geosynchronous flasher

Bruno Tilgner (
Mon, 6 Apr 1998 16:37:28 -0400

Tony Beresford wrote:

>I am in some doubt about identification as this was some 11min in RA Wes=
t, =

>and 1.2 degrees N of the 6913B ephemeris. Can somebody suggest the
>most accurate programme to use with such high satellites? The
>orbital elements for 6913B are now 19 days old, and I am uncetain
>of their validity.

The appropriate model would be SDP4 which takes into account earth
gravitational resonance effects and luni-solar perturbations. However,
for periods up to a week or two there is not much difference compared
with the SGP and SGP4 models.

The biggest source of error is in the TLE's. I would be highly suspicious=

of what USSPACECOM release for geosats, at least for those that cannot
be observed from US territory.

I have extensive astrometric measurements of the positions of the ASTRA
satellites during a period of 18 days, 12 hours a day. There are 7
satellites in this family, and the USSPACECOM TLE's are wrong for all 7
of them. The extreme case is ASTRA 1F where for several months the TLE's
gave a position up to 20 degrees too far east. They were obviously
tracking (or just calculating) the wrong object and missed the real one.

More on this (and other satellites) will be reported later this year.

Bruno Tilgner
Saint-Cloud, France