I hate spaceplanes

[paleorob]

So I've been playing since October and have yet to manage to get a spaceplane into orbit from a horizontal launch in one stage (SSTO). I can launch them on top of vertical launch rockets, but that kind of defeats the purpose. It doesn't matter if the design is my own homebrew or a copy of someone else's successful SSTO that they posted a photo of on the forum. It always ends one of two ways (or sometimes both): Not enough rocket fuel to circularize, although I am able to get out of the atmosphere AND/OR catastrophic loss of control from either an engine flame-out or shifting CoM causes the rocket to flip end-over-end.

Any remedy to solve one problem makes the other occur. Add more rocket fuel and BAM the formerly stable suborbital spaceplane now flips stupidly through the sky. Balance the fuel so there is no CoM shift (all the fuel lateral over the center of lift) BAM I can't circularize.

I'm sure someone will ask for photos of the craft I use. It doesn't matter. Pick any SSTO plane you have seen a picture of on the forum and I have probably tried to recreate it with one of the two results above occurring (sometimes in the same flight). My current homebrew flies happy as a clam all day long at ~900 m/s at 17km. Fire the aerospike and we get to do carwheels.

What sort of voodoo black magic do you all use to get SSTO aircraft TO? They breed nothing but hate and frustration in me. Anytime I see someone post one (especially orbiting somewhere other than Kerbin) I get this jealous/angry feeling inside that I know isn't healthy. AUGH! I need help, kerbonauts.

[ROFLCopter64bit]

I hate them too. The damn runway keeps loading my planes on an angle. That is the #1 killer of my space planes.

EDIT: Also, for the shifting CoM when you run out of air, use only one air breathing engine in the center (so when it flames out it cant shift CoM) and use two normal rocket engines of both sides of the first one.

EDIT 2: Assign action groups to the air breathing engies to toggle them, and a seperate one for the normal ones. When you get to about 25km high press 1 to toggle the air engines, then 2 to start the normal ones. Do 2, then 1, when re-entering.

Edited May 1, 2013 by ROFLCopter64bit

[paleorob]

I've got my engines set to action groups so that my aerospike isn't lit until we're about 20km up. Shortly thereafter I shut down the jets as intake air drops below .4 to avoid a flameout. My center-mounted single jet +2 lateral aerospike designs have not been successful either.

[bsalis]

I can launch them on top of vertical launch rockets, but that kind of defeats the purpose.

Far from it. Regardless of how it is launched, you can still land it back on the KSC runway. Also, you can carry a decent payload too.

It was good enough for NASA remember!

[Spatzimaus]

What sort of voodoo black magic do you all use to get SSTO aircraft TO?

Not to spoil too much, since the discovery is half the fun, but my own method involves a microwave oven, a cannibalistic guinea pig, a tube of super glue, and some Mountain Dew.

The flipping can be caused by two things: either you're getting too much drag from intakes (i.e., you're not closing them at the right times and/or you're not shutting off the turbojets explicitly), or your engines are misaligned. From what you've described, I'd guess it's the latter. When designing an SSTO, EACH set of engines (typically your turbojets and your rockets) needs to have its respective center-of-thrust line pass directly through the CoM, and the easiest way to do this is to make the plane extremely flat (to where everything critical lies in a single plane). And remember, the stock landing gear might say it weighs 0.5 tons in the SPH, but it actually weighs nothing, so take those off before balancing the mass. An ASAS or Avionics can help minor imbalances, but it won't overcome a major misalignment; your plane won't flip until you get to the upper atmosphere because the wings will keep it stable below that, but that doesn't mean the problem wasn't there from the start.

If you don't have enough fuel to get into orbit, you can add more, but remember that there's a practical limit to the weight of a spaceplane; it's not like normal rockets, where bigger is always better. Most standard SSTOs are around 10-15 tons, although a 2-engine plane can still get to orbit at up to 25-30 tons. (My own favorite spaceplane design is a 24-ton craft.) A 15-ton plane can make the final boost to orbit with a single 50kN engine, barely, but any heavier and you start needing something a bit more powerful. This is why the LV-N is a popular choice for the lighter planes, since it has unparalleled efficiency once you're outside the atmosphere. The problem is that if you go with a more powerful engine, your efficiency is so much lower that you'll waste a tremendous amount of fuel getting into orbit. You'll have to carry far more fuel to get the same amount of impulse, and that very quickly puts you into a downward spiral as you have to carry more fuel to feed the powerful engine you needed because of your heavier mass. In other words, the solution to your problem might be to go SMALLER; instead of just slapping more fuel on, try trimming your design down a bit. A smaller engine might still be sufficient to reach orbit, and the mass savings (including fuel) might be all you needed.

