Loading
D315-26 · Miscellaneous
late 1990s
743104314
capacitive aneroid
aluminium, cardboard, glass, paper, plastic, steel
Object Overview
Before us is a Meteorological Radiosonde model RS80-67 dating from the late 1990s to the first half of the 2000s, made by the Finnish company Vaisala Oyj. It is a single-use automatic measuring instrument, carried into the atmosphere beneath a free-flying balloon and transmitting its measurements by radio to a ground receiving station. Its task is to obtain a vertical profile of the principal properties of the air: pressure, temperature and humidity — a set designated in international practice by the abbreviation PTU. In addition, wind speed and direction at various altitudes are derived from the movement of the sonde; in equipment of this generation that was done by radio direction finding, by hyperbolic radio navigation systems (Omega, LORAN-C) or by a GPS receiver.
The operating principle took shape in the early 1930s and has not changed in essentials since. A latex balloon filled with hydrogen or helium ascends at about 5–6 metres per second. Suspended beneath it on a line hangs the instrument, which continuously interrogates its sensors and transmits the data downward once a second. At an altitude of around 30 kilometres, where pressure is about one per cent of that at the surface, the envelope bursts and the instrument descends by parachute several hundred kilometres from the launch site. It is not designed to be recovered and cannot be used again: the entire construction is subordinated to the requirement of minimum unit cost at a given accuracy.
The RS80 family was released by Vaisala in 1981 and remained in production until 2008 — twenty-seven years, an exceptional span for measuring equipment. It became the most widely used radiosonde of its time and was employed by the World Meteorological Organization as a transfer standard in international intercomparison trials, that is, as the instrument against which all others were compared.
The "-67" index in the model designation indicates the frequency configuration: this variant operates in the 1680 MHz band, used when the sonde was tracked by a ground radiotheodolite. Versions for the 400–406 MHz band were produced in parallel, while the letter suffixes (A, H, G, N) denoted the type of humidity sensor and the presence of a navigation receiver.
The set is supplied in a hermetically sealed bag of gold-yellow metallised polymer film. The seal runs around the entire perimeter, and a dashed opening line is printed along the upper edge. On the front are the VAISALA logo with the company slogan Measuring the Environment, the word RADIOSONDE and a label bearing the model designation: RS80-67. Alongside are the manufacturer's details: Vaisala Oyj, P.O. Box 26, FIN-00421 Helsinki, Finland, together with telephone and fax numbers, the help desk address and the website address.
The packaging here is not a formality but part of the measurement chain. A factory-made humidity sensor is hygroscopic by its very nature, and the sealed bag must keep it in a stable condition from factory calibration until the moment of launch — sometimes for years. It was precisely this packaging that later proved to be the cause of the well-known systematic dry bias of the RS80 series: the polymer film outgassed volatile compounds which occupied binding sites in the sensor's active layer intended for water molecules, so that over time the instrument began to understate humidity, and the longer it had been in storage the more it did so. The phenomenon was identified in the 1990s through comparison with independent measurement methods and was subsequently accounted for by correction algorithms.
Housing. The instrument's shell is made of heavy cardboard with a moisture-resistant coating, printed in white and pale blue with the legends VAISALA RADIOSONDE RS80 and the website address. The dimensions of the housing without the sensor boom are 55 × 147 × 90 mm. Inside the cardboard shell sits a moulded expanded polystyrene insert serving three purposes at once: structural (the boards and the battery are held in it), thermally insulating (the electronics must not freeze at −70 °C) and shock-absorbing on landing.
Suspension fitting. A metal eyelet is set into the upper flap of the housing, through which the suspension line is passed. Through it the entire mass of the instrument — about 220 g including the activated battery — is transferred to the suspension.
Unwinder. A separate assembly of pale blue plastic, attached to the instrument by the line: a reel with several tens of metres of suspension line wound on it, held by a retaining frame. At launch the sonde hangs directly beneath the balloon, and the line pays out in the air, gradually letting the instrument down to the full length of the suspension. The purpose is twofold: a short suspension is safer and easier to handle at ground launch, and in flight the instrument must be kept away from the balloon envelope so as not to enter its thermal and aerodynamic wake and corrupt the temperature and humidity readings.
Sensor boom. A thin cantilever of polished, slotted metal strip projects sideways and upward from the housing; the factory marking on it reads P.RS101B. The temperature and humidity sensors are carried at its end. The extension is essential: the sensors must be bathed in undisturbed oncoming airflow and kept as far as possible from the warm housing and electronics. The mirror surface of the boom and its narrow cross-section serve the same end — minimising solar heating and the thermal mass of the structure itself.
Antenna. A flexible insulated conductor — the transmitting antenna — emerges from the lower part of the housing.
Battery compartment. A hinged flap on the side wall carries a polarity diagram with "+" and "−" arrows; the end of the battery is visible through a window in the polystyrene.
Power source: the water-activated battery
The most unusual assembly in the instrument. Power is supplied by a reserve-type battery: in its unactivated state it contains no liquid electrolyte and therefore hardly self-discharges at all. The photographs show a flat stack of alternating yellow and dark plates clamped in a frame, with two terminals for a contact block; red and black conductors run from the battery to the board.
Before launch the battery is briefly immersed in water and then inserted into the sonde housing. The water acts as the electrolyte, the cell begins working immediately and delivers a nominal voltage of about 19 V for the whole of the flight — an hour and a half to two hours. After that it is not needed.
The solution looks archaic, but for a single-use instrument it is close to perfect. A dry battery keeps for years without loss of capacity, requires no maintenance or state-of-charge checking, is untroubled by storage at a polar station, tolerates frost and offers high energy density. Its one drawback is irreversibility: an activated cell cannot be returned to its original state, and its working life is limited. For an instrument that departs irretrievably in any case, this is not a drawback but an exact match to the task.
