Changing plane while periapsis kicking

[Aegolius13]

That's probably not a very helpful title, so let me explain the issue.  I was doing an ejection burn to Moho, using the launch-at-node method.  The ejection burn was about 2000m/s prograde, 1200m/s normal.  I planned to do about 40% on one pass, which would take me out to about the orbit of the Mun, and 60% on the next pass. Naturally, I did both the prograde and normal components at the same time to get that "Pythagorean" savings.  

Well, after the first pass, much to my chagrin, I realized I had done the WRONG normal direction, so my orbit was now inclined further in the wrong direction, and would seemingly require a huge burn to fix.  I had just quicksaved, so I despaired having to load a much earlier save.  But then I thought I might as well check how much the plane change would cost at the Mun-ish apoaspis in between my periapsis kicks.  And it was VERY cheap, as such burns tend to be very far out.  Like 80-something m/s to flip my inclination in the opposite direction -- i.e., to get twice as much plane change as I did in my first periapsis kick.  

It looked like the apoapsis burn was so cheap, it came out of ahead of doing the plane change burn at my Kerbin periapsis, even after factoring in the Pythagorean discount on the latter.  So this got me thinking: should I ONLY do the plane change this way?  That is, do a pure prograde burn out to the Mun-level or so (any further and I really don't like the delay before your second pass), then do ALL of the necessary plane changing, then do a final all-prograde burn to finish your ejection.  It seems like, in addition to being cheaper, this shortens the further splits up your primary burns, which is good if you have a low TWR and/or are using ions. 

Does anyone do it this way?  Am I missing something that makes it less effective?  

Two other side notes: First, you could probably save further by launching into an inclined orbit.  But for some reason, I never bother to do this, due to the hassle of figuring out when you need to launch and how much to tilt your orbit.  

Second, when I was doing the maneuver at apoapsis, the normal burn pushed out my periapsis a but further from Kerbin.  It appeared to be worth it to include a bit of retrograde in the plane change burn, to get that peri back to the edge of Kerbin's atmosphere and maximize Oberth for the final ejection burn.  

Edited January 1, 2017 by Aegolius13

[Plusck]

You're probably right about that.

Still, is it worth the hassle in most cases? Moho is a bit of an exception due to the 25° angle or whatever you need at Kerbin to match Moho's inclination on SOI exit.

And what were the "real" savings? 2000 prograde + 1200 normal gives a total burn dv of 2350 or so. If you were already in the right plane, what would it be? 2100? 2200?

For the second point - about bringing Pe down again - I'm not sure that that is a good idea. If you're going a burn in two passes, it means your TWR is pretty low. On that second pass you'll be actually at Pe for a very short time, and lowering Pe is only going to shorten that time. If you need to burn for 10 minutes, then a lot of that 10 minutes is going to be significantly off prograde, and you don't have the option of going round again once your Ap is in interplanetary space.
 

 

 

[bewing]

Oberth does not help plane changes at all. For creating a plane change, your velocity in the "y" direction is some ratio of your velocity in the "x" direction. So if you are going fast, then you need a lot of normal velocity. So it's always better to wait until the part of your orbit where you are going much slower to do plane change burns.

[WanderingKid]

In general, yes. If you're going to burn normal/anti-normal, you want to be doing it at the slowest part of your journey, which means getting your ascending/descending node onto your Apoapsis and doing your burn there.  However, under most circumstances, if your an/dn is NOT on the apoapsis, you're going to have to burn off point.

On a side note, a trick you can use to get your inclination straight(er) on liftoff is to watch the Prograde route of Kerbin and liftoff a little before you get to it with the inclination you're going to need for your very near future burn.  Obviously this doesn't work if you're launching a month or two ahead of schedule, but if you're lifting off and leaving in a day or two once you ferry up fuel and cargo, this can save you a massive amount of d/v by already being ready to plane transfer.

[SpeedDaemon]

The other thing about Moho is that it moves so fast in its orbit that splitting a burn can put you far enough off of your target transfer window that you spend more dV correcting later.

