binary image using a self adaptative threshold. The ring widths on this new image are too large and have to be skeletonized, assuming that for each side of every fringes the luminous flux is symmetrical. The static part of the record (reference rings) is detected, the axis of symmetry computed and the diameter of the reference rings measured. At this point, the program can find out the origin of the dynamic part of the record or allow the user to fix it by himself. Estimation of the initial velocity value has to be given. Most of the time, the program is working with the two innermost rings. From each line of the image, ARIDEL removes all the points that do not satisfy two sets of conditions : the first one in order to ensure coherence between them, thanks to analytical relations connecting the fringes diameters, the second one using continuity conditions with the previous lines. So, it is possible to take into account the appearance of a new innermost fringe or its disappearance. By this way, the computation al error of measured velocity is being kept at its lowest possible value. Today, ARIDEL prints out velocity vs time, visualize the results on graphic screens and on plotters. Routines computing other physical quantities are to be added to the main program, depending on the users' requests.

A technique is presented for automatic tracking of the three-dimensional motion of tracer particles in an incompressible fluid flow using high-speed photography and computerized processing. Particles in the fluid are illuminated so that they appear bright over a dark background. Two perpendicular views of the test section are recorded on each frame using a HYCAM camera and a mirror arrangement. The images on film are video digitized and suitably enhanced to allow automatic identification of the tracers even under less than optimal conditions. Corresponding pairs in the two views are matched to yield the spatial co-ordinates of the tracers, which are then tracked by an algorithm that uses the most recently computed velocity vector for a particle to predict its position in the following frame. The accuracy of spatial location obtained from orthogonal views compares favourably with stereoscopic techniques used by other workers, and tracking in three dimensions minimizes problems that may arise when following particles independently in each view of a stereo-scopic image.

Light-sensitive liquid films are used to produce cross-modulation between light beams in an all-optical device. In the first setup presented in this paper a laser beam (HeNe, CK, 1 mW) is switched-on (off) when another laser beam (HeNe, ON, 10 mW) is switched-off (on). The diameter of the interaction region in the liquid film is about lmm. In another setup, large-scale deformations of the liquid film (diameter about Smm.) are induced by heating with a white-light source. The deformed region behaves as a mirror whose radius depends on the intensity of the heating beam.

Distinction has to be made between image processing and vision Image processing finds its roots in the strong tradition of linear signal processing and promotes geometrical transform techniques, such as fi I tering , compression, and restoration. Its purpose is to transform an image for a human observer to easily extract from that image information significant for him. For example edges after a gradient operator, or a specific direction after a directional filtering operation. Image processing consists in fact in a set of local or global space-time transforms. The interpretation of the final image is done by the human observer. The purpose of vision is to extract the semantic content of the image. The machine can then understand that content, and run a process of decision, which turns into an action. Thus, intel I i gent vision depends on - Image processing - Pattern recognition - Artificial intel I igence

According to the filtering theory, we derive a functional expression of the optical field distribution, which is the filtering image of the intensity spectrum of the double-exposure speckle pattern passed through the filtering aperture in Fourier transform plane. Further, an affect on formed fringes to be yield by filtering due to the shape of the filtering raperture is discussed. An effect due to filtering is not appeared on the fringe pattern, it only affects the envelope and contrast distribution of the frienges, this influnence seems weak, however, it offeres the theory foundation, if we try to obtain an optimum quality of the fringe pattern.

A previous velocity interferometer used in high-explosive experiments, giving high resolution data by means of a rotating mirror streak camera, has been modified in order to transmit light in two optical fibers and to record on an electronic streak camera ; in both cases, Doppler shift is analysed by a Fabry-Perot interferometer. Experimental arrangement is described in two channels : The emitting channel is composed of : - An argon ion laser and a high speed mechanical shutter which allows light to pass only during the shot duration : so the experimental set up with explosive will not get hot and the camera noise will be reduced. - An optical system focusing the laser beam into the fiber. - An optical fiber, linking together the laboratory and the firing table, ended by a pick-up head included in the experimental set up. The receiving channel is composed of : - a second optical fiber coming from the above head, - a Fabry-Perot interferometer included in an optical system, - a streak camera with its slit and a lens. The optical system made of spherical and cylindrical lenses performs two functions : (i) it displays the fringe pattern onto the entrance slit of the camera, two or three fringes, (ii) it realizes a good light transfer between the geometries of the circular light source, the fiber, and the rectangular slit ; this is an anamorphosis. Light transmission has been separated into two different fibers in order to avoid, on the time-resolved interferogram, static fringes induced by the laser light reflected by dioptric surfaces. Experiments carried out with this new arrangement have given good results in terms of accuracy, homogeneity and duration of records. Utilization and adjustments are easier than in the previous case. Thanks to its small size, the pick-up head can be integrated in experiments of detonics in any environment, even at some distance from the laboratory.

