PERUN—Space shuttle external tank used as a space station

Space Shuttle External Tank Used as a Space Station Study Project Perun
Tomáš Svoboda, Tomáš Svítek, Jiří Vackář, Jiří Bárta, Karel Vítek, Michal Kirschner, Milan Urban

Munich 1979

Abstract
This paper describes the results of a study project which has been conducted by a Student Working Group for Astronautics of the Planetarium of Praha. It deals with the possibility of converting an expended Space Shuttle External Tank into a Space Station, a permanently manned orbital facility. The Space Station was designed economically by using much off-shelf hardware developed for earlier space projects. It is compatible with the Space Transportation System which will operate in the 80's. It is proposed that mission dependent experimental equipment be carried aboard Spacelab-modules which should be exchanged during periodic revisits by the Space Shuttle

The key functions of the proposed Space Station are discussed such as power supply, thermal control, trajectory corrections, logistics etc.

In our project we have arrived at a Space Station carrying a permanent crew of 8 people, orbiting the Earth at an altitude of 390 km, resupplied by STS once in 80 days.

Keywords
Space Station, Space Transportation System, Space Shuttle, Space Shuttle External Tank, Spacelab, Permanent, Universal.

Introduction
The Space Shuttle External Tank (ET) is the only part of the Space Transportation System which is designed as expendable. It will carry liquid hydrogen and oxygen to power the Space Shuttle Main Engines during launch. ET is 48.4 m long and has a diameter of 8.4 m. It is composed of two large propellant vessels. In the upper part the liquid oxygen (LOX) tank is located while the liquid hydrogen tank occupies the bottom of ET. The tanks are ended by ellipsoidal bulkheads with the exception of the upper portion of the LOX tank which forms the tip of the tank and has an ogive shape. The tank is welded from aluminum alloys and weighs 35 000 kg empty.

The tank is designed to be jettisoned from the Orbiter shortly before reaching orbit. It is possible however to keep the tank connected to the Orbiter and accelerate it to orbital velocity. The required delta v is about 100 ms-l and can very well be provided by the OMS engines of Orbiter. Once in orbit, the tank structure can be used in many ways, for example as a building block of a Space Station. This is the basic idea of the Perun Study Project.

The idea is not original: in recent years many proposals have appeared to utilize ET in orbit as a Space Tug (Seguin, 1978), a reentry vehicle (Kent, 1978) and even more recently as an orbital refuelling depot (Covault, 1979) Some of the most controversial proposals are those by G. O'Neill (1977, 1978) to use the structural material of ET in pelletized or powdered form as propellant for an electromagnetic reaction engine or to assemble a large space habitat of a large number of ET's (O'Leary, 1978).

Of special interest to us are the proposals to make use of the ET as a Space Station. These have been made by Grumman (1977) and by Marshall Space Flight Center (NASA, 1977; Dooling 1977). The Grumman have decided not to utilize the internal volume of the tank. They make use of ET only as a strongback of a multipurpose space facility. On the contrary the MSFC in their project Super Skylab have attempted to utilize the interior of the tank for habitation. However the project has not been finished and it's preliminary results are of small value.



Fig. 1. Principal parts of the External Tank


 * 1.Orbiter attach point


 * 2. LH2 tank


 * 3.Intertank structure


 * 4.LOX cylinder section LOX tank


 * 5.LOX tank ogive section

Basic Ideas
The shape and dimensions of ET make it quite advantageous to be used for habitation, providing a comfortably large living volume for the crew. However the research or processing equipment should better be placed in interchangeable modules derived from Spacelab. That way the mission dependent equipment could easily be exchanged during regular Orbiter revisits to suit the precise tasks undertaken by the crew. On the contrary, all the equipment which is mission independent should be located permanently inside the tank. That includes for example experiment support computers which are a part of research equipment but are the same for any kind of research - are mission independent.

This approach makes the station universal. It should also be a permanent facility with crews rotated by the Shuttle along with experimental modules and supplies.

Perun - the name of the proposed Space Station was arrived at by abbreviating the words PERmanent and UNiversal which are the essential features of the facility.

