Table of contents

An outline is given of the nature of aliasing errors, which are caused if the signal is not band limited before sampling. In the case of sensors with digital output, bandpass filtering before sampling is often impossible because of the direct coupling with the process to be measured. The errors are treated, leading to exact as well as to approximate results. Several ways of minimizing the errors are considered which take into account the influence of signal processing after sampling and the consequences of non-ideal demodulation filtering (interpolation). The investigations show among other things the new and important result that, in the case of a mean operation, the sampling frequency may only be half the value due to the sampling theorem. Beyond this, the pulse shape of the samples should be chosen in such a way that the spectral density decreases with frequency as fast as possible. A numerical example demonstrates the efficiency of these methods.

The pulsed-laser atom probe is a time-of-flight mass spectrometer which is used to study the structure and composition of solid samples in atomic detail. A high electric field and a short-duration laser pulse are applied to a field ion microscope tip to remove surface atoms as ions for time-of-flight analysis. This ion generation mechanism extends the capabilities of the conventional, pulsed-voltage atom probe used previously. The instrumentation associated with the pulsed-laser atom probe, the physical mechanisms which underlie its operation, and the unique applications of the technique are reviewed.

In response to a need by the aerospace industry to provide standards for the calibration of eddy-current conductivity meters, a measurement system based on a Heydweiller bridge and traceable to national standards has been developed to determine the resistivity of annular specimens with an uncertainty of the order of 0.7%. Sets of reference blocks have been produced covering the conductivity range 14 to 35 MS m^-1 and calibrated, with an uncertainty of 1%, by comparison with the annular rings at frequencies in the range 10 to 100 kHz; they are now available to the industry.

A design of a thermobalance for study of fluorination of carbon is discussed. A beryllium-copper spring balance and a linear variable differential transformer (LVDT) are used as the sensing mechanism to monitor weight change and thus measure the extent of the reaction. Corrosion problems are overcome by previous passivation of these items. The sealable nature of this thermobalance allows a complete mass balance for the kinetic study. Data are acquired by a microcomputer connected to the balance through a commercial data acquisition system, and the reaction rate profile is computed immediately after completion of the reaction.

A perfectly diamagnetic foil rolled up to form an open cylinder is shown to have only a spatially constant magnetic field in its bore, irrespective of whether the diamagnetism results from superconductivity or skin effect. This can be used to homogenise the field of a short coil or to generate a region of high homogeneity anywhere in an inhomogeneous field. If a superconducting foil is used, the method will work from DC to RF, limited only by the corresponding crucial field; if a normal conductor is used, the lower frequency limited will be at audio frequencies.

A magnetic susceptometer is described which, for the first time, allows the volume dependence of the susceptibility of weakly magnetic materials to be measured. Such measurements have been made possible by utilising the high sensitivity of a SQUID detector. The susceptometer has been designed to use He gas as the pressure medium, thus ensuring that the pressure is applied hydrostatically. Measurements can be made over the temperature range 4-300 K.

The authors have improved the properties of their eddy current shield by equipping it with an active compensation system to cancel the static and low-frequency magnetic field. The performance of the compensation system is measured very accurately. These measurements and the measurement of biomagnetic fields show that an eddy-current shield with an active shielding is a practical solution for clinical biomagnetic measurements because of its low cost and the small space it needs.

The use of Bragg cells for radio-frequency spectrum analysis is now well established and an extension of the technology into direction of arrival measurements using an optical interferometer was recently demonstrated. In an acousto-optic direction of arrival interferometer the signals from a pair of radio antennas spaced less than a quarter of the radio wavelength apart are amplified and fed to a pair of Bragg cells mounted in the two arms of an optical interferometer. The relative phase of the radio signals is transferred to the diffracted beams produced by the Bragg cells and when the beams are recombined the intensity modulation produced can be used to measure the relative phase of the radio signals. A prototype instrument to determine the feasibility of the system is now being built and a new interferometer using a pair of Kosters double-image prisms has been designed that has inherent stability due to its symmetrical optical system. The design of the interferometer is discussed, a brief description is given of the electronic control systems required to maintain the accuracy of measurement, and some results obtained with the equipment at VHF frequencies from 30 to 55 MHz are presented.

A simple, single-channel 2 mm microwave interferometer has been developed for plasma electron density measurements on the Joint European Torus (JET) experiment. A long (76 m) plasma-probing beam path in an oversized waveguide enables all sensitive components to be located outside JETs biological shield. A high sensitivity liquid-helium-cooled InSb detector compensates for the high waveguide attenuation ( approximately 65 dB). Detection and digitisation noise limit the minimum resolvable phase change with the interferometer to +or-1/50th of a fringe. Since the instrument measures the electron density integral, integral n_e dL, along a nearly central vertical chord through the plasma, this corresponds to an average electron density change along the chord of only +or-76*10¹⁵ m^-3 for a typical plasma with a 3 m vertical diameter. The maximum measurable electron density is expected to be limited by refraction to approximately 8*10¹⁹ m^-3 ( approximately 230 fringes in the interferometer).

An ion chamber has an angular response. The author studied a special mask to compensate for this angular response so as to make it constant and independent of the incident angle. The mask was composed of two sheets etched with an opaque triangular pattern.

