The Gemini spacecraft was a conical structure nearly 5.8 m high, 3.05 m in diameter at its base and weighing over 3175 kg.
The spacecraft was designed to endure the aerodynamic pressures, temperature loading, vibration and acoustical noise of launch; the temperature and vacuum of orbital flight; and the extreme heat of reentry, and the impact forces of water landing while providing life support for two astronauts and the necessary equipment for planned missions and experiments.
Gemini's design reflected the knowledge obtained from the development, manufacture and flight operation of the Mercury spacecraft, which, like Gemini, was produced for NASA by McDonnell Aircraft Corporation, St. Louis, Missouri.
Gemini was launched by an Air Force Titan II launch vehicle built by the Martin Company.
The spacecraft consisted of two major parts, a reentry module and an adapter module. The reentry module was designed to withstand the extreme heat of reentry. Its side walls were protected by heat resistant shingles and the large bulkhead by an ablative heat shield. The sections of the adapter module remained in decaying orbits and were burned up during reentry.
The spacecraft was primarily "skin-stringer" construction. Ring stabilized stringers carry near all axial loads. Structural materials and construction methods exhibited the influence of Gemini engineers' search for optimum strength-to-weight ratios. Titanium and magnesium were the principal metals used.
The reentry module (the dark portion of the spacecraft) had three primary sections: rendezvous and recovery (R and R), reentry control system (RCS), and a cabin section. Integral to the reentry module was a heat shield attached to the large end of the module. An aerodynamic cover protected horizon sensors at the midpoint of the reentry module.
Materials used in the conical section of the reentry module varied greatly because of the effects of reentry heating. Super alloys such as Rene 41 and L-605 were used for the outer skin and skin attachments, which were thermally isolated from the inner structure by Johns-Manville MIN-K, Fiberglas and Thermoflex RF insulation. The basic load- carrying shell was titanium. Aluminum was used inside the cabin where heat was not a structural problem.
Rendezvous and Recovery (R and R) Section
The R and R section provided sufficient volume for a rendezvous radar system and a parachute landing system. Structural rings, stringers, and bulkheads were made of titanium. The external surface was covered with beryllium shingles. The nose fairing was reinforced plastic and Fiberglas-laminate. The R and R section was attached to the RCS section with 24 frangible bolts, which were fractured to jettison the R and R section upon deployment of the pilot chute following reentry.
Under the beryllium shingles were Thermoflex RF blankets held in place by a titanium mesh attached to the stringers. The outer surfaces of the rings and stringers were insulated with 0.038 mm Inconel-foil-encased Min-K held in Fiberglas channels.
Reentry Control System (RCS) Section
The RCS section housed reentry control system fuel and oxidizer tanks, and thrust chamber assemblies. This section was located between the rendezvous and recovery section and the cabin section. The cylindrical RCS section was an inner skinned ring-stringer titanium structure with an outer skin of beryllium shingles. External heat protection was essentially the same as for the rendezvous and recovery section.
The cabin of the Gemini spacecraft was a truncated cone (a conical enclosure with approximately the top one-third cut away), which housed the Gemini crew, electrical, and life support equipment, and various experimental devices. The pressure vessel (crew compartment) provided adequate space for the two-man crew plus instrumentation and life support equipment.
The pressure vessel had a fusion-welded titanium frame attached to side panels and fore and aft bulkheads. The side panels and pressure bulkheads were double thickness, thin-sheet titanium (0.25 mm) with the outer sheet beaded for stiffness. A hatch was provided over each astronaut for entering and leaving the spacecraft.
Equipment bays, which contained a variety of electrical and electronic equipment, were located outside the pressure vessel. Unlike the Mercury spacecraft, which had nearly, all systems inside the pressure shell, the Gemini spacecraft had most system components located in unpressurized equipment bays. These components either required no pressurization or were internally pressurized. Since equipment was normally only one layer deep within the compartments, launch crews could remove a hatch, quickly pull out a malfunctioning unit and insert a new one, reinstall the hatch and proceed with the launch.
Experiments for NASA and the US Air Force were installed in bays on the "bottom" of the reentry module, in the adapter section, and in the pressurized crew compartment.
Two hatches, contoured to the shape of the conical cabin, were located in what was the top of the spacecraft during orbital flight. A hatch was located over each astronaut and was manually operated by handles inside and outside the spacecraft. The latching mechanism was mechanical. The hinge was on the outboard side of the door.
Each hatch incorporated an observation window consisting of one outer and two inner panes of glass with an air space between each pane. The outer panes were high-temperature 96% silica glass. The innermost pane was of temper-toughened alumino-silicate glass for structural strength. The surface of each pane, with the exception of the outer one, was coated to reduce reflection and glare and to aid in attenuating ultraviolet radiation.
Skin and beam construction made up the structural design of the personnel access hatches. A silicon rubber seal around each hatch sill and around the two inner panes of glass in the window prevented loss of cabin pressure when the hatches were closed.
A hatch curtain was stowed alongside the hinge of each hatch, which, after a water landing, prevented water from entering the cabin when the hatches were opened. In emergency situations the hatches opened by a 3-sequence operation actuated by a pyrotechnic (explosive) device. When initiated, these actuators unlocked the mechanical latches, opened the hatches, and finally supplied a hot gas that ignited a seat-ejection catapult rocket.
The heat shield was a dish-shaped structure that formed the large end of the reentry module.
The ablative substance of the Gemini heat shield was a paste-like material, which hardened in standard atmosphere after being poured into a honeycomb form.
Starting with a load-carrying Fiberglas sandwich structure consisting of two 5-ply faceplates of resin-impregnated glass cloth separated by a 0.65inch thick Fiberglas honeycomb core, an additional Fiberglas honeycomb was bonded to the convex side of the sandwich and filled with Dow-Corning DC-325 ablative material. The entire shield was encircled with a Fiberite ring. The basic ablative substance of the heat shield was developed by McDonnell and was marketed by Dow-Corning.
The surface of the reentry module was covered with overlapping shingles, which provided aerodynamic and heat protection and held in place shaped pads of flexible insulation. The composition of the beaded (corrugated) Rene 41 shingle (0.41 mm thickness) on the sides of the cabin was 53% nickel, 19% chromium, 11% cobalt, 9.75% molybdenum, 3.15% titanium, 1.6% aluminum, .09% carbon, .005% boron, and less than 2.75% iron. The shingles were identical in composition and manufacturing technique to those used on Mercury. Extra large holes in the shingles at the attachment bolts allowed each to expand during aerodynamic and solar heating. Oversize washers covered these holes to minimize heat and air flow penetration.
The R and R and RCS section surfaces were unbeaded shingles of cross-rolled beryllium. The plate was supplied to McDonnell, by Brush Beryllium Company, in sheets ranging in thickness from 7.6 mm to 14.1 mm and was finished by McDonnell to a thickness of 2.28 mm to 7.11 mm. The shingles were attached to the spacecraft by beryllium retainers fabricated from similar plates.
Beryllium was previously used on Project Mercury using sections fabricated from hot-pressed beryllium blocks rather than cross-rolling procedures. The new manufacturing technique permitted much higher strength and shock resistance. Gemini rendezvous flights required almost twice the strength and impact resistance available with hot-pressed beryllium blocks.
Both Rene 41 and beryllium shingles were coated with ceramic paint on the outer surface to permit high thermal radiation from the spacecraft. The inner surface of the beryllium shingles had a very thin gold coating to attenuate thermal radiation into the spacecraft.
The most obvious structural difference between the Gemini and Mercury spacecraft was the integral adapter module, which was part of the orbital configuration of Gemini.
The adapter module, 2.29 m high and 3.05 m across its largest diameter, provided volume for systems and equipments needed for long-duration orbital flights, as well as the mating structure between the reentry module and the launch vehicle.
The adapter was a ring stiffened skin-stringer structure consisting of circumferential aluminum rings, extruded magnesium alloy stringers, and magnesium skin. The T-shaped stringers had a hollow bulbous portion to provide a path for the flow of liquid coolant, which transferred heat to the adapter skin for subsequent radiation into space.
The outer surface of the adapter module was coated with white ceramic paint and the inner surface was covered with aluminum foil to reduce emissivity. The adapter was joined to the reentry module by three titanium tension straps external to the structure of both the reentry module and the adapter section.
At the small end of the adapter module was the retrograde section containing crossed aluminum "I" beams on which were mounted four retrograde rockets. This section made up the first 100 cm of the adapter module. One retrograde rocket was mounted in each quadrant of the section. In addition to retrograde rockets there were six orbit attitude and maneuvering system (OAMS) thrust chamber assemblies. Four of these assemblies permit orbital translation, up, down, left and right. Non-operating "dummy thrusters" were installed in place of these four assemblies on Spacecraft. No. 3. Two, whose nozzles face toward the reentry module, provided for the rearward or "backing away" maneuver of the spacecraft.
At the large end of the adapter module the equipment section provided volume and attach points for several system modules, including orbit attitude and maneuvering system propellant tanks, the environmental control system primary oxygen supply, batteries, coolant, and electrical and electronic components. A honeycomb blast shield between the two sections protected the equipment section and the dome of the Titan launch vehicle from excessive (explosion-causing) heat should it be necessary to fire the retrorockets in an abort condition.
Ten OAMS thrust chamber assemblies were mounted in the equipment section providing for roll, pitch, yaw, and forward maneuvering of the spacecraft during orbital flight.
A Fiberglas cover over the open end of the adapter protected the equipment inside from solar radiation after separation from the launch vehicle. A forged aluminum alloy ring mated the spacecraft and the Titan II launch vehicle, 3.05 m in diameter with 20 lugs through which bolts were fastened to secure the mating. When the spacecraft was separated from the launch vehicle, a pyrotechnic charge was fired to sever the adapter section approximately 40 cm above the launch vehicle/spacecraft mating point. This charge cut through the metal skin of the adapter section instantly but in the same way a metal shear would cut through sheet metal.
The adapter structure was constructed as a single unit. The two sections were separated by a shaped-charge pyrotechnic device prior to reentry.
Communications And Tracking
The communication and tracking system provided two-way voice communication. ground-to-spacecraft command link. spacecraft-to-ground telemetry transmission, radar tracking signals and recovery aids. Subsystems consisted of telemetry, tracking. voice communications, digital command, antennas. and recovery aids.
The voice communications subsystems included the voice control center, the UHF voice transmitters-receivers and HF voice transmitter-receiver, built by Collins Radio. The voice communication subsystem was operational from prelaunch through postlanding.
The voice subsystem provided communication between the astronauts, between the blockhouse and the spacecraft during launch, between ground stations and the spacecraft from launch through reentry, and between astronauts and frogmen during the water recovery. The voice subsystem also provided communication between the spacecraft and recovery forces during landing and postlanding.
During the reentry, voice communication would be lost on two occasions. Under worst conditions. the first extended from approximately 1310 seconds after retrofire to 1775 seconds after retrofire, nearly 8 minutes (under nominal conditions, this time was approximately 6 minutes). This loss of communication was caused by ion sheath formation around the spacecraft. The second period lasted about 30 seconds, starting at main parachute deployment, and was due to a delay between loss of the nose stub transmitting antenna and the erection of the descent antenna.
The voice control center provided for intercommunication between the astronauts, for control and distribution of audio to and from the transceivers, and supplied a tone for direction finding (HF-DF). The voice control center provided for individual selection and control of various functions in UHF, HF and intercommunication circuits.