To keep the price of the Perun Space Station as low as possible we have made effort to utilize in the design plenty of offshelf hardware which has been developed for earlier space projects. In fact even the tank itself is a piece of offshelf hardware. A number of subsystems is used from Skylab and Spacelab projects, some from the Space Shuttle. More details will be given in the appropriate parts of this report.

Launching the Tank
It is possible to launch the tank fully fuelled and use its internal volume for habitation by the crew. It would be necessary to introduce some modifications of the tank before launch, like grid floors, electrical and other installations. Most equipment would be brought by later Shuttle missions and plugged into prefabricated racks. This is reminiscent of the early history of the Skylab project. Skylab too should have been originally rebuilt from an expended S IV B tankage. In this configuration a small mass penalty (typically 2,6 t) must be paid for bringing the tank into low earth orbit.

We have also considered the concept of ET partial tanking, similarly to the Super Skylab project. The basic idea of the Super Skylab project is that ET can be orbited with as little as 90% propellant in it before launch. This limits the mass orbited on the flight but provides some vacant volume inside the tank where hermetic enclosures can be installed before launch. These enclosures should be used for habitation by the crew and for installation of equipment. The MSFC engineers have planned to build only one enclosure in the upper part of LOX tank. It's volume of 56 m3 would be a very limiting factor We have however found a way of increasing the internal volume of the enclosures which the Super Skylab team have not recognized: Any untanked volume in the LOX tank also creates a correspondingly larger vacancy in the LH2 tank. If 56 m3 is enclosed in the LOX tank then there is also additional unused volume of 156 m3 in the LH2 tank. An enclosure can be built in each of the vacancies. The two enclosures should be connected by a tunnel. The payload capability for this version would be 15 750 kg according to Super Skylab data and 16 250 according to our estimation.

Concept Selection
During the initial evaluation of concepts we have considered two versions of Perun space station, the MSS and MPL. These abbreviations stand for "Minimum Space Station" and "Multiple Purpose Lab" respectively. A comparison of the two versions of Perun and the Super Skylab is shown in Fig. 3. where the habitable volume is indicated by hatching.

MSS is launched partially tanked. It bears a superficial similarity to the Super Skylab project. In both of these configurations the habitable volume is located inside hermetic enclosures in the tankage. However the habitable volume inside Super Skylab is quite low ( 56 m3 ), while in Perun MSS this volume has been increased by providing enclosures in both the LOX and LH2 tank while Super Skylab makes use only of LOX for this purpose.

For MSS the shape of the LOX enclosure is identical with the Spacelab short module. The interior could therefore be adopted from Spacelab almost unchanged. The LH2 enclosure has the shape of two half-ellipsoids. Both of these enclosures are connected by a tunnel passing through the intertank through existing manholes used normally for ground servicing of the tank. The total living volume in the tank is 212 m3. On the top of ET a passive Apollo docking port should be installed, protected by the ET Nose Cap. After jettisoning the Nose Cap in orbit a Skylab Multiple Docking Adapter should be connected to the docking system. The two docking ports of the Docking Adapter would be used one for Orbiter docking and the other one for Spacelab berthing. This version should be manned by a crew of four.

MPL: In this version the tank is launched from Earth fully fuelled. The crew then occupies the volume which was used previously for propellant storage. In this version the whole LOX tank is inhabited but not the LH2. This gives a total volume of 552 m3 - that is 1.7 times as much as on Skylab. Several Spacelab modules should be connected by special berthing ports to the LOX cylindrical section. Electrical installations, plumbing and a mechanical secondary structure should be built in prior to launch. The crew would have to perform an extensive task of pressurizing the tank and installing plug-in modular equipment brought by later Shuttle missions. For this version a crew of eight is proposed.

Discussion: Evaluation and comparison of the two concepts have shown us that the performance of MSS would be very low compared to MPL. The decreased launch mass capability would present a very serious problem with MSS allowing only a limited mass of equipment to be installed before launch. Most of the on-board equipment would have to be installed on orbit. This steals from MSS its main advantage - the immediate readiness for use after launch. Also negative effects of the cryogenic environment would have to be expected on the equipment mounted in the LH2 enclosure as the temperature of LH2 is typically 20 K compared with 90 K for LOX. Also the mass available for experiments in the single Spacelab module present with MSS would be very low.

As the development costs of both versions would be marginally the same we have decided to stop work on MSS and prefer MPL in further study.

