Last update: March 30, 2012 08:41:55 AM E-mail Print


The Satellite Ground Station at Grootfontein


L.J.J. BOTHA, Animal Production,

Grootfontein College of Agriculture,




THE weather satellite ground station at the Grootfontein College of Agriculture is the result of a project of the local Scientific Society. This project was initiated in 1974 in an attempt to investigate the possibility of receiving satellite signals. Results with relatively unsophisticated apparatus showed that these radio signals could be tracked very well. Since very strong signals were received from the weather satellites ESSA 8 and NOAA 3, it was decided to develop the necessary apparatus for the demodulation of picture data from weather satellites.



The world's first weather satellite, TIROS 1, was launched into its orbit in April-1960 from Cape Canaveral, Florida USA. The TIROS 1 type spacecraft evolved to the TIROS Operational Satellite (TOS), which was first launched on February 3,1966. These "wheel-type" TOS satellites were designed to fill the need for a relatively low cost weather satellite system to provide worldwide photographic coverage of the earth and its atmosphere. In effect, this satellite rolls on its rim during orbit as a wheel would roll along the earth's surface.

The improved TIROS Operational Satellite, ITOS 1, launched on January 23,1970, ushered in the second-generation polar-orbiting satellite system. The third generation polar-orbiting system TIROS-N was initiated on October 13, 1978. This satellite contained advanced sensors to provide improved temperature soundings and multi-channel images. The latest satellites in this series are NOAA 6 and 7, and are at present fully operational. Several Russian METEOR satellites transmit good quality weather pictures in the so-called "weather satellite band" on 137 MHz.

The NOAA and METEOR satellites are known as polar orbiters, which mean that they revolve around the earth in approximately 102 minutes. For tracking purposes a steerable antenna, which follows the satellite for periods of up to 15 minutes as it passes over the tracking station, is called for.

More recent developments are the geostationary satellites, like METEOSAT, which is a European meteorological satellite orbiting at a height of some 36 000km above the equator at 0° longitude. Since the satellite turns around the earth at exactly the same angular velocity as the earth turns around its axis, it appears to be stationary in relation to the surface of the earth. This means that a fixed antenna can be used for receiving weather pictures.

The American NOAA satellites are in near polar sun synchronous orbits at an average altitude of 850km. Orbital parameters are such that morning passes occur at 07h30, while afternoon orbits are at 15h00 local time.

The METEOSAT satellite takes pictures of the earth in visual, infrared, and water vapour light every half hour, and then transmits these to the European Space Agency (ESA) Operation Center (ESOC) at Darmstad, Germany. After optical correction and dissemination at the ESOC, the pictures are retransmitted by METEOSAT to users in Europe and Africa.

The tracking of both polar-orbiting and geostationary weather satellites necessitates very high frequency (VHF), as well as ultra high frequency (UHF) receiving apparatus.

The VHF and UHF equipment, which was developed at Grootfontein, will be discussed under separate headings.



Radio and space science enthusiasts have been designing private satellite receiving stations since the early sixties and some interesting electronic circuits have evolved. 1, 2, 8, 9, 10 The most basic system developed for the display of satellite photographs consists of an antenna, preamplifier, receiver, demodulator, oscilloscope and camera (Fig. 1).



The apparatus depicted in Fig. 1 constituted the prototype on which all subsequent developments at Grootfontein were based.

A 22 element crossed yagi-type antenna was designed and constructed, as various polarization effects cause severe fading of satellite signals. Although a high quality radio receiver (Eddystone 990R) was used, noise on the satellite signals, especially on low passes near the horizon, was found to be excessive. This problem was solved by constructing a preamplifier3 and inserting it between the antenna and the radio receiver. Suitable electronic circuits were constructed to operate the vertical and horizontal scanning of a Tektronix 555 oscilloscope. The resulting line-by-line signal from the satellite was photographed from the screen of the oscilloscope by means of a 35mm camera mounted in a light-tight enclosure fitted to the cathode ray tube of the oscilloscope. This system has some inherent disadvantages, such as low resolution, as a result of the relatively large size of the light spot compared to the diameter of the cathode ray tube. These preliminary results however proved that weather satellite pictures could be received successfully with modest apparatus.

Since the low resolution of the photographs was attributed to the small diameter of the cathode ray tube, it was decided to use a TV -tube, which has a much greater spot to tube diameter ratio. A black and white portable TV set was converted with locally developed circuits to a 4 Hz scanning rate. Vertical framed rive was achieved by the use of a voltage integrator.

Photographs received with this system were of a high quality and resolution (photograph 1), but photographic processing was very time-consuming. An attempt was made to decrease the time factor by using a Polaroid camera, but since the photographic format was unpractically small, the idea was abandoned.