Dual controls permitted mode switching and volume controls for HF, intercommunications, and UHF. A common section in the panel provided squelch control of UHF and HF, receiver selection of HF and UHF, and keying.
The HF voice transceiver provided for over-the-horizon spacecraft-to-ground communication and a direction finding signal when in the HF-DF mode.
The voice control center was located in the pressurized cabin of the Gemini spacecraft. Two UHF transceivers and one HF transceiver were located in the reentry module of the spacecraft outside the pressurized cabin.
The Motorola built digital command subsystem consisted of a receiver-decoder and associated relay units, which permitted spacecraft utilization of ground commands. Located in the equipment section, this system was operational from pre-launch until jettison of the equipment section. The digital command system received and decoded command transmissions from the global network of ground stations and transformed them into a digital format.
Digital commands were categorized as either "real-time commands" or "stored program commands." Real-time commands operated DCS relays that controlled equipment input power or energized relays in the spacecraft electrical system to control equipment usage. The stored program commands were used to provide such units as the time reference system and the computer with updated data.
The antenna subsystem consisted of antennas, coaxial switches, a diplexer and a quadriplexer. The antenna subsystem was operational from prelaunch through postlanding and provided radiation coverage for all communication and beacon tracking signals between ground stations and the spacecraft. The system included C-band tracking helical, C-band slot, S-band, HF whip, UHF whip, descent, recovery, and UHF nose stub antenna.
The antenna system provided radiation coverage for the communication system during all mission phases. Coverage varied with the mission phase depending upon spacecraft stabilization mode and ground coverage requirements. During the launch phase, when continuous C-band and UHF coverage were required for flight safety reasons, the antenna system provided roll symmetrical antenna patterns to optimize the ground coverage.
During the orbital phase, S-band and HF were added. The orbital antenna system provided yaw symmetrical, horizon- oriented, hemispherical patterns for optimum coverage in stabilized orbit attitude. For drifting flight with uncontrolled spacecraft attitude, the antenna/communication system provided complementary coverage.
Complementary coverage was obtained by use of yaw symmetrical and roll symmetrical antenna patterns.
The communications systems used these patterns simultaneously.
The astronauts select the antenna system to obtain the optimum pattern for voice and telemetry. During the reentry phase, UHF and C-band coverage was identical to launch phase coverage. During the recovery phase, antenna capability was provided for HF and UHF.
The telemetry subsystem provided real-time, delayed-time and standby telemetry transmission.
The frequency modulated telemetry transmitters were employed during all phases of the Gemini mission when the spacecraft was in contact with the ground stations. These transmitters were energized either by the astronauts or automatically by the ground digital command system. The standby transmitter was used as a replacement for either the real-time or delayed-time transmitter in event of a failure. A delayed-time transmitter sent pulse code modulated information, which was stored in the tape recorder of the data transmission system. The real-time transmitter sends current information from the data transmission system programmer.
Recovery Aid Subsystem
The recovery aid subsystem included a UHF recovery beacon, a UHF rescue beacon transceiver and a flashing light. The HF voice transmitter in the reentry module was tone modulated in the HF-DF mode as part of recovery aid equipment.
A pulsed UHF output signal (energized upon landing impact) supplied continuous direction-finding information for recovery forces on the international distress frequency, and the flashing light provided the visible indication of the spacecraft location should recovery operations be conducted at night. The light was designed to be visible from an altitude of 12,000 feet at a distance of 50 nautical miles on a starlit, moonless night. The recovery aid subsystem was operational only during landing and postlanding phases of the Gemini mission.
The tracking subsystem included a C-band radar beacon, an S-band radar beacon and an acquisition aid beacon. The C- and S-band beacons provided tracking responses to interrogation signals from ground stations. Either or both of these beacons could be energized by ground command via the DCS. The astronauts could also energize either beacon, and select the antenna system for the C-band beacon to achieve a roll symmetrical or yaw symmetrical pattern. The acquisition aid beacon provided a radio frequency signal from the spacecraft to ground communication facilities for spacecraft acquisition.
The C-band radar beacon was operational from prelaunch through the landing phase, the S-band radar beacon was operational from prelaunch to immediately prior to retrograde and the acquisition aid beacon was operational from prelaunch to immediately prior to retrograde. S- and C-band beacons were built by ACF Electronic Division. The acquisition aid beacon was built by Vector Manufacturing Company.
The Gemini GT-3 spacecraft electrical system was a two-wire, grounded system using silver-zinc Eagle Picher batteries as sources of 25 VDC electrical power. There was no primary AC electrical power system on the spacecraft. Devices utilizing AC power obtain that power from self-contained inverters within the individual systems.
Prior to launch, external electrical power was provided to the spacecraft through the umbilical to prevent undue depletion of the spacecraft power supply. The battery subsystem was capable of supplying sufficient power to the electrically operated equipment for all phases of the planned GT-3 mission. In addition, sufficient power was available for a prelaunch period of two hours and a postlanding period of 36 hours for operation of necessary recovery equipment and for 12 hours of post landing suit compressor operation.
Ten silver-zinc batteries were provided, each activated and sealed at sea level pressure. The battery cases were vented to permit the escape of gases. Battery temperatures were controlled by mounting the battery cases in direct contact with spacecraft coldplates.
The reentry module main batteries supply a portion of the main bus electrical power during launch and all of the main bus electrical power during reentry, landing and postlanding. These four batteries were located in the right-hand equipment bay outside the pressurized area of the reentry module.
The squib batteries supply electrical power for squib-activated pyrotechnic devices throughout the entire mission. Squib batteries were isolated both electrically and mechanically from all other batteries. The three squib batteries were located ir the right-hand equipment bay outside the pressurized area of the reentry module.
It was possible to connect squid circuitry to the reentry main batteries in case of squib battery malfunction.
The three adapter batteries supply power to the main bus and were capable of supplying all of the electrical power necessary for spacecraft operation until separation of the adapter module.
To insure smooth system operation, complete electrical system management by the crew was provided. Extensive circuit protection was incorporated throughout the system and indicators were mounted on the instrument panel for use by the astronauts in systems monitoring.
Environmental Control System
The Gemini environmental control system was capable of providing life support for the biological systems of two astronauts. It provided for ingestion of appropriate gases and fluids, and the dispersal of by-products which were created, as well as for cooling of spacecraft equipment, and the cabin interior.
The system consisted of a water management subsystem, an oxygen supply subsystem, and a cooling subsystem. It provided gaseous oxygen for breathing, for suits and cabin pressurization, and for suit and cabin ventilation. It provided for the removal of small solids, carbon dioxide, odors, and moisture from the suit and cabin atmosphere. For the flight of GT-3, it provided a drinking water supply-. It provided for storage and disposal of water accumulated as a condensate, and for disposal of urine. It included a dual, recirculating coolant system for regulating the temperature of the suits, the cabin and items of electrical equipment.
Primary Oxygen Subsystem
The primary oxygen subsystem stored and dispensed oxygen for breathing and for suit and cabin pressurization. It supplied oxygen during the entire flight, commencing two hours prior to launch and terminating with the equipment section jettison at retrograde.
Oxygen pressure in the crew compartment was limited to 5 5 to 6.0 pounds per square inch above ambient by the cabin pressure relief valve. Primary oxygen supply capacity for a two-day mission was 15.3 pounds, located in a single spherical container in the equipment section, and was the primary source of oxygen during prelaunch, ascent and orbit. Oxygen was stored in this container in cryogenic form. It was heated to a gaseous state by a heat exchanger, passed to a pressure reducing regulator, then on to the cabin pressure regulator which automatically maintained cabin pressure as desired for the mission. Oxygen remaining in the primary supply was jettisoned with the tank when the equipment section was jettisoned prior to retrograde. The two-day absorber cartridge removed both odors and up to eleven pounds of carbon dioxide. The suit heat exchanger transfers heat from suit circuit oxygen to coolant flow. The heat transfer capacity was 1500 BTU per hour. The exchanger also removed moisture from the suit circuit oxygen and transfers it to the water management subsystem.
Pressure was maintained automatically- in the cabin through all phases of the Gemini mission. During the launch phase, cabin pressure relief valve permitted outflow of any overpressures which might exist. This valve seals the cabin when the ambient pressure falls to 5.5 pounds per square inch below cabin pressure. This would occur between 20 and 40 seconds after launch.
If the cabin should decompress for any reason, the supply of oxygen provided through the dual cabin pressure regulator automatically turns off when the pressure reached 4.0 pounds per square inch. When decompression occurs, either as programmed or as a result of a malfunction, the astronaut's pressure suit automatically took over the pressurization responsibilities previously provided as a cabin environment.
If one or both astronauts chose to work with the face-plate of their suits open, a manually operated valve permitted circulating cabin air through the suit circuit for carbon dioxide and water vapor removal. The same valve provided for relief of the vacuum created at the compressor inlet if the snorkel valve was momentarily closed by water after a water landing.
The crew compartment was maintained at a nominal 65 degrees F during orbital flight. It was expected to rise to a high of approximately 120 degrees F during reentry. The maximum acceptable temperature in the crew compartment during launch and reentry of the Gemini spacecraft was considered to be 200 degrees F.
Secondary Oxygen Subsystem
Secondary oxygen was contained in two tanks in the pressurized compartment of the reentry module. There was sufficient quantity in each of the secondary oxygen supply tanks to provide oxygen adequate for one orbit at a normal flow rate and reentry at a nominal oxygen high rate of 0.08 pounds per minute to each astronaut.
The secondary oxygen subsystem operated when the pressure in the primary oxygen line falls below an allowable 75 pounds per square inch. When the primary oxygen container was jettisoned. the secondary oxygen subsystem assumed the primary role.
The oxygen line from the suit-demand and cabin pressure regulators was common to both the primary and secondary supplies so that flow was continuous to the suit circuit if a malfunction occurs.
Egress Oxygen Subsystem
Oxygen for breathing and for suit pressurization in the event the astronaut's initiate ejection at 45,000 feet or below during launch or reentry was provided by the egress oxygen subsystem.
A tank containing approximately 1/3rd a pound of usable oxygen was located in the seat-mounted egress kit for each astronaut. The egress oxygen circuit was lanyard-opened with seat ejection.
Water Management Subsystem
The water management subsystem collected and stored water for drinking, dumps waste water overboard, and managed the water used for cooling. Components of the subsystem include a water tank, a urine receptacle, a drinking nozzle, controls and valves, an evaporator, a reservoir, and a water pressure regulator.
The water was stored in a tank located in the equipment section. For GT-3, it contained approximately 16 pounds of water stored at 7.5 psig. A second water tank in the reentry module also contains approximately 16 pounds of water. Another seven pounds of liquid was contained in the launch cooling heat exchanger reservoir. A tank in the equipment section was charged with oxygen at 1,000 psig and this pressure was available upon demand to pressurize the water storage tanks. Oxygen-pressurized diaphragms forced the water in the line to the cabin water tank and to the water dispenser.
Urine Disposal equipment was Government furnished, installed by McDonnell. It consisted of a urine line, bellows assembly, quick disconnect coupling, and a uriceptacle. On GT-3, urine was routed to the water evaporator for disposition upon actuation of the cabin water dump valve.
Temperature Control Subsystem
A distinctively new feature of the Gemini spacecraft was the fluid coolant system, which maintained cabin temperature, astronauts' suit temperature, and equipment temperature within acceptable limits.
