Vostok retro-fire attitude

Sven Grahn

Introduction - the solar orientation as a back-up mode

From the literature we know that Vostok spacecraft were originally designed with an attitude determination system based on an earth and sun sunsensor. However, this system repeatedly malfunctioned and the Vostok operational flights then relied on a sun sensor only solution to find the proper retro-fire attitude, a back-up mode devised by Igor Yatsunskiy (1). The idea was to use specific launch time to have attitude for de-orbit burn with TDU engine looking directly to to the Sun. .

The figure on the right has been published in "The Soviet Encyclopaedia of Space Flight" (2) to illustrate the principle of orienting Vostok for retro-fire. showing just how the back-up orientation worked. It shows the spacecraft adopting the solar attitude at exit from eclipse and then keeping this until the retro impulse vector was pointing slightly downwards. In (3) we can read:

"B E Chertok was in charge of the orientation system. It consisted of two redundant systems: an automatic/solar orientation system and a manual/visual orientation system. Either system could operate two redundant cold nitrogen gas thruster systems, each with 10 kg of gas .... The automatic solar orientation system consisted of solar sensors, DUS-L2 angle of flight sensors, and an SRB analogue computer unit. The TDU would only fire if the sun sensors - consisting of a slit arranged over three photocells - indicated correct orientation. The DUS-L2 angle of flight sensor utilized two-step double gyroscopes with mechanically opposed directions. The SRB used these inputs and generated impulses to carry out the burn."
In general, a single attitude sensor such as a sun sensor is not sufficient, you need at least two sensors to determine attitude (and then you may obtain two solutions and have to figure out a way to discard one). However, as Igor Lissov pointed out and Igor Yatsunskiy found out, if the sun direction coincides with the desired attitude a single sensor suffices. If you point the longitudinal axis of the spacecraft at the sun the roll orientation does not need to be controlled, only its rate.

Thus, it is interesting to study the orientation of the sun vis-a-vis- the orbital plane and the orbital path at the time of retro-fire for various Vostok missions.

Summary of sun orientation vis-a-vis the orbital plane and path

First, I have tried to establish the time-line of all Vostok mission, especially trying to find out the exact moment of ignition of the retro-rocket. This has been possible to do with great accuracy in the cases when that piece of data has been published by Soviet sources. In other cases I have estimated the instant using the rule-of-thumb that the retro-fire point is 8000 km from the landing point. Orbital data have been obtained from Jonathan McDowell's "Space Home Page".

The first analysis that I ran was to compare the right ascension of the sun to the right ascension of the ascending node of the orbital plane of the Vostok spacecraft at the time of retrofire. You would expect that the difference between these two numbers be zero, so that the sun would be in the orbital plane. But, as you can see from the figure below, this is not the case.

The next analysis that I ran was to compute the azimuth and elevation of the sun at the time of retrofire.  Azimuth is defined as the position of the sun in a plane perpendicular to the radius vector to the spacecraft. Zero-azimuth is in the direction of flight  and the positive sense is clock-wise as seen from above (analogous to the normal definition of azimuth on the ground). Elevation is the angle of the sun above this plane.

