While I have had a long-time interest in lunar and planetary exploration both as a scientist and a historian, for the past quarter of a century I have had a particular interest in the enigmatic mission of Zond 2 launched towards Mars in November of 1964. Before the fall of the Soviet Union, it was generally believed to have been a photographic flyby mission in competition with NASA’s Mariner 4 spacecraft (see “Mariner 4 to Mars“). But in the early-1990s I was able to provide new evidence (which culminated in a paper published in The Journal of the British Interplanetary Society) to support a view that it was intended to land on Mars instead. Unfortunately not long after the publication of my paper in 1993, new evidence from the former Soviet Union began to appear that eventually proved that Zond 2 did not carry a lander but was equipped with a state-of-the-art camera package.

Even though the mystery behind its odd choice of trajectory remained, the conventional wisdom has been for much of the past two decades that Zond 2 was intended to be a photographic flyby mission. But now new documents from old Soviet archives revealed a couple of years ago once again turns what we thought we knew about the Zond 2 mission on its head resolving some mysteries but raising new questions in the process. In this essay, I review what is currently known about the early history of the 3MV series of spacecraft which included Zond 2, provide historical context of what was known about Mars in the months leading up to the 1964 Mars launch window and analyze the 1964 Mars trajectory options to provide the latest view of the mission of Zond 2.


Origin of the 3MV Series

Like its predecessors, the 3MV series of planetary explorers were designed at OKB-1 led by famed Chief Designer Sergei Korolev. Externally, the 3MV series of spacecraft flown during the 1964 launch windows to Venus and Mars resembled the earlier 2MV series of spacecraft launched in 1962. The 3.6-meter tall 3MV consisted of two sections.  The first was an orbital compartment which contained control systems, power supplies, communications gear as well as some instrument electronics. On the the exterior of this compartment were mounted a pair of solar panels with hemispherical radiators for thermal control mounted on their ends, an umbrella-like high gain antenna as well as low gain antennas and a variety of sensors.  A number of design improvements had been made to the 3MV based on the experiences with the earlier 2MV design such as significantly more redundancy of key systems and the inclusion of a set of experimental plasma engines to serve as a backup to the conventional nitrogen gas jets used for attitude control which had failed during the flight of Mars 1.

The second section of the 3MV was the planetary compartment which was geared towards specific investigations of the target planet. As before, the planetary compartment came in two varieties: one contained a film-based camera system and a set of ultraviolet and infrared instruments designed to study the target planet during a close flyby while the other was a roughly spherical lander with a diameter of about 0.9 meters designed to detach from the orbital compartment before the encounter and touchdown on the target planet. Both the orbital and planetary compartments were pressurized to provide a laboratory-like environment for the internal equipment in order to simplify the design and testing of various systems as well as provide easier thermal control.


Russian diagram of the 3MV-1 spacecraft designed to deploy a lander on Venus in 1964. (RKK Energia)

As with the 2MV, the 3MV design had four design variants. The 3MV-1 carried a planetary compartment designed to land on Venus while the 3MV-2 carried instruments designed to study Venus during a close flyby. Because the launch window to Mars in November 1964 was so much more favorable than the earlier launch window used by Mars 1 in November 1962, the Mars-bound 3MV spacecraft would be significantly more massive than their earlier 2MV counterparts. The 3MV-3, designed to land on Mars, would have a launch mass of 1,042 kilograms while the 3MV-4 flyby craft would weigh in at 1,037 kilograms. These craft would carry a significantly more massive instrument payload as a result. The 3MV-3 and 3MV-4 would be four times more massive and much more capable that the pair of 261-kilogram American Mariner-Mars flyby spacecraft that were also under development at this time.

Korolev understood that there was a need to flight test the new 3MV design in order to iron out the problems that would inevitably crop up. To this end, Korolev envisioned flying two or three dedicated 3MV engineering missions to be launched starting in the summer of 1963. It was hoped that any problems uncovered during these test missions could be resolved in time for the 1964 planetary missions thus increasing their chances for success. Since these test missions would not be directed specifically towards Venus or Mars, after launch they would receive the generic name of “Zond” which means “probe” in Russian.

