How to calculate a orbit for space engineer's planets?

^^

 

i don't have the equations handy but I don't think we had the speed to make a real orbit without mods. Simulating eosync isn't hard though, just make sure you're high enough to get enough sun and have enough thrusters to counteract gravity.

 

several tests with radial gravity mods showed that you can with sufficient distance. The farther from the surface you go, the gravity strength drops. There's gotta be a point where it's balanced between the gravity's acceleration rate and the mediocre speed of 104m/s at a high enough altitude that it causes an orbit. I feel it's totally possible.

The only missing is some sort of drag coefficient. Objects should decelerate a lot faster in the atmosphere due to drag. Wonder when that will be added.

 

There isn't. Gravity cuts out entirely before that happens.

 

You need a speed mod.

 

well basically it's pretty easy ;-)

the centrifugal acceleration a from orbiting the planet (equal centripetal force) should be as large as the gravitational acceleration.

all you need are these equations:

1. Centripetal Acceleration a=v²/r (1)
2. gravitational Acceleration g*d (2) ->(which is given by the game as g= 9,8 m/s^2) at the factor (d) which is the decrease of gravitational acceleration due to the distence to the body's center

well... everything you have to do is

a=g*d -> v²/r=g*d (3)

r... the radius of the orbit (from the center of the body)
v... speed of ship (maximum 104,4 m/s)

d... factor that lowers gravitational acceleration... the gravitation decreases with the further away you are from the center of the body



-----> so for d you take (R/(R+h))² where

R=Radius of celestial Body
h= is the distance form the surface of the celestial body

and the formula for gravitational acceleration (a) at a given distance "h" from the surface of a planet with the radius R is:

a = g * (R/(R+h))² (4)




so (R+h) is the overall distance from the gravitational center, so you should substitute "r" from the centripetal Acceleration (equation 1) with this term so it is like this

a= v²/(R+h) (5)



Now... in the last step, combine (4) and (5)

v²/(R+h) = g*(R/(R+h))²



R, and g depend on the planet, Planets are 100km diameter, moons are 25km diameter, well and v is quite slow, so let's assume the maximum speed at 104. so we are looking for the height to the surface "h", let's rearrange the equation

h= R*(g-v²*R)/v²

R= 50 km (Diameter /2)
g= 9,8 m/s²
v= 104 m/s

unfortunately you will see that you won't be able to reach a stable orbit with the 104 m/s speed cap... maybe you could do it at a moon but didn't calculate it...

grz


edit:

:-( :-( :-( :-( :-( :-( :-( :-( :-( :-(

 

thank you

 

Small correction: d = (R/(R+h))^7. The in-game gravity isn't Newtonian.

 

Is that scaling valid in general or does it just cut off somewhere? Because it would allow a stable 104m/s@45km orbit, according to my calculator =P

 

Gravity cuts out at 40km, IIRC.

 

Perhaps not.
If I remember correctly...
The speed limit is not directional; you can go 104.4 m/s total, not 104.4 m/s in each direction. If you are going max speed aligned to the X axis, for example, when you gain speed on the Y axis you'll lose speed on the X axis to keep you from exceeding the limit.
This means that, if you're going at 104.4 m/s and your thrusters can accommodate for it, your ship won't be affected by gravity. SAGE has a ship that turns using the same concept.
So all one would need to do to orbit, is properly set a gyroscope (which admittedly might be very, very difficult, but possible thanks to the hand altitude marker/rangefinder) and set the thrusters to accelerate faster than the gravitational pull, thus keeping the ship at its max speed in a direction parallel to the planet's surface.

This is of course mostly theory and conjecture based on the game's mechanics, and I'm not entirely sure if it has been updated at all, but I digress.
Perhaps it would be worth a try.

 

Not really an orbit then; you could get the same effect with a bunch of waypoints and the autopilot.

 

Has anyone yet checked what is default gravity falloff in release?

 

Default falloff is 7. You can change it by manually editting <GravityFalloff> value in SANDBOX_0_0_0_.sbs savefile.

 

Ive succeeded in an orbit, however it is more "flying round a planet" than a real life orbit. However seeing your station move across the sky high up from the ground is intense as it is awesome. Build a station you want to a certain grid size and then add thrusters from there. It's not economical but for docking on it is the best 5mins I've ever had in SE yet.

 

True. But it should simulate orbit accurately enough without requiring too much energy and without filling up your screen with markers when you reach a certain distance from them.

 

So, I spent some time experimenting with orbits last night.

Procedure:
In a small ship, I started at velocity 0 m/s positioned at the transition between 0.19g and 0.18g above the earth-like planet. (You cannot use the HUD altitude to reliably position yourself relative to the planet because the HUD gives the altitude from the *surface*, not the *center of gravity* of the planet, but I digress.) I proceeded to accelerate to a given speed, parallel to the ground, with a speed mod to allow unlimited speed.

