Mars mission planning, eccentric orbits at Mars

[Laie]

I want to do a Mars mission that basically follows this scheme (never mind the calendar dates; it's the relative positions of the bodies that's giving me a headache, which is pretty similar for any launch window):

In order to save on dV, I don't want to circularize but rather enter into a very eccentric orbit around Mars. The problem then becomes that the possible orbits I could make at arrival are totally different from what I'd need at departure. Below you see the trajectories I'm coming in at, and I ought to go out on.

The pic makes it look as if the maneuvers were 180° apart, actually it's more like 120°. With a "reasonable" amount of radial (say, 10% of total insertion dV), I can tweak the incoming periapsis by 15° or so. Another 10-20° can be gained from midcourse maneuvers (which is a lot and possibly exploits patched conics). Still, that's only a few drops in an bucket.

Currently, I see no way of getting from arrival to departure that does not take me through a low circular orbit, or at least would require just as much dV. But maybe I'm missing something? If you have an idea, please let me know.

To clarify: I'm looking for a way of either shifting my argument of periapsis by a lot, or find an outgoing trajectory that requires less of an periapsis shift, or any combination thereof. The available dV budget for such shenanigans is pretty low, on the order of 300m/s or less.

Edited September 9, 2016 by Laie

[Streetwind]

I found your description to be a bit confusing, so let me sum this up in my own words to check if I understood you correctly:

The argument of periapsis you establish at arrival will be unhelpful when you want to leave again several months later. You are looking for the strategy to follow in order to achieve the minimum dV cost total for the sum of trans-Mars injection, Mars capture, and trans-Earth injection. The mission requires an orbit with a low periapsis, but the apoapsis does not matter. Correct?

Edited September 8, 2016 by Streetwind

[Laie]

1 hour ago, Streetwind said:

The argument of periapsis you establish at arrival will be unhelpful when you want to leave again several months later. The mission requires an orbit with a low periapsis, but the apoapsis does not matter.

The above is correct.

As for strategies, the chosen strategy is to enter into, and depart from, a highly eccentric orbit. I wonder if it can be made to work. While some adjustments to the mission plan are obviously in order, I'm not open to suggestions that would amount to an entirely different schedule (like aerobraking into a low orbit).

Hmmm. Maybe a *very* eccentric orbit and doing a 180 at the high apoapsis, turning it retrograde? (<- this ought to work well, but it feels totally wrong. Would a real space agency even consider it?)

Edited September 8, 2016 by Laie

[Streetwind]

Well, you have a very, very large number of possible solutions, if we include various ways to shift the argument of periapsis but chalk their cost up against the mission. This is probably not something you can easily calculate.

A first step would probably to determine the dV cost of just flying the mission straight: make a normal trans-mars injection, capture into your eccentric orbit, perform your mission, and then make your trans-Earth injection despite the woefully misaligned orbit. Then, once you know that, you can try the orbit reversal method instead. Or midcourse maneuvers to adjust the argument of periapsis. Or other such things.

[GoSlash27]

Laie,

Is there some reason why you need to do a highly eccentric orbit besides saving DV? I haven't run the numbers, but it seems to me that it'd be cheaper to circ into the gate orbit instead of doing it this way.

Best,
-Slashy

 

[Laie]

23 minutes ago, GoSlash27 said:

Is there some reason why you need to do a highly eccentric orbit besides saving DV?

None at all.

Some more self-imposed restrictions:

I can send the mission to Mars on hydrolox or possibly even with nukes, but once under way I've restricted myself to storable propellant and ISPs on the order of 315s. I want the whole mission in orbit before landing. Capture at mars definitely has to be done propulsively; later aerobraking to lower AP is an option of last resort, I'd prefer if I could avoid aerobraking altogether.

The option to park in an eccentric orbit seemed like an easy way to conserve dV: the ascent vehicle is very lightweight, the return vehicle is not.

[GoSlash27]

10 minutes ago, Laie said:

None at all.

Some more self-imposed restrictions:

I can send the mission to Mars on hydrolox or possibly even with nukes, but once under way I've restricted myself to storable propellant and ISPs on the order of 315s. I want the whole mission in orbit before landing. Capture at mars definitely has to be done propulsively; later aerobraking to lower AP is an option of last resort, I'd prefer if I could avoid aerobraking altogether.

The option to park in an eccentric orbit seemed like an easy way to conserve dV: the ascent vehicle is very lightweight, the return vehicle is not.

I see... Well, you may be able to save more dv by using the gate orbit instead of doing the high eccentricity thing. That would kill 2 birds with 1 stone.

Best,
-Slashy

[Laie]

25 minutes ago, GoSlash27 said:

you may be able to save more dv by using the gate orbit instead

Gate orbit?

Edit: never mind, wikipedia has it covered. Thanks for the hint!

Edited September 8, 2016 by Laie

[stratochief66]

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20150009466.pdf

See FIgure 6, 'Butterfly Orbit"

I saw something like the above with more detail in a printed book that chronicled past Mars missions and contained a bunch of documentation from studied crewed Mars missions, but I can't link that obviously.

[Laie]

2 hours ago, stratochief66 said:

http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20150009466.pdf

See FIgure 6, 'Butterfly Orbit"

"For the Hybrid mission design a set of maneuvers termed the “Butterfly” (Figure 6) have been developed that use third body effects and SEP thrusting near the edge of Mars sphere of influence for alignment. [...] The 10 Sol transfer in Figure 7 is 110 m/s total ∆V and 100 days for the transfer. [...] The duration of the butterfly maneuver varies and is up to 250 days, however the required ∆V is nearly the same."