The first Zond version was a stripped down model of the Venus lander craft designated 3MV-1A. Originally envisioned to have a launch mass of 800 kilograms, this spacecraft would carry little scientific instrumentation and a light-weight 275-kilogram entry probe.   The 8K78 launch vehicle (better known as the Molniya after the communication satellite series that made extensive use of this rocket starting in 1964) would be used to launch the 3MV-1A into a nearly circular solar orbit inclined about five degrees to ecliptic. This trajectory was designed to bring the Zond back to Earth after a flight of five to six months at which point the entry probe would separate from the orbital compartment and return to Earth. This reentry into Earth’s atmosphere at a speed in excess of 11.5 kilometers per second would simulate the conditions of an entry into the Venusian atmosphere and provide a realistic test of the 3MV-1 design for the upcoming Venus missions.

The second Zond variant was a modified version of the Mars flyby spacecraft designated 3MV-4A. Originally envisioned to have a launch mass of 996 kilograms, this spacecraft was to carry a planetary compartment equipped with an updated miniaturized film-based imaging system and other scientific instruments. The 3MV-4A would use an 8K78 to send it on a simulated trajectory towards the orbit of Mars. At a distance of 40,000 to 200,000 kilometers, the 3MV-4A would turn its camera back towards the receding Earth and acquire a sequence of photographs that would subsequently be developed automatically on board. The spacecraft would then transmit its scanned photographs and other data gathered on the interplanetary environment out to distances as great as 200 to 300 million kilometers as part of a long distance communications test. The 3MV-4A mission, if successful, would provide the first images of the Earth taken from deep space and give the Soviets another long distance communication record in addition to a much needed engineering test of the 3MV-4 design.

On March 21, 1963 the Soviet government officially approved the 3MV program. It would consist of one or two 3MV-1A flights and a single 3MV-4A flight to be launched in 1963 as well as a total of six operational 3MV spacecraft to be launched to Venus and Mars in 1964. As had happened all too frequently in the past with other projects, however, design and construction of the new 3MV took longer than expected. But by July of 1963, Korolev had set the launch schedule for the Zond tests: A 3MV-1A would be launched sometime between September 1 and October 15 (with the optimum launch date being October 12) on an Earth-return mission. The 3MV-4A Zond mission was scheduled for launch in March 1964 with sufficient time to make any needed changes to the Mars-bound 3MV design to be launched eight months later.

Further delays in the launch of the first Zond test flight ensued but finally on November 11, 1963 an 8K78 lifted off from the Baikonur Cosmodrome in Soviet Kazakhstan carrying 3MV-1A No. 2 on its Earth-return mission. Unfortunately, yet another malfunction of the Blok L escape stage stranded the craft in its low Earth parking orbit and it was designated Kosmos 21 – the first of many failed interplanetary spacecraft to be (rather badly) disguised with the “Kosmos” moniker.

With the launch failure of the first Zond engineering test mission and the continued unsatisfactory condition of the 3MV spacecraft under final assembly and test, by the end of December 1963 a second 3MV-1A mission was officially approved for launch in January 1964 on the first improved 8K78M. The 3MV-4A test flight would follow in the April-May time frame after the launch of the Venus missions during their late-March to early-April window. At some point, possibly in response to further delays in the preparation of the second 3MV-1A test craft, the decision was made to abandon the Earth-return option and launch it on a long Type II trajectory towards Venus during the beginning of the launch window in mid-February (instead of the faster Type I trajectories to be used by the 3MV-1 spacecraft starting in late-March). On February 19, 1964 the first of the improved 8K78M rockets lifted off carrying 3MV-1A No. 4A on a test flight of the Venus probe design. But a failure of the Blok I third stage prematurely ended this last chance at a Zond test flight before the upcoming Venera missions.

Without the benefit of a test flight, and with the Venus mission now scaled back to just a pair of 3MV-1 landers, 3MV-1 No. 5 was launched on March 27, 1964 and successfully made it into its parking orbit. But during the unpowered coast, attitude control was lost and the Blok L never ignited its engine. Stranded in Earth orbit, the rocket and its payload were designated Kosmos 27. Despite the failure, one of the many improvements made to the new Blok L escape stage of the 8K78M paid off this time. Data gathered by the new telemetry system were recorded and radioed back to ground controllers on the next orbit allowing engineers to diagnose the failure like never before. This failure (and probably a couple of earlier Blok L failures) were found to be the result of a fault in the design of the wiring of a key control system. Fortunately the fix involved only 20 minutes of a technician’s time with a soldering iron to resolve for the next launch attempt.

