Spatial responsivities for 3cm aperture and 22.5cm aperture absorbing glass volume calorimeters have been measured. It is shown that a simple correction to the calorimeter output can improve uniformity. The technique of electrical calibration is discussed and it is shown that calibration is achievable to within 3% of absolute.

In attempting to set up a system for measuring the power and energy of pulsed and c.w. lasers at a variety of pulse durations, power or energy levels, and wavelengths it is neither practical nor very desirable to have several absolute standards working at different points in the multi-dimensional space representing the problem. This paper is not so much a review as a description of the current NPL programme which has evolved under the influence of various external pressures and staff availability.

An absolute laser power meter, based on the momentum exchanged between the incident beam and a movable mirror is described. The force exerted by the beam on the mirror is proportional to the incident power. Thus from the measurement of the force value, we can deduce the incident power carried by the beam. But, generally speaking, the coefficient linking the exerted force and the incident power contains parameters which cannot be known with precision. The conceived device we describe, cancels these parameters.

Submillimetre-wavelength radiation is now used in a wide and growing range of research activities. Absolute power measurements are mainly required for the design and specification of components, and in the calibration of instruments used in emission spectrometry. Work at NPL on power meters for submillimetre lasers and similar sources is discussed, as are the problems of calibrating broad-band spectroradiometers used in stratospheric monitoring and plasma diagnostics.

The applications of infrared detectors can be divided into the two main areas of imaging and measurement, the latter including photometry, spectroscopy and range finding. The detector properties such as wavelength range, sensitivity and speed of response can be quite different for each specific requirement and these differences will be covered in detail. The two most important types of infrared detectors are thermal and photon devices. Thermal detectors utilize materials with a strongly temperature dependent property such as electrical conductivity or thermal expansion. The main characteristic of these devices is their broad, flat spectral response, however the major disadvantage is that their response times are relatively slow - typically of the order of milliseconds - which limits their use in certain applications. This situation has been improved recently, with the development of pyroelectric devices, which use insulating materials such as triglycine sulphate which have a permanent internal electrical polarization, the magnitude of which is temperature dependent. With suitable circuitry the response time can be less than 1 μsec. Photon detectors are fabricated from semiconducting materials in which the incident radiation is absorbed causing electronic transitions within the photosensitive material. These sensors have a spectral response with a precise cut-off wavelength, but generally have higher sensitivities and are much faster than thermal devices. Intrinsic photo conductors and photodiodes were first developed using PbS, PbSe, PbTe or InSb; however, these materials are not sensitive beyond about 7 μm. Subsequently extrinsic devices had to be used for longer wave-length detection. The most popular choice was doped germanium, which could be used to beyond 100 μm. The major disadvantage of these devices was the necessity to cool to much lower temperatures than intrinsic detectors. The recent development of the ternary alloy systems, such as cadmium mercury telluride and lead tin telluride, in which the energy gap may be varied by choosing an appropriate composition has now produced high performance intrinsic detectors in the wavelength range 2-20 µm.

AGA System 680 (2-5 µm) and 680 LW (8-13 µm) thermovision cameras have been interfaced to a high speed ADC and pulse code modulation recording system. The data acquisition system continuously collects the digitised output from both AGA linescan cameras, each at 209 Kwords/s, with a dynamic range of 60 dB for periods up to 15 minutes (tape reel size limited). The data is subsequently recovered from tape under minicomputer control at low replay speeds, converted into a frame sized block format, and dumped onto digital magnetic tape for further computer analysis. A range of peripheral devices, including a colour graphics system, provide facilities for off-line data processing, frame storage and display. Quantitative field of view measurements have been made using high intensity quartz halogen bulbs and the instruments have been radiometrically calibrated against reference black bodies in wave bands defined by interference filters.

Energy gains and losses from the surfaces of buildings, industrial process equipment, radomes and space vehicles, are examples of an area of predictive technology which is becoming increasingly important. Energy gains are predicted using the solar-integrated transmittance and absorptance of the surfaces involved. NPL uses measurements of the spectral hemispherical transmittance and reflectance over the wavelength range 0.3 μm to 2.1 μm to compute the solar transmittance and solar absorptance as part of its existing calibration services. In order to predict energy losses it is necessary to know the spectrally-integrated hemispherical or directional emissivity, and to cope with temperatures from -20°C to + 400°C the wavelength range which needs to be covered is 2.5 to 60 μm. By Kirchhoff's Law, the emissivity under stated conditions is equal to the absorptance calculated from the reflectance and transmittance under corresponding conditions. Because direct measurement of spectral emissivity is impracticable for calibration work at low temperatures, NPL has designed an infrared absolute reflectance/transmittance facility to measure spectral hemispherical/directional radiance factors over an adequate spectral range for a variety of angles of emission. The principles of this design and technique are discussed and illustrated.

