Rendezvous With an Elliptical Orbit

[Endeavour]

Does anyone have any tips for how to do this?

[Cykyrios]

The best I can think of right now (I haven\'t done any research on the subject yet) would be to get to the Pe/Ap of your target, and adjust the opposite point so that a full period will get you close enough to meet your target.

I\'m going to try that right now and I\'ll tell you how it turns out.

EDIT: So, I tried, and I can tell you that at least, it works.

I put a ship in a 150x600km orbit, and launched a second one to match the orbit. I wrote down the MET when my target passed at my Pe as well as when I passed it.

I was about 19min and 48s late, so I calculated the time difference between 2 orbits, one of which had a different Ap. I found out that a 85km Ap (which becomes the Pe then) will get me a 1 second error, if I were to actually perform a perfect burn.

My rocket ran out of fuel while I was 2.1 km away from the target, burning to match speeds.

TL;DR: Rendezvous is possible with elliptic orbits, but is more fuel consuming and difficult.

[Endeavour]

I never really doubted that someone on these forums could do it. There are some really skilled people here. I was mostly wondering if someone had developed a technique for getting it easily, the way they have for circular orbits. Until then i guess I\'ll have to keep trying the hard way.

[Kosmo-not]

I do it based on gut feeling. I don\'t have a spreadsheet for something that complicated yet.

[Cykyrios]

While I don\'t use a table for this, my lil\' spreadsheet allows me to make rough (as always) calculations, but then again, once I have matched the orbit and written the METs, I have to wait for Pe or Ap.

I can\'t imagine doing it on gut feeling as I can sometimes do with circular orbits, though, because ellipses are a deceptive bunch (due to Kepler\'s 2nd rule).

I think I can make a video if needed.

[Kosmo-not]

The gut feeling approach isn\'t all that bad. It\'s just about estimating (based on velocity and distance) if both craft will get to the orbital intersection at the same time. You just adjust your trajectory to your best estimate as you go along.

[Anachronda]

While I don\'t use a table for this, my lil\' spreadsheet allows me to make rough (as always) calculations, but then again, once I have matched the orbit and written the METs, I have to wait for Pe or Ap.

Well, since you folks already know I ain\'t the sharpest bulb in the drawer, I guess I don\'t have to worry about opening my mouth and being thought a fool.

Bear in mind that I\'ve not yet managed to accomplish a rendezvous (can\'t quite close those last couple of kilometers), so take it for what it\'s worth. I also know little about orbital mechanics.

Seems to me that if you thrust tangent to the ground, you\'re raising the point on the opposite side of your orbit.

If you line up your orbit such that one point is on the orbit of your target, you\'ll see a big hole on the other side that shows the difference in eccentricity. Thrusting at the aligned point will close the hole.

It\'d be a big help if I could put a marker on my orbit in the map display to remind me of where I want to accomplish a burn...

[Excalibur]

Bear in mind that I\'ve not yet managed to accomplish a rendezvous (can\'t quite close those last couple of kilometers), so take it for what it\'s worth. I also know little about orbital mechanics.

For your first rendezvous I\'d suggest launching your target into a circular orbit of 350km with as close to zero inclination as possible (reason for the relatively high altitude is so you can use a higher time compression factor). Make sure your ships are easy to control in orbit with lots of RCS fuel. You\'ll need it.

Obviously you\'ll now need to launch your chase vessel into as close an orbit as your target. Once this is done do as follows:

• 
  [li]Go into Orbital Map view. Focus the camera on Kerbin[/li]
  [li]Zoom the camera in as much as possible so that you can still see the relevant orbital tracks just \'in front\' of the camera[/li]
  [li]Now find one of the points where the two orbital tracks cross over each other (there will be two points 180 degrees apart)[/li]
  [li]Orientate your ship so that it\'s locked Northwards (best done with the \'+ NML\' button on MechJeb)[/li]
  [li]When your ship reaches one of the \'crossover points\' along its orbit make a small forwards RCS burn with fine controls on (\'Ctrl\' key then \'h\' key)[/li]
  

If you do the above while in the Orbital Map view you should notice that the orbital tracks begin to converge at the two points at ninety degrees of your orbital position. Simply keep firing forward with the RCS and you should be able to match up your orbital inclination with that of the target almost exactly. If you find burning Northward increases your separation, just burn your RCS in the opposite direction with the \'n\' key to give you a Southward burn.

