One of the most crucial phases of many interplanetary missions is orbit insertion. Everything must go right the first time, or the spacecraft fails to enter orbit around its target. And, all too often, there are no practical opportunities at a second chance, so an orbit insertion failure usually translates into a mission failure. With so many spacecraft being launched to orbit various targets throughout the solar system, it seems like this is a good opportunity to review orbit insertion failures over the last half century of planetary exploration to see what can and did go wrong. For this list, we are only considering missions which were destined to orbit another solar system body beyond the Earth and survived launch only to succumb to problems as they were in transit or arriving at their targets.
Mars 4 – 1974
The very first planetary orbit insertion failure involved a Soviet mission launched to Mars in 1973. Back in 1971, a pair of new 4,650-kilogram 3M spacecraft, Mars 2 and 3, met with limited success in their missions to deploy the first landers on the Red Planet and enter orbit as part of the Soviet M-71 mission. Both landers failed (although the Mars 3 lander did briefly transmit from the surface for 20 seconds) and both orbiters experienced issues which limited the usefulness of their data. In the case of the Mars 3 orbiter, a problem with the spacecraft’s new autonomous navigation system resulted in an abbreviated orbit insertion burn allowing the spacecraft to barely attain orbit just missing being the first orbit insertion failure of the Space Age (see “The Mars Orbiter That Almost Was Not”).
Mars 3, with its lander visible on top, shown before its launch.
In order to beat the planned landing on Mars by the American Viking mission in 1976, an armada of four improved 3M spacecraft were prepared as part of the Soviet M-73 mission. This group of spacecraft, which consisted of pair of 3,440-kilogram 3MS orbiters, Mars 4 and 5, as well as a pair of 4,470-kilogram 3MP lander-laden flyby craft, Mars 6 and 7, was the second attempt by the Soviet Union to reach Mars using the new generation of advanced 3M spacecraft employing the powerful Proton-D as a launch vehicle.
The Mars 4 orbiter in its cruise configuration. Note that the lander has been replaced with an instrument suite to study Mars from orbit. (NASA/NSSDC)
During the late stages of assembly and test of this quartet of spacecraft, engineers discovered that a key transistor used in systems throughout the ships, designated 2T-312, had not been manufactured to the established requirements. Someone had decided to cut corners and use cheaper aluminum leads instead of gold as had been specified and without the performing the appropriate tests to determine that the change would work. As a result, they found that the time-to-failure for this vital component had been cut significantly due to corrosion of the leads and thus there was even odds that any given spacecraft in the Mars armada would fall victim to a failure of a vital system before reaching its target.
With no time to replace the faulty components before launch, the decision was made to launch all four spacecraft and hope for the best. Mars 4 was launched on July 21, 1973, with its sister orbiter, Mars 5, following on July 25. Mars 6 and 7, carrying their landers, launched on August 5 and 9, respectively. As had been feared, the failure of the 2T-312 transistors affected all four spacecraft to some degree or another. In the case of Mars 4, two of the three channels of the main computer, including, unfortunately, the channel that controlled the main KTDU-425A engine, failed after the first course correction. This meant that Mars 4 could no longer use its propulsion system to make course corrections or enter orbit around Mars.
Mars 4 was reprogrammed to acquire image strips during its unplanned flyby of Mars such as this one taken using its panoramic imaging system. The circular Argyre Planitia in Mars’ southern hemisphere is visible just right of center.
Mars 4 flew 1,844 kilometers above Mars on February 10, 1974. While it was reprogrammed to acquire a swath of images and make other observations during its brief encounter with the Red Planet, its mission was largely unsuccessful and resulted in the first planetary orbit insertion failure. Mars 5 successfully entered orbit on February 12 but went silent after only 16 days. While a blow for Soviet aspirations for Mars, the 3M spacecraft design was adapted for a series of highly successful Venera missions flown from the mid-1970s to mid-1980s using the 4V design and its variants (see “Venera 9 and 10 to Venus”).
Phobos 1 – 1989
After a hiatus of 15 years of Soviet missions to Mars (and 13 years after the launch of the American Viking missions), the Soviet Union launched the 6,220-kilogram Phobos 1 on July 7, 1988 on a Proton-D launch vehicle. Along with its sister craft, Phobos 2 launched on July 12, these spacecraft were the first of a totally new, third-generation interplanetary spacecraft replacing the highly successful Venera design that had been used for a decade to explore Venus and adapted as part of the Soviet Vega mission to Venus and Comet Halley (see “The Missions to Comet Halley”).
