You Spin Me Round
April 24, 2015 4:22 AM   Subscribe

Here's a neat browser toy where you can play with gravitation interaction and make planetary orbits...or horribly destabilize them.
posted by polywomp (44 comments total) 51 users marked this as a favorite
 
(Requires Flash)
posted by ardgedee at 5:17 AM on April 24, 2015 [1 favorite]


What's flash?
posted by Samuel Farrow at 5:28 AM on April 24, 2015 [1 favorite]


Click the Huge or OMFG size at the bottom left, and checkmark Paths. Place a big object, then try to orbit various sizes by dragging a blue line and releasing it.
posted by jjj606 at 5:29 AM on April 24, 2015


Fun!
posted by sidereal at 5:41 AM on April 24, 2015 [1 favorite]


You can type in any arbitrary number in to the mass box, but anything larger than OMFG size seems to make the system incredibly unstable, either by sucking in all the other bodies or launching them far, far away.

Binary systems seem to be difficult to keep in one place. I can get two OMFG bodies to orbit each other, but the pair tend to wander. Getting a stable orbit for a planet with a moon also in a stable orbit is also pretty hard.
posted by backseatpilot at 5:54 AM on April 24, 2015


Oh that proto disk is awesome!
posted by Bugbread at 5:59 AM on April 24, 2015 [2 favorites]


Yep, I'll be playing with this all day.
posted by Foosnark at 6:09 AM on April 24, 2015


I so want to be able to make a moon that revolves around a planet that revolves around a sun, but man, that's incredibly impossible.
posted by Bugbread at 6:15 AM on April 24, 2015 [1 favorite]


> Oh that proto disk is awesome!

Came here to post exactly that. What I can't tell though, is if the objects already have some velocity applied or if the spinning of the disc is an effect of gravity on a disk of non-uniform density, or if it's something else entirely. Anyone know what's going on there?
posted by noneuclidean at 6:16 AM on April 24, 2015 [1 favorite]


I think the spin might be derived from the collapse, since some objects end up orbiting backwards. I'm not sure.
posted by polywomp at 6:20 AM on April 24, 2015


This was posted on metafilter before (not as an FPP, but I think as a comment in an ask somewhere... ahha, here) and it has kept me idly amused every now and then for the last 6 months.
posted by Just this guy, y'know at 6:22 AM on April 24, 2015 [1 favorite]


Neat! Been running this in the background for a while now.
posted by carter at 6:24 AM on April 24, 2015


I'm pretty sure they all start with a velocity at a right angle to the center of the disk.
posted by Bugbread at 6:24 AM on April 24, 2015


This is so nice!

It's wonderful and meditative to build a static system, turn on view paths and watch the slow mathematical dance of these gentle spiraling bodies play out on the screen, all of them oblivious to the fact that you are about to drop an OMFG right in the middle and destroy them all.
posted by dirtdirt at 6:31 AM on April 24, 2015 [7 favorites]


If only there were an app like this that modeled in 3 dimensions, that allowed you to fine-tune the properties of each object, save your creations, speed up and slow down time, and really play with the evolution of a sandbox solar system complete with moons and rings and stuff.

Oh hey, there is! Check out Universe Sandbox.
posted by General Tonic at 6:53 AM on April 24, 2015 [6 favorites]


It's neat, you can build an elliptical orbit by putting an object 1-2 orders of magnitude larger at a static point and starting a smaller object on a course perpendicular to its ultimate orbit. The smaller object cause deflection in the large object on the close end of the ellipse, which creates rather lovely spirals.
posted by graymouser at 6:55 AM on April 24, 2015


It is possible, but difficult, to have a central sun, a larger planet and then a smaller moon around that planet. I've been trying to have a planet/moon system around a binary star, but so far no luck.
posted by Just this guy, y'know at 7:46 AM on April 24, 2015


The guy who wrote this could sell art prints of "proto disc" with trails enabled.
posted by Alterscape at 7:56 AM on April 24, 2015


Can I do that gravity trick with, say, only Indiana?

(I'm just wondering...)
posted by mule98J at 8:13 AM on April 24, 2015


Well *I* won't be getting any work done today.
posted by Ben Trismegistus at 8:16 AM on April 24, 2015


You know what would be fun? An accretion disk simulator.
posted by symbioid at 8:50 AM on April 24, 2015


What I can't tell though, is if the objects already have some velocity applied or if the spinning of the disc is an effect of gravity on a disk of non-uniform density, or if it's something else entirely. Anyone know what's going on there?

Finally a question that you can answer with physics!

The key is the angular momentum of the system. If you let the system pan out for a while, you wind up with a number of bodies (say, 20-40) that all rotate in the same direction (i.e. clockwise or anti-clockwise). Thus, the angular momentum of the system is non-zero. But since the total angular momentum (=integral over all bodies) is constant, this means that it was non-zero to start with, which means that the objects already had some velocity applied to them in the beginning. (Well, theoretically, at least one of them had, but I think it is apparent that most or all of them have).

