So, how hard is a Return To Launch Site Abort?

[1101]

Anyone who has ever looked up Shuttle Abort Modes has probably found some sort of description of "RTLS is so near-suicidal, we're glad it never happened".  And while I agree with this statement, KSP gives us an amazing opportunity to try it....

And so began Operation Planned SSME Shutdown!  By giving a contract thicker than a phonebook*, we convinced our brave set of orange-suited kerbals to fly this mission, to boldly go up - and straight back down again.... Nah, we just told them they'd be going to space.

Initial setup was to make some modifications to the Stock Dynawing - payload was drained of fuel, and the front Monoprop tank shut off:

Liftoff followed shortly after, passing through MaxQ with no issues, somehow.  Then came SRB jettison:

At which point, someone switched off Engine #1.  Some immediate quick decision making by Jeb (and urged by some not very surprised mission control) later, and a massive pitch down maneuver began, rolling the shuttle all the way back toward the KSC:

As the craft began to slow down, certain irregularities began:

(Not seen, kerbals' becoming sick as the shuttle rotates like a bullet from a gun)  Fortunately due to the gyroscopic nature of the anomaly, Jeb was able to hold the shuttle back toward the runway, somewhat too successfully (height was still around 30km, speed around 600 m/s) until separation of the external tank, flung away by centrifugal forces:

A short but harrowing descent later, and the craft was near KSC, actually flying.  Jeb exceeded everyone's expectations by actually bringing the shuttle around to runway 09, despite being in control of a slightly aerodynamic brick:

In short order, it even landed:

 

Shortly thereafter, Jeb and crew got back to mission control, and the words "What do you mean, you planned it??" were detected by seismographs around Kerbin.....

In conclusion, RTLS = fun.  If you're not in the craft.

*Phonebook - outdated technology used to store large numbers of telephone numbers.  Superseded by Google.

 

[CatastrophicFailure]

Well, your run through turnout out considerably more successful than shuttle crews thought the real thing would, at least.  Nicely done. 

[53miner53]

I'll have to try this with my shuttle Columbia. I haven't come up with abort modes other than stage until external fuel tank separates, and fly over land and pull the chutes.

 

Edit: due to having liquid fuel boosters with SRBs attached to help the TWR, I tested after both final SRB separation and after final booster separation.

The after SRB separation test had some explosions, but the shuttle landed safely on the runway, though it slid near the end and ended up with its nose off the runway. I blame Jeb, he can't fly by probe control. I should have had him actually in the cockpit.

The final separation test had me at I think 1500 m/s and 62 km(normally it would be out of the atmosphere by this point, but it's somewhat unstable and it was a bad launch), where I used the remaining fuel to basically drop out of the sky and fly back to the runway and land. It would probably have worked better as abort to orbit.

Edited January 28, 2017 by 53miner53

[1101]

For reference, doing this in KSP is rather different to a hypothetical STS RTLS flight profile.  Staging for the above flight(SRB separation) is rather low and at a low speed (~500 - 600m/s).  As a result, and because even a single vector puts out a lot of power, the flight curves back above the earlier gravity turn.  My understanding of the actual RTLS abort is that it should go deeper into the atmosphere as it turns around, rather than higher as mine did.

Of course, the TWR is different to reality, so is the mass ratio, specific impulse, etc.  So these differences are to be expected, I guess.

[septemberWaves]

The difficulty of an abort sequence like the shuttle's RTLS abort depends on whether the vehicle was designed to be able to do it. For example, many space shuttles which I design tend to be incapable of a RTLS abort because I don't often use a NASA-style launch configuration; usually, the launch begins with just the powerful main engines on the tank that the shuttle is attached to, and the shuttle's own engines are not used until the vehicle is approaching MECO (at which point the center of mass is too imbalanced to maintain thrust from just the tank engines without flipping the vehicle). In the case of an abort on launch, the orbiter would have to jettison the external tank and glide down to the land mass that is east of the KSC, because the shuttle's own engines are designed for orbital maneuvering and don't usually have high thrust or a significant fuel supply. The stock Dynawing, by contrast, is designed to mimic NASA's shuttle in terms of launch configuration, and so it makes sense that it would be capable of achieving a RTLS abort.

[NISSKEPCSIM]

I have a shuttle which is capable of RTLS. I even had to use the abort sequence once: 

Cutting chutes to make a powered flight back to the runway:

Touchdown!

The launch configuration for this shuttle is more Von-Braun-style, with the shuttle mounted on top of a reusable rocket, capable of 140,000 of the 150,000 cost refunded by Stage Recovery.

[1101]

It's nice.  Is the lack of Vertical Stabiliser and Rudder the reason you aborted?

[NISSKEPCSIM]

Yes. I was ascending when suddenly the booster started to flip and lose control. When I decoupled the shuttle, it started to stall due to the lack of a vertical stabilizer. I fixed the problem by adding stabilizers and a reaction wheel to the booster.