Usually for Moho xfers, I spend the extra effort to get as high a TWR as I can for that first burn, so I can get it over with as fast as possible.

[GoSlash27]

Aegolius,

 You're doing it right. Plane changes cost proportionally to the orbital velocity.

The one thing I'd recommend changing: After the plane change (or even concurrently), I'd recommend raising your Pe to 680 km. It sounds counterintuitive, but that should save you about 40 m/sec on the final kick.

Best,
-Slashy

 

[Aegolius13]

6 hours ago, WanderingKid said:

if your an/dn is NOT on the apoapsis, you're going to have to burn off point.

Do you mean the node with respect to Kerbin or Moho?  On the former, you should be able to do it anywhere from an equitorial orbit.  On the latter, I usually don't try to hit the node exactly, just within a day or so.  And I guess you could do the first burn 2 days before the node, then get an orbital period of 2 days.

7 hours ago, bewing said:

Oberth does not help plane changes at all.

Right.  The reason to do the plane change burn at peri would not be Oberh, it would be the Pythagorean savings grin doing it at the same time as a ginormous prograde burn.

 

1 hour ago, GoSlash27 said:

Aegolius,

 You're doing it right. Plane changes cost proportionally to the orbital velocity.

The one thing I'd recommend changing: After the plane change (or even concurrently), I'd recommend raising your Pe to 680 km. It sounds counterintuitive, but that should save you about 40 m/sec on the final kick.

Best,
-Slashy

 

I.e, a bielliptic to the gate orbit?  I would have thought that was the way to go.  But when I put down some nodes, it appeared that the final burn to Moho intercept cost less the lower I went.  Maybe I had some other inaccuracy, though.

[GoSlash27]

4 hours ago, Aegolius13 said:

I.e, a bielliptic to the gate orbit?  I would have thought that was the way to go.  But when I put down some nodes, it appeared that the final burn to Moho intercept cost less the lower I went.  Maybe I had some other inaccuracy, though.

Interesting... In that case, nevermind.

Best,
-Slashy

[Aegolius13]

4 minutes ago, GoSlash27 said:

Interesting... In that case, nevermind.

Best,
-Slashy

I may have to recheck this, as it just seems odd.  But there was a thread awhile back about dropping down from Minmus to near Kerbin to do an interplanetary burn.  Might be a similar thing here (without even the minor Oberth provided by burning at a moon).

[WanderingKid]

5 hours ago, Aegolius13 said:

Do you mean the node with respect to Kerbin or Moho?  On the former, you should be able to do it anywhere from an equitorial orbit.  On the latter, I usually don't try to hit the node exactly, just within a day or so.  And I guess you could do the first burn 2 days before the node, then get an orbital period of 2 days.

I meant with respect to Kerbin for the angle you want to leave at for arriving at moho, and it depends on how much change you're trying to get in a single adjustment or if you've gone off-kilter and still need to adjust.  I was thinking aloud and to explain what I meant if it's not making sense would require pictures and much unnecessary typing.

[Aegolius13]

9 minutes ago, WanderingKid said:

I meant with respect to Kerbin for the angle you want to leave at for arriving at moho, and it depends on how much change you're trying to get in a single adjustment or if you've gone off-kilter and still need to adjust.  I was thinking aloud and to explain what I meant if it's not making sense would require pictures and much unnecessary typing.

I think I follow.  If you start in a wonky orbit, there can be problems.  But if you start perfectly equatorial, the new node will just be where do your burn, and at there opposite end.  Do doing the plane change burn at apo should not be meaningfully different than at peri.

It might be tricky to know exactly how much plane change you need when there's still the ejection burn to finish.  But this can be dealt with by putting the final prograde maneuver node up too, and looking at the combined result of the burns.

[Tyko]

I hate changing planes...especially in Dallas/Fort Worth and Philadelphia    

[Aegolius13]

56 minutes ago, tjt said:

I hate changing planes...especially in Dallas/Fort Worth and Philadelphia    

As in KSP,  cost is inversely proportional to speed.