Orbital or Direct Landing on Surfaces without an atmosphere

[humzaq]

Hello, I'm sure this same question has been asked many times, but i've looked through the forums and I'm unable to find a satisfactory answer.

I understand that it is better to enter into a circular orbit when leaving Kerbin rather than blasting right through to your destination due to the effects of gravity.

So does this same reasoning apply for landing on a non-atmospheric surface? A direct landing will cause you to suffer gravity's effects every step of the way while an orbital landing will cause you to use deltaV to cancel out horizontal velocity.

I have hundreds of hours in this game and I still have these relatively simple questions. I believe that the answer is somewhere in the middle, as eliminating horizontal deltaV is unique to landings and not ascents.

Any explanations/experiences would really be appreciated!

[laythelover17]

I'm pretty sure that for saving delta V you would want to go with orbiting but the more direct approach would be effective too. as it would mean less time to get there and land. However I do think that the landing would be harder to estimate but it would be easier to land when orbiting as you can go down as gently as you want. so I would say that if you have a lot of thrust and Delta V then go direct but if you have very little thrust/Delta V then go with orbiting.

[Motokid600]

You mean how efficient it is to perform direct drcents as apposed to entering into orbit first? I'd say it's just a tad more Dv needed for a direct descent. Unless you position your pe real low and burn just a few thousand meters above the surface I imagine it might be around the same.

I do know that it's been done in real life with the first us probes to the moon. They fired solid motors iirc to kill the horizontal speed and directly descended to the surface.

The difference only is if you happen to have a specific landing site.

[humzaq]

Thanks for the replies. So both of you believe that its slightly more efficient to enter into an orbit before landing. I also agree that it easier to land from an orbit as you can take it slow.

Also, it seems that the difference in landing direct vs from orbit compared to ascending direct vs orbit is much smaller due to the required horizontal deltaV kill and the shorter descent time under gravity. So much so that landing direct vs from orbit isn't discernible.

I love orbital mechanics. Thanks for improving my understanding!

[humzaq]

Thanks for the replies. So both of you believe that its slightly more efficient to enter into an orbit before landing. I also agree that it easier to land from an orbit as you can take it slow.

Also, it seems that the difference in landing direct vs from orbit compared to ascending direct vs orbit is much smaller due to the required horizontal deltaV kill and the shorter descent time under gravity. So much so that landing direct vs from orbit isn't discernible.

I love orbital mechanics. Thanks for improving my understanding!

[Orbital Vagabond]

Thanks for the replies. So both of you believe that its slightly more efficient to enter into an orbit before landing. I also agree that it easier to land from an orbit as you can take it slow.

Also, it seems that the difference in landing direct vs from orbit compared to ascending direct vs orbit is much smaller due to the required horizontal deltaV kill and the shorter descent time under gravity. So much so that landing direct vs from orbit isn't discernible.

I love orbital mechanics. Thanks for improving my understanding!

It's true that going to orbit first is slightly more efficient for Mun, compared to a vertical ascent (BTW "Direct Ascent" is not a correct term for that trajectory).

However, as the body's gravity increases (e.g. Tylo), the difference between the two methods increases as well, with the parking orbit always being more efficient.

This has been discussed in some depth in a very recent thread here.

[tutike2000]

A suicide burn is the most efficient way to land. If you first circularize then land, you're essentially making the same effective burn in orthogonal directions. It's shorter to walk the diameter of a rectangle (suicide burn) than to walk its sides (circularize to kill some of your horizontal velocity, then land).

Think of it in terms of the Oberth effect. Given a certain kinetic energy at your starting point (let's say a circular orbit at the edge of the SoI), the best place to use your delta-V is as close as possible to the surface.

- - - Updated - - -

A suicide burn is the most efficient way to land. If you first circularize then land, you're essentially making the same effective burn in orthogonal directions. It's shorter to walk the diameter of a rectangle (suicide burn) than to walk its sides (circularize to kill some of your horizontal velocity, then land).

Think of it in terms of the Oberth effect. Given a certain kinetic energy at your starting point (let's say a circular orbit at the edge of the SoI), the best place to use your delta-V is as close as possible to the surface.

Actually... wait... this may depend on your TWR. Or rather... how high up do you need to start your suicide burn?

If you need to start burning too high (TWR close to 1.0), circularizing may be a lot more efficient.

Edit: related discussion: http://www.reddit.com/r/KerbalAcademy/comments/1wxt6v/how_is_it_best_to_land_on_planets_without/

Apparently, somebody claims that constant-altitude landings are more efficient, performed after circularizing. However I think this doesn't account for the delta-v needed to circularize.

Edited June 22, 2015 by tutike2000

[cephalo]

I've tried to land on Mun from a very high altitude, thinking that since my orbital speed is so low I might save some dv, but what happens is that you pick up a ton of speed on the way down. I found that the dv requirement was around the same. You either have to kill that 600ms horizontally or vertically. The lower your orbit, the less time there is to build up vertical velocity. I guess that means the more expensive path is the middle ground between those two.

[tutike2000]

When in doubt, just remember that a landing is literally a takeoff in reverse.

Every maneuver or burn that you do to land, if you do in reverse you end up where you are now.

So what's the most efficient way to take off from an airless body?

