RO and spaceplanes

[xrayfishx]

First of all I'm going to make it clear for those who didn't notice that I'm using Realism Overhaul. So the appropriate mods are to be assumed

Currently I'm trying to create a shuttle-type spaceplane and no matter what I do, it flips over. My newest attempt is a lawn dart in the lower atmosphere, but during re-entry it still behaves like it had its wings attached to the nose or something. I didn't even know it's possible that a plane is unstable during re-entry, even though, should it survive that re-entry, it becomes so stable that I'm still able to recover and land it.

So, here are some pics. What is wrong in them?

Click for bigger versions.

Those things attached to the vertical stabilizer are airbrakes. Also, on the wings, the control surfaces further from the middle are set up as flaps with the intent that they would also act as airbrakes and would help stabilize the spaceplane. However, this didn't seem to help much.

Edited April 14, 2015 by xrayfishx

[Guest]

You need some way to move the center of mass to the forward part of the ship during reentry. If it's too far back it becomes uncontrollable. I generally try to put a tank of some kind right behind the cockpit that I can transfer fuel to or engineer the plane so that once it's empty the COM is way forward.

[Alshain]

It's absolutely possible for a plane's center of mass change and destabilize the plane. What is in the cargo bay? Do you remove it when you get to orbit? If so your Center of Mass will jump backward with how it is placed.

[capi3101]

Yeah - I'm with Alshain on this one. What's in the cargo bay, and what does the CoM/DCoM look like without it?

Also, since you've got RO you've got FAR; what strength level do you have your wing surfaces tuned to? A full-strength tail adds a hell of a lot of mass; the DCoM could easily creep upwards on you due to the wair of the fin. It has gotten me more than once.

[Deutherius]

In addition to checking the CoL/CoM, could you try running a FAR simulation at different altitudes and velocities? That might tell you why it's flipping out at hypersonic speeds.

[xrayfishx]

In my cargo bay I have my fuel tanks and optionally a couple tons of satellite which will be removed before re-entry. Really, the whole thing is designed so that neither cargo nor fuel would move the CoM. I enabled RCS balancer in those screenshots just to demonstrate that; you can see how the DCoM is slightly ahead of CoM.

I will experiment with some sort of system to move the CoM forwards. What's the densest transferable fuel in RO? Also, how to do it without making the plane several tons heavier?

[viktor19]

I have had a similar problem when I was constructing a shuttle that looks very similar to this one and with problems similar to this one. Luckily, I found inspiration on the real shuttle, and came to the conclusion that Flaps are your friends on reentry. Make big flaps that bend some 40 degrees and tie them to action groups. Flaps pull down your nose so you won't go out of control easily if your tail is heavy. I can send you the .craft file if you need it

[xrayfishx]

I have had a similar problem when I was constructing a shuttle that looks very similar to this one and with problems similar to this one. Luckily, I found inspiration on the real shuttle, and came to the conclusion that Flaps are your friends on reentry. Make big flaps that bend some 40 degrees and tie them to action groups. Flaps pull down your nose so you won't go out of control easily if your tail is heavy. I can send you the .craft file if you need it

I added flaps and they didn't change much, as I said in the OP. If you could send me that craft file, it'd be much appreciated. I'm getting closer to making my design fly, but it comes at a cost of carrying 3t of dead weight. I really just need to fix yaw stability to get my design working. And I have ideas about that.

[Alshain]

Well I'll be honest, aside from your satellite (which DCoM won't account for) I don't see a reason why it wouldn't re-enter. The good thing about shuttles is they can be overly stable because they don't go up as a plane. But your CoL is so far back, it should be stable... unless your dropping the satellite.

EDIT: I just realized your air brake is really high. Leverage-wise, that could be pulling it around (it would go nose up if that was it). Mount it closer to the center of mass on the vertical stabilizer.

Edited April 15, 2015 by Alshain

[viktor19]

I added flaps and they didn't change much, as I said in the OP. If you could send me that craft file, it'd be much appreciated. I'm getting closer to making my design fly, but it comes at a cost of carrying 3t of dead weight. I really just need to fix yaw stability to get my design working. And I have ideas about that.

I'll send you now. BTW nice sig

[Wanderfound]

Well I'll be honest, aside from your satellite (which DCoM won't account for) I don't see a reason why it wouldn't re-enter. The good thing about shuttles is they can be overly stable because they don't go up as a plane. But your CoL is so far back, it should be stable... unless your dropping the satellite.

EDIT: I just realized your air brake is really high. Leverage-wise, that could be pulling it around (it would go nose up if that was it). Mount it closer to the center of mass on the vertical stabilizer.

My vote's on the airbrake as well. Bulk torque from that when it's deployed.

Try a run with no airbrake at all; you don't really need them anyway. Just keep the nose 5-10 degrees above surface prograde until you level out, then gradually drop lower until the drag slows you down.

[xrayfishx]

EDIT: I just realized your air brake is really high. Leverage-wise, that could be pulling it around (it would go nose up if that was it). Mount it closer to the center of mass on the vertical stabilizer.

The air brake is positioned like that because I want it to hold the AoA quite high, like the Shuttle. It is supposed to pull the nose up some, but keep it down should it go too high. It isn't responsible for flipping it. I know because earlier versions didn't have an airbrake.

