Table of contents

This paper deals with design methods for Pirani and related gauges. The analytical approach is discussed, with particular reference to bridge classification and `gauge sensitivity'. The application of the load-line technique is also discussed, in some detail. It is shown that a useful insight may be gained, by this method, into the interdependence of the element and its associated bridge, that it may be used over the whole range for which the output is pressure dependent and with equal facility for both metallic filament and thermistor elements.

An automatic torque magnetometer is described in which a modified ballistic galvanometer is used as a torque head. Torque-free electrical contacts and a means for easy replacement of the suspending fibre are incorporated into the torque head. A compact vacuum furnace enables measurements to be made in the temperature range 300-1000°K and in fields up to 12 kOe using a 7 in electromagnet. A digitizing and tape punch unit permits rapid processing of the data. The measurable range of torques, 10⁻¹-10³ dyn cm, enables both natural rock samples and synthetic ferrimagnetic single crystals to be studied.

In order to carry out magnetic measurements it is often necessary to have demagnetized specimens. The theory is given for the time of magnetic flux change in cylindrical specimens, and the optimum frequency of demagnetization deduced. For an 8 mm radius, 3% silicon iron bar the optimum frequency is approximately ½ Hz. The construction of a demagnetizer suitable for such bars is described and test oscillograms given.

This paper describes an automatic torque magnetometer with a sensitivity of 1 nN m per cm deflection, for the measurement of very small magnetic anisotropies (of the order of 0·1 mJ kg⁻¹) in the presence of large values of magnetic moment. This high sensitivity is obtained because of a new frictionless method used to eliminate unwanted lateral displacements of the specimen suspended in an inhomogeneous magnetic field.

When a digital computer acts as an on-line computational device in a closed-loop measurement or control system, the performance is degenerated by the quantization errors incurred in reading continuous data into the machine. The analysis presented in the paper shows that the increase in steady-state mean square error introduced by this degenerative effect can significantly be reduced by employing non-uniform quantization of input data. The resulting deterioration in transient performance is negligible.

It is shown, under small signal conditions, that the spectrum of non-uniform quantization errors is essentially white. As a result, the expected maximum accuracy improvement in a given system can be easily evaluated. As a means of verifying the analysis, the performance of a practical direct digital control system employing non-uniform quantization is compared with a counterpart having uniform quantization. The steady state response to a spectral input shows an accuracy improvement equivalent to 7 bits.

The technique of non-uniform quantization described in the paper is readily implemented to improve the steady state accuracy of an existing system, or alternatively for a given design-consistent accuracy, to economize in analogue-to-digital converter bit capacity.

A derivative attachment for a recording spectrophotometer is described which gives a direct measurement of the first or second wavelength derivative of the light intensity transmitted by a sample. Such measurements can lead to a better detection of overlapping bands in an absorption spectrum. Methods of calibrating and evaluating the performance of the instrument are given together with a derivative trace of the MoS₂ exciton absorption spectrum which illustrates the advantages of the derivative method.

A double beam spectrophotometer is described, suitable for the measurements of very low attenuation coefficients in glass in the wavelength range 500-1000 nm. The instrument is shown to have a balance stability of ±1 × 10⁻⁵. The systematic errors involved in the use of the long samples required for these measurements are studied. The instrument is shown to be satisfactory in making measurements of attenuation coefficients of 0·0001 cm⁻¹ with an accuracy to ±0·000 01 cm⁻¹ when sample pairs of 20 cm length difference are used. Results of measurements on very low loss silica, which illustrate the performance of the instrument, are given.

The sensitivity of omegatron mass spectrometers is limited by the space-charge effects due to the electronic and ionic charges, although in usual conditions a partial compensation of these charges occurs. A new operation mode of the omegatron, characterized by the time separation of the ionization and ion collection processes, is proposed. Experimental data are reported which show that a considerable increase in sensitivity, together with a more stable mass calibration of the instrument, can be achieved in this operation mode, especially in low pressure measurements. The time separation mode may also prove to be a useful tool in the study of ion-molecule reactions.

A description is given of simple devices designed for the production of single bubbles from a submerged orifice. The design affords the introduction of small quantities of water-soluble dye into the air cell as it escapes from the orifice. The dye delineates the disturbed region on the downstream side of the ascending bubbles.

A scanning device is described which allows accurate localization in one dimension of the distribution of weak beta radiation (¹⁴C and ³⁵S) over a planar surface. A single uncooled photomultiplier tube is used with a very small anthracene crystal to achieve both high detection sensitivity and a low background counting rate. Methods of cutting small anthracene crystals and of making rectangular slit collimators of 0·1 mm width are described; the latter is adaptable to other designs. A geared drive from a constant speed motor moves the combined photomultiplier, crystal and collimator system over the labelled surface, and the distribution of activity is recorded automatically on a chart recorder. For a collimator 0·2 mm wide and 0·7 mm deep moving 0·7 mm above a straight planar edge, the instrument's resolution is 0·54 mm while the comparable value derived geometrically is 0·4 mm.

A solid-state sampling circuit (`boxcar integrator') is described which has a ratio of holding time to learning time in excess of 10¹¹. The circuit may be used to improve the signal-to-noise ratio for signals having repetition frequencies between 500 and 10⁻⁴ pulses per second or to sample a 200 ns portion of a single event and hold the sample to within 1% for 6 h. It accepts positive or negative signals of up to 5 v amplitude, has unity gain and has an overall linearity of better than ±0·1%.

