Luna 3 - the first view of the moon's far side


One of the origins of Luna-3 was the work by Boris Raushenbakh on attitude control started in 1955  in the NII-1 rocket research institute of the Ministry of Aviation. This work was co-ordinated with Korolev's design bureau and had as its goal stabilised photo-reconnaissance satellites. Raushenbakh's group was contracted by the lunar project section Korolev's design bureau OKB-1 to design the attitude control for the E-2 variant of lunar exploration craft intended to image the Moon's hidden side. The manager of the lunar probe project at OKB-1 was Gleb Maksimov. An excellent account of these early Soviet lunar missions can be found in (4) and (14). A cutaway sketch of the space probe with a Soviet-era translation of Russian captions added is shown below.

The radio system was designed by deputy chief designer of the NII-885 (Scientific Research Institute of Radio Instrument Building, established 1938), Yevgeniy Boguslavskiy at NII-885 and it provided trajectory measurements, telemetry and command and command verification functions. Boguslavskiy and Boris Chertok (1) (deputy chief designer at Korolev's OKB-1) preferred pulsed systems for telemetry and ranging ever since the 1940's when Boguslavskiy had criticized the German use of carrier based radio systems in their missiles. However for E-2A Boguslavskiy surprised everyone by designing a continuous carrier system. The explanation was that tried and true methods were quicker and safer to develop under a rushed schedule. Boguslavskiy later explained that the lack of contrast in the pictures from Luna-3 was caused by lack of onboard transmitter power. The low contrast of the Luna-3 images is partly due to the fact that the Moon was illuminated from above. There isn't much contrast in a full moon. The electronics of Luna-3 was transistorized, an innovation that was regarded as risky.

Boguslavskiy and the NII-885 was also responsible for setting up the ground station at Kochka mountain near Simeiz in the Crimea in cooperation with the military unit 32103. The station was placed on the southern slope of a mountain facing the Black Sea. It was formally a part of NII-4 under general Sokolev (1). The location probably had something to do with the fact that the Crimean Astrophysical Observatory and the Physical Institute of the USSR Academy of Sciences was, and is, located at Simeiz. The ground station and control center had been in operation since 23 September 1958, i.e the same day that the very first Soviet attempt to reach the moon was made (14).
 

The flight starts - everything looks good on 4 October

The E-2A probe was sent to Tyuratam in August 1959 even though it was not ready for flight. Final testing would be performed at the launch site. Chertok traveled to the launch site on 17 September and the probe was declared ready for launch on 25 September (1). In the small hours of 4 October 1959, a mere two years after the launch of Sputnik, the Soviet Union launched its most impressive space spectacular ever - Luna 3 - the first space vehicle to catch a view of the moon's hidden far side. The E-2A probe weighing 278.5 kg lifted off at 0043:39 UT on an 8K72 rocket from Baikonur. The space probe was placed in an earth orbit with perigee at 48280 km, apogee at 468300 km and an inclination to the equator of 55 degrees (5). During its flight, that lasted until 29 April 1960 when it decayed in the Earth's atmosphere, Luna 3 was to bring credit to its creators and put the British Jodrell Bank radio telescope into the media focus.

Korolev wanted that the TASS communiqué of the launch of the probe mentioned the objective of photographing the Moon's far side, but less courageous minds prevailed (1). Soviet media announced that Luna 3 transmitted signals on 39.986 MHz that consisted of pulses 0.2-0.8 seconds long sent at a repetition rate of 1 +/- 0.15 Hz. The transmitter on 183.6 MHz was also used for tracking (trajectory determination). Information from the probe was sent at intervals of 2-4 hours/day. On 4 October 1959 the transmitters were to be switched on at 1300 MT (1000 UT) to operate for two hours (until 1200 UT) at a point 108000 km over 80 degrees east and 5 degrees south. At this time everything appeared to be in good shape (1).

For Luna 3, the Soviet Union again provided Jodrell Bank with pointing information for its telescope during the Luna 3 flight. Such data were sent on 4,7,8,9 October 1959 and on several more days in October 1959. Therefore, Jodrell Bank successfully picked up Luna 3 less than ten hours after launch: Signals on 183.600 MHz were received at 1206-1220.25 UT on 4 October 1959 (this agrees reasonably well with the schedule announced by the Russians at launch - see above). Jodrell Bank kept their colleagues at NASA up to date with its tracking activities on a daily basis. Pointing information from the Russians were obviously also relayed to NASA (8). In general, observations were sent both to the USA and the USSR. All data from the USSR were forwarded to the U.S., but not vice versa. Comprehensive reports including tape recordings were sent to Moscow after the flight was over. The reports to NASA were sent to the telegram address SPACE CONN NASA WASH DC (for the attention of an individual called Bryant).
 

