In the Webmaster's last TLE report, there was this: "Launch time for STS-107: 16 JAN 03 @ 15:39:00.075 UTC"<br><br />On this site it said: Lift off time (UTC) : 2003/016/15:38:59.958<br><br />I know the difference is small but does anyone know why there is a difference?

Alexander,<br /><br />This is exactly what I received from Cape Canaveral Network Operations..<br /><br />Columbia and the STS-107 Crew departed Pad A 3016/15:39:00.075. The International Designator is 03003A and the Object Number is 20647. First<br />Vector following.<br /><br />However the launch time is used for calculating the rev number, so a small discression wont affect the TLE in anyway.<br /><br />I am guessing (as I do not know their source of 'launch time') they refer to launch time as actual SRB ignition, rather than physical liftoff.<br /><br />Our data courtesy of Cape Canaveral Network Ops is definately actual liftoff.<br /><br />If your interested, let me know and I can try to find out!

Well...<br />My interest in all-things-space is a bit obsessive so if you have the time to check that fairly obscure fact, sure!

I could be wrong here, but there usually is two actual time of lift-off, and it all depends on who you ask.<br /><br />The first lift-off time is the time at which the shuttle computers declare the lift-off time at command (usually the time at which the hold-down bolts are commanded blown) and the second is the actual physical time at which the SRB's leave the bolts on the pad.<br /><br />The computers guys use the first, while the rest of the real world use the later time of physical lift-off.<br /><br />Again, I could be wrong...

Well, we update the TLEs way before any difference that small could matter; even to people that need tracking accuracy, right?

Alexander,<br /><br />The liftoff time makes not a single difference to the TLE. The TLE is computated using SGP4 using X, Y, Z and X dot, Y dot, Z dot and epoch time of the state vector.<br /><br />Then the TLE is calculated with the spacecraft drag co-efficient. The launch time then comes into play simply calculating the REV number.<br /><br />Simply explained the state vector is simply a Earth Centered Inertia or Earth Fixed Greenwich 'snap shot' of absolute velocities and heights at a given instant with reference to the earth, there is NO reference to REV number.<br /><br />So even a minute inaccurancy in the launch time will not effect a object orbiting every 80 to 90 minutes.<br /><br />I guarantee our vectors from the cape used to generate TLEs are within 30 meters of the real position (depending on your clock accurancy and age of the TLE), infact quite often the cape send us differences which we may trim or increase the drag co-efficient. The differences can only be seen by using a tracking station and comparing the downlinked telemetry from the shuttle.<br /><br />Hope this helps you?

That was a great help, David!