Reaching the Planets

[Chads]

Hi there,

Round two of my questions regarding reaching the other planets.  I pulled down Astrogator, which shows me the optimal burn times, and necessary Delta-V based on those.  (If you don't know about Astrogator, it'll create your nodes, too.  Not bad!)

But, I'd like to know how to "eye-ball" these, as well, so that I'm not always dependent on mods, etc. for everything.

I'm assuming that some common sense comes into play, and some trial and error.  I've started the latter (I'm not sure how much of the former I have...).  I was able to successfully get to Eve, but burned up in its atmosphere when I misread the nav charts for how high it's atmosphere is.  Doh!  

The process I used was :

1.  execute a burn on the dark side of Kerbin, to get a Kerbin Escape moving toward the center of the solar system.  
2. Once out of Kerbin's SoI, I started to play with a node to get an intercept with Eve.  
3. Once I got a reasonable one (500,000KM--I thought it was reasonable), I executed the maneuver. 
4. About 1/3 of the way to the intercept, I played with another maneuver node, and got an Encounter.
5. Performed that maneuver, and smiled until I started hearing explosions....

The ship I built to get there started with 10k Delta-V (non-atmosphere value, and when my ship hit the atmosphere, I still had close to 3000 d-V left.  So, I see that it was overkill, but in the future, that may give me some idea of what it'll take to get back.

So, the question is, barring the use of Mods to do the calculations/flying for me, is the process I used roughly correct for inter-planetary exploration?  (Minus the explosions part...)

 

Thanks!

[Snark]

Well, as they say, any navigation that gets you to your destination is a good navigation.

However, it must be said that the way you describe is very dV inefficient.  The optimal way is to do one single big burn in LKO that aims you right at your destination, perhaps with a mid-course burn to match planes.  (As long as the source and destination planets are pretty close to coplanar. If you're going somewhere with a radically different inclination, like Moho, there are some other techniques that sacrifice travel time to save some dV.)

Here's a recent post I made explaining the process to a person who was new to maneuver nodes. The post describes going to Duna, but it's essentially the same technique you'd use to go to Eve.

Hope it might be useful.

[Chads]

Quite Useful indeed!  Many thanks!

Totally agree that it's dV inefficient.  I was about 4 days ahead of the ideal transfer when I did this.  More than anything, I wanted to see if I could even get intercepts (since I hadn't tried before).  Looking forward to visiting more places!

Duna was my first choice, but it is now about 60 degrees behind me in orbit.  It'll be a while before an efficient transfer to Duna comes back around.  Like, 2 years.

 

[Weywot8]

Just something to note in relation to your original method - the main inefficiency comes from splitting the steps of number (1) and (2).  Getting to Eve within the transfer window period shouldn't take >2000dV and can be done for ~1100dV at the best of times (not all transfer windows are equal) by planning and executing the entire burn from parking orbit. 

Transfer windows are actually very generous - for ~10% extra dV, you can still get there +/- up to 2 kerbin weeks deviation from optimal departure date. 

 of course, knowing when the transfer windows are without doing the calcs yourself means you are still using mods in some form or another.

Edited July 19, 2017 by Weywot8

[Kurld]

8 hours ago, Snark said:

Here's a recent post I made explaining the process to a person who was new to maneuver nodes.

Hope it might be useful.

Dang, even I learned something. I had no idea you could use the mouse wheel on the navigation widget to do fine adjustments! Six years playing this game!

[Snark]

8 hours ago, Chads said:

Totally agree that it's dV inefficient.  I was about 4 days ahead of the ideal transfer when I did this.

Actually, missing the ideal launch window by just a few days doesn't make all that much difference to the dV.  As long as you're pretty close, it's fine.

What does lose you dV-- I mean, a lot-- is if you do a big burn when you're in interplanetary space.  That's the kicker.  As much as possible, you want to do all of your "big burning" in low Kerbin orbit.  Oberth effect is the reason.  It's a whole lot better to do a great big burn in LKO than it is to do only a medium-big burn in LKO and then another big burn later in interplanetary space.

[Zhetaan]

6 hours ago, Weywot8 said:

 of course, knowing when the transfer windows are without doing the calcs yourself means you are still using mods in some form or another.

Not necessarily:  it's fairly easy to set up a node to go efficiently to any planet's orbit without using mods.  For example, let's go to Duna.

1. Start with a spacecraft that's in a low Kerbin parking orbit.  It doesn't have to be the interplanetary vessel, though of course that would help.
2. Create a node there and stretch the prograde modifier to get a transfer orbit that reaches Duna's orbit.  It's helpful to know whether you need to start your burn on the daylight side, but if you set up the node and start pulling on prograde, you'll figure that out on your own very quickly.
3. Tweak the transfer orbit by sliding the node along your low Kerbin parking orbit and reducing prograde to keep the apoapsis at Duna's altitude.  This can be done in two stages:  slide the node and observe whether the apoapsis rises or falls.  If it falls, go the other way.  If it rises, go until it reaches a maximum and then reduce your prograde to drop the apoapsis back to Duna's orbit.
4. This will get you close, but changing your prograde burn will require you to slide the node a bit more.  Keep doing these two steps to refine your burn delta-V until you cannot reduce it further without dropping the apoapsis below Duna's orbit.  Now you have your transfer time and delta-V.
5. Then it's just a matter of clicking the +orbit button on that node so the arrival time at Duna's orbit occurs when Duna is at that point on the orbit, and you've found your window.

