Structural failure on reentry

[blindcoder]

Hello.

I'm running KSP 0.24.2 on Linux with as x86_64. I have a construction that I can easily get into orbit, but trying to land it back on Kerbin always gives me structural failure at around 20-30 km and I can't quite understand why.

A screenshot of the part I'm trying to land is here: https://oc.anderdonau.de/public.php?service=files&t=4cdcb0d214c5bb674768695f95beeecb

At the bottom is a L2 Atlas engine connected to a KW Rocketry SB-2 LFT surrounded by 2x 2 KW TRQ-2 SAS Halves.

My Plugins GameData directory looks like this:

[09:14:38][blindcoder@flora:~/.local/share/Steam/SteamApps/common/Kerbal Space Program/GameData]$ ls
000_Toolbar              Champagne                FrozenPepper       KWRocketry               OpenResourceSystem    TreeLoader
ActiveTextureManagement  Chatterer                HyperEdit          MagicSmokeIndustries     RemoteTech2           TriggerTech
B9_Aerospace             CommunityResourcePack    Interstellar       MechJeb2                 ResGen                TweakScale
B9GearFix.cfg            Engineer                 KAS                ModStatistics            SafeChute             VNG
BoJaN                    EnhancedNavBall          Kerbaltek          ModuleManager.2.3.4.dll  SelectRoot
BoulderCo                ExsurgentEngineering     Kethane            NASAmission              Squad
build.txt                FerramAerospaceResearch  KineTechAnimation  NavyFish                 toolbar-settings.dat

The first thing to break off is always the KER ER-7500, but it vanishes into thin air. Then random parts detach from the LFT and crash into other parts with everything exploding (beautifully) leaving only the Mk1 Pod and its parachute behind.

I'm using FAR, so I would have assumed that the 2.5m expanded fairing base would divert airflow giving me some little aerodynamic. As such it surprises me that the parts that detach are _behind_ that base.

The Player.log can be found here (almost 10MB): https://oc.anderdonau.de/public.php?service=files&t=a4afd48e411f318ef873f7c343d8de43

Anyone got an idea or pointer to help me?

Edited September 18, 2014 by blindcoder

[ferram4]

This is intended behavior, brought on by very high reentry velocities. You're trying to land something that has very little drag for its mass, so it's not slowing down much. Eventually, that's too much for the parts and they start getting ripped apart.

Either lower the density of the vehicle, increase the drag a lot, or do a powered reentry.

[blindcoder]

Thanks for your reply! I'll try that!

Guess I misunderstood how the aerodynamics work here.

[MisterFister]

Another possibility is to do a much shallower reentry. Dynamic load stresses are proportional to both speed and air density (EAS). If unpowered, speed isn't very controllable, but air density is. You would need to bleed off speed while remaining at higher altitudes for longer, before descending to the lower strata of the atmosphere. Depending on how conservative your bleed-off passes are, you may need to perform multiple orbital passes. It's been a while (and a few FAR versions) since I had an assembly like yours that would have required this, but if I recall correctly, the sweet spot is when I can get the AP (on stock-size Kerbin) below 80km while still keeping PE above 60km. MUCH easier said than done, I'll grant, but once that scenario plays out, you likely won't have enough speed to remain aloft for more than one more orbital pass.