A new method for measurement, documentation and visualization of fast repetitive processes is presented. The method extends the principle of sampling--as is currently used in measuring fast repetitive electrical signals--to video recording; using a fast external shutter in front of an unmodified, free running, standard CCIR or EIA video camera. The result is a slow-motion display that can be viewed continuously, recorded on standard video equipment or further processed like any other video signal. In a number of experiments in the field of nonlinear optics, exposure intervals of down to 100 nanoseconds at a full-frame repetition rate of 25 Hz were used, thus slowing down the time scale by a factor of 400,000. Additional features increase the efficiency, which is otherwise limited by the short exposure time, increase the maximum time-scaling factor and facilitate polarization-resolved measurements.
Rates exceeding 1000 frames/s can be achieved with multiport CCD state-of-art video sensors. In order to provide sufficient spatial resolution, sensor configurations of 512 X 512 pixels are typical. Image area is divided into segments with individual video ports. Each port includes a photocharge sensitive amplifier, typically comprising sample/hold and charge reset circuits. Some amplifiers are even provided with a double correlated sample circuit for improving the signal/noise ratio. Frame rates are proportional to the number of ports, since the individual sensor segments are run in parallel. Unfortunately, the amount of external circuitry required for signal processing increases accordingly, 16-port sensors are a quite common configuration. Cameras with even higher number of ports are prohibitively expensive. Therefore, in order to achieve very high frame readout rates with a moderate number of ports, the sensor's charge transport clock frequencies must be increased to the limit. Horizontal charge transfer frequencies exceeding 30 MHz have been achieved. The quality of the video signal deteriorates with frequency due to bandwidth limitation of the photocharge detecting amplifier. Its sample/hold and double correlated sample circuits are useless at such rates. Methods and circuits for the processing of video signals under such conditions are described. The circuits include wide bandwidth video buffer amplifiers/level translator/line drivers, fast peak stretchers, 10-bit resolution (or more) A/D converters and fiber optic data links to a remote mass digital data storage and processors. Also, the circuits must satisfy a number of practical conditions (size, power dissipation, cost) in order to make such camera useful in applications where space is limited and multiple head high frame rate cameras are required.
In this article, we propose a new sensor architecture for snapshot video. This sensor has the distinctive feature of memorizing of a sequence of images. The realization of the sensor has permitted us to study the transfer operation in a full custom implementation. For this reason we have developed, at GERE laboratory in Le Creusot, in collaboration with LIRMM laboratory in Montpellier, the experimental transfer circuits. Firstly, we present in this article the internal architecture of the sensor circuit. Next, we describe the elementary cell which is composed of a photosensitive element and memory zones. Thirdly, we describe the transfer circuit and we discuss in details the study done on it. Finally, we conclude on the feasibility of the specific sensor circuit.
Recent and proposed improvements in spatial resolution, temporal resolution, contrast, and detection efficiency for x-ray framing cameras are discussed in light of present and future laser-plasma diagnostic needs. In particular, improvements in image contrast above hard x-ray background levels is demonstrated by using high aspect ratio tapered pinholes.
We present the results obtained with a prototype of a high speed gateable Vacuum Image Pipeline (VIP) for selection of non-repetitive images from a continuous stream. It allows snapshots with a very short exposure time (of the order of 10 nanoseconds) to be accepted (or rejected) after a decision time of a few microseconds. The VIP is a vacuum tube equipped with a photocathode, a system of metallic grids and a phosphor screen. Photoelectrons produced by the images focused on the photocathode are guided by a uniform magnetic field parallel to the tube axis. By acting on the grid potentials, the drift time of the photoelectrons inside the tube can be adjusted between 0.3 and 2 microseconds. An image among many others can then be selected by an external trigger with a time resolution between 4 and 30 ns depending on the delay time. The selected photoelectrons are finally accelerated by a high potential (+15 kV) onto the phosphor screen where they reproduce the triggered image. A spatial resolution of 33 lp/mm at a magnetic field of 0.1 T has been measured. The VIP is a useful tool for high energy physics and astrophysics experiments as well as in high speed photography.