A Phobos spacecraft prototype shown during assembly. (NPO Lavochkin/RussiaSpaceWeb)
The mission of this pair of spacecraft was to rendezvous with the small inner moon of Mars, Phobos, and study it at ranges as close as 50 meters with a suite of state-of-the-art instruments. Building on the experience of international cooperation fostered during the earlier Vega missions, the Phobos mission featured cooperation from 14 other nations providing instruments and expertise, with the US contributing use of its Deep Space Network. Included on both spacecraft was a small stationary lander that would be deposited on the surface during a low flyby of the moon. Phobos 2 also carried a small mobile “hopper” that would take advantage of the low-gravity environment of this moon to hop from one location to the next, making in situ measurements along the way.
But as had happened far too often in the past, the first flights of a new Soviet planetary spacecraft design inevitably uncovered problems that led to mission failure. On August 29, 1988, a software upload was transmitted to Phobos 1 but a single character error, coupled with the lack of appropriate safeguards in the flight operating system, inadvertently resulted in the spacecraft’s attitude control system being deactivated—something that would only be done during ground testing. Unable to keep its solar panels oriented towards the Sun, the batteries on Phobos 1 drained, resulting in the loss of the spacecraft. The error was not discovered until nothing was heard from the spacecraft during a scheduled communications session on September 2. Attempts to regain contact with Phobos 1 continued until the mission was declared a loss on November 3, less than three months before it was to enter orbit around Mars.
A view of Phobos acquire on March 25, 1989 by Phobos 2 just two days before contact was lost with the spacecraft. (Russian Academy of Sciences/Ted Stryk)
Phobos 2 fared better, entering orbit around Mars on January 21, 1989. Although orbit insertion was successful, ground controllers had discovered that one of the three processors in the flight control system had failed with another showing signs of trouble. While the mission continued in preparation for a scheduled April 7 rendezvous with Phobos, control of the spacecraft was lost during a session to observe the Martian moon on March 27 and was never regained. While it was not known at the time, these would be the last Soviet planetary missions before the dissolution of the USSR in 1991.
Mars Observer – 1993
Mars Observer was the first American mission to Mars since Viking was launched in 1975. It was the first in NASA’s new Planetary Observer series of missions that adapted existing spacecraft and instruments for planetary missions with the intent to keep costs low in a time of tight federal budgets. To this end, Mars Observer employed General Electric’s three-axis stabilized Satcom K communication satellite bus as well as subsystems from military and civilian polar orbiting meteorological satellites. Mars Observer’s mission was to perform a wide range of investigations of Mars using an extensive suite of remote sensing instruments from a circular 350-kilometer orbit.
Mars Observer shown being prepared for launch. (NASA)
After years of delays, cost overruns and a switch from the Space Shuttle to a Commercial Titan III/TOS for launch in the wake of the 1986 Challenger disaster, the 1,018-kilogram Mars Observer finally lifted off on September 25, 1992 and was sent on its way to Mars. In order to avoid issues encountered with the propulsion system on the Viking spacecraft during their long transit to Mars, the decision had been made seven months before launch to postpone the pressurization of the propellant tanks that fed a redundant pair of 490-newton bipropellant engines to be used for major maneuvers including the vital 28-minute, 50-second orbit insertion burn. Minor course corrections took place during the cruise to Mars in a “blow down” mode that did not require full pressurization of the system.
The launch of Mars Observer on September 25, 1992. (NASA)
On August 21, 1993, with just 68 hours to go before orbit insertion, the propulsion system was finally pressurized. Because of concerns about the effect the mechanical shock caused by detonating the pyrotechnics used to open the valves would have on the radio electronics in a powered state, the decision was made to shut down Mars Observer’s transmitter during this vital operation. Unfortunately, Mars Observer was not heard from as scheduled 14 minutes after pressurization and all attempts to regain communications failed.
The first image of Mars returned by the Mars Observer Camera (MOC) on July 27, 1993 as the orbiter was approaching Mars. (NASA)
Investigators never determined a definitive cause for the loss of Mars Observer because of a lack of telemetry during the pressurization of the propulsion system. Although there are many possible causes, the consensus seems to be that Mars Observer was lost as a result of a fatal failure of the propulsion system which, since it was borrowed from an Earth-orbiting satellite, was never designed to delay pressurization after months in space. The best guess is that a small amount of nitrogen tetroxide oxidizer leaked past a valve during the long cruise to Mars and came into contact with the hydrazine fuel during pressurization, causing a failure of the system’s plumbing. Mars Observer would have gone into an unrecoverable tumble as a result of venting propellants from broken pipes. This loss—which showed the weakness of the philosophy of adapting Earth-orbiting spacecraft to planetary missions—along with Mars Observer cost overruns essentially killed the Planetary Observer program and set the stage for the “faster, better, cheaper” Discovery program (which has had its own problems, as we will see).
