The SKYLAB Apollo Telescope Mount

SKYLAB and the attached Apollo Telescope Mount (ATM) were launched into near earth orbit by a Saturn V rocket on 14 May 1973. The Apollo Telescope Mount (ATM) contained two NRL ultraviolet instruments for photographing the Sun and its spectra. Further information of this and the other scientific instruments can be found in a special issue of Applied Optics, Volume 16, No. 4, April 1977.

Apollo Telescope Mount

Adapted from: R. Tousey, "Apollo Telescope Mount of Skylab: an overview", Appl. Opt, 16, 4,825 (1977)


Space research commenced in the United States in 1946 when captured V-2 rockets became available for use as high-altitude though short-lived observing platforms. As suggested by E. 0. Hulburt, then Superintendent of the Opt-ics Division, at the Naval Research Laboratory (NRL) a program of research began on the short wavelength electromagnetic radiations from the sun, long searched for from mountain tops and balloons but never detected because of absorption by the overlying ozone layer. The first far uv solar spectrum was, indeed, obtained on 10 October 1946, less than a year after inception of the program.
Jumping nearly three decades, with the overwhelming success of the experiments in the Apollo Telescope Mount of Skylab, 1974 saw the attainment of a climax in solar research. This introductory paper describes Skylab and the course of events that led to this most complex of all space projects. In particular, it covers the ATM and its instruments and the method of operation of the ATM mission.


In 1958 the National Aeronautics and Space Administration was created, a consequence of three successes: Sputnik; the Army's Jupiter C or Explorer; and NRL's Vanguard. Many of the NRL space research team decided to join the new organization. Among them were Homer E. Newell, Jr., Head of the Atmosphere and Astrophysics Division at NRL, and John C. Lindsay, then a member of Herbert Friedman's group that pioneered solar x-ray research using rockets. At the Goddard Space Flight Center (GSFC) Dr. Lindsay supported by Dr. Newell at Headquarters, initiated OSO, the Orbiting Solar Observatory. The first OSO was placed in orbit in 1962. Among other experiments it carried the GSFC scanning XUV monochromator that monitored the total solar flux from 171 A to 400 A with 0.85-A resolution. NRL had extensive instrumentation in the second of the series, OSO-2, securing interesting results on solar x rays, as well as the first solar images in Lyman-a of hydrogen, 1216 A, and in He 11 304 A to be recorded photoelectrically from orbit.
The OSO series of spacecraft, constructed by Ball Brothers Research Corporation (BBRC), included instrumentation prepared by GSFC, NRL, Harvard College Observatory, and many other institutions. The seventh OSO was launched on 29 September 1971 and ceased operation on 9 July 1974 because of reentry. This carried the first successful orbiting white light coronagraph together with several other experiments built by NRL and an excellent XUV spectrometer/ spectroheliograph constructed by GSFC."

As a result of a recommendation by the 1962 Iowa Summer Study conducted by the Space Science Board, NASA commenced to develop an Advanced Orbiting Solar Observatory (AOSO) capable of carrying much larger instruments than OSO. NRL, GSFC, American Science and Engineering, Inc., Harvard College Observatory, and High Altitude Observatory were successful in having proposals accepted by NASA in 1964. In 1965 contracts for construction of the various instruments were placed, and the work began. But soon it became apparent that AOSO was ahead of the times and beyond the NASA budget. The entire project was canceled in December 1965.
By then the Gemini program had been outstandingly successful in placing man in orbit. A few simple experiments that were designed for operation by manin-space were included, for example, the NRL Air-glow Photography project with M. J. Koomen,13 Principal Investigator (PI); but none was flown for the observation of solar radiation.

Rescue of the AOSO experiments must be attributed to opportunism. Although Apollo had not then been launched, consideration was already being given to the use for space research of hardware expected to be left over after the lunar landings, in a program first called the Apollo Extension System (AES) and later the Apollo Applications Program (AAP). Among possibilities was the Astronomical Telescope Orientation Mounting (ATOM), again one of John Lindsay's ideas, which had been developed during 1965 by BBRC to the stage of a proposal. This could have carried some of the lamented AOSO experiments on a spar, stowed in one bay of the Apollo Service Module. Erection of the spar and operation of the instruments were to be carried out by the crew.
The first firm plan for flight of the AOSO solar instruments in the AAP made use of the ATOM. But it was soon realized that there was a better way. Also the amusingly inappropriate acronym ATOM was shortened to ATM, for the Apollo Telescope Mount. The new idea came from the Marshall Space Flight Center (MSFC); it was to build the ATM and its control center into a Lunar Excursion Module (LM) and to use the command module (CM) as sleeping quarters. This second plan was accepted in mid-1966 and was pursued at MSFC with high priority for a target launch date of December 1968. But in July 1967 it became apparent that the full complement of ATM instruments could not be made ready in time for so early a launch date. The ATM PI's were urged in every way possible to accelerate their schedules; Harvard agreed to drop one of its two experiments, and NRL simplified its XUV spectrograph (SO82B) experiment. In return, the PI's were promised a second flight about 2 years later; this would include all the original group of experiments.

