Efficient Kerbin Escape

[ScottyDoesKnow]

Quick question about the best time to burn while escaping kerbin orbit. Normally I line up kerbin and the sun and then burn perpendicular in the direction I want to escape. Should I instead be burning earlier so that my escape curve ends up perpendicular to the sun-kerbin line? If so, are there any easy tricks for eyeballing this?

Edit: the answer is to burn earlier so that my escape curve ends up perpendicular to the sun-kerbin line. Only trick to it is adding the maneuver node and dragging it around until this happens.

Edited October 24, 2013 by ScottyDoesKnow

[Dedjal]

Most efficient would be if you could manage to get gravital assist from mun and minimus.

Besides from that, I have no idea.

[Ariadne]

Usually if I plan escape Kerbin's sphere influence, I launch the rocket and fly directly up from Kerbin. I think this way is effecient because thus I don't waste fuel to gain tangential velocity.

[Superfluous J]

I have found that burning straight up is NOT the most efficient, but instead attaining orbit and then burning again once there. Saves a decent amount of fuel.

You can easily try this yourself by doing launches with the exact same ship.

[Urb4n0ninj4]

Usually if I plan escape Kerbin's sphere influence, I launch the rocket and fly directly up from Kerbin. I think this way is effecient because thus I don't waste fuel to gain tangential velocity.

I would like to see this tested, because this sounds like a good way to do it, and at the same time sounds bizarre.

edit* Ninja'd...I figured it sounded too easy to be true

[Oddible]

Most of the advice in this thread so far is counter to what more experienced players have posted, for instance burning "directly up" (or Radial Out) is dramatically less efficient than a prograde burn using your existing momentum (though I am by no means one of those more experienced players). There are a few resources I'll point out to help. First read up on a Hohmann Transfer. Second, here is a great tutorial with Kerbal details of how and when to implement interplanetary transfers. And finally a graphical calculator to help you line up your transfer. You can also get the fantastic addon Protractor to help do this in-game (though I don't know if it is updated for .22 yet and probably needs the part file tweak to add it to the tech tree). MechJeb can also tell you when to launch. My personal favorite is Kerbal Alarm Clock to make sure I don't miss a launch window.

[wasmic]

Also, as an addendum to the above this web calculator will tell you the exact time in days and the angle to kerbin prograde that you have to burn at. It's great for use along with the graphical calculator in the above post.

[KvickFlygarn87]

Usually if I plan escape Kerbin's sphere influence, I launch the rocket and fly directly up from Kerbin. I think this way is effecient because thus I don't waste fuel to gain tangential velocity.

Uh, yeah, no. Do you know why you do gravity turn and why you don't do it near the ground?

It's basically because if you fly straight up, gravity wil be the strongest force acting on your craft. But you once you've cleared the thickest atmosphere you won't be affected by drag as much when turned, so you turn.

Or something. I'm not one of the smarter in the KSP fanbase, but I think I have a good idea of how stuff works.

I think. If I don't, yell at me.

[Alistone]

Here is a tool to tell you your phase angles for celestial alignment and ejection burn angles for when to start your burn. http://ksp.olex.biz/

You can also play with the parking orbits to figure out what sort of orbit is best (lowest delta V) for a refueling station before you depart the Kerbin SOI. Turns out that for closer targets a higher orbit is good, but for further targets requires LOTS of deltaV you want to be less than 100km when you start your burn for maximum oberth effect.

Great tool though...

http://ksp.olex.biz/

[Tank Buddy]

You should burn so that when you are done with your burn your escape trajectory is parallel to the direction of Kerbin's orbit. I believe this is called the ejection angle. This way your velocity (direction part of velocity, not speed) are the same.

[SolarLiner]

If you want to get to outer orbits (objects that orbits further away), then burn in the middle of the night side of Kerbin. If you want to go to the inner planets, you burn on the middle of the light side of Kerbin. After that, depending on the accuracy of the window, you will have to burn sooner or later to get radial velocity (perpendicular to your velocity).

