Why are transfers from solar orbit so inefficient?

[Marclev]

Up until very recently, the way I've always transferred to other planets has been to first get into solar orbit, and then drag a manoeuvre node around until I get an intercept.

Very recently however,I've been experimenting with a continuous burn from LKO to get an intercept orbit.

I'm not even in a transfer window, but even so, the new approach is very noticeably more efficient than the previous one.

Anybody know why?

[Wanderfound]

Up until very recently, the way I've always transferred to other planets has been to first get into solar orbit, and then drag a manoeuvre node around until I get an intercept.

Very recently however,I've been experimenting with a continuous burn from LKO to get an intercept orbit.

I'm not even in a transfer window, but even so, the new approach is very noticeably more efficient than the previous one.

Anybody know why?

Oberth effect. The deeper you are in a gravity well, the more efficient a chemical rocket is.

http://en.wikipedia.org/wiki/Oberth_effect

[Renegrade]

That's quite simple - when you're in Kerbin orbit, you can leverage the Oberth effect.

In layman's terms, the faster you're going at Kerbin orbit, the less time gravity gets to steal from you, so you end up going farther with the same acceleration/delta-v.

Otherwise when you just barely exit the SOI, Kerbin spent that whole time stealing velocity from you. If you go burning out faster, it takes fewer seconds (hours~), which is fewer seconds that gravity can act on you.

Note that you will have to burn a bit more in orbit than outside of Kerbin's SOI (it's like 950-ish to go to Duna from outside the SOI, vs. about 1055 inside)..but you have to cross that distance anyways so might as well pour on the speed and leverage that Oberth effect.

[No one]

You keep more of the velocity you had while you were orbiting Kerbin.

In case it sounds weird, imagine the following scenario:

Kerbin is orbiting the sun at K m/s.

You're orbiting Kerbin at O m/s.

You're carrying a magical bomb thing which will accelerate you really really fast to a speed at which you escape Kerbin's SOI in a fraction of a second and the losses due to gravity are negligible. Let's say the bomb gives you X m/s of delta-v, where X is some really big number.

You also have normal rockets which can accelerate you from your orbit to escape velocity, but only just enough to escape. Let's say they give you R m/s.

You have two choices:

Detonate bomb and escape then fire rockets

Fire rockets and escape then detonate bomb.

If you detonate the bomb and escape then fire the rockets you are going at (Approximately, you'll still lose a tiny bit due to gravity) K+O+X+R m/s. (To start with you're going at K+O m/s relative to the sun, then the bomb gives you X m/s, then the rockets give you R m/s)

If you fire the rockets and escape then detonate the bomb you are going at K+X m/s.

[Red Iron Crown]

The Oberth effect. You get more change in the energy of an orbit from a given amount of delta-V when you're traveling faster. This reduces the amount of delta-V required for a transfer.

[metl]

Damn. I've been doing it wrong this whole time. Whoops. Lol.

[NecroBones]

Damn. I've been doing it wrong this whole time. Whoops. Lol.

Well, not necessarily "wrong", just less efficient. The benefit of doing it the way you've been doing it, is that it's easier to do without any tools for calculating transfer windows and the like. However, it is of course less optimal in terms of fuel expenditure, and probably also in terms of elapsed flight time.

Granted, if you use a transfer window, and still do your burns from a solar orbit, then yeah, you're just wasting fuel.

[Wallace]

Gravity is acceleration. Soo, the more time you take to escape the gravity well, the more speed you will lose to gravity.

By doing your whole burn at the lowest point you are making your exit a lot faster and so there's less time spent with that gravity pulling you back.

Same reason why landing with one big ol burn just before impact is more efficient than slowing to a stop halfway, it's all about spending the least time having gravity accelerate you.

[dryer_lint]

Up until very recently, the way I've always transferred to other planets has been to first get into solar orbit, and then drag a manoeuvre node around until I get an intercept.

Oh, phew. I thought I was the only one who used to do this...

[Illumia-Star]

If efficiency isn't a problem, maybe you have tonnes of fuel (Bad planning maybe?), would it be more efficient to go straight from a solar orbit or to go straight from the SOI of the planet?

[severedsolo]

If efficiency isn't a problem, maybe you have tonnes of fuel (Bad planning maybe?), would it be more efficient to go straight from a solar orbit or to go straight from the SOI of the planet?

If fuel wasn't an issue why would you care about efficiency? Either way it would still be more efficient to go from LKO as explained above (in terms of fuel consumption at least).

If you mean which would be quicker, probably a LKO transfer anyway I imagine? As you can burn directly rather than having to get outside the SOI first.

[Lukaszenko]

If efficiency isn't a problem, maybe you have tonnes of fuel (Bad planning maybe?), would it be more efficient to go straight from a solar orbit or to go straight from the SOI of the planet?

Either way, going from solar orbit is inefficient.

Whether efficiency is a problem or not, does not change the efficiency. It just changes how much you care. Perhaps you should reword your question?

Going from solar orbit does take less planning however. Game time and efficiency might be worse, but it might make better use of your real world time.

[Kasuha]

I used this method (get out of SOI, then find an intercept from Solar orbit and burn there, also perform all inclination burns in solar orbit) in my early days of playing KSP. It worked (as long as I took enough fuel with me), it was easy, it did not require any external tools. I see no shame in using that method if you don't care whether you're efficient.

I don't use this method anymore. Instead, I perform ballistic transfers using alexmoon's planner. It requires an external tool and learning a few tricks how to deal with inaccuracies caused by long burns deep down in gravity well but in general I am more happy with it. Also my rockets don't need to be as big.

[Odielthen]

I've learned a lot from this thread! I'm going to have to try to get much closer to my target from LKO. Right now I just pick a target and go, and hopefully I have enough dv to get there!

[kanvil]

Here's another way to think about the Oberth Effect. Getting to a higher orbit or an escape trajectory is about getting your energy high enough to climb up the walls of the gravity well. However, rockets operate by changing your velocity. So the goal is to get the most change in energy from the smallest change in velocity, and that happens when you're moving as fast as possible. The fastest part of your trajectory is when you're deepest into the gravity well, so that's where you should make your velocity change in order to get the most change in energy.

Another thing I just realized is that when you want to go from a highly elliptical, inclined orbit (such as just after aerocapture) to a circular orbit for rendezvous, it's best to circularize first, then change your inclination. This is because both the ascending and descending nodes are going to be close in for a highly elliptical orbit and so the velocity will be very fast, making plane changes expensive. By circularizing first, you push the nodes further out, making the plane change less expensive.

[NathanKell]

It's not just the Oberth effect; usually concomitant with this method is not caring about *when* you do your transfers. But since both the origin and destination bodies are orbiting, not stationary, then you need to consider their relative alignment.

[Renegrade]

It's not just the Oberth effect; usually concomitant with this method is not caring about *when* you do your transfers. But since both the origin and destination bodies are orbiting, not stationary, then you need to consider their relative alignment.

Actually, if you transfer into a very gentle solar orbit, and you don't have any time limit, you can just wait for a window so you don't have to make corrections and a second pass.

You'd still have to pass on any starting Oberth savings, of course.