- Full throttle,

- Pitch upwards only so much that your vertical speed is positive and doesn't change

- Stop once your Ap is at the desired height

- Circularize at Ap

I think I just described a suicide burn in reverse

[Starhawk]

When in doubt, just remember that a landing is literally a takeoff in reverse.

Every maneuver or burn that you do to land, if you do in reverse you end up where you are now.

So what's the most efficient way to take off from an airless body?

- Full throttle,

- Pitch upwards only so much that your vertical speed is positive and doesn't change

- Stop once your Ap is at the desired height

- Circularize at Ap

I think I just described a suicide burn in reverse

OK, but broaden the question slightly.

What's the most efficient way to leave an airless body.

-Full throttle.

-Point at the horizon and pitch up just enough to have positive vertical velocity and to clear terrain.

-Burn until you've reached the desired escape trajectory.

In practice, this is impractical. It's very difficult to take off at just the right time so that your ejection angle will end up taking you where you want to go. It's much simpler to create an orbit and then set up your timing for your ejection from there.

Nonetheless, the opposite of the process described above would be the most efficient landing on an airless body. You would adjust your trajectory from your original ejection such that you have an extemely low periapsis. Very close to the surface while still clearing terrain.

You would then 'suicide burn' at the appropriate moment (near your periapsis) such that you are just above the surface as you finish removing all your horizontal velocity.

Once again, it is almost universal practice to circularize first for several reasons. Not least of these is that orbit is a stable situation, whereas a hyperbolic trajectory is not, requiring you to finish the landing before you can turn your attention elsewhere.

Starting from low orbit will allow you to start with much lower relative velocity and you won't have as long a burn required for landing.

Starting from orbit allows you a choice of landing sites.

tl;dr Mathematically it's more efficient not to circularize, but in practice circularization before landing is the way to go.

Happy landings!

[Prepper-Jack]

It's normally optimal to circularize and come in as low as possible to land, as you're working with gravity, not fighting against it.

If you come into the Muns sphere of influence in a collision course, approximately 2Mm away, traveling at 300 m/s - you will have gravity pulling on you the entire 2 million meters, leading to a massive sustained suicide burn in the end. I think I saw in some thread recently that someone calculated this to around 2,700 m/s of dV - and that doesn't sound too far off the mark (though I haven't done the calculations myself).

Alternatively, you could come into a high orbit at low speed, circularize for 50-100 dV, transfer to a very low orbit for 100 dV, then suicide from that orbit with around 550-600 dV. Quite a bit of savings.

[Snark]

Details below, but the TL;DR version is: on an airless world,

- The theoretical best is to come in almost horizontally (so that periapsis just barely kisses the surface) and then suicide burn to land.

- However, in practice, a more useful option is typically to come in with a very-very-low periapisis that doesn't quite touch the ground, circularize into the lowest possible circular orbit, then to land do a light retro burn followed by suicide burn to the surface. It costs a teeny bit more dV to do this than the theoretical optimum, but the difference is quite small, and in return you get a lot more flexibility in choosing landing site.

That's the "what"; the "why" is a bit more lengthy, details below.

The main principle for dealing with an airless body (whether you're landing or taking off):

1. Do as much of your burn at as low an altitude as possible.

2. Always burn prograde only.

3. Burning horizontally at low altitude is better than burning vertically.

#1 is due to Oberth effect.

#2 is to avoid cosine losses; it provides for the most efficient conversion of thrust to kinetic energy.

#3 is due both to Oberth effect and to minimize gravity losses. If you're burning horizontally, it means you can do a long burn while still at low altitude (thus maximizing Oberth effect); it also means that your burn is mostly at right angles to gravity rather than fighting directly against it, so you minimize gravity losses.

#1 and #2 apply equally to all rocket designs regardless of TWR. #3 becomes less relevant as TWR increases; the higher the TWR, the less of a difference it makes.

#1 and #2 simply say that a suicide burn is the best way to land; they don't make a lot of difference for vertical-versus-horizontal trajectory on land, or should-I-circularize-or-not.

#3 is where the difference comes in. For sizable bodies with significant orbital/escape velocities, a burn will typically take at least a few dozen seconds unless you have a freakishly high TWR.

Edited June 22, 2015 by Snark

[Laie]

I can't believe that this has been going for two pages without one of the elders chiming in.... what you are looking for is called a

constant altitude landing(at least in the KSP world)

And the OP is right, this has been investigated to death several times over.

There's a nice demo video from Kosmo-not, here:

Data tables: http://forum.kerbalspaceprogram.com/threads/39812-Landing-and-Takeoff-Delta-V-vs-TWR-and-specific-impulse

As a rule of thumb: plotting a collision course, then suicide burn is acceptable if you have a very high TWR. The lower your TWR, the better it is to do the constant altitude thing.

[SanderB]

In all my experiments with low TWR engines horizontal landings are always best. Coming in to your target landing zone at any angle other than almost horizontally induces gravity losses. Burning the engines in any direction that is more than 30° away from the horizontal is going to waste your dV unless you have a TWR of a lot more than 10. The only way that prevents a lot of vertical burning as possible is circularizing first and then burning retrograde at a distance (to target) dictated by Velocity^2/(2*Acceleration).

Edited June 22, 2015 by SanderB