[Oddible]

You should burn so that when you are done with your burn your escape trajectory is parallel to the direction of Kerbin's orbit. I believe this is called the ejection angle. This way your velocity (direction part of velocity, not speed) are the same.

Actually the ejection angle is just the angle you depart from Kerbin's SoI, so if you happened to be in a polar orbit you would still have an ejection angle if you thrust prograde away from Kerbin (of course there is nothing out there but deep relentless SPACE!).

[Godot]

Here is a tool to tell you your phase angles for celestial alignment and ejection burn angles for when to start your burn. http://ksp.olex.biz/

You can also play with the parking orbits to figure out what sort of orbit is best (lowest delta V) for a refueling station before you depart the Kerbin SOI. Turns out that for closer targets a higher orbit is good, but for further targets requires LOTS of deltaV you want to be less than 100km when you start your burn for maximum oberth effect.

Great tool though...

http://ksp.olex.biz/

I use it as well ...

definitely recommended ... it makes a huge difference (1000+ dV) to wait for the calculated planetary angles (and to use the calculated escape vector out of Kerbins SoI).

Normally I need less than 200 additional dV (after leaving Kerbins SoI) to finetune everything so that the target planets SoI really is reached

[SRV Ron]

With testing, I found the best way for a Kerbal escape is to launch straight up at dawn for getting to the outer planets, launch at dusk for the inner ones. Launching at midnight or noon is less efficient because you are not adding or subtracting your escape speed to the orbital speed of Kerbal. This is assuming that you have the ideal thrust so that you are not wasting excessive fuel fighting Kerbal's gravity during the early launch phase or getting excessive drag from the atmosphere.

[luchelibre]

SRV Ron, I would certainly be interested to see your testing data to see how it matches up with a standard delta-v map.

[ScottyDoesKnow]

You should burn so that when you are done with your burn your escape trajectory is parallel to the direction of Kerbin's orbit. I believe this is called the ejection angle. This way your velocity (direction part of velocity, not speed) are the same.

This seems to be the correct way. I just did a Jool mission so I slid around my maneuver node for a bit. If you make it line up perfectly with your future sun orbit, your orbit compared to kerbin's is expanded by the same amount on each side. I assume this is where you want to be. Ended up being about 90 degrees before where I would normally do it (exactly opposite the sun).

[Oddible]

With testing, I found the best way for a Kerbal escape is to launch straight up at dawn for getting to the outer planets, launch at dusk for the inner ones. Launching at midnight or noon is less efficient because you are not adding or subtracting your escape speed to the orbital speed of Kerbal. This is assuming that you have the ideal thrust so that you are not wasting excessive fuel fighting Kerbal's gravity during the early launch phase or getting excessive drag from the atmosphere.

With physics, even Kerbal physics, launching straight up with no gravity turn is a very inefficient burn. Show your data...

[SRV Ron]

With physics, even Kerbal physics, launching straight up with no gravity turn is a very inefficient burn. Show your data...

That testing was done from another designs that were unstable to attempt to get into a 100k orbit for a later escape burn. They all left Kerban's SOI at the same time but the probes launched at dawn or dusk had added or subtracted from Kerbal's orbital speed once they reached escape from its SOI. I have a career mode design that made Kerban escape upon a straight up dawn launch that is stable enough to test doing an orbit, then escape burn. The results of the efficiency should be showing the escape speed once the probe crosses Minmus orbit after it has consumed all of its fuel. I'll test it out later this evening.

[Oddible]

You aren't doing a straight up burn like you've been saying, you're burning prograde from orbit, that is exactly what you should be doing for an efficient transfer. If your statement is that your burn to orbit is more efficient going straight up rather than doing a gravity turn, again, there is no way that is possible unless the physics in this game are horribly broken. If you're saying that it was more efficient because you had a wobbly ship that you couldn't turn, I buy it.