Or, if you insist on going big, do what I did on my 24-ton design and add a third engine set to the plane; my plane uses two turbojets to get up to speed, one LV-N for orbital maneuvers, and two 110kN radial engines for the brief transition from turbine flight to suborbital ballistics. Once you're out of the atmosphere, you have plenty of time to use the LV-N to circularize.

Also, velocity is critical; 900m/s just isn't nearly enough, you need to be going quite a bit faster than that on your jets. This is probably why you're running out of fuel; you should be using the rockets to make a small final transition, not relying on them for the bulk of the thrust. Ideally you want a design that can reach 1500m/s before you have to hit the rockets at all, so that you only really need to add ~500m/s to reach a circular orbit. Realistically, you can aim for 1200ish. That'll require flying well above that 17km altitude, but if you can't handle that then it probably means you don't have enough intakes on your design and/or your plane is just too heavy. You should be able to cruise above 20km while not losing much air if you've done it right.

[Flixxbeatz]

My Spaceplane flights always went like either one of these:

Load → SPONTANEOUS DISASSEMBLY

Load → Spacebar →RUNWAY EXPLOSION

Load → Spacebar → Runway roll → VEERS OFF RUNWAY

Load → Spacebar → Runway roll → Runway end → FAILS TO TAKE OFF

Load → Spacebar → Runway roll → Runway end → Take-off! → EARLY FLAMEOUT

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → NOT ENOUGH THRUST

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → Sufficient thrust → NOT ENOUGH FUEL TO ORBIT

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → Sufficient thrust → REACHED ORBIT! → NOT ENOUGH FUEL TO DE-ORBIT

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → Sufficient thrust → REACHED ORBIT! → De-orbit successful → STALL SPIN DURING RE-ENTRY

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → Sufficient thrust → REACHED ORBIT! → De-orbit successful → Good re-entry → MESSED LANDING

...and that, folks, is why I stopped building SSTO Spaceplanes.

Oh - and the nearest I've went was #9.

Edited May 1, 2013 by Flixxbeatz
Outcome lists expanded.

[ROFLCopter64bit]

My Spaceplane flights always went like either one of these:

Load → SPONTANEOUS DISASSEMBLY

Load → Spacebar →RUNWAY EXPLOSION

Load → Spacebar → Runway roll → VEERS OFF RUNWAY

Load → Spacebar → Runway roll → Runway end → FAILS TO TAKE OFF

Load → Spacebar → Runway roll → Runway end → Take-off! → EARLY FLAMEOUT

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → NOT ENOUGH THRUST

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → Sufficient thrust → NOT ENOUGH FUEL TO ORBIT

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → Sufficient thrust → REACHED ORBIT! → NOT ENOUGH FUEL TO DE-ORBIT

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → Sufficient thrust → REACHED ORBIT! → De-orbit successful → STALL SPIN DURING RE-ENTRY

Load → Spacebar → Runway roll → Runway end → Take-off! → Transition altitude → Sufficient thrust → REACHED ORBIT! → De-orbit successful → Good re-entry → MESSED LANDING

...and that, folks, is why I stopped building SSTO Spaceplanes.

Oh - and the nearest I've went was #9.

You can fix #9 and #10 with parachutes. Infact, all space planes I make come with an abort sequence with decouplers and parachutes.

Edited May 1, 2013 by ROFLCopter64bit

[earth]

I made a space shuttly SSTO thats quite easy to fly, but i installed FAR. Else, i could not end with a viable plane.

[Flixxbeatz]

You can fix #9 and #10 with parachutes. Infact, all space planes I make come with an abort sequence with decouplers and parachutes.

Those spaceplanes was K-Prize entries, so a typical landing is a must.

Oh - and edited the list further, might as well re-quote that

Edited May 1, 2013 by Flixxbeatz

[Spatzimaus]

Infact, all space planes I make come with an abort sequence with decouplers and parachutes.

I used to do that, but it sometimes caused other issues, like excessive SAS wobbling. And once you get the hang of it, it's really not tough to land a spaceplane the right way.

