"Planet Moves Star"

by Ted Huntington
11/16/2005


Note: This is not a formal paper, but is instead my notes from experimenting with modeling a planet that can move a star, for example, how a Jupiter mass planet could be used to move a star the size of the sun out of the plane of the Milky Way Galaxy.


Planet Moves Star:
----start planet moves star model-----------
This model uses planets to move a star up or down (out of a spiral galaxy to an orbit of a globular cluster).

There is a way to move a star, using gravity. This experiment seeks to find how easy moving a massive star may be. Our star, the sun, is 1 million times more massive than the earth, the earth pulls on the star, but only in very small force. The Sun is 1000 times more massive than planet Jupiter, and Jupiter has enough mass to pull our sun back and forth in a larger amount. Some things seem clear when talking about moving a star with a planet, or some kind of star moving huge piece of matter:
1) A large amount of planning, and continuous checking probably must go into verifying the movement of the star, since the movement of all the atoms in stars and planets cannot be known, only estimates may be used and that requires constant adjustment.
2) A large amount of fuel is needed. For each thrust a planet gives, for example, from some massive Hydrogen+Oxygen engine at it's south pole, a huge number of photons are released from the fuel (H2 and O2), and that makes the mass of the planet less. Perhaps a small initial thrust is all that is needed to set a star moving in a plus or minus y direction, that is what I hope to understand through doing this experiment.

There must be massive trade-offs, probably the smallest star-moving matter will be wanted initially, although advanced civilizations may have massive objects with massive engines for just such purposes. Early advanced life would probably have to start small, with the smallest planet (or star-mover) possible. They would probably accept a very very slow change in star velocity. Perhaps there would be no hurry to get out of the plane of a spiral galaxy. In addition, there is the problem of where all the fuel will come from. Perhaps fuel will need to be imported. Even a tiny thrust might be all that is needed to send a star up or down out of the Milky Way, but then, without doubt that star will come back down some where else in the Galaxy. I would think that globular clusters put together by advanced life would want to stay relatively close to the plane of the spiral galaxy to get more matter to use as fuel, since there is no doubt that they are losing more matter (in the form of stars emiting photons, and any other matter they use that results in photons being lost, all matter emits photons) than they can get from photons they can collect from other distant stars and galaxies.

Notes:
1) possible to move star very quickly and orderly in -y direction with constant
-.1ypixel/frame constant thrust on 1/100 mass planet.
star100x_planet_up_const.avi