During 1981 it was decided to upgrade the existing weather satellite equipment to receive pictures from METEOSAT 2. By doing this, several advantages would be achieved, viz:

(a) A fixed antenna could be used, as the satellite remains stationary above a certain point on the surface of the earth.

(b) Access to more data, as METEOSAT also transmits water vapour and cloudtop height diagrams.


These advantages outweigh the fact that the complexity of the system would be considerably increased.

The complete METEOSAT receiving system consists of a parabolic reflector antenna, converter, radio receiver, electronics and a facsimile machine (Fig. 2)




At S-band microwave frequencies, parabolic antennas become practical and offer large gains in small physical size. An unorthodox approach was employed in the construction of the dish. This involved cantilevered beams, which were end-loaded to form the parabolic surface. For small deflections a beam supported at the one end will assume an almost perfect parabolic curve when a force is applied to the other end. This principle is used in the design of collapsible stressed skin dishes. A good size for a reflector suitable for weather facsimile (WE FAX) reception is in the 1,25 to 2m size range, resulting in a gain of 24dB to 27dB at 1691 MHz. Consequently, a parabolic reflector was designed with a diameter of 1,5m and an F/D of 0,3. A novel method was conceived for the construction and only easily obtainable materials were utilized. A wooden disk with a diameter of 12,5cm and 5cm thick was used as the central hub of the antenna. Eighteen equally spaced holes, 7mm in diameter, were drilled around the circumference of the disk and wooden dowels (carefully selected for straightness), 75cm in length, were glued into the holes. A vertical hole was drilled into the centre of the hub to accept a wooden rod 25mm in diameter and 50cm in length. The ends of the dowels were drawn towards the central mast by 18 equal lengths of strong nylon cord, so that the depth of the dish thus formed was 31,25cm (Fig. 3).



The frame was covered with fine aluminium gauze to form the reflecting surface of the antenna. After securing the gauze with thin nylon thread and smoothing it carefully, several layers of fibreglass were applied and allowed to cure. A light metal structure was used to mount the antenna and the dish was aimed at METEOSAT by calculating the elevation and azimuth by means of trigonometry4. The mounting co-ordinates can also be calculated by means of a very convenient locator charts. For Middelburg Cape, the METEOSAT co-ordinates are elevation 44,2 and azimuth 314,0 degrees.



Later, when a professional 4m diameter parabolic reflector became available, a supporting structure was designed and the reflector erected. It is interesting to note that the much smaller homemade parabolic compared favourably with this one.



Radio signals emanating from the satellite are collected by the parabolic antenna and focused onto a cylindrical feedhorn. The horn was made from a 750g coffee tin6 and fitted onto an adjustable mounting7 to facilitate tuning for maximum signal strength. Energy collected by the horn is conveyed to the converter by means of a very low loss RG 142 B/U semi-rigid feedline.



A microcomm Model RX-1691 S-band converter is used to change the microwave signal to 137,5MHz. This signal is then of a suitable frequency for detection on the Eddystone VHF receiver. The converter was mounted at the parabolic antenna to minimize the length of RG 142 B/U cable carrying the microwave signal, as losses at these frequencies are excessive.



Signals from the radio receiver must be electronically processed before being reproduced by a suitable display unit. As photographic processing was very time-consuming with the television system, it was replaced with a photographic facsimile machine, which is most probably the best approach in terms of image quality, while processing time is reduced to a matter of minutes. With a facsimile machine the image is printed directly on photographic paper using a modulated light source. The principle of this type of machine is not new, as the international press has been using it for many years for the transmission of photographs by landline and radio. The picture is produced as illustrated in Fig. 4.



A sheet of photographic paper is wrapped around a drum, which is rotated by a motor. A light source and lens arrangement (glow modulator tube) is mounted near the drum in such a way that a very small spot is focused onto the paper. The drum is mounted on a traverse platform and is slowly moved parallel to the light source by a threaded lead screw driven by a small motor. A fine helix is described by the spot on the photographic paper. The brightness of the light source is varied in accordance with the scene scanned by the satellite. After picture transmission, the paper is removed from the drum and developed.

The two motors as well as the modulation of the light beam are electronically synchronized. As photosensitive paper is used, a darkroom environment is needed while printing a picture.

Since surplus commercial facsimile machines are expensive and difficult to procure, it was decided to design and build a machine to suit the present requirements.



One of the most critical components of a facsimile machine is the drum motor. Home-built versions usually make use of synchronous motors. The problems associated with these motors, such as low torque and starting difficulties under certain conditions, however, led to the decision to deviate from standard practice and examine the possibilities of using a high quality stepper motor. A SLO-SYN Type MO61FC08 motor with its associated electronics was procured. This motor has a torque, which is more than adequate for the present application. The motor rotates the drum once for every 200 pulses applied and is coupled to the drum shaft by means of a short length of high-pressure rubber laboratory tubing.