Since the Gemini equipment and crew would generate heat at approximately three times the rate of the Mercury spacecraft, and would do so for almost ten times as long, it was necessary to provide a new method of heat rejection, namely, a space radiator. The entire outer skin of the adapter module served as the radiating surface, and the hollow stringers, through which the coolant passes, transferred the heat to the skin.
At the heart of this system was the space radiator in the adapter module, a coolant fluid reservoir, a low-level coolant-sensing device, two identical positive displacement pumps for each of two redundant loops of coolant lines, and two regenerative heat exchangers. The redundant systems provided protection against loss of a loop due to failure such as meteorite penetration.
A silicon ester coolant fluid (Monsanto's MCS 198) was pumped through each, then through heat exchangers, which heated the primary cryogenic oxygen supply, cooled the cabin and suits, cooled such equipment as electrical power supplies, and various electronic equipments. The cabin and suit heat exchangers were conventional heat exchangers, and the electronic equipment heat exchangers were coldplates on which the equipment was mounted.
In the cabin each coolant circuit was divided into two parallel paths, one for the cabin and one for the suit. Ahead of each heat exchanger was a manual valve, which permitted crew adjustment of the flow through the heat exchanger.
In the section involving coldplates (there were 24 of them in each spacecraft), as many of them as possible were arranged in parallel to minimize pressure loss. However, all of the flow was required through some high power density units, and these coldplates were arranged in series.
Two regenerative heat exchangers in the equipment section were provided for each of the loops. A temperature-sensitive valve modulated to maintain an outlet coolant temperature at between 36 degrees F and 42 degrees F. When the coolant temperature was 36 degrees F or lower, the valve caused all coolant to flow to the regenerative heat exchanger. When the coolant temperature was 42 degrees or above, the full coolant flow was directed to the space radiator. The adapter structural shell provided an area of 165 square feet to radiate the excess heat into space.
Each of the redundant coolant loops followed alternate stringers in the adapter, so that when operating on one loop, the tubes of the other loop were prevented from freezing by heat from the adjacent warm tubes. The coolant fluid, however, had a freezing point well below -100' F. and had a characteristic of low viscosity at extremely low temperatures.
During preflight operation a ground heat exchanger removed heat from the spacecraft by circulating a coolant fluid from a servicing cart through a secondary loop in the heat exchanger. During launch, and for a short period thereafter, the space radiator was too warm for effective cooling. Then, a water boiler cooled the coolant fluid by the simple process of absorbing the heat through vaporization.
In thermal balance tests run in the 30-foot space chamber at McDonnell, it was determined that the emissivity for the GT-3 spacecraft, which did not had fuel cells, was actually higher than desired. The GT-3 adapter was striped with .66 inch aluminum tape with black coating for deliberate and controlled reduction of the total heat emission of the radiating surface. There were 72 strips on the retrograde section, and 88 on the equipment section. A tape strip was aligned vertically between each stringer.
The temperature control subsystem circuits, which connected with the reentry module, were lost when the adapter section was jettisoned. Temperature control subsequent to initiation of retrograde was provided by cabin wall insulation and the heat shield.
Supercritical Storage of Cryogenic Fluids
Primary oxygen was stored in the equipment section of the Gemini spacecraft in a cryogenic state in especially designed cryogenic containers. Because cryogenic storage was at a lower pressure than tanks of equal size holding the same quantity of oxygen, danger of structural failure was reduced; associated components were lighter; and the heat absorption capacity of the fluid was available to help dispose of spacecraft waste heat.
The difference in weight between a vessel for storage of oxygen as a gas and in its cryogenic state was significant. Gaseous oxygen required a heavily constructed tank to withstand a pressure of 7,500 pounds per square inch. In the supercritical cryogenic system used in Gemini, oxygen was first stored as a liquid, then in a confined tank volume, it was heated to a point where it became a homogeneous compressed fluid. This process took place in the cryogenic storage container at a pressure of only 750 pounds per square inch. A cryogenic storage subsystem weighed less than four-tenths of a pound for each pound of oxygen stored. A high-pressure vessel of comparable capacity would weigh five times as much.
Gemini's supercritical storage of fluids insured a steady delivery of oxygen under all gravity conditions just as high-pressured gas does.
Guidance and Control
The guidance system included the inertial guidance system (IGS), information interfaces with the digital command system (DCS) and the data acquisition system (DAS), with the time reference system (TRS) functioning as an integral part of the guidance system. This system performed navigation and steering logic computation for trajectory management. The guidance system output provided guidance information directly to control systems and/or to instrument displays which the crew interpreted for manual control inputs. The Gemini guidance computer, guidance analysis, and guidance system integration subcontractor was IBM Federal Systems Division's Space Guidance Center.
The control system provided commands to the thrusters, acting in response to outputs from the horizon sensors, the computer, attitude control and maneuver handles, and the attitude control and maneuver electronics (ACME). The propulsion system was an integral part of the control system. Both automatic and manual modes of operation provided angular control of the spacecraft about its three major axes and for maneuvering. spacecraft translational maneuvers were manually controlled.
Inertial Guidance System
The inertial guidance system included an inertial measuring unit (IMU), digital computer, manual data insertion unit (MDIU), incremental velocity indicator (IVI), and an auxiliary computer power unit (ACPU).
The inertial measuring unit, built by Honeywell, was a stabilized inertial platform including an electronic unit and a power supply. The inertial measurement unit provided a stable attitude reference and incremental velocity data.
The inertial platform was a four-gimbal structure with an all-attitude capability. The gimbals, in sequence, starting with the innermost gimbal, were pitch, inner-roll, yaw, and outer-roll. The stable element contained three single-degree-of- freedom rate-integrating gyros and three accelerometers of the force rebalance type, mounted orthogonally (mutually perpendicular). Each gimbal contained two analog pickoffs for angle measurement between body and platform axes. One set of pickoffs operated in conjunction with the computer to generate digital representation of the angles. The second set operated with the attitude display group to provide attitude information to the flight crew.
The IMU subsystem electronics provided circuitry and equipment for alignment, stabilization, and gimbal torquing, and accelerometer rebalancing. IMU malfunction detection circuitry was also included.
The IGS power supply operating from 28-volt DC spacecraft power provided the alternating current and direct current power necessary for operations of the IMU, the computer, the MDIU, the IVI, the ACME, and the horizon sensors. The attitude control and maneuver electronics had a separate power supply, which could be operated when the inertial guidance system was not operating. The IGS power supply provided alternating current for components normally supplied by the attitude control and maneuver electronics unit, in the event of a ACME power supply failure.
The onboard digital computer was a binary, fixed point, stored program, general purpose, digital computer that provided executive functions, computations, timing and signal processing for spacecraft guidance and control. The Gemini computer had a memory of 4,096 words (39 bits/ word). It was random access with a non-destructive readout. Division of the memory into three syllables provided flexibility in instruction and data storage location assignment. It could add, subtract, and conduct a transfer operation in 140 microseconds. Multiplication (full precision) was accomplished in 420 microseconds. Division (full precision) was achieved in 840 microseconds. Multiplication and division could be programmed concurrently with addition and subtraction or transfer operations. The clock rate of the Gemini digital computer was 500 KC arithmetic bit rate; 250 KC memory cycle rate.
Computer programs included an executor program, operational programs and standard computational sub-routines. The executor program performed diagnostic checks, determined elapsed time, selected the desired operational program, and performed all data input/output sub-routines.
The basic computer program was inserted at McDonnell prior to computer installation in the spacecraft. After installation, minor updating of constant words and time variable words was accomplished by the digital command system or the manual data insertion unit and was monitored by the data acquisition system. Operational program requirements were functions of the computer mode. These modes were varied in accordance with the missions to be accomplished by the individual spacecraft. For Spacecraft 3, four modes were available: Prelaunch, Ascent, Catch-Up, Reentry.
In the prelaunch mode, the computer was programmed to perform diagnostic checkout routines and sub-routines. In this mode, sum checks were performed in the section of the computer memory that contained constants and fixed programs that could not be varied during the mission.
Guidance in the ascent mode served a backup function in the event of malfunction of the radio guidance system (RGS).
In the ascent mode, information was displayed for flight crew evaluation of mission status. The computer also provided navigational data for reentry guidance in the event an abort should become necessary during the launch phase.
While the RGS had primary responsibility for launch vehicle guidance from liftoff through launch vehicle/spacecraft separation, the inertial guidance system provided backup guidance by effectively duplicating RGS functioned in an inertial frame, defined by the launch pad vertical and the desired azimuth at insertion. Stage 1 of standby guidance consisted of a time-sequenced roll and pitch program based on the design constraints of the RGS. Stage 2 guidance also satisfied the guidance constraints defined for the RGS. The ascent program could be updated by the digital command system with ground-determined velocity data to reduce the effects of IGS uncertainties. In the event of a guidance switchover, the IGS provided faded (smoothed) attitude error signals to the launch vehicle secondary autopilot. Switchover could occur at anytime. Should switchover occur before liftoff, the switchover signal simultaneously directed engine shut-down, preventing liftoff on backup guidance.
The IGS provided orbit insertion guidance so that the effect of insertion errors on apogee and perigee altitude was nullified. The velocity increment to be added for proper insertion was computed from IGS navigational data and desired insertion condition data stored in the onboard computer. The velocity increment to be added was displayed on the incremental velocity indicator. The attitude error command necessary for proper vehicle orientation was displayed on the flight director indicator. The inertial guidance system then performed navigation computations as the insertion sequence proceeds, driving the incremental velocity indicator display in accordance with the velocity increment added. Orbit insertion guidance began immediately after a normal second-stage engine cutoff (SSECO), or in the event of a premature engine shut-down, at least 300 seconds from liftoff.
After the incremental velocity was applied at perigee, the flight crew could read out of the computer, via the MDIU, an incremental velocity to be added at apogee to correct the perigee altitude. The time to apply this velocity could also be read by the MDIU.
In the event of a launch abort after a velocity greater than 21,000 feet per second was achieved, reentry navigation equations were initiated from the position and velocity information generated by the inertial guidance system. Thirty seconds after equipment section separation, an event initiated by the flight crew, the computer would generate transformation coefficients to relate the platform coordinate system to that required for reentry guidance equations. Following initiation of reentry navigation, the data flow was identical to that which occurred in the reentry mode.
Between orbit insertion and reentry, the inertial guidance system was operated in the catch-up mode. Various maneuvers were performed in this mode on the GT-3 flight to study the spacecraft's capabilities relative to future missions. The catchup mode provided a means of displaying and applying a ground-command velocity change. The time-to-go to apply thrust and the velocity change to be applied were computed at the integrated mission control center on the ground, based upon knowledge of the time of liftoff or upon ground tracking information or both. Time and velocity change components were transmitted to remote tracking sites for transmission to the spacecraft by voice and digital communication. Velocity change was normally transmitted by the digital communication system; however, it could be transmitted by voice and inserted into the onboard computer by the astronauts via the manual data insertion unit. In addition to the MDIU and DCS capability, the velocity change to be applied could also be displayed by the incremental velocity indicator. Spacecraft maneuver commands were manually applied using orbit attitude and maneuvering system thrusting.
Reentry Guidance Application
The Gemini spacecraft was designed for guided reentries from orbit to permit spacecraft landing at pre-selected, prepared sites. This capability was provided through the use of the inertial guidance system to develop steering commands for proper orientation of the aerodynamic lift vector arising from an offset center of gravity position.
