Swedish Space Corporation
25 years, 1972-1997

Part 3: The second decade: 1982-1992
By Stefan Zenker

Colour sketches by Lasse Eklund
 

Table of contents

Go to Part 1: Ancient history - 1972: How it all began
Goto Part 2: The first decade: 1972 - 1982: The sky is no limit
Part 3: The second decade: 1982 - 1992: Making dreams come true
Go to Part 4: New horizons: 1992 - 1997: Change in all things is sweet


The second decade: 1982 - 1992: Making dreams come true

Esrange: A big boost
Sounding rockets forever!
A ballooning activity
A rising sun over Esrange
The wolf man
Mariane
Don’t touch my antenna!
The equator? From Esrange? No kidding!
Space night
MAXUS
Salmijärvi
Remote Sensing: A tough business
High spirits
Shifting sands
Rolling hills
PROMIS - doing something about the weather
EBBA - image processing for the masses
MIMA - the infinite table
Peace is our profession
Mapping the Philippines - an early success
Digital Terrain Models - the battle of the bulge
A digital satellite map of Sweden
10 years of hard labour in Solna
The Maritime Surveillance System
Krylbo - a train wreck
New faces
Sobering up
Satellitbild: The pioneering years
Now, here is the plan...
And the winner is...
Pomp and circumstance
Have another cigar!
Finally in business!
Teething problems
Your mission, should you decide to accept it...
Selling bits to BITS

Space Science: Going into orbit
The challenge of Viking
Doubts
Triumph
Slither
Microgravity
It’s a GAS!
PIROG
Freja

Tele-X: A dream come true
The SSC project team is formed
Playing for high stakes
Talk to my lawyer!
The bear’s hug
Rolling the dice
Early customers
Go to Part 4: New horizons: 1992 - 1997: Change in all things is sweet


The second decade: 1982 - 1992

Making dreams come true

Esrange: A big boost

Sounding rockets forever!

Contrary to what many had believed when Esrange was "nationalised" in 1972, scientific research using sounding rockets showed no signs of decline by the end of the first decade. Of course, the growing capabilities of satellites attracted many scientific groups. On the other hand, satellite projects also took longer from inception to launch into orbit, and they became more expensive. As a result, fewer satellite missions were flown than had been projected, and fewer flight opportunities than predicted became available.

Experiments using sounding rockets had the great virtue of short turn-around times - something that was very desirable to research groups that included doctoral students.

At Esrange, new research fields were added to the traditional atmospheric and auroral research. IR astronomy was one; microgravity research was another. The anticipated availability of the Space Shuttle and the International Space Station (which was approved by President Reagan as early as 1984!) to do research in orbit for extended periods only accentuated the interest of "pathfinder" activities using sounding rockets.

During this period, 15 sounding rockets were launched for the Swedish Space Science programme, utilising Nike-Orions and Black Brant combinations in upper atmospheric and magnetospheric research, with apogees from 100 to 500 km. During the same period seven sounding rockets for the Swedish microgravity research programme were launched, including two MAXUS.

A ballooning activity

In 1985 the first high altitude scientific balloon within the Swedish Space Science Programme was launched from Esrange. The Pointing Infra-Red Observing Gondola (PIROG) aimed at studies of star formation and carried a telescope to an altitude of more than 30 km. The gondola weighed 400 kg and was as complex as a small satellite, with a cold gas system and video cameras for pointing of the telescope, and telemetry and telecommand systems for communication with the computerised ground station.

After some struggle during the first launches, good scientific data was obtained during launches 4-6. During flight no. 5 the separation system by radio command failed. The balloon started a long journey up over the Arctic sea, turned around and approached the Murmansk area.

Unfortunately, the Soviet authorities, who may have suspected that its purpose was not purely scientific, decided that the balloon should be shot down. A Soviet fighter did the job. Fortunately, the payload parachute deployed, and the gondola landed close to the Finnish border. The Soviet authorities claimed that the gondola had disappeared in a swamp, but science colleagues in the area took a picture of the gondola in good shape in a forest and sent it to Sweden. With this evidence, high-level negotiations led to the return to Sweden of the gondola. It had been stripped of high-tech parts by the Soviets, however.

In the mid-1980s, a big surge of interest in the use of balloons for atmospheric research came when the ozone holes over the poles were discovered.

A rising sun over Esrange

Already in the early 1970s, the Japanese space organisation ISAS visited Esrange to discuss business and co-operation. However it took almost ten years until the first contract was signed in 1980. A balloon campaign was carried out for Prof. Ejiri, National Institute for Polar Research.

This became the start of a very fruitful co-operation. We performed services for many different Japanese space institutions, including a number of balloon campaigns.

In 1983, we were contracted by ISAS to receive data from the EXOS-C satellite. This was followed by a contract in 1989 to receive data from the EXOS-D satellite. This satellite is still in operation, and the contract has been prolonged on a yearly basis. ISAS intends to receive data from the satellite during a complete sun cycle until the year 2000 if it continues to work well.

NASDA (the National Space Development Agency of Japan) placed a TT&C station at Esrange in 1990 to support Japanese Remote Sensing satellites. SSC operates and maintains the station under contract to NASDA.

The wolf man

Early in this period, a fascinating project briefly touched Esrange. A private German project to build an inexpensive satellite launcher was announced - OTRAG. The rocket was going to be assembled from a large number of modular boosters. Testing would start with the launch of a single rocket, burning red fuming nitric acid and diesel oil. Gradually more powerful combinations would be assembled and tested.

The flamboyant director of the project visited Esrange in his search for a suitable launch site. He came in his LearJet and was dressed in an impressive wolf fur coat. At the Ferrum hotel he asked to buy the piano! Another small detail that impressed us was that ordinary car windscreen wiper motors were going to be used in the rocket design. What could be more reliable in relation to cost?

Unfortunately, in the end the launch site was established in Zaïre, where an enormous land area was made available. Only one or two launches to low altitude of the basic rocket module took place. There was a lingering suspicion that the whole project might have been a cover to acquire some very valuable real estate in Zaïre... Then the launching activities were suddenly moved to Khadaffi’s Libya. A number of launches were performed from there, until an article in the German magazine Stern revealed the operation. Of course, there was concern that Khadaffi might get hold of a rocket for military purposes. But OTRAG also survived this affair.

In 1983 the next launch took place - this time from Esrange. The rocket flew normally for about ten seconds, when it suddenly exploded. The commission of inquiry found out that the reason was a payload problem. An opening had been made in the nose cone in front of the lens of a photometer. There was no glass cover. When the rocket accelerated, the rectangular hole acted just like an organ pipe. This led to vibrations which caused the explosion.

The failure became the final straw for the investors, and the OTRAG company was dissolved. - So far, OTRAG is the one and only liquid rocket we have launched from Esrange!
 

Mariane

In 1985, a scheme to turn Esrange into a satellite launch site was briefly considered. At first, various configurations based on the old Scout, the Conestoga, and on Chinese and Japanese rockets were considered. Volvo Flygmotor flirted with the idea of promoting the Conestoga.