Erratic signals on 5 October - Korolev dashes to the control centre

In the morning of 5 October Chertok returned to Moscow from the launch site (1).  He returned to work in the morning of 6 October and was in a meeting about the upcoming Venus shots when Korolev called him on the phone and told him to "come along, bring no documents, don't count on returning home today...". When Chertok entered his boss' room Korolev was ordering up a plane. After hanging up he explained that "radio contact with the probe are very bad. Telemetry cannot be received properly and commands are not received. We will fly to the Crimea to be there in time for the next communications session that starts at 1600 MT (1300 UT). There are two cars at the gate. Take one, go home, fetch what you need and then come to Vnukovo. The plane leaves at 1200 MT (0900 UT)".

Korolev had taken the decision in the morning after reviewing the results of the previous day's communications session. Weak signals had been received in 5 brief periods: 1212-1221, 1227-1233, 1239-1244, 1250-1309, 1341-1409 UT (see diagram below taken from (17)). During 5 October Jodrell Bank could not receive any signals on main telemetry frequency, 183.6 MHz. In a message to NASA dated 5 October 1959, Jodrell Bank described observations made on 5 October 1959 described: "Signals received on 39.986 MHz from 1035 to 1412 UT. At 1301 UT the frequency is fluctuating rapidly and at 1351 UT the frequency starts drifting down rapidly." This was the only occasion on which Jodrell bank failed to pick up the main communications link on 183.6 MHz despite the fact that the British radio telescope had an aperture almost ten times that of the little station in the Crimea. Clearly there was reason for Korolev to be worried!.

To organize the hurried flight Korolev had to mobilize the air force, Aeroflot, the Communist party in the Crimea and the Soviet Council of Ministers! The flight to the Crimea was made with the most modern airliner at the time, a Tu-104 jet plane. There were six passengers on board including Korolev, Academician Keldysh, Boris Chertok and Arkaiy Ostashev (senior engineer in OKB-1) who arrived so late that the plane had to interrupt taxiing to bring him on board. The plane landed in the Crimea (probably Simferopol, although Chertok does not say so). They were picked up by helicopter but could not land at Ai-Petri as originally planned because of "wet snowfall" and zero visibility. Instead they diverted to a helicopter pad at Yalta. The SIM and Pobeda cars from the local party committee were redirected to Yalta. The local party bigwigs wanted to take Korolev and his party on a tour of the area, but Korolev pointed out the need to proceed immediately to Simeiz. Korolev and his party arrived at the control center near the "Pavlov observatory" at 1430 MT (1130 UT).

The antenna system for communicating with the probe had an aperture of 100 square meters. It consisted of an array of crossed dipoles (turnstile antennas). It may have looked something like the WW2 German Freya radar station. (see image on the right). The receiving antennas for short-wave telemetry on 19.993 MHz and 39.986 MHz (see "The radio systems of the early Luna probes") were probably also located at Kochka mountain or nearby. For 20 MHz two stacked three element Yagi antennas with a total gain of 7-10 dB, while 40 MHz was received with a crossed Yagi (to pick up the two different polarisations) using ten elements and having a gain of 13 dB. In (16) it is stated that "the receiving antennas were set up on a steep seashore".

Radio equipment were housed in trailers and the control center itself was housed in a temporary wooden barrack. In one of the rooms was the image recording equipment where the pictures from the moon were to be recorded on thermo-paper for real-time readout and film, later to be developed in Moscow. The staff lived in tents and there was a field kitchen cart steaming away among the tents. It all gave the appearance of a military camp(1), which it was to some extent. Koshka (Cat) mountain is located at approximately 34.02 North  44.54 East longitude. There was a back-up station located on the Kamchatka peninsula as described vividly in (19).
 

Korolev takes command - reliable contact re-established on 6 October

Boguslavskiy offered the opinion that the problems the following day was caused by an unfortunate "dip" in the antenna diagram of the probe. Korolev interviewed the operators of the control center and found that the work of various operators was inadequately co-ordinated and that military officers formed a second tier of direction for the operators. Korolev now ordered that only Boguslavskiy would direct operations during communication sessions with the probe. At 1600 MT (1300 UT) telemetry was received and to everybody's surprise everything worked well onboard the spacecraft. During this communications session the probe made its closest approach to the Moon at 1416 UT on 6 October 1959 when Luna 3 passed over the lunar south pole at a distance 7884 km from the moon's centre.

