When should I launch to my station?

[photogineer]

So, I've been building myself a space station in LKO the last few weeks, and it's been going well, in a nice 100km circular equatorial orbit. After getting the habitation pod and ship parking module docked, I decided I wanted a challenge, and that I wanted to boost the orbit in to a higher, inclined orbit. You'd think that little voice of self preservation in my head might have chimed in at this point, but apparently Jeb had fed it some snacks, and it was quiet. So I sent up a new fuel module consisting of a Jumbo tank and some RCS tanks and thrusters, and docked it to the station. After that I used the station's propulsion module to slowly boost the orbit and inclination over several burns to ~27 degrees inclined, 350km circular-ish. I intend to trim that a bit more, still have plenty of propellant left in the station, but for now it's good.

So, here is the question, I hadn't given much thought to this until now (of course), when I'm ready to actually crew the nearly completed station, I have no idea how to launch in to the same orbital plane as the station. When it was equatorial it was easy, but now I am at a loss. I understand it is much preferred I launch straight to the correct inclination as changing on orbit is dV-intensive, and I can understand why. I even (think I) understand the concept of a launch azimuth, but it seems that I need to wait until one of the nodes of the station's orbit is over KSC when I launch, is that correct?

Assuming that is correct, is there any way to predict when that will happen? Anything better than just timewarping until it happens? I am trying to do stuff by the book-ish, and calculate launch windows and such for these sort of things.

So, in summary, questions are:

1. If I want to reach an inclined orbit, it is always better to launch straight in to it at the right time then try to match it from orbit?

2. If I wanted to reach a 27 degree orbit, my launch azimuth should be 63 degrees instead of the 'typical' 90 degrees?

3. I also need to launch when the ascending node of the station is over KSC to end up on a coplanar orbit with the station?

3a. Is this a way to calculate that properly?

3b. If I launched when the descending node was overhead, would I be able to just launch with an azimuth of 117 degrees?

Thanks,

Matt

Edited December 9, 2013 by photogineer
Removed the space station screenshot because the link keeps breaking

[Tank Buddy]

If you wanted to have the same inclination by launch, you would have to launch at the ascending or descending node. Your answers for 3 are correct. Launch in the correct heading for either nodes. Technically it should take less D-V without an in orbit plane change.

I don't believe #2 is correct. You would either have to launch at the ascending or descending node or do and inclination change when rendezvousing.

[Dogface]

If I may ask, was there a reason for the inclined orbit? I have put a number of satellites into inclined orbits, to simulate weather or spy satellites. The inclined orbit provides larger ground coverage.

Why does one have to launch on the node?

[razark]

You don't need to worry about nodes. You just need to wait until KSC is directly under the orbital path of your station. Once you have KSC under the orbital path, launch on a heading that matches your station. This would either be 63 or 117 degrees, depending on the direction the station is traveling. (Although, when launching, you might want to keep in mind that you already have a sizable amount of speed towards 90 degrees. It may help if you aim somewhat north of 63 or south of 117.

[photogineer]

If I may ask, was there a reason for the inclined orbit? I have put a number of satellites into inclined orbits, to simulate weather or spy satellites. The inclined orbit provides larger ground coverage.

Why does one have to launch on the node?

The reason is the same reason I built a space station in sandbox mode, it seemed like fun! Equatorial orbit was getting crowded too, so I figured it was time to learn how to do things in other orbits.

You don't need to worry about nodes. You just need to wait until KSC is directly under the orbital path of your station. Once you have KSC under the orbital path, launch on a heading that matches your station. This would either be 63 or 117 degrees, depending on the direction the station is traveling. (Although, when launching, you might want to keep in mind that you already have a sizable amount of speed towards 90 degrees. It may help if you aim somewhat north of 63 or south of 117.

Correct me if I'm wrong here, but wouldn't the orbital path passing over KSC, which is on the equator, be the same as when the node is overhead? I do see your point though, I don't need to wait for the node, it just happens that the node will be overhead when the orbit is.

Does anyone know how to calculate when the orbital path and/or node will pass over KSC? Timewarping while eyeballing the map view is kind of a pain and I'd like to work on other stuff between launch windows, without having to worry about missing them.

Matt

[diomedea]

You find that the orbital path of your station will cross the KSC latitude (almost 0°) twice per orbit (ascending and descending node). Size of the orbit determines the period, and you compare the period with the rotation time of kerbin (6 hrs). The time difference is easily translated in longitude difference (with 6 hrs rotation time, 1° longitude = 1 min). That means you will find the crossings (nodes) coming at different longitudes. You have to wait for a launch window, when (generally every number N of orbits) either of the two nodes longitude will return to match with KSC longitude.

For a perfect orbital insertion, the longitude of the node should be a bit east of KSC: while your rocket rises to the orbital altitude, it still is initially tied to Kerbin's rotation and therefore it will move a bit to the east in relation to the orbital path (perfect calculation would require timing your ascent and taking in account altitude and Kerbin's radius).

