Re: [amsat-bb] Accuracy of tracking programs

[John Wright <g4dmf@xxxxxxxxx>]

Further observations:


At 23:31 07/06/01 +0100,I wrote:
>I have 2 computers in my shack,that are networked. Both are mid to high 
>range pentiums.
>On my main PC, ( this one ) I run LogSat 5.2 on my other PC I run Nova. I 
>have two programs,
>because each is better at one or another thing.
>
>LogSat is great for forward predictions,and is prettier, but I never did 
>get it to drive my KCT card.
>Nova is not so good at forward predictions, not so easy to use, but drives 
>the KCT great.
>
>Now here is the point, Nova consistently puts LEO satellites about 1 
>minute in front of LogSat,
>this makes no real difference to the high orbit satellites, but *is* 
>significant on LEO's

I though what was different between yesterday being one minute out, and 
today being spot on. The only thing changed was the KEP's. So I thought a 
little more... the AGE of the KEP's. Yesterday I was using 8 day old KEP's, 
today I'm using 1 day old KEP's. Could it be cumulative errors creeping 
into the calculations with truncation or rounding having a compound effect, 
making calculations less accurate as the KEP's aged?

So I ' artificially ' aged the the KEP's by advancing the machine clock. 
Below is the outcome.
Both programs run on the same machine, using the same observation location 
data, and the same KEP set.
KEP's for the ISS dated day 158 June 7th 2001 the elevation given at the 
nearest whole minute after AOS for a visible late evening pass

                 LOGSAT  NOVA            Difference
8 june          0.6             0.7             0.1
9 june          0.4             0.7             0.3
10 june 0.4             1.0             0.6
11 june 2.7             3.5             0.8
12 june 0.5             2.0             1.5
13 june 1.7             3.2             1.5
14 june 0.7             3.4             2.7
15 june 0.5             2.0             1.5

The errors on 13th , 14th and 15th June actually amount to 1 minute 
difference in AOS of the ISS.

So it looks like compounded rounding errors are generating differences in 
the computations. It still remains to be seen which program is most 
accurate, or if both are wrong to some extent. It's not out by a great lot, 
but enough to make a difference. I think the outcome of this thread 
probably has to be that you must use up to date KEP's, even over 1 week 
errors do creep in. If I remember, I will leave these KEP's unchanged and 
check on the 15th visible pass, just to see if one or the other is more 
accurate.

John



____________________________

      G4DMF  QTHR   IO93HA
      Amsat-UK Member 5372
          g4dmf@amsat.org
      http://www.g4dmf.co.uk
____________________________

----
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org