Preparation for reentry began significantly ahead of retrograde to precisely establish the inertial platform reference and the initial conditions for navigation computation in the reentry mode. Following retrograde, the retrograde section was manually jettisoned and the spacecraft manually controlled to a flight crew "heads down" full-lift attitude. Reentry steering was initiated when the spacecraft entered the atmosphere and total acceleration achieved a prescribed predetermined value. Roll commands controlled horizontal and vertical components of lift as a function of downrange and cross-range errors. Roll errors were supplied to the attitude control and maneuver electronics for automatic control and also to the flight director indicator for manual control capability. When the density/altitude parameter reached a pre-selected point within the atmosphere, a maximum lift attitude was commanded and reentry guidance was terminated.
The reentry configuration for Gemini had a center of gravity location resulting in angle-of-attack conditions producing an appreciable lift vector during the atmospheric regions of the reentry. The roll attitude of the spacecraft established the direction of lift application; zero bank angle produced range extension, banking to the right a left turn, banking to the left the opposite.
For standard reentry, ground computers determined the future Earth tracks of the spacecraft, determined approximately when the desired landing site would become available, and then performed iterative trajectory solutions to establish a retrograde time that would require about one-half the downrange extension provided by the reentry lift. This time, along with the position and velocity associated with the retrograde time and the target coordinates, were transmitted to the spacecraft and inserted into the IGS.
The spacecraft computer, using the initial conditions transmitted from the ground, maintained a knowledge of the trajectory through conventional navigation equations. It then predicted a touchdown on the basis of "zero lift" and compared the predicted touchdown point with the desired target site. Downrange and crossrange errors were used to develop bank angle commands.
The system utilized the ground complex to develop a very accurate orbit position for use in subsequent reentry guidance. Ground stations locations were planned to maintain excellent coverage of the spacecraft during the majority of its flight.
The control system included the attitude control and maneuver electronics and the horizon sensors. It was used in conjunction with the propulsion system and associated guidance systems to provide spacecraft orientation about its three major axes and for translational maneuvering. The orbit attitude and maneuver thrusters were employed to assist in spacecraft/launch vehicle separation and for attitude control and maneuvering prior to adapter section jettison. During retrograde and reentry, control thrusting was provided by the reentry control system thrusters.
Attitude control and maneuver electronics (ACME) included the attitude control electronics (ACE), orbit attitude and maneuver electronics (OAME), power inverter and two rate gyro packages. Input signals to the attitude control and maneuver electronics were supplied by the IGS, the horizon sensors, or by crew manipulation of the attitude control handle or the maneuver controller, depending on the operational mode the astronauts had selected.
The attitude control electronics accepted input signals from the inertial guidance system, from the attitude control handle, from the rate gyros, and from horizon sensors. Input signals were converted into drive commands for the reentry control systems solenoids and to logic commands for the orbit attitude and maneuvering electronics subsystem. Honeywell, Inc. was subcontractor for ACME, ACE, and OAME.
The orbit attitude and maneuver electronics accepted input signals from the attitude control electronics and the maneuver controller for conversion to drive commands for the OAME solenoids.
Rate gyros sensed angular rates about the pitch, roll and yaw axes of the spacecraft. Rate signals were supplied to the ACE in the rate command mode plus all other modes where rate damping was employed. The rate gyro packages also provided inputs to attitude displays and to the telemetry system.
Attitude control and maneuver electronics modes provided for maneuver control to effect spacecraft/launch vehicle separation, for translational maneuvering during the catch-up mode operation, and for maneuvering in all planes by attitude maneuvers and forward/aft thruster operations. Forward or aft displacement of the maneuver controller from the neutral position produced a direct command to the respective solenoid valve driver. The astronauts could select any one of six control modes for spacecraft orientation during orbit and reentry. These modes were rate command, direct, pulse, reentry rate command, horizon scan and reentry.
Rate command mode: Spacecraft angular rates were proportional to rate command signals initiated by flight crew displacement of the attitude control handle. The rate command signals were compared in the attitude control electronics with rate gyro outputs and when the difference between the two signals exceeds the damping dead zone in the system, proper reaction control jets fire for attitude control. This mode was utilized during manually initiated attitude control operation and during retrofire and other velocity change maneuvers. In this mode it was possible to maintain attitude within +/- 1 degree using the flight director indicator for reference.
In the direct mode, switched on the attitude hand controller directly control the ACME solenoid drivers. Control switches were actuated when controller displacement was greater than approximately 1/4 of its full travel.
The pulse mode permitted the astronauts to make fine attitude corrections of the spacecraft about its three major axes. In this mode, angular rates were incrementally changed by single thrust commands of fixed duration. Each displacement of the hand controller by the astronaut triggers a single pulse generator in the attitude control electronics and sends a single pulse command to the proper reentry control system or orbit attitude and maneuvering electronics solenoid valve drivers. The astronaut must return the hand controller to the neutral position before he could initiate another pulse to increase his angular rate or effect a braking action in the other direction.
Reentry rate command mode was utilized for manual attitude control during reentry. This mode provided similar operating characteristics to that of the rate command mode, except that damping dead bands were wider and roll rate crossfeed was included in the yaw damping loop, providing for conservation of fuel because the fine control provided by rate command was not necessary for manual performance of this phase of the mission.
The horizon scan mode provided for automatic control of the spacecraft about the pitch and roll axes during the orbit phase of the mission. ACME received attitude information from the horizon sensor and generated an output to the proper thrusters to maintain the attitude within the damping dead band. When in this mode, the ACME supplied a nose-down pitch bias which enabled the flight crew to view the horizon out of the window.
In the reentry mode spacecraft pitch and yaw angular rates were automatically- maintained within a damping dead band by the ACME. A roll attitude was determined by inputs from the computer to the ACME. The computer either provided a bank angle command or a fixed roll rate depending on the relationship between the predicted and the desired touchdown points. ACME would not accept rate commands from the attitude control handle when in the reentry mode.
Two horizon sensors, one primary and one secondary or standby unit, provided reference signals for alignment of the inertial platform and error signals to the ACME for controlling the spacecraft attitude about its pitch and roll axes. Horizon sensors operated by tracking the Earth's infrared horizon.
Two attitude displays, each incorporating a three-axis attitude reference ball with 360 degrees of rotation about each axis, were provided on the right and left instrument panels. These displays, built by the Lear-Siegler Corporation, were slaved to the positions of the normal inertial platform gimbals and provided a continuous all-attitude reference of roll, pitch, and yaw.
Integral rate and command flight director needles displayed control movements required to position the spacecraft in a commanded attitude or rate. When the commanded attitude or rate was achieved, the needles were centered.
An attitude hand controller was mounted on the console between the astronauts and provided a means of manually controlling the spacecraft attitude and rate in three axes. The controller could be operated by either astronaut, while either was in the restrained position, through simple wrist articulation and palm pivot motion. The controller was spring loaded to provide an increasing resistance as the handle was moved away from neutral. The total travel of the hand controller was +/-10 degrees from neutral in all three axes. Displacement or rotation of the controller caused the spacecraft to turn in the direction of displacement or rotation.
A maneuver hand controller initiated translation of the spacecraft. The controller contained centering springs and six switches, one for initiation of spacecraft displacements in each of six directions along with three major axes. Movement of the handle in any of these six directions initiated corresponding spacecraft translation. The handle could be removed and stored when not in use, providing clearance in the event of seat ejection.
Guidance System Functions
|Mission Phase||Applicable Guidance Mode||Function|
|Launch||Ascent|| Provided backup guidance in event of a radio guidance system (RGS) malfunction. |
Provided attitude error data for astronaut evaluation of mission status.
Provided navigation initial conditions for reentry guidance in event of an abort during boost.
|Post-SSECO||Ascent||Provided capability to correct apogee and perigee altitude deviations resulting from insertion velocity, flight path angle, and altitude errors.|
|Orbit||Catch-up||Provided the capability for displaying and applying ground determined velocity changes.|
|Orbit||Prelaunch||Provided a diagnostic sum check of computer syllable two instructions. |
Provided standby guidance mode wherein digital acquisition was provided and MDIU/DCS locations were telemetered and provided different telemetry quantities.
|Retro||Reentry|| Provided a real-time navigation during retrofire sequence. |
Display of retrograde incremental velocity to astronauts.
|Reentry||Reentry||Performed real-time navigation. |
Performed continuous prediction of the spacecraft zero lift impact point.
Provided control logic to deliver steering commands to control crossrange and downrange travel.
Provided an automatically controlled reentry in conjunction with reentry attitude control mode.
Attitude Control System Characteristics
|Mission Phase||Applicable Attitude Control Mode||Function|
|Launch||Direct||Rate information to flight crew.|
|Post-SSECO||Direct||ON-OFF commands to attitude thrusters.|
|Rate Command||Vehicle angular rates were proportional to attitude control handle displacements.|
|Orbit||Horizon Scan||Automatic pitch and roll attitude control to a horizon sensor reference.|
|Pulse mode capabilities retained for manual over-ride about all axes.|
|Five-degree small-end down pitch bias included to maintain astronaut view of Earth's horizon.|
|Pulse||Angular rate could be changed in incremental steps by commanding thrust for a calibrated period of time.|
|Effective about all three axes and could be used for fine attitude control.|
|Rate Command||Same as described for Post-SSECO.|
|Direct||Same as described for Post SSECO.|
|Retro||Rate Command||Same as described for Post SSECO.|
|Direct||Same as described for Post-SSECO.|
|Reentry||Automatic rate damping about pitch and yaw axes.|
|Roll attitude and rate controlled by computer and ACME.|
|Roll rate cross-feed was included in yaw damping loop.|
|Reentry||Reentry-Rate Command||Astronaut control of roll orientation.|
|Rate damping provided about all axes.|
|Roll rate cross-feed was included in yaw damping loop.|
|Direct||Astronaut control of roll orientation.|
|Manual rate damping required in all axes.|
ACME Inputs and Outputs
Time Reference System
The time reference system consisted of an electronic timer. a time correlation buffer, an event timer, and a clock.
An interface existed between the time reference system and the digital command system, the digital computer, and the data transmission system.
The electronic timer recorded elapsed time in 1/8th-second increments from liftoff through impact; it counted time-to-retrograde from liftoff to zero in 1/8th-second increments; and it counts time to equipment reset on command in 1/8th second increments. The electronic timer exchanged signals with the digital command system, the digital computer, and the data transmission system.
The electronic timer had a crystal controlled time reference accurate to 35 parts in 1 million for 24-hour period. Stability over a 3-hour period was 10 parts in 1 million at 25 degrees C 10 degrees C. The timer was mounted behind the center instrument panel.
Updated or revised time-to-go was forwarded to he electronic timer by the digital command system.
To prevent inadvertent or premature countdown to retrofire, the electronic timer was provided with a lockout set at 512 seconds. It would not accept any time to go quantity of less than 512 seconds.
The time correlation buffer accepted elapsed time and clock information from the time reference system pulse code modulation input. The time correlation buffer provided outputs to the voice tape recorder and the two biomedical tape recorders. Information to the recorders from the time correlation buffer was updated every 2.4 seconds. It provided serial data and clock data outputs to a buffer register every 2.4 seconds.
The event timer provided a cockpit decimal time displayed in minutes and seconds to a maximum 99 minutes and 59 seconds. This display permitted countup and countdown timing by the astronauts. The display could be manually positioned or it could be started by face-mounted switching or independent electrical remote signal. The unit operated completely independent of the electronic timer.