With its wiring fault fixed just in time for the end of the Venus launch window, another 8K78M lifted off on April 2, 1964 carrying 3MV-1 No. 4. This time all four stages of the 8K78M worked sending what was to be called “Venera 2” on its way to Venus and the first attempted landing on another planet. But before the triumphant launch announcement could be made, a major problem was discovered during the first communication session with the probe. The pressurized orbital compartment was leaking and all its gas would be lost within a week severely compromising the ability of its equipment to operate. With bleak prospects for success, the Soviets announced the probe simply as Zond 1 making no mention of its mission to Venus.

The leak in the orbital compartment of Zond 1 was eventually found to be caused by a bad weld near the quartz window for the probe’s star and Sun sensors. While it would not help Zond 1, future 3MV craft would have their welds X-rayed as a new quality control check. In the mean time as ground controllers made valiant attempts to keep Zond 1 alive and made excellent use of the various redundancies built into the new 3MV design, the craft finally succumbed to its growing list of problems and fell silent on May 24, 1964 some two months shy of its encounter with Venus.


Change of Plans

With the ongoing quality control issues in the 3MV spacecraft revealed by the failure of Zond 1, the engineering flight of the 3MV-4A to test the Mars flyby spacecraft in April or May was cancelled. But aside from a government resolution dated August 3, 1964 calling for the continuation of the 3MV series into 1966 (which would include not only the Mars launch window in the fall of 1964 but the next Venus launch window a year later), the exact details of what transpired after the failure of Zond 1 have yet to be revealed.

However, it is clear that the plans for the Soviet’s 1964 Mars missions had been changed dramatically after the spring of 1964. Instead of the pair each of 3MV-3 landers and 3MV-4 flyby spacecraft that were planned at one point, now plans called for the launch of as many as three (according to some sources) modified versions of the 3MV-4A test craft. Part of the change certainly must have been due to the fact that the development of the 3MV-4A was much further along and that these craft were more likely to be prepared in time for the upcoming Mars launch window. I would speculate further that part of the decision to cancel the launch of the lander portion of the 1964 mission might have been motivated by new revelations about the nature of the Martian atmosphere during the year leading up to the 1964 launch window.


One of the best pre-Space Age images of Mars acquired from the Mt. Wilson Observatory in 1956. (Mt. Wilson)

When design work on the first Mars landers started at OKB-1 in 1960, it was widely believed that the Martian atmosphere was composed primarily of nitrogen with about 10% carbon dioxide. As a result of numerous analyses of various photometric and polarimetric measurements made during the 1920s to 1950s, the overwhelming consensus of the astronomical community was that Mars had a surface pressure of about 85 millibar.

Based on computer simulations I performed back in 1990, I had determined that a spherical entry probe with a diameter of about one meter (about the same shape and size as contemporary Soviet Venus landers) and a mass of less than about 350 kilograms could safely land on the Martian surface with such an atmosphere via typical Mars approach trajectories used at the time. This mass limit compares well with the stated 250-kilogram mass (or the 305-kilogram mass given in another source) of the 2MV-3 Mars lander unsuccessfully launched during the same window as Mars 1 in 1962. According to Vladimir G. Perminov, a leading designer of Mars and Venus spacecraft at NPO Lavochkin (which officially took over this task from OKB-1 in April 1965), it seems that Soviet designers were even more optimistic about the density of the Martian atmosphere than the astronomical community generally was and they were apparently assuming surface pressures in the 100 to 300 millibar range! But like the properties of the atmospheres of all the bodies in the solar system, the exact figures for Mars were still quite uncertain during this time.

Of course today we now know that the Martian atmosphere is composed primarily of carbon dioxide with a mean surface pressure of 6 millibars – less than a tenth as dense as had been generally assumed at the beginning of the Space Age. As a result, the original Soviet Mars lander design was simply inadequate in such a thin atmosphere. The realization in the scientific community that the Martian atmosphere was much thinner than expected started to become known at about the same time as the 3MV-3 landers would have been in the process of being developed and manufactured.

The first indications of trouble actually came from Soviet astronomer Vassili I. Moroz at the Sternberg State Astronomical Institute in Moscow. A pioneer in infrared spectral studies of bodies in the Solar System, his analysis of the IR spectra he had obtained of Mars during its opposition in early 1963 showed that the surface pressure of Mars was much less than 100 millibars and was likely only about 24 millibars. His analysis was submitted for publication in the Russian-language Astronomicheskii Zhurnal on September 21, 1963 and surely came to the attention of the engineers at OKB-1 who were probably a bit skeptical about the unusually low surface pressure results.