#7 is most likely caused by having too few vertical control surfaces, especially if you're trying for a purely unpowered (gliding) landing. What I do, instead, is turn my turbojets back on once I drop below about 15km, but I leave them at the lowest possible power for which I can still hear their sound. This little bit of thrust is just enough to keep the plane from tumbling or spinning, while not adding enough velocity to make landings difficult.

The usual reason I found for #8 is bad information. If you're not landing right next to the ocean, it's too hard to know how far above the terrain you are, and how quickly you're descending. Information mods, like Flight Engineer, make this one MUCH easier; I can regularly land on the sides of mountains now without any real problems. But of course, this depends on exactly what is meant by "Messed-up landing", since there are a lot of ways you can botch a landing.

[Flixxbeatz]

I used to do that, but it sometimes caused other issues, like excessive SAS wobbling. And once you get the hang of it, it's really not tough to land a spaceplane the right way.

#7 is most likely caused by having too few vertical control surfaces, especially if you're trying for a purely unpowered (gliding) landing. What I do, instead, is turn my turbojets back on once I drop below about 15km, but I leave them at the lowest possible power for which I can still hear their sound. This little bit of thrust is just enough to keep the plane from tumbling or spinning, while not adding enough velocity to make landings difficult.

The usual reason I found for #8 is bad information. If you're not landing right next to the ocean, it's too hard to know how far above the terrain you are, and how quickly you're descending. Information mods, like Flight Engineer, make this one MUCH easier; I can regularly land on the sides of mountains now without any real problems. But of course, this depends on exactly what is meant by "Messed-up landing", since there are a lot of ways you can botch a landing.

You mean #9 and #10 Also, I define messed-up landing as a ground landing so hard that the plane breaks apart, or that it encounters a sudden land elevation and explodes.

Might as well helpful if someone knows the fix from #1 up to #10, but yea, just like what the OP said, a fix to another problem could give you another problem.

[evilC]

The best advice I can give is to get mechjeb.

Test the viability of a plane on autopilot - that way you have reproducible results and can test out theories safe in the knowledge that each test run is vaguely similar.

See my thread here for a spaceplane plus instructions on orbiting it and re-entering without flipping.

I also have a new 100% stock one that can lift 70 tons SSTO and drop off the cargo space shuttle style, will be posting that soonish.

[ComradeGoat]

I'm trying the "give them ludicrous amounts of power with turbojets and shut all but the central one down as you run out of air" trick. When it works, it results in planes getting to stupid speeds at lowish altitudes. However, I have found it makes the plane very prone to uncontrolled pitching up, presumably from the high mass of fuel and engines at the rear.

Anyone got a good way to counter this?

[RoboRay]

Well, I was going to try to offer you better help, but since you're refusing to post pictures of some specific craft you've tried then I guess I can't. Nobody can give effective specific answers to general questions; not without a lot of luck in hitting the right answer, anyway.

But the usual problems are:

#1 It's too big and heavy.

#2 It doesn't have enough wing area to lift it (which is often related to #1).

#3 It doesn't retain positive stability throughout the flight (usually due to center of mass shifting or simply poorly positioned aero controls.)

#4 Thrust is not applied symmetrically to the center of mass.

Having no idea what kind of planes you're trying to build, I'd suggest staying very small. You should be able to make a single-seat SSTO spaceplane that weighs considerably less than 10 tons, using a single turbojet, two or three ram inlets, a jet fuel tank, a couple of smaller rocket fuel tanks, two to four of the little orange 24-77 radial rocket motors, and a modest set of wings that are positioned in a way that maintains positive stability.

[Flixxbeatz]

Case in point: This thing.

This was the spaceplane I was talking about that reached until #9 (refer to my post earlier below). It was my entry to the K-Prize before (and earned me a spot in the "Party Crashers" list). 6 airbreathers, 3 spikes. No SPH shots available.

Lifts off easily, but kinda hard to go punch through the atmosphere. Made it to orbit once, but #9 happened. Also, always in the brink of #7 (evident in first screenshot) and #8 (de-orbited using RCS once)

Edited May 1, 2013 by Flixxbeatz

[RoboRay]

Well, looking back at that list, if you're not getting into the air before going off the end of the runway, you're in the "too heavy" or "not enough wing area" categories. Possibly the "rear landing gear is too far aft of the center of mass" one, too, which I didn't mention above.