The dimensions of the drum must be such that the eventual picture has an aspect ratio (height to width) of 1:1. The drum was constructed from an aluminium tube, 5,2cm in diameter and 17,0cm in length. An aluminium shaft was welded to the drum and lathed to accept standard 10mm roller bearings. The drum was also lathed to remove any surface inaccuracies, which could have a defocusing effect on the light source.



Many articles have been written on the design of suitable drum carriages. Some of these were tried, with mediocre results. The problem is that 800 lines of picture information must be accommodated on a drum length of 170mm. Each line can thus only occupy 0,21 mm. Any erratic movement of the drum carriage will be manifested as interference lines across the picture. The problem was eventually satisfactorily solved by modifying a redundant typewriter carriage, which runs on ball bearings. The drum and its motor were mounted on a wooden platform and are moved in front of the light source by means of a 16 tpi lead screw coupled to a 250rpm synchronous motor. The rotation of this motor is reduced to the required speed by an 8,33:1 gearbox.



The electronics necessary to operate the facsimile machine consist of two separate sections, viz:

(a) The light source electronics

(b) The drum motor circuits.


The complete circuit diagram is represented in Fig. 5.



Satellite video signals from the receiver or tape recorder are processed in this circuit, so as to vary the brightness of the glow modulator tube in accordance with the scene scanned by the satellite. These signals are stepped up to a suitable voltage by the LT-790 transformer. The signal to noise ratio of the system is improved by tuning the secondary winding of the transformer to the satellite subcarrier frequency of 2400 Hz by means of capacitor C1. The diode bridge D1-4 allows full wave rectification of the carrier. Maximum lamp current (black level of the image) is set by the 10K ohm potentiometer. The expensive type 1130 glow modulator tube is protected by a 50 mA fuse in the collector of the EGG 124 transistor. The lamp intensity, which is proportional to its current, is thus varied in accordance with the video input and is monitored by means of a 0-50 mA meter.



The purpose of this circuit is to lock the motor drive signal to the satellite subcarrier. This is achieved by applying the input signal from the satellite to an LM 565 phase locked loop (PLL) integrated circuit (IC). The 2400Hz output of the PLL is buffered by a Schmitt Trigger before being digitally divided by three. The resultant 800Hz signal is applied to a type STM 103 translator module which in turn drives the stepper motor at the required 4rpm.



Without phasing, the picture margin could be anywhere on the photographic paper. The margin must therefore coincide with the edge of the paper. Trigger pulses for the oscilloscope, used for visual adjustment of the margin, are obtained from a small magnet glued to the drum at the point where the paper edges meet. Each time this magnet passes a tape recorder head, an electrical current is induced and stepped up by the drum pulse amplifier11. The resulting pulses are used to trigger a Tektronix 555 oscilloscope. Video from J1 is applied to the Y -input of the scope. Each line of picture information can thus be viewed on the screen and the margin is shifted to the correct position by means of switch S1.



Several different types, grades and surfaces of photographic paper were tried and eventually a grade 1 matt surface was found to be most suitable for the present application. Plastic-coated paper is more convenient as no hot glazing is required. Matching photographic processing chemicals are used.



Since 1981 good quality satellite images have been received as required from METEOSAT 2 at the Grootfontein College of Agriculture. In addition, cloudtop heights, water vapour diagrams, as well as administrative data are available (photograph 4).




After several prototypes, a satellite receiving station capable of printing good quality weather pictures has evolved. Cost was kept to a minimum by using existing apparatus, improvising and home-building some of the more expensive pieces of equipment.



A special word of thanks to Mr. A.J. du Plessis for his valuable advice with regard to the photographic processing and to all other persons who assisted me during the development of this system.



1. KENNEDY, G.R. Weather satellite ground station. Wireless World, November 1974.

2. OSBORNE, J.M, Receiving weather pictures from satellites. Wireless World, October 1971.

3. ARRL. The Radio Amateurs' Handbook, 55th edition. Improving FM receiver performance.

4. TAGGART, R.E., Microcomputers and your satellite station. 73 Magazine, February 1980.

5. OBEREM, G.E., Locator for geostationary satellites, Radio ZS, July 1981.

6. NESS - NOAA publication, prepared by N.M. Seese., Ground stations to receive GOES WEFAX.

7. TAGGART, R.E., Feed horn mounting made easy. 73 Magazine, July 1979.

8. STRONG, C.L., The Amateur Scientist, Scientific American Magazine, February 1974.

9. KENNEDY, G.R., Weather satellite picture facsimile machine. Wireless World, December 1976.

10. TAGGART, R.E., Direct Printing FAX. 73 Magazine, November 1980.

11. KENNEDY, G.R., Weather Satellite Picture facsimile machine - 3. Wireless World, February 1977.



Karoo Agric 2 (4), 18-23