Gradually, the interest focussed on an all-European combination. Volvo Flygmotor and SSC visited ESA, CNES, Arianespace, Aerospatiale, SEP, BPD and MBB. The proposal engendered a lot of interest. The most promising configuration studied was based on a scaled-down version of the Ariane rocket, a mini-Ariane or "Mariane". The second stage of the Ariane would become the first stage of Mariane. The second stage would be a French military solid-fuel rocket named P6, and the third stage would be the Italian IRIS rocket. A phase A study was endorsed by the interested parties.

If Esrange was going to be used as a launch site, the rocket would have to cross Norwegian territory on its way into orbit. This was actually the third time that the issue of overflights of Norway from Esrange had been raised. In the 1960s, the Norwegians had opposed ESRO launches of the Dragon rocket into the Atlantic. At that time they, unlike Sweden, claimed national jurisdiction over their country all the way into the vast reaches of the universe! In the mid-1970s, a similar scheme was blocked when the Norwegian authorities withheld their approval on political grounds after conceding that the launches met the agreed safety criteria.

This time, the problem turned out not to be Norwegian opposition but rather the difficulty of organising the venture and finding a market that could sustain the Mariane development effort. The project was put on ice.

Don’t touch my antenna!

The ELS antenna for Landsat-2 and -3 was upgraded to X-band for Landsat-4, launched in 1982. The reception of Landsat data was performed on behalf of ESA. In the same period, Esrange found a new customer for satellite support: CNES, which was making plans for its SPOT satellite, and which was ready to make Esrange one of its two principal receiving stations.

In the interest of economy, we wanted to use the same antenna to serve both customers. This quickly created a difficult situation. ESA demanded an absolute priority for antenna use, while CNES was equally adamant that it should have an unconditional priority. - The problem was not a substantial one, but a lot of prestige was involved. Somehow, a face-saving formula was found that both customers could accept. - A second X-band antenna was soon installed, justified by the projected growing traffic over Esrange, and the problem disappeared. Similarly, the office building on "the Hill" at Esrange where the Landsat station is located, was expanded.

The equator? From Esrange? No kidding!

Once the Tele-X project was underway, the sensitive question of where to locate the ground control station arose. Our principal customer, the Swedish Telecommunications Administration, favoured placing it at their ground terminal in Ågesta near Stockholm. SSC, as always, was looking for additional tasks for Esrange.

After some heated discussion and lobbying, it was decided to split the roles, so that the actual traffic carried by the satellite would be handled by Ågesta, while the vital satellite control functions were executed at Esrange. Of course, the solid experience of Esrange in TT&C was an important factor in the decision.

The construction work, involving the extension of the main building, and the procurement and installation of a dedicated antenna, started in 1984.

Space night

In many ways, 22 February, 1986, represented the culmination of the "pioneering years" of SSC. That night, we rolled the dice and launched our own Viking satellite together with our "adopted" SPOT-1 satellite on an Ariane rocket. - If the basket had been any larger, would we have put our Tele-X egg in it too??

With remarkable aplomb, our incoming Managing Director Lennart Lübeck chose that date to assume his new office, not knowing whether he headed for Esrange to deliver a victory speech or to pick up the pieces after disaster. After all, the previous Ariane launch had ended in failure (and the second one following the Viking launch was another failure). And the launch itself was only one of the many things that could go wrong.

In the event the launch was a success, the satellites were in excellent health, and Esrange had found another area of activity. It gave support to the launch itself, and to the TT&C and data reception for the SPOT satellite, and was wholly responsible for the TT&C and data reception for Viking.

Big festivities were organised at Esrange and at Satellitbild in Kiruna. In Stockholm some two hundred people gathered in the middle of the night to witness the launch from our Solna office.

MAXUS

In the mid 1980s, several launch vehicles were grounded after failures (Space Shuttle, Ariane, Delta). The microgravity research community desperately looked for experiment flight opportunities in preparation for the Space Station Columbus. SSC proposed a Long Duration Sounding Rocket to ESA for these demands and received a study contract. The result was an agreement in 1988 between ERNO and SSC for the development of a single-stage rocket based on the Castor strap-on boosters for the Delta 2 launch vehicle. A large MASER/TEXUS rocket became MAXUS and could take a 700 kg payload to 700 km, giving 14 minutes of microgravity.

ERNO was responsible for the procurement of the motor and service systems. SSC was responsible for the development of the rocket systems (guidance & control by Saab, interstage adapter including the separation and thrust termination, telemetry & tracking unit, and fins) and new installations at Esrange. Both parties were responsible for experiment modules contracted by ESA and national agencies. ERNO and SSC shared the investments (and future revenues). SSC invested a total of 56 MSEK. ESA and the German Ministry for Research and Technology committed to contract ERNO and SSC for 2-4 launches per year during a 5 year period.

A MAXUS-test 3-stage Skylark 12 rocket was launched successfully in November 1990 to an altitude of 534 km and with an impact point 5 km from that predicted. The objectives of the test flight were to verify guidance and control systems and safety operations systems onboard and on the ground. The first MAXUS launch took place in May 1991. As a result of bad workmanship in the Thrust Vector Control System delivered by Thiokol, the rocket disintegrated at 40 km altitude and fell in pieces to the ground. The reflight in November 1992, MAXUS 1B, was successful and the rocket reached 716 km. A second successful flight occurred in 1995.

A MAXUS 3 has been decided by ESA for launch in late 1998, but the future for the MAXUS programme beyond that point is uncertain.

Salmijärvi

When ESA prepared its first Earth Observation satellite project, ERS, it was quickly discovered that the existing network of ground stations was inadequate to serve the mission. A station at high geographic latitude was needed - both for TT&C and data reception.

Sweden of course offered Esrange for this role. Only a modest amount of lobbying was needed to win the agreement of the ESA executive - but there were some conditions. Sweden had to pay for the infrastructure, including buildings. In addition, any risk of radio interference must be removed by the physical separation of the new site from Esrange. It soon emerged that this involved more than just a physical separation - ESA wanted to "run its own shop" with a clear, separate identity from Esrange. This caused no difficulty from Sweden’s point of view; in fact it might even be advantageous to invite ESA to be the owner of the new facility, and not just a tenant.

Construction on the new site, just 6 km from Esrange, started in 1987. The official inauguration in the presence of the King and the Queen took place in September 1990. On that occasion, the author felt like the director of a theatre play, as he had drafted or reviewed the speeches of the King, the Under-Secretary representing the Government, the Director General of ESA, and the Director General of SBSA. Only the Mayor of Kiruna and the Managing Director of SSC were "loose cannons" from his point of view. - The newly appointed ESA manager of Salmijärvi, Björn Eriksson, discharged his first official duty brilliantly by complying with royal etiquette to everybody’s full satisfaction, including the provision of a table cloth of the correct size from Skövde!

25 SSC employees had found a new home!

Remote Sensing: A Tough Business

High spirits

Around our 10th anniversary in 1982, those responsible for developing remote sensing as a business area were brimming with optimism. We had in short order established the Landsat station at Esrange, set up an image processing system in Solna, and created Satellitbild. We had developed the inexpensive EBBA image analysis system as a spin-off from our airborne maritime system development and sold a dozen of them. Technologically, we were miles ahead of any competition. We were "on a roll"!

Clearly the time was ripe to start cashing in on our foresight by offering our services to an eagerly waiting world - or so it seemed.

Shifting sands

An early opportunity presented itself in Saudi Arabia. The country had ambitious plans to set up an Earth Observation satellite station. SSC hastened to offer its services as a consultant.