Third body effects are not available to me, however, just inverting the orbit at high AP can be done for ~110m/s an will work similar enough in our patched conics. Can't say that I'm happy with that, but it's as good as it gets.

[maccollo]

21 minutes ago, Laie said:

Third body effects are not available to me, however, just inverting the orbit at high AP can be done for ~110m/s an will work similar enough in our patched conics. Can't say that I'm happy with that, but it's as good as it gets.

Well, there is principia. Weak boundry can be quite tricky, but this one is very easy. Simply burn to get captured. Right as you cross into a captured orbit, the trajectory will end up looking like this, so you only have the do  the burn to get kicked down into an elitpica trajectory again, which should be about 175-200 m/s, so you save a couple of 100 m/s compared to doing the same maneuver with patched conics.

[Laie]

1 hour ago, maccollo said:

Well, there is principia.

I thought of that myself, and from the look of your screenshot, a lot has happened since I last used it. How usable has it become? I mean, this is RO -- I can't just pack a little extra dV and enjoy the ride as Principia tosses me around. Before I even build my vessels I need to have a pretty good idea of the entire trajectory from launch to landing. Are there any planning aids?

[GoSlash27]

Laie,

 Did you check the numbers for a full circularization at the gate orbit? How does the DV budget stack up in comparison?

I can run the numbers on that if you'd like.

Best,
-Slashy

[maccollo]

The gateway orbit, which would be at about 4000 km above Mars, will only lower the round trip delta V from 4000 to 3600, while the elliptical orbit approach will lower it to less than 1600 m/s... If it is executed correctly, and it might be more on some windows since Mars is like Moho when it comes to consistency in transfer windows.
The MAV would need 1 km/s to reach the gateway orbit from LEO, compared to 1.35 km/s to reach the ERV in a highly elliptical orbit.

6 hours ago, Laie said:

I thought of that myself, and from the look of your screenshot, a lot has happened since I last used it. How usable has it become? I mean, this is RO -- I can't just pack a little extra dV and enjoy the ride as Principia tosses me around. Before I even build my vessels I need to have a pretty good idea of the entire trajectory from launch to landing. Are there any planning aids?

 

Principia is quite stable now. I tested it quite a bit by running it doing a career run with it, and it works quite well.

n-body orbits only gets really tricky when you are dealing with low energy trajectories, but for the most part you don't actually use those. Orbit that are set firmly inside a gravity well don't suddenly go crazy, and the flight planner allows you to predict years into the future. 
As an example, here is a rough flight plan starting from getting captured into an eliptical orbit, all the way back to Earth using 1 km/s of deltaV. It's not quite as efficient as it could be. The flight planner is a bit awkward for these multiple maneuver plans, but it is still sort of manageable.
http://imgur.com/a/ApwVL 

 

To plan more generally for transfer windows and such, I'm not sure the regular transfer planner is going to be accurate, I've been using this one that was made for orbiter.
http://www.orbithangar.com/searchid.php?ID=5034

 

Edited September 9, 2016 by maccollo

[Laie]

10 hours ago, GoSlash27 said:

Did you check the numbers for a full circularization at the gate orbit?

I did not, because even at first glance it was obvious that it would need a lot more dV than I've been accounting for. The problem is that I still need to bring down the lander. Given that we're talking about a 18+-month surface stay, it's not exactly lightweight. Gate orbit might be marginally better for a mere there-and-back (Macollo did the numbers and says it isn't), but once you factor in the landing I don't need no calculation to know that it won't work out.

Full-on bragging mode (also, for context), the lander:

5 hours ago, maccollo said:

Principia is quite stable now. I tested it quite a bit by running it doing a career run with it, and it works quite well.

That was not my main concern. I know that even low orbits change over time (spent stages from moonshots that used to be coplanar to the moon, after a year, were somewhere else entirely), and for a highly eccentric one I'd expect both argument of PE and inclination to wander quite a bit while I'm on the ground.

I guess I'll have another look at Principia, but only *after* I'm done with this mission.

Edited September 9, 2016 by Laie

[GoSlash27]

Laie,
 Ah. I didn't see anything earlier about a 'lander', else I wouldn't have mentioned it.
 

Good luck!
-Slashy

 

[Palaceviking]

I'm a bit of a noob at this but I'd flip it just after apoapsis and put your periapsis exactly where you want it.

[Laie]

Excuse the poor quality, there's probably a setting somewhere to make the projections stand out more but I couldn't find it.
The bright yellow thing is the markup probe, scheduled departure maneuver on the right.

The purple PE is from the incoming orbit; the blue/cyan one is after high orbit maneuvers.

The late capture burn already moves PE by a bit, it's hidden behind the yellow one. But the most important part is the maneuver at high AP that not merely swaps the inclination, but goes into a much less eccentric high orbit. Which is later brought back down. The high orbit maneuvers are quite cheap, and doing retrograde burns quite a bit away from PE likewise comes a lot cheaper than I'd have expected. But of course it takes a good long while to move around mars at high altitudes.

The NASA paper cited by stratochief assumes that the return vessel will do independent maneuvers in mars orbit (cheaply on electric engines, it has almost two years after all), while my mission was designed such that I have to bring it into it's final orbit *before* I detach the lander. Having the crew sit idle in space for another four to six weeks is painful to watch.