Designing detailed Astronomy?

[Greason Wolfe]

Greason Wolfe: Cool program- Thanks! I had a play about with it. Need to explore more, but that could be just what Im looking for now.

No problem, glad to hear that it might be useful to you. I found it by accident one day when I was researching some stuff for a javascript application I was writing for some personal astrophysics stuff. I thought it was a pretty sharp program and I'm going to have to go get it again one of these days.

[Talroth]

This would be so much easier if my last physics class that covered the math I actually need for designing the simulation wasn't over 4 years ago,...


But by playing with fairly low density moons (2/3 that of our moon) and the orbitsimulator program Greason Wolfe posted, I've managed to get 7 moons ranging in visible size from 50% to 150% of our current moon. Sadly that simulation has several issues in step size, as patterns highly depend on speed the simulation is run at. That is, running ultra high speeds produces very stable systems, where as lower speeds have far more wobble to them.

Mind you, I didn't actually try it with standard density moons, and it would have produced several small moons that would travel across the sky at fairly high speeds. I'll try to find the time to play with these problems over the next week or so and see if I can't come up with some better modeling tools.


*Edits*
Very odd, systems can die very violently after displaying very steady patterns for several thousand years. I'm still thinking it might be an issue with their stepping algorithm, as running a burst of high speed, then restarting and waiting for it to run at a slow speed produces very different positions. Anyone else that uses the program noticing this?

[EDG]

Looks like I joined just in time .

Orbit Simulator (aka Gravity Simulator, or GravSim) is a godsend for this sort of thing, I've used it a lot (I post on their forums as Mal).

Celestia is also a good visualisation tool.

Gravity Simulator is a tricky one to use properly, but is pretty much unparalleled and incredibly useful. The timestep size has to be much smaller than the smallest orbital period involved in the simulation to get accurate results - I think I usually go for 5 to 10% of the shortest orbital period. So if you have a moon with a 10 day orbital period, you'll need to have your step size at a maximum of 32k-64k seconds. The step size does affect things a lot since for larger step sizes two bodies can influence eachother for only one or two steps, whereas for smaller step sizes they'd influence eachother for many steps - and that changes everything that comes afterwards.

So you need to run systems slowly in GravSim - and if that means running it for a few hours (or overnight) then that's what you'll have to do. Fortunately you can save a data file, and there's a really good visualisation tool for the results available that one of the users wrote and has posted on the gravsim discussion boards too.