High voltage pulse of 140 ps in width and 2.7 kV was generated to gate the multiframe images on a meander shape microstripline on MCP. The measurement time range was extended to 1.1 ns while the exposure time of each image is 60 ps. The measured spatial resolution of the framing camera is 25 lp/mm. New method to reduce the exposure time down to 10 ps was simulated numerically.
The photoelectron throughput in streak tubes may be understood by using the brightness theorem to couple the photoelectron emission from the virtual cathode, through the anode aperture, to the recording screen. The virtual cathode is generated by the immersion lens formed by the photocathode-accelerating electrode structure. The throughput is limited by the anode aperture that acts as a system field stop. We have calculated the throughput for a variety of streak tubes, given the photoelectron energy distribution of some typical photocathodes. We conclude that higher throughputs are obtained when using a slot, rather than a mesh, for the accelerating electrode. The variable magnification design of the Philips P850 streak tube allows one to optimize the throughput for arbitrary photoelectron energy distribution.
Material opacities are of interest in many fields. We have developed a Bragg reflection spectrometer that is gated for imaging samples in a laser heated environment for opacity measurement. A micro-channel plate is coated with a photocathode material and a fast pulse is launched across it. Electrons are converted to photons in a phosphor and recorded on film. Optical gate pulse widths of 100 ps are achieved. Some optical pulse width and sensitivity enhancements are noted at launch and termination. Events of interest are 200 ps long. The framing window is approximately 250 ps in length. Timing jitter is a problem. The instrument timing networks have been examined, and the source of jitter is still unknown. Timing to 50 ps resolution is desired. Close in proximity to the laser-driven event leads to complications in shielding from hard x-rays, hot electrons and shock-driven damage. High Z materials provide shielding from hard x-rays. Magnets screen out hot electrons produced by laser-matter interactions. Filters provide energy fiducials. PCD's provide high resolution timing measurements. Data is recorded on film in a specially designed film pack. The instrument is designed to be used in the NOVA Laser Facility at Lawrence Livermore National Laboratory.
We have recently developed a gated monochromatic x-ray imaging diagnostic for the national Inertial-Confinement Fusion program. This new imaging system will be one of the primary diagnostics to be utilized on University of Rochester's Omega laser fusion facility. The new diagnostic is based upon a Kirkpatrick-Baez microscope dispersed by diffraction crystals, as first described by Marshall and Su. The dispersed images are gated by four individual proximity focused microchannel plates and recorded on film. Spectral coverage is tunable up to 8 keV, spectral resolution has been measured at 20 eV, temporal resolution is 80 ps, and spatial resolution is better than 10 micrometers .
The light emission from the bow shock around the tip of a metal jet formed by the collapse of a shaped-charge linear was computed for tip speeds up to 15 km/s and the laser energy needed to overwhelm this emission for a front-lit photographic application has been determined. Upon approximating the nose of the jet as a hemisphere with a 2 cm radius, a 3D inviscid flow field code was used in conjunction with a nonequilibrium air radiation code to compute the shocked air properties including the temperature, pressure and emission. For comparison, an analytical calculation of the shocked air properties and visible radiation at the flow stagnation point was made. Both calculation methods yield results which indicate that at a tip velocity exceeding 10 km/s the emission from the bow shock is equivalent to blackbody radiation. Additional values for the emission at tip velocities below 10 km/s are also contained in the paper. These results specify that a laser pulse energy of 10 mJ would be required to match this background luminosity for the 10 km/s case assuming a 1000 cm2 illuminated object area, a 1.5 nm spectral bandpass and 50 ns exposure time for a camera.