The spacecraft clock displayed Greenwich Mean Time (GMT) in hours and minutes. Launches were conducted at Cape Kennedy on GMT. The clock included an additional minute hand and a second hand, which could be stopped and reset to zero mechanically at any time. A calendar day display was also provided.
Instrumentation and Recording Systems
Gemini instrumentation and recording systems monitored specific spacecraft systems, conditions, and events, and sensed, conditioned, encoded, reproduced, and transmitted data to ground stations.
The instrumentation and recording system performed the following:
The instrumentation system had both an operational and a diagnostic function. The system provided real-time telemetry transmission of data generated in the spacecraft and required at ground stations for monitoring the progress of the spacecraft mission, for assessing the spacecraft status, and for making decisions concerning flight safety. In addition, Gemini instrumentation included the necessary hardware for generating signals, not available through individual spacecraft systems, for operation of the spacecraft cabin displays. In its diagnostic function, the spacecraft instrumentation system provided a means for documenting significant events and data, throughout the entire mission, by three methods: Real-time transmission, delayed transmission, and onboard recording. Electro-Mechanical Research supplied the data acquisition system for Project Gemini.
Data Transmlssion System Programmer
The data transmission system programmer provided for data multiplexing, analog-to-digital data conversion, and digital-data multiplexing.
The programmer included a high-level multiplexer, a low-level multiplexer and a delayed transmission recorder/reproducer.
The high-level multiplexer functioned as a high-level analog commutator and on-off digital data multiplexer. It provided for the sampling of 32 high-level data channels, 24 bi-level signals and 16 inverted bi-level pulse signals.
The low-level multiplexer functioned as a differential-input analog input commutator and provided a sequential sampling of 32 low-level signals. The delayed transmission recorder/reproducer recorded data during the time the spacecraft was in orbit and out of range of worldwide tracking stations. When the spacecraft was within range of a tracking station, the recorder/reproducer was triggered by either a ground signal or by the astronauts, reversed its tape direction, and plays back the recorded data at an accelerated rate.
Photographic recording was accomplished by two cameras, one a 70-millimeter camera for still photography and the second a 16-millimeter movie camera. These cameras were hand held and operated by the astronauts and were stored in the cabin when not in use.
Voice recording was accomplished on a tape recorder installed in the center pedestal. Recording was limited to astronaut conversations and would occur at any time that the mode selection switched on the voice control center was placed in a record position.
Biomedical recorders were installed in the spacecraft and accept and recorded data from the biomedical instrumentation centers attached to the astronauts.
Secondary propulsion systems (as contrasted to the primary propulsion systems of the launch vehicle) were installed in the reentry module and the adapter module to provide capability for separation from the launch vehicle either under normal conditions or in emergency conditions; for translational maneuvering in six basic directions: up, down, left, right, forward and rearward and for attitude control about the pitch, roll and yaw axes. One system, the retrograde rocket system, provided the necessary velocity decrement to initiate reentry.
Propulsive thrust to perform these functions was generated by three individual systems: the orbit attitude and maneuver system (OAMS), the reentry control systems (RCS), and the retrograde rocket system (RRS). The orbit attitude and maneuver system was installed in the equipment and retrograde sections of the adapter module. The reentry control systems were located in the reentry module. The retrograde rocket system was clustered in the center of the retrograde section, just aft of the reentry module heat shield.
Orbit Attitude And Maneuver System (OAMS)
The OAMS was a liquid bipropellant rocket engine propulsion system constructed on a modular basis consisting of thrust chamber assemblies; pressure storage tanks; pressure regulators; propellant tanks; propellant shut-off valves; propellant and pressurant lines; propellant line cutter-sealer assemblies; and five component packages which provided means of ground testing, pressurant and propellant tank filling, burst diaphragms, relief valves and instrumentation. The components were divided according to function into a thrust chamber assembly group, an oxidizer/fuel group and a pressurizing group.
The OAMS thrust chamber assembly group consisted of 16 engines, each mounted in a fixed position and operated at a fixed thrust level. (The amount of force desired from each engine for maneuvering was obtained by varying the time it was operated at the fixed thrust level.) Eight of the engines develop about 25 pounds; two produce about 85 pounds and six develop approximately 100 pounds of thrust.
The engines provided attitude and maneuver control from spacecraft separation from the launch vehicle until the adapter equipment section was jettisoned. The OAMS engines were operated by signals from the orbit attitude and maneuver electronics system. Pitch, roll and yaw torques were obtained by firing the 25 pounds thrust engines in pairs. Maneuvering was accomplished by firing the 100 pounds thrust engines for lateral, vertical and forward movement. Two of these fired to the aft to provide thrust for spacecraft separation from the launch vehicle. The two 85 pounds thrust engines fired forward to provide rearward motion.
Each engine assembly consisted of two propellant valves with calibrated orifices and filters, a fuel and oxidizer injection system, a combustion chamber, and an expansion nozzle. The propellant valves were quick acting and were normally closed and operated by solenoid action; that is, the valves open upon application of an appropriate electrical signal to permit the flow of fuel and/or oxidizer to the injector to which they were fitted. In the event of electrical malfunction of any kind, no signal was transmitted and therefore the solenoid valves remain closed. Valve construction was such that the parts in contact with the fuel and oxidizer were not adversely affected by the corrosive nature of the propellants.
The combustion chamber and the expansion nozzle were lined with an ablative material, which maintained internal geometry and protected the external wall from temperature damage.
The engines operated on storable hypergolic propellants. The oxidizer was nitrogen tetroxide; the fuel was monomethylhydrazine. Fuel quantities varied according to mission requirements.
The oxidizer and fuel propellant tanks were all-welded titanium spherical structures. Propellant tank volume and arrangement varied in accordance with mission requirements. Two propellant shut-off valves permitted isolation of the propellants from the engine chamber assemblies in the event of engine malfunction. One valve was in the oxidizer feed system and the other valve was in the fuel feed system. The oxidizer/fuel group valves were electric motor operated.
Since the OAMS functioned under a weightless condition during much of its operational lifetime, it was necessary to use a pressurization system to expel propellants on demand.
To achieve positive expulsion of propellants, both the oxidizer and the fuel were in bladder containers inside storage tanks. Gas was induced between the tank wall and the bladder to provide a "squeezing" pressure that forced propellant to the engines.
The pressurant, which forced the oxidizer and fuel to the engine, was helium. The pressurization system included a storage tank, component packages and pressure regulation group.
The pressurant tanks also were all-welded titanium spheres each with a volume of 1700 cubic inches. The pressure in the propellant tanks was regulated at 295 pounds per square inch. The number of tanks installed varied with mission requirements.
Reentry Control Systems (RCS)
The reentry control systems were completely independent of the orbit attitude and maneuver system and were installed in the reentry module just forward of the pressurized cabin. There were two completely independent reentry control systems. Each was a liquid bipropellant rocket engine propulsion system constructed on a modular basis. Both were identical and provided redundancy in the event of a malfunction of one system.
There were eight fixed-thrust-level, fixed mounted engines in each of the two systems. As in the OAMS, they operated on storable hyperbolic propellants supplied by a cold gas, pressurized, positive expulsion feed system. Electrically operated valves controlled the oxidizer and fuel flow in the thrust chamber assembly. The basic operation of each RCS was identical to that of the orbit attitude and maneuver system. The reentry control systems responded to electrical signals from the orbit attitude and maneuver electronics system, the same unit which operated to provide input to the orbit attitude and maneuver engines.
In the event of a malfunction of one system, the remaining RCS had sufficient total impulse capacity and thrust to assure attitude control during retrograde, stabilization for a safe reentry, and thrust as necessary to steer the reentering spacecraft to the desired landing point.
Attitude was controlled through the attitude control and maneuvering electronics from inputs of the astronaut's hand controller. Pitch, roll and yaw torques were obtained by selective firing of pairs of engines. Each had a nominal rocket engine thrust output rated at about 25 pounds.
The fluids involved in the operation of the RCS were identical to those in the OAMS except for the pressurant, which was nitrogen in the case of the RCS. The pressurizing group and the fuel packages were contained in the non-pressurized section of the reentry module. The pressurant tank had a fluid volume of 185 cubic inches. The pressure regulator maintained a 295 pounds per square inch to the propellant tanks.
The propellant tanks, one oxidizer and one fuel, were all-welded titanium cylindrical structures. The oxidizer tank had a fluid volume capacity of 439 cubic inches; the fuel tank had a fluid volume capacity of 546 inches. The engines of the reentry control systems had characteristics similar to those of the orbit attitude and maneuvering system 25 pounds thrust engines.
Retrograde Rocket System (RRS)
The Thiokol RRS, an independent system, consisted of four solid propellant rocket motors mounted in the retrograde section of the adapter and located symmetrically about the longitudinal axis of the spacecraft.
The retrograde rockets, each with approximately 2500 pounds of thrust, provided an impulse to the reentry module resulting in a sufficient velocity decrement to initiate reentry into the Earth's atmosphere. The retrograde rockets were fired automatically by an electrical signal from an onboard electronic timer. There was also a manual backup initiation capability. The rockets fired sequentially at a nominal 5.5-second interval.
The spent retrograde rockets were jettisoned with the retrograde section of the adapter approximately 45 seconds after rocket firing initiation. In the event of an abort before orbital altitude and velocities were achieved, the retrograde rockets could be salvo fired by the flight crew to aid in separation of the spacecraft from the launch vehicle.
Each rocket consisted of a motor case, a partially submerged contour nozzle, and dual pyrogenic igniters with removable pressure cartridge initiators. The motor case was an all-welded titanium alloy sphere slightly greater than one foot in diameter. The nozzles include an expansion cone a throat insert and a nozzle bulkhead. The pyrogenic igniter was a small rocket of short burn duration, which fired into the charge of the retrorocket thus igniting it.
A parachute landing system provided for final descent of the spacecraft to the Earth's surface. Deployment of the parachutes reduced the reentry module trim angle and decelerated the reentry module to a rate well within acceptable limits for water impact.
The parachute landing system included a high altitude drogue parachute, cable guillotines, a pilot parachute, a main parachute, a bridle assembly, attachment and disconnect assemblies, mortar assemblies, reefing cutters, displays and controls.
Landing system operation began at approximately 50,000 feet, with the astronaut deployment of the high altitude drogue parachute. It was deployed by firing mortar cartridges, which ejected the drogue parachute from its container.
When the spacecraft reached approximately 10,600 feet the pilot parachute was deployed by the flight crew. Approximately 2 1/2 seconds later the rendezvous and recovery section automatically separated from the reentry module by action of a mild detonating fuse. The pilot parachute then pulled the R and R section free of the reentry module, and as it was pulled free, it drew the main parachute out of its container. The main parachute deployed in a reefed condition.
After landing, the parachute was jettisoned by the astronauts. Transfer from the single point suspension to a two-point suspension was effected by the astronaut depressing the landing attitude switch.
High Altitude Drogue Parachute
The high altitude drogue parachute was an 8.3-foot diameter conical ribbon chute of nylon material. It was packaged in a deployment bag stored in the drogue parachute mortar tube at the top of the R and R section. Approximately 16 seconds after the high altitude drogue parachute was deployed two redundant, lanyard-actuated reefing cutters unreefed the drogue parachute. The riser from the drogue parachute was attached to the R and R section by three steel cables. The unreeled drogue parachute stabilized the reentry module within 23 degrees of the vertical axis as it descended from 50,000 feet to 10,700 feet.