As to safely reaching orbit and tumbling out of control during reentry, most likely you've burned off enough fuel that your center of mass has shifted too close to or even past the center of lift. A contributing factor is almost certainly all those draggy jet intakes way up on the nose. Lots of drag on the nose? The craft naturally wants to turn around and point the other direction. This is why the arrows you shoot from a bow have the fins on the tail and not up at the front.

That thing is also, frankly, way too big and heavy for a one-seat plane. It's got more engines and more fuel than my cargo-hauler SSTO. If you cut the weight of a spaceplane in half, it's at least four times as efficient. Try smaller designs. Much smaller. They are far, far easier.

Edited May 1, 2013 by RoboRay

[Flixxbeatz]

I have tried small and compact designs before, but I always ended up experiencing #7 and (at one time) #8 and #9 (basically the same fate). That's why I have came up with that over-engineered monstrosity.

Nevertheless, I'm done with Spaceplanes. Call me a noobie, but I'm done. Might get back working at it when I'm bored or something.

[evilC]

its not about the jets, its about the intakes.

For larger craft, you will want 4:1 - I recommend a pod with 1x jet fuel, then a turbojet and 3x bi-couplers to split out to 4 for your ram intakes.

Mount these on the wing as near the CoG as possible, in pairs (top and bottom). To lift a jumbo-64 fuel tank use 12 of these pods (3 per wing, top and bottom)

Be aware that in 2x warp when in an atmosphere, forces are exaggerated. The wings will tend to flex up and inwards (front of wing pushing inwards) so strut accordingly.

A good tactic is to make a string of i-beams and then place them thru the center of the craft at the trailing edge of the wing, then attach control surfaces to that and strut the trailing edge of the wing to it - this should keep larger wings rigid.

Another piece of advice is to keep the wings and landing gear as separate as possible - you do not want them to influence each other.

[paleorob]

I knew people would want pictures. Here they are, but I'm not sure how helpful they'll be. Note: these are just a small sample of the types of spaceplanes I have tried.

This one apogeed out at around 100km. The rockets weren't enough to circularize. More rocket fuel killed it. Then I tried going small.

Flipped.

Many more that I didn't photograph in .18.X. In .19 this unconventional design is bar-none the best airplane I have built. Period. Unfortunately it appears to be a lousy SSTO but it is rock-solid at any altitude below the flame-out limit. 4:1 intake-engine ratio, avionics. Only hard part is taking off without smashing the lower wings.

Ends up backflipping once you are only on the aerospike.

You know what might be helpful? A graph of a flight profile for a SSTO.

[evilC]

You know what might be helpful? A graph of a flight profile for a SSTO.

I already linked the thread for my craft which contains an SSTO and full instructions on orbiting / landing it.

[Spatzimaus]

You know what might be helpful? A graph of a flight profile for a SSTO.

The profile will really depend on your specific plane's capabilities, but the general trend is this:

0> Before launch, close most of your intakes.

1> Climb at a 45-degree angle until you reach 12km altitude.

2> Pitch down to a ~10-degree angle, open all of the intakes, and start accelerating.

3> Watch your airspeed. At first you'll start gaining speed quickly, but eventually the acceleration will drop off to where you're only gaining a fraction of a m/s at a time. This means you're too low for your speed, so pitch up to gain some altitude.

4> At each altitude, level off to gain speed again. Go back to #3 until you're almost out of intake air.

(Once you're more experienced, you can combine 3 and 4 into a single smooth ascent, but the method I'm saying is better when you don't know your design's exact capabilities.)

5> Once your air is about to run out (~0.01 per intake), pitch back up to 45 degrees, turn on your rockets, turn off your turbojets, and close most (or even all) of your intakes again. This NEEDS to be done at a high altitude (>25km) to prevent your wings and intakes from destabilizing you.

6> You will be going to space today.

But this assumes you've done a few things with your designs, which it appears you didn't:

A> You need more intakes than engines (I have a 7:1 ratio on my own plane, and I'd consider 4:1 to be the practical minimum)

B> You need to have ASAS or Avionics on as much as possible, with enough control surfaces (both horizontal and vertical) to offset any roll/pitch/yaw you don't want

C> Most importantly, your intakes have to be far enough back that they don't flip you when their drag gets too high.

