Need Tips on Duna Aerobraking

[Geschosskopf]

I'm about to make my 1st crewed trip to Duna and would like to aerobrake (although I THINK I have enough fuel to do without). I've looked all over and can't find any particularly clear general guidelines so would like some help on this and related issues.

The ship that'll be doing it is shown in the last pic of this album:

Javascript is disabled. View full album

So my questions are:

1. The central axis of the ship is held together with 3 of the standard size FusTek CBM docking ports, no struts or quantum struts. It's not very aerodynamic and I use FAR. Will this stand up to aerobraking?

2. I hear a 50km orbit is what I want to end up with. I would rather do the aerobraking gently over the course of several orbits than do it all at once. I'm not that interested in saving fuel, just learning how to do it and what effects it has so I can plan to use it in the future. What's a good periapsis altitude to shoot for?

3. At what point during the process, as my apoapsis comes down, should I start panicking? I mean, I want it to come down, but I'd like to know in advance how much and how fast is good and how to recognize when things are going badly wrong.

4. The front part of the ship is a MapSat probe intended for Ike. It has about 2500 delta-V and needs to end up in a 185km polar orbit. Is it better to release this on the way in or after I've achieved LDO with the mothership?

Thanks in advance.

[GJames]

That's a pretty well put together craft there, quite a lot of RCS though!

1. I don't personally use FAR, but Duna's atmosphere is quite thin, especially up near 15km, so it should be fine (If anyone who does use FAR could elaborate that would be appreciated).

2. I think for Duna you want to aim for about a 15km periapsis, though it does very much depend on your velocity as you arrive. 15km should get you at least in some kind of orbit, or very close to one so a tiny burn should be able to capture you. Once you've done this adjust your periapsis to something between 30-40km, depending on how far you want to bring your orbit down.

A 50km orbit is quite low, as Duna's atmosphere starts around 41km. When aiming for such a low orbit, it's usually best to do the last bit using only rockets, just to be on the safe side.

3. You should probably start panicking if you get captured before you've reached your periapsis, but as it seems like you have a decent amount of fuel. you'll probably be okay as long as you don't get caught out by it.

4. 2500km/s dV is plenty, so you could release it after your in a stable orbit, but you could also release it a couple of weeks before you get there and give it a little boost so it gets there a few hours before the main craft. Up to you I suppose, first option is safer, second is a bit more interesting...

Unfortunately there is no hard and fast rule to aerobraking so remember to quicksave first...

Edited June 30, 2013 by GJames

[Richy teh space man]

If your using mechjeb 2, then go to the landing autopilot and there's an option on there to show your predicted orbit after aerobraking.

As to your questions:

1 Should be, F5 and F9 if it doesn't work

2 See above on mechjeb 2

3 Just needs practice.. F5 and F9 if things go badly

4 May as well release it in low orbit, you can use the Oberth effect then to get to Ike

[Geschosskopf]

Thanks for the tips.

OK, 15km periapsis. I assume this will decrease somewhat as I go through, and that I'll have to raise it again before making another pass. What is too low?

I got the 50km LDO from the delta-V map on the Wiki. It that's too low, what's better?

Hold the satellite until after orbit. F5/F9. Got it . I take it ships held together with only docking ports are OK for this?

@ Richy:

Blame the copious RCS on FAR. If you looked at the various launch vehicles shown in that album, you'll note they're all built like arrows, long and skinny with lots of tail feathers. This is pretty much a requirement for rocketry with FAR. Being so long, they have HUGE moments of inertia to pitch and yaw so until you ditch the ascent stage, you need LOTS of RCS to maneuver at all. This is especially true to aim prograde at the apoapsis burn when you're above the atmosphere and all the tail feathers become useless. I've found through much trial and error that a good rule of thumb is that for each big stacked tank in the ascent stage, you need 4 RCS on each end of the rocket.