The complete radiometric analysis of a fluorescent opaque sample is described. Conventionally, the spectral analysis of reflected radiation from a non-fluorescent opaque object is made by comparing the radiation reflected from the sample under study, with that from a near-perfect reflecting diffuser, dispersing the radiation either before or after reflection. This approach using one monochromator cannot provide a complete analysis of the radiometric properties of a fluorescent substance. The NPL Spectrofluorimeter described here has been developed to tackle this problem. The instrument uses two monochromators to enable the fluorescent radiation to be separated from the reflected radiation and to allow the spectral distribution of fluorescent emission to be studied, for a given excitation. Three types of spectrofluorimetric measurement are possible enabling the sample to be characterized in terms of a set of radiance factors, or in terms of its excitation and emission spectra, and permitting the behaviour of the sample under any stated illuminant to be predicted.

Polarization pyrometry is discussed and it is shown that under certain conditions this thermometric technique does not rely on a knowledge of the emissivity of the surface whose temperature is to be measured. An instrument has been designed which operates over a range from 200 - 500 °C, and some preliminary results are presented for tests on stainless steel, aluminium and copper surfaces.

At Sira, a number of pyrometers have been evaluated, both portable and fixed installation, covering the temperature range 50 to 1200°C. Performance testing pyrometers of the radiation type requires accurate and stable black body sources covering the temperature range of interest and having a sufficiently large aperature for the particular instrument undergoing evaluation. The same equipment may be used for optical (disappearing filament) pyrometers, although special test lamps are available at lower cost if this is the only type of pyrometer to be considered. The paper describes tests carried out at Sira during the course of several evaluations and the problems encountered.

Development of instrumentation to measure high heat-flux levels (≥ 1.0 MW/m2 ) is described. Progress has followed the provision of arc-image thermal irradiation facilities of power inputs between 2-30 kW. Fact response radiometers have been constructed and after - calibration against an absolute calorimetric measurement used to indicate incident radiant intensity. Constantan-foil disc radiometers can confer a high degree of spatial as well as temporal resolution with a high degree of precision. Delicate disc-centre thermocouple joints were necessary as well as robust peripheral welding to a cold solderless junction. About ten different coating blacks have been assessed for burn-off tendency over the linear region of response, up to thermal intensities of 5.0 MW/m2.

Crops that are supplied with adequate water and soil nutrients grow at a rate that is proportional to the intercepted solar radiation. The fraction of solar energy stored in a crop by photosynthesis may be considered as an efficiency, and typical values for various farming systems range from 0.04 to 1.0 %. One factor limiting this efficiency is the ability of the crop to intercept light, which may be measured by a combination of tube solarimeters, or other spatially averaging instruments, above and below the crop canopy. Photosynthesis is sensitive to radiation in the 400-700 nm band, but broader band measure-ments may be used in practice because the spectral distribution of incident solar radiation varies very little with weather conditions. Many plant development processes respond to the spectral quality of light. The ratio of light in the red band (centred on 660 nm) and the far-red band (centred on 730 nm) determines the concentrations of two forms of phytochrome which control plant development. Light quality in crop canopies may have important ecological implications and measurements can be made using a portable spectroradiometer.

A leaf typically has a small reflectance (about 0.1) in the visible spectral region and a larger reflectance (about 0.6) in the near infra-red region. Spectral differences are due to selective absorption by the chlorophyll in the red and internal leaf scattering in the near infra-red. The reflectance of a crop canopy is similar to leaf reflectance but other factors such as the orientation of leaves, shadows, background reflectivities, solar zenith angles and the physiological status of the plants modify the process. The ratio of light reflected in the near infra-red and red spectral bands is well correlated with the amount of vegetation cover. Normally, light intercepted by vegetation is determined from a knowledge of the leaf area index. It is proposed that the ratio far red/ red (or visible) can give a direct measure of the amount of light intercepted by a crop, thus bypassing destructive leaf area determinations.

The preliminary results of a recently initiated programme of photocell comparison at wavelength down to 120 nm are presented. Emphasis has been placed on uniformity of the photocathode and on the repeatability of measurements. For a few cells, measurements have been made over a period of one year. For predictions about how the cells will behave over several years, the results obtained on UV and visible cells at NPL and on VUV cells at the National Bureau of Standards are used.

The majority of spectroradiometers make measurements at a number of discrete wavelength settings spaced evenly across the spectrum. Many modern light sources such as fluorescent or metal halide lamps have complex line spectra which may not be properly evaluated by this method. An automated spectroradiometer system involving a non-stop spectral scan with continuous integration of the output signal from the detector is described. The method is designed to make accurate measurements of all types of spectral power distribution whether made up of lines or continuum or any mixture of the two.