Planning Interplanetary Trajectories with Plane Changes?

[Srpadget]

Looking generally for tips and tricks for more-efficient planning of burns, here.

I'm just starting to foray out beyond Kerbin's SOI, with a mission partway to Duna and in the planning stages for an upcoming launch window to Dres. The latter of which has me wondering about how to plan the burns given the significant plane change involved.

Duna was pretty straightforward, partly because of Duna's very small orbital inclination and partly because I happened to luck out timing-wise and launched just a few days prior to passing a node line -- so it happened that an in-ecliptic transfer orbit was 'good enough' to get me into Duna's SOI (which of course will be sitting right near the opposite node line when I arrive in a couple hundred more days).

But that's not going to cut it for Dres. It seems I'm going to have to set up the Hohman transfer "blind", with the apoapsis far above or below Dres SOI (due to the difference in orbital planes) and thus no feedback regarding Dres' position when I reach apoapsis...until after I've done a plane change well into the voyage. And at that point, fixing a badly-planned (or -executed) burn will be terribly terribly dV-costly.

So there seems to be substantial risk to my default plan of "burn out of Kerbin orbit into Hohmann transfer, then plane change at node line, then fine-tune as necessary". Even though I plan to practice with a couple Mun-to-Minmus runs in the run-up to my Dres launch window, Dres will be much much trickier (tiny SOI, ever so far away, by comparison).

And hence, I'm wondering: how do other people plan the burn sequence for trips to planets with highly-inclined orbits? (Manual only, please; eventually I may make life easier with Mechjeb and/or other automation mods, but for now I'm doing everything manually just to prove that I *can*. Clever ways to extract more usable info from the guts of KSP, that's fine; anything that does the work FOR me is right out.)

Edited June 20, 2014 by Srpadget
Answered!

[Taki117]

Have a look at this.

http://alexmoon.github.io/ksp/

[allmhuran]

If you're willing to drop the requirement of launching during a window of minimum transfer time, things get a lot easier.

Just launch when Duna is passing through the ascending/descending node of Dres, and get your intercept on the other side. You will still have to "pay for" the cost of the inclination difference because your insertion burn will be larger, but the intercept itself becomes much easier.

Now, clearly the timing may be off if you do this, and Dres probably won't be there when you get there. Easy solution that uses minimum delta-v:

1. Put a maneuver node down at the intersection of the two orbit lines (ie, the opposite side of your solar orbit from your Duna escape). This is where you are going to get the intercept, and will be your closest approach. Don't put in any delta-v changes yet.
   
2. Put a second maneuver down anywhere on the orbit resulting from the first maneuver.
   
3. Now collapse the second node and click the lower right circle, which sets the maneuver to be one orbit ahead. The closest approach position won't change (if the orbit lines are intersecting, that's clearly the closest approach). But watch the target position shift.
   
4. Keep clicking "one orbit ahead" on the second node until the target position at closest approach is "pretty close" to the closest approach. Then drag the handles of the *first* node as required to change your orbital period such that you get an intercept.
   
5. The required delta-v will be smaller and smaller the more "orbits ahead" you look, because the change in your period will accumulate over all of those orbits, but it will mean looping around in solar orbit for longer. Conversely you can make a larger up front change for more delta-v, with the upside of not having to wait as long.
   

Lots of examples of this technique on my youtube channel (see below).

Edited June 19, 2014 by allmhuran

[Specialist290]

Seconding Taki117's post. The Launch Window Planner should be able to tell you everything you need to know about when and where to plot your maneuver from orbit. You may still need to make a few slight adjustments to get spot-on, but it'll definitely put you on the right track.

EDIT: I'll also vouch for allmhuran's technique as well.

[Srpadget]

OMG...the Launch Window Planner is a THING OF BEAUTY.

I had just gotten out my old astrodynamics text from college, and was going to build a utility much like that. (Or rather, a utility like a much less-capable, less-user-friendly, less-versatile, lobotomized version of that....)

I begin to suspect that NASA is working with the devs on KSP at least partly because this community has developed better preliminary-mission-design tools than NASA has. (And I get to say that, because I've worked with some of those tools.)

Okay, so now for the 'forum newbie question': How do I tag this thread as "Answered"?

Edited June 20, 2014 by Specialist290
Merging sequential posts

[Apollo13]

I use MechJeb for planning, not execution. NASA/ESA engineers don't use a clumsy, imprecise graphical tool to plot trajectories. Why would you? They use flight computers to model solutions. The KSP flight computer is called MechJeb.

Don't want MechJeb? use Protractor.

Okay, so now for the 'forum newbie question': How do I tag this thread as "Answered"?

Edit in Advanced mode. At the top is a drop-down where you selected "Unanswered".

And, as cantab says below, PreciseNode is an excellent tool for maneuver nodes. MJ has a maneuver node editor which is very precise.

Edited June 20, 2014 by Apollo13

[Specialist290]

Okay, so now for the 'forum newbie question': How do I tag this thread as "Answered"?

I've gone ahead and done it for you since you requested it, but for future reference: Edit your first post, then choose "Go Advanced" and change "Unanswered" to "Answered" in the drop-down menu.

[cantab]

But that's not going to cut it for Dres. It seems I'm going to have to set up the Hohman transfer "blind", with the apoapsis far above or below Dres SOI (due to the difference in orbital planes) and thus no feedback regarding Dres' position when I reach apoapsis...until after I've done a plane change well into the voyage.

If your projected apoapsis is beyond Dres's orbit, I believe you should still see the intersect distance markers when you pass near its orbit, even if you're not set to hit the SOI.

I use MechJeb for planning, not execution. NASA/ESA engineers don't use a clumsy, imprecise graphical tool to plot trajectories. Why would you? They use flight computers to model solutions. The KSP flight computer is called MechJeb.

Don't want MechJeb? use Protractor.

And if you want to plan your manoeuvres with numbers instead of draggy handles, try Precise Node.

[Srpadget]

NASA/ESA engineers don't use a clumsy, imprecise graphical tool to plot trajectories. Why would you?

Because I'm reveling in a universe with basically-correct physics in which realistic space travel is merely 'tricky' rather than 'd*mn near impossible' and hence CAN be done manually, of course. KSP has made my old job into something FUN again! YAY! :-)

At some point, when the newbie-shininess of it all has worn off, I will likely get MechJeb and/or other labor-saving tools. But for now, I'm rather enjoying doing it all by hand.