the most "efficient" suicide burn.

[datdenkikniet]

Since I started playing KSP, I have always landed on other planets using Suicide Burns (except for the ones with an atmosphere). I was thinkink "what is the most efficient way to do this?" and had no idea, so I decided to ask it on the forums. (With suicide burn, I mean: killing all of your horizontal velocity whilst on an escape trajectory)

I have two examples:

Example 1: Doing the suicide burn up high, with a longer decent (Where "here" is the place where I do my suicide burn) .

(I've drawed everything, because I am not that good at explaining things in english)

Example 2: Doing the suicide burn (almost) at PE, with a shorter descent, but a suicide burn which costs more deltaV. (Where "here" is the place where I do my suicide burn)

Thanks in advance!

- Datdenkikniet

Edited March 16, 2014 by datdenkikniet

[maccollo]

Example two is always more efficient, preferably just a couple of km above the surface or as close as you dare.

[datdenkikniet]

Example two is always more efficient, preferably just a couple of km above the surface.

Thank you very much!

[Superfluous J]

I believe you're using the wrong terminology. A "suicide burn" is a specific technique for burning at *exactly* the last second, full burn right before crashing into the ground. Not 1 second before crashing into the ground, but 1 second before the last moment you can start burning and not die. If you wait 1 second, you'll crash. If you burn one second too early, you'll end up stopping dozens of meters (at least) above the terrain.

They're very hard to time (without mod help or lots of F5/F9) and very catastrophic if you make a mistake, henc the name.

To answer your question, you want to do #2. The burn to slow down at that point is a lot more than the burn to slow down at the point in #1, but once you've slowed down at the periapsis in #2, you're essentially landed. If you slow down at #1 you have to fall into the planet and you'll end up using more fuel before you come in for a landing.

IMO though the more important thing is pinpointing your landing, and that's far easier to do by setting up a collision course and then burning well ahead of time, and then coming down the final KM or so more or less vertical. I'd rather be where I'm aiming than land with a few extra dozen m/s of delta V.

EDIT: I shoulda known typing this long a response I'd get ninja'd

[datdenkikniet]

-snip-

Okay, thanks for telling me this! (I feel like writing more, but I have no idea what to write :s)

[Superfluous J]

Oh one more thing: I just looked at your diagrams again and I was answering as if the arrow on #1 was at the other side of your orbit. whatever you do, you don't want to burn to come in for a landing as you're moving AWAY from Mun.

[datdenkikniet]

Oh one more thing: I just looked at your diagrams again and I was answering as if the arrow on #1 was at the other side of your orbit. whatever you do, you don't want to burn to come in for a landing as you're moving AWAY from Mun.

oh, It most certainly wasn't my intention to do that! The arrow has to go the other way.

Could you also please tell me how on earth I change the prefix of this post?

edit: oh, nvm, You have to go in "advanced" mode to do that :>.

Edited March 16, 2014 by datdenkikniet

[Pds314]

Neither method is either suicide burn or highly efficient. Remember that whenever you moving slower than you could be and burning, you are wasting fuel. That is what real suicide burns are for. Stopping at the ground. It takes some planning and number-crunching (or quick guessing if you are near-suicidal).

Being 1 second off on an initially 1000 m/s 5 G burn will leave you either stopping 1 km from the ground or still moving at about 50 m/s on impact.

[Rickenbacker]

A maneuver node placed roughly at ground level and adjusted to zero out the speed at the node helps. All you have to do to then is burn at 100% when you reach the beginning of the estimated burn time. Even if you put the node a few hundred m above the surface, you'll still get pretty damn close to a real "suicide burn".

Didn't the Apollo landers use one continuous burn from orbit to touchdown, by the way?

[Kasuha]

Suicide burn is not in general the most efficient way of landing, but it is (IMO) the most efficient way of landing if you get all parameters (periapsis height and the exact moment when to start the burn) exactly right. It is extremely hard to get them right. If you don't get them right, you either spend more fuel than you would if you used more efficient method of landing, crash into surface at high speed, or stop moving at certain, usually significant distance from the surface.

