TKS short-wave telemetry

Sven Grahn

Kosmos-929

This spacecraft transmitted frequency-shift-keying (FSK) pulse-duration modulation (PDM) signals on 19.954 MHz continuously during the flight. An example of a PDM frame is shown below.

During the frame duration of 16 seconds sixteen equal word-length intervals are available. Fifteen of these are used to transmit data, whilst the other, word zero contains a synchronization pulse-train. The FSK transmitter is switched between two radio frequencies, fH and fL, 1 kHz apart. (Listen to Kosmos-929 here). At the beginning of a data-word interval, fH is turned on. At some time before the end of the interval, fH is turned off and fL simultaneously turned on, and remains on until the end of the interval. The ratio of the duration of fH to fH+fL is a measure of the data value being transmitted. There are 32 possible times within a word within a word when the fH to fL transition can occur (2). The analogue data being sampled has been quantised in one of 32 possible levels. The permissible fH to fL transition times are defined in the synchronization pulse-train which normally contains 29 pulses. In this text word-values will be expressed in the form of integers ranging from 1 to 32, where 16 corresponds to fH being on for 50% of the word interval.
 

Two separate telemetry modes were transmitted on a time-sharing basis. Mode A had a 30-pulse synchronization train and word 14 taking on the word value 15. Mode B was characterized by having 29 pulses in the synchronization train and word 14 had the word value 23 (1). One telemetry mode was transmitted for 30 minutes and the other for 30 minutes and then the cycle was repeated.

Mode changes occurred at 28 min 55 sec and 58 min 55 sec past the hour. The commutators in the two modes were not synchronized anthe switch-over between modes could occur at any word within the frame. Although  not synchronised, the cycling-rates of the two formats were equal to within +/- 20 milliseconds.

Different transmitter frequencies were employed for modes A and B. In mode A, fH=19.954 MHz and fL=19.953 MHz. For mode B, fH=19.955 MHz and fL=19.954 MHz.

In general, the transmited data words in modes A and B were quite similar. The tables above show word-values up to 17 September 1977. Word 3 and 4 in both modes were those that varied most during the first two months of the flight. In the graph on the right values of words 3 and 4 in both modes are plotted as a function of time.

It is still a mystery what these two slowly varying values (3 and 4) signified. Similarly it is still unknown why there were two different, but quite similar, telemetry modes. Did they represent two redundant chains of subsystems?
 

Kosmos-1267

19.954 MHz
Just as for Kosmos-929, Kosmos-1267 transmitted two telemetry modes, A and B, on 19.954 MHz. The two modes replaced each other evry ten minutes, as opposed to every thirty minutes for Kosmos-929. The mode-change occurred at 8, 18, 28 .... minutes after the hour. Both modes had 30 pulses in the synchronization train. Words 3,4 and 5 were very much different between the modes as can be seen from a sample of the telemetry in the table below.
 
Word 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Mode A 14 14 17 16 18 14 6 22 6 6 17 30 23 21 14
Mode B 14 14 2 2 2 14 6 23 7 7 17 30 22 22 15

Otherwise there were negligible changes in word values between modes. However, word 11 varied much - between 15 and 27, but there was no systematic variation over time. Word 8 was always at 22-23 except during the orbit when the craft docked with Salyut-6. Then and only then did word 8 drop to 14.

19.942 MHz
During this flight a second short-wave frequency was also observed, 19.942 MHz. Signals on this frequency were not continuous but rather quite rare. It was discovered by Jan-Ola Dahlberg in Malmö in southern Sweden. Several stations, including myself, confirmed the signal. It was not picked up after 11 May 1981. I have called this signal mode C. To compare with modes A and B I have repeated the table above with mode C included. One can then see that modes C contained many more short words than the two other modes. It had 29 pulses in the synchronization train.
 

Word 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Mode A 14 14 17 16 18 14 6 22 6 6 17 30 23 21 14
Mode B 14 14 2 2 2 14 6 23 7 7 17 30 22 22 15
Mode C 2 20 2 2 2 2 2 2 2 2 29 2 6 21 2

Jan-Ola Dahlberg also reported that on 6 August, 1977, he heard a signal near 19.942 MHz while listening to Kosmos-929 on 19.954 MHz. The word values of the signal near 19.942 MHz were similar to those from Kosmos-1267 on this frequency. Perhaps Kosmos-929 also used this frequency. The short words sounded like "blips", so Geoff Perry initially called the three modes the "no blip", "three blip" and "eight blip" mode! In this way they were easily identifiable by simply listening with the "naked ear"!
 

Kosmos 1443

This craft had only a single telemetry mode on shortwaves. It turned out that the telemrty was extremly similar to that of Kosmos-1267, Mode B. The only difference was in word 7, where Kosmos-1443 had the vakue 30, while this word had the length 6 or 7 for Kosmos-1267.
 
Word 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Kosmos-1443 14 14 2 2 2 14 30 23 7 7 17 30 23 22 15
Kosmos-1267, Mode B 14 14 2 2 2 14 6 23 7 7 17 30 22 22 15

Kosmos 1686

This craft also had only a single telemetry mode on shortwaves. A recording from 3 October 1985 has been converted to values in the 1-32 range. It shows some similarities to that Mode B of Kosmos-1267
 
Word 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Value 22 22 2 1 2 21 15 22 7 7 20 29 22 22 15
References
  1. The Kettering Group, Observations of 1977-66A, Cosmos 929
  2. Perry, G.E. and Flagg, R.S., Journal of the British Interplanetary Society, 23, 451-464, 1970.

Back to "The TKS ferry for the Almaz space station"
Back to "Space History Notes"