Custom Work

[Ceili]

I'm new to this. I'm creating a gaming system for a forum I'm working on, and thought Celestia would be great to show the universe it's set in, so you could have a visual representation of all the working elements.

I'm currently trying to track down any programs that'd help me create the files to add stuff to Celestia. I got one called System Maker, however it always locks up when I try to load the solarsys.ssc file, and any other files. It gives me some runtime error, so I think the program is bad.

Anyway, I'm not sure if this is possible. I wanted to know if it's possible to create a section in space where there is two stars (suns) that are a set distance apart. And the primary planet of the region has a unique orbit that is setup like an infinite symbol (where it starts directly between both stars, swings out on a path around one star, comes back to the center point, then swings around the other star, and continues along this path.

I'm not sure it can be done with this program. If so, could anyone more knowledge help me figure it out? I'm just guessing as it stands, I think it could work, if the stars were just far enough apart, so that when the planet got to the center point, it could break the orbital path of one star, but get caught in the path of the other star. So it's like it keeps getting passed back and forth.

Also, could anyone point me to some good programs that would be useful in creating the stuff required in the various file formats used in Celestia? Or do I need to learn how to write them by hand?

[Ceili]

I believe I partially know how to start this sequence.

In looking through some files, I've learned that .xyz files determine an orbit that's a bit off, like for the satellites. I saw about Galileo and Cassini, how it shows the various points they go to.

So I think it would be possible, if I had the two stars created first and then do the planet in question like this:

First I name the planet and attach it to one of the stars. Rather than giving it an elliptical orbit, I link it to an .xyz file. Then I have it set to swing around the first star, then around the second, and then back to the first.

The only problem I see is that using an .xyz gives it a start and a finish. I don't think I can setup an infinite loop of points for it to cross.

But that should be ok, if I were able to just give it the orbit pattern for say 10 full turns (so that it's went around each star and returned to the center point exactly 10 times.

[Chuft-Captain]

You may also be able to do this with a Timeline statement containing 2 separate definitions of EllipticalOrbits. You'll have to set the Epochs of each orbit such that the end of one orbit meets the beginning of the other, but unfortunately you'll have the same issue with trying to make a Timeline repeat.

CC

[Guckytos]

A general hint on binary systems.

You can also define system barycenters.
And the have both stars rotate around it. (Or not have them rotate around it, up to you)
You can also place any object in orbit around this barycenter. (i.e. a planet on an xyz file)

On how you would do with the repeat cycles, I can't give you tips.

Guckytos

[BrianJ]

I wanted to know if it's possible to create a section in space where there is two stars (suns) that are a set distance apart. And the primary planet of the region has a unique orbit that is setup like an infinite symbol (where it starts directly between both stars, swings out on a path around one star, comes back to the center point, then swings around the other star, and continues along this path.?

This is an interesting idea. I'd be surprised if it hasn't already been created for Celestia by someone, but I can't find anything on the Motherlode or by googling variations on "celestia binary star planet system".

So, a few thoughts:
To take a simple case, your binary stars will be orbiting each other around a common barycentre in circular orbits. To simplify even more, the stars could be the same mass, with the barycentre exactly mid-way between them. This thread could be useful:
viewtopic.php?f=15&t=13528

I'm not sure that a "figure-of-eight" orbit for a planet would be stable or even possible in reality (we're into some heavyweight celestial mechanics here, way beyond my mathematical capabilities, but I'll continue anyway), to generate a "figure-of-eight" orbit in Celestia you could use the "ScriptedOrbit" function:
http://en.wikibooks.org/wiki/Celestia/T ... iptedOrbit
The sample code given for the "lissajous" orbit could be pretty close to what you want.

Here's a couple of links to items relevant to the subject:
http://www.atlasoftheuniverse.com/orbits.html
http://www.dam.brown.edu/people/mumford ... w7/HW7.pdf

OK, I think I might have a go at creating a binary star system with a planet in a (vaguely plausible) "figure-of-eight" orbit. See how far I get before my brain melts.
I'll let you know how I get on ;-)

Cheers,
Brian

[Chuft-Captain]

I'm not sure that a "figure-of-eight" orbit for a planet would be stable or even possible in reality

Brian, I read somewhere that it is possible for a group of stars to follow a figure of eight orbit around each other in a stable fashion. I can't remember exactly where, but I remember that it had some JAVA applets demonstrating some possible configurations.
I'll try to track down the site again if you like.

EDIT: Not the site I saw, but this has some background on it: http://www.thefreelibrary.com/Orbiting+ ... a061617781

Animation here: http://www.atlasoftheuniverse.com/orbits.html

More on the subject:
http://merganser.math.gvsu.edu/david/re ... node8.html

and GOOGLE is your friend http://www.google.co.nz/search?hl=en&as ... =&safe=off


As for planets doing this around a couple of stars, that's quite another story I think.

CC

[BrianJ]

Thanks for the links CC, lots of interesting stuff there.

Looks like a "figure of eight" orbit for a planet between two stars can be stable.

Whether the x,y,z position can be specified with respect to time using a (relatively) simple equation (like calculating state vectors from Keplerian elements), I'm not sure. I think you'd end up turning Celestia into a numerical-integration-gravity-simulator. I dunno, I'm not a mathematician.

A ScriptedOrbit "lissajous" type function might be a reasonable aproximation though.

Surely, someone must have done something like this for Celestia already?

Cheers,
Brian

[Chuft-Captain]

Here's the original site that I saw about this: http://burtleburtle.net/bob/physics/eight.html

Turns out that I did re-find this site yesterday, but I didn't recognize it because I had JAVA disabled in my browser, so no applets appeared.

CC