Launch at 63° azimuth on the ascending node, 117° azimuth on the descending node. However, as your rocket moves in latitude (northern than KSC in the first case, southern in the second) it will also continue to move to the east with the same initial speed of KSC, and that speed is greater than the speed of the ground at higher latitudes. This means you see the rocket drifting east (due to Coriolis effect) in relation to Kerbin's frame of reference, but that does not effect the orbital frame of reference. Heading has to be adjusted as it changes along the orbital path (as a sine wave, from 63° at 0° latitude, to 90° at 27° latitude north, then 117° at 0° latitude descending, then 90° at 27° south).

[razark]

I'm not sure, but does the node actually exist if you're sitting on the pad waiting to launch? If the node does exist, then it would be the point at which you would launch. You could use the Kerbal Alarm Clock mod, which would allow you to get the time until node.

Otherwise, you'd need to know the angle between launch site and node, and then figure how long it will take the planet to turn that far. Short of using a protractor on the screen, I don't know of an easy way to get that angle if you're looking for a stock way.

[diomedea]

@ razark: two different planes always have a line in common, and the nodes are the opposite directions along that common line. In orbital mechanics, the two planes are the ones containing orbits or a reference frame. In this case, one plane is tied to kerbin's equator (equatorial plane) the other the orbit of the station. Thence, yes, we have ascending and descending nodes; they have nothing to do with the rocket, the launchpad or else.

Yes, the nodes (either of them) are the points to match for a launch. I did not even consider previously that we don't need to wait for the station to actually occupy one of the nodes or be in any specific position along its orbit. Though, most efficient rendez-vous launches are conducted when the "station" already in orbit is just on the node, if the rocket is powerful enough to get to orbital altitude in a fraction of one orbit. If not, best insertion would also consider how fast the rocket will gain on the station while ascending (so, the station has to have crossed the node already by the same time that will be gained by going on a lower orbit).

I definitely use Kerbal Alarm Clock (KAC) for most time-related functions in KSP, and definitely to wait for a node passing. Even if I could calculate the time myself. As I wrote before, time calculations on Kerbin are really simple due to the rotation period of 6 hrs. It takes as many minutes to rotate as the number of degrees of the rotation (longitude difference). Given that an orbit in space is fixed (does not rotate with the body) and therefore the direction of the nodes is fixed, I only have to consider Kerbin's rotation to calculate where a node will be found with any orbit.

[razark]

Thence, yes, we have ascending and descending nodes; they have nothing to do with the rocket, the launchpad or else.

I understand the nodes would exist in reality. I was questioning whether they are considered to exist in the game, so that you could time launches off of them. I believe I may have worded it badly.

Based on what I'm looking at on the map screen now that I'm actually in the game, the nodes do not appear. Either they do no appear because they are underground, or they are not drawn because the game knows your are on the ground. I'm also not seeing the dashed lines that accompany them, so I'm guessing they simply aren't calculated/displayed for a landed ship.

[diomedea]

@ razark: what you wrote may be revealing where the problem is. It is with KSP itself. It is a game and shows only what is relevant to the game. But to consider how planets and vessels move in space, I use astronomical concepts. KSP has them embedded, but does not show what is not relevant.

When you have a ship on the launchpad (or landed), it actually moves with the ground of the planet. That point on the ground actually revolves due to the rotation of the planet, so it describes a circle at its latitude. That circle is parallel to the equatorial plane; in case of a point on the equator, it is on the equatorial plane, so I can say that the landed ship has an orbit on the equatorial plane itself. However, KSP does not show any orbit for a landed ship, so you can't see the nodes I am talking about (those would appear if you could ask KSP for a plane change for that ship, but KSP doesn't let that for landed ships). KSP is no astronomical software per se, so it does not provide references to let show purely astronomical concepts. For me it is easy to visualize the equatorial plane by seeing the equator, and can figure where it intersects any other orbit.

I studied astronomy on books, and had to learn to calculate the relationships of angles, time, distances. I can determine my exact position on Earth using only the stars. To teach the whole of what I am writing about, it would take a course. However, there are some good computer programs that may help visualize something, in case you are interested in the subject. My preferred one is Celestia http://216.231.48.101/celestia/index.html.

If you still have doubts, or would like to continue on this subject, I am happy to try and explain what I know.

[photogineer]

Well, thanks everyone, I just managed to put the first long term crew on to my KSS. I was able to predict about when the orbit would be aligned with the launch site based on the 1 minute per degree figure and some Kerbal Engineer readouts on the station while in orbit. I then timewarped ahead to that time, and launched a crew rocket up, in to a nice 350km x 300km phasing orbit. Took a little playing with the engine but I successfully rendezvoused and transferred my crew.

Next step, precision landing!

Thanks all,

Matt

[Dogface]

Ascending node = The point at which the orbit crosses the reference plane moving north. Here, "north" means the direction of the orbit normal of the reference plane.

Descending node = The point at which the orbit crosses the reference plane moving south.

http://wiki.kerbalspaceprogram.com/wiki/Terminology

[diomedea]

http://wiki.kerbalspaceprogram.com/wiki/Terminology

The KSP Wiki is right, though still not very clear.

For more: http://en.wikipedia.org/wiki/Orbital_node

For hard-core academic correctness: http://astrowww.phys.uvic.ca/~tatum/celmechs.html, e.g. @ Chapter 10, Page 2: "The two points where the plane (of the ecliptic) and the plane of the planet’s orbit intersect are called the nodes, and the point marked ☊ is the ascending node."