Re: 'Official' TiPS elsets are very accurate

Jim Varney (sat_watcher@rocketmail.com)
Tue, 12 Aug 1997 21:16:54 -0700 (PDT)

---3432P@VM1.CC.NPS.NAVY.MIL wrote:
>
> A week or so ago there was a brief mention of the TiPS elsets as
published
> by the Naval Research Lab (NRL) at their website:
> 
> http://hyperspace.nrl.navy.mil/Tips/latest_tips.els
> 
> The NRL says that the accuracy of their elsets has improved greatly,
and I
> have seen no evidence to the contrary.
> 
> A claim was made here that the NRL elsets shouldn't be considered
reliable.


I made that claim based on a careful observation I made and a
discussion on SeeSat, both back in May.  Rather than fill up the
archives with repetition, see the comparison of Mike McCants' and
NRL's elsets at

http://www2.plasma.mpe-garching.mpg.de/sat/seesat/hyperold2/0479.html

Essentially I used a small refractor at 33x and timed TiPS' transit. 
I found Mike's elsets, then 24 days old, to be more accurate than
NRL's at only 2 days old.  The NRL's elset had a mean motion too large
and a negative drag term.

Back in May I exchanged e-mails with a few people at NRL, and they all
said they were tracking TiPS very accurately and were shocked that I
observed TiPS ten seconds late.  I tried to explain how SGP programs
need 'mean' elsets that are smoothed with short term periodics
removed, but they didn't seem to understand what I meant.

> Perhaps that was true in the past, but I offer another opinion based
on
> my recent observations, bolstered by some statistics published on the
> following page also hosted by the NRL:
> 
> http://hyperspace.nrl.navy.mil/Tips/Tb_drift.html
> 
> An excerpt from that page reads:
> 
>    "With increased SLR tracking during April '97 and the
>     addition of tracking from the ALTAIR radar on Kwajalein
>     Atoll, we were able to refine our drag parameters to the
>     point where we can now issue predictions of the orbital
>     motion which are accurate to a few tens of milliseconds
>     over a week.  


Yes, but that is largely irrelevant to SeeSat observers if NRL
achieves these tracking results by running professional numerical
integrators on  high-end workstations.  The trick is for them is to
correctly translate their highly accurate observations into mean
elsets compatible with SGP (Quicksat) and SGP4 (SkyMap, STSPlus, etc.).

> I've been using the NRL elsets for quite some time, and have found
them
> very accurate.  I track the tether at 40 power, with a 1 deg. field
of
> view, and the predictions have been within a second of USNO time. 
As a
> test, I used a 12 day old NRL elset last night and observed two
passes,
> right on time and track.  

That's great!  I hope your observations indicate that they've got the
right elset-machine cranking at NRL.

__________________________________________________
Jim Varney   121.398W  38.458N 8m   Sacramento, CA
jamesv@softcom.net, sat_watcher@rocketmail.com

Goodness is the only investment that never fails. 
--Thoreau
 

_____________________________________________________________________
Sent by RocketMail. Get your free e-mail at http://www.rocketmail.com