Gravity Assists in post 1.0

[MassoudGL]

A few questions.

So I've been reading guides and watching the Manley video on gravity assists. But most of these are two years old and I'm not sure if things have changed or not. I've used gravity braking and small assists often enough that I'm comfortable with them and I understand the concept. I am currently trying to send stuff to Jool via an Eve gravity assist. I've seen that you can save somewhere around 1000k+ÃŽâ€v if done correctly and even sloppily. This doesn't make sense to me because the ÃŽâ€v to get to Eve is around 1080, while Kerbin->Jool is ~1900, according to the Olex calculator.

I actually have a near perfect transfer window lined up currently, I'll arrive at Eve within 5 days of Eve->Jool. But even still, regardless of how close I get to the planet, I can only manage to get near Duna's orbit without a retroburn.

So I'm now assuming that planning transfers doesn't really matter because you would need several encounters to get to Jool? As it stands now, I can get to Jool with 1080ÃŽâ€v ->Eve + ~1200ÃŽâ€v powered assist near Eve periapsis. More than 1900ÃŽâ€v, so why bother?

Otherwise I assume my journey will look more like Kerbin ->Eve, then back to Kerbin, then back to....Kerbin again? Gaining slightly more energy each time? Until I finally swing up to intersect Jool's orbit with small correctional burns?

[Empiro]

There haven't been any major changes in the game that would impact how gravity assists work. I haven't ever tried using Eve for a gravity assist to Jool, but you're correct that generally speaking, you'll need to spend at least as much as it takes to get to Eve in the first place.

One key thing is that to get a proper assist from Eve, you'll want your initial burn to be slightly higher delta-V. You need to have a radial component (e.g. your orbit crosses Eve's instead of just touching it). This video does a pretty good job of explaining it:

[xtoro]

You could do several Kerbin flybys to get you there. It's similar to the real life mercury missions where multiple Venus flybys were done. It takes more time, but you save dV. For these kinds of maneuvers, you spend either dV, or time. You can't get it done without using one.

[PLAD]

Using Eve to get to Jool can save you a lot of dV, but it is tricky. The simplest way, going from Kerbin to a flyby of Eve and then directly from Eve to Jool, is not practical. Eve does not have the gravity to throw you to Jool all by itself. You can flyby Eve really fast (leave Kerbin at over 2900m/s from a 75x75km orbit, way more than the ~1950m/s that would get you straight to Jool) and if you time it right you will continue on to Jool from Eve. Or as you point out, you can do a huge thrust at Eve to turn your orbit more than Eve can do by itself and add energy. Either way is more expensive than a direct Kerbin-Jool route.

I know of two cheap ways to get to Jool:

1) K-E-K-J. You flyby Eve and then flyby Kerbin before heading on to Jool. There is always a K-E-K-J path available in any given 350-(Earth)day period. The required start boost varies from about 1230m/s to about 1530m/s depending on the window. Using a flyby of Mun would save another 50-70 m/s. Read this post here for some examples.

2) K-E-K-K-J. Using two flybys of Kerbin allows Kerbin to point your solar orbit in just the right direction to get to Jool with a minimum of starting dV. I fly this path with lots of pictures and explanation of what I'm doing here. Note that on my second try (further down in that post) I got from the 75x75km orbit around Kerbin to an orbit around Jool for 1011m/s by using a Mun flyby as well. Without the Mun flyby it would be about 1070m/s. This path is not unique either, but the only other one I've found costs about 1160m/s. I don't know of a great tool for finding double flybys so these are quite hard to find.

There may be ways that involve a double flyby of Eve out there, but I haven't found one. I only know of two programs that search for flybys in KSP, my Flyby Finder and Arrowstar's TOT.

Neither one searches for double flybys.

And regarding your other question, the Kerbol planets have had the same orbits and masses since at least version 0.22 so anything that worked back then would work now. The big change is that time used to be kept in 'Earth style' with 24-hour days and 365-day years, nowadays Kerbin time is used with 6-hour days and 426-day years, be careful to figure out which time style is being used when someone describes a departure or flyby date. I always used to use Earth time, but recently I redid the K-E-K-K-J mission using Kerbol time here.

Also, it is very, very hard to fly multi-flyby missions without a node editor, if you're not using one and you want to try these paths then you should probably try one. I use Mechjeb.

[MassoudGL]

Thanks for all the advice. My game time is pretty far, around 8 years in. I managed last night to get a K-E-K-K-J path but it took quite a while and wasn't cheaper than 1200m/s. But practice and hours of experimentation make perfect? I also use Mechjeb, I noticed that changes of 0.1m/s make a huge difference haha.

[xtoro]

I might give that a shot next time I send a ship out there just for the heck of it. So far I've just been sending them directly to either Laythe to lightly aerobrake or Tylo for a gravity swing.

[PLAD]

Thanks for all the advice. My game time is pretty far, around 8 years in. I managed last night to get a K-E-K-K-J path but it took quite a while and wasn't cheaper than 1200m/s. But practice and hours of experimentation make perfect? I also use Mechjeb, I noticed that changes of 0.1m/s make a huge difference haha.

Executing KEKKJ for something around 1200m/s is really good, especially if it's a window you found yourself. Randomly selected windows seem to cost in the 1100s, add 50m/s or so minimum for course corrections and you are around 1200. My 1011m/s was a result of finding one particular window that is exceptionally good, and practicing it many times with variations to figure out the tricks. That second pass of Kerbin and the deep-space correction around Duna's orbit are particularly tricky, the first path I find often needs 100m/s or more of adjustments, but extensive tweaking (30-60 real-world minutes!) ultimately finds one around 5-10m/s if I timed the earlier flybys right.

In any case it sounds like you are doing it right. I'd enjoy seeing any good windows you find if you ever choose to show them!

-PLAD