Best Way Into Non-Atmospheric Orbit?

[Streetwind]

Very nice challenge, I see now the error of my ways. Had great fun using one of the linear RCS ports to keep the craft from dropping back to surface while burning in a completely flat trajectory Managed to get my Kerbin periapsis to disappear into the planet on the first try with fuel left that way, while with direct ascent I couldn't get it into the atmosphere.

[maccollo]

Very nice challenge, I see now the error of my ways. Had great fun using one of the linear RCS ports to keep the craft from dropping back to surface while burning in a completely flat trajectory Managed to get my Kerbin periapsis to disappear into the planet on the first try with fuel left that way, while with direct ascent I couldn't get it into the atmosphere.

Ah, I forgot the pods have monoprop now XD. I needed to reduce the delta V a little bit so I added the linear ports to add a bit of mass.

That's pretty creative though

[ComradeGoat]

Also worth pointing out that the reverse of this procedure, burn as close to horizontal as close to the ground as possible, is a very efficient way to land.

Especially on Tylo.

[SSSPutnik]

Sorry, I am probably just extremely dense, but I still don't see how not burning direct for SoI departure is not more efficient.

In my reading on the Oberth effect it seems it has little to no relation to takeoffs.

From what I understand, the Oberth effect comes into play most strongly at high velocities, like at periapsis. This is counter intuitive to low velocity at takeoff.

Surely most efficient is to get as far away from a gravity well as quickly as possible so gravity has less time to act on you.

Or is this applying only to KSP physics? Getting into orbit, yes, I can see horizontal as best, but to leave SoI?

[tavert]

If you have infinite TWR, it doesn't matter which direction you burn in once you're past escape speed. If you have a given speed at a given altitude, your orbital energy is the same no matter which direction your velocity vector is pointed.

But for finite TWR, your net acceleration is lower when you burn up than when you burn perpendicular to gravity. So it takes less time to reach a given speed by burning sideways. The lower your TWR, the more the difference.

[Kasuha]

Sorry, I am probably just extremely dense, but I still don't see how not burning direct for SoI departure is not more efficient.

You can try the mini-challenge so you can see it is true, the only question is why is it true.

First, I don't know any explanation which would be completely clear and obvious.

Second, I don't think it's consequence of Oberth effect.

The most clear explanation I can get together is this:

First you must understand that TWR of the ship plays important role in this. If your ship was able to deploy the whole impulse instantly, there would be no difference whether you straight up or whether you burn sideways to get to circular orbit first and then at suitable point burn for escape. Explanation of this point is a bit technical so we can return to it later if necessary.

What plays the key role here is therefore not the energy you need to get away (it is the same) but the energy you waste on inefficient thrust.

When you lift off the surface, you are already in orbit. Very elliptical orbit with periapsis deep underground and apoapsis right where you are.

When going straight up, you start raising your apoapsis from where you are and as the apoapsis rises, your position on the ellipse of the orbit moves towards the periapsis. Relatively to the size of the orbit, of course, your real distance from periapsis does not change much on the course. But as you burn, you never burn anywhere close to the point where the maneuver is the most efficient, i.e. the periapsis. It only gets slightly better over time.

You waste a lot of dv by burning at places which make the burn inefficient.

When going to orbit first, you put a lot of effort to burn at optimum points whenever possible. I.e. you raise your periapsis first, only wasting energy to make sure you don't get too far away from apoapsis. And when in orbit, you start the burn at the periapsis and raise the apoapsis from there without getting significantly far from it before the burn ends.

The only losses here are from off-prograde burning when you're getting to orbit and make sure you don't fall back on the surface.

Apparently losses during liftoff and circularizing are smaller when added up than losses caused by inefficient burn straight up.

[maccollo]

Sorry, I am probably just extremely dense, but I still don't see how not burning direct for SoI departure is not more efficient.

In my reading on the Oberth effect it seems it has little to no relation to takeoffs.

From what I understand, the Oberth effect comes into play most strongly at high velocities, like at periapsis. This is counter intuitive to low velocity at takeoff.

Surely most efficient is to get as far away from a gravity well as quickly as possible so gravity has less time to act on you.

Or is this applying only to KSP physics? Getting into orbit, yes, I can see horizontal as best, but to leave SoI?

The oberth effect states that you get more efficient the faster you move, since kinetic energy is the square of the velocity.

This means the effect is very small initially, but at every point during a vertical take off you will be moving slower than in a horizontal take off, thus the oberth effect starts kicking in during the later parts.

You can ignore the oberth effect and just think about it in terms of gravity losses.

Your total burn time will be the same regardless of how you achieve escape. So you have to fight gravity equally long whichever way you do it. The difference is in how hard you have to fight gravity.