Several laser fusion applications require to transmit laser pulses of different wave-lengths (X. = 1,06 µm - 0,53 µm - 0,35 wri) through optical fibers onto a streak camera. Laser pulses of 10 ps duration have been propagated in a short length of optical fiber (12 meters). We describe in this communication the experimental set up of a high bandwidth liaison between a high power laser (104 to 109 W cm-2) and a picosecond streak camera. The utilisation of step index optical fiber (600 µm to 1500 μm diameter) is compatible with transport of high power density an much more convenient than the use of gradient-index or single mode fibers. Measurements have been done at 1,06 µm and 0,53 µm wavelengths ; we have demonstrated that 100 ps laser pulses can be transmitted without any significant temporal dispersion through these step index fibers.

Shift registers and random access memories (RAM) may be used as buffer memories for allowing shorter time increments in conventional multichannel analysers. We describe circuits containing 64-bit shift registers or 1024-bit RAM's with minimum channel advance time intervals of 20 ns and 50 ns, respectively. These circuits are applied for fluorescence decay and energy transfer studies in laser crystals, but they are usable for LIDAR (light detecting and ranging), too.

A gating grid was used for nanosecond gating of image tubes in the early sixties (1,2). With the improvement of switching techniques and impedance matched image tube design, 300 picosecond framing time was reported using biplanar diode image tubes (3). The present paper reports the extension of these ideas to 100 picosecond framing in a magnetically focussed image tube. A strip line photocathode-gating grid structure with discrete areas of photo-sensitive properties equally spaced along the strip line is used to to switch successive frames. The framing time is defined by the duration of the switching pulse, the length of the photosensitive region, and the gating voltage which defines the transit time spread of the photo-electrons. The operation of a prototype tube using an ultra violet sensitive photocathode and using an UV light source is described. Spatial resolution of this magnetically focusssed tube is better than 20 line pairs per mm.

The study of the process of light propagation in random inhomogenious media leads, as a rule, to the solution of two principal problems. The first one is to find out the optical properties of the medium from the results of detection of a scattered field of a certain structure. The second problem is to find out the structure of light field, that propagates through the scattering medium, with its optical characteristics being known. These problems are solved, as a rule, by spectrometric methods; and the solution appears either in the form of determining the integral characteristics, or eliciting the properties of a sought value in a small volume /1/.

In order to get a clearer view of a wire explosion through use of oscillographic recording and Schlieren high-speed photography, experiments were performed with medium energy input which show all the stages of the wire explosion. Oscillographic recordings display four of the distinct stages of the wire explosion including the heating stage, explosion, current dwell, and the reignition stage. The Schlieren-framing photographs show the cylindrical symmetry of the contact surface, primary shockwave, arc channel, and the secondary shockwave. The Schlieren-streak photography yielded the trajectory curves of the contact surface and of the shockwaves from the beginning of the explosion stage to the end of the restrike stage. Photographs with a strong illumination reveal the clear view of the unexploded core residue. The experimental results obtained during the explosion stage follow the predictions of the "metal explosion model".1 By taking oscillographic traces of current and voltage during wire explosions with variation of the initial discharge voltage, a converging value of the explosion time at the high voltage limit is obtained. The steady state supercritical ex-plosion speed is then determined by dividing the wire radius by the limiting explosion time, and found to be 350 m/sec for aluminum. There exist two distinct exponential relations between the dwell time and the dwell voltage. The more rapid exponential decay of dwell time with dwell voltage in the lower voltage region is attributed to the interruption of the development of the arc discharge chan-nal along the central line of the cloud of product particles by the debris of unexploded core residue. For explosions free from unexploded core residue, empirical relations between the speed of shocks and the specific energy of product above the heat of explosion are found. The primary shock speed is found to be proportional to the square-root of the specific energy of the explosion product. The speed of the secondary shock which is generated by the arc discharge through the explosion product is found to be less sensitive and proportional to the cube-root of the specific energy of the arc discharge product.