I'll grant that it is tedious, I'll grant that it solves for both phase angle and ejection angle without telling you the values for either, and I'll even grant that it doesn't do anything about inclination corrections, but it does find the window--and since the solar system always starts in the same configuration on Day Zero, you only need to find the transfer window once.

The trick to it that makes it difficult for people to do it is that they usually don't separate the idea of reaching Duna's orbit efficiently from the idea of reaching the planet.  They try to encounter the planet first and then play with the node to make it more efficient, or else launch first and tweak later, but that's not really how it works.  The important thing to remember about interplanetary encounters is that transferring to the orbital altitude you want is all about the initial burn; timing doesn't matter because if you do it correctly, then you will reach the correct orbital altitude whether or not you intercept anything there.  Intercepting the destination after that point--once you have a burn that puts you on the correct orbit--is then a matter of timing alone because you already have the burn.

To illustrate with an example that is in Kerbin's sphere of influence (and thus easier to see without a lot of zooming), I don't know how any of you set up encounters with the Mun, but the way I do it is to set up a node to go to the Mun's altitude first, and then slide the node along my parking orbit until I get an encounter.  I may make minor tweaks after that but for the most part, initial burn setup is done before I have the encounter.  The equivalent to the less-efficient way most first-timers try for interplanetary is to start with a node that gets the encounter and then start tweaking that encounter (by both changing the amount of prograde and sliding the node) until the required delta-V stops decreasing.  That can work sometimes but it involves solving for two variables at once (burn start time and total burn duration) and it gets confusing--especially if you hit the right value for one variable but not the other, and you don't know which one is correct.  The way the original poster tried for Eve is equivalent to burning for a transfer that takes you up towards the Mun's orbit (possibly one that gets close to an intercept, though that was not made explicit), waiting for an hour, and then trying to get an encounter.  I'm not saying that can't work but it's horrifyingly inefficient.

[bewing]

1 hour ago, Kurld said:

Dang, even I learned something. I had no idea you could use the mouse wheel on the navigation widget to do fine adjustments! Six years playing this game!

It only works sometimes. I haven't bothered to figure out the pattern yet.

 

[Weywot8]

1 hour ago, Kurld said:

Dang, even I learned something. I had no idea you could use the mouse wheel on the navigation widget to do fine adjustments! Six years playing this game!

7 minutes ago, bewing said:

It only works sometimes. I haven't bothered to figure out the pattern yet.

It depends on the speed and direction of the scrolling. A smooth scrolling mouse can be very helpful for this. It's also rather picky, mouse cursor has to be pretty much center of the direction marker being adjusted.

 

[Cpt Kerbalkrunch]

11 hours ago, Chads said:

I was able to successfully get to Eve, but burned up in its atmosphere when I misread the nav charts for how high it's atmosphere is.  Doh!

So you had a couple of fights and decided to take on the champ? No surprise you got knocked out, bud. 

[HebaruSan]

2 hours ago, Zhetaan said:

1. Then it's just a matter of clicking the +orbit button on that node so the arrival time at Duna's orbit occurs when Duna is at that point on the orbit, and you've found your window.

Note that further movement of the node may be needed if you click +orbit more than a few times, because Kerbin keeps moving around the Sun during those orbits, which changes the ejection angle of the node. As an extreme example, if you wait half a year, the "same" node will send you towards Eve instead of Duna.

[Zhetaan]

@HebaruSan:

True, but if that happens, the solution is to back up a step and refine it again.  The point is to break up the transfer into manageable pieces; however, just as everything else in KSP is an approximation, so too is breaking up the transfer planning.  There is no single action you can do that affects only one variable, so the only way to break the problem into manageable pieces with the tools at hand in stock is to work out a way to limit the effects of those changes on all but one of the variables and then make corrections to the approximation.  It's Newton's method, in space.

[Chads]

OK, need something clarified.  I'm reading through Snark's full post, and he mentions starting on the night side of Kerbin to go to Duna.

I thought it was the other way around--start on dark side to go in, start on light side to go out.

I'm assuming I have it backwards?  (I thought I picked that up from Scott Manley videos....)

1 minute ago, Chads said:

OK, need something clarified.  I'm reading through Snark's full post, and he mentions starting on the night side of Kerbin to go to Duna.

I thought it was the other way around--start on dark side to go in, start on light side to go out.

I'm assuming I have it backwards?  (I thought I picked that up from Scott Manley videos....)

 

My bad.  Just looked.  I messed up with Mr. Manley said.  

Might explain why my dV requirement was so high getting to Eve....

 

[GarrisonChisholm]

On 7/19/2017 at 8:41 AM, Kurld said:

Dang, even I learned something. I had no idea you could use the mouse wheel on the navigation widget to do fine adjustments! Six years playing this game!

Wha wha what???......  O.O

 

(...that's a +1 )