Actually, playing around with it a little more, it seems that the rotation of the resulting set of bodies is always counter-clockwise. Also, it seems that the initial velocity depends on the distance to the central star.
posted by sour cream at 9:04 AM on April 24, 2015 [2 favorites]


Turning trails on is like making Spirographs. Vector a Huge one diagonally, then pick a spot and generate a constant stream of Tinys from there.
posted by zinon at 9:14 AM on April 24, 2015


Actually, playing around with it a little more, it seems that the rotation of the resulting set of bodies is always counter-clockwise

I too noticed that it always spins left round, unlike a record, baby.
posted by aubilenon at 11:15 AM on April 24, 2015 [2 favorites]


Ooh, this is nice.
posted by cortex at 11:36 AM on April 24, 2015


Its all fun until you realize we are just a "tiny" dot spinning around a "OMFG" and then you spend the rest of the day huddling in the corner waiting for the inevitable......
posted by inflatablekiwi at 12:03 PM on April 24, 2015


Actually, playing around with it a little more, it seems that the rotation of the resulting set of bodies is always counter-clockwise

If you run it from a computer in the southern hemisphere, does it go in the other direction?
posted by eykal at 1:01 PM on April 24, 2015


So is the precession of one tiny object orbiting a bigger object caused by some physical process they are simulating or is it some numerical artifact?
posted by Zalzidrax at 1:16 PM on April 24, 2015


"There are a variety of factors which can lead to periastron precession, such as general relativity, stellar quadrupole moments, mutual star–planet tidal deformations, and perturbations from other planets."

Apsidal precession
posted by intermod at 2:42 PM on April 24, 2015 [1 favorite]


I was just looking at that page too, intermod.

The closer you get the satellite to the larger mass, the more pronounced the precession becomes. Additionally, the orbit also appears to be decaying.

I'm fairly sure this is just a Newtonian simulation, and the precession and decay are both due to the fact that it's just an approximation (Euler's method, it says on the top of the screen). There's a very small discussion of various ways of approximating this in the reddit thread.
posted by aubilenon at 3:06 PM on April 24, 2015


(Euler's method, it says on the top of the screen)

Hah! That's about par for the course for my reading comprehension.

Though it looks like he means the Symplectic Euler method according to the reddit thread, which shouldn't decay in a two body system...
posted by Zalzidrax at 3:20 PM on April 24, 2015


This might be a good place to note that voar.io - previously - is back online! And it doesn't take Flash.
posted by Pronoiac at 10:38 PM on April 24, 2015 [3 favorites]


I found this one fun for a short while; but it bogs down horribly with larger number of particles (understandably so I guess). And I think also the side-effect of that bogging down is that the integration gets quite inaccurate: particles pass through each other, particles get randomly fired out of the system during flybys rather than holding their orbit.

The proto disk is by far the most fun.

This might be a good place to note that voar.io - previously - is back online!

Oh yay; I loved voar.io!
posted by We had a deal, Kyle at 11:23 PM on April 24, 2015


Loved this. I found the best results for making a small system of one Large body and several satellites seems to be to start with one Large, then randomly click a bunch of Tiny, Small and Mediums around it, increasing in size the further out you go. The Tinys get absorbed quickly, but so do the Mediums further out. Once the chaos settles, voila. Three-moon planet!
posted by Aya Hirano on the Astral Plane at 7:39 AM on April 25, 2015


Ha! Two planets with stable orbits, each with a moon in a stable orbit.
posted by backseatpilot at 9:46 AM on April 25, 2015 [1 favorite]


Holy crap, how did you manage that?

I've been playing with creating a cloud of 30-50 Small particles, with a Large or Huge at each end. I am determined to create a binary system somehow.
posted by Aya Hirano on the Astral Plane at 3:10 PM on April 25, 2015


Incredible luck, I think. The most success I've had is to start with a planet in a wide, circular orbit (or as near as you can get) - highly elliptical orbits seem to throw moons at the periapsis. You also need to lead the planet slightly when launching a moon, plus the moons need to be launched at a higher velocity than the planet is at so that they don't just fall in to the planet. Trying to eyeball it is really hard. It seems like it might be a little easier to try to get the moon into a revolution that's opposite to the revolution of the planet.
posted by backseatpilot at 8:44 AM on April 27, 2015


Here's two moons orbiting a planet, orbiting a star. Getting more than one moon around another orbiting body is hard - the first moon tends to absorb other particles that get within its sphere of influence (much like our own Moon does!).
posted by backseatpilot at 9:40 AM on April 27, 2015 [2 favorites]


Planet with a moon, orbiting a binary.
posted by eruonna at 9:58 AM on April 27, 2015 [2 favorites]


This is one of the coolest things I've played with in a LONG time. Thank you.
posted by gauche at 11:43 AM on May 4, 2015


Is there anything like this but more full-featured? Something where I could more precisely set the velocity of objects, or zoom in and out, or save states for experimentation? I've been playing with this for the past week and it really has me captivated.
posted by gauche at 8:03 AM on May 6, 2015


There's a thing called Universe Sandbox that might satisfy?
posted by Just this guy, y'know at 8:12 AM on May 6, 2015 [1 favorite]


I feel like I played with something like this on the Andrew system at CMU in 1987. For hours and hours.
posted by not_on_display at 11:58 PM on May 6, 2015




« Older Immediate Family   |   Man Healthy Newer »


This thread has been archived and is closed to new comments