A sample IR spectrum of Mars acquired on February 20, 1963 by Soviet astronomer Vassili Moroz. Key carbon dioxide lines are indicated in this spectrum that runs from wavelengths of 1.1 (right) to 1.8 (left) microns.  Analysis of IR spectra like these indicated that the Martian atmosphere was much thinner than had been previously estimated using other techniques.  (Astronomicheskii Zhurnal)

But a couple of months before Moroz’s paper was formally published in March-April 1964, another analysis of IR spectra of Mars by three American astronomers was published. On January 1, 1964 Lewis D. Kaplan, Guido Munch and Hyron Spinard of JPL, the Mt. Wilson and Palomar Observatories published an analysis of IR spectra of Mars obtained by Gerard P. Kuiper of the Lunar and Planetary Laboratory during April 1963. Their analysis showed that the Martian surface pressure was 25±15 millibars. Again, this was much less than had been generally assumed by the astronomical community up until that time but it agreed well with Moroz’s findings. Kuiper, working with then-graduate student Tobias C. Owen (who today is with the University of Hawaii’s Institute for Astronomy), finally published their own analysis of their IR spectra of Mars on August 24, 1964 which showed that the surface pressure on Mars was only 17±3 millibars.

The uncomfortably large disagreement of the newer IR spectral studies with the earlier photometric and polarimetric-based estimates for the Martian surface pressure began to be resolved with the publication of a paper in August 1964 by American astronomer, Steven Musman. Part of the disagreement with the earlier photometric and polarimetric measurements of the Martian atmospheric pressure had to do with the assumed optical properties of the Martian surface. Musman used better estimates for the albedo of the Martian surface at a wavelength of 330 nm (where Rayleigh scattering in the atmosphere is quite strong and the Martian surface quite dark) to estimate that the Martian surface pressure was 19 millibars if the atmosphere were dominated by carbon dioxide and 27 millibars for one dominated by nitrogen. These figures were more in line with the estimates based on IR spectral results and hinted that some of the assumptions made in the past analyses of photometric and polarimetric measurements that led to the higher surface pressure estimates were wrong (in fact, much later it was realized that scattering from the large amounts of fine dust in the Martian atmosphere significantly skewed these earlier estimates).

While future results using better analytical models and especially the results from the radio occultation experiment performed by Mariner 4 in July 1965 would continue to drive down the estimates of Martian surface pressure towards the value accepted today, it is clear that by the fall of 1964 that the earlier accepted value was seriously in doubt and possibly too high by a factor of four or possibly more. These fresh doubts about the properties of the Martian atmosphere were certainly being reflected in American Mars mission planning during the last half of 1964. Results from the NASA-sponsored “1964 Summer Study” at Stanford University on future Mars missions showed that the participants were being influenced by the latest reports with their best estimate of the Martian atmospheric surface pressure being in the 10 to 80 millibar range – much lower than had been considered earlier. Proposed standard models to be employed in future studies of NASA’s Mars missions were in the 10 to 40 millibar range by November 1964 compared to values in the 54 to 136 millibar range that were typically used just a year earlier.

Just as in the US, it seems likely that these new data were also available to Korolev and his engineers at OKB-1 long before the opening 1964 Mars launch window. The likelihood that the Martian atmosphere was much thinner than had been believed would have surely cast grave doubts about the assumptions used in the design of the 3MV-3 Mars lander and their chances for success. It still remains to be seen what role, if any, these revelations played in the decision on the mix of spacecraft to be launched towards Mars in the fall of 1964.


Russian diagram of Zond 2. (RKK Energia)

In the end, perhaps as many as three camera-bearing 3MV-4A spacecraft with a mass of 950 kilograms each were prepared for launch during the 1964 Mars window, according to most sources. After the experience with the earlier Soviet Venus and Mars probes, there were apparently no illusions about the chances of these spacecraft actually surviving the long voyage all the way to Mars. From the start they were to be billed as engineering test flights and would receive a “Zond” designation with the hope that they would still be functional once they reached their target. In the end, only Zond 2 was launched on November 30, 1964 with an intended encounter date of August 6, 1965. And leading the way to Mars was the sole survivor of the American Mariner-Mars 1964 project, Mariner 4, launched two days earlier with an encounter date of July 15.


3MV-4A Trajectory Analysis

Figure 1, taken directly from Clark et al., shows a plot of the minimum C3 launch energy as a function of launch date during the 1964 Mars window with both Type I and II trajectories shown (with the two types defined as transfer trajectories where the spacecraft travels through less than 180° and more than 180° around the Sun before encountering Mars, respectively). Even though for a given value of C3 the launch window for Type II trajectories would be significantly longer than for Type I, the times of flight for the latter were much shorter and would have been preferred a half a century ago when the longevity of spacecraft was an important consideration.


