Random destruction of wings

[Claw]

Yes. It seems to always be with the small triangular wing pieces. I think the Type B might be the other one, but I can't remember for sure off the top of my head.

Cheers,

-Claw

[Stealthie]

Maybe it's just me but I don't think I've EVER had one of these failures when I'm making a "quick & dirty" landing out in the middle of nowhere.

They ALWAYS seem to happen when I'm on final approach to the KSC, and usually when I'm being all sensible and boring, coming in on a sensible glide-slope at a sensible speed.

Is it possible that proximity to the KSC has something to do with this?

[Jon144]

This Needs Fixed

This bug has happened to my friend Flying Beaver and I as we are trying to land my stock space shuttle from orbit. And it happens after 50% of missions. This is what I classify to be a game-breaking bug as it makes building space-planes pointless. One weird change that most people haven't noticed is that it happens anywhere on Kerbin. Over the water, Near the KSC, the island airfield etc. all under about 15K. The triangular pieces are generally what goes first but then almost every wing part on the entire craft.

It doesn't seem dependent on the approach at all because you don't even have to be near the KSC for it to happen. Again this was on a 100% stock install. Screenshots from this case and a quick-save of the mission it always happens on will be shared in the future. But this is the worst bug I have ever come across in the game. And it is def. not rare in my case.

Here is my friend's persistence file with the quick-save of the instance this bug occurs. https://drive.google.com/file/d/0B0KMCEnqKrRETVZDRG83VGpPS2s/view

It is a stock space shuttle descending from the upper atmosphere. Once it enters the lower portion of the atmosphere wing parts start breaking up. It is a pretty tricky plane to fly in the first place so make sure to keep it steady if you want to replicate it.

It is very frustrating having missions you spend hours or days working on end in failure from something that isn't under your control.

My first hypothesis is that Squad accidentally left some collision polygons in the air that were originally from the KSC buildings but since they were so far away forgot to delete them afterwards. And they are thin enough but stretch over long distances. Perhaps some wing parts are more susceptible to being broken by collisions with them. But for some reason these collision meshes are not always present in the game but are only activated during certain conditions.

To test this it would be interesting to try and launch a spaceplane after another spaceplane triggers the bug during descent. And see if the launched spaceplane runs into the same collision structure. Because the bug sure doesn't happen on launch or atmosphere-only missions. It always seems to happen after a reentry from orbit.

Important note: All buildings within the save file are in an operational condition and do not require repairs.

The easiest way to see if it is a Unity-related bug is to get Unity 5 KSP pushed out the door ASAP. That way we can distinguish between what bugs are due to Unity-5's code and then what bugs are due to KSP's bugs. Good luck to everyone working on this bug.

Edited September 19, 2015 by Jon144

[Majorjim!]

Agreed, this is an old bug that should have been squashed before version 1.0.

If you leave a spaceplane in orbit for a while it will collide with the pad, a building or non existent clamps.

Squad sort this out please!! It is game breaking.

[Jon144]

Agreed, this is an old bug that should have been squashed before version 1.0.

If you leave a spaceplane in orbit for a while it will collide with the pad, a building or non existent clamps.

Squad sort this out please!! It is game breaking.

In my case the space-plane was only in orbit for a day before reentering.

[Mad Rocket Scientist]

Oh, wings. I've hit water module slider.collider so many times I've lost count, I've had the standard canards REVERSE mid-flight after I've reverted them in some way, on a almost regular basis...

[Artimus]

@Claw I've got new interesting information at hand. I just flew a spaceplane to Gilly and back, having *no* small winglets attached but still got into sudden "aerodynamic destruction effects".

The flight included 2 landings on Gilly and several aerobreak turns back at Kerbin. As I was finally approaching KSC the plane started to shake just as if the flight-direction vector was far off the nose-pointing vector, but they were almost the same and the speed was quite low with ~150 m/s and less. When comming lower the shaking increased until the plane always got pushed hard to the right at the nose and thus got torn apart in the middle section. There was absolutely no reason for all of that since the plane was perfectly balanced and flying just straight.

So the whole thing sounds to me as if there are some iterating calculations running for the aerodynamic parts that do.. like "seamlessly" calculate air-based-stuff along ascend/descent through the atmosphere where rounding-errors pile up through time (these can even occur through binary calucation issues at the and of the fraction-part of the variable). This is the only thing that comes to my mind why the plane suddenly gets pushed to one side even when being in perfect balance. Something like one wing has received different aerodynamic values though all that process. This may lead to the high areodynamic stress applied to the parts. Maybe the small triangular wings are just the most fragile of all parts and thus affected first and the tremble caused by their desintegration ripps off the rest. But now I did this very long flight and even the very strong big parts got affected so strongly that they couldn't handle it.

Just to make this post complete you may take a look at the plane used this time:

Edited October 31, 2015 by Artimus

[Artimus]

@Claw Even more information. Yesterday I performed another flight to Gilly with the same ship and returned with very similar effects. The effects were much weaker, but the second travel just took 4 hours compared with 6 hours for the first flight. Thus the plane was not shaking but still had a drift to the right which was noticable by the A.I.R.B.R.A.K.E. on the left (viewd from behind) was going up to work against the draw to the right. Thus this supports my previous conclusion and the key to solve this issue seems to be uninterrupted mission duration. More fragile parts will receive the effects within less time, stronger ones will need several hours.

So... the rest is up to you.

[Commander Jebidiah]

This what cancelled my shuttle program, PDS Columbia was renentering when all of a sudden half of the port wing disappears. After touchdown brakes go on and half of starboard wing goes and no more port wing then the tail fell off. Glad it was only jeb on board so if he wanted to eject he could, dropped Bill off on Alpha Station.