For a reference, let's say that this planet is where Jupiter is 800e6 km from the star. Each pixel is then 8 million km. This is a planet 10x the mass of Jupiter, thrusting continuously to produce a velocity of .1 pixel a frame= 3 pixels/sec=2.4e6 km/s 2.4 million km every second (that sounds like too much).
2) 1/1000 mass planet, I can see that moving a star might require very little thrust indeed.
star1000x_planet_down_30f.avi
In this video, a planet the mass of Jupiter (1/1000x the star) moves a star down, you have to compare the star at the beginning and at the end of the video to see the change in location of the star, which is large, but the velocity of the star is so slow, that the movement is basically unnoticable. The planet thrusts this time for only 1 second at the beginning. The thrust is still huge, resulting in .1 pixel/frame 2.4 million km/second.
3) 1/1000 mass planet, this time with a more realistic initial thrust of 1000km/s for 1 second, this translates to .000033 pixels/frame...a very small change in velocity. I won't bother to show the video, because the position of the star changes by only a fraction of a pixel. The velocity (in the y [down] direction) of the star changes from 0 to: 1.28e-7 pixels/frame = 997 million km/year. At that rate it would take 10,000 years to move 1 light year. But then, only the tiniest amount of thrust was used, a thrust that produced a velocity of 1000 km/s on Jupiter in the y direction for only 1 second. The fastest jet can move 3x the speed of sound ~1 km/s, I don't know how to calculate the amount of fuel needed, and the size of the engine needed to move a planet the mass of Jupiter.
I am not sure what a realistic goal would be, but it seems like, the velocity of the star would be the thing being adjusted. I guess a civilization might want the velocity of the star to be 1 light year/an earth year...that is kind of fast...in four years we would sail by Centauri (4 ly away)...in fact it is the speed of a photon, so is very unrealistic. More realistic would be a tenth of that. We would go past Centauri in 40 years.
4) 1/1000 mass planet, 1 second of thrust that results in planet velocity of .1c 30,000 km/s. again, no real velocity is imparted to the star. I have learned more. Here is some cool stuff: The velocity added to the planet is indirectly tranfered to the star. The planet loses y velocity (after that initial 1s thrust) while the star gains velocity. In this example, the sun y velocity went from 0 to 3.8e-6 (30 km/s) and the planet's y velocity fell from 1.25e-4 (1000 km/s) to 6.8e-5 (544 km/s). Again there was no noticable difference in the location of the star since it did not move even 1 pixel.
5) 1/1000 mass planet, 10 seconds of thrust that makes the velocity .1c (30,000 km/s). This was closer to seeing actual movement in the star. The star velocity went from 0 to 3.7e-5. After only 10 seconds the star had moved .006 of a pixel=48,000km 4,800km/s just over 1% c, which would get us to Centauri in 400 years. That means 10 seconds of a relatively small thrust on Jupiter, could move our star to Centauri in 400 years, not too bad.
6) what about bringing Jupiter in closer, to the distance of Mercury? 5 pixels away at 46 million km from the sun. Actually, things get kind of nasty at that distance because the planet is too close to the star...it is tough to control that close. Because the closer it gets to the star the more gravitational influence (inverse distance squared, so the force is exponential). 1/2 the Jupiter distance is easier to control, 400 million km away, about 100 million km outside of Mars orbit.

I am starting to think about the perfect ship to build, and I think that a good idea would be a ship that has 4 main engines, one on top, bottom, and 2 in the middle on opposite sides. Perhaps even two engines is all that would be needed, one main engine for thrust and the one side engine, the side engine is pulled out to spin the ship on the axis of the main thrust engine. Actually a third engine would be needed to spin the ship at a 90 degree angle to the other steering thruster. Anything else about the ship is open to anything...it could be a sphere, a hollow sphere, a cylinder, a cube, rectoid, ...anything. Probably the ship would have many living compartments, perhaps an all transparent surface so that the inhabitants (at least in the prized outermost units, perhaps many would be public shared spaces) could look out at the universe. It is difficult to know how the future will play out, but potentially, the real business of feeding people may result from atom separating and constructing machines, and have very little to do with growing plants. It is a startling possibility. Any old matter could be turned into any new matter. Probably the valuable matter of trash and human waste would be converted to new delicious food dishes, air, water, etc. How interesting it would be to be living on a huge human-made ship that was thrusting to move the central star (and all accompanying planets, comets, etc...basically every smaller piece of matter in orbit) to a better location. Ofcourse when a star moves up, the much more massive center of the Milky Way Galaxy will pull the star back down eventually, and that equation is even more complex, because of the hundreds of millions of stars, but it too could probably be realtively accurately generalized.

ok the simulation is done. Again, the star did not move 1 pixel (8e6km) in 10 seconds, it moved: .007 of a pixel, so better than before (only .006), but not the real noticable difference I was hoping for. The star had a final velocity of: One problem is getting the initial velocity of the planet to make a perfect unchanging ellipse.

7) ok 2 planets, with the same thrust above in 6).

These models are kind of interesting in that a small object is using wisdom to move a bigger object. Ideally, if we could put the engine in the star we would be getting the most efficient star system movement, but since that is not easily possible, moving a large piece of matter as close as possible is the next best thing.