The intake ratio is the big one; with 5+ intakes per engine you can level off at certain altitudes and GAIN air while you accelerate, which removes a lot of the time pressure from your ascent. Note that those silver radial intakes count as 5 apiece (they store 1.0 air instead of the 0.2 of the others), so if you don't want to stick ridiculous numbers of ramjet intakes, you can just slap a couple radials on. It's possible to still do an SSTO ascent without this many intakes, but it makes things MUCH easier... as long as you remember to put the intakes far enough back. Ideally they'd all be lined up with the center of mass, so that they can't flip you at all, but they can be a bit forward as long as there are control surfaces at or in front of them. And your designs just don't seem to have enough flaps; for that first plane, two large flaps and two canards are just NOT enough horizontal control to prevent pitching if there's an imbalance somewhere else on the design.

[RoboRay]

Nevertheless, I'm done with Spaceplanes. Call me a noobie, but I'm done. Might get back working at it when I'm bored or something.

Nobody's calling you a noobie. Spaceplanes are hard. I attempted, failed, and gave up on spaceplanes several times before I finally figured out the very narrow band of tolerances that will result in a working design. There's nothing wrong with going back to having fun with vertical rockets until you're ready to give the planes another try.

[RoboRay]

this unconventional design is bar-none the best airplane I have built. Period. Unfortunately it appears to be a lousy SSTO but it is rock-solid at any altitude below the flame-out limit. 4:1 intake-engine ratio, avionics. Only hard part is taking off without smashing the lower wings.

Ends up backflipping once you are only on the aerospike.

That one is a pretty decent start. Here are some things I believe may allow it to perform better...

• Straighten the wings so they are flat to the ground.
  
• Add wing area, preferably toward the tail (a rectangular wing section on each side, shifting the triangular sections outward might work).
  
• Replace the vertical stabilizer with a smaller, less draggy one.
  
• Remove the forward canards unless you simply cannot control the craft without them (forward-mounted aero controls tend to be destabilizing).
  
• Add more engine intakes. A bicoupler or tricoupler on each side, where the current intakes are, is usually an easy way to fit more in.
  
• Replace the aerospike with an LV-T45 so you've got gimbal-control. (Alternatively, keep the aerospike and add an RCS system that you activate when entering the thin upper atmosphere.)
  
• It looks like you've got four tanks of jetfuel, which is excessive. Replace one of those tanks on each side with a nacelle or fuselage section.
  
• Adjust your landing gear positions so that the plane sits on the runway with a slight nose-upward attitude (this helps you get off the ground more easily).
  

Note that ram inlets are the only intakes worth using on spaceplanes... the others work much better than the rams down low, but only the rams collect useful amounts of air at high altitudes. Using the radial scoops, for instance, you'll find that they gather almost no air at 20km up (right click them in flight to see how they are doing in the current conditions, instead of just looking at their stats on the ground) but still produce vast amounts of drag.

Use the Avionics Package for your entire ascent, especially once you get into the thin air where your aero controls don't have much effect. It will really help keep you pointed in the right direction.

If you're still not getting into space, please return with pics of your changes.

Edited May 1, 2013 by RoboRay

[Torham234]

I love spaceplanes, and planes in general. I think I have spend more time building and flying planes than I did building and flying rockets in KSP.

Edit: And to help you out a bit:

- Start small. Smaller SSTO are much easier to make than big ones.

- Use robot probes for control to keep the weight down.

- use 1 centralized turbojet - 1 turbojet can get anything up to 18 - 20 ton spacecraft into orbit.

- Aerospike is very heavy. You can get away with 2x LV909, if your craft is light enough. You can mount the to the sides of the 1 turbojet for very balanced thrust.

- with 1 turbojet, you do not need dedicated jet fuel tanks. It can happily sip a little of the fuel from the main rocket tanks.

- you do not need ram intake spam. I can get a small SSTO to orbit with just 2. every intake is essentially a huge aero brake, the more you have, the bigger problem you have flying fast. Efficiency is the key.

After you get a small SSTO to orbit, you pick up a lot of the tricks and the bigger ones will be much easier to design/fly.

Happy flying.

Edited May 1, 2013 by Torham234

[SteevyT]

The profile will really depend on your specific plane's capabilities, but the general trend is this:

A> You need more intakes than engines (I have a 7:1 ratio on my own plane, and I'd consider 4:1 to be the practical minimum)

I'll work on it again in a week or two, but I have a plane that can just barely hit orbit with a 1:1 intake to engine ratio with very careful piloting.