Cartridge-actuated guillotines severed the high altitude drogue parachute attach cables. Actuation of these guillotines was switch-controlled by the crew. With the attach cables severed, the high altitude drogue parachute pulled the pilot parachute from its mortar, resulting in pilot parachute deployment.
The pilot parachute was an 18.3-foot diameter ring-sail parachute constructed of nylon material. The parachute, its reefing line, reefing cutters and risers were packaged in a deployment bag stored in the pilot parachute mortar tube in the rendezvous and recovery section. The pilot parachute was deployed in a reefed condition and within 2.5 seconds 0.125 seconds after deployment, the pilot parachute drew the R and R section (released by action of a mild detonating fuse) away from the reentry module. As it did so, it deployed the main parachute in a reefed condition.
Within approximately 6 seconds after pilot parachute deployment, two redundant lanyard-actuating reefing cutters severed the pilot parachute reefing line. The unreefed pilot parachute lowered the R and R section to the sea.
An auxiliary landing sequence was available in event of malfunction of the high altitude parachute. Use of this switch-controlled sequence enabled the astronauts to actuate drogue parachute attachment guillotines and the apex cable guillotine, to fire the pilot parachute mortar and to initiate R and R section separation, which resulted in main pilot parachute deployment.
The main parachute was an 84.2-foot diameter ring-sail parachute constructed of nylon material with alternating white and international orange gores. The main parachute was attached to the reentry module by a riser and a bridle arrangement. The parachute reefing lines, reefing cutters and risers were packaged in a nylon web cotton sateen bag and stored in the main parachute, a Fiberglas cylinder inside the R and R section. The main parachute was deployed in a reefed condition at approximately 9,700 feet. Within 10 seconds 2.2 seconds after main parachute deployment, redundant lanyard-actuating reefing cutters severed the reefing line. Operation of any one of three reefing cutters resulted in the unreeling of the main parachute. The unreefed main parachute provided a rate of descent that resulted in allowable loads upon water impact.
Only one main parachute was provided in the Gemini landing system. Crew escape was possible through use of the ejection system in event of spacecraft main parachute malfunction.
The spacecraft was suspended from the parachute by means of a "bridle" assembly. The bridle included a forward and an aft leg. The forward leg was stored in a Fiberglas tray in the reentry control system section of the reentry module. It incorporated a loop at one end for connection to a disconnect assembly. The aft bridle leg was stored in a trough in the reentry module skin and incorporated a loop at one end for connection to an aft disconnect assembly. The trough, located between the hatches, was covered with a frangible insulating material, which allowed the aft bridle leg to tear free when two-point suspension was initiated. Following main parachute deployment, the initial mode of descent was by single-point suspension from a single riser. Transfer from the single point suspension to a two-point suspension occurred when an astronaut depressed the landing attitude switch. After landing, the parachute was jettisoned by the astronauts.
After the water landing, the astronaut depressed the parachute jettison switch, which released both the forward and aft bridle leg attachment/disconnect assemblies. This also energized a cartridge-actuated guillotine, which severed door restraints permitting a spring loaded hoist loop and attached flashing recovery light to extend. The Gemini reentry module, unlike the Mercury spacecraft, had no attached landing bag to absorb landing shock. The Gemini spacecraft was brought to a "pilot heads-up" position causing it to impact the water at the edge of the heat shield. Water landing forces resulting were less than the touch-down forces of Mercury, even though Mercury had a landing bag installed.
Postlanding and Survival Equipment
Postlanding and survival equipment provided recovery personnel with visual and radio reference aid in locating the reentry module after landing. It provided the crew with water, food and survival equipment and included a mooring lanyard assembly and hoist loop for recovery of the spacecraft after sea landing.
The equipment included a flashing recovery light, dye marker, survival equipment, UHF and HF rescue communications and beacons, splash curtains, a hoist loop, flotation material, and electrical power supplies.
The recovery light was a high intensity flasher deployed upon parachute jettisoning and located on top of the reentry module as it floats in the water. The light had a minimum flashing rate of 15 flashes per minute and was visible on a clear night at a distance of 50 nautical miles.
Fluorescent dye marker was installed in the forward end of the reentry control system section just below the flotation line. The dye was green-yellow and stored in a container having openings covered with a water-soluble film for automatic deployment upon exposure to the water.
The radio recovery aids included a UHF recovery beacon, UHF voice transceiver, HF voice transceiver and the rescue beacon in the survival pack.
The mooring lanyard assembly consisted of a lanyard with a fitting at one end for attachment to the "D" ring on the astronaut's parachute harness. A life raft was attached to the lanyard at a point about 8 to 12 feet from the astronaut.
During an ejection sequence the lanyard assembly extracted the life raft and a survival pack from containers in the seat as the backboard of the seat falls away from the astronaut. A cylinder charged with carbon dioxide was packed with the life raft for inflation. The life raft could be manually extracted following a normal landing in the spacecraft.
The Gemini escape system for safe escape of the astronauts through use of the ejection seats while the launch vehicle was still on the launch pad during boost and after spacecraft reentry. While ejection capability was available throughout the mission, actual usage was determined by the altitude, the type of emergency, the system condition, the mission phase, and the astronaut evaluation of the problem. Subcontractor for the ejection system was Weber Aircraft.
Two ejection seats were provided in the Gemini spacecraft for emergency escape. These seats included the seat structural assembly, a backboard assembly and an egress kit assembly. The seats were arranged side-by-side facing the small end of the crew compartment. They were constructed to be compatible with crewmembers wearing fully inflated pressure suits. The escape system incorporated a survival kit containing water, food, life raft, fishing gear, a radio transmitter and a machete, all packed into the seat.
The ejection seats functioned as one complete system. Should the need to abort arise, the decision to eject was made by the astronauts themselves. Once it was determined to eject, either man could pull the escape ring located between his knees, ejecting both astronauts. Only when manually initiated by an astronaut would the ejection sequence occur.
When the system was actuated, the remainder of the operation was fully automatic. First, both hatches of the Gemini spacecraft were opened simultaneously, then the rocket powered escape seats were propelled out of the vehicle.
Then, 1.1 seconds after ejection, the seats and men separated; 2.3 seconds later, a drogue gun fired, extracting a pilot chute from the astronaut's back pack. The deployed pilot chute then pulled a 28-foot diameter main parachute out, allowing for full canopy inflation.
In a pad abort, just 10 seconds after leaving the Gemini spacecraft, both astronauts descended to a safe landing in a nearby cleared landing area. After reentry the astronauts would have had the option of riding the spacecraft to water impact beneath a Northrop-Ventura ringsail parachute or ejecting themselves and landing much like a paratrooper.
In most instances the latter method would not be used unless the spacecraft entered the Earth's atmosphere at a point where a dry landing had to be performed.
The Gemini rocket catapult, furnished by Rocket Power, Inc., had a total impulse of 2650 pound-seconds. Burning time for the rocket was approximately 1/4 second. Rocket ignition occurred 0.2 seconds after catapult ignition. The astronauts would be subjected to maximum ejection acceleration of 24 g.
Trajectories that could be expected for "off-the-pad'' ejections would land the astronauts at least 500 feet from the launch vehicle. Tests indicate that the landing point would actually be closer to 800 feet from the launch site. The height of trajectories on 'off-the-pad' tests was approximately 350 feet above the terrain when launched from 150-foot height.
Escape systems testing was begun at the US Naval Ordnance Test Center, China Lake, California, in 1962, using a 150-foot-high tower. The 150-foot tower was used to simulate escape from the spacecraft while still on the pad. During the field testing operation, environmental studies were conducted on the system's related components, including harnesses, back pack, straps and pyrotechnics to determine component reaction under extreme heat, cold, humidity, shock, and vacuum.
High speed track tests were conducted with a full-scale Gemini boilerplate spacecraft mounted on a rocket powered sled. During sled tests, every possible escape condition was simulated by firing each seat at different attitudes to determine how the system would perform under adverse conditions. The sled was accelerated to 550 mph and the escape system actuated, causing the seats to be ejected out and away from the vehicle to qualify the system for high speed ejections. High altitude tests were conducted at the Naval Parachute Facility, El Centro, California, utilizing an F-106 supersonic fighter flying at Mach 1.75 and 20,000 feet. More than 100 studies and tests were conducted in the laboratory and field before the system was man-rated as operational Gemini equipment.
If the returning astronauts used their ejection seats at high altitude, the ejection sequence would function as in a pad abort situation with one exception. At altitudes above 7,500 feet a device called a ballute were employed. Designed and built by Goodyear, the ballute would stabilize and decelerate the ejected astronauts during the free fall before parachute deployment.
The ballute, a contraction for balloon-parachute, was a stabilizing device included as part of the ejection system. The ballute was packaged in the ejection seat and was utilized during high altitude aborts to achieve desired stabilization of the astronaut.
The ballute was a balloon-shaped device looking much like a child's spinning top. It was constructed of inflatable rubberized fabric and was packaged in a deflated condition in the ejection seat during flight. The inflated balloon was approximately 48 inches in diameter and 54 inches long. The astronauts could use the ballute at heights up to 14 miles. When an altitude of approximately 5700 feet was reached, the personal parachute would open automatically and bring the astronaut safely down to Earth.
Pyrotechnic (explosive) devices were used throughout the Gemini mission for switching operations and major configuration changes and in recovery and escape systems. All pyrotechnic systems were designed with inherent redundancies. Pyrotechnic devices were initiated by either pressure, electrically, or by means of a lanyard.
The pyrotechnic devices were installed in such a manner that explosive effects were confined within the device or directed outward and shielded in such a manner as to preclude damage to nearby equipment or inward release of shrapnel
Location and Function of Pyrotechnic Devices
|1||Separation Assembly, FLSC||Severed the spacecraft from the launch vehicle and the adapter equipment section from the adapter retrograde section.|
|2||Guillotine||Cut cables and wire buddies.|
|3||Cutter-Sealer||Cut and crimped seal the OAMS fuel and oxidizer lines.|
|4||Horizon Sensor Fairing||Jettisoned the horizon sensor fairing.|
|5||Horizon Sensor Release||Jettisoned the horizon sensors.|
|6||Ventilation Door Release||Jettisoned the ventilation door.|
|7||Cartridge, Pyrotechnic Valve||Opened or closed pressurant, propellant, or oxidizer lines.|
|8||Pyrotechnic Switch||Opened circuits prior to guillotine operation.|
|9||Separation Assembly, MDF||Separated rendezvous and recovery section from the reentry module.|
|10||Retrograde Rocket igniters||Ignited the retrograde rocket motors.|
|11||Retrograde Rocket||Provided thrust for velocity decrement necessary for reentry.|
|12||Parachute Mortar and Reefing Cutter||Parachute mortar deployed parachute in reefed condition, reefing cutters allowed complete parachute deployment.|
|13||Parachute Disconnect||Changed reentry module suspension and jettison parachute.|
|21||Separation Assembly, FLSC||Severed the reentry module from the retrograde section.|
|14||MDF initiation System||Initiated the hatch actuators.|
|15||Hatch Actuator||Opened hatch before ejection seat operation.|
|16||Rocket Catapult||Provided ejection seat propulsion during ejection seat operation.|
|17||Harness Release Actuator||Unlatched seat restraint harness after seat ejection.|
|18||Seat-Man Separator||Separated astronaut from seat.|
|19||Ballute Deploy and Release Assembly||Deployed and release ballute according to altitude.|
|20||Drogue Mortar-Backboard Jettison Assembly||Deployed astronaut drogue parachute.|
Electrically-initiated devices incorporated cartridges and detonators. They were used in the flexible linear shaped charge assemblies, guillotines, cutter sealers, release assemblies, valves, pyrotechnic switches, mild detonating fuse separation assembly, retrograde rocket initiators, parachute mortars, and parachute disconnects. The cartridges and detonators were provided with one or two independent electrical bridge wire circuits as reliability dictates.
