How to plan a slingshot manuever? Possible from the ground?

[Arch3rAc3 80]

Hey, all.

I never tried a slingshot maneuver and was looking for some tips on how to even begin thinking about it.

While the definition seems simple enough -- using another planetary body's gravity to propel me further towards my target, saving fuel/Delta-V -- I'm totally clueless on how to put it into action.

Using only stock tools (as I'm playing on the Xbox), what's the best way to plan for a slingshot with the Mun to reach Duna for example? How do I even figure out how much Delta-V I'll require beforehand?

I was thinking the best way to plan it would be by trial and error in the Orbit screen, plotting burns until it gets to where I want. But if this is the best way to do it, it means I must have my craft already in orbit to begin with, right?
Is there no way for me to plot this from the ground before launching my craft? And - only 1 more question, I promise xD - how do I know when is the best time to launch my craft - is it when my target planet is closest to Kerbin, or when the planet is a bit retrograde so it will be closest to Kerbin by the time my spaceship gets there; or am I'm overthinking way too much and the Delta-V difference is just negligible in both cases?

Thanks!

[Zhetaan 797]

Let's go through this in order:

18 hours ago, Arch3rAc3 said:

Using only stock tools (as I'm playing on the Xbox), what's the best way to plan for a slingshot with the Mun to reach Duna for example?

Your example is not the best; the Mun cannot give enough of an assist to get you all the way to Duna.  However, it can take 500-ish m/s off of the delta-V cost.  The problem is one of trade-offs:  at minimum closing velocity (which is also minimum fuel, i.e. a Hohmann transfer), the Mun gives enough of an assist to throw you into deep space, but not enough for you to reach anywhere but Kerbin.  You can burn towards the Mun for faster closing velocity, but then your trajectory will turn less and you'll get less from the assist.  The following post offers a lot more information.  It's old, but gravity hasn't changed much in the past six-and-a-half years:

You might consider a form of practice to test the theory:  consider that if the Mun can throw you into deep space, then it can get you an assist to Minmus.  Played the other way, a flight from Minmus to the Mun can give you braking to a lower Kerbin orbit.  Remember that all of the rules for interplanetary transfer apply to trips from the Mun to Minmus and vice versa.  Adding assists to it is not difficult, and the best part is that the Kerbin system repeats its configuration so quickly that you can get a good intuition for how to put together these kinds of gravity assists before you try it on planets orbiting the sun.

18 hours ago, Arch3rAc3 said:

How do I even figure out how much Delta-V I'll require beforehand?

You'll need a delta-V map (or to calculate the transfer cost yourself) and you'll need some idea of how much you'll save from the assist.  Let's assume that you found a great assist or series of assists.  Then you need the delta-V to launch to the first fly-by, and the delta-V necessary to brake at your destination, and ... that's it.  Ignore the rest of the delta-V map.  You'll need a manoeuvring reserve for corrections, but not having to burn it is rather the point of saving propellant.

Put differently, gravity assists are not about delta-V so much as they are about timing.  Normal interplanetary transfers are about timing, too, but your use of propellant allows you to shift the variables to favour your mission:  you can set the correct direction and speed by the use of propellant.  With gravity assists, you are totally dependent on arriving at your destination at the right time so that your flyby sends you in the correct direction and at the correct speed without further input from you.

18 hours ago, Arch3rAc3 said:

I was thinking the best way to plan it would be by trial and error in the Orbit screen, plotting burns until it gets to where I want. But if this is the best way to do it, it means I must have my craft already in orbit to begin with, right?

That will definitely work.  I'll caution you to ensure that you set the patched conics limit to something high, like six or more:  if you don't, then the map view will stop displaying your predicted trajectory before you know whether it will actually get you to your destination.

If you want to plan it by trial and error, then yes, you'll need to have something in orbit.  It need not be the vessel that will make the journey; you can use something else that is already in orbit and it won't be a problem.  This is generally desirable, though:  starting from low Kerbin orbit ensures that you can launch within about fifteen minutes of the ideal transfer window (since low orbit has a period of about thirty minutes).  Starting from the ground leaves you with a minimum margin of three hours.  This is usually surmountable, however, because you can build corrections into the launch burn.  More important is the fact that starting from orbit offers more potential savings via the ability to split the launch burn into several well-timed apoapsis-lifting burns, which saves propellant in low-thrust vessels such as ion-powered probes.

18 hours ago, Arch3rAc3 said:

Is there no way for me to plot this from the ground before launching my craft?

There are two ways.  One involves the use of a tool called Flyby Finder.  It's a good tool, albeit it comes with some limitations.  It cannot calculate double assists, and it is not the most easy-to-use.  Read the manual first.  The initial post for Flyby Finder also has a link to a spreadsheet that can calculate double assists, but it's definitely not user-friendly.

