Sat Track

[Andre Leao]

Preserved...



https://planetasinclair.blogspot.com/20 ... k-mia.html

✓ PROCESSED

New ID created (ID: 39043). TZX and screen added. DH 02/01/22

[smurphboy]

Love this. It has a phone number to ring for the orbital updates! As a licensed amateur radio operator, I quite fancy updating this for the next / esxDOS machines using .http to pull the orbit data.

[R-Tape]

Love this. It has a phone number to ring for the orbital updates! As a licensed amateur radio operator, I quite fancy updating this for the next / esxDOS machines using .http to pull the orbit data.

I'm glad this still makes sense to people here! This is uMIA from G4RWT. I didn't know if we even still had amateur radio. All I can tell from this is that it somehow tracked old, now kaput, satellites Radio Sputnik 5 - 9 and Oscar 1 (Uni of Surrey).

Any chance you could give a duffer's guide guide to how this program worked, how would it have been used, and in what way would the Speccy have been involved?

[Guesser]

You can predict a satellite's position from its orbital elements and lots of maths (computers are good at maths ).

This works well for things like the moon, but small things in low orbits get pulled around a lot and get slowed down by the atmosphere, so the orbital parameters change hence needing updates.

[Guesser]

All I can tell from this is that it somehow tracked old, now kaput, satellites Radio Sputnik 5 - 9 and Oscar 1 (Uni of Surrey).

RADIO 5-9 are still in orbit, OSCAR 9 (not 1) re-entered in 1989

[Andre Leao]

eheh, great you both enjoyed!

[R-Tape]

You can predict a satellite's position from its orbital elements and lots of maths (computers are good at maths ).

This works well for things like the moon, but small things in low orbits get pulled around a lot and get slowed down by the atmosphere, so the orbital parameters change hence needing updates.

Cheers. I'm slightly the wiser. I assumed this involved hardware rather than 'just' calculating from inputted data, but after you explained I see there are no INs or OUTs in the code.

So could this still be used? And how well would it work compared to contemporary alternatives (don your Spec-friendly tinted glasses please). Having said that this copy of the program exits to BASIC a lot, so bugs may have come in during copying (or been there in the first place!).

RADIO 5-9 are still in orbit,

So are they still active, or they just haven't re-entered yet? And are they Radio Sputnik?

OSCAR 9 (not 1) re-entered in 1989

Sheesh another one I'll never live down. It's almost as bad as when I suggested that Class 45 locos were on the Scottish line

[Guesser]

So could this still be used? And how well would it work compared to contemporary alternatives (don your Spec-friendly tinted glasses please).

I was just trying to figure out the orbital elements to update it
Compared to just looking on one of the million satellite tracking websites that give you a pretty map and details for your exact location... not very well

So are they still active, or they just haven't re-entered yet? And are they Radio Sputnik?

They are amateur radio satellites and no longer functioning, so tracking them's a bit academic, though I suppose you could hack new names into the menu and enter the orbital data of newer ones, but that would be extremely pointless and nerdy.

[Guesser]

hence needing updates.

(that and the software thinking year 23 comes before 83 for some silly reason, tsk eh!)

[R-Tape]

extremely pointless and nerdy.

That's exactly what I come here for. Get to it man!

[R-Tape]

(that and the software thinking year 23 comes before 83 for some silly reason, tsk eh!)

Heh yep. It looks like this was made in 1983, so G4RWT were either fans of the millenium bug or they'd taken Orwell's 1984 to heart.

[Guesser]

To be fair, the orbital elements are still specified by a two digit year now. Since no artificial satellite existed before 1957, earlier numbers must be in the future and it's all good until 2056 when Somebody Else will have to Deal With It

[smurphboy]

They are amateur radio satellites and no longer functioning, so tracking them's a bit academic, though I suppose you could hack new names into the menu and enter the orbital data of newer ones, but that would be extremely pointless and nerdy.

I fully intend to do this. perhaps using .http from esxDOS for the Next or Spectranet machines...

[smurphboy]

G4WRT has a profile on QRZ.com, so may well still be with us...

[smurphboy]

Nowadays you'd want ISS, SO-50, AO-91, AO-73 and QO-100, although QO-100 is pretty easy

Or the Weather Sats like Meteor-M2 or NOAA 19

[Guesser]

Can anyone get this to actually make correct predictions? I spent a while last night trying to enter keps for the ISS etc and it would get some handful of the times vaguely close but wildly wrong bearings and elevation, suggesting it's more luck than science.
I don't really want to step through all the calculations in BASin to reverse engineer it
You've thoroughly nerd sniped me here you basts

[smurphboy]

My guess is that it doesn't have leap years quite right... so try + - 1 day and see if that is closer?

[Guesser]

Would probably be easier to just start over and write a new one in C that takes two line elements for all the current sats and works for any location... Aaargh I mustn't!

[Guesser]

My guess is that it doesn't have leap years quite right... so try + - 1 day and see if that is closer?

I can't see how that should matter (or rather that's a problem for later) as it can't predict the orbits immediately after the reference it's been given. Whatever error it has in its date/time frame should be the same in both the orbit and the output passes. Unless I've completely misunderstood how it works

[Guesser]

I think it's got orbital parameters or something hard coded in - it does different things with the RS and OSCAR sats to decide if they're "out of range" and calculate azimuth etc. so changing the period to the ISS was never going to work.

But it also doesn't work for the four remaining original satellites either
Would have to go through line by line working out what all the variables mean and then checking its calculations. That's too much work for me

[smurphboy]

I might just make a front end to something like the n2yo.com API. Although I'd have to bridge https to http for the speccy.

[Guesser]

That's cheating though!

You can download the TLE keps in plain http from http://www.amsat.org/tle/current/nasabare.txt

[smurphboy]

Good luck converting TLE's into predictions in reasonable time

Although that wouldn't be 'as cheaty' - I should get my old text books out...

[Guesser]

Good luck converting TLE's into predictions in reasonable time

Why? That's surely what this software attempts to do, in a roundabout way.
It's not exactly exactly the same format, but as far as I can make out it's the same data. The TLE format is derived from punch card data. Calculating keplerian orbits shouldn't tax a modern computer like the speccy!

[smurphboy]

Calculating keplerian orbits shouldn't tax a modern computer like the speccy!

Keplerian orbits aren't what you use, they only model two bodies (Earth and the satellite) and rapidly depart from the actual orbit. You need model it as a three body problem and estimate drag. This means numerical methods and I'm trying to find an approach that might work on a speccy.

The standard perturbation algorithms used like SGP4 / SDP4 won't really be computable on a base speccy in real time.

I have found a paper that models drag but might work on a slower machine - like a Spectrum Next.