Lanyard-initiated devices were the parachute reefing cutters, mild detonating fuse initiation system, harness release actuators, ballute deployment and release systems, and drogue mortar-backboard jettison assemblies.
Pressure-initiated devices were utilized in the hatch actuators, rocket catapults, and seat-man separators. These devices fired when exposed to pressures between 500 and 3000 pounds per square inch except for the rocket catapult which fired when exposed to 300 to 9000 pounds per square inch.
Flexible linear-shaped charge assemblies were sections of electrically initiated charges capable of separating sheet metal, wire bundles, straps, and tubes. Three such assemblies were provided to separate the launch vehicle, the adapter equipment section, and the adapter retrograde section from the reentry module. The assemblies completely severed inter-connecting tubes, wires, bundles, titanium straps and structural skin of the sections.
Guillotines were provided for cutting wire bundles, cables, and bolts. These were knife-like devices explosively driven through the item to be severed. Redundancy was provided through use of two guillotines, one on either side of the separation plane. The guillotines used for severing cables or bolts contained two cartridges. Function of either cartridge was sufficient to sever the cable or the bolt.
Cutter-sealers provided for sealing and cutting of the orbit attitude and maneuvering system propellant lines prior to section separation. The lines were sealed to prevent leakage of fuel or oxidizer residue at the same instant that the tubes were severed. Redundant cutting was provided by using two cutter-sealers, one on either side of the separation plane. Each contained one cartridge.
The horizon sensor fairing release assembly was secured to the spacecraft by a cartridge-activated single-point hold- down device containing two cartridges The cartridge was sufficient to jettison the fairing. The horizon sensor fairing release assembly was enclosed to preclude the escape of any pyrotechnic by-products, thus protecting the delicate horizon sensors against damage during fairing jettison.
The horizon sensor was secured to the spacecraft by a two-cartridge, single-point, hold-down device. The function of a single cartridge was sufficient to jettison the sensors and the cartridges from the spacecraft. This was accomplished following retrograde and prior to reentry. The electrical connectors, through which the cartridges were initiated, were disconnected by the force of ejection.
Normally open or normally closed pyrotechnic valves were provided for isolation and control of pressurants and propellants. Normally closed pyrotechnic valves isolated the pressurants and propellants in the tanks of the orbit attitude and maneuvering system and the reentry control system from the remainder of the system during prelaunch. Orbit attitude and maneuvering system valves were pyrotechnically opened shortly before launch. The normally closed reentry control system valves were opened prior to reentry. The orbit attitude and maneuvering system was provided with both normally closed and normally open pyrotechnic valves actuated in the event of pressure regulator malfunction.
Pyrotechnic switches were provided for opening electrical circuits in the wire bundles prior to severing. Each pyrotechnic switch contained a single cartridge with dual bridge wires.
A mild detonating fuse separator was provided to break all bolts that attached the rendezvous and recovery section to the reentry module. The separation assembly contained two detonators and an explosive ring assembly. Booster charges initiated by the detonators caused simultaneous detonation in both mild detonating fuse strands of the explosive ring assembly. Either strand had sufficient energy to break all the attachment bolts.
Each retrograde rocket contained two independent cartridge-actuated igniters mounted adjacent to the rocket nozzle. Each igniter contained an electrically actuated cartridge, 20 boron/potassium nitrate pellets, and a 0.04-pound solid propellant charge of polysulfide/ammonium prechlorate. When fired, the cartridge ignited the boron/potassium nitrate pellets which in turn ignited the solid propellant charge. The propellant would burn for 0.35 second, discharging its exhaust gases into the retrograde rocket cavity and igniting the basic retrograde rocket. Either of the two igniters provided for each rocket was sufficient to ignite the retrograde rocket.
The parachute mortar which deployed the pilot parachute, contained two electrically-initiated cartridges. Function of a single cartridge was sufficient to eject the pilot parachute from the mortar tube. The pilot parachute was deployed in a reefed condition.
Two lanyard-initiated reefing cutters, incorporating a 6-second pyrotechnic time delay, were provided to complete pilot parachute deployment. Three lanyard-initiated reefing cutters, incorporating a 10-second pyrotechnic time delay, were provided to complete main parachute deployment.
After the main parachute had deployed and the spacecraft was suspended from a single point, a parachute disconnect was actuated to allow the spacecraft to rotate to a two-point suspension. Two parachute disconnects jettisoned the main parachute upon impact. Each of these disconnects contained two electrically initiated charges.
Manned spacecraft fabrication techniques developed at McDonnell for the Mercury spacecraft were extensively applied in the Gemini program. Weight limitations combined with heat resistance, air load, and acoustic requirements necessitated improved fabrication practices and placed even greater demands on assembly techniques. New welding techniques were developed at McDonnell to meet many of these demands.
The adapter skin was a magnesium-thorium alloy of 0.032-inch thick sheet stock. Welded on the inside was a continuous magnesium-thorium tee-bulb extrusion that served both as a stiffening stringer and a closed passage for the coolant that converted the adapter to a radiator. Intimate contact between skin and extrusion to facilitate heat transfer was accomplished by using a weld-through sealer for conductivity and as a moisture barrier. Seam welding then mated the parts.
These skins were fabricated in quarter panels of two sizes. Radiator extrusions on each panel were joined on assembly by using filler wire and a small hand torch to make fillet welds at each end of the sleeve joint. Since the radiator extrusion was already seam welded to the adapter skin, a mirror was necessary to make this difficult weld in the confined space.
Penetrant and X-ray inspection was then made and a sustained pressure check on the system to assure no leaks exist. A pressure drop test assured that no weld burn-through had occurred to restrict coolant flow.
Most welding problems centered around the pressurized cabin. The cabin was made up of a spherical bulkhead on the large end and a flat circular bulkhead on the small end. Eighty-five percent of the cabin section, which included equipment bay doors and hatches, was made of welded titanium assemblies.
A bonded honeycomb structure was considered but discarded in favor of the welded assembly. An airtight cabin was required to hold the life-sustaining atmosphere for the astronauts and a welded assembly was ideal from this standpoint. It also provided a relatively smooth surface, even in welded areas, upon which to mount many of the over eight hundred stiffeners, brackets, clips and equipment holders that made up the completed cabin section.
The flat walls of the cabin consisted of two titanium sheets resistance-welded together - 0.010 inch beaded skin on the inside, 0.010 smooth skin on the outside. Brackets and doublers were attached by spot welding and the wall panels were attached to the mating structure by seam welding.
Advanced techniques, improved equipment and higher weld reliability from experience gained on the Mercury project were important design factors.
Large Pressure Bulkhead
Strengthening beads appeared on the outside of' the large pressure bulkhead (the backs of the astronauts' seats were next to this bulkhead), while the skin on the inside was smooth. Both skins were 0.010-inch titanium and similar in construction to the cabin walls except the contour was spherical. A special "birdcage" weld fixture held the contour while skins were being spot welded.
Unique problems were presented by the hatch sill. A groove 3/4 inch wide and 1/2 inch deep around each of the two hatch sill openings accommodated the hatch door seals. Part of the cone-shaped cabin was "scooped out" to allow the astronaut to look out the window in the hatch. This presented reverse contours in the "eyebrow" area. A complete "double" hatch sill was made up of eighty-eight welded details and required over seven hundred linear inches of automatic and hand-fusion welding.
Hatches for crew access to and from the pressurized cabin area mated with close tolerance to the hatch sill. Fabrication problems were similar to the hatch sill but, in addition to the conical shape and reverse contour of the "scooped out" area, an observation window installation was provided. The window frames were welded assemblies and in turn were welded into the hatch opening. Two hundred eighty-five inches of hand-fusion welding were required to mate the thirteen titanium pieces of each hatch.
Several of the titanium cabin subassemblies were fabricated by fusion welding, which consisted mainly of hand or automatic applications of tungsten-inert gas with or without filler wire.
Application of McDonnell-designed welding chambers speeded production. The weld fixture containing the production assembly was put into a weld chamber, chamber air was purged by inert argon gas, and a skilled welder outside the chamber made the fusion weld while peering through windows and holding the welder unit in rubber pressure gloves.
Several sizes of welding chambers were designed by McDonnell to accommodate the various sizes of weld fixtures and production assemblies. Features of these chambers included replaceable flat plastic windows to minimize visual distortion, elimination of the argon cylinder by piping the gas from a central source, and a small positioning fixture that held the assembly at any desired angle.
Some production assemblies presented unusual problems and required special weld chambers. One of these was a special weld chamber made to hold the hatch sills in the proper relationship to the environmental control system (life support and equipment cooling) box while the structural skeleton of the Gemini pressure vessel was welded to them. This tool was affectionately called the "green-house" because of its many plastic windows.
The amount of automatic and hand welding on the cabin assembly alone was about two hundred fifty feet. This did not include spot, stitch or seam welding.
Automatic Fusion Welding
Two automatic welding units with the boom-mounted torch extending over the fixtures moved on rails behind the line of automatic weld fixtures. This gave the TIG welding head three-axis movement. Each weld fixture was supplied with air for clamping and argon gas for backup. These units produce burr welds, angle welds, "T" welds, and corner welds in straight or contoured configuration. One unit made burn-through "T" welds of two pieces and also produced "T" legs at various angles. This fixture was the only one requiring cooling water in addition to copper chill bars for temperature control.
Strain relief fixtures were necessary after most welding operations to prevent warpage of materials involved.
The innovation of air pads on weld fixtures was of great importance in Gemini spacecraft production. Air pads supported a heavy weld fixture, making it possible to move a heavy assembly with the touch of a finger, while an assembly was spot or seam welded in a Sciaky welder. Six Sciaky machines were installed in a line. A smooth aluminum jig plate floor was located in front of each two welders. The air pads replace casters on the weld fixture and. with standard shop air. supported up to 400 pounds per pad. A ten-inch diameter pad was developed at McDonnell to ride 0.003-inch above the jig plate floor.
Fusion Welding Inspection
All welds got a visual (size and shape) inspection. A penetrant inspection was done on non-magnetic materials. One hundred percent radiographic inspection was done with few exceptions. Inspection fixtures were designed to check tolerances at various stages of assembly.
Resistance Welding Inspection
The welding machines were certified for specific material-thickness combinations. Test specimens made prior to production runs simulated the production spotweld and were used for shear and microscopic evaluations. Inspection for nugget penetration was made on spot, stitch and seam welds. When shear test samples were not made on stitch and seam weld, examination was made for minimum nugget diameter. Production welds were penetrant inspected before final acceptance.
Many of the manufacturing aspects of spacecraft were typical of the aircraft industry. However, many were peculiar requirements that require significant refinement of old techniques or the development of new techniques.
Some of the conspicuous special procedures received considerable publicity. These were usually the ones involving assembly and test operations in the "clean rooms". There were other operations that necessitated the improvement of old or the development of new techniques, however.
Materials used in the spacecraft were either inherently corrosion-resistant or were processed to resist corrosion in the various environmental conditions which could be encountered by the spacecraft. Also, materials which were not encapsulated or contained in hermetically sealed enclosures were either inherently fungus-resistant or were processed to resist fungus attack.