Jodrell Bank observed Luna-3 again (11): "Signals received Oct. 6 183.6 Mc/s only. 1342-1405.40 UT:15 sec on 15 sec off (probably the off period represents a frequency shift). 1405.40-1409.30 UT: No signals in band. 1409.30-1510 UT: usual bleeps. 1510-1538 UT: Modulation as in first period." There were two signals in the band with about 5 kHz separation (11). Was the silence at 1405.40-1409.30 UT caused by the probe's passing behind the edge of the moon?

Chertok (1) recalls that after the communications session all went out with Boguslavskiy to smoke. An ice-cold wind was blowing while the men watched the coastal resort area in the light of the setting sun. Boguslavskiy pointed out a naval vessel at sea that was there to check for radio interference. Korolev demanded that a plan was drawn up for operations the following morning when the pictures of the Moon were to be taken. At that point the designer of the Yenisey imaging system, Brazlawez, worried that the available film in the image recording system was not sufficient and pointed out that the idea was that Korolev's flight from Moscow should have brought extra film. Korolev was furious, but arranged via Rudnev that another Tu-104 would take the film from Moscow to Simferopol where it would be fetched by helicopter.

Korolev finally left Koshka mountain for a hotel in Nishnaja Orenada near Yalta but left Ostashev at Kochka overnight. A sumptuous dinner was ready even thorough Korolev and his party arrived near midnight. Korolev ordered that nobody should drink alcohol because they needed to leave at 6 a.m. They got only four hours of sleep.

The 29 exposures of the moon's far side were taken at 0330-0410 UT on 7 October 1959 when the probe was 65567-68785 km from the moon's centre (7). Intermittent telemetry was received at the start of the picture-taking session, but to save power telemetry was switched off. Keldysh called NII-4 to check on the trajectory. Ochozimski at the department of applied mathematics at the Steklov institute of the USSR Academy of Sciences was responsible for trajectory calculations together with Elyasberg at OKB-1 and the NII-4 institute. The computations were made on a computer at the Steklov institute. The reply from NII-4 was that "to be on the safe side we have re-calculated the trajectory data three times and the probe has not approached closer to the Moon than 7000 km". (The actual closest approach was 7884 km from the moon's centre).
 
 

Film from American "spy balloons"!

Academician Andrei Severny from the Simeiz observatory had been invited to the control center by Keldysh. Severny now spread some panic while everyone was waiting for the picture-taking process to be completed. He was of the opinion that the probe would have needed 5-6 cm of lead shielding to avoid fogging of the film. Chertok only writes in his book "We shall see"! (1).

What Severny did not know, but probably Chertok, was revealed 45 years later in Novosti Kosmonavtiki (13) of August 2000: radiation-tolerant film from U.S. high-altitude "Genetrix" spy balloons was used in the "Yenisey" imaging system!: (rough translation by Babelfish edited by S.G.)"....The history of that used in the camera the "Yenisey" of the film of the type ASH with a width of 35 mm is sufficiently amusing. According to the evidence of the deputy chief designer of the "Yenisey" Bratslavtsa and of leading engineer of the onboard camera Yu.P.Lagutina, our industry by that time yet did not master the production of the film, which satisfied all requirements for "Yenisey". But the situation was rescued by the fact that in the second half of the 50's the USA used balloons for reconnaissance purposes. The possibility of their application for the reconnaissance was based on the special features of the airstreams above our country - constant displacements of air masses from the West to the east. The balloons, supplied with special photographic equipment, were started from the military bases of the USA in the countries of West Europe and, carried by air streams, appeared above the USSR, photographing the territory of our country along the path of motion. .... Certain quantity of film from these balloon spies proved to be in the academy of the name of A.F.Mozhayskogo, with which collaborated with VNII television. After a study of the mentioned film it turned out that it from its parameters was suitable for the use in onboard equipment "Yenisey". Then was accepted, secretly from the high authorities, the solution to cut it into the required dimension, to perforate and to use for photographing of the hidden side of the moon. Hence becomes clear a somewhat mischievous designation of the type of film ASH - "American ball(oon)s"." 

The Luna-3 team


I took this picture in May 2008 at the Musuem honouring the chairman of the Academy
of Sciences during the heroic space era, Mstislav Keldysh. It shows the team behind
Luna-3 gathered at the ground station at Simeiz in the Crimea on 8-9 October
1959. Don P Mitchell helped me straighten the picture and removing the reflections
in the cover glass. From left to right: N.S. Lidorenko, I.I. Pikovskiy, M.V.Keldysh,
A.V.Belousov, S.P. Korolev, Y.Y. Boguslavskiy (deputy chief designer, NII-885),
M.S. Ryazanski (chief designer, NII-885)

 

The images are transmitted - but when?