With two planets, there is little change. The star's velocity goes to: 7.5e-5 (I was hoping for 14e-5 or twice the amount, but there was no such result). But again, we are only using a tiny thrust, and only thrusting for 10 seconds. star1000x_2planets_down_30f.avi

8) I think a good goal would be to get the star to move 1% to 10% the speed of light, that would move us:
1% = 400 years to Centauri
5% = 80 years to Centauri
10% = 40 years to Centauri
1%c=3e4 km/s To move 1 pixel at this scale would take (8e6km/3e4km/s) 266 seconds and would be 3.75e-3 pixels/s, 1.25e-4 pixels/frame (.000125). The most I could get in 7) with 2 planets at 10% the speed of light was .000075, about half that. I am not going to make a 300 second (5 minute) video just to see the star move 1 pixel, but the principle is neat; that it is not too difficult to imagine a relatively advanced civilization (I would say, like us, perhaps in a few thousand years) to actively move their star system without too much trouble. In fact, there is no doubt that with the creation of large ships, the cumulative effect on the star would definately have to be modeled and planned carefully. Ships would be built very small at first, but no doubt they would be modularized, so that they could be combined to form larger ships. The largest ships might be built of many millions of smaller ships.

To get a star to reach that velocity in only 10 seconds is perhaps unrealistic; to get the star to reach that velocity in 10 years might be more realistic...to have 10 solid years of thrust...it would make for an exiting decade for those in that star system, in particular anybody living on the star mover.

This is a basic idea that we can use S=v*t, the distance will equal the velocity*t, so 1-10% c velocity for the sun is probably a good goal for any star system. Actually, I need to see how the velocity of the star is changing over time. Vstar=Vstar+X each second given a constant thrust, so what is
X? V=at is the equation, so I see that for 6) the change in velocity (acceleration) of the star is 8e-6 pix/s/s or 64 km/s/s
V will equal
.01c in [3000 km/s=64*t]=47 seconds (so 47 seconds of thrust on Jupiter should put the star at .01c)
.05c in [15000 km/s=64*t]=234 seconds ~4 minutes (about 4 minutes of thrust that results in .1c velocity on Jupiter)
.10c in [30000 km/s=64*t]=470 seconds ~8 minutes (about 8 minutes of thrust, and the star is going .1c the desired velocity)

Going back to 1 planet, the same mass as Jupiter and same distance. Let's see if we can get the star to go .01c in 47 seconds [1410 frames] of constant thrust. ok it ran past 1410 frames, and at 1749 the velocity of the star is 7.66e-5 pix/sec = 613km/s, still not enough (so the acceleration on the star must be less than 64 km/s/s, here it is only 17km/s^2)....ok I just realized that I stopped the planet thrust after 10 seconds so that explains the lower acceleration. I will keep the simulation going until it reaches .01c, .05c and .10c using constant thrust that results in .1c on the planet. Actually this thrust can be looked at as an acceleration of .000125 pix/frame or 30,000 km/s/s since I am adding this to the velocity of the planet each frame. Perhaps a more realistic acceleration for a planet would be only 1,000 km/s or even less. Yes, it is very unrealistic, since adding .1c to the velocity of the planet quickly puts the planet near the velocity of light. I will test that acceleration
Here are the results for 1 jupiter accelerating at .1c every second:
To reach:
.01c [3000km/s] took 1504 frames = 50 seconds (acceleration=60 km/s^2)
.05c [15000km/s] took [estimate at 60 km/s^2= 250 s = 4 minutes] 7516 frames = 250s=4 minutes, an exact match to the estimate

I can see that, no doubt advanced life would go for a lower constant long term thrust on the planet/star mover. That is a nice capability to have for adjusting velocities later too. In addition, the nature of fuel rationing would favor smaller thrusting over longer times.

Here is a more realistic acceleration on the planet of 1000km/s^2. This could perhaps be thought of as 1000 jet engines, since the fastest jet can go 1km/s.
.01c [3,000 km/s]
.05c [15,000 km/s]
I am going to skip this because it is too unrealistic.