A schematic of the Mars Observer propulsion system with the location of the probable nitrogen tetroxide leak indicated. Click on image to enlarge. (NASA/JPL)
NEAR – 1998
The American NEAR (Near Earth Asteroid Rendezvous) mission, which was the first of NASA’s Discovery class of small planetary science missions to be launched, experienced a propulsion system failure that nearly scuttled its mission to orbit the near Earth asteroid 433 Eros for one year. The 800-kilogram NEAR launched using a Delta II 7925 rocket on February 19, 1996 was later renamed NEAR Shoemaker after the late American planetary scientist Eugene Shoemaker (1928–1997). On June 27, 1997, NEAR flew 1,212 kilometers from 253 Mathilde, gathering vital new data on asteroids in the process. On July 3, NEAR performed a two-part deep space maneuver using its primary 450-newton engine to set it on course for a 540-kilometer flyby of the Earth on January 23, 1998 and subsequently on course to rendezvous with Eros on January 10, 1999.
The NEAR spacecraft being enclosed in its fairing prior to launch. (NASA/APL)
The rendezvous with Eros planned to use a series of four propulsive maneuvers over the course of three weeks. The first and largest one, on December 20, 1998, required the main engine to burn for 15 minutes to change NEAR’s velocity by 650 meters per second roughly matching the spacecraft’s orbit with that of Eros in the process. Three subsequent maneuvers would adjust the spacecraft’s approach trajectory and finally establish an orbit around the low-gravity target. However, NEAR’s main engine shut down a fraction of a second into its burn and communications with NEAR were lost as the spacecraft began to tumble. After controllers finally reestablished communications 27 hours later, they found that its thrusters had fired thousands of times as the spacecraft tried to regain control of its attitude, expending 29 kilograms of propellant in the process and reducing its propellant margins to zero. Investigators never definitively determined the cause of the problem, but software and operational errors apparently only made the problem worse.
The alternate trajectory flown by NASA’s NEAR mission to reach 433 Eros. Click on image to enlarge. (NASA)
Ground controllers immediately put a new plan into effect to salvage the mission. It was already too late to rendezvous with Eros this time around, so scientists had to content themselves for the moment with a quick reconnaissance as NEAR flew by Eros at a distance of 3,927 kilometers on December 23. Under the new plan, NEAR performed the major burn of its main engine on January 3, 1999, to roughly match the orbit of Eros. After another 13 months in solar orbit, NEAR finally completed its rendezvous with one final small maneuver on February 14, 2000. NEAR was able to successfully execute its one-year mission at Eros ending it with a successful touchdown on the asteroid’s surface (a mission-ending task for which NEAR was not originally designed) on February 12, 2001, where it remained until it was shut down 16 days later. Unlike most orbit insertion failures, NEAR was fortunate enough to recover and salvage its mission.
A view of 433 Eros acquired on February 28, 2000 by NEAR Shoemaker in orbit around the asteroid. (NASA/JPL/JHUAPL)
Mars Climate Orbiter – 1999
Probably the most infamous orbit insertion failures involved the loss of the Mars Climate Orbiter (MCO), whose mode of failure could only happen in America with its stubborn refusal to go metric like the rest of the world. MCO was one of NASA’s 1990s-era missions touted as being “faster, better, cheaper.” Unfortunately, one of the ways to make the mission cheaper was to cut back on testing and oversight of the contractors. The 638-kilogram MCO carried a pair of instruments to study the atmosphere of Mars from a circular 421-kilometer orbit including a copy of one that was originally carried by the ill-fated Mars Observer spacecraft lost in 1993. Initially, MCO would enter an elliptical orbit around Mars with a nominal periapsis of 210 kilometers and a period of about 15 hours. After entering orbit, MCO would use an aerobraking technique first employed by one of its predecessors, the Mars Global Surveyor launched in 1996, to gradually lower its orbit over the course of a couple of months. One final propulsive maneuver would then raise its periapsis out of the Martian upper atmosphere and circularize its orbit for science operations.
