Aerobraking - when its actually *advantageous*?

[RidingTheFlow]

In lot of tutorials & discussions floating around I almost universally see suggestion that you *should* aerobrake if possible (basically any time you want to slow down while under SOI of body with atmosphere).

However, yesterday I played "make orbital station @Gilly" contract, built specially designed craft to withstand Eve aerobraking, did several attempts to find proper braking altitude, etc.

And after quite some time spent I now suspect it was wasted effort - why? Because I had to spend a lot of dV on *raising periapsis* after successful aerobraking (and Gilly is in quite high orbit). I now suspect that I would've been better just capturing to high orbit by burning.

I googled it a bit but cant find any previous guide/discussion on topic,- basically when it makes sense *not* to aerobrake? (because it will be less efficient this way). Anybody encountered/discussed this issue before?

Edited May 12, 2015 by RidingTheFlow

[SanderB]

You can quicksave and try different altitudes, or go to higher altitudes than normal and use the aerobrake parts to generate immense drag and turn them off again after you've lowered your speed as much as you needed. You need to be able to control of predict how much you aerobrake. That said, aerobraking is almost free, or at least a lot more efficient than burning almost any kind of fuel.

[Streetwind]

Have you actually compared how much dV it would have cost you to slow down propulsively vs. raising the periapsis?

One easy way to do this is on approach to your target planet, put your periapsis just above the atmosphere (even if only simulated, with a maneuver node). Then create a maneuver node at periapsis and drag retrograde until you have achieved your desired orbit.

You will find that for a typical, ideal Hohmann transfer to Eve, your dV cost to brake into orbit will exceed a kilometer per second. If you want to brake into a low orbit, it will likely exceed 1.5 km/s.

Conversely with aerocapture, you skip this insertion burn entirely, but you may have to invest 100-150 or so m/s in dV to raise your periapsis out of the atmosphere afterwards.

Saving 90% sounds like a pretty good deal to me, to be honest

[RidingTheFlow]

Raising periapsis just above atmosphere is not expensive, true. But this leaves you either with low circular orbit or highly eccentric orbit.

Afterwards it did cost me about 1 km/s to raise periapsis for Gilly circular rendezvous eject orbit, then do transfer and capture. That's why I thought it was mostly wasted effort.

It appears that aerobraking only worth an effort if your final goal orbit is relatively low.

Edited May 8, 2015 by RidingTheFlow

[KerikBalm]

I think that is wrong...

For comparison, plot a trajectory where your PE is very high over Eve... about the saem as for gilly, then plot a maneuver node to capture....

The dV will be even worse than a simple capture, due to the lack of an oberth effect.

If you aerocapture with a high apoapsis, then raising your perapsis from a high apoapsis doesn't cost much.

Assuming a capture with no aerobraking at 31,500 km above Eve, your insertion burn will require 988 m/s according to this site:

http://alexmoon.github.io/ksp/

Now if you aerocapture with an Apoapsis at about 50,000 km, raising your PE is not going to take nearly that much.

Don't aerocapture into a low circular orbit... aerocapture into a highly eliptical one

[RidingTheFlow]

Assuming a capture with no aerobraking at 31,500 km above Eve, your insertion burn will require 988 m/s according to this site:

http://alexmoon.github.io/ksp/

It says insertion dV is 513 m/s for me, how did you get 988?

And I would use even higher orbit, more likely about 40Mm+, to be *outside* eccentric Gilly's orbit to rendezvous.

However, assuming with aerobraking we get eccentric orbit of 110/31500 I've checked cost of raising periapsis to 31500 and its about 220m/s. So, yes, its more efficient than pure insertion though not insignificant by any means (e.g. calculators can't really "omit insertion cost if aerobraking is available").

Tricky part is getting apoapsis you need with aerobraking so you will only need to raise periapsis. Getting aerocapture that simply does not smash you into ground is fairly easy, but making it achieve precise orbit is much trickier - and with more eccentric orbits it becomes more tricky.

Edited May 8, 2015 by RidingTheFlow

[Lukaszenko]

You're obviously overdoing the aerocapture/ aerobraking, so no wonder you lose delta-V to make up for it. WHEN DONE RIGHT, aerocapture/ aerobraking can be a very efficient method, but again, you have to do it right. This is easier said than done, as you're basically walking a tightrope. In real life aerobraking is done VERY conservatively, where it takes many months and many passes through the atmosphere to get the final orbit. Aerocapture has never been performed, presumably due to its difficulty.

That said, in KSP you can always save/revert, so that removes most (all?) of the difficulty.

[KerikBalm]

It says insertion dV is 513 m/s for me, how did you get 988?

Good question... I can't get that number again... I don't know what I was looking at or what I did

I think its basically going to come down to the ratio of dV needed to capture, vs the dV needed to go from low orbit to the destination.

The relative velocity when arriving at Eve isn't so high.