Maybe I should just go for maintaining a constant velocity on the planet, since adding up acceleration may be unrealistic since, as something starts to accelerate, [think of a jet, for example] it becomes more difficult to accelerate more...it takes more fuel to maintain a constant acceleration...like in a car...it is easy to maintain a constant velocity, but you can imagine trying to accelerate at even 1 mi/hour per second more, because the engine is "topped out" and going as fast as it can at 200 mph and can not go any faster. So how long would it take if a constant velocity of 1000km/s could be maintained on Jupiter? I think a constant velocity on Jupiter would mean a constant velocity on the star, so again, we are stuck with the basic idea of: the star will move slower (perhaps .5 the velocity of the planet...this I will check).

Ok, the Y velocity of the star is roughly 1/10 that of the planet and does increase (perhaps because the planet is at times closer to the star and so the velocity is felt more strongly there).

Vplanet=3000km/s (.01c) Vstar=30 km/s (.001c) (3.7e-6 pix/frame) the star has a velocity a tenth the velocity of the planet [this velocity would take 4000 years to Centauri...that would probably be too slow] [the most accurate thrust is probably some where in here between 1 and 5% c] Vplanet=15000km/s Vstar= 6480km/s (2.7e-5 pix/frame) 43% planet and 2%c...not bad...reach Centauri in 185 years
the average velocity after 46s was 13,955km/s, 93%
after 495s the average velocity of the star was 14,329 km/s 96% efficiency, that is phenomenal, but other stars and planets in the Milky Way would make the efficiency less. Centauri would be reached very near 200 years.


ok at 11 minutes the velocity of the star is 15,549 km/s, at 16 minutes the star reached .10c, so something is definitely wrong with my model here. At 54 minutes the ave velocity of:
star: 14,771 km/s
planet:
after 20 hours:
star: 14,679 km/s
planet: 14,748km/s


One interesting thing I noticed after 20 hours of simulation is how little the elliptical orbit of the planet changed, there was no recognizable rotation called "procession", where the ellipse rotates (the path the earth takes, an ellipse, rotates very slowly). I can only imagine what billions of years of rotating does to matter and what that would look like sped up into a few minutes.

so I guess this is accurate, since neither velocity averages over 15,000km/s and they are both very close to that velocity. It says to me that almost all the velocity applied to a planet is echoed in the star with almost 100% efficiency because there is very little friction between planets and star. But I have to think that distance from the star when thrust is applied does make a large amount of difference.

This high a velocity is kind of unrealistic, maybe a constant velocity would not be wanted, but a constant thrust instead. I am not sure how to add in thrust to the velocity that results from gravity, because thrusting an engine produces an acceleration that is related to the existing velocity of the engine relative to the rest of the universe. The faster the planet moves, the less acceleration thrusting will add. This phenomenon, I don't completely understand, I think it has to do with the interaction of the photons in the engine, the thrust is produced by photons being thrown in one direction against those photons that are part of the engine that results in movement in the opposite direction. Most people don't describe rocket engines (or any engine) in this way.

Perhaps I am doing the calculation wrong here. I calculate the force (using inverse distance square m/dist^2) of each other piece of matter on each individual piece of matter, add those together (so in theory, the more matter, the larger the force, and the larger the velocity...). I was doing m1*m2/d^2 but quickly saw that one any given piece of matter only the other mass matters, because I am calculating the force on that one piece of matter. In addition, I am not using the gravitational constant = 6.67300 × 10-11 m3 kg-1 s-2, but I think that is ok because it relates to the mass, and I am simply using proportianal masses, but perhaps I should include the actual masses and the gravitational constant. Since multiplying these is linera, it would be the same result but at a different scale.

