Finally! Secrets of SSTO spaceplanes, unlocked!

[ComradeGoat]

Then you are doing something wrong in the build or fly.

My plane lifts 30 tons with a 4:1 ratio and the thrust hits 50% at 30km, at which point I turn on the aerospikes.

Aerospikes (I note plural) have considerably more thrust than a single NERVA, which is rather gutless.

[ComradeGoat]

Part of my learning curve has been to discover that the drag matters, a *lot*. If you have more drag on the bottom or the top, it will pitch uncontrolably in hypersonic flight. Similarly, if your centre of thrust is too far behind the centre of mass, the plane will try to pitch forward under power, and it will do it more the faster you get.

[Ryu Gemini]

My first and laughable mistake with spaceplanes...

It was three crashes before I decided to look at a stock one and go "(facepalm) Oh, I need to actually add the landing gear! Duh!" Interesting to note that without such gear, they just sort of skid along a bit before veering off to the right.

[ComradeGoat]

I accidentally a Swiss Army plane! Better than my previous attempt, it'll do full throttle at 30K without the strut intake mounting trick and doesn't need a "kick" from fuel thirsty high thrust rockets to get into orbit: it'll do it on a Nerva and the 2 jets only. Enough Delta V to reach Duna and beyond, can land on the Mün and return, and unlike my previous attempt, isn't an utter dog to try and get off the runway (previously had to put the wheels further back than I'd have liked to avoid smacking the NERVA into the ground on rotation. Have solved by pulling the NERVA forward into a central cavity I left for it).

[SkyRender]

What I'm learning is that, unsurprisingly, you do in fact have to account for all vectors (not just the four cardinals, but also up and down). Center of lift needs to not only be slightly behind center of mass, but also level with center of mass. Also, it's good to not angle it too far upwards. Though that makes for easier initial lift, it makes for much harder high-atmosphere maneuvering.

[ComradeGoat]

What I'm learning is that, unsurprisingly, you do in fact have to account for all vectors (not just the four cardinals, but also up and down). Center of lift needs to not only be slightly behind center of mass, but also level with center of mass. Also, it's good to not angle it too far upwards. Though that makes for easier initial lift, it makes for much harder high-atmosphere maneuvering.

There seems to be quite a trick to designing one which will handle nicely at low altitude and hypersonic flight. Part of the fun I'm having is working out why stuff that works when I do it intuitively works in theory, so I can improve the next design. The poster who said this is *hard* is right. It seems to be far harder than anything else in the game; it's almost a full game in its own right, but I am having *so* much fun!

[CalculusWarrior]

Every so often I am amazed at the spaceplane designs that are on this forum, I go and try and make a SSTO myself, and fail miserably. I give it a few more tries, then when I don't get anywhere, I go back to rockets. A little while later, the cycle begins anew...

Is there a comprehensive design criteria post for spaceplane anywhere? I know about the centre of lift and mass, and I have a rough idea of how to launch one. How many wings should I add for the weight of the craft? Should I aim to have the lift rating balance out the mass, or should I add more/less? Also, I've created planes that fly, that get up to ~1600 m/s at about 30km, but when I light the rockets no matter what I do, the apoapsis stays quite low, and I lose speed until I'm doing 1.2km/s. Then I start gaining speed again, but by that time I've exhausted my fuel reserves, and make a landing on that large continent to the east of KSC. (I've dubbed it the Spaceplane Failure Peninsula)

[RoboRay]

Oh, you should see the countless wrecks and abandoned craft of mine dotting that peninsula.

There aren't any comprehensive guides to building SSTO spaceplanes, I don't believe. Just general rules...

The lighter the better. Doubling the weight makes it at least four times harder to design and fly. You don't add parts (often even fuel) to improve performance... you remove them.

Use multiple ram inlets for each jet engine, and ignore the other air intakes... those are for airplanes, not spaceplanes.

Put your aero control surfaces as far back as you can. Flight controls near the middle of your craft do nothing useful (just roll, and most KSP planes have more roll authority than I want anyway due to pod torque). Don't put controls at the nose unless there's no other way to make the thing controllable. Front canards make it easier to handle but also much easier to tumble out of control.