One thing I thought of myself is that maybe at a high AoA the CoL changes because the lifting body effect becomes stronger relative to wing lift. In that case I would need to have most of my cross-section behind CoM. What I mean is that maybe at 40 degrees angle wings don't produce much lift and most of what's affecting the plane is drag. So really I need to be worried about my center of drag, which, I believe, could easily be in front of the CoM in my design. I'll keep experimenting.

Edit: Now that I think about it,.that could explain the yaw instability as well.

How could I move the center of drag backwards? Moving CoM forwards more than near the end of the cargo bay is not feasible, because carrying around 5t of dead weight isn't how I like to play.

Edited April 16, 2015 by xrayfishx

[Wanderfound]

The Shuttle used its RCS to maintain stability during the extreme AoA early reentry; if you're determined to keep the high AoA reentry (which generally isn't necessary in KSP, even with DRE), half a dozen strategically placed Vernors would replicate the Shuttle RCS.

[funk]

Delta wings and their secrets... I've built a lot of them with FAR but not using RO. Besides things already mentioned, here's my attempt to explain whats happening. I'm not natively speaking english, so pls be lenient with me, my non-existing drawing skills and my technical expressions...

When flying transsonic or supersonic it's recommended to use arrowed wings, but there are some special facts about them:

Air flows like this at arrowed wings:

Javascript is disabled. View full album

As you can see, this could decrease vertical control when the tail is in the middle.

For increase of control capabilities use multi-delta-wings and a bigger or two tails (I prefer All-moving-wings from B9 proc. wings) in the middle and/or like a winglet. In reality thickness of the wing and edge shape would effect airflow too, but I think FAR doesn't recognize it - changing edge and thickness has no impact to the stability numbers.

Javascript is disabled. View full album

Now airflow is closer to the fuselage. Pay attention to the lenght of the leading edge. It is greater which leads us to hypersonic flight:

At arrowed wings and high AoA lift is created by air vortexes at the upper side of the wing. Thus even higher AoA up to 25-30 can be handled easily and increases lift in comparison to the linear method. This effect is known as vortex lift. I hope this link can help: http://soliton.ae.gatech.edu/labs/windtunl/classes/unstaero/vortflo1/vortflo1.html

Don't hesitate to search for other stuff, it's really interesting, because the behaviour of the vortexes has to be modeled numerically for special usecases, eg. edges of scramjet inlets.

Now long story short the vortexes are mainly related to the length and shape of the leading edge and the shape of the wing. Until the vortexes break down (at very high AoA) they are symmetrical and have the following pressure distribution:

So for your instabilities:

1. Increase leading edge length - change wing shape - you can fly at higher AoA.

2. Use vertical control where the highest pressure occure.

This is a simple example of your craft and FAR shows nice numbers (don't pay attention to Xw it's unimportant):

Javascript is disabled. View full album

I hope, you're able to build a stable shuttle now.

Edited April 16, 2015 by funk

[xrayfishx]

*Insert funk's explanation here*

Now that is the answer I was really hoping for. Thank you so much! I won't call it [Answered] just yet, but hopefully I'm close now.

Edited April 16, 2015 by xrayfishx

[NathanKell]

Body lift indeed does not respond the same way to AoA as wing lift does. As your AoA increases, the percent of lift you get from your body compared to the your wings will increase, thus shifting CoP forward. This is why, for example, the SR-71 could not exceed something like 8 degrees AoA or it would become unstable ("flip out" in KSPeze).

[Tetragon213]

Have you tried shifting fuel around?

[funk]

Body lift indeed does not respond the same way to AoA as wing lift does. As your AoA increases, the percent of lift you get from your body compared to the your wings will increase, thus shifting CoP forward. This is why, for example, the SR-71 could not exceed something like 8 degrees AoA or it would become unstable ("flip out" in KSPeze).

Indeed, especially because SR-71 has a long sharp edged body, which should create lift, compared to relative short wings at the back. It's the same vortex lift mechanism for the body. If I remember correctly they were very surprised during flight simulations that critical AoA is 14 subsonic and 8 degrees while in supersonic flight. But the transition between stable and unstable vortex lift isn't at a fix AoA. Eg. maneuvers like rolling can create non symmetric lift, which causes yaw torque at the body. But that doesn't mean, that the vortexes breakdown completly.

As I said above the shape of the leading edge has a great influence to vortexes. Here is an example craft I´m working atm:

Javascript is disabled. View full album

It's the first version of a ZEHST (

) alike crew lifter HOTOL, in KSP SSTO of course. Pictures are from the first test flight. I forgot air breaks, but it's rockstable during ascent and reentry, even at high AoA.

Look at the wing shape and leading edge of the wings, from thick root to thin ends, sharp leading edge and a falciform overall shape of the wing.

Can anyone confirm, that FAR doesn't recognize thickness and leading edge shape? Or is it B9 proc. wings, because at least in the VAB the colliders of leading edge aren't updating.

Perhaps one of you might know, if there is a mod which contains ramjets? I'm also searching for high thrust turbofans like the ones of Tupolew Tu-22M which have 245 kN IR.

Edited April 17, 2015 by funk

[NathanKell]

FAR indeed does not recognize wing thickness or LE shaping, just 2D wing shape.

You want AJE for engines.