The second involves figuring out what you want to do by hand.  This is not quite so daunting as it seems, because gravity assists are not so much difficult as they are picky about timing.  Provided that you know the orbital periods of the planets, you can set up simple assists just as easily as interplanetary transfers--which are not easy, of course, but it is possible with the stock tools.  Once you have chosen a path, it's simply a matter of picking a launch time that fits the window.  You'll need to do a bit of correction, simply because there's no way to reliably repeat a launch without a launch script or other automating mod, but it is absolutely possible.

18 hours ago, Arch3rAc3 said:

And - only 1 more question, I promise xD - how do I know when is the best time to launch my craft - is it when my target planet is closest to Kerbin, or when the planet is a bit retrograde so it will be closest to Kerbin by the time my spaceship gets there

Neither.  You want to time it so that the planet is where your spaceship is by the time your spaceship gets there.  Once you leave Kerbin, Kerbin's location ceases to be relevant.  It is the shape of your vessel's orbit that is the important part.  You are correct that timing that launch is something that can be related to the relative positions of the planets; the angle between your origin planet, the sun, and your destination planet is called the phase angle and it's different for every interplanetary path.  This is because the orbital periods of the planets are different, and more importantly, the ratios of the orbital periods are different.

18 hours ago, Arch3rAc3 said:

or am I'm overthinking way too much and the Delta-V difference is just negligible in both cases?

There is a Kerbin-Eve-Kerbin-Kerbin-Jool path that can get you to Jool for less than 1,100 m/s; the typical transfer cost is about double that.  Provided that you get the timing correct, there is a lot of delta-V to be saved by using Eve to go anywhere and Tylo or Laythe to go anywhere at Jool.

For example, you can do this:

I'll include a link to the general tutorial on the subject:

This is a YouTube video that covers a lot of the basic principles, and it also has the benefit of being less than a year old:

I know that we're on version 1.9 rather than 1.7, but a lot of the other tutorial videos out there are from version 0.90 or earlier.

Good luck!

Edited February 24 by Zhetaan

[Snark 13,525]

On 2/23/2020 at 6:36 PM, Arch3rAc3 said:

Using only stock tools (as I'm playing on the Xbox), what's the best way to plan for a slingshot with the Mun to reach Duna for example? How do I even figure out how much Delta-V I'll require beforehand?

Are you doing this for the sheer challenge of it (i.e. "hey, how can I make this harder and therefore more fun")?  Or are you trying to make things easier (i.e. "if I slingshot I don't need as much dV and that will make the mission easier")?

If it's the former, sounds great.  If it's the latter... honestly, as a matter of practicality, I think most folks find it more trouble than it's worth.  Going to Duna from LKO only takes about 200 m/s more dV than it takes to go to the Mun.  That means that you can save at most 200 m/s of dV by doing a slingshot (because you have to at least reach the Mun)... and in exchange for that, you make the navigation much more challenging.  Seems easier just to stick another 200 m/s of dV onto the ship, honestly.

But if you're just into it for the challenge, sounds like a great idea!

[Arch3rAc3 80]

23 hours ago, Snark said:

But if you're just into it for the challenge, sounds like a great idea!

More for the fun of it. I understand it's not really necessary to be done in KSP, but it seems to be a pretty common thing in real life.

Watching Juno's trajectory for example: I have no idea how to even plan something like this in KSP (much less how they can even calculate it in real life xD), but would love to hear some tips in regards to it for KSP. 

 

Edited February 26 by Arch3rAc3
Better link

[AHHans 603]

8 hours ago, Arch3rAc3 said:

More for the fun of it. I understand it's not really necessary to be done in KSP, but it seems to be a pretty common thing in real life.

Well, in real life adding more dV to an interplanetary probe is way more expensive than computer and scientist time. So they spend serious amounts of computer an manpower to optimize the flight path to use as little dV as possible. (Well, probably within the boundary of not extending the flight time too much.) Planning such a trajectory is probably as much an art as it is science. I think one thing they do - and which would be hard to do in KSP - is that they plan the path as much backwards as they do forwards, i.e. they not only start with the trajectory around Earth and what maneuver would get them closer to the target, but also with their target orbit (around Jupiter or wherever) and what maneuver would have gotten them there from an orbit closer to Earth. And then try to match both up.

I'm sure you also could do this in KSP - which would be a lot easier because KSP doesn't use multi-body gravity - it is not what I consider "fun".

[Curveball Anders 1,492]

There's also an out of game tool.

 

[Arch3rAc3 80]

On 2/27/2020 at 5:17 AM, Curveball Anders said:

There's also an out of game tool.

 

That looks amazing!

I need a better PC ASAP, the game looks 100x better and more complex than in my limited Console version.

[Arrowstar 773]

9 hours ago, Arch3rAc3 said:

That looks amazing!

I need a better PC ASAP, the game looks 100x better and more complex than in my limited Console version.

Hey there, I'm the author of KSP Trajectory Optimization Tool.  Please let me know if you have any questions about the tool, either PM or a message in the KSPTOT tread will work fine.  The maneuver planning tools are pretty simple to get the hang of, and the full up mission planning tools are not too bad either once you get the hang of them.