Logistics
As a part of the Space Transportation System the Perun Space Station is to be resupplied by periodic Space Shuttle missions These missions would change crews, rotate interchangeable modules and replenish the station's supply of expendables.

At first we have considered bringing the supplies to Perun in a dedicated Logistics Module (a modified Spacelab) which would be attached to Perun and rotated by the Shuttle just like the experiment modules. That way some of the launches would be dedicated to carrying only supplies and nothing else. A similar concept has been used in many space station designs by Grumman, McDonnell Douglas and other companies.

Before commitment to such a concept we have had to carefully examine the mode of operation of the Space Shuttle. Two limits are imposed on the mass transported by the Shuttle - the launch capability (up to 29 500 kg) and the reentry limit (always 14 515 kg). That means that for low trajectories more payload can be taken into orbit than can be returned to Earth. Normally this difference is used to orbit small non-retrievable satellites. Our approach however is to use this difference to bring to Perun the expendables (oxygen, water, food, propellant) which are not to be returned to Earth. On reentry the Shuttle would carry only empty containers and an experiment module which are both fully reusable. That way we can make best use of the Shuttle's launch and reentry capability.

As a consequence every mission must bring expendables to Perun as well as a mission module. As it is above the Shuttle capability to carry both a mission module and a Logistics Module, it is clear that the Logistics Module concept is a mistake. This is true for any Space Station operating in conjunction with the Space Transportation System.

Bulk cargo (food, films, clothes etc.) should be carried on the Orbiter lower deck in containers and transferred to Perun manually by the crew. Water and gases which it would be difficult to transfer from Orbiter to Perun should be carried on board the mission modules. The containers would therefore be rotated along with the mission modules. We believe that the most favorable location for water storage is in the subfloor space of the Spacelab module structure while compressed gases should be stored in spherical bottles mounted on the outside of Spacelab endcones.

Trajectory
The selection of trajectory for a space station depends on a large number of factors. Three parameters are to be selected: the orbital inclination, altitude (presuming that the orbit is circular) and longitude of ascending node (LOAN). For Perun we have selected an orbital inclination of 56° for there reasons:

l) This inclination is big enough to permit extensive Earth observation activity, while the STS launch capability for this inclination is still sufficient.

2) It is the only inclination to which Space Shuttle can be launched from both STS launch sites - the Kennedy Space Center and the Vandenberg AFB. This is a positive factor for the operational capability of the Space Station.

The determination of orbital altitude (390 km) is the result of further trajectory/logistics optimalization to be described in section 10 of this paper.



The Longitude of Ascending Node (LOAN) although it may seem unimportant at first sight has a very strong influence on the length of exposure of the station to the sun. We have expressed a formula where the percentage of solar illumination is a function of LOAN and time of the year. The presumptions for this function are: circular orbits of Perun around the Earth and of, the Earth around the Sun, spherical shape of Earth, orbital period substantially shorter than the year and negligible angular dimensions of the Sun. Fig. 5 shows the percentage of solar illumination versus time for three cases of LOAN: 0°. 90°,and 180. 180 is the orbit with minimum * * o illumination however the most stable. LOAN of 0 gives an orbit with sharp variations of illuminations ranging from 60% to 100%. In the 90 case the illumination varies slightly from 60%. to 75%. We have used the 90° case as a baseline for all the calculations in this paper.

The altitude of the Space Station will decrease gradually due to atmospheric drag. Presuming that the average density of atmosphere is 6.2 * 10~12 kgm-3 at 390 km we have calculated an average drop of 25 m per orbit. However this value can differ substantially due to the day/night cycle and solar activity. It can vary from as much as 90 m per orbit down to 1.3 m per orbit. The above data were calculated for the standard orientation of Perun described in section 13. The values correspond with those published by Grumman (197?) who foresee an average force of 0.5 N-acting on the ET. The lifetime of the facility according to our calculations is 4-8 months. This is significantly less than for most other spacecraft because the frontal area of ET is very large. The frontal area to mass ratio for Perun is about 4 times greater than for Skylab. This justifies the quicker downfall of Perun. A Propulsion System is necessary for cancellation of the atmospheric drag to keep the station at a stable altitude. This system is described in the next section.