Soon after this, however, NASA produced an even better plan: the LM would still be used but would remain connected through a multiple docking adapter tQ the expended Saturn IVB hydrogen tank that had placed it in orbit; the crew would be launched in a command and service module (CSM), dock with the LM/ATM, and set up housekeeping in the hydrogen tank used in launching the LM/ATM. This involved converting the hydrogen tank into an orbital workshop (OWS). After the first crew had finished their work and returned to earth, another crew would come up, make use of the completed living facilities, and operate the ATM.

This third plan was followed until mid-1969, when NASA had yet a better idea. Instead of fitting out the OWS for living while actually in orbit, an empty Saturn IVB hydrogen tank would be fitted out before launch with every possible aid for living and working in orbit. This was called the dry, or Saturn Workshop (SWS), in contrast to the previous version, which was called the wet workshop. The ATM would be linked to the docking adapter and workshop in such a way that, once in orbit, it could be unfolded into the operating position. The entire complex was christened Skylab and would be placed in orbit with a Saturn V. Three visits were planned, for 26, 56, and 56 days; it was hoped to operate the ATM and other experiments over a period of 9 months. This was, indeed, the final plan, and it was found possible to extend the visits to 28, 59, and 84 days.

Skylab (SL-1) was launched on 14 May 1973 and went into orbit as planned. But when the various parts were unfolded from the stowed configuration of launch, it was discovered that something was wrong. Intensive studies quickly led to the conclusion that (1) the meteorite and thermal shield and one of the two solar cell arrays had been torn from the SWS during the boost phase; and (2) the other solar array had failed to deploy fully. It appeared that the SWS would be too hot for the crew to live in unless something could be done. This lead to a crash program, in which JSC designed and constructed a large parasol, to be deployed through the solar-facing instrument airlock, and MSFC prepared a sailtype sunshade to be set up during extra-vehicular activity (EVA). All this, and the necessary crew training, were accomplished in little more than 1 week, so that this gear was carried up by SL-2, launched on 25 May.

The crew of SL-2 was the first in history to perform major repair work in space. They erected the parasol, and their living quarters gradually became habitable. On 9 June Astronauts Conrad and Kerwin in EVA succeeded in deploying fully the remaining SWS solar array and so added about 5 kW to the 1.1 kW of electrical power provided by the ATM array. The MSFC sail was not used during the first mission, but during SL-3 it was deployed on top of the parasol to provide additional cooling. This picture shows Skylab in a photograph taken from SL-4 with one SWS solar array gone and with the MSFC sail covering the parasol except for some small sections that can be seen projecting beyond the sail's edge.

In addition to this remarkable feat, which rescued Skylab, a large number of important repairs and changes were made during the mission. SL-3 and SL-4 both carried up replacement subsystems and some new components. The success of the Skylab crews in making repairs and changes of aR kinds was astonishing. It has proved the value of man in space.
Thus AOSO-to-ATOM-to-ATM/CM/LEM-to-wet OWS-to-dry OWS (Skylab)-to-launch in May 1973 to final splashdown in February 1974 covered a period of about 10 years. During the decade spent preparing Skylab a large number of individuals at each institution having an experiment in ATM and also at BBRC became involved, and there were many whose roles were critical. The final ATM teams, which included the key personnel involved in the actual mission, numbered something like 150 individuals. Many of them are now heavily involved with the analysis and interpretation of the results.

Those of us who have been involved in the ATM program from its commencement are keenly aware of the large number and great diversity of the experiences we have been through. We have weathered crisis after crisis, found solutions to seemingly unsolvable problems, and learned a great deal in many new fields, especially about large scale engineering methods. I believe that we now understand a little better how engineers think; and on the other hand, I hope that the engineers have learned something of the stringent re quirements of scientific experimentation and have come to understand the way scientists think. It has been a fascinating and highly rewarding experience.

ATM Solar Observatory

In its final version Skylab consisted of four principal parts. Largest was the Saturn Workshop (SWS), which provided living and working quarters for the three-man crews. It contained also most of the experiments other than the ATM. The hydrogen tank of a Saturn IVB booster (second stage of Saturn IB) that was converted to become the SWS was a cylinder 15 m long and 6.7 m in diameter. This was where most of the corollary experiments were located, some of which are described in other articles in this issue.