Figure 1: Plot of minimum C3 launch energy as a function of launch date for Type I and Type II trajectories (taken from Clark et al). The three red “+” show the conditions for the following trajectories: 1) Zond 2, 2) a hypothetical 2nd 3MV-4A launch and 3) a hypothetical 3rd 3MV-4A launch. Click on image to enlarge. (JPL/NASA)

If the C3 values in the 12.4 to 12.5 km2/s2 range used by Kosmos 27 and Zond 1 earlier in 1964 (which had the same 950-kilogram launch mass as Zond 2) are representative of the performance limit of the 8K78M rocket also used by Zond 2, then the launch window to Mars using Type I trajectories would have extended from about November 10 to December 9, 1964.


Figure 2: This plot, taken from Clark et al., shows contours of C3 launch energy plotted as a function of launch date (X-axis) and time of flight (Y-axis). The three red “+” show the conditions for the following trajectories: 1) Zond 2, 2) a hypothetical 2nd 3MV-4A launch and 3) a hypothetical 3rd 3MV-4A launch. The blue “+” indicates the trajectory of the following: 4) Mariner 4. Click on image to enlarge. (JPL/NASA)

Figure 2, also taken directly from Clark et al., shows a contour plot of C3 launch energy as a function of launch date and time of flight. As can be seen in this plot, higher energy Class I transfer orbits (where the encounter takes place before the spacecraft reaches the aphelion of its transfer orbit) offered significantly shorter times of flight compared to Class II transfer orbits (where the encounter takes place after the spacecraft reaches the aphelion of its transfer orbit). This was the tactic used by NASA in its choice of a Class I trajectory for Mariner 4.


Trajectory followed by NASA’s Mariner 4 to Mars. Click on Image to enlarge. (NASA)

The trajectory characteristics for Zond 2, with its known launch and encounter dates, are summarized below in Table 1 based on the information from Clark et al. along with the characteristics for two hypothetical launch dates for the pair of 3MV-4A spacecraft that some believe were originally meant to accompany it. The characteristics for these hypothetical launches were based on two sets of assumptions: First, it was assumed that the launches of the 8K78M were spaced three days apart which represents the theoretical minimum spacing for such launches during this era (a spacing that was seldom achieved in practice because of a variety of factors). Second, it was assumed that the encounters with Mars were spaced two days apart starting with the August 6 encounter date of Zond 2. This preferred spacing (which was used for the 1962 Mars missions and again for the 1965 Venus missions, for example) was used in early Soviet planetary missions probably for logistical reasons.


Table 1: Summary of 3MV-4A Trajectories to Mars

Zond 2

Hypothetical 2nd Launch

Hypothetical 3rd Launch

Launch Date & Time

Nov 30, 1964

13:12 UT

Dec 3, 1964

Dec 6, 1964

Arrival Date

Aug 6, 1965

Aug 8, 1965

Aug 10, 1965

Time of Flight (days)




Launch Energy, C3 (km2/s2)




Trajectory Type/Class




Asymptotic Speed WRT Mars (km/s)




Entry Speed (km/s)




Distance to Mars at Encounter (million km)





What is immediately apparent when looking at the characteristics of the transfer trajectory used by Zond 2 as plotted in Figure 2  is that it purposely followed a longer Class II trajectory unlike Mariner 4 which followed a faster Class I trajectory. Zond 2 could have been launched on a Class I trajectory with the same C3 launch energy of 11.7 km2/s2 and would have reached Mars around June 21, 1964 with a time of flight of about 203 days. This time of flight is 46 days shorter than the trajectory actually flown and would have allowed Zond 2 to beat Mariner 4 to Mars by over three weeks! Had Zond 2 been launched with a higher C3 of about 12.5 km2/s2 (the same C3 employed by Kosmos 27 and Zond 1 which had the same mass as Zond 2) on a Class I trajectory, an additional ten days would have been saved on the travel time to Mars allowing Zond 2 to beat Mariner 4 to Mars by a full month despite having been launched two days later.