Another time after 5 days in orbit STS Enterprise lost its whole starboard wing and both Bill and Bob were stranded in space and Jeb, and another unknown kerbal, were sent up in RS Atlantis to take them home and dispose of the now dead STS Enterprise, with a HE-KV-1 Missile.

Unfortunatly only;

- STS Enterprise,

- PDS Columbia,

- CTS Challenger and

- RS Atlantis ever saw flight.

HS Endeavour and SS Explorer were built but never saw flight.

Only Columbia and Enterprise were lost.

[Majorjim!]

This is why I have only made one SSTO. Has Squad acknoledged this issue? Because it has persisted for many updates and now, well into release. Due a fix.

- - - Updated - - -

@Claw Even more information. Yesterday I performed another flight to Gilly with the same ship and returned with very similar effects. The effects were much weaker, but the second travel just took 4 hours compared with 6 hours for the first flight. Thus the plane was not shaking but still had a drift to the right which was noticable by the A.I.R.B.R.A.K.E. on the left (viewd from behind) was going up to work against the draw to the right. Thus this supports my previous conclusion and the key to solve this issue seems to be uninterrupted mission duration. More fragile parts will receive the effects within less time, stronger ones will need several hours.

So... the rest is up to you.

Yes these are my findings too. It seems to be related to the amount of time you spend in orbit or on mission.

[Claw]

This is why I have only made one SSTO. Has Squad acknoledged this issue? Because it has persisted for many updates and now, well into release. Due a fix.

Much like the claw bugs, it's a known problem that is incredibly difficult to replicate. To be quite frank, it took me nearly a solid month of begging for saves and replication steps for the claw, then painstakingly debugging my way through the code to find the cause.

Do people run into it? Yep. Some people apparently run into it so often they feel they can't continue with playing the game. Unfortunately the only bits of the profiles that I can gather is that it somehow involves a plane, leaving the SOI, and never quickloading. Once it happens, it seems to repeatedly happen...until you restart the game. So often quicksaves do no good in replicating the problem.

If some of the above few posters actually has a repeatable save file (that persists through KSP restarts), then I'm happy to help dive in. With these complex and difficult to replicate bugs, I see lots of people upset about them. But it's just very difficult to track down. There are a few others, such as the docking ports not wanting to dock/undock, that I still haven't narrowed down. I have suspicions about this one getting triggered by colliders from KSC getting skewed during a world coordinate reposition, and that's why it happens at about the time that you get within range of KSC loading up...but that's impossible to really test without playing the game for hours on end in the off chance that it might happen.

Please note that I share in everyone's frustration. It's very aggravating when it happens, and I'm sorry if you feel that this one bug ruins all of KSP for you. All I can say is make periodic quicksaves. And if it happens, restart KSP.

For anyone who can find repeatable cases, that persist across quicksaves, then I'm all ears. Or if you find some specific details, please post. I would love to get to the bottom of this as much as you guys. I do read this thread, and I've been paying attention to what Artimus has been posting.

Cheers,

~Claw

[Artimus]

So please put away the thought that it is caused by collisions, it simply does not fit! A collision does not apply air shake effects and the likes to a plane. A collision that only destroys a single part without "rushing" through the whole plane seems very unlikely.

In addition, I really doubt you will get ANY save game to replicate the problem as everyone states that the problem is solved by reloading save states - sometimes only after restarting the game, so maybe there are 2 different bugs but who knows.

As others now adviced, too, the only key to replicate the problem seems to be real time mission duration. I would advice you to create a plane with different triangular wing parts, from the small ones up to the Big-ASS Wing and implement a continous log file about the (aerodynamic) stress that is applied to those parts. I'm pretty sure you will get different values - I'd even say different direction vectors (as far as I can just imageine how things work internally) - between start and return.

Thus I fear you have to bite the bullet and "brute force" into the bug on your own by running missions for a few hours.

[Broco]

On 19.1.2015 at 9:47 AM, Farex said:

Greetings!

I also experienced this. Please check this, could be the reason for the random wing-crashes:

When your Wings crash, please go to your Spacecenter and check your launchpad. There should be a little bit of damage to it, like a destroyed tank besides the pad or a destroyed flagpole. The launchpad itself is intact, but parts of it's model are damaged /destroyed.

Every time I have a damaged launchpad, my wings keep on getting crashing when my SSTOs come back to the KSC for a landing. After repairing the damage to the launchpad, this bug never occurred again until my launchpad was damaged again.

Needs confirmation.

I can confirm this, this bug "bugged" me for a long time. Sometimes it appeared, sometimes it didn't and I couldn't find any patterns. I had some Launchpad damage and repairs in between but didn't make a connection to wing crashes (why would I the launchpad was far away). When I read your post I immediately tried it. It works. This is definitely the source of (at least my) random wing explosions.

 

Again for all to read: This bug is caused by a damaged Launchpad. Reproduce by slightly damaging the launchpad (e.g. the tank) and then try to land a spaceplane on the runway.

Workaround: Fix the damage to the launchpad (rightlick launchpad in overview --> Repair).

 

Edit: Maybe I was wrong... I just landed a Spaceplane and after Touchdown on the runway (luckily I could land it) one wing just evaporated going hand in hand with a big explostion at the VAB. It seems that the part collided with the VAB. Would be an ok explanation if the runway wasn't about 50 meters away from the VAB and my plane wasn't ON the runway.

Maybe this helps sort things out though.

 

TLDR: The random explosions of parts may be caused by collisions with buildings even though you're nowhere close to the building.

Edited March 7, 2016 by Broco
Additional Information