Various experiments are underway at the Materials Research Laboratories, Australian Department of Defence, to develop a theory for the behaviour and propulsion action of plasmas in rail guns. Optical recording and imaging devices, with their low vulnerability to the effects of magnetic and electric fields present in the vicinity of electromagnetic launchers, have proven useful as diagnostic tools. This paper describes photoinstrumentation systems developed to provide visual qualitative assessment of the behaviour of plasma travelling along the bore of railgun launchers. In addition, a quantitative system is incorporated providing continuous data (on a microsecond time scale) of (a) Length of plasma during flight along the launcher bore. (b) Velocity of plasma. (c) Distribution of plasma with respect to time after creation. (d) Plasma intensity profile as it travels along the launcher bore. The evolution of the techniques used is discussed. Two systems were employed. The first utilized a modified high speed streak camera to record the light emitted from the plasma, through specially prepared fibre optic cables. The fibre faces external to the bore were then imaged onto moving film. The technique involved the insertion of fibres through the launcher body to enable the plasma to be viewed at discrete positions as it travelled along the launcher bore. Camera configuration, fibre optic preparation and experimental results are outlined. The second system utilized high speed streak and framing photography in conjunction with accurate sensitometric control procedures on the recording film. The two cameras recorded the plasma travelling along the bore of a specially designed transparent launcher. The streak camera, fitted with a precise slit size, recorded a streak image of the upper brightness range of the plasma as it travelled along the launcher's bore. The framing camera recorded an overall view of the launcher and the plasma path, to the maximum possible, governed by the film's ability to reproduce the plasma's brightness range. The instrumentation configuration, calibration, and film measurement using microdensitometer scanning techniques to evaluate inbore plasma behaviour, are also presented.

Several phenomena limiting the performance of a pulsed supersonically cooled CO-laser with 10 4s discharge excitation are investigated. Among these are density perturbations, which result from nonuniform heat release within the discharge volume due to the extreme high power density in the cathode fall. They are studied by high speed shadowgraphy. In addition, discharge instabilities, which also interact with the flow and lead to glow to arc transition are observed by high speed image converter photography. It is found that for an excitation pulse as short as 10 4m the density perturbations play a minor role, whereas the discharge instabilities limit the laser performance.

Photographic sequences of bubbles induced in water by a Nd:YAG laser and of the flow generated in front of a solid target by a CO₂ laser are presented. These were carried out with the aid of an electronic image converter camera at framing speeds of 2,106 F 2.10⁷ fr/s and a streak speed of 10 μs/cm.

With the help of a diagnostic system designed for "DELFIN-1" installation we have measured the velocities 250 km/s. The agreement between the attained compression velocities and the obtained experimental scaling is shown. The compression velocity of shell targets is one of the most impotant parameters determining the feasibility to reach thermonuclear temperatures of the compresse:. gas. The compression velocity of shell targets at "DELFIN-1" installation has been measured by three independent methods. Two of the noted methods are based on studying the spectral and spa-tial characteristics of 2Gc..)c. harmonic emission ( is the laser frequency) generated2fn plasma. As 2ed.-, harmonic occurs in the vicinity of the critical density region (nc=10²¹ cm-3)the measuring of Doppler spectral shift of the harmonic line with respect to nominal value λi/2 c5299 Å allows the determination of the n, region motion uc. On the other hand, the registration of the harmonic luminescence spatihl localization a/lows one to directly observe the n, region and to define its motion.

Dynamical behaviour of picosecond soft x-rays generated from laser-produced plasmas were studied both theoretically and experimentally. A grazing incident suctrometer with a flat field focal plane is developed and time resolved spectra near 100 A are obtained by using a newly designed XUV streak camera with 10 ps time resolutions. Temporal changes of intensities of soft x-rays from laser produced plasmas is investigated for various targets. Decay times are about 100 ps and decrease as targets are changed from low Z to high Z materials. Computationally, temporal changes of x-ray intensities is predicted in the framework of the simplified model. A qualitative agreement between experimental and computational results is obtained.

Paper reports a successful application of a holographic interferometry to a flow visualization study of a shock transition over wedges in a shock tube. Experiments were conducted on a 60 mm x 150 mm shock tube of the Institute of High Speed Mechanics, Tohoku University for Mach numbers from 1.1 to 3.0 in air. Using a pulsed laser double exposure holographic interferometry the shock transition over concave or convex walls and multiple straight wedges was quantitatively observed. Consequently the critical transition angles over these wedges were determined and in addition to this a hysterisis effect in the truly non-stationary transition phenomena was clarified. Advantages of applying the holographic interferometry to the high speed gasdynamics were demonstrated.