The prevention of corrosion in the magnesium extruded stringers used as a coolant loop in the space radiator posed unique problems. These stringers were procured as extrusions of varying lengths up to 60 feet. Protection started with the handling of the stringers at the supplier through the receipt of the material at McDonnell, its subsequent processing in fabrication both at McDonnell and at a West Coast chem-mill subcontractor, and up to and including its completed installation as a spacecraft plumbing system ready for servicing and functional application. A corrosion preventative compound was applied to provide protection of the material until completion of the assembly of the coolant loop within the adapter. Prior to the baking of the spacecraft adapter exterior paint finish and coupling spacecraft operational components and modules to the coolant loop, the system was purged to remove all traces of the corrosion preventative compound.
Subsequent corrosion protection was provided by pressurizing the closed system with dry nitrogen gas until the spacecraft was serviced with coolant.
Heat Shield Non-Destructive Testing
An extremely important requirement was internal structural integrity of the spacecraft heat shield. The basic configuration consisted of a honeycomb core of approximately one-quarter million cells which were filled with a poured- plastic compound. The completed shield was radiographically evaluated for internal structural soundness. The complete structure was recorded on X-ray film documenting out-of-tolerance conditions, i.e., lack of bond, inclusions, or voids.
The size of the assembly and number of cells to be evaluated dictated the need for a coordination technique that could achieve accurate and consistent correlation between exposed film and the heat shield structure. A 0.020 clear Mylar cover was tailored to heat shield dimensions and contour and provided with locators, which mated with index markings on the heat shield. Horizontal and vertical grid lines were marked on the cover to establish sections. Each section carried a location identity. This identification recorded on the X-ray film during radiographic exposure and become permanent location information. Discrepancies affecting structural soundness were readily visible on exposed X-ray films, but a Polaroid film pack was used to more accurately locate a discrepancy in a given grid section. A series of lead arrows were placed in the proximity of the affected area, the Polaroid film pack was laid up on heat shield and exposed with the radiographic unit. The affected cell could be pinpointed in a series of three to four exposures by reviewing the previous Polaroid "shot", repositioning the lead arrows closer to the defect and reshooting. The defective cell was then marked with a map pin for re-work.
Clean Room Production
Preparation of functional equipment, installation into the spacecraft, and subsequent testing took place in clean rooms. The structural assemblies of the spacecraft sections were completed in an assembly area outside the clean room. Equipment to be installed was tested in the clean room. When the structural assembly of a section was complete, the section was thoroughly cleaned and brought into the clean room, where the installation of equipment began.
Clean room requirements for testing and installation of equipment in the spacecraft stemmed basically from the fluid and gas systems, where small foreign particles or small amounts of corrosion could prevent or degrade the functioning of valves, regulators, pumps, etc., or could cause loss of fuel, oxygen, coolant fluid, pressurizing gas, or other expendables, by preventing the complete sealing of valves.
Specific cleanliness requirements were defined for manufacturing such equipment, and the suppliers had clean rooms with controlled temperature, humidity, and air filtration. Equipment was protected during shipment by capping of openings and sealing in plastic enclosures with necessary dessicants included. Suppliers also provided information tags in the enclosures to define the degree of cleanliness of the part so that no one would inadvertently break the seal in an uncontrolled area.
Depending on the exact level of cleanliness deemed necessary these items were uncovered, tested, and installed in either the McDonnell Class 10 (the most extreme requirement) or the Class 6 Clean Rooms.
The occupants of the Class 6 room were clad in white. Those in coveralls and sneakers were full-time clean room personnel. Those in smocks and disposable booties were various personnel who entered the clean room for only short periods of time.
In the Class 10 Clean Room, personnel wore hoods and gloves which provided even more complete coverage than the clothing used in the Class 6 room.
A third controlled-cleanliness area was maintained for fabrication of electrical subassemblies such as wire harnesses and relay panels.
Spacecraft manufacturing quality control procedures differed from aircraft quality control procedures in degree rather than nature. McDonnell examined the condition of parts more thoroughly and maintained more exacting criteria for acceptance. For instance, no sampling inspection procedures were used for anything except standard nuts, bolts, and fittings. McDonnell tested 100% by X-ray, in 3 views, all transistors procured for the Gemini and required major Gemini electronics equipment suppliers to do the same. These X-ray photographs were examined for foreign inclusions or other detectable internal abnormalities.
In maintaining cleanliness of components and tubing used in the gas or fluid systems, the flushing fluids were inspected until microscopic examination verified satisfactory cleanliness, that is, contaminant particles in the 2-10 micron size range. As a graphic illustration of the degree of refinement this meant, a typewritten period is approximately 500 microns in diameter.
A necessary facet of quality assurance on a program such as Gemini was the amount of record keeping required. Most equipment items down to the level of switches, relays, pyrotechnic cartridges, etc., were controlled by serial number. All serialized items required logs which recorded significant events in their life. Likewise, higher-level equipment necessitated sets of data sheets recording exact results of all tests that they underwent. These data provided a permanent record of the behavior of each item of equipment in the spacecraft. Should last minute concern arise as to the suitability of a particular equipment for an imminent mission, its total manufacturing and performance history could be examined.
Sealing Of The Gemini Spacecraft
While much of the pressure vessel of the Gemini spacecraft was welded titanium, which eliminated leakage, there were a number of instances where construction involved bolts and rivets with their attendant possibilities for loss of cabin pressure. In addition, the egress hatches require sealing as they open directly into the internal pressure vessel.
Each hatch was manually operated by a mechanical latching mechanism from either the inside or the outside of the vessel. The latch forces dictated the use of a soft silicon rubber seal. For proper alignment for the hatch striker and seal in the manufacturing process, the channel was filled with soft putty to determine the striker contact. When the desired alignment was obtained, the putty was removed and the rubber seal was installed in the seal channel. The seal channel frame-to-structure joint was sealed with a room temperature vulcanizing General Electric silicon sealant.
Each of the ingress/egress hatches incorporated a visual observation window consisting of inner and outer glass assemblies. The outer assembly was a single flat pane gasketed on each side with a 0.04-inch Fiberfrax paper. A hollow metal O-ring provided the seal between the periphery of the glass and the frame. The inner window assembly consisted of two flat panes and was sealed with silicon rubber flat gaskets and silicon rubber O-rings.
Other doors which provided access to the equipment compartments within the pressure vessel were sealed by means of a gasket design.
In cases where it was necessary to add a shelf, a piece of equipment, or provided an attachment after major assembly was complete, the area was reinforced by spot welding a stiffening member to the structural wall. A seal problem was created if a hole was necessary under the angle or doubler. Normally the hole in the double wall could be sealed by overlapping spot welds. The hole for a late attachment was sealed by placing a piece of 0.005 silicon coated Fiberglas tape in the hole area under the attaching member. The bolt pressure then sealed off the air entrance to the hole in the wall and prevented pressure loss. Stat-O-Seal washers were used on both sides of a joint of this type to prevent leakage around the bolt.
The transmission of electrical power and signals in and out of the pressure vessel was accomplished by terminating electrical wire bundles at the walls of bulkheads with sealed connectors. The connectors were sealed at the structure surface by an O-ring grooved flange. All connectors were potted with a room temperature vulcanizing silicon sealant compound to provide a moisture and pressure seal and to provide support for the wires at the soldered terminals.
Gemini Thermal Radiation Control Coatings
The Gemini spacecraft required a number of thermal radiation control coatings to reflect and/or re-emit external and internally generated energy when the spacecraft was subjected to heat loads during ascent, orbit or reentry. The Gemini spacecraft utilized super alloys, beryllium, high temperature thermal insulation, ablative heat shields, in addition to the thermal radiation control coatings, to alleviate these high and low temperature extremes. During orbit, the adapter section also served as a space radiator or heat exchanger for the dissipation of internally generated heat; therefore, the exterior surface of that part of the spacecraft had a very low solar absorption and a very high infrared thermal emittance to maintain its desired temperature characteristics.
The internal surface of the adapter module walls required a very low thermal emittance to reduce the heat transfer by radiation between the skin of the adapter and the interior equipment. Also a flexible gold-plated fabric thermocontrol cover prevented escape of heat from the interior of the adapter and kept solar radiation off the internal equipment.
The Rene 41 shingles on the side of the reentry module were heat oxidized to provide a stable adherent, high temperature resistant high emittance surface finish. The beryllium shingles were chemically oxidized to provide a high emittance surface. An air curing silicate bonded black ceramic coating that had a significant degree of toughness at room temperature, good thermal shock resistance, and would withstand temperatures as high as 2300 degrees was used to repair any scratched or damaged areas that could affect the surface of the shingles prior to launch. The inside of the beryllium shingle was coated with a very thin layer of gold to reduce the heat emittance radiated from the shingles to the interior of the rendezvous and recovery section and the reentry control system section.
Proper thermal balance of the spacecraft and its equipment was not complete with only heat rejection features. Proper temperature of the equipment in the adapter section was the result of balancing several conflicting phenomena: namely, radiation from their surfaces to the adapter structure, radiation of heat from direct exposure to solar absorption when the spacecraft was on the daylight side of the Earth, and heat emission from the base of the adapter when the spacecraft was on the dark side of the Earth. The effect of these two phenomena was controlled by coating the interior of the adapter structure with a coating, which was highly reflective at relatively low temperature and covering the open end of the adapter with a surface which was relatively absorptive at the temperatures of solar heat. The interior of the adapter module of GT-3 was coated with aluminum foil tape, which had a silicon pressure resistant adhesive. A gold coating deposited at room temperature with a water base mixture was utilized in later spacecraft due to it ease of application and its reduction in weight.
An intermediate coating of white epoxy was applied to the treated magnesium prior to the application of the gold spray. This coating provided coating protection for the sub-structure and also provided a smooth surface so that the lowest emittance could be obtained with the gold coating. Very few silicon and acrylic bonded coatings were available that would meet the 600 degree requirement for the primary radiator coating. A potassium silicate bonded zinc oxide was selected as the primary space radiator coating on the outside. A silicon-bonded zinc oxide was selected for minor touch-up of areas that would not require the maximum 600-degree ascent temperature resistance and as a primer for the Fiberglas fairings.
The porous Dow 17 Type 1 treated magnesium skins of the adapter to which the coatings were applied were kept clean during assembly by a drillable low adhesive protective paper, which was applied immediately after the treatment. This paper allows rivet patterns to be laid out and drilled with minimum damage or contamination of the clean surface. After assembly, the protective paper was removed and the adapter was cleaned to remove any adhesive. Frayed surfaces were sealed to prevent any entrapped chromate solution resulting from the processing of the assembled parts or resulting from the weld-through sealer used during the welding of the magnesium coolant tubes to the magnesium skins.
After the adapter coating was completed, it was cured at 310 degrees and then scrubbed with a special cleaner and steam cleaned until a water break free surface was obtained. The adapter was then sprayed with alcohol to remove surface water. Masking was accomplished with a low adhesive tape and the Fiberglas fairings were coated with a silicon bonded zinc oxide coating and allowed to dry. The adapter was then coated with a silicon bonded zinc oxide coating in several coats until a total thickness of 4 to 5 mils was achieved. This required several successive coatings with each coat applied immediately after the carrier flashes off. The coating was cured at 310 degrees after several hours of air-drying.
The coatings were selected after being subjected to tests in simulated launch temperatures using furnaces and vacuum chambers and ultraviolet radiation for a period equivalent to two weeks.