The fundamental studies on a flash vacuum-ultraviolet (VUV) generator for producing water- window x rays are described. this generator consisted of the following essential components: a high-voltage power supply, a polarity-inversion-type high-voltage pulser having a 15 nF condenser, a thyristor pulser as a trigger device, a turbo molecular pump, and a VUV tube. The VUV tube employed a mercury anode, and the ferrite cathode was embedded in the anode. The pressure in the tube was primarily determined by the steam pressure of mercury as a function of temperature. The condenser in the pulser was charged from -10 to -30 kV by the power supply, and the electric charges in the condenser were discharged to the radiation tube after closing a gap switch by the thyristor pulser. As the high electron flows from the cathode electrode evaporated the anode electrode, VUV rays were then produced. The maximum output voltage from the pulser was approximately -1 times the charging voltage, and both the tube voltage and current displayed damped oscillations. The maximum values of the tube voltage and current were 14 kV and 2.0 kA, respectively. Since the effective accelerating voltage was substantially decreased by the ferrite cathode, soft x rays were easily generated. The pulse durations of the VUV rays including water-window x rays were nearly equivalent to those of the damped oscillations of the voltage and current, and their values were less than 15 microsecond(s) .
The development of a high-intensity kilohertz-range pulsed x-ray generator and its application to dental radiography are described. The pulsed x-ray generator consisted of the following major components: a constant high-voltage power supply, a high-voltage main condenser, a hot-cathode triode, a DC power supply for the filament (hot cathode), and a grid controller. The main condenser of 0.5 (mu) F - 100 kV in the pulser was charged from 50 to 70 kV by the power supply, and the electric charges in the condenser were discharged to the triode by the grid controller. To be exact, the tube voltage decreased during the discharging for generating pulsed x rays, yet the maximum value was equivalent to the initial charged voltage of the main condenser. The maximum values of the tube current and the repetition rate were about 0.5 A and 30 kHz, respectively. The pulse width of the x rays ranged from approximately 20 to 400 microsecond(s) , and the x-ray intensity with a charged voltage of 70 kV and a total resistance of 5.1 M(Omega) was about 0.83 (mu) C/kg at 1.0 m per pulse. Using this generator, high-speed dental radiography, e.g., delayed radiography and multiple-shot radiography, was performed.
An optical technique, which measures directly reflection ratio of a rough target surface in shock compression, is introduced. This technique uses laser as light source to illuminate target surface and measures light power scattering from randomly rough surface with two differently axial angles simultaneously. Reflectivity and lightness function about tungsten target surface in shock compression are deduced with variation of light power. In experimental system, polarizing beamsplitters separate laser from radiation of heat partially. The radiation of heat, which produces noise in signals, is eliminated completely by subtraction of S polarizing light beam from P polarizing beam. Reflection coefficient of tungsten target in shock-compression has been investigated and initial experiment results show: when tungsten target is shock- compressed at 165 Gpa pressure, relative change of reflection ratio compared to unshock- compression is only several times. Discussions and analyses about the technique and experimental results are also done in the end of the paper.