Holographic interferometry, with a Q-switched Ruby laser, has been used for laboratory study of dynamic problems in geophysics and detonics. The holograms provided not only the images of transient phenomena not visible in ordinary high speed photography, but also the data needed to analyze wave speeds and particle motions in solids, or density and pressure in gasses. Explosively driven cracks in rock plates have been studied dynamically by a modification to the usual timing sequence, in which both laser pulses occur after the explosive loading, and within less than a microsecond of each other. Changes in surface displacements, due to the stress wave motion in the model between exposures, cause interferometric fringes which are discontinuous along crack boundaries in the material. This method has successfully located cracks which would not be visible by normal optical methods. Surface and near-surface disturbances propagating in the earth from blasting sites were studied. Three-dimensional models were explosively loaded at points on the model surfaces. Holograms made at various delay times after loading showed the model's surface motions due to waves propagating from the source. Interactions of waves from two neighboring sources and wave interactions with steps and trenches in the model surface have been analyzed. Rayleigh waves crossing trenches were shown to undergo significant reductions in total particle displacement, particle velocity, and frequency content. It was also shown that as the diltational wave crosses steps or trenches, a substantial new Rayleigh wave is generated and propagates ahead of the source Rayleigh wave. Air shock due to explosive detonations has also been studied. Transmission holograms were made of the shock wave at the open end of a fine tube internally lined with an explosive coating.

A method for measuring the shape of a liquid surface ("hump") perturbed under effect of ultrasonic radiation pressure is proposed. The method is based on using the "hump" as a "liquid lens" with "applied" interference structure formed by the incident sound waves. The "hump" shape is calculated from the laser shadowgrams.

It is a well known fact, that a phase difference of mutually coherent wave fields sets the type of functional dependence of the formed interference pattern on space coordinates /1/. It seems to be useful to develop a method of interference pattern formation, depending not on the phase difference, but on the sum of the phases of interfering wave fields. This method could be used as a background for further development of interference techniques. In this paper we give a solution of this problem, based on addition to two interfering wave fields of a third one. The latter is used as a reference field during pattern recording in a nonlinear light sensitive medium and during image reconstruction in one of higher diffraction orders that followed.

The Photodeposition Effect (PDE) of Selenium hydrosols was investigated for transient photoactivation properties. It was found that the Volume Photoprecipitation (VP) has a faster response time as compared to Surface Photodeposition (SP). SP can be used for permanent recording of plane images. The faster VP process may be used in large volume display applications in near real time.

It is well known that different modes of electromagnetic waves in a waveguide have different attenuation for a waveguide of fixed size. For example, a TE8.1(Transverse Electric in circular guide mode has decreasing attentuation with increasing frequency. This property is shared by all TE0.n modes, and because of this , property some waves have received a good deal of attention for possible long distance propagation of en9r4ykl) and for low ohmic losses in rsonators of microwave and millimeter wave power sources such as gyrotrons.2) One of the major problems in the use of TEOn in circular waveguides arises because it is not the mode of lowest cutoff frequency, and therefore must always be used in a guide capable of propagating a number of modes (known as oversize guides). For the TEo.1 mode in a circular waveguide there are at least four other modes propagating if the TE mode is above cutoff T1,10.1, TE21). There are few ways to identify the mode of propagation 0.1 in an oversize waveguide, and even to measure the fractional power in each of them.

During shock waves experiments, integrety of the expanding free surface is required. Moreover, various mechanisms such as surface geometrical defects, metallurgical inhomogeneities, oxide, melting temperature, shock pressure, contribute to the deterioration of the so called free surface. For measurement of localized or distributed ejected mass due to those parameters, we use differents technics The first with ultra-light speed framing camera Barr and Stroud CP5 for optical recording of phenomenous.

This flash unit has been designed to illuminate short duration events to photograph with high speed cameras in both streak and framing modes. Flash duration can be changed to either : 25 - 50 - 75 or 100 microsecondes (1.350 to 5 400 Joules output). Various sources, such as Xenon flash tubes can be triggered by a 10 v positive pulse, 10 ps half width, 50 input loading impedance. One to four sources can be used according to necessities. Power requirement : 110 or 220 v - 50/60 Hz. With its extremely wide operating speed range, power and versatility enabling it to photograph anything in explosive field, this unit had been very usefull for optical recordings of localized ejected mass measurements with an ultra high speed framing camera Barr and Stroud CPS.

