How to ditch a plane?

[InterCity]

Hello, kerbonauts.

I've recently learned how to target KSC upon landing with my spaceplanes, but I'm still quite sloppy at that and often have to ditch my planes in the ocean due to pilot errors/whatever. I am terrible at these things as well - I lost 2 of 3 flights I attempted to ditch. I don't know whether I was doing it right, but I reduced my airspeed to borderline stall, lowered the landing gear and pitched all the way up just before touchdown in vain hope the tail will soak up most of the damage.

What can I do to save my brave astronauts? Is there a better way to land stuff in water? My planes have stall speeds around 70-80 m/s; if that helps...

Edited November 7, 2015 by InterCity

[magnemoe]

60 m/s is to high most of the small cockpits has 45 m/s and the mk 3 has 60.

Problem with water in KSP is that bottom parts don't act as crumble zones at least in far less way than on ground so you can destroy even an high rocket completely coming down just a bit above crash limit.

best way is probably emergency parachute(s). Mounted far forward they will land you tail first and you could even use this to do an powered decent on the jet.

Else the only objective is to get speed down as much as possible.

[Tex_NL]

Pictures of your plane could be useful.

70 to 80 m/s is rather high. I usually land my planes at that speed and prefer them to be able to fly significantly slower than that. You could try adding some more wing surface.

If you tend to overshoot the runway you obviously started your landing sequence too late. Start earlier and don't aim for the spot you try to land. Instead aim for a spot slightly in front of the runway. That way you'll end up on the runway when you flare (pull up) the nose of your plane.

In contrary to real life land is more forgiving than water. A rough landing on land will most likely break your craft in several pieces but many will survive. A rough splash-down will most often result in a total loss of craft and crew.

Edit: Ninja'd. Once again I took too much time to be clear.

[Skorpychan]

Parachutes and some height. Water's pretty much as hard as the ground.

[FancyMouse]

Nope 139m/s isn't an issue

EDIT: Oh you mean water? I don't know - hope for 1.0.5, then. I never trust water in KSP.

Edited November 7, 2015 by FancyMouse

[Tex_NL]

Nope 139m/s isn't an issue

This is an excellent example how at least part of a craft can survive a high speed land impact. The same landing on water would guaranteed have been fatal.

P.S.

If you don't know where the term 'rendered speechless' comes from. Now you know.

[jarmund]

Better Buoyancy is a mod that makes water behave more like water and less like liquid black hole. It makes dirhing planes a lot easier and more realistic. What you described sounds like the kind of emergency ditch i often do during test flights. Hopefully 1.0.5 will be similar.

[FlyingPete]

I recommend having some abort mechanism on the spaceplane. With planes built around circular fuselages, you can just use a decoupler to separate the cockpit from the rest of the plane and have that parachute to splashdown. On anything else, fit radial parachutes to several points around the plane and tie them into the abort action group. Deploy them high enough and they should bring the entire plane to a survivable splashdown.

[InterCity]

Thanks for the answers, guys.

Someone asked for the plane. It's this:

when full

when empty

... my most successful ditching attempt so far

I've probably been overreacting before, since this attempt was at 60 m/s. Still, I'd love to know what to do...

[Mastikator]

Water is pretty much the worst place to land, land on the runway or just the ground. Hell, landing on a mountain is better than the ocean.

[FungusForge]

Stock water is evil as it is. I can suggest the Better Buoyancy mod, it makes ditching aircraft a whole lot safer. Still better off with landing in the hills, but it makes a water landing not suicide.

[Tyren]

maybe use the stock radial intake on the bottom of your plane. Afaik it is sturdy enough to function as a oh-....-water-inc-gear.

[AeroGav]

Hi Jeb,

I'm also going to agree with what the others have said, landing speed is too high. Most of my designs have a stalling speed below 40 m/s when returning empty. As for the first Mk 3 I ever built ...

Crash landing at 25 m/s doesn't even frighten Bob Kerman, well not much anyway.

You can see this one just stitches together four Big-S deltas to make one very large delta. Effective and low part count, but not pretty. Your shuttle is rather short and fat , compared to my aircraft, so it needs a shorter span config.

How about the approach used on my Minmus mining ship?

I've clipped some Big S wing stakes to the leading and trailing edge of the main delta, to make a larger surface. It results, surprisingly, in a rather nice looking ogive (?). The strakes are longer than the edges of the delta , so they are clipping into the fuselage a little. You could use some rectangular pieces to make the delta start further out from the cargo bays.

