Results from TLE & SGP4 propagation - don't seem right & with interpretation

[thor36]

Greetings all !
I have this problem with SGP4 propagation, that I hope someone can help me out with.

I acquired a TLE of the ISS from internet and used the C++ SGP4 propagator to compute future position and velocity vectors of the Station. I am unsure about some aspects of results though and would like to get some explanations.

The TLE data looks like this :

1 25544U 98067A 14296.89020256 .00021128 00000-0 36676-3 0 2914
2 25544 51.6464 181.6097 0002420 294.7068 172.8232 15.50795889911298​

The settingss I use for the propagator are :

opsmode = i (improved)
typerun = m (manual)
typeinput = m (mfe - minutes from epoch)

I set SGP4 to propagate for 120 minutes after the initial moment in 10 minutes steps. The results are stored in 13 rows ( one for each computation from Tstart to Tend, once every 10 minutes ). It would look really messy to paste the results here, so I will only copy the last line.

R(x) = 4628.43055298
R(y) = 2891.92246591
R(z) = -3530.26416981
V(x) = -5.504313283
V(y) = 3.379764650
V(z) = -4.417716694
Time = 1995 12 61 23:31:53.501202​

So, my questions now :

1. Position
   
   I reckoned calculating magnitude of R will result in distance of ISS from the Earth center ( or altitude, if you subtract Earth radius ). But calculating magnitude of this final set of data (posted above) results in a magnitude of approx. 6500 km, which is obviously too small value since we know ISS doesn't orbit the Earth at an altitude of about 100 km. Moreover, if I propagate just two more time steps ( +20 min ) I get error 6 - satellite has decayed. Why such a quick drop in altitude after propagating 140 minutes in future, which is a little more than an orbit and a half for ISS ?
2. Time
   
   The last column is supposed to give the time of that particular computation. TLE epoch cell ( 14296.89020256 ) says the year is 2014 at 296th day of the year. But results seem to set the time at 61st December 1995 . Why such result ?
   

I may be missing something very fundamental, or cannot interpret results properly. I would like to ask for explanations / clarifications that will help me to make use of these ( and all subsequent ) results gained by TLE and SGP4.

Thank you very much and kind regards
T

 

[thor36]

After doing more "experiments", it seems that the particular TLE I used was erroneous. This was my first testing of SGP4, and many things could be wrong with the code itself, so obviously, the TLE I got from online source wasn't the first thing I blamed for the errors. It seems that even generally reliable online TLE sources can occasionally produce an unfortunate bad instance of a TLE set.