Our efforts seemed to pay off. A Saudi official visited us and was taken on a trip to Kiruna in a rented business jet by Fredrik Engström, who had meticulously read up on Saudi culture and understood the importance of showing hospitality. We were invited to Saudi Arabia in 1982 and reconnoitered suitable sites for the receiving station. We offered our support in setting up image processing services.

Unfortunately, shifts in the Saudi administration negated much of our efforts to build close personal relationships with decision makers. The new officials decided to "play it safe" and turned to the Americans for technical support.

We got some follow-on business worth 12 MSEK in 1985, however, when a Norwegian company, SysScan delivered a digital mapping system to Saudi Arabia. SSC was responsible for the remote sensing part of the system. SSC sent staff for operations and training over two years.

Rolling hills

We had better luck in Pakistan. Again, we offered our services as consultants in the setting up of a satellite station. Our offer was accepted. Some of our most skilled engineers spent several months at the station site in the Pakistan hills near Islamabad. We also sold our EBBA-II image analysis system.

Despite these successes, and some very satisfied customers, the idea to sell our consulting services in setting up satellite stations had only yielded a modest profit. This business had to carry a lot of overhead in the form of attention from our top management, and the customers expected our top engineers to be constantly available, something that was difficult to reconcile with their "normal" duties.

We decided that in the future we would not let this kind of business crowd out more pressing matters, such as turning Satellitbild into a success, and attracting new customers to Esrange.

PROMIS - doing something about the weather

In the early 1980s, the Swedish meteorological office, SMHI, decided to use modern technology to develop a next-generation weather service for the 1990s, PROMIS-90. SSC played an important role in that effort by preparing a report on how space technology could support the meteorological services, and by endorsing the industrial policy aspects of PROMIS-90. We also developed software for satellite image processing. Before long, TV viewers could enjoy attractive satellite colour images with the daily weather forecast.

EBBA - image processing for the masses

Based on our experience in developing a presentation system for maritime surveillance, and in image analysis on our newly acquired IAS system, in 1981 we completed development of a simple interactive image processing system - EBBA - hosted on the tabletop microcomputers available at the time: ABC-80 and Metric-85. A number of these tiny systems were made available to Swedish research groups and stimulated their interest in remote sensing.

In 1984 we introduced EBBA-II. The image planes had now been expanded to 512 * 512 pixels (picture elements). The IBM PC and DEC VAX were chosen as hosts.

EBBA-GIS was introduced in 1986. Its development cost 3,2 MSEK. We had now moved up to 1024*1024 pixels. Some fairly advanced features were included: Raster-to-vector conversion; geometrical corrections using automatic correlation and resampling techniques; the ability to "flicker" between two registered colour images to more easily detect changes; the ability to roam a "window" over a satellite image, showing an underlying map; a pop-up menu and help system. An American GIS system supplied by a company in North Carolina was bundled with the system.

EBBA-GIS was a brilliant technical achievement. It was developed and sold at a fraction of the cost of equivalent systems. Yet it failed to reach its sales targets and was discontinued at the end of 1987. (It continued to form an important part of the Maritime Surveillance System.) At the time, we estimated that we had a technical lead over the competition of two years. In retrospect it may have been closer to four. So what happened?

In essence, we failed to identify and understand EBBA’s market. It was originally designed as a very inexpensive system to put in the hands of remote sensing scientists. By raising its price and targeting professional customers, we put it out of reach of many "small" users. On the other hand, the size of the professional market was very limited at the time, so volume was low. We sold fewer EBBA-GIS than EBBA-II. Moreover, the professional market required a professional level of service and support.

In the end EBBA was "axed" by managers who perceived it as mainly a special-purpose processor with an unpromising future.

MIMA - the infinite table

In 1982, SSC started a pioneering project with IMTEC (later Teragon), a company started as a spin-off from the labs of the technical university in Linköping. They had developed a very advanced (for its time) pipeline processor to manipulate large images. Their business idea was to apply this new technology to newspaper production. Text and images would be edited electronically. The type-setter’s profession was about to be eliminated!

SSC saw an opportunity to combine self-interest with support to Swedish industry. A joint development project was set up, "MIMA" (named after the intelligent computer in Harry Martinsson’s epic poem "Aniara"). It was financed by the Swedish National Space Board (SNSB). Its purpose was to apply IMTEC’s image processing technology to remote sensing. SSC would develop applications software for IMTEC’s VAX-hosted platform.

The core of MIMA was the concept of the "infinite" table. No more would we be constrained by any particular image format. Lightning-fast handling and processing of any image would be possible, within the constraints of available disk space.

IMTEC’s priority was the newspaper application. They quickly ran into problems in this area. The newspapers were understandably nervous about committing themselves to a new technology before it had been thoroughly tested in an operational environment, and the unions resisted its introduction. The net result was that the availability of the software tools needed for SSC’s development was delayed by two years.

Our development team telescoped their development of applications software into a short and hectic year. Several MIMA systems were sold, but basically the window-of-opportunity had been missed. The market for image processing systems gradually moved away from special-purpose processors, as standard platforms became more powerful.

Peace is our profession

Around 1985, SSC spotted an opportunity to offer satellite technology in connection with disarmament treaty verification. A camp of non-aligned states sought to put political pressure on the superpowers to agree to disarmament proposals. The Swedish Government under Olof Palme was very active in this group, which also included Canada, India, Mexico and Indonesia. The question arose of how compliance with any disarmament treaties would be monitored.

Olof Palme was also personally engaged in a peace initiative in the war between Iran and Iraq which raged at the time. Here too, an independent verification system was desired.

A preliminary study of the possibilities was carried out in 1986-87 by the Swedish Defence Research Establishment (FOA) with the support of SSC. The Ministry of Foreign Affairs decided that the study should be brought forward into a "conceptual phase". A technical study was included. It was performed by SSC under the project name "Tellus" (which could also be read as "Tell us"...). The study was managed by Per Zetterquist, with the participation of Mats Rosengren as remote sensing expert. It was completed in 1988.

The study showed that a SPOT class satellite with an optical 2-metre resolution instrument would do the job under certain assumptions. If phase A was started in 1989, the satellite could be launched in 1995.

The political interest of the proposal waned, however. A Swedish proposal for a de-militarised zone through central Europe lost its relevance after the fall of the Berlin Wall. The idea of a "neutral" group of countries monitoring compliance with disarmament agreements lost its appeal as new constellations and security concerns emerged.

The basic idea of Swedish participation in a verification satellite system might yet be realised some time in the future, however!

Mapping the Philippines - an early success

The first major project using SPOT data was started in 1987. It involved the mapping of the land cover of the entire Philippines in just over a year. Some 200 SPOT images were received and processed in Kiruna, and analysed and interpreted in Kiruna and Solna. Extensive investigations in the field were made by our project team. - The project was led by Hans Rasch at SSC in Solna.

The project attracted a lot of attention internationally. It became the model for a lot of our projects in developing countries during subsequent years: financing by the World Bank and BITS; co-operation with the local mapping authority, including joint field investigations and technology transfer.

Later, we found out - to our regret - that it was far from typical in another respect. Most projects in developing countries require several years of preparatory work before they are approved and can be executed.