Edited October 24, 2013 by Oddible

[Superfluous J]

I think, Ron, you are comparing launching straight up at dawn to launching straight up at noon. I agree in that case that your method is better.

However, both methods are fuel costly compared to launching into an orbit (at any time) and then doing a Hohmann Transfer from orbit.

[Superfluous J]

While this image is awesome, it doesn't match the data from http://ksp.olex.biz. Duna looks close but Jool is way off compared to this. Any idea why this is? I'd love to be able to just eyeball my transfers in this way and not have to check every time.

[Steve-S]

One can throw a stone further with use of a sling to impart more kinetic energy than simply by throwing. That's why you gravity turn. Going straight up is throwing, gravity turning and orbiting is using Kerbin's gravity as your sling.*

*all standard disclaimers apply but I think this is a reasonably accurate simplification.

[SRV Ron]

You aren't doing a straight up burn like you've been saying, you're burning prograde from orbit, that is exactly what you should be doing for an efficient transfer. If your statement is that your burn to orbit is more efficient going straight up rather than doing a gravity turn, again, there is no way that is possible unless the physics in this game are horribly broken. If you're saying that it was more efficient because you had a wobbly ship that you couldn't turn, I buy it.

Actually, straight up to Kerbal escape. I will test later as the design I posted earlier is stable enough to do a gravity turn to orbit and the escape burn from there to prograde or retrograde to Kerbal's orbital direction around Kerbal. The other design had spin issues making it impossible to do an orbital turn.

[allmhuran]

You should burn so that when you are done with your burn your escape trajectory is parallel to the direction of Kerbin's orbit. I believe this is called the ejection angle. This way your velocity (direction part of velocity, not speed) are the same.

This is exactly right.

It's not so hard to understand if you understand the process.

Consider what you are trying to do, by, for a moment, ignoring the fact that you are in orbit around Kerbin. Instead, just think about the fact that you are in Solar orbit.

Well as we all know, in order to raise your apoapsis, you thrust prograde. If you want to go to Duna, you need to raise your solar apoapsis so that it intersects the orbit of Duna. So you need to thrust prograde relative to your *solar* orbit.

Now let's re-introduce the fact that we are orbiting around Kerbin. When you do your burn, Kerbin is going to bend your trajectory. So if you start your engines at midnight on kerbin, you will end up flying away from Kerbin heading in towards the sun a little bit. But you want to escape Kerbin heading prograde relative to your solar orbit, not radially in! So this is not ideal.

Instead, you need to start firing your engines some time before midnight Kerbin time. How long before midnight depends on how much delta-V you need to apply (in other words, where you intend to go). Why? Because the more delta-V you add, the less your escape trajectory will be bent by Kerbin's gravity, and the straighter (closer to tangential) your escape trajectory becomes. Imagine if you had infinite TWR and could instantly add 50,000 m/s of delta V. Well then your escape trajectory would be tangential to the point you did your burn, with Kerbin having almost no effect on the trajectory.

This means that to burn for closer planets (Duna), you start your burn earlier (closer to sunset Kerbin time). To reach further planets (Jool) you start your burn closer to midnight Kerbin time.

Edited October 24, 2013 by allmhuran

[SRV Ron]

OK, just ran the test. Launch weight is 44.84 tons for this rocket design with SRB, LVT-45 and LVT-909. Empty weight is 1.54 tons. For the stats, I will use probe weight at Kerban escape.

The test rocket which is easily steered into an orbital turn after booster staging neat 10k;

First test results are the straight up launch at dawn; Probe weight, 2.75 tons. That works regardless of direction launched.

This is the test results of achieving a 100k orbit. The maneuver mode has been set to about the same path as the direct launch.

And, after the escape burn, the probe weight is now 3.07 tons, a savings of 0.32 tons of fuel.

So, it appears that Kerbal Physic work where getting into orbit first allows you to take advantage of the Hohmann Transfer effect for the more efficient escape.

Edited October 24, 2013 by SRV Ron