In my experience, a single jet engine per 10-12 tons of craft mass is sufficient.

Throttle back as you reach high altitude and begin to run out of IntakeAir. You can still accelerate with the throttle reduced, and your engines need less air to keep working. The faster you go, the more air the rams gather, so the higher you can go, which allows you to go even faster as drag decreases.

Ensure your craft has sufficient controls (aero, RCS, whatever) to get the nose pointed upwards at around 45 degrees even in the thin air at 25-30km. Once you start the rockets, you need to climb up out of the goo as fast as you can, until you can level out again around 50km+ to continue accelerating. Once you shut down the jet engines, forget about flying a plane... fly the same kind of profile you would for a conventional rocket at that altitude and speed.

Edited April 24, 2013 by RoboRay

[SkyRender]

Okay, significant discovery that not a lot of spaceplane designers even think about: the overall center of mass of your craft is essential, but so too is the distribution of that mass. You don't tend to have to worry so much with rockets about that (at least until you're in space, but it's less pronounced there), but with a spaceplane, a craft that is back-heavy or front-heavy will fly horribly no matter how you muck about with center of mass/lift.

I'd say we need a fourth indicator: center of structure. Center of lift appears to need to be the same distance from center of mass as center of mass is from center of structure. The closer all three are to equal, the better your craft will fly. My attempt to add more speed to my spaceplane-in-progress by adding more engines to the sides backfired horribly when the center of mass was effectively shoved much further back. The end result: even with center of lift reasonably behind center of mass, the plane still pitched upwards ridiculously once it got into the air after a bit.

[RoboRay]

I don't know if I would agree with that. I've built longer craft where the CoM and CoL were both very near the tail and were perfectly stable and handled nicely.

Like I said above, I think a fundamental problem with your design is that it's short and wide, and you seem to be trying to fix it by making it wider rather than longer. While that kind of design is workable, it's much more challenging to make stable than a more conventional design that's longer than it is wide.

I really suggest setting that design aside and learning with a smaller, simpler craft, then returning to perfect that design when you've mastered the underlying principles.

Edited April 24, 2013 by RoboRay

[ComradeGoat]

Every so often I am amazed at the spaceplane designs that are on this forum, I go and try and make a SSTO myself, and fail miserably. I give it a few more tries, then when I don't get anywhere, I go back to rockets. A little while later, the cycle begins anew...

Is there a comprehensive design criteria post for spaceplane anywhere? I know about the centre of lift and mass, and I have a rough idea of how to launch one. How many wings should I add for the weight of the craft? Should I aim to have the lift rating balance out the mass, or should I add more/less? Also, I've created planes that fly, that get up to ~1600 m/s at about 30km, but when I light the rockets no matter what I do, the apoapsis stays quite low, and I lose speed until I'm doing 1.2km/s. Then I start gaining speed again, but by that time I've exhausted my fuel reserves, and make a landing on that large continent to the east of KSC. (I've dubbed it the Spaceplane Failure Peninsula)

I started by building SSTO VTO craft in the VAB. Taught me a lot without throwing the whole "and now you're building planes" thing at me too. Probably a shallower learning curve.

[AmpsterMan]

Something that I find helps is a "less is more" approach, sort of like building a rocket in the VAB.

That is ofcourse, unless you go for the "MOAR! is more" approach; this is a perfectly legit way to design craft as well

Edit: Here is mine that I JUST got into space and back with

Edited April 24, 2013 by AmpsterMan

[CalculusWarrior]

That looks awesome, AmpsterMan!

Thanks everyone for the tips, I hope I'll be able to show off a spaceplane on the forums one day.

[SalmonellaDingDong]

Good luck! There are plenty in the screenshot thread if you need inspiration.

[Van Disaster]

Caveat: I am a FAR user. This is true of designing atmo planes too, but: sort your CoL positioning out before you add any control surfaces other than ailerons ( and the rudder, I guess ), that way they should have the same authority in both directions. Canards can be especially deceptive, if you're not careful you'll balance everything with them acting as wings and you'll have rather reduced ability to pitch up. Don't worry if the SPH shows the CoL forward of CoM once you add them.