AlliedSignal Aerospace Propulsion Engines produces propulsion engines for business class executive jets and auxiliary power engines for both commercial full•size airlines and business jets. As cinematographer for ASAC, it is my responsibility to provide high speed motion analysis of the catastrophic testing sequences that the engines must pass in order to receive FAA certification. This paper will outline and explain several of the tests that propulsion engines must undergo to receive FAA certification. The test parameters and setups concerning photography will be explained herein. Within the testing profile we have established our cameras of choice are Redlake hi-cams and 10-cams. These have been used exclusively because they have demonstrated the durability to withstand the harsh environments at our San Tan test facility 35 miles southeast of Phoenix. In addition to the Redlake compliment we also make use of 2500W HMI type illumination. Our test regime is structured so that the illumination for the tests permit the test to be conducted at any time of the day and regardless of weather. Testing programs at AlliedSignal have evolved from early years where seldom more than two or three cameras were used to photograph an event. Today, with the commitment to engine development time tables being so strict, we have substantially increased the photographic coverage of an engine involved in catastrophic testing like foreign object ingestion or full blade out testing. On a blade-out test, wherein a single blade is, with a small charge, explosively released at a Max-Power setting, 16 cameras are utilized to provide redundant coverage's of flight inlet areas, aircraft mounts, accessory components, exhaust behavior, and then for primary overview of the test. Optical considerations on hazardous testing require some level of risk to cameras on some positions of the testing, like blade-out events, hi-cams are at times located within 1-2 feet of the engine itself. Blast shields with lens cutouts have been used to protect cameras, but shooting in close, and lighting in close has proved challenging. The HMI lighting used is generally oriented to the flight inlet area. On a blade-out event or similar severe test, accessory and mounts are critical subjects so they are examined in a 3k to 7k frame rate. We have utilized 2k Tungsten light sources to provide illumination in these tight areas. Our lenses utilized range from 50mm lens for inlet perspectives, 25mm for overview perspective, and 10mm to 15mm lenses for extreme tight areas underneath the engine. The only real difficulty encountered on under-engine cameras has been with the effect of oil/fuel misting on lenses. on a blade-out test, and the majority of other extreme testing, the preferred frame rate is from 7K PPS to 9K PPS. Our efforts to coordinate cameras and events have proved to be very challenging so we have integrated a camera control that will initiate the cameras from a firing computer, wait for a specific % frame rate signal from the hi-cam shutter pulse circuit, and then trigger the blade charge once the desired number of cameras are at speed. This effectively has given us a fail-safe condition where the control of the event still remains secure in the event of any technical anomalies. Once the predetermined sampled cameras convey the required number at speed the firing control then commits the event to explosive blade separation. In addition to the cameras controlled by the sequencer there are cameras both high speed and real time that are initiated manually. The blade out test is by far the most severe of the testing segments and the high speed coverage directly reflects the critical aspects of the test. In addition to the blade out test, high speed photography utilizing cameras from 500 PPS us to 7K PPS are also used on the Ice Slab Ingestion, Hail Ingestion, and finally the Bird Ingestion segment. Included here are test plans that will detail the specific criteria necessary to perform each of these tests starting with the Fan Blade Containment Test, followed by the Ice and Hail Ingestion Test, and concluded by the 1.5 pound Bird Ingestion Test.
One of the most important objectives in nuclear tests is the precise and accurate measurement of the physical phenomena involved. Very fast and wide dynamic range imaging is among the multiple and other diagnostics used. A major development in scientific imaging is the increasing capabilities of diagnostic systems for gamma ray, X ray, visible and neutron spectra. However, the different sub-assemblies making up an imaging chain introduce a number of attenuations and distortions in the acquired image. The main objective of the work described here was the accurate measurement of these perturbations to correct the raw recorded image. For this, a test bench was set up and calibration methods were used to precisely adjust the spatial position of the imaging detector. To provide automation, all the instruments were controlled by a microcomputer. The quantities measured were: spatial uniformity, amplitude transfer function (linearity), background noise and noise on signal, geometric distortions, spectral response, 2D time aperture and 2D Modulation Transfer Function.
We have developed a charge transport model for predicting the effects on Charge Transfer Efficiency (CTE) of Charge Coupled Devices (CCDs) as functions of number of transfers, pixel charge flow rate, and magnitude in the CCD's vertical and horizontal charge transport mediums. The model uses carrier lifetime and mobility criteria to establish pixel speed arguments and limitations for various CCD architectures. The model is compared with experimental measurements obtained using strobed single pixel illumination and a variant of the deferred charge tail technique while independently varying the CCD pixel rates for both the vertical and horizontal readout phases. The generic model is discussed and applied to specific real CCDs. Agreement between predicted performance and actual measured performance is presented.
In this paper, the interaction process of high-power Q-switched YAG laser and MNOS-type charge-coupled devices (CCD) is studied with the help of plasma shape and structure under the repeated actions of laser pulses. Mach-Zehnder interferograms of plasmas and related experimental results produced by a 1064 nm laser beam with a pulse width of 15 ns acted upon the MNOS-type CCD are obtained for the first time.