Big S strakes and Deltas are the min-maxer's choice of wing part, because they have the same 10:1 lift:mass ratio of the best wing parts, heat tolerance and tankage for liquid fuel. However I don't see any airbreathers on your design, so maybe this fuel capacity goes unused? In which case, use any wing pieces you like, except one of the swept wings needs to be avoided as it's lift : mass ratio is poorer than the others.

Of course, lack of wing area is not the only reason for your high landing speed. If you fly with F12 (aerodynamic forces display) on, you will see that your elevons are raising the nose by applying downforce to the trailing edge of the wing. Relatively close to the plane's centre of gravity, they have to use a lot more (down) force to get the nose up than a conventional tailplane further back. But the best thing would be a canard surface up front, which adds lift rather than subtracting it when trying to get the nose up.

As a minimal change, I'd say

1. add canard

2. as this moves the centre of lift forwards, move the main wing back slightly if necessary

3. remove the elevons from the trailing edge and attach a strake instead, for a .... tapered trailing edge.

Why strake over elevon? Because control surfaces like elevons only have 5:1 lift : mass ratio , and no fuel tankage.

If you're worried about the effect of all this on your ship's performance,

http://forum.kerbalspaceprogram.com/threads/138291-Wing-Lift-Wing-Lift-to-Drag-Ratio-Charts

you can see that best lift:drag ratio occurs at only 2-5 degrees angle of attack, which needs a lot of wing at high alt. There are counterarguments, but i'd say a larger wing won't noticeably worsen performance, at least. My "goose" doesn't want to go fast below 10km. However it can climb to 16km subsonic, then transition to supersonic. In the very thin air above 20km, it flies very well.

Finally, on the subject of crashworthiness, I'm not yet convinced by the Mk3 cockpit. Yes it has higher impact resistance than others, but its positioning means Jeb & Co are always the first to arrive at the scene of an accident. This type II/III mining hybrid has an inline cockpit with landing gear below, canards either side, a crew cabin, cargo bay with ore tank, type 2 to 1 adapter (rocket fuel), service bay with 2 batteries, and a shock cone in front as "crumple zone". Which is just as well as horizontal landings on Minmus are pretty terrifying.

[Tex_NL]

Thanks for the answers, guys.

Someone asked for the plane. It's this:

http://i.imgur.com/46rayUI.png

when full

http://i.imgur.com/LRLydjk.png

when empty

http://i.imgur.com/2dHmugQ.png

... my most successful ditching attempt so far

I've probably been overreacting before, since this attempt was at 60 m/s. Still, I'd love to know what to do...

Many of my shuttles look nearly identical except for the wings. I've used TweakScale to increase them by at least 20%.

When empty your CoL is quite far behind your CoM making it hard to fly. Moving your wings forward will solve this. Yes, this will move your CoL in front of your CoM during launch but that's completely irrelevant as you won't be launching it from the runway anyway. Just make sure your complete launch vehicle (including tanks and boosters) is aerodynamically stable.

[alpha tech]

I would recommend VNG parachute\eject system

[JTpopcorn]

Land on land, water is deadly. Have enough parachutes to slow yourselft down. My highspeed bomber has a regular chute used as a drogue, and worse case it slows the whole craft to within the limit of the cockpit, and it bounces off its nose onto its landing gear.

[Scarecrow88]

I find the best way to make sure my spaceplanes come back safely is to land them back at KSC. If you make your de-orbit burn so that your projected orbit touches the surface of Kerbin about twice the distance to the east of the KSC as the island with the second runway is, as long as you have a small amount of fuel left you should be close enough to the KSC to be within flying distance to land there.

[Tex_NL]

@ JebKerboom

Based on your pictures I tried to recreate your design as best as I could and my earlier suspicions were confirmed. Your design with empty tanks is a nightmare to fly. I have landed roughly a hundred of my shuttles nearly without incidents but yours was not one of them. I lost control during re-entry and ended up in a 60 degree nose dive unable to pull up.

Then I moved the wings forward as I suggested (see the picture below). I horribly overshot the runway (my shuttles have airbrakes, you too might want to add a few) but it flew a lot smoother and was able to divert to an island north east of KSC. Even as slow as 40 m/s it was still flying and not stalling.

Escape systems and parachutes can be useful but you do not need them. All one needs to land this shuttle is practice.

Edited November 7, 2015 by Tex_NL