If you burn straight up you fight the gravity at maximum strength for the whole duration of the burn.

When you burn efficiently into a low orbit and escape from there you will only suffer gravity losses during the initial stage of take off.

Once you're halfway to orbit the losses are negligible.

*Edit

Probably not the best explanation, which is why I made the challenge. It's easier to demonstrate that it is true than to explain why it is true

Edited February 13, 2014 by maccollo

[Streetwind]

Sorry, I am probably just extremely dense, but I still don't see how not burning direct for SoI departure is not more efficient.

In my reading on the Oberth effect it seems it has little to no relation to takeoffs.

From what I understand, the Oberth effect comes into play most strongly at high velocities, like at periapsis. This is counter intuitive to low velocity at takeoff.

Surely most efficient is to get as far away from a gravity well as quickly as possible so gravity has less time to act on you.

Or is this applying only to KSP physics? Getting into orbit, yes, I can see horizontal as best, but to leave SoI?

Being in orbit is in essence just one big, neverending gravity assist. Specifically, you gain exactly as much acceleration from the slingshot maneuver as it costs to counteract the downwards acceleration from gravity. That is why your vehicle is always weightless in orbit, no matter how much gravity the orbited object might have and how close you are. By performing a direct ascent towards a destination, you do not get any gravity assist. You must pay extra fuel for what you usually get for free, namely counteracting the acceleration towards the local center of gravity. Therefore, burning for escape velocity out of an orbit is cheaper than a direct ascent, even if you were starting from the same height.

This makes me want to paint a picture. BRB.

in addition, the Oberth effect comes into play because you are moving faster during a horizontal takeoff than you would during a vertical one. After all, if you have a finite amount of acceleration from your engines, then the less you must subtract to account for gravity, the more is left to actually put you into motion. Therefore, after 5 seconds of accelerating horizontally, you are moving faster than after 5 seconds of accelerating vertically. This is especially true directly at liftoff, when your vehicle is chock full of fuel and you are deepest in the gravity well. The horizontal takeoff thus builds a strong early lead and benefits sooner and in larger amounts from the Oberth effect while simultaneously increasing it, which is quickly snowballing into more and more free acceleration.

As it turns out, the combined effect of the gravity slingshot and the Oberth effect is so large that it is still cheaper to build a fully circularized orbit and take a roundtrip around the entirety of the celestial body rather than burning directly upwards. I personally thought the effect was not as strong, and there would be a point where the effort of building the orbit would be more costly. But maccollo handily proved me wrong.

EDIT: Wow, massively ninja'd.

Edited February 13, 2014 by Streetwind

[SSSPutnik]

OK, I can understand that. So in effect, you're purely gaining speed then in the direction you wish to go.

But I guess then, to be purely optimal, your SoI exit must be planned from the moment you hit the launch button to avoid Deltav losses due to course correction.

This may be easier if you launch straight up in the desired direction, but you then lose Oberth effect boost...

Horizontal launch to an elliptical orbit sounds good, with course correction and final SoI departure boost at the next periapsis?

[KerbMav]

Sorry, I am probably just extremely dense, but I still don't see how not burning direct for SoI departure is not more efficient.

Surely most efficient is to get as far away from a gravity well as quickly as possible so gravity has less time to act on you.

Or is this applying only to KSP physics? Getting into orbit, yes, I can see horizontal as best, but to leave SoI?

Take a lesson in Judo/Karate.

You "can" (try) and catch - Bud Spencer style - an incoming fist aimed at your nose, but it is more efficient to simply you move your nose out of the way, grab the opponents wrist, continue his forward movement somewhat and pull him to the side.

You can also try to pull your hand out of someones grip who is pulling you towards him into a dark alley, but it would be more efficient to follow his pulling and let your elbow/knee meet his nose/stomach.

What am I trying to say?

You can thrust against the pull of gravity with all the power of your engines, but gravity will always cancel out your acceleration by its own force/acceleration - but if you fly horizontally you are using the same principle that keeps you in orbit to some extent. You cannot go parallel to the surface until you have reached orbital velocity (on your way to escape velocity in this case) or have a sufficiently high apoapsis, but you can accelerate to the side unhindered by gravity more and more. Gravity will pull your movement vector towards the ground - well, like the way your orbit always looks like - but it cannot take away your horizontal speed.

So it is more efficient to only invest as much vertical acceletation as is necessary to not fall back down to the ground.

[Streetwind]

Did I say I was going to paint a picture? Yes, that means exactly what you think it means.

Behold!

[maccollo]

If you've landed on the equator on the mun for example, that isn't really a problem.