The Mars Climate Orbiter shown undergoing pre-launch testing. (NASA)
MCO was successfully launched using a Delta II 7425 rocket on December 11, 1998 and performed its first course correction ten days later. A second minor course correction took place on March 4, 1999. But by the time of the third course correction on July 25 just 60 days before arrival at Mars, the mission’s navigation team was beginning to suspect that something was wrong with the design and execution on the course correction maneuvers since their calculations consistently showed MCO tracking closer to Mars than their predictions. On September 19, a fourth course correction took place, but tracking showed that MCO would pass 173 kilometers from Mars instead of the nominal 210 kilometers. Since MCO could come as close as 85 kilometers of Mars and survive, controllers elected not to perform a fifth course correction.
But with only hours to go before orbit insertion on September 23, 1999, JPL navigators were reporting that MCO would pass only 110 kilometers from Mars. This would be survivable but MCO would need to raise its periapsis altitude during its first orbit. The full magnitude of the navigation error did not become apparent until it was too late. About five minutes into its 16-minute, 23-second orbit insertion burn, contact with MCO was lost as it passed behind Mars 39 seconds earlier than predicted. MCO was never heard from again. Apparently, the navigation error had continued to grow and MCO passed as close as 57 kilometers above the Martian surface, where it was destroyed.
A diagram showing the actual and planned approach trajectories of the doomed Mars Climate Orbiter. Click on image to enlarge. (NASA)
An investigation into the loss of MCO later revealed that the error was embarrassingly trivial. In order to predict accurately the path of MCO, JPL navigators needed to take into account all the forces acting on the spacecraft, including the small impulses of its attitude control thrusters. Instead of being supplied with a table of thruster impulses in the metric units of newton-seconds that the JPL navigation programs required, the spacecraft contractor, Lockheed Martin, had supplied a table in units of pound-seconds. As a result, the navigational corrections for the attitude thruster firings were too small by a factor of 4.5. This error, combined with a greater number of thruster firings than originally planned because of the asymmetric arrangement of solar panels and other appendages, caused the fatal navigation error. The loss of MCO and NASA’s Mars Polar Lander two months later from unrelated causes resulted in a major review of mission engineering practices for such smaller missions and a two-year delay of further missions to Mars.
Nozomi – 2003
The US and USSR were hardly the only countries to experience planetary mission failures. Japan’s 259-kilogram Nozomi (also known as Planet-B) was launched using the all-solid M-V rocket on July 4, 1998, with the intent of entering orbit around Mars on October 11, 1999 to study the Martian atmosphere and, with a periapsis as low as 150 kilometers, further investigate the magnetic patterns originally observed in the Martian crust by Mars Global Surveyor. Instead of being sent directly into a transfer orbit to Mars, engineers devised a complicated series of maneuvers in the Earth-Moon system over five and a half months that included a pair of lunar flybys to boost the payload sent to Mars.
Nozomi shown being prepared for launch. (ISAS)
The M-V launch vehicle successfully placed Nozomi into an extended 703 by 489,382-kilometer orbit around the Earth. Nozomi then completed a pair of lunar flybys on September 24 and December 18, 1998. The plan was to make one last pass 1,003 kilometers above the Pacific Ocean on December 20, coupled with a seven-minute burn of its bipropellant main engine, to finally send the probe on its way to Mars. A malfunctioning valve resulted in more fuel being used than planned and Nozomi gained insufficient velocity to reach Mars. To compound the problem, a pair of course correction burns the following day also used more propellant than planned and left Nozomi with insufficient propellant to complete its intended mission.
An image of the Earth and Moon acquired by Nozomi as it maneuvered through the Earth-Moon system before departing for Mars. (NASA/NSSDC)
Engineers developed an alternative plan to salvage the mission. Nozomi would stay in solar orbit for an extra four years and make a pair of flybys of the Earth in December 2002 and June 2003 to permit the probe to approach Mars at a significantly slower velocity in December 2003 allowing it to enter orbit with its remaining propellant. Unfortunately, the backup plan began to unravel on April 21, 2002, when powerful solar flares damaged Nozomi’s communications and power systems. While technicians devised workarounds to maintain contact with the craft, an electrical short in the system used to control the temperature of its hydrazine propellant meant that its vital fuel supply would freeze solid when the temperature fell below 2° C, as would happen when the spacecraft travelled beyond Earth’s orbit.
Nozomi’s alternate trajectory to reach Mars. (JAXA)
While the hydrazine thawed out in time for the first Earth flyby at a distance of 29,510 kilometers on December 21, 2002, and the second flyby of 11,023 kilometers on June 19, 2003, it froze solid on its way to Mars, preventing the propulsion system from working. These problems were compounded when all contact with Nozomi was lost on July 8, 2003 and never regained as a result of the damaged communication system. Nozomi silently flew an estimated 894 kilometers above the surface of Mars on December 10, 2003 and continued into a two-year solar orbit.