The project also provided a focus for the co-operation between the Solna and Kiruna parts of the Remote Sensing Division. Some of the expertise and experience of the pioneers in Solna was made available to the Kiruna team and resulted in improved production methods and quality, the development of satellite image maps, and later topographic image maps.

Digital Terrain Models - the battle of the bulge

In 1986-87 we developed an automatic system to derive terrain height figures from parallax measurements in SPOT stereoscopic data. It was based on the MIMA system and its Teragon processor hosted on a DEC VAX computer. The system was developed by SSC, taking as its starting point a unique algorithm devised by Dan Rosenholm.

The new tool was installed at Satellitbild in Kiruna in 1988.

In 1986, the Swedish Land Survey had established a development unit in Space House ("Rymdhuset") in Kiruna. Its main task was to spearhead the adoption of new digital technology in map production in Sweden, but it also became a partner with SSC in developing and marketing satellite data products world-wide. The digital terrain model development was the basis for a joint venture, "TOPSAT", to market and produce topographic maps.

Although technically very successful, the capability to create topographic maps from satellite data was not a big commercial success. We quickly discovered that stereo pairs from SPOT were difficult to acquire from most parts of the world, due to cloud cover and operational constraints. In addition, the market took some time to reach the "take-off" point.

A digital satellite map of Sweden

In January 1990, thanks to the successful development work and sales pitch by Mats Rosengren, SSC Satellitbild received an order from Swedish Telecom and the Swedish Defence Materiel Administration for the creation of a digital database of terrain and vegetation types covering the whole of Sweden.

Landsat Thematic Mapper data and some SPOT data was used. All data was precision corrected to the National Grid with sub-pixel accuracy. The digital database contained 13 classes with 50-metre pixel size. The complete dataset contained 225 Mbytes.

This was the first operational nation-wide application of high-resolution satellite data within Sweden.

10 years of hard labour in Solna

Sounds like a prison sentence, doesn’t it? But nothing could be farther from the truth. Throughout these years, there was a continuous flow of low-profile work, laying the foundation for new applications and business opportunities, not just for SSC. It involved a series of small studies, investigations, development projects - often in co-operation with other organisations. It was largely done as part of the SBSA sponsored "remote sensing application programme", but there were also many small external contracts.

Most of the work concerned investigations and demonstrations of the applicability of new satellite sensors and systems to "real-world" information needs. Landsat TM, SPOT, ERS SAR, AVHRR... Erosion studies in Africa, forest mapping in Sweden, the marriage of satellite data with GIS technology, the Geovision geology project, ice mapping in the Baltic Sea, forest decline in central Europe, environmental problems in Estonia... Some of the more technical work was done for CNES (software development for the SPOT-4 ground segment, largely sub-contracted to Satellitbild), and ESA (a study of their contribution to a "Global Environmental Network", GENIUS).

A lot of effort was also spent on spreading information through seminars, meetings, visits, and the "Remote Sensing" newsletter.

The Maritime Surveillance System

In the 1970s we had successfully developed an airborne surveillance system for the Swedish Coast Guard, which had also been sold to the Netherlands. It was soon followed by further successes on the export market; Great Britain, Germany and others.

Later, a microwave radiometer was added to the system. This gave the capability to provide quantitative estimates of the thickness of oil spills.

By 1987, the MSS system had been sold to ten countries. That year a new generation of the system based on the EBBA-GIS image processing system was introduced on the market. With the new system, the operator could for the first time enhance and analyse the information collected from the various sensors. The result was displayed in full colour on high-resolution monitors in real time. - The Swedish Coast Guard became the launching customer. The system was installed in the Spanish CASA-212 aircraft. In 1991 three MSS systems were delivered to Portugal.

Although we enjoyed a clear technical and commercial supremacy in the market throughout this period, and the MSS business unit was profitable overall, the market developed more slowly than expected. This made it difficult to achieve a steady production flow. As good years and bad years alternated, there was a lot of pressure on the group which they handled admirably.

Krylbo - a train wreck

By the end of the 1980s, with a decade of Landsat operations and a few years of SPOT operations under our belt, and with new major investments coming up, we felt ready to look for improvements in the way we procured and operated our processing systems for satellite data in Kiruna. It did not seem reasonable that we should use completely separate systems to support different satellites, run by completely different teams, down to separate photographic laboratories - especially when a complete processing system for a single satellite could cost 50 MSEK or more.

As a minimum we ought to strive for a multi-mission philosophy, where "plug-in" modules would be used to adapt a generic processing system, including archiving and catalogue systems, perhaps, to particular missions, hopefully at a fraction of the cost of the traditional approach. We should strive for portable software solutions, independent of any particular hardware implementation.

Perhaps we could even use our own extensive experience in defining and developing image processing systems to assume the role of system architect and/or provider of modules of the overall system?

This concept came to be known as the "Krylbo" project. Krylbo is a small Swedish town known for its railway station, where all the major railroad lines are supposed to meet. Kiruna would become the Krylbo of Earth Observation satellites, if you can follow the line of reasoning. - Unfortunately, as Lennart Lübeck prophetically pointed out, Krylbo is also the place where a huge explosion occurred during World War II, possibly caused by sabotage...

A committee was set up to explore the idea of a multi-mission ground system. Its efforts were quietly buried. It was suggested that our customers were not wild about the idea, and that we would do well to listen to them.

New faces

The second decade of our remote sensing activities in Solna saw some important changes in staff. Just before Fredrik Engström left SSC in 1985 for ESA (amid speculation about what would happen when an irresistible force encountered an immovable object), he hired Håkan Kihlberg from IBM to head the newly created Remote Sensing Division. Håkan instilled some much needed discipline and methodology. When he was summoned to even greater tasks in 1986, Claes-Göran Borg, who had been managing commercial remote sensing services and methodology development until 1985, took over. Christer Andersson headed methodology development from 1987 until the early 1990s. Olov Fäst, who had succeeded Lars Backlund as technical manager for the Maritime Surveillance System in 1982, became Head of the MSS unit around 1990.

Sobering up

By the time of our 20th anniversary, we could look back with pride on some of our accomplishments:

  • Satellitbild had been established and had become fully operational
  • We had successfully completed major mapping projects in developing countries using satellite data
  • The Salmijärvi station had been established
  • Applications of the new generation of Earth Observation satellites had been developed and demonstrated
  • Old rivalries had been turned into strategic alliances.
However, some of the most optimistic expectations of 1982 had dissipated, and morale had been adversely affected by the relative lack of commercial success, by the absence of a dramatic "breakthrough" for remote sensing applications in Sweden, and by the perception among some developers in Solna that they had somehow been "boxed in". Turnover in personnel reached unprecedented levels.

In retrospect, it is clear that the root problem was that the world market for remote sensing products and services was developing more slowly than anticipated. Our possibilities to change that situation were somewhat restricted, of course. But then, patience was never one of the corporate virtues of SSC!

Claes-Göran Borg summed up the prevailing mood in a 1992 "Remote Sensing" editorial:

Now we are facing the really difficult task; making use of these fabulous data volumes... If we do not succeed, what we see now could well prove to be just a short boom. Space is easy - the challenge is on the ground!

Satellitbild: The pioneering years

Now, here is the plan...

