Finally! Secrets of SSTO spaceplanes, unlocked!

[ComradeGoat]

Every so often, while exploring the solar system, I would notice the funny little building with its funny little runway, and decide to have a play.

This would invariably result in something that accelerated along the runway like a demented thing, didn't actually take off, and careered into the sea as it turned into a fireball.

"Spaceplanes will never work for me", I thought, and returned to rockets, impressed by all those people with their amazing SSTO designs.

Then I stranded a couple of Kerbals on Laythe after botching my insertion, and feeling the need to go and rescue them, started to try and build an air breathing lander to descend to Laythe's surface and return to space, all in one stage.

But how to test such a thing? The only realistic way was to develop it on Kerbin, which has a larger gravity well. If it works there, it will work on Laythe.

My initial design just about worked, but my ascent profile was awful. I switched off the jets at a mere 600 m/s and clawed my way to orbit on a pair of aerospikes, arriving there with almost no fuel left, with most attempts not even doing that and falling back to Kerbin.

I got a few tips here, and my final design is something I'm actually rather proud of. It takes off vertically, gets to around 1,200 m/s before firing up the rockets, and arrives in orbit with plenty of fuel to rendezvous and dock.

During development of this though, I learned a lot abut how to get an SST vehicle into orbit, and slowly figured out the variables I had to balance to get the thing to behave through hypersonic flight into the high atmosphere, and then transition to the orbital insertion rocket burn.

I played around with one or two vertical take off SSTO designs, managing one that was able to get out beyond Dres' orbit without refuelling, all in the VAB.

And then it occurred to me that I'd probably climbed most of the learning curve I needed to climb to build an actual spaceplane.

So I'm pleased to report that, after returning to the SPH, I have now built, from scratch (not modifying a stock design), an SSTO, piloted spaceplane that can ascend to orbit, dock, reenter and land. My first practical design is nicely balanced and behaves beautifully in hypersonic flight, unlike my first attempts where touching the controls between 18,000 and 35,000m would typically be followed almost immediately by an abort to surface, if the vehicle actually stayed together!

No-longer am I terrified of the dark arts practiced in the SPH; things seem to have "clicked", and I have a feeling that my standard way to get kerbals into space will now involve taking off from a runway.

It feels like a whole new area of the game has opened up! Is it just me though, or is building a working SSTO spaceplane just about the hardest learning curve in the game? Docking was a piece of cake compared to this.

[FritzPixel]

I felt the same way about SSTOs. It can be a huge challenge! I've found that it's less about what you fly and more about how you fly it. You should post some pictures of your first successful design!

[Czman007]

IMO spaceplanes are definitely the hardest part of the game. I usually don't bother with the runway at all because rockets are so much easier. Still haven't been able to make an SSTO. You should post some pics of your plane so everyone can see it.

[ComradeGoat]

I felt the same way about SSTOs. It can be a huge challenge! I've found that it's less about what you fly and more about how you fly it. You should post some pictures of your first successful design!

Gladly! This is the first one I few to orbit. It's a prototype and doesn't have the docking capability, or much of anything. It's now evolving into a practical design with a more fuel-efficient ascent (intake air, intake air and MOAR intake air!) and all the comforts of home, like properly balanced RCS, a docking port, lights, etc.

[SalmonellaDingDong]

Check out FAR (Ferram Aerospace Research) if you're interested in more realistic air modeling and flight physics. It makes nosecones aerodynamic and such. I installed it and now my planes fly like they should IRL.

http://forum.kerbalspaceprogram.com/showthread.php/20451-0-19-Ferram-Aerospace-Research-v0-9-1-Aerodynamics-Fixes-For-Planes-Rockets

[ROFLCopter64bit]

I keep seeing ram intakes on SSTOs. Why is this?

[reptile2001]

Congrats! SSTOs are something I had been working on since I first started playing. Landing on the Mun, rendezvous and docking were all milestones in my Space Program, but nothing compared to the first time I successfully launched an SSTO.

To ROFLCopter - the Ram Air Intakes are necessary on an SSTO because it allows the jet engines to operate at much higher altitudes, allowing the craft to pick up much more speed before the more fuel hungry rockets kick in.

[reptile2001]

Also, I agree with what FritzPixel said - your launch profile can make a MASSIVE difference in the performance of any SSTO. Practice practice practice!

[SalmonellaDingDong]

How else are you going to get air?

[ROFLCopter64bit]

...I don't know. Shut up.

[jostanman]

Here is the real secret

[RawChicken]

I keep seeing ram intakes on SSTOs. Why is this?

In case this is no trolling attempt, here;

A ram-air intake is any intake design which uses the dynamic air pressure created by vehicle motion to increase the static air pressure inside of the intake manifold on an engine, thus allowing a greater massflow through the engine and hence increasing engine power. -Wikipedia.

There are 3 types of air intakes in ksp, ram air intakes would allow you to fly at higher altitudes with jets as they can gather more air than the other air intakes. This is crucial for a high altitude flight, which is what gives you enough speed to exit the atmo when rockets start boosting.

[ComradeGoat]

As someone said above, a lot of getting it to work is about how you fly them. If I explain my ascent profile, the ram intakes may become clear:

Take off from runway with just the ram intakes open (scoops closed due to drag), climb at 45° and terminal velocity (ish) to around 14,000. Reduce angle of attack to 30°. Climb to around 20,000. Level off.

