In message <3a3j8GAqJbJzEwqC@wingar.demon.co.uk>, Alan Pickup <alan@wingar.demon.co.uk> writes >With a few exceptions, tles are for a point close to (but not exactly >coincident with) a northbound equator crossing. I doubt if the object is >under active observation at that position - rather I imagine the tle is >an extrapolation from its observed motion at a later time on that rev. >However, to get reasonable mean motion and drag terms, observations >prior to the epoch must (usually) be included in the derivation of the >tle. Of course someone (are you there Mike?) might suggest that some >tles are "predicted", being based entirely on observations prior to the >epoch. Does anyone have the answer? > >Alan Yes, some of TLEs are "predicted" indeed. I guess such TLEs produces for some object which are hard to track due to their orbits and probably due to SGP4/SDP4 theory that is using on tracking stations couldn't be use to propagate correctly orbital motion for such satellites. I know that such procedure is normal for 13901,19688,20413,21139, 22653,23331,23632,23686,23842 etc. For example, currently you can found in OIG BBS database TLEs for 13901 with epoch 97073.26759740, for 23632 - 97072.95002340, for 20413 - 97071.56690600 despite of it is approx. 97071.0000000 now. To the point, deorbiting of Progress M-33 is scheduled today, Mar 12. Deorbit burn should take place at 02:35:00 UT, duration is 2 min 44.6 sec, dV=87.90 m/s, current mass of Progress is 5592 kg. Best regards, Vladimir Agapov.