To achieve a stable orbital altitude trajectory corrections should take place approximately every 10 days.

Propulsion System
This system serves for compensating the atmospheric drag and for other trajectory corrections, not for attitude control. We have examined 4 different systems for this purpose: cold gas thrusters, hydrazine monopropellant, monomethylhydrazine nitrogen tetroxide and electric ion propulsion. An evaluation of the mentioned propulsion systems has shown us that hydrazine monopropellant is the most likely candidate system for this purpose because of its high specific impulse and simplicity of hardware.

We have also assumed that the propulsion system should be detachable from the space station to be taken to Earth for refurbishment and refueling.

These presumptions have led us to design the Propulsion Module (PM) for this purpose. In accordance with the trend to use offshelf hardware in the design we have derived the PM from the Teleoperator Retrieval System (Martin Marietta, 1978) - the spacecraft which should have reboosted Skylab.

The Propulsion Module for Perun (Fig.6) consists of a pressurized tunnel surrounded by 4 TRS propulsion kits. Each of these kits weighs 177 kg empty and holds 695 kg of hydrazine monopropellant. The total thrust of the 4 propulsion kits is 3500 N. The pressurized tunnel has an international docking system from ASTP on one end and a special berthing port (to be described in sect.13) on the other end. The berthing port allows the PM to be attached to Perun while the ASTP docking system serves for docking with the Orbiter. PM would be carried on the initial Perun launch. As it is not possible to berth the PM to the top of ET by Remote Manipulator System the following sequence was adopted for the berthing1: PM would be connected to the docking system of Orbiter while the ET and Orbiter are still mated in the launch configuration. After this the Remote Manipulator System would get hold of some suitable part of ET and fire explosive bolts to separate from the ET. By movement of the manipulator the Orbiter would approach the top of ET and perform berthing.

PM has a total weight of 4400 kg fully fuelled and 1600 empty. It should be exchanged by Orbiter periodically and taken to Earth for refuelling, refurbishment and reuse. How often PM should be exchanged is a subject of study within the trajectory/logistics optimalization in section 10. The Propulsion Module could serve not only with Perun but also with free flying Spacelab modules, space telescopes and other spacecraft It should be noted that a very similar Propulsion Module could be designed through modification cf the aft section of USSR's Salyut space station.

Trajectory / Logistics Optimization
For the operation of a space station to be as inexpensive as possible it is necessary to reduce the number of STS flights for it's support as space transportation still is the most expensive part of the space station budget. This can be done by designing the operational pattern of the space station so as to increase the interval between regular Orbiter revisits. This interval equals the length of time which it takes for the crew and the propulsion system to consume all the expendables (consumables + propellant) brought by the Shuttle. The total mass of expendables equals the difference between the launch capability of STS and the reentry limit as written in section 7« Interval T between Shuttle revisits then equals the ratio of total expendables mass to the daily consumption of expendables.

$$ T = \frac{Launch\ Capability - Reentry\ Limit}{Consumables\ Per\ Day + Propellant\ Per\ Day} $$

The fixed parameters for this function are: reentry limit of 14 515 kg and 12 kg consumables per man day. The functions for launch capability versus altitude and for propellant consumption versus altitude were obtained by interpreting graphical information from references (NASA, 19?6j Grumman, 1977), The resulting function T(A) appears in Fig. 7 in full line, compared for three different crew sizes.

This function is not exhaustive^ because it deals with the mass of PM propellant as with a continuous function, while the propellant is brought only in integer multiples of full tank capacity. The dashed lines show the time till propellant depletion versus altitude. The numbers above the dashed lines indicate after how many Shuttle revisits should PM be exchanged. It is desirable for the propellant to last just as long as other consumables. This condition is fulfilled only in points of intersection of the full line with the dashed line.



Fig.7



This way we obtain 4 points for every crew size which indicate combinations of orbital altitude and resupply interval depending on how often PM is to be exchanged. The goal now is to select that point in which T is biggest. For a Perun MPL the results of the optimalization are indicated in the graph by a circle:

crew size 8 men altitude 390 km revisit after 80 days PM exchange after 160 days

This optimalization with slight modifications can be applied to any space station operating in conjunction with the Space Transportation System. As far as logistics is concerned, there is nothing specific about Perun.

