Mars 4, 5, 6 and 7

Earth and Mars line up for a transfer every 26 months, but the line-ups are not equal, and 1973 was a bad one. Reaching Mars that year needed a departure 250 to 300 metres per second faster than in 1971, and a Proton-K could not lift an orbiter and a lander together on that trajectory. So Soviet planners split the standard pairing in two. Mars 4 and Mars 5, of the M-73S type, would fly alone and brake into orbit around the planet. Mars 6 and Mars 7, of the M-73P type, would each carry a descent module, drop it on the way past, and let the orbiters relay its signals home. Each type was flown twice for redundancy, which is why four spacecraft left Baikonur in 19 days: Mars 4 on 21 July 1973 at 19:30:59 UT, Mars 5 on 25 July at 18:55:48 UT, Mars 6 on 5 August at 17:45:48 UT, Mars 7 on 9 August at 17:00:17 UT. Soviet paperwork records those moments in Moscow time, three hours later. Lavochkin's own history calls it the first occasion on which four Soviet spacecraft took part in a single interplanetary expedition. The pressure was not only celestial. NASA was building the Viking landers for the following window, and the Soviet Union badly wanted a working machine on the Martian surface first.

The flotilla was compromised before it left the ground, and the people who launched it knew. Ground testing found that one component, the 2T312 transistor made at the Voronezh semiconductor plant, was degrading. Academician Mikhail Marov, who worked on these missions and described the episode to RIA Novosti in 2011, said a cost-saving change had replaced the transistors' gold leads with aluminium, and that the aluminium oxidised after about six months. Asif Siddiqi's NASA chronology tells it differently: testing less than four months before launch showed failure rates climbing after 1.5 to 2 years of operation, which was roughly when the spacecraft would reach Mars. The accounts disagree about the clock and agree about the consequence. The transistors were scattered through the onboard electronics, replacing them would have taken about six months, and the window would not wait. Marov remembered the argument being settled at a meeting chaired by Mstislav Keldysh with Lavochkin representatives present, with the decision pushed down by the Central Committee and the Council of Ministers: fly them anyway, na avos, a Russian phrase meaning on the off-chance. Siddiqi records that roughly even odds of success were accepted. Marov's verdict decades later was blunt. They saved kopecks on precious metal, he said, and destroyed the colossal sums spent on the project.

Mars 4 arrived first and could not stop. Two of the three channels of its onboard computer had already failed, the second midcourse correction could not be loaded, and the braking burn never happened. Sources differ on whether the engine failed or the command was simply never given, but the computer was the cause either way. On 10 February 1974 the spacecraft swept past at 15:34 UT and 1,844 kilometres, and the ground team salvaged what it could: the cameras came on at 15:32:41 UT for about six minutes, taking a 12-frame photographic cycle from 1,900 to 2,100 kilometres out, plus two scanned panoramas begun 27 minutes before closest approach. Passing behind the planet afterwards gave a dual-frequency radio occultation, which produced the first detection of Mars's night-side ionosphere. Mars 5 did better, briefly. Its engine fired at 15:44:25 UT on 12 February 1974 and put it into an orbit of roughly 1,760 by 32,586 kilometres, inclined about 35 degrees, taking nearly 25 hours to circle. Almost immediately, controllers saw the pressurised instrument compartment losing air, most likely punctured by a particle during or just after the burn. The estimate was three weeks. The science team tore up its schedule and ran everything at once: five imaging sessions between 17 and 26 February, gamma spectra of the surface, water vapour, ozone, infrared radiometry. The last session was on 28 February 1974, about two weeks after arrival.

Mars 6 had been half-deaf since 3 September 1973, when its telemetry system failed and controllers lost the ability to see what the spacecraft was doing. For the rest of the cruise they sent every command two or three times, blind, and inferred success from side effects. The machine did its job regardless. On 12 March 1974 it released the descent module at a range of 48,000 kilometres; the module's own retro fired about 15 minutes later to bend it onto an impacting path; and at 09:05:53 UT it met the atmosphere at 5.6 kilometres per second, 11.7 degrees below horizontal. The heat shield took the worst of it, and at 09:08:32 UT, at about 600 metres per second and roughly 20 kilometres up, the parachute opened. From then until 09:11:05 UT the module measured pressure, temperature and deceleration and radioed the numbers to the bus overhead for immediate relay to Earth. NSSDCA counts 224 seconds of data; Lavochkin's own account says 150 seconds on the parachute leg. Both describe the same thing: the first measurements ever made inside the Martian atmosphere and sent home. Much of what arrived was unusable. NSSDCA says a large part of the data was unreadable because of the degraded computer chip, and Lavochkin records that useful information could be pulled only from the descent module's own radio set, the second link being too weak to decode. The last telemetry confirmed that the soft-landing engine had been commanded to fire. The next signal was expected 143 seconds later. It never came, and the Soviet investigation never settled between a crash, a radio failure, and the module swinging too violently under its parachute in a dust storm.

Mars 7 had already failed, three days earlier. The Russian record blames incorrectly computed settings for the turn the spacecraft made before releasing its descent module; Siddiqi's account says the module's main retro-rocket never fired. Either way it sailed past Mars, missing by 1,300 kilometres in the American telling and 1,400 in Lavochkin's, and went into orbit around the Sun, transmitting for a while to a bus that could do nothing with it. Contact with the bus held until 25 March 1974. Lavochkin's own summary of the expedition reads like a ledger of loss: Mars 4, programme not fulfilled; Mars 5, not fulfilled in full; Mars 6, partially fulfilled; Mars 7, not fulfilled. Yet real science survived. Mars 6 gave the first vertical profile of the Martian troposphere and a surface pressure near 6 millibars, and its numbers underpinned the first Soviet model of the atmosphere. Mars 5's photographs, some of them comparable to Mariner 9's, showed channels cut by running water. Most of the harvest filled a special issue of Kosmicheskie Issledovaniya in 1975, although the index to that issue notes that Surkov's gamma spectrometry and Krupenio's 3 cm radio measurements were left out of it. The descent module itself, meanwhile, is still missing: HiRISE went looking in December 2011 and found bedrock, and a candidate crater proposed by Russian enthusiasts in 2018 has never been confirmed. The flotilla also settled the Soviet Mars programme's fate. Two years later NASA's Viking 1 and Viking 2 landed and worked for years, and the USSR did not send another spacecraft to Mars until Phobos 1 and 2 in 1988. Mars 6's descent was the last time a Soviet spacecraft entered the Martian atmosphere at all: Mars 2 crashed in 1971, Mars 3 made the first confirmed soft landing on Mars later that year and fell silent within seconds, and Mars 6 went quiet with the ground in sight.