TAC fuel balancer mod is really useful for larger craft. You can rarely go wrong by adding more wing area. I notice someone's having issues transitioning to space; it sounds like you don't have enough rockets, simply. One nuke per jet is my rule of thumb unless I'm taking two jets just to be symmetrical. Nukes are reasonably efficient in the upper atmosphere so you can use them as boosters with drop tanks rather than add more jets.

[CalculusWarrior]

I guess that would help with interplanetary transfers as well, thanks!

[smunisto]

I noticed you are all linking and discussing quite light SSTOs.

I spent the last week trying to get a heavy SSTO Cargo craft in LKO, so far highest I have reached with the varying changes has been 55km Apo, unable to circularize due to the lack of fuel.

Being a perfectionist, i want it to be aesthetically pleasing, as well as capable of reaching LKO.

The craft itself is around 115t, 110 without the payload.

I will post some pics of the most recent version later today.

Has anyone been able to figure out a reasonable ratio of TurboJets per ton?

[numerobis]

Typically I max out at 20t per turbojet, which is around 0.5 TWR on liftoff, and around 1 TWR at 1000 m/s. Then you need lots of wings to produce lift, of course.

[Van Disaster]

Yeah, my 75t/25t payload plane has 6 of each engine type ( note - FAR turboramjets ) - 20t is probably quite optimistic for those, they're a bit scaled down from the defaults. My 180t+ tanker has 12 of each .

[Spatzimaus]

I don't know if I would agree with that. I've built longer craft where the CoM and CoL were both very near the tail and were perfectly stable and handled nicely.

Correct. It's not a function of where your CoM is relative to the center of the structure, it's about where your control surfaces are relative to your center of mass. I mean, here's my SSTO, which I've posted in other threads:

The center of mass is directly between the visible RCS jets, just in front of the vertical tail assembly. (It's actually slightly in front of those RCS jets, but as fuel starts burning off it moves slightly aft.) Clearly, most of the structure is well in front of that point, and yet it's completely stable, because I've got control surfaces far forward and far aft of the CoM. The forward control surfaces are the two canards embedded in the front edge of the wing, and they are a big part of why the design's stable.

Basically, control surfaces exert a torque around the center of mass, and their distance from the CoM makes all the difference. Canards at the front of a vessel make your ship much more maneuverable than if you'd put them barely in front of the CoM, and the same goes for flaps at the back. The further from the CoM your flaps are, the more pitch/roll/yaw you'll get from them. On my design, you'll notice the little winglets out at the ends of the wings; they're designed to be directly in line with the CoM so that using them WON'T generate pitch; they're entirely for roll movements. Now, put too many control surfaces too far from the CoM and your ship will feel unstable, but tune it right and you'll have a design that handles well in-atmosphere at any speed. My own plane is actually too maneuverable, but that's because it was designed to keep Avionics on at almost all times for better precision.

Center of Lift isn't enough, because it's dominated by pure lift surfaces that can't be steered. Conversely, a "Center of Flaps" statistic would be meaningless, because a flap at the back doesn't cancel out one at the front, they add to each other. What we really need are straight roll/pitch/yaw totals for the vessel, which don't need any graphical representation to be useful.

[Toxicterror]

i fail at ssto's and that makes me upset

always need to activate infinite fuel to get to LKO

[barrenwaste]

Ok, guys, I'm noticing some things here. First, most of these designs are having trouble because of similar problems. You don't need a ton of wing and fuel to make it to space. In the case of SSTOs, at least light ones, less is moar! One turbojet engine with an atomic engine or aerospike clipped in will get you into space easily on half the fuel from an FL-t400 fuel tank, and that's including the weight from rcs and a clampotron. Each part you add also adds drag, which means more fuel used to achieve orbit. Keep your part counts as low as possible, including wings. With the above build all you need are a couple of structural wings, a delta deluxe winglet, and a few small control surfaces. A lot of the really smooth looking ssto designs funtion entirely because they are smooth, with few added parts and only as much fuel as is truly needed. You can always refuel in orbit for longer trips.

[Vanamonde]

Since this thread was created, SAS got its major overhaul, we've got new hybrid engines, and other things have changed, rendering the info in this thread no longer applicable. Thread locked so as not to keep outdated info floating around.