Power Supply
Power supply for Perun is based on photovoltaic solar arrays. We have selected a very versatile type of array which is now being developed for use on the Solar Electric Propulsion System ( Freatag,1978; Bekey, 1978). This deployable solar array gives an output of 12.5 kW and measures k by 32 meters. It is particularly suited for our purpose because it can be retracted for maneuvers and subsequently redeployed. It is very light (200 kg) which permits it to be mounted on the tank exterior (using preinstalled locks) by a crewman riding a Manned Maneuvering Unit the so-called space scooter (Martin Marietta, 1978). Another alternative is to mount the arrays into position by the Shuttle Remote Manipulator System arm (NASA, 1976).

A large number of these arrays could be mounted on the ET exterior. For our project, however, 6 arrays are sufficient with a total output of 75 kW. Presuming a 65% exposure to the sun according to sect, 8 the average output is 45 kW. We have calculated the full average energy consumption of Perun to be 40 kW.

A set of accumulators is necessary to supply power during eclipses, maneuvers and moments of peak energy demand. NiCd batteries are very advantageous for this purpose because of their long lifetime. A battery with a fairly large capacity of 100 kWh would weigh about 2500 kg.

Thermal Control
Thermal control of the space station was a very challenging problem because ET is covered with polyurethane foam insulation prior to launch which chars and turns black during ascent due to aerodynamic heating. The black surface complicates the thermo-optical properties of the structure and the insulation prevents the dissipation of waste heat from the walls of the tank.





The main purpose of the insulation is to prevent ice forming on the cold walls of the tank which might fall off during launch and cause damage to the tiled surface of Orbiter. Therefore the insulation is necessary primarily on the tank surface adjacent to the Orbiter. That means that on some parts of the tank which are far enough from the Orbiter insulation could be replaced by white paint. This is the main idea of our design of Perun's thermal control. The uninsulated portion of the tank would serve as a radiator to dissipate waste heat from the space station. Ve have designed the uninsulated surface to cover one half of the LH£ tank - that half which is opposite to the Orbiter of course. During orbital operation the radiator would be oriented away from the sun to be kept as cold as possible. A surface of this size is capable of radiating at least 75 kV of waste heat which is the full energy production of Perun. Although it seems unlikely a white radiator is better for this purpose than a black one because the i.r. Emissivity of the. two is approximately equal while the absorptivity of the white surface is much lower (Zerlaut, 1965). Heat should be transferred to the radiator through, a cooling loop operating with a mixture of water and glycol. Heat exchange between the fluid loop and the radiator would take place in a system of aluminum pipes welded on the inside of the LHp tank. Pump packages from Spacelab should be used for fluid circulation.

The absence of insulation on a part of the LH2 tank would lead to prelaunch heating of propellant. Our preliminary estimations were quite favorable however more exact calculations show an increase of temperature of approximately 2.5 K per hour which is not acceptable. This problem could be solved by circulating liquid helium through the pipes of the thermal control system before launch.

We know that excess ice forming on the ET surface has been considered a problem recently during the development of the Space Shuttle System. However we believe that the uninsulated surface of ET is smooth enough for all ice to fall off immediately after liftoff without any negative impact on the operation of the Space Shuttle.

It is also necessary to protect the station from direct sunlight which could cause sharp variations in temperature between the light and dark orbital periods. For this purpose we have designed a sunshade of metallized foil to be erected on the sunny side of the LH tank (see Fig.8). It has the shape of a long rectangle and is supported by the Orbiter attach points (another case of using existing hardware).

Another cone shaped foil should be deployed to surround the inhabited LOX tank. It not only prevents sun from shining on the station but also minimizes radiative heat losses from the inhabited section which could cause low temperatures inside. This sunshade would be brought into orbit folded in toroidal shape, wrapped around PM and subsequently  deployed by motor pulled strings to wrap the ogive section of the LOX tank. The cylindrical section is left uncovered because it is the place of Spacelab module berthing. EVA Support by at least two crewmembers would be necessary for the deployment of both sunshades.

