Unexpected Adventure

[Tarmenius]

So here I was, about to begin another Mun mission. I\'d been there and back a few times already, but only with the objective of landing and returning safely. For this mission, the goal was to attempt a return from the surface using only the RCS. I knew others had done it, and with successful Mun missions under my belt, I was confident. I never would have guessed what was about to happen.

I rolled out my proven Mun rocket, Prometheus.

The launch went smoothly and although I could have done a better job establishing my parking orbit, I was still within the rocket\'s considerable fuel margin. The Trans-Munar-Injection burn was the same as it had been the other handful of times I had performed it. And the orbit plane needed no adjustments, so without a Mid-Course-Correction burn, the crew had plenty of time to catch up on their reading.

As soon as the Mun\'s Sphere of Influence became the dominant gravitational body, it was time to prepare for capture. I turned retrograde and gave a short burn to set my desired Periapsis then waited for my Lander to reach it. Once at Periapsis, I began the circularization burn. But before I could complete it, the engine shuts off.

In a moment of confusion and mild panic, I exited the map and found four full tanks unused and inaccessible. Looking at a still-elliptical orbit, I assessed the situation.

Realising I have no way to salvage the mission, I decided to jettison the Lander stage and return home. After a visual inspection of the Lander, the cause of the engine shutoff was clear.

I had failed to ensure the fuel lines had been properly connected after re-arranging some parts in the VAB.

Left with only the Command Capsule, 2 RCS tanks and 14 thrusters, I needed to finish reducing Apoapsis so that I could plot a more efficient return to Kerbin.

This seemed more efficient to me, but I don\'t really know for certain

Once I was back in Kerbin\'s Sphere Of Influence, I made my Periapsis at 50,000m to prepare for aerobraking in case I ran out of RCS fuel. Having never before needed to rely on it, 50km sounded like a nice round number just inside the atmosphere where friction wouldn\'t cook my crew but still provide enough drag to eventually bring the Capsule down. When I got to Periapsis, I spent my remaining RCS fuel to bring Apoapsis down to 1,082,300m. And then I waited.

And waited.

Finally, after 11 orbits of aerobraking, and with a total MET of 1:04:40:00, the crew is returned safely. Good thing they can\'t starve or suffocate (yet).

Now, I\'m betting that the astute reader will notice some inconsistencies between the written story and the supplied pictures. Most notably, fuel levels and MET. Sadly, while the original mission was unfolding, I did not have the presence of mind to take screenshots. So, I recreated the scenario. Oddly enough, during the recreation I was much more efficient with my fuel use and had plenty left to deorbit without relying on aerobraking. So I didn\'t rely on it.

But the whole situation got me thinking, and this brings me to the true purpose of this post. Has anyone purposefully designed critical system failures to test emergency abort procedures or ensure that the crew could be saved if something horrible went wrong? If so, what scenarios were run and what did you learn from them?

Thanks for reading, and I look forward to hearing about your scenarios so I can test them myself

[wired2thenet]

Nice thread, and it\'ll be interesting hearing from everyone!

For me, I generally design all my rockets to allow me to 'Panic!' if needed if anything goes wrong. Usually this happens during launch. In one case, I actually managed to land the munar rover on the launchpad at KSP after my SRBs accidentally broke apart and destroyed much of the rest of the craft (I failed to put a decoupler under the command module, so the parachute would have been useless with that much weight). In another instance, I was able to get the crew into a stable low altitude orbit after a critical error on the pilot\'s part, then used aerobraking to get the craft back down, due to no fuel (and splashed down a few km from KSP ... through no design of my own )

One of the things I like to do is save at least 1 RCS tank unused at all times, just for the return mission. Careful management (RCS fuel bug aside) of the RCS has saved MANY a Kerbal.

[Max Grant]

as anyone purposefully designed critical system failures to test emergency abort procedures or ensure that the crew could be saved if something horrible went wrong?

Naw. They respawn on the launch pad . . .

[blossm13]

can someone explain areobraking to me please

[wired2thenet]

Basically, it\'s dipping into the atmosphere so that the friction causes a loss of velocity. If you run out of fuel returning to Kerbin, but can put your orbit so that it dips into the atmosphere (70k or so above Kerbin\'s surface), you\'ll eventually splash down without using any fuel. It just takes a TON of orbits, depending on your Apoapsis.