The construction and the radiographic characteristics of a plasma flash x-ray generator having a molybdenum-target (anode tip) triode are described. This generator was primarily designed in order to perform soft radiography in dental medicine and employed the following essential components: a high-voltage power supply, a low-impedance coaxial transmission line with a gap switch, a coaxial oil condenser of 0.2 (mu) F, a turbo-molecular pump, a Krytron pulser as a trigger device, and a flash x-ray tube. The high-voltage main condenser of 0.2 (mu) F was charged from 40 to 60 kV by the power supply, and the electric charges in the condenser were discharged to the tube after closing the gap switch. Because this tube employed a long target, the plasma x-ray source which consists of molybdenum ions and electrons was easily produced by the target evaporating. The maximum tube voltage was nearly equivalent to the initial charge voltage of the main condenser, and the maximum current had a value of about 25 kA with a charging voltage of 60 kV. The average width of flash x rays was less than 1 microsecond(s) , and the time-integrated x-ray intensity with a charging voltage of 60 kV was approximately 20 (mu) C/kg at 1.0 m per pulse. The characteristic K-series intensity substantially increased according to increases in the charged voltage. High-speed dental radiography was performed using a laser timing switch and a trigger-delay device.
The constructions and the fundamental studies of two types of kilohertz-range harder pulsed x- ray generators are described. The multiple-pulse generator was primarily designed in order to increase the x-ray intensities even when the x-ray duration increased. In contrast, as the damped oscillation of the tube voltage was prevented by using two high-voltage diodes, we designed the single-pulse generator to obtain short x-ray durations. Each generator employed the following essential components: a thyratron pulser, a high-voltage double transformer, a storage battery for the hot cathode (filament), and an x-ray tube. The main condenser in the pulser was charged from 8 to 16 kV, and the electric charges in the condenser were repetitively discharged to the primary coils of the transformer. Because the high-voltage impulses from the secondary coils were then applied to the x-ray tube, repetitive x rays were generated. The x-ray tube was of a diode having a hot-cathode with a maximum temperature of about 2,000 K. The tube voltage increased in proportion to the charged voltage, and the maximum value was about 170 kV. The tube current was primarily determined by both the filament temperature and the tube voltage and had values of less than 1.5 A. The maximum intensities of the multiple and single types were about 48 and 16 nC/kg at 0.5 m per pulse. The x-ray pulse widths obtained by the single generator were less than 250 ns, and the maximum repetition rate was approximately 10 kHz.
For many years 16 mm film cameras have been used in severe environments. These film cameras are used on Hy-G automotive sleds, airborne gun cameras, range tracking and other hazardous environments. The companies and government agencies using these cameras are in need of replacing them with a more cost effective solution. Film-based cameras still produce the best resolving capability, however, film development time, chemical disposal, recurring media cost, and faster digital analysis are factors influencing the desire for a 16 mm film camera replacement. This paper will describe a new camera from Kodak that has been designed to replace 16 mm high speed film cameras.
Main features and some applied cases of Model DKF-250 High Speed Wide Framing Rate Camera are introduced. The speed of rotating mirror is variable continuously in the range of 0.12 X 104 approximately 30 X 104 rpm and the corresponding framing rate is variable on the range of 1 X 104 approximately 250 X 104 fps. These features broaden the application of high speed rotating mirror camera and make it `a camera for multiple purposes'.
We, in this paper, investigate resolution in various confocal imaging methods based on the use of an ultrashort pulsed laser beam. For bright-field confocal microscopy, time-resolved and time averaged imaging methods are considered, while resolution in two-photon and single- photon confocal fluorescence microscopy is examined. It is shown that the time-resolved approach can give a 3D image of a thick object with axial resolution approximately 30% higher than the limiting value under CW laser illumination of the same wavelength as the pulsed beam, while the transverse resolution is slightly improved.