Take off towards the 90, and once you are in orbit you plot the escape maneuver somewhere around 45 degrees of the prograde of the moon. You then get a perfect escape.

If you however landed say, right between the equator and the north pole it becomes a little bit harder. However, you can always get into an orbit which will get you a perfect escape.

Finding which direction to launch is the hard part.

[KerbMav]

OK, I can understand that. So in effect, you're purely gaining speed then in the direction you wish to go.

Always, as you do not loose any, you take your orbital velocity with you, so to speak, and it is maintained as always in an orbit.

Spent acceleration into building an orbit is a 100% return investment into escape velocity.

But I guess then, to be purely optimal, your SoI exit must be planned from the moment you hit the launch button to avoid Deltav losses due to course correction.

According to what was mentioned by Streetwind you do not have to plan anything. You just launch and build a low circular orbit, then pick your desired departure time.

[SSSPutnik]

Yup, I get it now, would be more complicated, say leaving Mun for Jool, but principle holds.

[Kasuha]

Did I say I was going to paint a picture? Yes, that means exactly what you think it means.

Behold!

https://dl.dropboxusercontent.com/u/44754370/slingshot.png

Gravity assist is a maneuver where you gain or lose orbital energy in orbit around object A by near miss of object B which is also orbiting object A.

This case is not gravity assist.

[Kasuha]

If you however landed say, right between the equator and the north pole it becomes a little bit harder. However, you can always get into an orbit which will get you a perfect escape.

Finding which direction to launch is the hard part.

It is actually very simple. You launch in the direction where the Mun orbit (line along which Mun center moves around Kerbin) intersects Mun surface.

Of course it is not 100% optimal because to be perfectly optimal you need to launch towards the point where that line will intersect Mun's surface when you'll be leaving the Mun SOI but it's rather good approximation.

[SSSPutnik]

Yup, I get it now.

Some of you are amazingly bad at explaining and some very good but I'm not naming names and the net effect is I now get it. I think. Haha

There seems to be some argument here as to the amount of effect the Oberth effect actually plays in this.

Its scary that googling this subject returns mostly KSP hits.

There are some good NASA docs on lunar ascent and descent.

Gravity drag does matter, high TWR is important in reducing dV used.

Wiki on gravity turns is very informative.

[maccollo]

It is actually very simple. You launch in the direction where the Mun orbit (line along which Mun center moves around Kerbin) intersects Mun surface.

Of course it is not 100% optimal because to be perfectly optimal you need to launch towards the point where that line will intersect Mun's surface when you'll be leaving the Mun SOI but it's rather good approximation.

I thought of that initially, but then for some reason I thought it wouldn't work on other bodies, but now I realized it will.

If you do the process in reverse, any orbit you get into will intersect an imaginary line that goes from the center of the body to the point where you made final trajectory adjustments.

If your orbit intersects that line you can escape in that direct with no course correction.

Edited February 13, 2014 by maccollo

[SSSPutnik]

Hmm, wiki says Mun escape velocity is 807 but I'm hitting soi departure at around 660? (Launching vertical).

Can we use Munar assist to cheat on return?

Awww no electricity!

Edited February 13, 2014 by SSSPutnik

[Streetwind]

Gravity assist is a maneuver where you gain or lose orbital energy in orbit around object A by near miss of object B which is also orbiting object A.

This case is not gravity assist.

Maybe the term was applied loosely, but the point was to illustrate the difference between hovering and being in an orbit. In the latter case you are being helped by the celestial body's gravity because it maintains your trajectory for you indefinitely, for free. Thus by burning for escape out of an orbit, you can apply the full power of your engines to that goal, while the directly ascending vessel can only apply that portion which isn't spent on hovering. The result is exactly the same as a gravity assist: you performed a maneuver wherein you save fuel by taking advantage of a celestial body's gravity and/or motion. Which is the wikipedia definition, and consistent with what turns up if I google across NASA pages. Maybe a studied aerospace engineer has it defined a bit more strictly, I wouldn't know...

[Kasuha]

Hmm, wiki says Mun escape velocity is 807 but I'm hitting soi departure at around 660? (Launching vertical).

Can we use Munar assist to cheat on return?

Awww no electricity!

807 m/s it is at 0 m altitude. As you progress towards SOI boundary, your speed decreases as your kinetic energy transfers to potential energy within the gravitational field.

Also in KSP there are two kinds of escape velocity. The "normal" escape velocity which means you're on parabolic orbit, and "KSP" escape velocity which will just ensure you'll exit the SOI.

[SSSPutnik]

Ahhh.... Well I learnt a lot tonight! Thanks people... I just hope I don't dream about optimum ascent paths, or worse, Oberth!

I forgot about escape velocity from centre of planet... Shame we don't have n body physics