Figure 3: This plot (taken from Clark et al.) shows contours of launch energy plotted as a function of launch date (X-axis) and asymptotic velocity with respect to Venus (Y-axis). The three red “+” show the conditions for the following: 1) Zond 2, 2) a hypothetical 2nd 3MV-4A launch and 3) a hypothetical 3rd 3MV-4A launch. Click on image to enlarge. (JPL/NASA)

So what was the motivation for the odd choice of a purposely long trajectory? Figure 3, taken directly from Clark et al., shows a contour plot of C3 launch energy as a function of launch date and asymptotic approach velocity with respect to Mars. As can be seen, the particular trajectory choice for Zond 2 minimizes the approach speed compared to the faster Class I trajectories. The pair of hypothetical launches would have followed similar trajectories that would also minimize their approach speed. This is exactly the tactic that would be employed by a Mars landing mission which led myself and others to conclude that Zond 2 was intended to land on Mars.

However, we now know that Zond 2 did not carry a lander but a camera package instead. The only conclusion is that Zond 2 was purposely placed on this trajectory in order to simulate the flight of a future Mars landing mission. And with a 249-day time of flight, which was longer than that attempted in any previous Soviet or American planetary mission (save for the failed Mariner 3 with its intended 254-day flight time to Mars – see “50 Years Ago Today: The Launch of Mariner 3“), it must have also been meant to be an endurance test of the 3MV design. These facts support the view that the primary objective of the Zond 2 mission was to be an engineering test flight with the encounter with Mars being only a secondary objective. If the primary objective had been to return data from Mars (preferably before the Mariner 4 encounter to score another Soviet space first), Zond 2 would have been launched on a faster Class I trajectory that would have a time of flight up to 8 weeks shorter and reach Mars up to a month before NASA’s Mariner 4.

Looking at Figure 3, it is also apparent that the minimum approach speed with respect to Mars decreases with launch date. If the objective was simulating the trajectory of a Mars landing mission, the latter part of the Mars launch window would have been preferred for Zond 2 and its sister craft. However, it appears that the launch of Zond 2 may have been delayed for any one of a number of reasons and occurred very late in the available launch window. With a launch on November 30, there would have been only about a week left for the pair of sister craft to be launched into trajectories that minimized approach speed. And launching a 3MV-4A into any trajectory to Mars would have been out of the question after about December 9 due to the quickly increasing launch energy requirements and the performance limits of the 8K78M.

If the preparations and launches went off without a hitch with a high cadence rate of a launch every three days, a total of three 3MV-4A spacecraft could have just made it during the time left in the 1964 Mars launch window. But looking at the history of 8K78 and 8K78M launches during periods of high cadence launch rates from 1962 to 1965, there was maybe about a 15% chance of actually launching all three spacecraft in the time left but about an 88% chance of launching Zond 2 and just one more 3MV-4A. This suggests that launch of Zond 2 might have been delayed or that there never was any intention of launching all three 3MV-4A spacecraft that had been allegedly prepared.


The Flight of Zond 2

As final preparations were being made to launch Zond 2 towards Mars, a recently revealed letter dated November 28, 1964 from Leonid Smirnov (who was the chairman of the Military Industrial Commission or VPK which oversaw the Soviet space effort) was sent to the Central Committee which gave details about the upcoming mission. According to this letter, a single spacecraft was to be launched to test elements of the 3MV probes and perform scientific studies of outer space. In addition, the letter clearly stated that if all systems functioned normally, an attempt would be made to alter the probe’s path with the goal of impacting Mars and delivering commemorative pennants on the surface.

With this new revelation, the choice of trajectory for Zond 2 begins to make sense as does its late launch. Zond 2 was an engineering test flight that would follow a simulated trajectory for a Mars landing mission to provide an endurance test for the 3MV design. This was basically the original 3MV-4A test flight plan which would end either with the malfunction of the probe or, this time, with a Mars impact. Since the spacecraft would have been destroyed upon reaching Mars, Zond 2 was never meant to photograph the Red Planet even though its advanced film-based camera system could have returned about an order of magnitude more imaging data than NASA’s Mariner 4. Instead it would have probably tested its new camera system by taking photographs of the receding Earth just as had been originally planned for the 3MV-4A test mission with the images being developed and then repeatedly transmitted back to the ground at ever increasing distances. The only data returned from Mars would have been magnetic field and radiation measurements made during the final hours before impact. The other two spacecraft that some sources claimed were also prepared for launch in the November-December 1964 launch window, if they actually existed, were likely just backups for Zond 2.