We are aware that fine thermal control is essential to a succesful space project. To achieve good thermal behavior of the space station we have developed the computer program TERKOT, a mathematical model of the thermal structure of Perun. It was one of the most important parts of our study. The title Terkot is a Czech abbreviation of the words Thermal Control of Space Object. The program takes into account the radiant heat exchange, thermal conduction, thermal capacity and cooling fluid convection.

For purposes of math modelling we have broken the structure down to 240 nodes. Every node has been assigned a thermal capacity, heat input(from Sun, Earth and internal sources), radiating areas to the interior and exterior and thermal conductivity toward the adjacent nodes.

The program works in a cyclic mode. During every cycle it computes the heat input or output for every node and makes appropriate temperature adjustments. The greatest problem we have faced during the development of this program was the selection of the optimum length of the time cycle. The problem was that for long cycles the temperatures tended to oscillate instead of approaching a steady state. On the contrary, very short cycles increase the necessary computing time to unacceptable levels. We have solved this problem by using a series of succesively shorter time cyckles. During the initial long cycles the temperature approached the steady state roughly while still oscillating and then during shorter cycles the oscillations were damped to give a real steady state.

We have also developed a special program for the graphical output of data from the Terkot program. It performs a nonlinear polynomic interpolation of temperatures between the nodes and then charts the isotherms e.g. the lines of constant temperature. An additional transformation of coordinates has enabled us to depict a system of isotherms on the tank surface as shown in Fig. 9.



It should be emphasized that the software described in this paper was developed exclusively by undergraduate students without the help of a professional and that it may be modified for applications in the near future. The programs are universal and can be used for many different purposes by just changing input data.

We have used the Terkot program to conduct a thermal analysis of the Perun Space Station. The analysis has verified that the Perun thermal control system was capable of providing comfortable working conditions for the crew in the inhabited section (22°C) while dissipating into space the full output of waste heat Fig. 9. shows steady state temperatures during normal operation of the Space Station. For the simulation we have presumed the facility to be exposed to the sun for 65% of time (see sect. 8) and a constant thermal radiation from Earth.

The algorithm of simulation of the Terkot program was tested on a Hewlett Packard HP 9825 desktop computer. The final program has been run on ICL-4-72 and written in Fortran IV. The Hewlett Packard HP 9845 S system with a HP 98?2 plotter was used for the graphical outputs.

Operations
ET is launched into orbit having been somewhat modified. The modifications include the installation of Life Support System, Attitude Control System, a modified LOX tank interior, berthing ports in the LOX cylindrical section and on the top of the tank etc.

After orbital insertion the crew install sunshades and solar arrays and berth the Propulsion Module to the station (PM is necessary for the next docking). Perun is then left unmanned till next visit. Four more Shuttle launches are required to bring to Perun 4 Spacelab modules carrying materials for rebuilding the tank. These include plug-in equipment mounted in 78 Spacelab type small racks, 60 prefabricated panels 1OO by 250 cm to serve as partitions, excercise equipment etc. The initialization period of rebuilding should take about 5 weeks for an initial crew of k. Then regular operation may start. The crew for Perun is 8 persons. This crew size has the advantage of potential rescue by a single Orbiter (providing that no mission specialist is present).

The crew of 8 is divided into 4 subcrews of 2 men. One subcrew is technical, responsible for the functioning of the Space Station and the 3 remaining subcrews are mission specialized. Three Spacelab modules are always present with Perun, one for each mission subcrew. The operational pattern is designed so that Spacelab exchange is coherent with subcrew exchange meaning that the subcrew would work only in their assigned Spacelab - this simplifies training. This pattern of exchanges is to be seen in Fig.12. where exchange of a subcrew or of a Spacelab is indicated by a dot on the appropriate line. It can be seen that up to two successive subcrews would work in the same Spacelab because the interval between Spacelab exchange is typically 320 (sometimes l60) days while crews are exchanged after 160 days. This duration of orbital stay has recently been demonstrated by the Salyut 6 cosmonauts. Orbiter visits take place regularly every 80 days to bring consumables, exchange Spacelab (and sometimes PM) and rotate two of four subcrews.

600px|Operational Pattern of Perun

As Perun is equipped with an international docking system docking could be accomplished not only with the Shuttle orbiter but also with other spacecraft e.g. USSR's Soyuz or Progress or the proposed French minishuttle Hermes.