When Satellitbild was born in 1982, its childhood had been meticulously planned. A business plan had been prepared, detailing its various products and services, the estimated market, the anticipated staffing and investment needed, the projected revenue, and the financial requirements. Break-even would be achieved within two years after the launch of SPOT-1. Based on this plan, the necessary financial guarantees had been given, including the basic income from the French for supporting SPOT. The French participation as a shareholder of Satellitbild had been agreed.

During the setting up of the company, a Godmother in the shape of the Swedish Land Survey demanded to be represented on the Board of Satellitbild. At the time, the Land Survey and SSC looked upon one another as competitors, so Fredrik Engström flatly refused the proposal, which had already been backed by the Government. If we were going to operate Satellitbild as a business, it must be free from outside interference. He threatened to dismantle the whole project rather than give in on this point. - Luckily, the Ministers concerned (or the concerned ministers...) relented, and our future relations with the Land Survey developed on the basis of common interest rather than government decree. A few years later, the Land Survey under its new Director General Jim Widmark launched a joint venture with Satellitbild, and Jim eventually became a much respected member of the Board of Satellitbild.

The first order of the day was to find a Managing Director for the new company. After a brief search, Svante Astermo was appointed. He had a background as production manager at the Land Survey which fit well with Satellitbild’s intended role as a "Remote Sensing factory". - Lars Bjerkesjö was appointed Administrative Manager, and Karin Lindholm was recruited as Secretary from 1 January, 1983.

In the business plan, we had identified the recruitment of qualified staff to Kiruna as the largest risk factor. Our fears turned out to be exaggerated. A lot of talented engineers with the right background were willing to take part in this pioneering endeavour. Some of them worked at Esrange or LKAB. Others had roots in the region and were happy to return after a period of work in southern Sweden. A core team of a dozen people was assembled in the first months.

A new building was going to be constructed for the company. Meanwhile, a large apartment in central Kiruna was rented on a provisional basis. This became the first home of Satellitbild.

Here, the core team had their first taste of the remote sensing business. In the SkyEye project they used reduced-resolution "quick-look" Landsat images to generate fresh information on crops in Europe and the Soviet Union on a daily basis for an American customer. Vegetation indices were generated on the EBBA system, and the results were sent by fax to the customer!

The next task was important: to procure the facilities needed to archive and process the data. After all, SPOT-1 was planned to be launched in 1984!

And the winner is...

As the very existence of the company depended on our co-operation with the French, it seemed obvious that we would procure the facilities to generate SPOT standard products from French industry. This would also reduce risk levels and promote full compatibility between our system and the one installed in Toulouse. We received good assistance in preparing the technical specification from a CNES specialist, Philippe Delclaux, nowadays the Technical Manager of Spot Image.

However, as negotiations with the French supplier SEP started in Stockholm, where most of our expertise resided, the perspective slowly changed. Our engineers resented the idea of using the ageing 16-bit computers that were offered. The design seemed inflexible, and the cost outrageous. Worst of all, there was very little of the usual give and take in a "real" negotiation. We felt cornered.

Finally, we threatened to open up the competition. This would have been a severe blow to SEP on the world market. A senior director was flown in, and within a few hours the necessary concessions had been made and the contract signed.

We vowed that we would not get into the same situation with the next procurement: a system to produce "higher-level" precision-corrected products. This was an open competition, and we got several interesting offers. In the final round, two companies were invited to Stockholm for negotiations in parallel. In fact, during several days, we spent the morning with one of the finalists, and the afternoon with the other one. We now felt that we had the upper hand. Our expertise was in many respects superior to that of the bidders, and we enjoyed superb legal assistance from our professional solicitors.

The situation changed dramatically when suddenly one of the bidders withdrew their offer and went home! Within their team, they had unearthed such inconsistencies and doubts in their proposal that the financial backers of the company had lost faith in the bid!

What to do? It was too late to summon one of the dismissed semi-finalists, and anyway this would have been interpreted as imminent victory by the remaining finalist! We decided to go on as if nothing had happened, and prayed fervently that the two finalists were not staying at the same hotel! In fact, we improved our negotiating position, as we could now use the afternoons to prepare for the next morning’s negotiations!

The strategy worked, and we got a watertight contract with the remaining bidder at a reasonable price!

Pomp and circumstance

The new Space House in Kiruna was inaugurated in September 1984 by the Finance Minister Kjell-Olof Feldt. About 3000 people took the opportunity to visit Satellitbild for an "open-house" inspection. The Municipality of Kiruna acted as host for a big party in the Town Hall. We were slightly embarrassed to find that an American astronaut had been flown in as the guest of honour; how should this be explained to our French partners?

By this time Satellitbild had reached about 20 employees. The cosy days of operating a VAX computer in a room with a tiled stove were over!

Have another cigar!

The launch of SPOT-1 had been postponed repeatedly. Thanks to the foresight of Klas Änggård and others, sufficient margins had been built into the financial model of Satellitbild to accommodate the delays without serious repercussions.

In preparation for the launch, Satellitbild gradually built up its technical staff, trained them on the new facilities to be installed, and performed various acceptance tests. The hiring of operators in the photographic laboratory and the computer hall was delayed as much as possible, so that expenses could be minimised until we knew that the launch had been successful.

In parallel with these preparations, all the detailed agreements of how the work would be arranged were negotiated with CNES and Spot Image. The original agreement between SBSA and CNES had been brief. It had set out the principles for the co-operation. Now a lot of thinking had to go into the implementation of those principles.

This was not your everyday business negotiation! A whole new set of rules and principles had to be invented along the way - as much in co-operation with the French as in negotiation. Exactly what was a standard product, where exclusive licenses would be granted? What was a "value-added" product? Could a simple poster or a postcard be considered to fall into that category? What if someone produced a merged SPOT/Landsat product; could and should a royalty apply? How would programming conflicts be resolved? What about income from foreign stations? Who would pay how much to whom when a Swedish user required a scene, and it was received in Toulouse, or in Kiruna? How would copyright infractions be prosecuted? - The topics seemed endless.

In fact, this negotiation probably was just as helpful to our French colleagues in setting up their distributorships on the world market as in defining our business relationship. It lasted for several years. It was headed on our side by Klas Änggård and on the French side by Gérard Brachet. Gradually, small rituals developed between them, such as offering each other a cigar to finish each working lunch. Their skill and ingenuity finally built a solid foundation for the future.

Finally in business!

The dual launch of SPOT-1 and Viking on 22 February, 1986, was a spine-tingling, nail-chewing event for all of us in SSC, but the atmosphere was especially tense at Satellitbild. A failure would have been a very hard blow for the young company. Luckily, everything went according to plan. We soon had some marvellous first images to look at. They showed a barren landscape in Algeria and the snow-covered Po plain in Italy in unbelievable detail.

An early coup was scored just after the satellite was commissioned at the end of April after two months of intensive testing and calibration in orbit. The Chernobyl blow-up occurred just before the 1 May holiday. Thanks to some intensive work over the holiday, we managed to deliver the first satellite images of the accident site - both SPOT images and a thermal Landsat image - to the media ahead of our competitors. This put Satellitbild "on the map" and gave a boost to its distributor Space Media Network.

