Rule of thumb for interplanetary injection?

[evxxvi]

Hi all, the title says it: what is your rule of thumb for planning interplanetary injection maneuvers? To be more specific: only with in-game available tools, i.e., maneuver nodes. After some practice I can easily dock my space station from whatever kerbular orbit or go from Mun to Minimus or vice versa by intuition+maneuver nodes. I didn't get a deterministic rule of thumb for getting to Duna or Eve, even after some interplanetary practice following tutorials which magically bring me to the right trajectory. So how do you plan your interplanetary travel?

Edited November 25, 2014 by evxxvi
Answers were given

[Norpo]

To be honest, I know mun and minmus like the back of my hand by now, and for interplanetary I just use the Kerbal Alarm Clock transfer window calculator thingy.

Really though, I just escape kerbin prograde compared to it's solar orbit enough to give me a intercept with duna/outer planets for those, retrograde for Eve and Moho. All doing so when the transfer window is open.

[evxxvi]

Uhh, Kerbal Alarm Clock can do such things? Have to take a closer look...

[Norpo]

Uhh, Kerbal Alarm Clock can do such things? Have to take a closer look...

Yeah, you can set an alarm for transfer windows. I think it's the little icon with the blue and green arrows.

If that doesn't work, there's another calculator I know about, here: http://alexmoon.github.io/ksp/

It also gives you additional information about when exactly is the absolute best time to leave, and how much delta-v it would take if you wanted to go there faster. Nice, but I don't use it, because most of my games end up around year 8 and I don't want to have to type in my exact time whenever I use it.

EDIT: And the more advanced one I was talking about near the end, was actually turned into a ingame mod by the person who made kerbal alarm clock: http://forum.kerbalspaceprogram.com/threads/93115-0-25-0-Transfer-Window-Planner-v1-1-0-0-(Oct-16)

And since it can use the times ingame, it works without having to add that! Yay! (also works for RSS and stuff so that's nice if you use that)

Edited November 24, 2014 by Norpo
oooh, I didn't know about this

[Wanderfound]

Try to think of Kerbin itself as the spaceship that you're trying to fly to Duna or wherever. Figure out what direction the planet would have to burn to make the Hohman transfer (i.e. Kerbol prograde to raise orbit, Kerbol retrograde to lower orbit). Set your exit burn so that you're thrusting in the same direction as the planet would be if it was flying itself.

Edited November 24, 2014 by Wanderfound

[Crown]

Uhh, Kerbal Alarm Clock can do such things? Have to take a closer look...

Yes, Kerbal Alarm Clock can alert you when a transfer window for interplanetary travel opens. But it doesn't say how long it is and when it closes.

Wanderfound's hint is good if you want to calculate that stuff yourself (which is fun) or estimate the angle. And the base for all the rendezvous maneuvers you already did.

[Somtaaw]

I tend to eyeball, and use the commonly mentioned "pro/retrograde escape maneuvers", with a slightly overbuilt rocket. I'm usually doing my deceleration and orbital burn with the stage intended to start me going.

[Laie]

the title says it: what is your rule of thumb for planning interplanetary injection maneuvers?

To be more specific: only with in-game available tools, i.e., maneuver nodes.

Quite rule of thumb: place a note either on the sunny side or the oppsite. Pull prograde until your projected orbit touches on that of the target planet. Adjust normal until you get markers for "clostest approach". Then twiddle with the node until it all comes together. Two minutes of experimenting will tell you more than a thousand words could.

For a planned maneuver, I use this: http://alexmoon.github.io/ksp/ -- note that you can click the delta-V figure to get separate numbers for normal and prograde.

In the game, I'd first pull the normal until the node dV is as large as the normal component of the burn. Then pull prograde until you have the full dV. Then move the node back and forth until you get an encounter, or at least come close. Then refine the node. I'm actually using mechjeb to type in the numbers, but pulling the node's arms should be functionally similar.

[OhioBob]

So how do you plan your interplanetary travel?