A device is described with which tensile test specimens may be irradiated under a well-defined tensile stress with fast neutrons in low temperature irradiation facilities in order to stabilize the dislocation arrangement present in the stress-applied state.

The measured e.m.f. of some saturated Weston cadmium cells is raised while an alternating current of a few microamperes is passing through them. The rise is approximately proportional to the square of the amplitude of the alternating current; over the frequency range studied (50-300 Hz) it diminishes as the frequency is increased. The magnitude of this effect is correlated with the internal resistance and with the age of the cells. Attention is drawn to the relevance of these results to the design of mains-driven current supplies for precision potentiometers.

A capacitive proximity gauge is described, which has a novel configuration of electrodes and makes use of the fringing capacitance between them. The device is shown to have certain advantages, particularly in the linearity of its output and the down-grading permissible in the performance of connecting leads. The variation of characteristics with geometrical details has been established experimentally, allowing optimum designs to be made to meet any chosen performance specification.

Several methods are reviewed for generating electron channelling patterns from small selected areas on bulk specimens in the scanning electron microscope. Procedures are given for indexing patterns, and it is shown that Kikuchi maps can be used for this purpose. To illustrate one possible application, the crystallographic misorientation across a low-angle boundary is determined.

An automatic apparatus has been constructed for weighing in different atmospheres at pressures between 1 μtorr and 1 ktorr and temperatures between 76 and 1500°K. In the pressure range between 10 mtorr and 800 torr an adsorption isotherm can be measured in 100 isobaric steps upwards, and subsequently the desorption isotherms can be determined by the same stepwise measurement downwards. The pressure control system is operated via a (mercury-free) buoyancy gauge; programmes and measuring times are variable. From the results obtained it is possible to calculate specific surface area, pore size distribution, average particle size or surface roughness of solids, kinetics of chemisorption, and other parameters.

A small vacuum hot-press is described which is easily operated and maintained within a Harwell MK II glove box. The press incorporates a sliding vacuum seal.

Superconducting, saddle-coil dipole magnets are used for bending and magnetohydrodynamic generator magnets. The uniform transverse fields may be produced by coils of tubular form with a sin θ current distribution or by uniform current density coils with cross sections of overlapping circles or ellipses. It is shown that overlapping circles and ellipses use less conductor. A simple new approximation to the overlapping circle or ellipse form is shown, in which pancake windings are used rather than the layer-type windings used previously. Pancake windings have an advantage in that the large transverse mechanical forces are transmitted through the pancakes without passing through the cooling channels as in layer-wound dipole coils. Pancake coils are simple to wind and permit an additional important innovation in that bell-shaped ends can prevent the field at the windings exceeding the uniform central field anywhere, thus allowing conductors to be optimized for actual working field rather than a higher peak field. Reduced end effects and high homogeneity are obtainable with these techniques. Models and tests are described.

A readily constructed pressure transducer is described which, with suitable electronics, is capable of detecting pressure changes equivalent to 0·2 mm of water.

This note describes a relatively simple method of measuring the anisotropy field and the angle between the easy axis and the geometric axis in the case of cylindrical films plated on nonferromagnetic wires. The method is based on the Procopiu effect.

In this note a novel form of spectrometer table is described. It has easily demountable arms and an incremental accuracy to 10⁻⁵ radian.

A method using a light-sensitive potentiometer to facilitate accurate measurement of the displacement of an optical lever system has been described. The output characteristics of the potentiometer may be recorded concurrently on a potentiometric chart recorder. The use of the instrument in extension and torsion experiments on single wool fibres has been illustrated and discussed. The technique provides a convenient and automatic method of accurately measuring and recording the displacement of an optical lever system.

A description is given of a calorimeter suitable for measuring the output power from continuous wave carbon-dioxide lasers. The calorimeter is absolutely calibrated and may be used to measure powers from 1 W to 1 kW, with an error of not more than 3% at 100 W.

Gamma detector arrays, in the form of precut mosaics on single blocks of semiconductor material, are being considered for use in nuclear medicine, but have generated a collimator aperture-indexing problem, because of the need to match the coordinates of each aperture and detector element. Previously, multichannel collimators could have random hole positions, and many techniques are available for their production. In this note is described a proven technique for the production of an aperture-indexed collimator, and other methods are suggested that may prove applicable.

A balance is described which eliminates some of the disadvantages of the `classical' design of skin friction measuring devices. By revising the motion of the sensing plate, one of the major sources of error is removed. Further advantages of the new design are the relocation of the displacement coil remote from the sensing plate, and the displacement amplification possible with the torque spring configuration.

A new method of controlling electron beam evaporation of beryllium utilizing the light emitted by the vapour as a measure of evaporation rate is described. This method has been used to give films with thickness accuracies to three per cent.

A goniometer for ultra-high vacuum is described. It allows rotation about three independent axes by angles of 270°, 20° and 15° respectively with an accuracy to 1/10°. The goniometer has been used successfully for six months.

A simple cryostat, developed from a tension-compression rig for plastic deformation, is described. The apparatus is based on the heat balance principle, and operates between 117 and 253°K if isopentane is used as a cooling medium.