Apollo landers did not use suicide burn for landing. Their landing consisted of two parts, braking and approach. During braking the lander was burning above retrograde to kill horizontal velocity while keeping the lander sliding above terrain without crashing to it. Only when horizontal velocity was low enough the lander entered approach phase during which the trajectory was smoothly changed from almost horizontal to vertical and speed was reduced down to zero as it was getting near the ground. Similar approach is referred to as "horizontal landing" and it is about the safest and most effective practically achievable way to land in KSP as well.

Decent approximation to it is:

1/ get a low periapsis (as low as possible but still above terrain)

2/ put a circularizing maneuver at the periapsis

3/ burn the maneuver the standard way (half time ahead)

4/ during the burn, delete the maneuver and continue killing your horizontal velocity while keeping your vertical speed low or zero by pitching above the retrograde marker

5/ when your speed decreases sufficiently, let the ship approach the ground and land

Edited March 17, 2014 by Kasuha

[SSSPutnik]

You get more Oberth boost in option 2 due to higher speed when you start braking.

[FREEFALL1984]

Set yourself onto a collision course with the planet, and then place a maneuver node on (or a couple of hundred meters above) ground level to kill all of your velocity. You can even get MJ to burn your node for you and it'll always leave you few feet above surface for a soft touchdown.

[maccollo]

Set yourself onto a collision course with the planet, and then place a maneuver node on (or a couple of hundred meters above) ground level to kill all of your velocity. You can even get MJ to burn your node for you and it'll always leave you few feet above surface for a soft touchdown.

That is terrible for efficiency. Burning vertically means you are getting high gravity losses during the burn, even if you get it perfectly right.

More likely your'e going to start way to early, which increases gravity losses further.

Edited March 17, 2014 by maccollo

[FREEFALL1984]

That is terrible for efficiency. Burning vertically means you are getting high gravity losses during the burn, even if you get it perfectly right.

More likely your'e going to start way to early, which increases gravity losses further.

Haha I never said it was pretty but it's effective. Also you shouldn't need to burn vertically, I didn't mean collision course like a 90degree impact, just a glancing blow with a trajectory only barely touching surface. you'll only be killing orbital speed so should't be too bad

[maccollo]

Haha I never said it was pretty but it's effective. Also you shouldn't need to burn vertically, I didn't mean collision course like a 90degree impact, just a glancing blow with a trajectory only barely touching surface. you'll only be killing orbital speed so should't be too bad

Ah, well in that case it's not as bad. It's not to unsusual for people to think that going straight for the planet is more efficient. I apologize for the misunderstanding

I think it's still better to have the periapsis a few km above the surface. That way you can control the descent rate during the burn and end up a few 100 meters above the surface when you come to a near stop.

[Kasuha]

Actually the "set collision course and put a maneuver" method is not that bad. Yes, it is less efficient but depending on your TWR you will spend 5%-25% fuel more (the less TWR the worse efficiency) and it is rather good to land at pre-selected spot.

This is Kerbal Space Program, you don't need to fly with least necessary fuel, not in most cases at least. So it's good to know what the efficient approach is in case you get into tight situation, but otherwise it's usually better to do things the easy and straightforward way.

[maccollo]

This is Kerbal Space Program, you don't need to fly with least necessary fuel, not in most cases at least. So it's good to know what the efficient approach is in case you get into tight situation, but otherwise it's usually better to do things the easy and straightforward way.

Sure, but the question asked by OP concerns efficiency, not which method is most straight forward.

[Joe32320]

Sure, but the question asked by OP concerns efficiency, not which method is most straight forward.

It is a related question though, even if it is mathematically the most efficient, there is also the practical aspect of being able to do it and not have your kerbals splattered across the Mun's surface etc.

[seyss]

You get more Oberth boost in option 2 due to higher speed when you start braking.

in real life yes, but do you think Squad implemented this?

[Warzouz]

I think it's not about implementation, but about eliptic orbits and acceleration.

[thadisjones]

in real life yes, but do you think Squad implemented this?

No, they didn't implement it deliberately. What KSP does is simulate a system of physics accurately enough that a basically correct analogy to the real-world Oberth effect appears in a significant manner inside the simulation.