The film was developed, fixed and dried on board. Thereafter it could be scanned with a light beam with up to 1000 lines per image for transmission to Earth. Radio observations at Jodrell Bank on 7 October 1959 were described as (12): "Signals on 183.6 Mc/s 1400-1459 UT. 39.986 MHz for about 3 minutes before 1505 UT. The probe appeared to be 1-1.5 degrees above the elevation given by the Russians. CW at one of several frequencies near 183.6 MHz between 1400 and 1412. Bleeps from 1412 to end. Bleeps on 39.986 MHz. The two frequencies were not on together." This description of the signal does not seem to indicate that a picture transmission similar to facsimile was being made.

Chertok's book indicates that the images were received on 7 October and that Korolev was present while the images appeared on the thermo-paper.  Chertok sat next to Boguslavskiy as the first pictures appeared but they were disappointing - only showed that the Moon was round! Later during the transmission the image quality improved. Whether or not a later transmission of images was made is not clear from Chertok's book. Very late that evening  (7 October) Korolev and his party returned to the hotel in Nishnaja Orenada and this time Ostashev came with them and the government wine and vodka cellar was utilized.

On Monday 19 October 1959 newspapers (9) reported that Moscow media circles had started to talk about the fact that pictures had been taken of the moon. These reports appeared after the probe had come close enough to the Earth to provide sufficient picture quality. Had the transmission of pictures failed the rumors would never had started, it was speculated (9). The pictures of the moon's far side were distributed by tele-photo from Moscow late on Monday (UT) 26 October 1959 and created another space age media frenzy. According to newspaper reports at the time the picture transmission to Earth took place on 18 October 1959 (10).

Jodrell Bank kept up contacts with the Russians even after their publication of the far side pictures. On 1 November 1959 a telegram was sent to AN USSR asking for "co-ordinates for the next transmission from Luna 3", and the first batch of tapes of signals from Luna 2 and 3 were sent to Prof. Nesmeyanov of AN USSR on 9 December 1959. In the accompanying letter Prof. Lovell thanked the Academy for the data with pointing information for the Jodrell Bank telescope. The final batch of Luna 3 tapes were sent to Akademi Nauk on 14 January 1960. The story about Jodrell Bank's reception of Luna 3 signals could have ended here, but there is an interesting epilogue: On 17 January 1961 the director of the Jet Propulsion Laboratory, William Pickering, wrote to find out if Jodrell Bank had recorded pictures from Luna 3 and if JPL could have copies of the tapes. Prof. Lovell replied on 20 January 1961 by saying that "Yes, we have a considerable number of tape recordings", and that he had asked Dr. J.H. Thomson to find the tapes, copy them and send them to JPL. Probably the CIA analyzed the tapes sent to Dr Pickering, because in 1994, three decades later, a secret American assessment (6) of Jodrell Bank's Luna 3 recordings was declassified:

"The very weak signals returning to Earth from Lunik III could not be picked up by standard Elint collection facilities and equipment. Through the co-operation of the University of Manchester in England, therefore, use of the 250-foot radiotelescope at Jodrell Bank was obtained. The Soviets announced that pictures of the far side of  the moon were taken between 0330 and 0410 UT on 7 October 1959. Jodrell Bank succeeded in getting a signal on the announced frequency of 183.6 megacycles between 1410 and 1445 UT that date. A ten -hour lag behind the actual picture-taking is not excessive: the film had to be developed, and then the electronic scanning and transmission had to wait until rotation of the Earth brought the Soviet deep-space station in the Crimea into position to pick up the signal. The graduate students working for Sir Bernard Lovell at Jodrell Bank, however, in recording their intercept, narrowed the receiver bandwidth so far, in order to increase signal-to-noise ratio, that they "clipped off" much of the video information. Confirmation that the Lunik III pictures released by the Soviets were valid thus rested on a single poorly recorded intercept from which it could be judged only that the picture was more likely of a test pattern than of the moon. The signal bandwidth, however, was compatible with the parameters announced by the Soviets, and on the basis of technical extrapolation it could be concluded that the Soviet lunar pictures were authentic."

This assessment is interesting. In my opinion the picture transmission took place on 7 October, but also perhaps some 10 days later after which date the rumors in Moscow about the reception became rampant (9). In (7) it is mentioned that two transmission rates were available: "a slower transmission at great distances, and a faster transmission on approaching the Earth" and that image transmissions were made from distance of 467000 km (near apogee of the post-flyby trajectory). Perhaps the pictures sent at apogee were not good enough for publication but those received later were. We know that the pictures were transmitted in some kind of facsimile format, frequency-modulated on the carrier. So it seems that there were pictures sent on 7 October, but exactly when did the second transmission - if any - take place?

