UnIDs, unseen, elsets, errors

Walter Nissen (dk058@cleveland.Freenet.Edu)
Tue, 1 Aug 1995 10:10:15 -0400

Jeff Hunt, 
>      That was flight 345 from Cleveland to Charlotte, N.C.:-)  ! 
> Seriously though, I could not find an exact match for what you described. 
> I ran simulations on all 4600+ elements maintained on my program.  Please 
Thanks very much. 
> Three additional possibilities were 12331, 19445, and 6845 for the site 
> and time mentioned.  What's your assessment? 
I agree and don't think these match very well either, especially the 
Molniya, because its motion was REALLY slow.  If there was no known 
satellite moving like what I saw, I would guess that it would likely have 
been an airplane, even if one whose motion appeared strikingly like a 
Bjoern Gimle, 
I have attempted to see a couple of fairly unfavorable northwestern passes 
of C* 2264 on the past two evenings and was not able to locate it.  Maybe 
you are hogging my light. 
I have another mystery observation to pass along.  When I observed 
DMSP B5D2-5 
1 20978U 90105A   95202.13708376  .00000094  00000-0  48945-4 0  2736 
2 20978  98.6387 270.2562 0080404  69.6375 291.3435 14.32564057242175 
at 950731 0311 from -81.8637, 41.3735, 256m, I found two 4th-magnitude 
objects moving along nearly the same track at the same speed very much as 
the leader and the trailer do in a NOSS 2-n cluster, except that they were 
about 7 degrees apart.  I looked so extensively for a sidecar or for 
objects ahead or behind these that by the time it penetrated my 
consciousness that it really would be desirable to get an astrometric 
position for at least one of them, I couldn't find any suitable stars. 
When I attempted a similar observation at 950801 0238 I found only one 
object and no trace of the second.  Nothing in my 1000 elset file matches 
even remotely.  Internal reflection seems impossible.  Any ideas?? 
Ed Light, 
I put together a file of 55 elsets from STS70 and tried to make some sense 
out of the various ndot2's and MM's, looking for an early elset that 
successfully "predicted" later behavior.  No luck.  Have you or anyone 
else tried to do so? 
If not, I'm going to have to ask Mike McCants or Victor Slabinski or 
somebody to tell me exactly how to calculate the true, or empirical, MM 
from successive MA's.  (I guess the theory here is that if the ndot2 is 
bad, you can try relying on the successive MM's.  If they are bad, you can 
try relying on the MA's.  Always seeking something that is true). 
Bill Krosney, 
> Of particular interest are the comments about "ndot2" values.  What exactly 
> are these values?  Are they present in the TLE listing (i.e. are they the 
> drag term)?  Or is it a derived value from info contained within the TLE? 
> In any case, I would be interested in a brief explanation how one uses 
> a ndot2 value along with other parms to deduce a probable re-entry date. 
> And what are "exceptional" values. 
I'm not sure where the best source for the explanation of the TLEs would 
be.  I think TS Kelso posts it to sci.space.news once a month.  He has it 
on his Celestial BBS.  I'm pretty sure it is in the SeeSat-L archive. 
Some version or another appears in the documentation for most tracking 
programs.  It must be available from the OIG BBS.  OIG used to provide 
them on paper. 
Now, concerning drag factors and "exceptional values": 
Bjorn Gimle, I, and others have speculated about the error rate in elsets. 
Please understand that these comments are definitely NOT intended as OIG 
or SPACECOM bashing.  I've often commented that if I were responsible for 
reporting on 8000 objects, I'd no doubt emit some real howlers.  But 
errors are a fact of life and we users are entitled to try to figure out 
what we are up against. 
                                         drag factors 
                                    ndot2   ndotdot6   bstar 
LandSat 5 
1 14780U 84021A   95202.12845011  .00041820  00000-0  92164-2 0  2942 
2 14780  98.1086 257.6508 0006168  99.6752 260.5320 14.57184412605574 
1 14780U 84021A   95209.13252142 +.00000386 +00000-0 +95407-4 0 03237 
2 14780 098.1084 264.4695 0006254 080.6787 279.5117 14.57162352606595 
C* 1763 
1 16860U 86052A   95202.12244895  .00251447  00000-0  83803-1 0  8507 
2 16860  74.0383 100.2248 0032314 338.7127  21.2847 14.35434183471988 
1 16860U 86052A   95207.69812214 +.72702812 +00000-0 +19483-2 0 08599 
2 16860 074.0367 089.9999 0032094 325.6790 034.2288 14.35371278472789 
Helios A 
1 23605U 95033A   95202.21741673  .00005193  00000-0  10021-2 0   417 
2 23605  98.0706 138.3095 0001881 110.6074 249.5326 14.63877293  1983 
1 23605U 95033A   95209.18944942 +.02999963 +00000-0 +51951-0 0 00609 
2 23605 098.0713 145.1485 0001898 070.6340 289.4519 14.64301080003008 
I obtained the second elset of each pair from OIG yesterday.  The earlier 
ones were already in my 1000 elset file.  I think it is safe to say that 
not all of these ndot2s correspond to reality. 
I'm not the only one who has looked skeptically at the various elsets 
emitted for Mir and its various components, trying with some difficulty to 
reconcile the inconsistencies.  The worst Mir ndot2 I've seen was an 
eye-popping .55.  I think Bart may have something about this in the 
forthcoming Flash. 
Ted Molczan must have examined the ndot2 situation with some care before 
he began averaging them for his file, but I haven't seen any extensive 
analysis from him. 
Theoretically, ndot2 can only be valid for unpowered flight, and hence, 
ndot2 is, strictly speaking, undefined, if you are trying to "predict" 
past a burn.  Depending on your time frame you might choose different 
artificial values. 
I'd appreciate any comments about these elsets. 
Walter Nissen             dk058@cleveland.freenet.edu             216-243-4980 

Astronomy is looking up.