Teething problems

In the early months, although the satellite and the ground processing system worked beautifully, there were some teething problems in SPOT operations. It quickly became clear that the French had underestimated the amount of work needed to programme the satellite as the orders started to arrive. They soon found out that they needed to hire and train additional operators to assign priorities and manage system resources. - To some extent this created difficulties for us in meeting user requests. For instance, scientists doing in situ measurements in the Baltic Sea found out that the satellite had been looking at Norway during the overflight! In the case of Nordic users, we also had to deal with an unfavourable, rainy summer season. We got lots of clouds in our archive!

We also had our own start-up problems. Our photo lab had reject rates up to 75 %. Quality control of digital products was also inadequate, and production times in the beginning were too long and unpredictable. - When one considers the experiences of the Americans with Landsat, the French and ourselves with SPOT, and later ESA with ERS and NASDA with ADEOS, it seems to be a universal truth in Earth Observation that, somehow, insufficient attention is always given to the problems of running an operational system, as opposed to solving the purely technical problems...

Your mission, should you decide to accept it...

In the autumn of 1986, Lars Bjerkesjö took over as Managing Director of Satellitbild. He quickly took steps to improve the control of production flow. As a side effect, we got a better handle on the "real" costs of generating different products. This information was desperately needed to estimate profitability and set prices when the new "value-added" chain was finally delivered and gradually taken into production in 1987.

In the autumn, the performance of the overall SPOT system gradually improved, although the global catalogue system left much to be desired. Our volume of work for the French reached and surpassed the revised targets.

A basic problem quickly became manifest - a problem that persists to this day. Our early expectations had been that Satellitbild’s core business in "value-adding" activities would be to produce standard products in a "Remote Sensing Factory" on behalf of organisations acting as "middle-men", who would produce maps and information for the end users. It became evident that such middle-men were lacking, and that we would have to offer our services in the form of complete projects to the end users.

Naturally, this re-orientation created some headaches. What was the right mix between standardised volume production and customised low-volume production? How should our limited development resources be deployed when there were pressing needs to tune and improve the efficiency of the production system, to create better management tools for planning and control of order processing, and to develop new products? Could we afford to tie up resources in domestic pilot projects? Which contracts were worth pursuing?

The urgency of finding some good answers was underlined by the fact that Satellitbild’s net result for 1986 turned out to be 10 MSEK below budget despite the successful launch of SPOT-1. The jobs of some 50 employees were at stake!

Selling bits to BITS

So we intensified our efforts on the world market! SSC in Solna, which was responsible for the marketing of international projects, called in some seasoned professionals from the international civil engineering field: Ulf Kihlblom and Hans Rasch.

They scored an important, early victory by winning a contract to map the entire Philippines. The job was completed in just a year as a joint effort by the integrated SSC-Satellitbild team. The whole-hearted support of Spot Image in programming the acquisition of images was an important prerequisite for the success.

This project set an example for a string of similar, usually less grandiose, projects during the years that followed. There were vast, unfulfilled needs for geographical information in developing countries. The major problem in satisfying these needs was that developing countries usually were hard-pressed to finance large mapping projects. Normally, the financing had to come from international aid organisations, such as the World Bank, and in Sweden from SIDA and BITS.

The aid organisations wisely demanded that the recipient countries should carry the main responsibility for specifying the requirements, and provide resources for complementary domestic activities. The projects should also involve a substantial element of "technology transfer", i.e. of training of the local operators in the processing and use of satellite data.

Naturally, these different aspects contributed to a relatively complex decision procedure. We soon discovered that the start-up of a project often requires several years. The Philippines project had been atypical in that respect!

Despite the difficulties, during the initial years of Satellitbild operations, a steady flow of projects kept our operators and processing chains busy. In 1991, the marketing group in Solna was formally transferred to Satellitbild.

Space Science: Going into orbit

The challenge of Viking

The Viking project presented a major challenge to our young and relatively inexperienced organisation. The basic concept had its origin in studies of a Swedish scientific satellite back in the late 1960s, as we have seen. We borrowed heavily from the experience of a small-satellite team at Boeing, and profited immensely from the presence of Boeing systems engineer Larry Mitchell in our own design team.

Still, we now had to move beyond the definition phase to actual design and development. As if this was not difficult enough, we had to complete the job under severe budget constraints and under time pressure, as we had SPOT as co-passenger. And we could not depend on the SPOT development team to delay the launch!

On top of this, we had a complex project on our hands, with critical interfaces to be worked out both with the SPOT team at CNES and with Arianespace. We had to manage the activities of our relatively inexperienced Prime Contractor at Saab, and we had a number of interfaces with the instrument providers at different international research establishments. Here, our Project Scientist, Prof. Kerstin Fredga had an important role.

And we should not forget another critical task - to develop the ground segment that would be used to control the satellite and receive and process the data.

And by the way - as we were procuring the big Tele-X satellite in parallel, our technical resources were already stretched to the limit, and there was no assurance that our top management would be available to assist in negotiating the many critical agreements that had to be ironed out.

So Per Zetterquist, Sven Grahn and their young team had their plate full. The only way to stay sane in this situation was to apply the principle: "I don’t get ulcers, I give ulcers!", so the pressure must have been felt just as much by Johnny Andersson and his team at Saab, and by all the instrument providers.

Another way to cope was to disregard the "conventional wisdom" in satellite projects. The key to success was to adopt a "sounding rocket approach". We could not afford to have extensive paperwork in the project, or triple redundancy, or extensive testing at many different levels along the lines of major space organisations such as NASA, ESA and CNES. Sven introduced a "no-nonsense" approach, relying on a small project team, personal trust, and direct communications, to cut down on the layers of paperwork and testing. Of course, this meant cutting some corners, and taking some risks, but this did not mean that we were reckless. The risks were carefully analysed, and the increased level of personal trust and responsibility served to counteract some of the risks of reduced double- and triple testing and checking.

We tried to instil a similar attitude in our contractor teams. A typical exhortation borrowed from our sounding rocket background was: "If a hydraulic line breaks, the spray should hit the project manager!"

Doubts

A lot of our in-house development effort concerned the ground segment to be installed at Esrange. Our sounding rocket experience was useful, but it had its limitations in this context. This bird had to be controlled for many months rather than a few minutes! The design had to support routine operations while providing sufficient flexibility to handle emergencies.

Our ground segment manager Lennart Björn visited several foreign ground stations as part of his preparatory activities. In the process he got a fair account of all the things that could go wrong in operating a scientific satellite. Out of this he established some simple "self-evident" rules: No commands should be entered by the operator in hexadecimal code; no two commands should differ by just a single character; etc. - The human engineering part was the subject of many heated debates between Lennart, Joe Armstrong, and Bengt Holmqvist. One heated debate on a Friday evening turned into harmony on Monday morning, after Joe had prototyped a whole new operator environment over the weekend!

Another area of concern was the simple task of finding the satellite after it was put into orbit following launch. The lobe width of our big antennas is rather narrow, so unless you know exactly where to look, chances are slim that you will find the target. In our case, the main source of uncertainty was the exact pitch angle of the Ariane vehicle at the moment Viking was released. A French specialist at CNES was responsible for providing us with the correct antenna angles as a function of that pitch angle, which would only be known during launch.