Jool is another matter... (and you can aerobrake at jool, then aerobrake at laythe, but I just aerobrake at laythe...I think its still OK with the rentry heating as it is now).

I'm pretty sure it also is easy at Duna... Ike has a big SOI, so its easy to aerocapture at duna, and enter Ike's SOI... at which point, its easy.

Also, keep in mind, your Apoapsis doesn't need to be the right height, just pretty far out there.

For example, the difference between getting a Mun encounter, and a Minmus encounter, is 90 m/s.

Aerobrake and get a large apoapsis, and its easy to adjust. Its also better to have the apoapsis be too big, than too small.

You can fine tune your apoapsis via multiple passes, and do small maneuvers to get an intercept even if your Apoapsis isn't exact.

Aerocapture has never been performed, presumably due to its difficulty.

Unless you count direct capture and descent to the surface.

The rovers (and I think most/all of the probes that landed) directly captured and landed.

There's just never been a capture into an orbit.

[z26]

I think aerobraking is useful if

a) you have a lot of speed to get rid off and/or

your desired periapsis is relatively close to the atmosphere (your wanted apoapsis however can be wherever you want)

In many situations at least one of those statements is true. However, there are some situations, like going from a kerbinsynchronous to a semikerbinsynchronous orbit, where neither of these conditions apply and aerobraking isn't that great.

Also, when you aerobrake you may want not to wastefully lower your periapsis. To do that, you can of course take many shallow passes, but that's boring... A solution is to do just like what you do when thrusting:

To avoid affecting your periphasis you want to gain/bleed off speed as close to it as possible.

You can use airbrakes/parachutes close to the periphasis, then desactivate them when you are too far from it, to remove unwanted drag. That way you can lower the apoapsis without affecting too much the periapsis, at least in theory.

Edited May 8, 2015 by z26

[metaphor]

You can capture propulsively into Eve orbit for less than 200 m/s usually. (how to get to Gilly easily)

Capturing at Jool also takes about 200 m/s or less, or can also be done for free by a gravity assist from Tylo or Laythe.

Aerocapture has never been performed, presumably due to its difficulty.

Aerocapture is not performed in real life because it's usually just not worth it. The heat shield needed has similar mass to the amount of propellant needed for propulsive capture. It's only really worth it for missions to Uranus or Neptune or Titan orbit, it's borderline for missions to Mars, and counterproductive for missions to Jupiter, Saturn, or Venus.

[JedTech]

Steps I use:

1. At the end of your aerobrake maneuver you should have the highest apoapsis possible (without leaving the SOI).

2. at your super-high apoapsis, burn prograde to raise the perhiapsis to just above the atmosphere (very minimal DV required)

3. Plane Change (minimal DV required due to elongated orbit)

[eddiew]

Aerobraking is always advantageous, unless you have a stage to discard after orbital capture, that has enough delta-v for the job

As others have said, you just went too deep this time and got a bit more than you wanted. Since the game has no inbuilt tools for it, I always recommend a quicksave and a few quickloads until you find one that dumps you out with an apoapsis at the altitude you want to be at.

Burning into a circular orbit at high altitude... will really hurt your delta-v, you don't want to do that when you have a nice atmosphere to catch you

[Jovus]

As Eddiew has said, aerobraking is always advantageous. There are theoretical arrangements where it might not be, maybe, but none of those exist in KSP.

[RidingTheFlow]

As others have said, you just went too deep this time and got a bit more than you wanted. Since the game has no inbuilt tools for it, I always recommend a quicksave and a few quickloads until you find one that dumps you out with an apoapsis at the altitude you want to be at.

I understand this now, thanks. Unfortunately, achieving precise eccentric orbit with aerobraking looks extremely difficult even with save-load, since any m/s at periapsis aerobraking makes massive difference in apoapsis height. I have controllable airbrakes on spaceship so can control amount of braking, but its still difficult because I don't see predicted path... Probably I am not going to bother for ~300m/s of dV saving (esp. if whole budget couple of thousands).

[Old Foxboy]

I would strongly recommend using the new Mechjeb to have at least an idea where the journey might be going.

As for the validity of aerobraking: Given the fact you want to have a heat shield, maybe airbrakes and parachutes on a craft only if you are going to land on a planet with atmosphere, i would only do an aerobrake maneuver in those situations. And then you would only need it to bleed off some speed if a direct reentry would toast you in an instant. There's no sense equipping a craft with all this stuff if you are going to Gilly. It's wasted weight, which means wasted fuel from the first second after liftoff, and you can save all this fuel for a nice, controlled, safe, delta-V maneuver in high Eve orbit without the need to aerobrake.

But hey, it's stylish, isn't it? That's why people want to do it.

[Juicy]

I found aerobreaking very advantageous when Jeb's fuel ran out on his descent back home, and he went only as low as about 65 km, while his Ap was still about 700 km out.

An afternoon making circles later, he finally dipped below 59 km and his AP went below 70km.