Call me crazy but I enjoy the math. I calculate all my interplanetary trajectories using the method described in the following web page:

http://www.braeunig.us/space/interpl.htm#gauss

I put all the formulas into a spreadsheet. I input a launch date, a time of flight, and I then compute the trajectory that connects the departure and arrival points. It takes several iterations to find the energy optimum trajectory.

I know that there is a online tool that does the same thing, but that almost seems like cheating. I don't use any modes or tools created by somebody else. To me the fun comes in figuring it all out myself. I also like to compute my own delta-V. I find it very satisfying to plan a mission, compute the required delta-V budget, design an spacecraft to meet that requirement, and then successfully fly the mission as planned.

[MarvinKitFox]

Kerbal alarm clock has a decent transit calculator.

Mechjeb has one kick-*** porkchop-graph advanced transfer trajectory thingy.

But if you want to eyeball it, using pure stock only....

Using maneuver nodes for planning, take your ship's orbit to **just** outside kerbin's Sphere Of Influence. Absolute minimal residual speed out there.

It will still basically be in the same orbit as Kerbin, and such a teensy weensy distance ahead or behind that it doesn't really matter.

Now, on this Kerbin-lookalike orbit, add one more maneuver node. Make it loop out(or in) to the orbit of the planet you intend to visit.

Now pick up the node, and drag it around your Kerbin-orbit, until it gets an intercept with your target planet at per/apo gee.

Congratulations, you have just determined WHEN your departure window is, and the exact speed you need to be going when exiting Kerbin SOI.

Now, wipe all those fake maneuver nodes you just made.

When the time comes to depart, simply make a node that will send you in the right direction, and that will give you the correct speed at Kerbin SOI boundary.

The "right direction" will always be directly pro-or-retrograde along Kerbin's orbit line.

.

[tsotha]

Mechjeb has one kick-*** porkchop-graph advanced transfer trajectory thingy..

I've always resisted installing Mechjeb, but I might do it strictly for that feature. The porkchop plot tells you everything you need to know about going to another planet.

[AlexinTokyo]

If you want an in-game pork chop plotter and don't want MechJeb, try TriggerAU's Interplanetary Transfer Window Planner.

[technion]

I seriously don't understand how people do these without Precisenode. I mean, Duna is one thing.

But hitting Eeloo, you can find + or - one dv takes you from a good intercept to nowhere near it.

[cantab]

Kerbal alarm clock has a decent transit calculator.

Mechjeb has one kick-*** porkchop-graph advanced transfer trajectory thingy.

But if you want to eyeball it, using pure stock only....

Using maneuver nodes for planning, take your ship's orbit to **just** outside kerbin's Sphere Of Influence. Absolute minimal residual speed out there.

It will still basically be in the same orbit as Kerbin, and such a teensy weensy distance ahead or behind that it doesn't really matter.

Now, on this Kerbin-lookalike orbit, add one more maneuver node. Make it loop out(or in) to the orbit of the planet you intend to visit.

Now pick up the node, and drag it around your Kerbin-orbit, until it gets an intercept with your target planet at per/apo gee.

Congratulations, you have just determined WHEN your departure window is, and the exact speed you need to be going when exiting Kerbin SOI.

Now, wipe all those fake maneuver nodes you just made.

When the time comes to depart, simply make a node that will send you in the right direction, and that will give you the correct speed at Kerbin SOI boundary.

The "right direction" will always be directly pro-or-retrograde along Kerbin's orbit line.

.

Seconding this approach, I tried it the other way and it was no problem.

When it comes to tricky-to-reach places like Dres, you may have to just get a close approach and accept making a mid-course correction.

[Streetwind]

I seriously don't understand how people do these without Precisenode. I mean, Duna is one thing.

But hitting Eeloo, you can find + or - one dv takes you from a good intercept to nowhere near it.