Accelerate hard up to about 1,000 m/s in straight and level flight. Open air scoops, start to climb at 50-100 m/s rate of climb.

At around 25,000 metres and something like 1,200 m/s, the air starts to give up (people better at this than me can get this higher). Throttle back to about 50%.

It's usually about now that I kick in the aerospike. The plane is still climbing, but under reduced power and the acceleration has slowed. The aerospike, together with the jets at lower power, keeps the acceleration and ascent going. Keep reducing power to prevent flameout

Somewhere around 30,000 metres, it starts to get tricky. There is almost no air left. The jets are sipping fuel, as is the aerospike. Using them much more risks a flameout. At this point the plane is traveling at about 1,500 m/s. if we wait any longer then we will catch up to the apoapsis and start to descend. This is bad.

Kill the jets, close all air intakes. Full throttle immediately. The aerospike roars to life. Check map view. The apoapsis should be in front of you and running away. If it looks like you're going to catch it up, increase your angle of attack. BE CAREFUL. If your plane isn't well balanced, changing the angle of attack here could cause it to spin out of control. My earlier attempts were prone to this and needed RCS assistance at this stage. Now I'm building them somewhat better!

Watch the navball change from "Surface" to "Orbit". With the plane still below 40,000 metres, the apoapsis will shoot away and a periapsis will appear. Let the apoapsis climb to around 100,000 metres and cut the power.

Atmospheric friction will now start to bring the apoapsis back down, but we have built up enough speed and have enough vertical momentum to carry us clear of the atmosphere before it drops below 70k. Glide to orbit, circularise, feel smug.

That's why you need the air intakes. The jets formidable ISP and power means they can push the plane through the atmosphere and attain speeds that would require very heavy rockets to match. If you have to switch to rockets at altitudes where the air is still producing a lot of drag, you will burn all your fuel trying to keep the apoapsis in front of you, and the fall back to Kerbin. It's all about going as high as possible and as fast as possible on jets alone, so that the fuel-thirsty rocket has to do as little work as possible.

[Haemogoblin]

Many thanks Comrade

I've not braved spaceplanes yet but I've a feeling that info you just gave will be invaluable

[rosenkranz]

So what does a good spaceplane assent profile look like?

[UNSC]

I got halfway through the OP when I realized I was reading it in poem

[ComradeGoat]

I posted a pic of the prototype. Here's my practical spaceplane, in orbit round the Mün, after refuelling:

[RoboRay]

So what does a good spaceplane assent profile look like?

1. Go up fast.

2. Level out and go faster.

3. Go up again.

But seriously, the specific altitudes and speed vary with your craft design. Try to build something that can climb quickly to 15-20km, then accelerate to 1200-1500m/sec while reaching 25-30km before firing the rockets and resuming the steep climb.

You can design craft to reach higher speeds and altitudes on jets, but I value my time more than the fuel that saves.

[ComradeGoat]

OK, really getting the hang of this now. Just finished the final tweaks for this, the Minerva-DSV (Deep Space Vehicle) SSTO. Takes off from a runway, operates to 28,000m on jets alone under full thrust, enough delta-V to go from the runway at the KSP to Duna orbit without refuelling. Bit of a princess below 8,000 metres, but handles beautifully in hypersonic flight and has pretty decently balanced RCS for something that's such an awkward shape.

Very pleased with this:

Edited April 23, 2013 by ComradeGoat
Typo fixed

[SalmonellaDingDong]

Just built this in stock after a fresh install:

[ComradeGoat]

Wow! How are you mounting all those intakes on the leading edge? I seem to have to stick them on fuel tanks.

[SalmonellaDingDong]

Attach the smallest struts to the wing and rotate them so they face straight, then add intakes.

A 10:1 intake to jet ratio gives the best performance. Full throttle at 30k altitude feels great.

[ComradeGoat]

Attach the smallest struts to the wing and rotate them so they face straight, then add intakes.

A 10:1 intake to jet ratio gives the best performance. Full throttle at 30k altitude feels great.

Ah! Cunning. Thank you. Best I have so far is 6:1 (the one pictured above), which gets me to 28K. Not *quite* enough to get to orbit on just a NERVA afterwards

[SalmonellaDingDong]

I can usually make a 400km apoapsis on jets alone, saving a lot of rocket fuel.

[evilC]

I can usually make a 400km apoapsis on jets alone, saving a lot of rocket fuel.

You mean 40km.

To ascent well with jets, use mechjeb throttle control and enable manage intakes / prevent flameout.

A 4:1 ram intake:turbojet ratio is plenty, any more is abuse of mechanics IMHO.

@ takeoff, pitch is 40

@ 10km, go to 20

@ 20km, go to 10

@ 30km, turn on rockets, but leave jets on.

Using this, I can lift 30 tons to orbit - see my ship here: http://forum.kerbalspaceprogram.com/showthread.php/25772-SSTO-Spaceplane-Fuel-Lifter-%28Almost-stock-part-included%29

Other design tips:

With heavy planes, try and make the undercarriage nothing to do (Not connected) with the wings - this will cause less problems with it tipping over on the runway etc - see linked ship above for a great undercarriage technique.

If your plane flips end over end (say nose goes up) then often a good solution is to put more canards on the front - this will lock it down more.

If you have problems with spins on reentry - stay pro grade until under 700m/s and/or add more canards to front.

Also drogue chutes on wing tips will kill a spin.