Shuttle wings rip off on landing

[Sandman1330]

Hey everyone, long time lurker first time poster.

 

Having managed to visit Duna and Eve, I've decided to turn my sights onto my nemesis in game - spaceplanes.  My first attempt was a Mk2 that turned out surprisingly well.  It can take off, go to orbit and land all on internal fuel, and is quite easily controllable.

https://steamcommunity.com/sharedfiles/filedetails/?id=2071387439

I then decided to try a go at a heavy Mk3, obviously as you can see heavily influenced by the space shuttle.  However, while I can launch, get to orbit, re-enter and fly a final approach just fine, the second I touch down on the runway the wings shear off, even on fairly gentle landings. I usually touch down at about 180m/s, any slower and it wants to stall.

https://steamcommunity.com/sharedfiles/filedetails/?id=2071385507

Without the tank:

https://steamcommunity.com/sharedfiles/filedetails/?id=2071396584

As you can see, I've used both autostrut and rigid attachment, and heavily strutted the wings externally as well.  However, nothing seems to help - on touchdown, the wings fold up like a cheap paper plane.

 

Edit: OK guys, I've made a video of the landing. In this one I set it down about as gentle as I think is possible in game, given KSP's awful controls for flying spaceplanes.  As you can see, as soon as the wheels touch, the wings fold right up.

 

 

Can anyone help?

 

Thanks!

 

Edited April 23, 2020 by Sandman1330
Typo

[4x4cheesecake]

3 hours ago, Sandman1330 said:

rigid attachment

This will make your wings breaking even faster though. A little wobble may looks dangerous but it will act like a spring/damper to your construction. Anyway, the easiest way to solve this issue is to attach the gears not on the wings but the body of the shuttle/spaceplane and use the offset tool to move them to the wings. It will look like the gears are attached to the wings but actually they are still connected to the body which is a lot less fragile.

If you don't like to use the offset tool in this way, you can always try to add just more gears to distribute the stress on the wings/gears.
Also, at least in the video, you seem to carry back a lot of fuel, you should consider to start with less fuel so your spaceplane becomes much lighter during the landing.

[Sandman1330]

2 minutes ago, 4x4cheesecake said:

This will make your wings breaking even faster though. A little wobble may looks dangerous but it will act like a spring/damper to your construction. Anyway, the easiest way to solve this issue is to attach the gears not on the wings but the body of the shuttle/spaceplane and use the offset tool to move them to the wings. It will look like the gears are attached to the wings but actually they are still connected to the body which is a lot less fragile.

If you don't like to use the offset tool in this way, you can always try to add just more gears to distribute the stress on the wings/gears.
Also, at least in the video, you seem to carry back a lot of fuel, you should consider to start with less fuel so your spaceplane becomes much lighter during the landing.

Interesting, thank you!  I will try all of these things.  I did try with near empty fuel already, it was not nearly as bad but still ended badly. I left the fuel in intentionally in order to simulate returning with a payload, as in the video and all of my testing so far I've always come back with an empty cargo bay.

[Cavscout74]

Just going by the 180 m/s landing speed makes me think you have too little wing for the craft.   For comparison, the cruise speed of a turboprop airliner is typically around 140 m/s.  Landing speed for a 747 is only going to be around 80-85 m/s.  Or a better comparison, the space shuttle's final approach speed was only around 100 m/s.

[Sandman1330]

3 hours ago, Cavscout74 said:

Just going by the 180 m/s landing speed makes me think you have too little wing for the craft.   For comparison, the cruise speed of a turboprop airliner is typically around 140 m/s.  Landing speed for a 747 is only going to be around 80-85 m/s.  Or a better comparison, the space shuttle's final approach speed was only around 100 m/s.

Yeah, I thought it high myself but as my ship is effectively based on the real life space shuttle, using the KSP Big “S” parts, I thought the wing size should be proportionate.  Also, on review of the video, while I was targeting 180 on approach, after the flare the actual touchdown speed was 135.  Landing with much lower fuel I got it below 100m/s, with the same effect, but that weight would be unrealistic if returning with a payload in the bay, and after installing other equipment like the arm, etc.