?) 1/1000 mass planet, this time with a constant thrust of 1000km/s, perhaps this too, is relatively realistic, although, as fuel was used, the mass of the planet would go down (I do not account for this, in this model yet).

major points
1) Should the advanced civilization go for a large initial thrust or a constant (or long term) smaller thrust? I have to think that smaller thrusts from time to time may be used to correct or adjust the star movement. A large initial thrust that gives the star a larger initial velocity, might add up over time, where with smaller thrusts the velocity would be built up over a longer time.
2) Use one big piece of matter or many smaller pieces of matter to move star? I really didn't answer this one, in the one simulation I did there was no significant gain from adding an extra Jupiter-sized star mover, but I find it hard to believe that two planets would not double the velocity of a star.
3) I think the equation can be generalized by v=at. The longer the planet accelerates in the Y, the higher the end Y velocity of the entire system will be. So, for example, to aim for a 1%c end velocity, a variety of "accel*time" combinations exist:
a t (to get to a velocity of .01c)
3000km/s^2 1 sec
300km/s^2 10 sec
30km/s^2 100 sec (1.6 min) *
3km/s^2 1,000 sec (16 min)
300m/s^2 10,000 sec (167 min or 2.7 hours)
30m/s^2 100,000 sec (1,667 min or 27.7 hours)


* let's run this simulation. 100 seconds of 30km/s^2 acceleration. Do we (star and planet) reach a velocity of .01c (3,000km/s)?

after about 6 minutes (12,000 frames), long after the thrust has stopped, it looks like the Y velocity of the planet is oscillating between 2,000km/s and -2,000km/s (it's going up and down in it's periodic orbit). The star is oscillating but much less, between 74km/s and 105km/s. So potentially, it will take more time for the system to settle down and oscillate less. Perhaps then, the remaining portion of the Y velocity that appears in the planet will be more evenly distributed between the star and planet, resulting in something like 150km/s for both star and planet. So it looks like the velocities are far less than the desired 3,000km/s. I guess acceleration on a tiny planet 1/1000 the size of the sun goes 1000 times less on the star. Currently the star only took 3% of the 3,000km/s. If the star will only take at most 5% of this planet's velocity we would need to accelerate for 20x longer [2,000 sec] or 20x larger [600km/s^2] in theory.

looking again, both star and planet are still oscillating but the planet picked up velocity (from only planet-star distance differences I guess) to 88,000km/s to -96,000km/s, and the star 256km/s to -56km/s. So, perhaps the star may reach a final steady-state velocity of more...at least as much as 256km/s almost 10% of the original goal. ok I just measured and the star has an average velocity of 2,504km/s over the last 745 seconds, 83% of the goal. I guess, the velocty of the star of planet at any given time is very misleading, and the average velocity (simply Ystar/time of simulation) is a better guide. As a guide 2,500km/s would go .83%c would go 1 light year in 120 years, 4 light years to the stars of Centauri would be 480 years... it seems unbearably long to people used to our life of flying from city to city, but for a long term plan of moving out of the star system it might be acceptable...in addition, we might not have any choice, given our lack of matter and size of engines, etc. That is simply thrusting with an acceleration of 30km/s^2...in theory even an acceleration as tiny as 30m/s^2 for a few hours could achieve the same result. I don't doubt that advanced life will be going with small engines that constantly thrust (although you have to remember that the matter used for fuel would probably be finite, and have to be highly conserved). I don't doubt at the scale of moving a Jupiter-mass planet, that separating mass from fuel to photons would show a measurable difference in mass on the planet.