Akatsuki – 2010
The Nozomi mission to Mars was not the only Japanese mission to encounter problems at orbit insertion. Akatsuki, also known as Planet-C or the Venus Climate Orbiter, is a Japanese mission originally meant to study the atmosphere of Venus for two years from a 300 by 80,000-kilometer orbit with a period of 30 hours using a suite of a half dozen instruments to observe Venus at wavelengths from the infrared to the ultraviolet. Launched on May 20, 2010 using Japan’s H-IIA rocket, the 518-kilogram orbiter reached its target on December 7. The plan was for Akatsuki to fire its 500-newton bipropellant main engine for 12 minutes to enter an initial four-day orbit with a periapsis of 550 kilometers and an apoapsis of between 180,000 and 200,000 kilometers. Subsequent maneuvers would eventually place the spacecraft into the desired 30-hour orbit for its primary mission.
The launch of Akatsuki on May 20, 2010. (JAXA)
Akatsuki’s engine ignited on schedule and flew behind Venus as planned resulting in an expected loss of signal. Unfortunately, when it reemerged from behind Venus and communications were reestablished, the probe was found to be in a safe mode. Because of the slow communication rate through the spacecraft’s low gain antenna, it was not until the following day that controllers found the cause of the safe mode and confirmed that Akatsuki had failed to enter orbit around Venus. A subsequent investigation showed that Akatsuki’s main engine had only fired for about three minutes before shutting down. It is believed that deposits formed in a valve between the helium pressurization and the hydrazine fuel tanks, decreasing the fuel flow to the main engine in the process. The high temperatures resulting from the combustion of the oxidizer-rich mixture damaged the engine, forcing an early shutdown.
A photo montage showing Akatsuki’s ceramic engine and nozzle above, and the intense heating (lower left) the throat sees during operation especially when fuel-starved. Excess temperatures caused the ceramic to crack typically at the throat region (lower right). (JAXA)
After test firings of the main engine on September 7 and 14, 2011 showed that it was too damaged to be used again, engineers developed a backup plan to use the spacecraft’s smaller monopropellant thrusters (which share the hydrazine supply of the main engine) to enter orbit when the spacecraft once again reaches the vicinity of Venus in November 2015, a close approach made possible by a series of course corrections performed in November 2011. Unfortunately, because of the lower efficiency of its monopropellant thrusters and the fuel used to alter its course, Akatsuki would only be able to enter an orbit with an apoapsis of between 300,000 and 400,000 kilometers and an orbital period of 9 days compromising its science mission as a result. Another major issue is that this five-year delay in reaching Venus exceeds the original 4½-year design life of the spacecraft, which could be further reduced by the higher-than-expected temperatures Akatsuki experienced as a result of its unplanned stay in a 203-day solar orbit largely inside that of Venus.
An artist’s rendering of Akatsuki entering orbit around Venus using its monopropellant thrusters. (ISAS/JAXA)
Despite the myriad of issues, Akatsuki survived its lengthened transit to Venus and fired its monopropellant thrusters for 20 minutes on December 7, 2015 to enter an initial 400 by 440,000 kilometer orbit around Venus with a period of 13.6 days. A trim maneuver executed on March 26, 2016 lowered the apoapsis to 330,000 kilometers and shortened the orbital period to 9 days. Akatsuki finished its primary mission in April 2018 and is currently in its extended operation phase to become a rare example of a mission recovering from an orbit insertion failure.
A UV view of Venus created by Damia Bouic using Akatsuki data. Click on image to enlarge. (JAXA/ISAS /DARTS/Damia Bouic)
Follow Drew Ex Machina on Facebook.
Related Reading
“The Mars Orbiter That Almost Was Not”, Drew Ex Machina, May 22, 2014 [Post]
General References
Brian Harvey, Russian Planetary Exploration: History, Development, Legacy and Prospects, Springer-Praxis, 2007
Wesley Huntress, Jr. and Mikhail Ya. Marov, Soviet Robots in the Solar System: Mission Technologies and Discoveries, Springer-Praxis, 2011
Paolo Ulivi with David M. Harland, Robotic Exploration of the Solar System Part 2: Hiatus and Renewal 1983–1996, Springer-Praxis, 2009
Paolo Ulivi with David M. Harland, Robotic Exploration of the Solar System Part 3: Wow and Woes 1997-2003, Springer-Praxis, 2012
Paolo Ulivi with David M. Harland, Robotic Exploration of the Solar System Part 4: The Modern Era 2004-2013, Springer-Praxis, 2015