An industrial installation is said to be safe when protection of persons and property can be guaranteed under all circumstances. Fast neutron nuclear reactors are dimensioned so as to contain the consequences of major hypothetical accidents. Considerable calculation methods are used to achieve this measurement ; these programs, especially developped, have been substantiated by an important number and variety of tests which are described in this paper which also attempts to show what fast cinematograhy has contributed to these tests : complementarity and highlighting of new phenomena.

Since years there is the demand of the automobile-industry to get a light-weight, compact and small luminary for the inside lighting of cars or other details at crash-test vehicles. As the crash-tests are made with high accelarations, it is a necessity that such a luminary must withstand at least 25 G.

A loss of efficiency of P-11 and other ZnS phosphors at current densities below 200 mA/cm is well known for relatively long exciting electron pulses. There is no such direct results for electron pulses of shorter duration. It is necessary to have such data if streak tubes and cameras of the photo-electronic type are to be used for quantitative photometry. The authors in two previous papers have furnished evidence of phosphor efficiencies in streak cameras very considerably below the manufactureers quoted values. In this paper, we submit experimental data on the response of phosphors when used in streak cameras and excited by electron pulses only tens to thousands of picoseconds long. Both the rise and decay characteristics appear to be intensity dependent and the overall sensitivity of the phosphor is found to be low.

Despite a great deal of effort to improve time resolution of streak cameras, the best reported time resolution remained just under one picosecond for more than a decade. However, the sensitivity and resolution of these devices have improved substantially over this time. In this paper the design of a very high performance streak camera, capable of resolving one picosecond on 200 parallel channels with single photo-electron detection capability: the Dellistrique 2S. The system comprises of a streak tube type Picotron 200 with a photocathode resolving 200 spatial channels along a slit and with an extraction field near the photocathode of greater than 16 KV per cm and with a non saturating phosphor screen as the main component. There is post streak tube intensification of 30,000 at the streak tube output wavelength which increases both the sensitivity of the system. The intensifier is coupled to a CCD readout device fibre-optically for producing a two dimensional image with a dynamic range of greater than 500. The camera can be operated at a repetition rate of between 80 MHz and 240 MHz using R.F. scanning methods. Upto 20 parallel channels have been tried with parallel information recording, which is well below the tube's capability of recording 200 independent channels. The image is virtually distortion free across the whole usable screen length of 45 mm for a large screen tube.

The paper describes an image converter streak camera system developed in China. It consists of a streak tube, an intensifier, sweep and synchronizing circuits, a film recorder and a real-time readout system. The streak tube has a spherical photocathode and a spherical extraction mesh to correct the transit-time distortion. The measured highest time resolution is 2.5 ps. The real-time readout system with two windows consists of a SIT TV camera, a digitizing circuit, a CPU, a monitor and hardcopy devices.

Cine X-ray is technically the ideal method for recording fast events where the subject is optically obscured. However, practical problems have restricted its use to small subjects at speeds of a few hundred frames per second. Flash X-ray sources have been used instead and, although producing high quality images, the different aspect angle of each photograph complicates the analysis of the results. Several high speed eine X-ray systems have recently been developed and their performance is discussed.

High speed image converter cameras used in the streak mode of operation are being increasingly used as multi channel recorders to record large numbers of independently positionable data channels. Monofilament fibre optic light guides are used as optical switches for each data channel providing flexibility in the spatial positioning of each data measuring point and excellent synchronicity between channels. This paper describes some examples of the applications to which this technique may be put.

A new soft x-ray. and V-UV sensitive streak camera has been developed to meet the fusion research requirements of Los Alamos and Livermore National Laboratories. The main performance parameters called for were-Large number of spatial channels Longest practical cathode Undistorted streak. image

Solid state imagers with integrated scanning circuits offer many advantages over beam-scanned camera tubes in cost, reliability, size, and range of applications. Solid state electrical imagers are the realization of available methods for organizing and transferring charges from an image sensing area to electrical readout ports. Surface acoustic waves (SAWs) travelling on a piezoelectric medium have been shown to support charge transfer in a solid state device: charges can be injected into electric potential wells which travel at the acoustic speed and may be collected at a different location on the surface. The principles of SAW charge transfer devices (SAW-CTDs) are reviewed. Design concepts for SAW-CTDs and CCD-CTDs for use as electrical or optical waveform recorders and as imaging devices, e.g. streak and framing cameras operating in single-shot and continuously cycling modes, are presented.