Beneath the cardboard shell and the polystyrene insert lies the electronics, arranged as two printed circuit boards set at right angles to one another.
The main board carries the factory marking PCB0018 Rev.M. On it are the measurement transducer, the encoding logic and the transmitter, the last of these enclosed in a one-piece metal shield soldered to the board around its perimeter. Assembly is predominantly surface-mount with some through-hole components; the packing density is moderate and the layout designed for cheap mass production.
The auxiliary board, PCB0521A, is mounted horizontally in the upper part of the structure. It carries the barometric sensor — an aneroid capsule secured by sprung tab clips above a shielding plate — with the elements of its transducer alongside. A thin wire bracket above the capsule acts as a travel stop.
The design idea underlying the whole series is that all three quantities are measured in the same way: through a change in electrical capacitance. The sensors have matched dynamic ranges, so a single transducer and a single reference capacitor for drift cancellation suffice for all of them. This sharply simplifies and cheapens the electronics, which for a disposable article is the governing consideration.
Pressure sensor: BAROCAP
It is a capacitive aneroid with a measuring range from 1060 hPa down to 3 hPa, a resolution of 0.1 hPa and an accuracy of 0.5 hPa, the latter expressed as the standard deviation of differences under the repeated calibration method. The external diameter of the capsule is 35.5 mm and the complete assembly weighs 5 g.
The sensor consists of a small aneroid capsule with the plates of a capacitive transducer housed inside it. The plates are supported on membranes of a special steel alloy; their supporting rods are fixed to the membranes by hermetic glass-to-metal seals. An inverted construction is used, giving maximum sensitivity at low pressures — that is, precisely where accuracy matters most, since with altitude the relative error in determining pressure translates directly into an error in geopotential height.
The transducer senses only the capacitance between the plates; stray capacitances between the plates and the membranes are excluded from the result because the membranes are grounded.
Temperature sensor: THERMOCAP
It is a capacitive bead type covering a measuring range from +60 °C to −90 °C, with a resolution of 0.1 °C and an accuracy of ±0.2 °C. The lag is less than 2.5 s in a 6 m/s flow at 1000 hPa.
The sensing element is based on dielectric ceramic materials whose temperature dependence can be controlled precisely through the choice of composition and processing parameters. Metal electrodes are formed on both sides of a tiny ceramic chip measuring 0.5 × 0.5 mm and 0.2 mm thick; the capacitance between them is a function of temperature.
For complete protection against moisture the element is hermetically sealed in a glass capsule 1.5 mm in diameter and 2.5 mm long, with two leads 0.4 mm in diameter. To exclude uncontrolled stray capacitances — which could arise, for instance, from water droplets on the glass — a grounded thin-film aluminium coating is deposited on the capsule and the leads. That coating also has favourable radiative properties and reduces the radiation error of the measurement: at a pressure of 10 hPa and a solar elevation of 45° it does not exceed 2 °C. An insulating layer on the leads prevents short circuits.
Humidity sensor: HUMICAP
It is a thin-film capacitor covering the full range from 0 to 100 % relative humidity, with a resolution of 1 % and an accuracy of ±2 % expressed as the standard deviation of differences under the repeated calibration method. The lag is 1 s in a 6 m/s flow at 1000 hPa and +20 °C, and the element measures 4 × 4 × 0.2 mm.
The sensor is a thin-film capacitor with a polymer dielectric about 1 micron thick. Its capacitance is determined by the amount of water absorbed in the dielectric layer: the polymer reversibly sorbs water vapour, changing its dielectric permittivity.
The element is fabricated by thin-film technology similar to that used generally in microelectronics. Its small size gives it a small thermal mass, so that the sensor follows the temperature of the surrounding air quickly and closely — a condition without which true values of relative humidity in the atmosphere cannot be obtained. Its further merits are fast response, good linearity, low hysteresis and a small temperature coefficient. It remains reliable at low temperatures at least down to the −60 °C level.
HUMICAP technology was introduced by Vaisala in 1973 and was the world's first thin-film capacitive humidity sensor, displacing the hair hygrometer from meteorological practice. Developed initially for the radiosonde, the sensor unexpectedly attracted the interest of industrial customers and gave rise to a separate line of business for the company.
The Vaisala RS80 radiosonde is a finished specimen of engineering thought subordinated to a single requirement: to obtain a metrologically sound vertical section of the atmosphere by means of an instrument that will certainly be lost after its first use. Hence the cardboard housing and polystyrene insert in place of a cast chassis, the electrolyte-free battery in place of an accumulator, and a single capacitive principle for three different physical quantities in place of three specialised measurement chains. Each of these decisions taken singly looks like a simplification; together they form a system that delivered an accuracy of 0.5 hPa in pressure and 0.2 °C in temperature at an instrument mass of 220 grams and at a cost that allowed national weather services to launch such instruments twice a day for decades.
The historical significance of the specimen extends beyond measuring technology as such. Instruments of the RS80 series generated a substantial part of the world's upper-air observational record for the 1980s to the 2000s, while the systematic bias found in their humidity readings became the subject of a literature of its own and compelled the meteorological community to take seriously the problem of homogeneity in long observational series across changes of instrument type. In this respect the specimen is doubly interesting: as an example of mass-produced measuring equipment from the late analogue period, and as material evidence of an episode that influenced the methodology of climate research.
The specimen presented retains its factory hermetic packaging, the full set of accessories — unwinder, sensor boom, unactivated battery — and the original calibration strip with its individual coefficients, which makes it a complete and documented example of its type.