Proc. SPIE 2549, Three-dimensional measurement of moving objects using a multiple-camera vision system, 0000 (1 September 1995); doi: 10.1117/12.218313
Machine vision systems utilizing lighting triangulation technique are often used for the depth measurement or profiling. The surface depth information is extracted from a 2D image and require the object in stationary form. For most of the industrial gauging applications, several measurements of a large moving object must be done simultaneously from different perspectives with high accuracy. This requires a system with multiple cameras that can provide the true 3D measurements in a world coordinate system. A unique solution based on IDAS vision system has been designed and developed for measuring large fast-moving objects with high accuracy. The vision system developed consists of three camera systems. Each camera system is calibrated to obtain the information in all three dimensions through a very detail calibration procedure using a common reference jig. Information obtained from each of the three cameras can be converted into a world coordinate system that is created during the calibration. Because of the unified coordinate system, objects can be accurately measured independent of their orientation and position. This enables the system to perform the 3D measurement of the objects moving at a high speed without the occlusion problem. The technique can be expanded to n-camera system to support the measurement of complex objects. The details of the system configuration, optics selections, calibration procedure, and the image processing algorithms will be included in this paper.
Proc. SPIE 2549, Three-channel diagnostical system for the time-space measurement of the magnetic fields in the laser-produced plasma, 0000 (1 September 1995); doi: 10.1117/12.218314
The spatial-temporal structure of magnetic fields in a laser plasma has been studied by a Faraday-effect method by means of three-channel polarointerferometer system. The diagnostic method is based on simultaneous measurements of the polarization plane rotation angle and the interference phase shift of the probing laser pals. The diagnostical system allows to record simultaneously Faraday, shadow, and interference images of the plasma with a high spatial resolution, approximately 5 micrometers , and a high time resolution, approximately 50 ps, both in the framing mode (with an exposure time approximately 1.5 ns) and in a dynamic mode (in this case we used two image converter cameras operated in the streak mode). The contrast of the polarimeter was approximately 3(DOT)10-5, so it was possible to measure the angle through which the polarization plane rotated within +/- 0.1 deg. The error in the determination of the phase shift of the probe wave was +/- 0.1 line.
The emergence of lasers opened a new epoch in optical diagnostics of fast processes in many fields and in gasdynamics in particular (holography, holographic and speckle interferometry, etc.). However, as to classical optical methods traditionally applied in gasdynamics, these new light sources, side by side with numerous advantages, have a certain demerit. High spatial coherency of a laser beam leads to appearance of speckle structure in an image obtained with the help of interferometers or schlieren devices, which can not be eliminated even with scatterer application. A new semiconductor laser with electron-beam pumping is free from such a demerit. Its quality and capabilities are illustrated by flow pattern obtained at an experiment of gasdynamics, ballistics, laser plasma diagnostics.
Proc. SPIE 2549, Theoretical comparison between image intensifier tubes using EBCCD and phosphor readout, 0000 (1 September 1995); doi: 10.1117/12.218316
This paper discusses issues pertaining to the use of Electron Bombarded Charge Coupled Devices (EBCCDs) as readout structures for image intensifiers. Theoretical models are presented that form the basis for comparison of image tubes using EBCCDs and those using phosphor screens in terms of several figures of merit including: collection efficiency, signal to noise ratio, dynamic range, and spatial resolution. Some discussion is included with regards to the current state of EBCCD production and difficulties in realizing all of the benefits that EBCCD readout structures could provide.
The authors have developed a second generation CCD image sensor for high speed motion analysis. The 2/3 inch format device is constructed on a 16 micrometers square pixel pitch with 512 (H) X 512 (V) active elements. Design is based on a progressive scan interline transfer architecture with a vertical overflow drain for blooming and exposure control. Full resolution is achieved at 1,000 frames per second by use of eight parallel outputs operating at a data rate of 40 MHz per tap. Other performance parameters include dynamic range of 62 dB, less than 1% image lag, and very low smear. This work details the essential design features and reports results of our preliminary evaluation.