The retrofire problems of Korabl-Sputnik-1 - still more mysteries

My first finding in running this analysis is that the retro-fire of Korabl-Sputnik-1 (Sputnik-4) occurred in eclipse at 2352 UT on 19 May 1960. The spacecraft had entered into eclipse at 2346 UT with the sun at an azimuth of 176.2 degrees. This is an obvious result if you consider the fact that Sputnik-4 was launched at around 0000 UT, while later flights typically were launched around 0700 UT. But, the surprising fact is that the sun would be in the exactly opposite direction as that required for orienting the retro-rocket engine for ignition. Also, since the retro-fire impulse was initiated in darkness, it was not possible to use the sun as an attitude reference anyhow. This makes the following passage from (4) a bit hard to understand:
"Prior to the scheduled firing of the TDU-1 engine, the control group at Tyura-Tam (Group T) had detected anomalies in the primary system of attitude control, which used the infrared sensor. Although the system as a whole seemed to functioning fine, the sensor itself was not responding correctly. The Tyura-Tam team reported the problem to the control group at Moscow (Group M), but the designer of the system, Boris V. Rauschenbakh, refused to agree to Group T's recommendation to use the back-up system of orientation. OKB-1 Deputy Chief Designer Boris Ye. Chertok, who was the head of Group T, quickly called a meeting at Tyura-Tam and reached a consensus that the primary system not be used, in favour of the still-operaring solar-based Grif sensor. He then passed this recommendation on to Moscow. Although it seems that Korolev agreed with Chertok at first, he gave in after persuasive arguments by Keldysh and Rauschenbakh to go ahead with the primary system. Unfortunately, on the sixty-fourth orbit, the primary system malfunctioned and the fourteen thrusters working on compressed nitrogen inserted the spacecraft in the exact wrong attitude. The TDU-1 retrorocket automatically fired on time at 0252 hours Moscow Time, but the  spacecraft, instead of re-entering the atmosphere, was boosted into a new high orbit of 305 by 690 km..."
If the solar orientation system had been used it would certainly have caused the spacecraft to be oriented the wrong way (with the same orientation of the sun sensor as implied by the sketch at the top of this page), and at the time of retrofire the sun sensor was not possible to use as we have seen!

Korabl-Sputnik-3 - didn't it use an earth sensor?

The second observation from this analysis is evident from the figure on the right. The sun orientation was, in general, quite far from the expected azimuth = 0. That the elevation was larger than zero is quite as expected when one compares to the sketch at the top of the page. There is one point in the polar plot that is very much extreme - close to 80 degrees azimuth and 50 degrees elevation, i.e. nearly pependicular to the orbital motion. This represents the solar orientation at the time of retrofire for Korabl-Sputnik-3, a.k.a. Sputnik-6, on 2 December 1960. According to p.259 of (4), this craft was only equipped with the solar orientation system. This causes me severe problems! Korabl-Sputnik-3 would have fired its retrorocket almost perpendicular to orbital path creating almost no delta-v along the orbital path. I propose that it was equipped with the infrared system, Clearly it is difficult to use the solar orientation system in the middle of the winter because the sun is below the equatorial plane. It produces precisely the type of solar attitude determined for Korabl-Sputnik-3.

The average of all the other values of azimuth and elevation is azimuth = 29 degrees and elevation = 38 degrees. This is marked with a blue dot in the polar plot.

Gagarin's debriefing

Can we find any reference to the retro attitude of Vostok from the literature? Well, let me quote from the post-flight debriefing (5) of Gagarin:
Question: During braking in the segment when the RRS (Retro-Rocket System) was in operation, did the Earth run from your feet to the instruments, or did you pay attention to that?
Answer: The Earth moved a bit not from my feet to the instruments, but from my feet to the left, at an angle.
Question: What kind of angle?
Answer: It moved at an angle of about 30 degrees. ...

More launch time constraints than solar orientation at retro-fire?

The value from my calculations is 46 degrees. The fact that the sun is systematically NOT in the orbital plane, shows that there was some other factor than retro attitude that also dictated the launch time. Perhaps the fact that there must be enough time before sunset to look for the landing capsule upon landing.


  1. Igor Lissov, e-mail to Sven Grahn dated 23 January 2003.
  2. "The Soviet Encyclopaedia of Space Flight" , Mir Publishers, Moscow 1969, p. 495.
  3. Mark Wade's Encyclopaedia Astronautica, Vostok-3KA article
  4. Siddiqi, Asif, "Challenge to Apollo", NASA  SP-2000-4408, pp.253-254.
  5. "Questions asked of Comrade Yu. A. Gagarin. Answers of Comrade Yu. A. Gagarin at a meeting of the State Commission , 13 April 1961", Izvestiya, 9 April 1991, pp. 123-128

Back to Space History Notes