The detection quantum efficiency (DQE) of microchannel plates for electrons in the energy range from 5 to 50 keV was measured with enough spatial resolution to resolve the single channels using a scanning electron microscope. Between 10 and 50 keV the DQE is found to decrease with increasing energy from 0.4 to 0.2. Microchannel plates for image intensification in electron microscopy have been found inferior to photographic recording and well designed image pickup devises using phosphor screens on fibre plates. Nevertheless their DQE is sufficient for observing low intensity electron images at a high brightness level if some additional noise can be tolerated.

A commercial 50 mm microfocus cathode ray tube has been redesigned to incorporate a carbon-fibre field-emission electron source that can be thermally processed in situ to avoid 'tip explosion'. With the aid of an external optically controlled feedback system, it has been possible to eliminate the characteristic low frequency (<or=10 Hz) fluctuations in the emission current associated with this type of source, and thereby obtain a high-definition and stable display image at pressures <or=10^-6 Pa.

Describes the construction of a set-up for energy-dispersive X-ray diffraction measurements on weakly scattering (amorphous) samples. To minimise the effects of air scattering the diffraction chamber can be evacuated to pressures below 10 Pa. The set-up is fully radiation safe; the X-ray optics can be aligned by means of small DC-driven motors, while the X-ray beam is switched on. Some results of a thin (3 mu m) sputtered amorphous Fe₅₀B₅₀ film are shown to illustrate the performance of the set-up.

The construction of a Blumlein line for the power supply of a Nd-glass laser emitting a periodic pulse train of 200 J and approximately 1 ms duration is described. Methods for the simulation of the pulse shape by a low voltage experiment and by numerical modelling are described and checked by experiment. The described Blumlein line has an energy of 26 kJ and delivers a pulse with a fairly flat top for 0.7 ms at a voltage of 3 kV.

An experimental arrangement is described which uses a synchronised probe laser to make measurements of the small-signal energy gain in a KrF laser amplifier. An optical system allowing spatial resolution of the gain within the laser aperture is detailed. A simple but precise energy ratiometer using modified commercial vacuum photodiodes is described. The accuracy of the technique is shown to be better than 2%.

The authors report on a new compact pulsed dye laser cavity arrangement with a U-shaped resonator with an echelle grating and a prism deflecting beam expander with a magnification of 112. The double functional grating is a resonant reflector of the magnified and the non-magnified beam. With this apparatus the broadband background is suppressed leaving up to 0.8% of the small line width signal. The spectral line width of the laser radiation without an intracavity FP etalon is 0.15 cm^-1. The line width with a 5 mm air-spaced FP etalon is 0.02 cm^-1; and with a 15 mm FP etalon one-mode operation is possible. The passive wavelength stability is better than +or-0.2 pm for 2.5 min or better than +or-1 pm for 3 h. Wavelength tuning is possible by turning the grating and/or the FP etalon or by changing the pressure. The optical energy efficiency when using one amplifier stage is 5% for Rhodamine 6G at 580 nm.

A capacitance transducer is described which is particularly suitable for the measurement of ultrasonic surface acoustic transients. The transducer has a small spherical electrode in a simple construction which is straightforward to use in comparison with other designs. A typical transducer has been compared experimentally with a parallel-plate capacitance probe using laser-generated acoustic transients as a standard source. The transducer is shown to have a frequency response in excess of 5 MHz with a sensitivity of 1.5 mV nm^-1 for Rayleigh surface transients a few nanometres in amplitude.

Describes the development of a multi-element hot-wire anemometry system designed to measure the fluctuating velocity and temperature in non-isothermal boundary layer flows. This system utilised only standard constant-temperature anemometers and commercially-available sensors. Although the maximum frequency response for the present system was limited by the available signal digitisation rates the probe-anemometry combination itself was shown to be capable of resolving signals to approximately 50 kHz. The performance of the system was evaluated by comparing velocity and temperature statistics measured in an equilibrium turbulent boundary layer flow with similar results from other investigations.

Previous researchers have observed turbulent transport of scalars in liquid jets by fluorescence. However, the technique has not been thoroughly investigated. This work examines a fluorescence technique in detail. Fluorescein sodium in a buffer solution was used in bench tests and in an experimental jet flow. Instrumentation and properties of the dye were considered. The author measured the variation of fluorescence intensity with temperature, pH, concentration and laser intensity. Concentration of reaction product in an acid-base reaction by laser-induced fluorescence is also discussed. The distinction between requirements for qualitative flow visualisation and for quantitative concentration measurements was clarified.

Short exposures of a conventional Clark cell to oxygen-containing sample gases yield time-dependent peak-like signals. The integrals of these peaks are strictly proportional to the oxygen concentrations in the respective samples, providing exposure times are equal. The linear dynamic range comprises five decades. By the passage of small plugs of sample gases which have been introduced into a stream of an appropriate transport gas, the following applications are possible: (i) the oxygen content of any gas mixture can be readily analysed with millilitre-sized samples; (ii) the amount of oxygen physically dissolved in any liquid can be determined independently of its thermodynamic potential and the viscosity of the sample; (iii) by the application of a pulsed transport gas stream measurement of the de-oxygenation of solid samples is also possible; (iv) sensitive oxygen measurements with different diffusion resistances are possible.

A simple and inexpensive portable digital photometer system for the 350-1000 nm band has been developed. The TTL level digital output from its unique photon transducer has been calibrated with different sources for measurements in absolute units of W m^-2. At constant irradiance of 10 W m^-2, the relative standard deviation was found to be better than 0.14% of the mean.