 

Using Cheesecake's tips, in 2 tries I’ve managed to keep it together once, though I missed the runway and hit the grass.  The other time I landed on the runway, but still lost one wing.  This time it said the wing collided with the landing gear, which makes me think it flapped downwards on touchdown and hit the gear, which was in effect hanging below the wing due to the offset.  Tomorrow I’ll try tweaking a few more things, maybe turning rigid attachment back on will stop the wing from flapping.  I may be getting closer.

[4x4cheesecake]

In my opinion, vertical speed is more important during landings then horizontal speed. If you go too fast horizontally, the worst thing that can happen is running off the runway but you can always add chutes and tweak the breaks to slow down faster. But if you hit the ground with more then 10m/s vertical speed, you will most likely crash immediately or at least some parts will break apart.
This becomes more easy to handle with larger wing areas because your AoA doesn't need to be very high at low velocities to keep the vertical speed below 10m/s but it is definitely doable with small wings as well, it may take some practice to perform the flare maneuver correct though.

[Sandman1330]

12 hours ago, 4x4cheesecake said:

In my opinion, vertical speed is more important during landings then horizontal speed. If you go too fast horizontally, the worst thing that can happen is running off the runway but you can always add chutes and tweak the breaks to slow down faster. But if you hit the ground with more then 10m/s vertical speed, you will most likely crash immediately or at least some parts will break apart.
This becomes more easy to handle with larger wing areas because your AoA doesn't need to be very high at low velocities to keep the vertical speed below 10m/s but it is definitely doable with small wings as well, it may take some practice to perform the flare maneuver correct though.

Yeah, the biggest thing I'm struggling with is the controls.  Trying to flare smoothly when your pitch control is all or nothing (as it is with the keyboard) is extremely difficult.  I've tried an xbox controller, but I always seem to get a bit of roll in with the pitch.  Maybe I'll try separating the axis into two sticks...

[Sandman1330]

*Sigh*

 

I have to confess that I am starting to get very frustrated. None of the advice above worked for my current shuttle. Attaching the gear to the fuselage and offsetting it only resulted in the gear staying in place while the wing ripped off anyway.  The gear was clearly attached to the fuselage, and stayed in place when the wing went flying, so what stress was the wing under to cause it to rip off so easily???

 

So I gave up on that shuttle, and built the following Mk3 SSTO:

 

https://steamcommunity.com/sharedfiles/filedetails/?id=2072732104

 

And the wings kept falling off that one too.  I'd accelerate level to supersonic (to get the most thrust out of the Rapier engines), then even the slightest, and I mean slightest (using fine control and an Xbox controller) pull up to start a climb would cause the wings to rip off.  I tried a dozen struts, nothing would work.

I repositioned and reinstalled them several times, and one time it finally worked. I looked at the difference - I had forgotten to enable autostrut.  With autostrut on, the wing joints are as brittle as glass.  However, now on my new SSTO, the joint between the Mk3 cargo bay and Mk2 fuel fuselage is very loose, it jiggles around like crazy when I spawn in on the runway, and it lets go as soon as I pull up too.  I had it work once, it was very strong, but then every time after that it has ripped off. But this time at the fuselage, not the wings. Again, the most minimal of pullups causes it to detach, and even on spawn in when it bounces on the runway the joint jiggles all over the place.  I've even tried manually putting on several struts, and it still rips off. I'm starting to get very frustrated with that one.

The good news is, armed with my newfound knowledge of autostrut problems, I disabled them on my shuttle - and low and behold, it lands and the wings stay intact. In fact, it stayed intact on a hard landing that it definitely should have come apart. So one problem fixed, and my shuttle should now be good to go and usable. Now I need to figure out why my SSTO is falling apart...

 

TL;DR: turning off autostrut fixed the wings coming off my shuttle!

[Sandman1330]

OK, so I tried autostruts again on my SSTO... And it works.  It holds together (with some expected wing flex), no external strutting required.

Next question: The aircraft rolls left a few degrees any time I pitch up.  It's almost like one side flexes more than the other or something... I built absolutely everything using mirror symmetry, so I don't think that's it.  I have read about a stiffness bug, where each side gets a slightly different stiffness value, which would explain the problem I'm seeing.  Any way around it?