One of the Spacelabs (a double module) is always equipped with an international docking- system and a self contained life support system so that it can serve as a lifeboat in case of emergency. Because the docking- system is not to be used during normal operation a garbage container would be berthed to it. This container (full) should-be jettisoned before returning the Spacelab to Earth.

Perun is designed to be oriented by its long axis toward the Earth. This is advantageous for various Earth-oriented applications. It can rotate about its axis for its solar panels to track the sun. This orientation is shown in Fig. 8. One other advantage of this type of orientation is that it can be kept passively by gravity gradient forces. However if a precise orientation is required for experiments or for maneuvers Control Moment Gyros have to be used.

The mass of Perun MPL in its complete configuration is approx. 125 tons with the center of gravity located in the LOX cylinder section about 0.5 m from the cylinder/intertank junction.

Interior






The interior of the Space Station is designed with respect for the fact that it must be rebuilt from a tankage with a minimum of manual labor which is quite exhausting in weightlessness. The tank is launched from Earth with a modified secondary structure inside including grid floors, electrical wiring, plumbing and racks for the installation of plug-in equipment.

Access to the tank is through an existing manhole at the top of the tank which should be equipped with a special berthingport (described in section 15). At the top of the tank is a very small enclosure (13 m3) which is pressurized before launch. This enclosure contains the Life Support System which is ready to use because it's operation is necessary before rebuilding of the tank can commence.

The rest of the tank is divided into 4 floors: the habitation floor, the control center, lab and the sleep + hygiene section (See Fig.14)

The habitation floor is used by the crew for eating and leisure It also holds much subsystem equipment mounted in racks surrounding the tank walls. The racks in all parts of Perun are identical with those used on Spacelab. It would therefore be possible to transport most of the onboard equipment for Perun in Spacelab modules mounted directly in the Spacelab racks.

The purpose of the Control Center is clear. A new generation of Command and Control System is envisioned here. This floor also holds a medical lab and exercise equipment.

The lab is situated in the LOX cylindrical section where 4 berthing ports for Spacelabs are located. This provides direct access from the lab (where mission independent equipment such as computers is located) to Spacelabs which carry mission specific equipment.

The sleep and hygiene section provides each crewmember with an individual room (floor area 4 m2, height 2.5 m). Every two crewmembers share a hygiene compartment (area 2.5 m2) which includes a toilet, a shower and other necessities.

All the floors are separated by Skylab type grids except the sleep and hygiene section where a solid floor, ceiling and partitions are to be installed using the 60 prefabricated panels brought in the initialization Spacelabs.

Systems
The onboard systems were designed with the maximum use of offshelf hardware. Some of the systems are installed into the Space Station before launch and some are added subsequently in orbit.

The systems which are installed prior to launch are the ones which are crucial for the operation of the facility. This includes the Attitude Control System, Remote Control System, Thermal Control System, Environmental Control System and the Berthing System.

The Attitude Control System comprises 6 Skylab type Control Moment Gyros located in the intertank section. This utilizes the large unused volume in the ET intertank. The gyros can be desaturated conveniently by firing off-axis engines of PM. As a back-up Attitude Control System we propose a set of cold gas thrusters mounted on the intertank walls and near the tip of ET. Compressed nitrogen bottles can still be located in the intertank. Other systems located in the intertank are the Remote Control System which provides telemetry and control during the initial unmanned phase of operation and the Thermal Control System which operates the cooling loop leading to the LHg tank radiator. It should be noted that the temperature in the intertank will be very low before and during launch. Any electronics or precise mechanical equipment mounted in this section must be equipped with adequate insulation and possibly electrical heating. Also the storage of cooling fluid in this section would present a problem just like the on-orbit filling of the fluid loop.

The Environmental Control System is located in the small enclosure at the top of ET. Because of safety reasons it must be located inside the pressurized part of the Space Station. This system uses a set of air ducts which lead air to and from every floor. Air can pass freely through the grid floors to prevent the forming of any "dead spots" in the atmosphere. The velocity of air circulation we have selected at 6-10 meters per minute. For this we have drawn experience from the results of the Czechoslovak experiment "Tepelná výměna" (Heat Exchange) which was performed aboard Salyut 6 and for the first time ever has provided a comparison between objective measurements of skin temperature and subjective feelings of the crewman. The Environmental Control System can be composed mostly of Spacelab hardware.