If Zond 2 had the objective to photograph the Martian surface, it could have acquired about 25 images of quality comparable to this photometric version of a near-encounter image returned four years later by NASA’s Mariner 6. (NASA)

The Mars-bound engineering test flight, 3MV-4A No. 2, lifted off on November 30, 1964 from the Baikonur Cosmodrome at 13:12 UT with its 8K78M launch vehicle placing it into a 153-by-219-kilometer parking orbit. After a short coast, the Blok L escape stage came to life and successfully injected Zond 2 into a trajectory towards Mars. But problems were already apparent during the first communication session with Zond 2 with only half of the expected power being generated by the solar panels. As the problem was being diagnosed, measures were taken to conserve power including the apparent cancellation of the Earth imaging session as well as postponing the first scheduled course correction maneuver.

In the end, it was found that one of the two solar panels on Zond 2 failed to deploy as intended. When the Blok L finished its burn, protective shrouds that were connected to the solar panels with lines were suppose to jettison pulling the solar panels out of their stowed position. It seems that one of the two lines had broken resulting in the failure of one of the solar panels to deploy. After several engine firings were made to shake the spacecraft, the stuck panel was finally deployed on December 15. Unfortunately, by December 18 communications with the probe were apparently becoming increasingly erratic. Some accounts suggest that a course correction was made around February 17, 1965 that further refined the path of Zond 2 towards Mars. At some point in time after late February, contact was finally lost with Zond 2 with an official announcement being made on May 5, 1965. Three months later on August 6, Zond 2 flew silently past Mars at a reported distance of 1,500 kilometers.

With this latest failure it was obvious that the 3MV design was still in need of much improvement and that better quality control was required to ensure success. One or two additional Zond test flights that were also mentioned in Smirnov’s letter to the Central Committee had already been approved for launch during the first half of 1965 before the next batch of Venera spacecraft were to be launched in November 1965.  Obviously, this hardware was modified based on the experience of Zond 2 before the launch of the Zond 3 test mission on July 18, 1965. In the mean time, responsibility for the Soviet lunar and planetary probes was transferred from the overworked teams at OKB-1 to NPO Lavochkin in April 1965 under Chief Designer Georgi Babakin – an organization known for its intensive testing of the hardware it built. Hopefully future revelations from old Soviet era archives will be able to fill in more details about this troubled chapter of the 3MV spacecraft series history.



The details of the decisions that led to the cancellation of the original 3MV-3 and 3MV-4 missions to Mars still remain to be revealed.  What if any effect the scientific community’s downward revision of the estimated surface pressure of the Martian atmosphere after 1963 had on these decisions also remains to be seen.  What is clear is that the original plans for the 1964 Mars window were scrapped and instead a single modified 3MV-4A spacecraft was launched.  Unlike what had been previously believed, the primary objective of this spacecraft was to perform a long endurance engineering flight test of the 3MV design to support future missions. If the spacecraft remained functional, the secondary objective was to redirect the spacecraft to impact the Martian surface and deliver commemorative pennants.  This spacecraft was never meant to acquire images of Mars during a flyby using their advanced imaging systems and at best would have returned only magnetic field and radiation data from the vicinity of Mars before impact.  In the end,  Zond 2 was successfully launched but suffered a series of problems that led to its failure several months before reaching Mars on August 6, 1965.


Note Added After Original Publication

A few hours after the original publication of this essay, I received clarification from Bart Hendrickx on the contents of Smirnov’s November 28, 1964 letter to the Central Committee (of which I had only limited translated excerpts from a couple of published sources by Bart Hendrickx). While I was originally working under the common assumption that three 3MV-4A spacecraft were intended to be launched in the November-December 1964 window, he stated:

“I can add that in the 28 November 1964 letter to the Central Committee it is clearly stated that only one mission would be launched in the November-December 1964 window, so there was no plan to launch three probes during the window. The document also mentions a plan to launch “one or two” 3MV vehicles in “Zond” configuration in the first half of 1965 before the opening of the Venus window in November 1965″

Oddly enough, the fact that only one spacecraft was to be launched confirmed an off-hand speculation of mine at the end of the 3MV-4A Trajectory Analysis section of this essay as originally published. Mr. Hendrickx also added:

“And one more thing from the same document : the original plan was to launch four (not three) Mars probes during the 1964 window, two landers and two flyby probes. This was before new findings about the density of the Martian atmosphere changed plans.”

I have slightly altered this essay especially in the last two sections, The Flight of Zond 2 and Conclusions, from what was originally published to reflect the clarifications. My thanks to Bart Hendrickx for his input.


Follow Drew Ex Machina on Facebook.