Normally, orbits are somewhat forgiving; while its almost impossible to establish a perfectly circular orbit, a stable elliptical orbit should be fairly easy. This is not what I found in Space Engineers...

Results:
Accelerating uniformly to a speed of 393 m/s yielded a slow escape into space after less than two complete orbits of the planet.
Accelerating uniformly to a speed of 385 m/s yielded a descent, slow at first but accelerating rapidly, into the planet.

I hypothesize that the tendency to either fly off into space or crash into the earth, with a razor-thin difference of less than 8m/s, exposes the nature of gravity in Space Engineers.

** Conclusion: **
The effect of gravity in Space Engineers does not add/remove enough kinetic energy to allow for a stable orbit.

Put another way:
- on the fly-off-into-space side, gravity does not slow you down enough to prevent you from flying into space.
- on the crash-into-the-ground side, gravity does not accelerate you fast enough to miss the planet. You always hit the planet, every time.

Recommendations:
- For those wishing to repeat this experiment, I believe the earth-like planet in "easy start earth" is centered at the origin [0,0,0]. I recommend you make a GPS position on one of the primary axes to act as your starting point, and maybe a second coordinate some ways away to aim your ship at. This will allow you a consistent start each time.
- If we know the position of the planet (eg: earth is centered at the origin [0,0,0]), we may be able to write a Control Block script to plot a circular orbit at a given altitude+velocity and basically guide the ship along that route. At the right velocity, the only energy expenditure should be to nudge the ship back on course when it starts to drift; most of the acceleration to remain in orbit is provided by the planet. Also on the plus side, a script is better than a bunch of GPS coordinates. Also on the plus side, a script could rotate the ship such that its orientation to the planet remains the same (ie: the same side always faces the planet surface.)

Cheers

 

Actually, in advanced mechanics class a couple weeks ago, we talked about the fact that only a select few powers of r yield a stable orbit. Sadly, r^(-7) is not one of them. Thus you cannot orbit planets.

 

using a way point system i've found you can make a solar satellite and set it to orbit the world endlessly. you do have to keep an eye on the power levels though as batteries and solar cant be called from a programmable block anymore.

i have a feeling that if you were to stay at the right height speed and use a programmable block to check how high you were you could have it climb up every so often to make sure it held the right height. power should not really be a problem as well. you may need a mod to make solar panels follow the sun though.

 

Well the problem is if you're orbiting and not geosynchronous then you'll hit the darkside of the planet and probably have trouble recharging. Even with a battery you'll eventually run dry. Course toss 100 uranium in a basic Sat that just does laser comms or something and you'll be good for years.

 

ive not seen it drop below about 30% but its just 10 blocks 4 batts a programmable block and 6 solar panels and there's a single laser antenna and 4 cameras and some sensors. i've just got a script that turns off everything but the thrusters programmable block and sensors. i'm using a mod to disable the second 2 batts as backups as well and re-enable them under 5% power. it seems to work perfectly so far.

 

Something is wrong in this... It's always negative.

 

I had some time to think today and I was thinking about how orbits in SE might be possible. I came to the conclusion that due to SE's non-Newtonian gravitational physics, we can't set in place an orbit using traditional models. My idea was to make a script. First, the user would need to manually find out the acceleration due to gravity and the distance from the center of mass of the planet (good luck with that). After that, the ship would start moving at the required speed tangent to the planet, calculated from manual input. This should set up a basic orbit, but it's not perfect and only half of the solution. As soon as the script is activated, the ship could simulate the circular orbit pattern it is supposed to follow based on its distance from the planet and the speed the ship should be moving around the circle. Then, if the ship detects that it is falling behind or is drifting slightly off course, it will fire the thrusters to redirect itself to the desired orbit. The ship would remain in orbit, and wouldn't rely on thrusters to propel itself in a circle around the planet, but let it actually orbit around the planet and only use thrusters to correct small errors in its orbit.

Sorry if that didn't make any sense.

 

Did you see this thread started yesterday...

https://forums.keenswh.com/threads/r...ble-orbit-at-given-planetary-gravity.7381638/

 

oh i forgot this one existed lol

 

Well if you have a speed mod I guess you could slingshot yourself but aiming that would be difficult.

 

Clearly didn't have that speed mod when you replied to this thread

 

I actually used math on a MP world with a speed mod to calculate an earth-gravity-assist manoeuvre to aim a very large ship towards the moon, and to my surprise it actually worked with almost perfect accuracy. If only my calculation of the stopping distance had been so precise, it would have ended very well...

 

I'm sure that's rule 4 or 5 of Space Engineers. Don't point your ship directly at the object you are travelling too

 

Turn the gps markers off in your gps tab then they wont show up.