The Berthing System is very specific for this project. We believe that in the near future any Space Station will need a means of berthing dedicated modules. There are indications that a berthing port for this purpose is already under development in the USSR. Existing docking systems are not suitable.for this purpose bacause of their high mass and insufficient dimensions of passage tunnels. On the contrary the berthing port does not need their capability to damp great angular deviations and excess velocity because the berthing operation can be performed with the precision of several centimeters by the Shuttle Remote Manipulator System. These presumptions were used by us when designing the Berthing System for the Perun Space Station. Ve have also come to the conclusion that one part of the system must be completely passive with no moving parts. We have had to observe a maximum outer diameter of 147 cm dictated by the dimensions of ET structure below the Nose Cap where one of the Berthing Ports is to be located. The Berthing Port must permit not only the passage of crewmembers but also electrical and hydraulic connection of the, two segments through connectors mounted on the berthing interface. In our design the segments are joined by four probe/droque units.

The probes are mounted on the passive segment and have no moving parts. All the mechanical movements associated with the berthing take place in the active segment where the probe is first aligned by the drogue and then captured by 6 thrust arms which do not allow it to excape. In this moment the two docking interfaces are still 3 cm far from each other. The final contact and compression is accomplished by DC torque motors which pull the structural ring to which the thrust arms are connected so as to force the probe deeper inside the active segment. The unberthing sequence is reverse plus at the end 6 Disconnect Levers (one for each thrust arm) go into action to free the probe. The system was calculated for a compression force of 40 tons (lO tons per probe). The passage tunnel has a maximum diameter of 130 cm and minimum of 95 cm which allows single Spacelab racks to be transported through it.

Cost Estimation
To estimate the costs for the development of the Perun space station we have used the following' method:

We divided the structure into parts of high, medium and low technology and offshelf hardware. We have then used statistical development costs per kilogram of hardware: 10 000 5 000$ and 1000$ for high, medium and low/offshelf technology respectively. Other costs e.g. space transportation costs, crew training, ground support etc. were added subsequently.

TABLE 1 - Cost Estimation for Perun MPL

A large part of Perun's operational capability could be made available to commercial users for a nominal price. This money could pay back most of the investments into the project.

Conclusion
In this paper the members of the Student Working Group for Astronautics of the Planetarium of Praha have explored the possibility of converting an expended Space Shuttle External Tank into a space Station. During the preparation of this paper we have concentrated especially on the following subjects: Thermal control, logistics, trajectory and economy as well as on-board systems of the Space Station.

Though our concept is not original we believe that several ideas do appear in this paper which deserve further attention.

We have examined just one version of a space station utilizing the Space Shuttle External tank structure. This space station should carry a small permanent crew and it's main role is to support several Spacelab modules serving for scientific research. The research program could include experiments in biology, medicine, astronomy, space physics, materials processing, Earth observation and many other fields.

With the advent of space manufacturing coming near one can foresee a version of Perun operating in low inclination orbit (to increase STS launch capability) with the Spacelabs replaced by space fabrication modules (Spacefabs ?) as described by D.M.Waltz (1977). Such a space station would become the base for commercial exploitation of space.

Perun could also become an orbital depot serving for refurbishment, refuelling and repairs of satellites, space tugs, space telescopes, free flying Spacelabs and other vehicles. A space hangar could be established in orbit by jettisoning the aft bulkhead of the LH2 tank. The large size of the tank would permit the servicing of any spacecraft which is a part of the Space Transportation System. These are only a few of the great number of applications of the Space Shuttle External Tank used as a Space Station.

Acknowledgments
This paper has been prepared by the members of the Student Working Group for Astronautics of the Observatory and Planetarium of Praha. The authors would like to express their gratitude to the directoriate of the institution. Thanks are due to Ing. Marcel Grün for supervision of the project.

Also the support for the preparation of this paper should be accorded to Prof.Ing. R. Pesek, RNDr. P.Lala, RNDr, A. Vitek, J. Kroulik, RNDr. L. Lejcek and others whose assistance and consultations have been very helpful. Recognition is due to the representatives of the Hewlett Packard Corporation and of the Computing Centers for permitting the authors to use their facilities.