For some reason, our scientific wizard Dr. Dr. John Murdin (I am not kidding, he has two doctorates!) became suspicious. "If the pitch angle is this and that, what will the antenna angles be?" The result looked even more suspicious. - During the night before the launch John developed a complete mathematical model of his own, and discovered that the predicted antenna angles were significantly different from those provided by the CNES specialist.

For Lennart and John, the high drama of the launch 22 February, 1986, was deepened by nagging uncertainty: would we be able to find Viking? During the launch itself, SSC colleagues in Kourou and Toulouse read the critical parameters off the computer screens, standing behind the French operators, and telephoned the results to Esrange. John fed the pitch angle data and tracking data from the Bermuda downrange station into his model. Still, we tried the parameters transmitted by our French colleagues. When Viking was overdue at Esrange, and there was no contact, we switched to John’s data. Lo and behold, we immediately got acquisition of signal! The same thing happened on the next orbit.

Triumph

In the early morning of 22 February, we could monitor the critical events: the pre-programmed ignition of the kick motor that boosted Viking into its elliptical orbit, the deployment of the booms, etc. We could even monitor unforeseen events, such as the slight shift in rotation speed when Viking entered the Earth’s shadow and its moment of inertia changed due to thermal contraction of the booms.

Soon, the scientific groups received their first data sets. The mission was a complete success, and pioneering scientific results were obtained, to the great satisfaction of our Project Scientist, Prof. Kerstin Fredga. For the first time, the scientists and the general public could see what the aurora borealis looked like from above, extending over large distances in the polar region.

Viking had a design lifetime of eight months. It kept operating for well over a year, delivering valuable scientific data until its power supply subsystem gradually declined, as anticipated. It should stay in orbit for some 14,000 years, well into the next ice age...

Slither

When our launch team returned in triumph from Kourou in French Guyana, they had some interesting stories to tell. For instance, Anders Björkman brought a snake skin several metres long. He had been out into the jungle, "the green hell" (not to be confused with Esrange in the mosquito season!), with a guide. A large snake suddenly appeared on the jungle path between the guide and Anders. The guide calmly turned and shot the snake.

If Anders had not brought the snake’s skin, nobody would have believed him. - According to legend, once when he was driving on the road to Esrange in winter at his usual reckless speed, a cigarette dangling from his mouth, he suddenly met a big truck. As the road was too narrow for both of them, Anders used the packed snow on the shoulder of the road to squeeze by in a vertical position. Afterwards, his car showed scrape marks on the roof! - If you believe that, you will also believe that elephants can fly!

Microgravity

During the 1980s, SSC developed scientific equipment to support research in microgravity using sounding rockets. In particular, we developed ovens for materials research. During the brief flight of a sounding rocket, metal samples had to be heated until they melted. They were then allowed to cool under controlled conditions, so that they solidified while still in the "free-fall" part of the trajectory, before g forces started to build up during atmospheric re-entry.

Quite a tricky business! But then, that’s all in a day’s work at SSC...

Our microgravity work not only opened a new area of research for Swedish materials scientists. It also contributed to developing a new area of sounding rocket applications.

It’s a GAS!

One of our endearing weaknesses is that we never could say no to a good flight opportunity! So when NASA offered cheap flight opportunities on their Space Shuttle, we immediately started thinking of ways to take advantage of this.

What they were offering was space for self-contained experiments which could fly unattended during normal missions. This was referred to as a "Get Away Special", or GAS. The experiments were flown in canisters stowed away in the Space Shuttle.

SSC started preparing microgravity experiments on behalf of Swedish scientific groups for GAS flights. Our first GAS experiment had to wait several years for a flight opportunity, after the Shuttle was grounded in 1986. But we finally got our first flight in 1992. Another microgravity GAS experiment was conducted in 1994.

PIROG

The PIROG (Pointed InfraRed Observation Gondola) platform is a general purpose balloon platform capable of carrying a 200 kg payload for stratospheric flights lasting up to 12 hours. The platform concept was developed by SSC in the 1980s. A stabilised platform is used to permit astronomical observations of the interstellar medium in the infrared spectral range to be performed above the lower layers of the atmosphere. The attitude is controlled to within 1 arc minute, using a gas jet system with the sun as a reference. The telemetry includes image data from two TV systems and position data from an onboard GPS system.

Although some of the early flights in the 1980s ran into technical difficulties, the design is now fully mature. It has been used successfully in a number of flights. Typically, the 500 kg gondola is launched to 40 km altitude using a 400,000 m³ balloon filled with hydrogen. The flight is terminated by cutting the balloon line via ground command. The gondola descends to ground by parachutes. A helicopter brings it back to the launch base.

Freja

As soon as the Viking mission had been successfully completed, studies began in earnest of our next Swedish scientific satellite Freja, a magnetosphere research satellite. By now our "design bureau" was a seasoned group of experts, so we no longer had to rely on outside assistance.

A launch opportunity had already been found. Originally, we had planned to include a message forwarding experiment on Tele-X: "Trucksat". The idea had been that truck drivers should be able to communicate with their home office via satellite. We had been forced to abandon the experiment due to funding limitations, but the idea was resurrected a few years later in the form of "Mailstar", a small low-orbit store-and-forward messaging mission. Mailstar reached a rather advanced stage. We even registered the name as a daughter company of SSC. But again, the financing could not be arranged, and we were forced to abandon the project.

Mailstar was a far-sighted and viable project as later developments have demonstrated. Again, SSC had come up with a great idea before the world was ready for it! - But I digress. The point is that we had booked a launch option for Mailstar with a Chinese launcher, and that this option could be used for Freja.

Freja was designed as a sun-pointing spinner with a 2.2 m diameter and 214 kg mass. It was roughly the same size and shape as Viking. Swedish, German, Canadian and U.S. instruments were flown on the satellite. The structure consisted of a central tube machined from cast magnesium. Inside this, the solid rocket motors were mounted on a Kevlar adapter to limit heat soak into the satellite from the spent motors. Four radial walls connected the central tube with the launcher interface ring, and on the top and bottom of the walls the top and bottom platform decks were mounted. Platform and science equipment were mounted on the walls and the platforms. Freja had redundant transmitters, receivers, batteries, solar array shunts, telemetry systems, telecommand decoders, computers, pyro circuits and attitude sensors and actuators.

The system design was completed in 1988. The protoflight satellite structure was completed in the fall of 1989. The structure qualification vibration test was performed in China in March 1990. Satellite integration was finished in August 1991. Different tests were performed in 1991. The satellite was shipped to China for launch in August 1992. - This is where we leave our reader hanging from a cliff, saving the launch for a later chapter of this chronicle!

The Freja project budget was less than half the cost of the low-cost Viking project. As the missions were quite comparable in scope and level of ambition, Freja represented a further significant reduction in cost compared to Viking. This was partly accomplished by SSC acting as the Prime Contractor for this mission. To a certain extent we could also now benefit from the "learning curve" from Viking. – Still, the project could not have been carried out without the unwavering support of the Swedish National Space Board, and the additional financial support of the Knut and Alice Wallenberg Foundation.

Tele-X: A dream come true

The SSC project team is formed

The go-ahead for the Tele-X project in 1982 represented a tremendous challenge to our still young and relatively inexperienced organisation. True, most of the actual work was going to be performed under contract to SSC by some major aerospace companies, of which Aerospatiale was already an accomplished builder of telecom satellites. But in the end it was going to be SSC who prospered or suffered depending on the success of the project.