Once you\'ve got the inclinations synced nicely match up either Ap or Pe with your target. Leave the other at least 10km higher or lower. This will enable you to catch up the other ship (or the other ship to catch up to you). When you both eventually meet at your Ap/Pe then all you have to do is circularise and you\'ve got a nice easy rendezvous - no Rincomp required.

See this thread for some extremely close intercepts I\'ve achieved with this method: http://kerbalspaceprogram.com/forum/index.php?topic=10570.0

P.S. I plan on doing a more in-depth guide on this subject in the near future...

EDIT: Not tried it but I think the above method should also work for rendezvousing with other ships that themselves have an eccentric orbit. Probably a good deal more difficult I should imagine though!

[Anachronda]

EDIT: Not tried it but I think the above method should also work for rendezvousing with other ships that themselves have an eccentric orbit. Probably a good deal more difficult I should imagine though!

I suspect (but am not math-worthy enough to prove) that what happens is that you\'re pivoting the orbit around a line between where you are and the focus of the orbit (yeah, yeah, an ellipse has two foci; what I mean here is the focus that holds the body around which you\'re orbiting. Don\'t know the term for it. Probably center of mass or somesuch).

Once you\'ve got the inclinations synced nicely match up either Ap or Pe with your target.

Which is a nice sentence that hides a lot of trouble, since you can\'t see the Ap or Pe of the target orbit.

Furthermore, on an eccentric orbit the Pes can be quite close while the Aps remain far apart. The Pe is on the short end of a lever arm with the fulcrum at the center of mass, so a small change in Pe can make a *huge* change in Ap.

In my experience, the biggest problem is running out of patience, especially when the target gets close enough that you can see it. I got within 90m of my long-suffering orbiting Kerbonauts last night, but then tried to close the remainder of the gap by sight with the RCS. Didn\'t go well.

[Cykyrios]

If you\'re within 90m and your speed isn\'t too different from your target\'s, then RCS is more than enough. You just need to close the distance at, say, 5 m/s, and you\'ll be done in less than a minute.

When you want to match the orbits, you don\'t need the exact position of the Pe, you can eyeball its location by zooming in the orbital map view, since your orbit intersects your target\'s orbit. Sure, your Ap will be off as well, but this isn\'t too important as you\'ll probably want to rendezvous at the Periapsis, and you can get within 5km without trouble if you know the formula of the period of an orbit.

I personally did it in quite a hurry, but still managed to end up 2 km away from the target in a grand total of 3 or 4 periods.

I\'ll probably make a vid by Wednesday to show how I do it, might be useful for everyone, even though I doubt my method is the best there is.

[Anachronda]

If you\'re within 90m and your speed isn\'t too different from your target\'s, then RCS is more than enough.

Oh, sure. RCS is plenty. It\'s my *coordination* that is a problem.

[Excalibur]

Which is a nice sentence that hides a lot of trouble, since you can\'t see the Ap or Pe of the target orbit.

I should probably make myself a bit more clear on that point. I made an assumption that your target would have as perfect a circular orbit as possible (it\'s not difficult with fine RCS control). Then basically you make your chase ship\'s Ap or Pe coincide exactly with your target ship\'s altitude. Then within a few periods (depending on how elliptical the chase ship\'s orbit is) both chase and target ship should meet at the same point and you can circularise. I\'ve never had any trouble seeing Ap/Pe for both ships in these circumstances.

You mentioned in an earlier post you\'d not yet accomplished a rendezvous - did you mean target in elliptical orbit or circular? I assumed the latter so directed my advice that way. Apologies if I got the wrong end of the stick (and a little off topic!). =P

My edit referred to the fact that you should be able to invert this for chasing a target in an elliptical orbit. Instead of your chase ship using an elliptical \'chase\' orbit, use a circular one with an altitude the same as the target\'s Pe. Then once both ships inevitably meet at the target\'s Pe instead of circularising you burn until your orbit\'s eccentricity matches the target ship\'s orbital eccentricity.

The increased difficulty of rendezvous with a target in an elliptical orbit would probably be using the Orbital Map view to synchronise inclinations, especially if the target\'s Ap is miiiillles away. Which I\'ve realised is what you\'re getting at now.

