SSTO flips out of control once high up

[AHeroReborn]

Whenever I perform the last step of the ascent profile (high altitude), I flip out despite having a reaction wheel and pitch an yaw control surfaces.

[HvP]

Hello, @AHeroReborn.

It would be helpful to have a photo of your craft. If you can upload a screencap (press F1 in game) of your spaceplane to a site like Imgur.com then you can paste the URL for the photo into a forum post for us to have a look at. Screenshots are kept in a folder called "screenshots" in your KSP directory. A picture of your craft in the SpacePlane Hangar with center-of-mass, center-of-lift and center-of-thrust markers turned on would be great - preferably one with tanks full and with tanks empty.

How high up are you when you flip? If you are above most of the atmosphere then you won't get much help from your control surfaces since there's no air for them to work with. Does it happen after you ignite your space engines? If so, you probably have a misalignment between the direction of thrust of your space going engines and your center-of-mass. If they are not in line with with your mass then it will cause your craft to pivot around the center. This can be tricky because you have to have the thrust not only balanced when the jet engines are firing alone, but when the vacuum engines are firing by themselves, too. Depending on the thrust of your vacuum engines your reaction wheel might not be able to compensate.

The easiest way to check this is to load your craft in the SpacePlaneHangar turn on COM and COT markers, and turn the thrust all the way down to zero for your jet engines, but leave the vacuum engines at full thrust. Does the purple center-of-mass marker go up or down? Keep in mind that if your fuel tanks are also mounted above or below the center line of the craft, then your center-of-mass will rise or fall as you burn fuel. It can be very tricky to balance.

Edited May 19, 2017 by HvP

[steve_v]

4 hours ago, HvP said:

you probably have a misalignment between the direction of thrust of your space going engines and your center-of-mass.

Yeah, this ^.
Every instance of this behaviour I have seen is a misalignment of vacuum engines WRT COM. High wings or a large vertical stabiliser are the usual culprits.
Either tweak your engine position or brute-force it with a vernor on the nose. The latter can make even horribly misaligned piggy-back payloads flyable, provided you have the surplus fuel to feed it.
 

[AHeroReborn]

to clarify: it starts when i'm high in the second layer of the atmosphere/ early in the third.

JX-4 "whiplash" engines are still producing sufficient thrust for flight

 

[HvP]

29 minutes ago, AHeroReborn said:

it starts when i'm high in the second layer of the atmosphere/ early in the third.

Your engine placement looks good to me. Assuming those engines and wings are even with the center line of the craft then I'd doubt the gear and vertical fins are offsetting your mass enough to be the culprit.

My next guess is that it's the canards on the nose of your plane that are pulling you around. If you are pulling up to a high angle of attack at high speeds while still in atmosphere then the canards will induce too much drag at the front of your craft. Even the MK2 command pod will produce a significant amount of drag and lifting force, and when you couple that with the drag and lift from your canards it will act like a parachute in the wind if your angle strays too far from the prograde.

After weight reduction, minimizing drag is the single most effective way to get a spaceplane into orbit. If it were me, I'd take the forward canards off and instead add a very slight amount of rotation to your main wings to give them a small upward angle of attack. That should allow you enough lift to take off without the canards and help keep you from having to pitch your fuselage upwards as much to generate lift. Smaller gear in the back and larger gear in the front might be the next step if you still can't lift off without the canards. Otherwise, you'll have to just try maintaining a flatter ascent profile at such high speeds until you leave the atmosphere behind you.

If you are willing to try a couple of mods you might investigate "RCS Build Aid" which will show you if your engines are creating rotational torque, and "Correct CoL" which can show you how your lifting surfaces affect your lift during various angles of attack.

Edited May 20, 2017 by HvP

[AHeroReborn]

so, your advice is angling the wings (down if their behind the COM and up if in front) and getting rid of the canards? If it doesn't work,i may just experiment witht he stock craft or download a good one from somewhere in the forums.