I believe Aerobraking was used in a NASA Venus and Mars mission.

I sometimes do it just for fun

[ferram4]

What wired2thenet says is correct, including the part about doing it being fun. NASA also used it for the return to Earth on the Apollo missions.

<on-topic>I\'ve never purposefully done this, but I have accidentally designed rockets that always leave the Command Pod around even if it explodes. Normally the result of an SRB taking a LFE with it or a rolling moment causing the entire thing to spin and making the widest stage give a nice demonstration of what happens if there isn\'t enough centripetal acceleration.

I once had a moon rocket, a huge monster, that required a lot of fussing with during vertical ascent because if it didn\'t get that babying it would start to spin, and the last one would inevitably happen. No casualties from that rocket though. I think one of the updates broke it...

[deadshot462]

Having a ship named Prometheus, I\'m sure some kind of Xenomorph sabotaged your fuel lines

Fun story to read, always exciting to return home after a near-disastrous end.

[Tarmenius]

ferram: I had a few early rocket designs that were prone to out-of-control spins if I didn\'t pitch over juuuust right (hell, some that spun out if I went anywhere other than straight-up). Took a while to strike that balance between power and controlability.

deadshot: Whew. And here I thought it was just an oversight on my part. Damn Xenomorphs *shakes fist*. Anyway, I named it that because it was my first successful design. Stole fire from the gods and all that...

In any event, I now design all my rockets with that 2-RCS-tank-14-thrusters emergency stage. And I take screenshots when anything interesting happens... just in case.

[Newt0570]

I had a near identical incident. Except I landed and took off from the mun using RCS only. I had a total of 4 RCS tanks, and when I switched to my last RCS tank during takeoff, there was a visual bug that caused it to look like I had two RCS tanks left. Of course, I didn\'t notice this bug utill I was out of fuel. I ran out with a Pe at around 45,000m and a Ap around 10,000,000m. I think it took 14 orbits to land. But I did it successfully.

Attached is an image of a small ammount of one RCS tank of fuel left to spend.

I recorded the whole flight, but I\'m pretty sure I deleted the footage. It was taking up quite a bit of space.

[Awaras]

The best altitude for aerobraking is between 30 and 45 km. I find that, if after returning from the mun i set my periapsis between 40 and 45 km, I will usually be able to leave the atmosphere and establish a nice circular orbit that allows me to pick and choose my landing point later. 30 km will let you burn off all your speed and land immediately.

[DonLorenzo]

One of my TMI stages consisted of four tanks bolted to the side of the lander stage. One of the fuel lines had gotten connected to an RCS thruster instead of the tank underneath; three tanks would drain instead of four increasingly unbalancing the craft. I finished TMI burn but burned up a lot of RCS for attitude control of the unbalanced craft. Then I landed and had the RCS run out during stabilization, still landed OK. From there on the mission was as usual, except having no RCS on the return stage but that was small and controllable anyway.

[Tarmenius]

Newt, that would have thrown me for quite the loop. Good thing you already had your Periapsis set!

Thanks for the tip, Awaras. I\'ll try those altitudes next time I need to aerobrake.

DonLorenzo, that gives me ideas for testing independant tank failures, thanks!

Had another incident happen this morning. I was testing a different lander design, with an Emergency/Reentry Stage of only one RCS tank and 8 thrusters attached directly to the Command Capsule.

I was showing the game to a friend of mine and was successfully landed on the Mun. He thought it would be funny to reach over and hit the throttle. Not knowing the controls, he hit the spacebar and separated the Lander Stage from the Emergency/Reentry Stage. I botched the ascent, scattering debris all over the side of a mountain. He felt bad, but I quoted Max Grant\'s post above saying 'Don\'t worry, they respawn on the launchpad.'

[Shadowsun]

Reading stuff like this just really makes me wish that there was some sort of mechanic or probability of random mechanical errors in KSP. The potential for something like that randomly happening could make things so much more interesting if it were implemented well.