To manage industry and make sure that the project was completed on time and within the agreed budget, we had to develop our own capability in the most modern satellite technology. We also had to prepare and negotiate watertight contracts with industry and with the Nordic governments. The total budget for the satellite, launch, and ground equipment, was 1250 MSEK at 1982 economic conditions. This included a 5 MSEK contingency margin...

The core of the project team was made up of the SSC engineers who had been involved in the ill-fated Nordsat project. Per Zetterquist was appointed Project Manager.

Playing for high stakes

The industrial negotiations for the Tele-X development project were prolonged and difficult. The original intention had been to appoint Saab prime contractor. Their offer was awaited with great anticipation at SSC.

An office pool was set up to allow betting on how high the offer would be. The winner was Stigbjörn Olovsson. This was unfortunate, because he had entered the highest bid of all, so much higher than the target figure that the idea of making Saab prime had to be given up!

Instead, an industrial group led by Aerospatiale, and with Saab and Ericsson as major subcontractors, finally signed the contract for the development.

In parallel, the Swedish and Norwegian governments in 1983 made the final binding commitments allowing the project to proceed.

Talk to my lawyer!

In the two-sided negotiations with industry and ministries, we received first-class support from Swartlings Advokatbyrå in the shape of Rolf Olofsson. To many young engineers, the idea to spend a considerable amount on lawyer’s fees in a satellite development project probably seemed quite alien. With the benefit of hindsight, it turns out that this may in fact have been one of SSC’s wisest investments.

Our negotiating team headed by Klas Änggård and Rolf Olofsson gradually overcame all obstacles and managed to nail down watertight contracts and agreements.

It would be a mistake to simply describe these as "victories" over our counterparts. The essence of a successful business relationship is to ensure that both sides benefit from an agreement, and this is what was accomplished. In particular, by foreseeing everything that could go wrong or become contentious, and resolving these issues, we managed to avoid endless future conflicts and litigation.

The bear’s hug

The Tele-X project had been endorsed by the Swedish Telecommunications Administration (STA), and our relations had improved dramatically after the early conflicts during the Nordsat study. STA was genuinely interested in Tele-X as an experimental satellite where the new technology could be thoroughly tested before any long-term commitments were made.

However, it was also clear that STA was deeply committed to the expansion and modernisation of the ground network. Tens of billions of SEK were going to be invested in building the new network of "electronic motorways" across the country. Communications satellites were still seen mostly as a complement to meet special needs. An early operational use of Tele-X was not on the agenda. It would either force STA to take the responsibility for the operational use of an immature system, or foster an unwelcome competitive activity if some other entity became responsible.

This boiled down to another conflict in 1985. If an operational system based on Tele-X was going to be set up, it was time to start building another satellite, "Tele-Y". STA went public with a statement that Tele-X should not be made available for commercial use. Instead, experiments involving non-commercial entities such as "scientific groups, sports clubs, and Scandinavian Airlines System" were advocated.

Although STA had embraced the Tele-X project, to SSC the embrace seemed to be that of a bear more than of a tender lover. Opinions varied as to whether STA was a cross we had to bear, or a bear we had to cross... Fredrik Engström pleaded for support from industry to set up an operational system: "STA may be building electronic motorways, but we are going to pave the whole country!" But the support was not forthcoming. It looked as if Tele-X was destined to become a one-off technological demonstration.

The prospect that Tele-X might actually find some practical use dimmed further when the launch had to be postponed for almost two years. An Ariane failure caused a moratorium on launches for an extended period, and failures of other launch systems in the same period caused traffic to accumulate on the Ariane waiting list.

Rolling the dice

In 1989, the launch could finally take place from French Guyana. Some last-minute difficulties led to the familiar "Spencer’s Game": who would be the first one to concede that he was not ready? This occurs frequently in the sounding rocket business, when some scientific group has last-minute technical problems in its payload. Usually the red "chicken" button is not pressed until the very last minute in the hope that some other group will be forced to admit that it is not ready for launch. - In this case Arianespace tried to persuade our project manager that we should ask for a delay in the launch. It turned out that they themselves had problems but wanted us to cover the cost of postponement!

The moment of truth finally arrived on 2 April, 1989. Over a billion SEK and almost ten years of effort were on the line. The tension was palpable at the launch centre in Kourou, and at the late-night gatherings in Solna and Kiruna watching the TV transmission. But the launch was successful and the satellite functioned perfectly!

Our pride was not diminished by the fact that we had completed the project well within the agreed budget despite the considerable delays. How many other billion-plus SEK development projects could make the same claim?

Six months later, the author got a poignant reminder of how we had turned science fiction into reality, when he encountered Ola Jirlow on a plane to Kiruna – "What brings you there?" – "I’m going to assist the control team in eclipse operations." - Twenty-five years earlier, in a student project at Stanford university, the author had been responsible for calculating exactly how long a geosynchronous satellite would remain in the Earth’s shadow during the equinox periods!

Early customers

During the latter half of the 1980s, there was a growing unease over the apparent failure to take full advantage of an otherwise successful project. Plans to use Tele-X for a "Mini-Nordsat" concept were discussed, but scrapped in 1988. The take-over of Tele-X by the Nordic Telecom Administrations for operational use, which was a part of the original agreement, met with a considerable lack of enthusiasm.

All of this strained the relations between SSC and the main financial supporter of the project, the Swedish Ministry of Industry. Lennart Lübeck was quietly working to find a solution, and an important step was taken just prior to launch, when the Swedish, Norwegian and Finnish governments agreed that Sweden should become the sole owner of the satellite in exchange for certain rights for the others to use its resources. The Swedish government awarded the rights to SSC to market all Tele-X services except public TV (which was handled by the Government’s holding company NSAB).

This was still a very difficult task, as there were no concrete plans to provide backup capacity in the form of a second satellite. Valuable time had also been lost in preparing the market and developing ground terminals for business communications.

Our first customer was the newspaper Aftonbladet. They wanted to print the paper locally in Gothenburg. By transmitting the data via Tele-X, this became technically possible - outside broadcasting and video conferencing soon followed.

In 1990, TV4, a commercial Swedish TV channel, started broadcasting to the Nordic countries from Tele-X. This was 15 years after SSC had started promoting the use of direct-broadcast satellite technology. - In late 1991, another commercial channel, TV5 Nordic, became a customer.

Finally, in 1992, the Swedish Government decided to privatise NSAB, which owned Tele-X. SSC now owned the satellite jointly with FilmNet International Holdings AB. Four public TV channels were broadcast via Tele-X: TV4, TV5 Nordic, FilmNet, and Norwegian state television (NRK).

Our delight at commercial success was tempered by the realisation that long-term viability was not yet assured, and by lingering regret that the grand vision of Nordsat had been defeated.

End of Part 3

Go to Part 1: Ancient history - 1972: How it all began
Go to Part 2: The first decade: 1972 - 1982: The sky is no limit
Go to Part 4: New horizons: 1992 - 1997: Change in all things is sweet

 

Home page
News
Gallery
Curriculum Vitae
Araguacema
Christofer
Kerstin Amanda
Space
Family tree
History
Arts
Books
Chess
Mountaineering
Things that surprise me
Web stuff
Funny quotes
Contact