[HvP]

1 hour ago, AHeroReborn said:

so, your advice is angling the wings (down if their behind the COM and up if in front) and getting rid of the canards? If it doesn't work,i may just experiment witht he stock craft or download a good one from somewhere in the forums.

Only angle the main delta wings upwards. Looking from the side, if the nose of your craft is facing left, then use the rotate gizmo, turn off angle snap, and rotate them only a few degrees CLOCKWISE. You want the leading edge of the main wings to be a little higher than the trailing edge of the wings. This will, of course, change your center of lift -- especially without the canards. So you'll have to slide your main wings forward again to bring your center of lift back towards the middle of your craft again (perhaps by a lot). The MK2 fuselages can generate a lot of lift so it might not be so bad to keep the COL marker a little further behind the COM in this case. It's more stable that way.

You may find that you no longer have as much pitch authority from the control surfaces on the trailing edge of your wings because they are now closer to the middle of your plane. Some small elevators at the back might be necessary, but it's hard to say right now. You'll have to experiment to find the right placement that will give you the correct balance between stability, maneuverability and lift.

What this does is it allows the main wings to generate upwards lift even when you are in level flight. That means you don't have to pull the nose up as hard, which in turn reduces your drag at the front. Again, I'm not 100% certain this will solve your flipping, but I am 100% certain that canards way at the front can cause flipping at high speeds at high angles of attack. Rotating the main wings is the best way I know of to get the lift you need without the canards.

Or, simply adjust your ascent to stay close to prograde for longer until you are above more of the atmosphere.

Edited May 20, 2017 by HvP

[HvP]

Update: I made a replica of your craft to test some things.

I definitely got it to fly better without the canards and having moved the delta wings forward slightly with a few degrees of upwards rotation on them. Also, I think your rear landing gear are too far back. Push them up into your fuselage a bit and slide them forward closer to your center of mass so that it can rock backwards a little when taking off.

Update to Update: I was getting a lot of side slipping so I added a wing strake to back of the fuselage and moved the two winglets backwards above the jet engines. Also, enabled the advanced tweakables to autostrut the two main wings so that they won't warp under stress.

Got it into 80km orbit with 500m/s delta-v to spare.

Edited May 20, 2017 by HvP

[bewing]

Two things to note: In ver 1.2.2, you can only have at most one air intake per jet engine. So the other two are doing you no good and adding drag. Replacing them with NCS adapters and small nose cones will reduce your drag.

Also, there is a bug right now where the cargo bay doors open/closed drag state is backwards. So if you open your cargo bay doors during your ascent, you will have less drag and may get to orbit more easily. Doing this will expose anything inside the bay to drag, though.

 

 

[AeroGav]

I'd say the OP simply has centre of mass a bit too close to the centre of lift, he needs to allow a greater margin of safety given that the centre of lift indicator of the stock game is not 100% accurate (doesn't calculate aero forces from parts that don't have a lift rating in their description).  Deleting the canard removes some lift from the front of the ship but you'd get the same overall effect just moving the wings back a little.

The jet fuel tanks are slightly in front of CoM so it's possible the plane's becoming unstable from the slight shift that occurs towards the end of the jet powered phase as most of the jet fuel is used.

If you're having issues making the craft sufficiently stable in some phases of flight, while not sufficiently controllable in others,   you're better off putting the cargo bay right over the CoM, instead of ahead of it as he has here.   Might run into issues with not taking off with a full cargo bay and flipping out on an empty one.

Also, the main fuselage is a mk2 but it's transitioning straight into a mk1 sized reaction wheel at the back, without an adapter.  Need a mk2 to mk1 adapter or you end up with a ton of drag.

 

[Ignacio Urteaga]

I also have severe stability problems with spaceplanes.

Rockets are so more easy to design and fly!

But,

As you asked about spaceplanes I would try to share my humble (and humiliating) experiences.