So, the bigger rockets in my pictures have 8 RCS on both ends. That requires a lot of RCS fuel, and the payloads needed some themselves for docking anyway. Thus, I need 2 RCS tanks, 1 on the ascent stage and 1 on the payload. I design my ascent stages to reach orbit with just a bit of dV to spare, so once I'm up and stable I pump any remaining mono into the payload before staging.

The benefit of doing things this way is that you can reach LKO with rather less than 4000 dV on the ascent stage. But OTOH, the long tall rockets run you into the ceiling so your payloads have to be short, and also in fairings if they're too lumpy.

[Johnno]

Try this: http://alterbaron.github.io/ksp_aerocalc/

[Geschosskopf]

Try this: http://alterbaron.github.io/ksp_aerocalc/

Does that work with FAR or stock air?

[Vanamonde]

Coming from Kerbin, a 13,000m pass at Duna will drop you into an orbit with a periapsis of about 40% of Ike's orbital altitude. This ship has done that 3 times without taking any damage or losing any parts,

so yours should be fine.

[Geschosskopf]

@ Vanamonde:

Thanks for the encouraging example. 13km it is, then. But what altitude would you recommend for the final orbit that gives reasonably easy access to both Duna and Ike for the lander and probe?

Quite an impressive ship you have there! I'm intrigued by the stacks with the ports between them, looking like lathe-turned bedposts. Obviously you made them that way for a specific purpose but I can't figure it out .

[entaran]

Duna's atmosphere is so thin you can come in at interplanetary speeds into a 5km periapsis and still get spat out the other side (then adjust periapsis at apoapsis) and repeat. You can "brake" at duna with like 120ms d-v total because it's literally impossible to stop on your first pass if you're coming in from kerbin/another planet (not ike) due to the atmosphere being too thin. The only thing that will actually stop you is hitting a mountain... which whilst pretty, is not exactly the best "option"

Short version, there's no "panic" point on the first aerobraking pass. Ever.

Also, your final apoapsis is really dependent on the weight of your craft and a few other things like inbound velocity etc. Your best bet is to f5 it as you enter the final plane change/burn before SOI then see what happens with 4-5 passes at different heights and maybe take some notes. Then you can toy with the exact height. Duna makes pretty flames

[Vanamonde]

Duna's atmosphere is so thin you can come in at interplanetary speeds into a 5km periapsis and still get spat out the other side

Not meaning to be argumentative, but that doesn't sound right to me at all. While the upper reaches are thin, the effect ramps up extremely rapidly, and when I tried 10,000m it ended up as a ground impact.

At any rate, 13,000m drops you in an elliptical orbit with an apoapsis roughly halfway between Duna and Ike. From that first pass, it's pretty easy to alter the orbit to whatever you want it to be.

As for the ship in my pic, the shape is due to that being a multipart mission. It consisted of the ship that made the round trip,

in this instance carrying a lander,

a shuttle to run crew from high orbit to low,

and two robotic fuel tankers.

[K^2]

Alright, so in simulation, for periapsis of 12,745m after entering Duna SOI on transfer from Kerbin, I'm getting an apoapsis of 240km. Acceleration never exceeds 1G during the maneuver. This is consistent with what Vanamonde suggested, so I say go for it.

[Geschosskopf]

Tbanks again for the info....

@K^2:

How do you simulate stuff in KSP? When I say "simulate", I mean I have a separate saved game where I playtest out my ideas, but that's just as much trouble as doing it in the real persistent game. Is there another way?

[K^2]

Yeah. I punch the equations used by KSP into Mathematica and solve them there. Fortunately, they are very simple equations, so NDSolve method has no trouble with them.

[Meithan]

Yeah. I punch the equations used by KSP into Mathematica and solve them there. Fortunately, they are very simple equations, so NDSolve method has no trouble with them.

May I bother you to elaborate?

I've been wondering just how KSP models aerodynamics. You say it's simple? Something like quadratic drag with a fixed coefficient of drag (around 0.2 I hear)? And after that, how do you couple this with the orbital mechanics to determine the final orbit? I'd really be interested in the details, if you have some time.