Satgen 214  Doppler Part 7   by GM4IHJ                    1st May 1993
BID of this msg is SGEN214  Please use that BID if you retransmit the msg

Talking to G3IOR , he brought up the subject of the sudden rise in
frequency of the RS12 beacon as the sat approached the LOS loss of signal
at horizon on several recent orbits. The reason for these jumps is very
involved , so I have told Pat I will perhaps do a Satgen bull on it one
day, but first I must lay a foundation by discussing " ordinary doppler"
from RadioSports. 
Is the frequency shift we hear a true representative of the changing line
of sight angle to the satellite ? No it is not. Signals which start out
towards us from the satellite never get to us. The 3 ( D, E and F) layers
of the ionosphere bend the signal path in an irreglar manner and at low
angles of elevation as we acquire or lose the satellite the path the low
elevation signal has to take through the thick, turbulent atmosphere can
also twist the signal about a great deal.
All you need to see this happening is an audio freqency meter. Match it to
the phone or record socket of your HF receiver, transmit a constant mark
to the sat and tune the satellite audio to show about 1200 Hz on the
meter. Set the meter to report frequency ever 20 seconds or so , and write
down the signal frequency you read. Do Not tune the receiver , just record
the descending frequency.
A typical run on RS12 starting at AOS horizon acquisition was :- 
  F Hz  Hzchange      F Hz  Hzchange      F Hz  Hzchange    F Hz  Hzchange
1 1200    na       8  1230     - 50   15  1200    - 10    22  1080    - 20
2 1150    - 50     9  1250     + 20   16  1190    - 10    23  1050    - 30
3 1200    + 50    10  1250        0   17  1170    - 20    24  1000    - 50
4 1240    + 40    11  1220     - 30   18  1160    - 10    25   960    - 40
5 1240       0    12  1210     - 10   19  1150    - 10    26   910    - 50
6 1220    - 20    13  1220     + 10   20  1140    - 10    27   840    - 70
7 1280    + 60    14  1210     - 10   21  1100    - 40

Taking a reading every 20 seconds, it is clear that the first 14 readings
show a very turbulent picture consistent with the signal coming through a
thick low elevation path in the atmosphere. We expect in theory that the
signal will drop very slowly, it clearly does not. Then in the last 13
readings the pattern of decreasing doppler we have been expecting all
along begins to exert itself but even here it is not a smooth descent .
The rate of change in these right hand columns is of the correct order ,
but it is far from smooth.
The satellite is not dopplering in these random jumps. What is happening
is that the bent path we hear it on, at the start is bent away from us and
quickly lost and we hear it via another path which has been bent towards
us. Each different path left the satellite at a different angle to its
course through the heavens . So as actual doppler on a path = max doppler
x cosine of the angle this particular track had with respect to the
satellites bow, on leaving the sat. As we change path , we change angle on
bow and hence doppler frequency. Between dead ahead and sideways on the
change can be as much as 1 kHz (2way doppler). As the results in the above
table show we are not getting this change every 20 seconds , but we are
getting slight changes of path / angle , all the time due to atmospheric
and ionospheric anomalies. Indeed the table above shows that this pass
encountered a very disturbed atmosphere during first 4 mins and a slightly
disturbed ionosphere during the second 4 minutes. 73 de GM4IHJ@GB7SAN