[Anachronda]

I should probably make myself a bit more clear on that point. I made an assumption that your target would have as perfect a circular orbit as possible ... I\'ve never had any trouble seeing Ap/Pe for both ships in these circumstances.

Well, sure, for a circular orbit there\'s not much difference in Ap and Pe to begin with. Your Ap and Pe are marked on the map, but those of your target aren\'t. Doesn\'t matter much if you\'re both in circular orbits, since there\'s not much difference to begin with.

You mentioned in an earlier post you\'d not yet accomplished a rendezvous - did you mean target in elliptical orbit or circular?

I\'m not the original poster of the thread. I poked my head in here because the target I use is in a mildly elliptical orbit; it\'s something I put up there when I was still arguing with getting to circular orbits. I also have a much more elliptical target out there that I\'ll work on once I get it down with this mildly elliptical one.

It is the case that I have not yet succeeded. I can easily get within 20Km of the target. With more time and patience, I managed to get within 90m last night. At which point I got all enthused and tried to tweak the last little bit by sight.

As I said, my primary barrier is patience. I have no doubt that with sufficient patience I can succeed.

The increased difficulty of rendezvous with a target in an elliptical orbit would probably be using the Orbital Map view to synchronise inclinations, especially if the target\'s Ap is miiiillles away. Which I\'ve realised is what you\'re getting at now.

Aligning inclinations is not really a problem, because there\'s a clear point where the orbits cross. With sufficient patience, you can approach that point then do the burn as you describe.

Aligning the major axes is a bit trickier. As I mentioned, the Ap and Pe of your orbit are marked on the map view, but those of your target are not. When you target Kerbin and zoom out enough to see the whole orbit, you can\'t see enough detail to see small misalignments. When you target your ship and zoom in to see the difference between close orbits, you can\'t see enough of the orbit to find the difference. And small differences in Pe position translate to large differences in Ap position.

It\'d be nice if you could scroll around the orbit while you were zoomed in really close to the ship.

A rendezvous with debris in a highly elliptical orbit will be even trickier, as the orbits of debris aren\'t even displayed unless you mouse over the debris. It\'d be nice if there were some way to select a target and have its orbit displayed with Ap and Pe markers as long as it\'s selected.

[Excalibur]

Posting after bed-time isn\'t a good idea for me, I seem to get a bit muddled. ;D

You\'re right about the main challenge of elliptical rendezvous being lining up the major axes. I think that\'s what I was trying to get at.

I must admit I haven\'t put anywhere near as much effort into rendezvous as the other aspects of KSP, I guess I don\'t see much point until docking is released (but lets not get into that discussion).

Just had a thought.

What if you placed your chase ship into a polar orbit such that its Pe matched the target\'s Pe altitude figure. Perform a burn whilst over one of the poles that is normal to your orbit, if in the right direction that should move your Pe\'s together. Burn at Pe until your Ap matches the target\'s. Then next time you\'re back at Pe perform another normal burn to bring inclinations together.

Won\'t be easy and you\'ll need MechJeb for the precise burns. I\'ll give it a try.

[Kosmo-not]

Just had a thought.

What if you placed your chase ship into a polar orbit such that its Pe matched the target\'s Pe altitude figure. Perform a burn whilst over one of the poles that is normal to your orbit, if in the right direction that should move your Pe\'s together. Burn at Pe until your Ap matches the target\'s. Then next time you\'re back at Pe perform another normal burn to bring inclinations together.

Won\'t be easy and you\'ll need MechJeb for the precise burns. I\'ll give it a try.

That sounds like it would require a ton of delta-V.

[Excalibur]

Oh it would. Just an exercise in manual rendezvous. I\'m sure there\'s a much better way to do it!

[Cykyrios]

If I understand that correctly, what you want to do is line up the Pe of the two orbits, right?

In that case, I suggest you 'simply' get into a circular orbit at your target\'s Pe altitude, then eyeball the best moment to burn to raise the Ap.

As I said in a previous post, you don\'t need to line them up exactly, as you will want to rendezvous at Pe, which gives you a bigger room for error.

Going all the way to a polar orbit seems a bit like overdoing it to me =P

[Excalibur]

It\'s the most precise way I can think of doing it manually. Never claimed it wasn\'t long-winded or inefficient!