1. When you have not enough atmosphere for your control surfaces to bite then you lose control

2. In some designs the control surfaces revert the effect when you go faster than mach 1.

3. The relationship between your center of mass and center of thrust changes when you consume fuel and/or oxidizer making your vessel harder/impossible to control.

4. your non atmosphere engines are somewhat of center respect with the center of mass of your vessel.

 

My non elegant answer is to use vernor engines to stabilize my spaceplane once in space.

It is NOT elegant. You consume a lot of oxidizer and fuel.

I have also tried to pump oxidizer and/or liquid fuel from one tank to another in order to bring the center of mass where I need it. Some times it worked.

I use to adjust the height and inclination of wings. The height to align center of mass with center of thrust. The inclination to reinforce the stability of the plane.

As far as I can mention the main problem will appear when you try to re-enter in the atmosphere.

Best luck!

Ignacio 

 

[Spricigo]

@Ignacio Urteaga rockets or spaceplane, it's all about symmetry.  

When designing a plane there is several parts we place one side or another, messing with the symmetry.  Landing gears only underside,  vertical stabiliser only on top,  engines at the rear, wings at the sides.  The trick it's to restrain the effect of each of these to get a balanced result. 

[Wanderfound]

On 20 May 2017 at 4:14 PM, HvP said:

 

That's a good revision, but it's still massively oversupplied with intakes.

One ramscoop will feed two jets, one intercooler will feed three. There's enough there for fourteen engines.

Swapping out some intakes for tankage and nosecones would reduce drag and improve its range substantially.

It also wants a Mk2->1.25m adaptor on the back of the central fuselage.

Edited May 21, 2017 by Wanderfound

[HvP]

6 hours ago, Wanderfound said:

That's a good revision, but it's still massively oversupplied with intakes.

One ramscoop will feed two jets, one intercooler will feed three. There's enough there for fourteen engines.

Swapping out some intakes for tankage and nosecones would reduce drag and improve its range substantially.

It also wants a Mk2->1.25m adaptor on the back of the central fuselage.

All good points, but I was trying to specifically address @AHeroReborn's original problem about the plane flipping over without suggesting that he completely change the entire design. I guess it's just my style, but I like to take a systematic approach and work on one problem at a time.

The plane can get to orbit mostly as it is, but if he's looking for further improvements then what you and others have mentioned are certainly good options.

[AHeroReborn]

8 hours ago, Wanderfound said:

That's a good revision, but it's still massively oversupplied with intakes.

One ramscoop will feed two jets, one intercooler will feed three. There's enough there for fourteen engines.

Swapping out some intakes for tankage and nosecones would reduce drag and improve its range substantially.

It also wants a Mk2->1.25m adaptor on the back of the central fuselage.

aren't more intakes better for high altitude flight? They keep you're jet engines running.

[GDJ]

18 minutes ago, AHeroReborn said:

aren't more intakes better for high altitude flight? They keep you're jet engines running.

To a point, but if you go high enough there isn't enough air to keep the engines running, no matter the intakes.
Basically if you have the 100% requirements for resources that's all she wrote.

[HvP]

1 hour ago, AHeroReborn said:

aren't more intakes better for high altitude flight? They keep you're jet engines running.

It used to be common in older versions of KSP to get unrealistic altitude by spamming lots of air intakes. But they rebalanced the way intakes supply air to the engines after 1.0 so adding more than the minimum requirement is no longer helpful. Having one intake for every two air breathing engines is usually fine now.

Edited May 21, 2017 by HvP

[Wanderfound]

10 hours ago, AHeroReborn said:

aren't more intakes better for high altitude flight? They keep you're jet engines running.

In very early versions of KSP, yes, but that hasn't been the case for a few years now. No matter how many intakes you have, turbojets will die around 26km and RAPIERs around 30km. So long as you have enough intake to keep it at 100% air supply at that altitude, you have enough; any more is a waste.

For a two-jet ship, one good intake is all you need.