It is interesting to think about how the members of some advanced civilization would react to the idea of moving a star. Even after many simulations and assurances, I am sure that many, and even most would be reluctant to try and change the natural order of the movement of the planets and stars in their system...no doubt a smaller and newer test star system would be experimented on first...or would lead the way for the rest of the systems. After seeing the theory successfully performed on one star system, others would probably feel less worried or unsure of success for their star system. Things move on such a slow time scale, that there would probably be large amounts of time and fuel to fix any errors...those living objects would have plenty of time to observe the results of the experiment.

the average velocity of the star is increasing:
850s = ave velocity 2,525km/s
1048s = ave velocity 2,553km/s
23m = ave velocity


the sun reached .01c (3,000 km/s) at 1048 seconds, and .05c (15,000 km/s) in 24 minutes, so clearly there is some kind of cumulative feedback loop happening. The planet now has an oscillating velocity with maximum 1,000,000+ km/s so there is definately some amount of feedback (when a planet can have a fractional distance [less than 1.0], the force may be larger than the planet's initial force).

Perhaps I am using these equations inaccurately (perhaps I should use the gravitational constant and the actual masses, and perhaps something is wrong with simply adding the force exerted by each other object as velocity). I am not sure what to conclude other than that moving stars with planets is definitely possible, but that I need to experiment more to understand the basic relationships between thrusting a distant planet and what effect that has on the star in the center. I thought that I would be able to conclude that over time any force applied to a planet would be applied to the star with almost no loss, in other words the average velocity of the above star and planet would eventually go to 3,000km/s, and I have not ruled that out, but it looks like the velocity of the star (and planet) are exceeding the initial velocity because of close interactions, or for some other reason. Perhaps that is something to watch out for...close interaction with the star that will amplify the Y velocity component.

Perhaps one lesson is that any tiny change in movement may be amplified over time and that may be something to watch out for.
[trying 3km/s^2 for 1,000 sec (16 min) same results?]

finally, for some fun simulations:
1) fast down to centauri direction
star1000x_planet_down_const.avi
2) diagonal
10 seconds of thrust in the +X-Y
star1000x_upleft_10sth_5xy.avi

3) sideways shuffle
10 seconds of thrust in the X:
star1000x_right_10sth_5x.avi


notes:
1) There are only 151 known globular clusters of the Milky Way Galaxy. That is an interesting number. It could mean 151 advanced civilizations have arisen since the creation of the Milky Way Galaxy. Is the largest cluster the oldest? But there may be even more advanced civilizations that simply joined an existing cluster.
2) http://mb-soft.com/public/globular.html has a few good points...if globular clusters crossed the plane of the glaxy wouldn't they pick up diverse stars? are the clusters close to the central mass of the Milky Way?
3) Globular cluster 47 Tucanae is moving in our direction at 19km/s (.006%c=would take 15,789 years to move 1 light year...to me that seems like a long time, but maybe that is the best a globular cluster can do). M4 is the closest globular cluster to us at 6800 light years. I doubt we could get there in less than 68,000 years and that would be going .1c which is very fast. Perhaps we will make our own globular cluster. I don't doubt that advanced life in the globular clusters would be very observant, and may take a strong interest in knowing what planets are growing life that is advanced enough to go to other stars...or perhaps even advanced enough to leave it's planet of birth. It would be to their advantage to try and steer that new developing life in a way that they want...it's like seeing somebody that is lost...many people would probably want to give them a map. Perhaps they want to control those new civilizations, to stop them from growing uncontrollably...with a civilization that moves star systems around, I seriously doubt there would be any real resistence from new comers like us, ligitimate resistence could probably only come from other medium-developed globular clusters. Are globular clusters built by advanced life? maybe no, but why not pursue both ideas? It is interesting to think about communication with those advanced objects. We certainly would want info...scientific info and history about the galaxy. Some of that could be learned by watching them (capturing photons emiting from them). Everybody wants matter probably...to grow more stars, living objects, etc. It must be a constant quest for more matter, to grow more, in order to reproduce, just like it is for bacteria and most living objects even on the earth. it is important to track the globular clusters, are they following regular motions, or unusual motions that might indicate the presence of advanced life? Do they cross the plane of the Milky Way? What has their motion been? That should be modeled and shown to the public. In addition to the individual stars in globular clusters and elliptical galaxies.
---end planets move star model------------