Related Reading

“The Mission of Zond 2″, The Space Review, Article #2745, May 4, 2015 [Article]

“The Mystery of Zond 2”, Electronic Journal of the Astronomical Society of the Atlantic, Volume 2, Number 9, April 1991 [Issue]

“Mariner 4 to Mars”, Drew Ex Machina, July 14, 2015 [Post]

“A Cautionary Tale of Extraterrestrial Chlorophyll”, Drew Ex Machina, October 5, 2014 [Post]


General References

J.M. Brayshaw, “Mars Atmospheric Entry Parameter Study”, Technical Report No. 32-458, JPL, September 15, 1963

Boris Chertok, Rockets and People Volume III: Hot Days of the Cold War (ed. Asif Siddiqi), SP-2009-4110, NASA History Division, 2009

V.C. Clark, Jr., W.E. Bollman, R.Y. Roth and W.J. Scholey, “Design Parameters for Ballistic Interplanetary Trajectories Part I. One-way Transfers to Mars and Venus”, Technical Report No. 32-77, JPL, January 16, 1963
p. 72, Fig 4-8 “Mars 1964: Minimum injection energy vs launch date”
p. 271, Fig 12-1 “Mars 1964: Time of flight vs launch date”
p. 284, Fig 12-14 “Mars 1964: Asymptotic speed with respect to Venus vs launch date”

Edward Clinton Ezell and Linda Neuman Ezell, On Mars: Exploration of the Red Planet 1958-1978, SP-4212, NASA, 1984

Samuel Glasstone, “Chapter V: The Atmosphere of Mars”, The Book of Mars, SP-179, NASA, pp. 73-98, 1968

Brian Harvey, Race Into Space: The Soviet Space Programme, Halsted Press, 1988

Brian Harvey, Russian Planetary Exploration: History, Development, Legacy and Prospects, Springer-Praxis, 2007

Bart Hendrickx, “Managing the News: Analyzing TASS Announcements on the Soviet Space Program (1957-1964)”, Quest, Vol. 19, No. 3. Pp. 44-58, 2012

Wesley T. Huntress and Mikhail Ya. Marov, Soviet Robots in the Solar System: Mission Technologies and Discoveries, Springer-Praxis, 2011

Lewis D. Kaplan, Guido Munch and Hyron Spinard, “An Analysis of the Spectrum of Mars”, The Astrophysical Journal, Vol. 139, No. 1, pp. 1-15, January 1, 1964

Andrew J. LePage, “The Mystery of Zond 2”, Journal of the British Interplanetary Society, Volume 46, Number 10, pp. 401-404, October 1993 [Abstract]

Andrew J. LePage, “Trajectory Analysis of the Soviet 1964 Venus Missions”, Drew Ex Machina, April 2, 2014 [Post]

Andrew J. LePage, “Trajectory Analysis of the Soviet 1962 Mars Missions”, Drew Ex Machina, May 2, 2014 [Post]

George M. Levin Dallas E. Evans and Victor Stevens (ed), “NASA Engineering Models of the Mars Atmosphere for Entry Vehicle Design”, NASA Technical Note D-2525, November 1964

V.I. Moroz, “The Infrared Spectrum of Mars (λ 1.1 – 4.1 μ)”, Astronomicheskii Zhurnal, Vol. 41, No. 2, pp. 350-361, March-April 1964 (submitted September 21, 1963)

Steven Musman, “An Upper Limit to a Rayleigh Scattering Atmosphere on Mars”, Planetary and Space Science, Vol. 12, No. 8, pp. 799-800, August 1964

T.C. Owen and G.P. Kuiper, “No. 32 A Determination of the Composition and Surface Pressure of the Martian Atmosphere”, Communications of the Lunar and Planetary Laboratory, Vol. 2, Part 1, pp. 113-132, August 24, 1964

V.G. Perminov, The Difficult Road to Mars: A Brief History of Mars Exploration in the Soviet Union, Monographs in Aerospace History No. 15, NASA History Division, July 1999

Timothy Varfolomeyev, “Soviet Rocketry that Conquered Space Part 5: The First Planetary Probe Attempts, 1960–1964”, Spaceflight, Vol. 40, No. 3, pp. 85–88, March 1998

Timothy Varfolomeyev, “Soviet Rocketry that Conquered Space Part 6: The Improved Four-Stage Launch Vehicle, 1964-1972”, Spaceflight, Vol. 40, No. 5, pp. 181-184, May 1998

HORIZONS Web-interface, JPL [Web site]