Even with PreciseNode, you're not getting exactly the Eeloo periapsis you wanted anyway, because you can't execute the burn to that level of precision, especially if you have virtually invisible normal and/or radial components. Heck, MechJeb will probably struggle to do it precisely enough.

As such, the precision you can get with freehand maneuver nodes is more than enough. Just get any sort of intercept, make a burn for it (which may or may not loose you the intercept) and then one year down the road, make a mid-course correction that nails your intercept smack-dab where you want it. And then another couple years later, you arrive in Eeloo's SoI and perform another course correction because floating point errors accumulated over time and ruined your desired periapsis once more... PreciseNode is good for the comfort of editing maneuver nodes in a different, more detailed way, but it can't fix the inherent imprecision in the game itself, or that of its human users. The result is often no different from a haphazardly hand-calibrated node coupled with a mid-course correction of 2-3 m/s worth of dV. I used to obsess a lot about getting nodes just right, but nowadays I just fix the intercept on the way there. If that single-digit amount of dV screws up my mission, then I planned it wrong in the first place

But yeah, summary on "how to eyeball transfer windows, 100% stock":

- Put something in an equatorial low Kerbin orbit

- Create a maneuver node using that orbit

- Add 1000 ms/ prograde to start it off

- Drag the maneuver node around Kerbin until the escape vector lies parallel to Kerbin's orbit around the sun:

--- if you want to go inwards, the escape vector should point "backwards"

--- if you want to go outwards, the escape vector should point "forwards"

- Add some more dV to your maneuver node until the relevant apsis of the orbit around the sun just touches the orbit of the planet you want to reach, as seen from top-down view

--- Don't forget to occasionally move your node to keep the escape vector parallel to Kerbin's orbit as you add more dV

- Take note of the projected time to reach the relevant apsis, let's call it x

- Find out your target planet's orbital period by focusing it and checking the little info window, let's call it y

- Compare these two times:

x is the time you will be moving towards your target, but it is also the time your target will be moving along its orbit while you are in flight. Therefore, if you want your target to be at the apsis of your transfer orbit at the same time as you arrive there, you need to figure out where the target must be at the time you are leaving. You must "rewind" its position starting from the transfer apsis (the desired intercept point) by x amount of time. Conveniently we also know y, which is the time it takes the target to make one full orbit. By simply dividing x/y, you get a fraction which is equal to the fraction of the orbit that the target planet moves while you are in transit.

Super simple example: your target planet has an orbital period of 4 years, and you have a projected travel time of 1 year. Calculate: 1/4 = 0.25, which means that the target planet completes one quarter of its orbit while you are in transit. That means your transfer window, i.e. the ideal point of departure, is when the target planet is one quarter of its orbit away from reaching the desired intercept point.

Now comes the part where inner planets orbit quicker than Kerbin which orbits quicker than outer planets:

- If you're going to an inner planet and the target is closer to the intercept than you need it to be, you missed your transfer window and you need to wait for your target to come around the sun again.

- If you're going to an inner planet and the target is further away from the intercept than you need it to be, the next transfer window is coming up soon because your target planet is "closing in" on the desired intercept point, relatively speaking.

- But if you're going to an outer planet, the situation is reversed:

--- If the target is closer to the intercept than you need it to be, the next transfer window is coming up soon because the target planet is "receding away" from the desired intercept, relatively speaking.

--- If the target is further away from the intercept than you need it to be, you missed your transfer window and you need to wait for Kerbin to go around the sun once more.

Obviously this doesn't take into acount things like inclined and eccentric orbits, but it should give a rough estimate of where the transfer window is. At least with this method, you know that you don't even need to bother trying to get an Eve transfer while Eve is ahead relative to Kerbin (what people mean when they talk about "phase angles"). Unless of course you want to spend a lot of extra time or a lot of extra dV... generally even both. If Eve was in roughly the right place, though, you could get an intercept realtively easily, experiment with the transfer node and decide for yourself if leaving now, or waiting a little, or leaving even sooner next time would be beneficial to you.