In (17) there are graphs showing the signal strength on 183.6 MHz during the flight of Luna-3. Clearly, signal levels on 5 October were very low (see figure below). However, the interesting thing about this reference is that transmission times after the images were taken near the Moon can be deduced (see table above left). The signal strength on the 16 and 17 October was very much better than nearer the Moon. Probably, this is when the good pictures were received.


 
 

Imaging system

This is what Don P. Mitchell writes at his web site: "On October 7, 1959, Luna-3 returned the first images of the hidden side of the Moon. Using a phototelevision camera, pairs of images were simultaneously exposed through 200mm and 500mm lenses. The Luna-3 camera was developed by P.F. Bratslavets and I.A. Rosselevich at the Leningrad Scientific Research Institute of Television. The images were processed and analyzed by Iu.N. Lipskii and his team at the Sternberg Astronomical Institute.

The camera held 40 frames of film, and 12 images were received via frequency-modulated analog video (some reports claim 17 images were received). The full moon appears to have very little detailed texture, because the lunar mountains and terrain casts no shadows when lit from overhead. The fine stiple seen in some images may be the video subcarrier. .."

Six of the Luna-3 frames have been published and can be found at Don P. Mitchell's excellent site.

References and Notes

  1. Chertok, Boris E., "Raketen un Menschen - Der Sieg Koroljows", Elbe-Dnjepr-Verlag, 2001, pp. 278-292.
  2. J.G Davies, A.C.B. Lovell, Observations of the Russian Moon Rockets Lunik II and III, Space Research 1, Proceedings of the first COSPAR conference, 1960.
  3. Letter from Professor Lovell to Prof. Nesmeyanov of AN USSR, 9 December 1959.
  4. Asif A. Siddiqi, "First to the Moon", Journal of the British Interplanetary Society, Vol.51, pp.231-238, 1998.
  5. Leonid I Sedov, "The Orbits Of Cosmic Rockets Towards The Moon", presented at American Rocket Society's 14 the Annual Meeting, Sheraton Park Hotel,Nov 16-20,1959. ARS paper no 1051A-59.
  6. Henry G. Plaster, "Snooping on Space Pictures", Studies in Intelligence, Fall 1964.
  7. "The Other Side of the Moon", Translated from the Russian by J.B. Sykes, Pergamon Press, 1960.
  8. A telegram from Prof. Davies to SPACE CONN NASA WASH DC dated 8 October 1959. The telegram concluded "RUSSIAN PREDICTIONS FOR OCT 9 FOLLOW".
  9. Swedish Daily Expressen for Monday, 19 October 1959, p. 12: "Lunik saw 2/3 of the moon's far side"
  10. Swedish Daily Expressen for Tuesday, 27 October 1959, centre spread: "Pictures from space: the far side of the moon".
  11. A telegram to SPACE CONN NASA WASH DC dated 7 October 1959
  12. Second telegram to SPACE CONN NASA WASH DC from J.G. Davies dated 7 October 1959
  13. V. Efimov, Novosti Kosmonavtiki, August 2000.
  14. Timothy Varfolomeyev, Journal of the British Interplanetary Society, Vol.52, pp.157-160, 1999.
  15. Flagg, Richard S., analysis made in private e-mail to the author 14 November 2004.
  16. V.A. Kotelnikov, .M. Dubrovin, O.N. Rzhiga and A.M. Shakhovsky. "Reception and investigation of the properties of radio signals from Soviet Space Rockets".
  17. V.V. Vitkevich, A.D. Kuzmin, R.L Sorochenko and V.A. Udaltsov, "Results of Radio Astronomical Observations Obtained With Soviet Space Rockets", Iskusstvennyye Sputniki Zemli, 1961, No. 7, pp. 23-31.
  18. The Department of Radioastronomy of the Crimean Astrophysical Observatory with its 22-m radio telescope for mm and cm radio waves is located at the foot of mount Koshka (The Cat) in Katsiveli (near Simeiz) (lat=+44:23:52.6, long=2h 15m 55.1s). One of CrAO's Departments - Simeiz station (lat=+44:24;12.3, long=2h 15m 59.3s) is located on mount Koshka at an altitude of 346 m and a distance of 3 km from the shore. A laser device which has been mounted at 1-m telescope for the location of satellites is under operation here.  Could this be the old Luna-3 station site?
  19. V. Yefimov, ANNIVERSARY OF THE SPACE TELEVISION, Tele-Sputnik No. 3(5) March 1996

Back to Space Tracking Notes