The second major section was the Multiple Docking Adapter (MDA), center of the space complex, 5.2 m long and 3.2 m in diameter. This was attached to the SWS through an airlock module (AM) and contained the control and display panel for the ATM. It had two docking ports: the first port, located on the long axis extending through the SWS, was for docking the CSM when on each visitation it came up with its crew of three; the second port, 90' away, was an alternate dock for a rescue mission, should it be required. Opposite the latter port was the ATM. The third part was the CSM itself, much the same as the Apollo Spacecraft developed for the lunar landings. These sections can be seen in this photograph .

The principal subject of this paper, however, is the ATM and its solar instruments. As shown this fourth part of Skylab consisted of two concentric elements. The outer framework, called the rack, was an octagonal structure 3.4 m across and 4.4 m in length. Inside the rack was a cylinder or canister, 2.1 m in diameter and 3 m in length. This canister housed the solar instruments. The canister was attached to the rack through a pair of large gimbal rings carried on a large ring bearing that permitted rotation of the entire canister around its axis. There was also a set of four large solar cell arrays that provided as much as 1.1 kW power. This comprised the solar observatory known as ATM. Inside the canister was a stiff cruciform structure that ran down its entire length, dividing the space into four quadrants, and forming the optical bench to which the ATM instruments were attached.

The essential feature of the ATM was its ability to keep the instruments within the cannister aimed steadily and precisely at the desired point on the sun regardless of disturbances such as those caused by crew movement. The specifications were +2.5-sec of arc stability in yaw and pitch and 5 min of arc in roll. Although we never believed this precision achievable, it was actually exceeded. Short-time (minutes) yaw and pitch stability was @ +0.5 sec of arc, and short-time roll stability was excellent.

To achieve this stability and freedom to point it was necessary first to stabilize the entire Skylab. This was done with three control moment gyros (CMG), with axes mutually perpendicular, one being redundant. Each CMG was a double-gimbal-mounted, electrically driven, 53-cm diam rotor of 144-lb mass, that required 14 h to spin up to 9100 rpm. Each CMG stored 2300-ft-lb sec angular momentum. By energizing the proper torque motor, a CMG could be torqued by command around either or both gimbal axes by as much as 122 ft-lb in steps as small as.16 ft-lb. The entire control system was automatic, operating from gyros but with manual overrides. This system maintained the orientation of Skylab to 3 min of arc. Dumping the excess momentum was a maneuver executed at ni,-ht, usually against gravity-gradient torques.

But the solar instruments required an order of magnitude more stable pointing than could be achieved with the CMG'S. This was accomplished with a solar pointing control system (PCS). The PCS sensed the sun's center to a few tenths of a sec of arc and sent error signals into the torque motors that controlled the rotational positions of the ATM canister gimbals. Offset pointing in yaw or in pitch by steps of 1.25 sec of arc up to 24 min of arc could be introduced by counterrotating a pair of quartz wedges placed in the solar beam incident on the yaw solar sensor, or a similar pair for pitch. (These solar sensors were one of the few items inherited from AOSO.) Control of offset pointing by rotating the prisms was accomplished by the crewman with his panel joystick. Digital indicators read out yaw and pitch to I sec of arc and roll to I min of arc.

A most important subsystem was the digital computer. This computer enabled the crewman to initiate complex commands of all sorts quickly and simply. The digital computer could also be operated by ground command, while the crew was sleeping or in case of emergency. It made possible the operation of many of the experiments while the crew was asleep and also during the unmanned periods between missions. To take care of all possible failures and to provide flexibility, backup modes of operation of many kinds using the digital computer were worked out. The degree of sophistication of the ATM and Skylab system was a matter of astonishment to those of us not acquainted with the power and elegance of modern engineering.

A problem in the design of the ATM, recognized from the beginning, was the great thermal stability required to preserve the focus of ultrahigh resolution optical instruments such as in ATM. It was discovered rather early, but not until much work had been done, that passive thermal stabilization was not sufficient to meet our requirements. This led to the construction of an active fluid-cooling system for the ATM canister. The refrigerator consisted of a loop, located inside the skin of the canister, through which water/methanol was circulated to radiators on the outside of the canister and facing space. This system maintained the temperature of the canister wall to 50 5 F with cyclic variations of 3'F. In addition, several experiments had their own thermal-control heater systems, designed to maintain the temperature at all locations to within 2.5 F of the design and calibration temperature and to limit the rate of change to no greater than 0.0 F/5 min.