The purpose of the space planes

[Jouni]

Anybody who assumes SP design will depend on rocket engines and staging has not appreciated the fundamental advantages of using an oxygen-rich atmosphere to reach space. If you flew a rocket-powered SP the way you fly a rocket, you'd find it's performance overall was lower because the wings would be dead-weight and subtract from max payload mass and TWR.

My basic point was, for a given payload mass, an optimised SP will always get it space for less fuel than an optimised rocket.

Have you tried jet-boosted rockets? Those things are essentially high-thrust spaceplanes without wings.

The optimal launch mass for a rocket powered by a single turbojet is about 10.9 tonnes. If you build a staged rocket, at least 8 tonnes of the launch mass can be payload. If you want to build a reusable SSTO, you'll need a bit more fuel in the rocket stage and a lightweight parachute, reducing the payload to around 7.5 tonnes.

Precision landing is also easier and cheaper using a SP because you can use aerodynamics to glide towards your preferred landing site. If both designs are entitled to use parachutes, the SP can pick it's landing site at a lower speed, and corrections can be made by gliding and circling, potentially using no fuel. Rockets, by comparison, can only make corrections by thrusting. Without parachutes, it's even more fuel efficient, albeit marginally less precise on one axis (as in rolling distance), to land a SP.

I don't know about you, but this is enough precision for me. You aim the reentry roughly at the right area, start the jet engines once you're low enough in the atmosphere, fly over the landing zone, deploy parachutes to kill horizontal speed, and use the jet engines to soften the landing. Due to the fuel usage bug, KSP turbojets are around 100x more fuel efficient than rocket engines, so you'll only need a few drops of fuel for this.

[Geschosskopf]

Just a note: In the stock game (unlike IRL), thrust isn't altitude dependent. A turbojets thrust at sea level is equal to it's thrust at any other altitude. ISP changes with altitude.

Actually, this is not true. In KSP, jet engines lose thrust with altitude.

Whichever update it was that introduced the RAPIER engine also introduced an auto-throttle thing so players wouldn't have to keep tweaking the throttle down with altitude to avoid flameout as long as possible. This automated system doesn't move the throttle indicator by the navball but under the hood, it's actually throttling back. The result of this is that a jet's thrust decreases as it starts running low on IntakeAir, until thrust reaches zero at flameout. Because the player isn't working the throttle but the thrust is decreasing because the game itself is moving the throttle, the effect to the player is as if the thrust decreases with altitude.

[Red Iron Crown]

Actually, this is not true. In KSP, jet engines lose thrust with altitude.

Whichever update it was that introduced the RAPIER engine also introduced an auto-throttle thing so players wouldn't have to keep tweaking the throttle down with altitude to avoid flameout as long as possible. This automated system doesn't move the throttle indicator by the navball but under the hood, it's actually throttling back. The result of this is that a jet's thrust decreases as it starts running low on IntakeAir, until thrust reaches zero at flameout. Because the player isn't working the throttle but the thrust is decreasing because the game itself is moving the throttle, the effect to the player is as if the thrust decreases with altitude.

They lose thrust with high speeds and air starvation, which is not quite the same thing as losing it with altitude (though it's closely related to air starvation). The defined thrust curve only respects speed, the altitude curve only affects Isp. It is entirely possible to have a turbojet delivering its full rated thrust at 30km altitude or higher.

[LethalDose]

Okay, I'm only addressing this question because it's tangentially related to the OP's question. I am NOT discussing this further in this thread.

FOR THE PURPOSE OF THIS POST ONLY I am using the term "SSTO" to describe a spaceplane that can go from the surface of Kerbin to LKO and back in one stage for brevity.

I didn't miss your point at all. I was just pointing out to the OP who is asking about spaceplanes to begin with. Perhaps he is not aware that it's possible to create an SSTO capable of flying from Kerbin to Lathe, landing, and returning back. All without staging. It also depends on what your definition of "efficient" is. If efficient means saving money (which is what it always means to me, now) then you're going to want to take a SSTO along anytime you plan on returning to Kerbin. Or some sort of reusable lander.

Okay, fine, then if you didn't miss my point, then you're trying to make the explanation way more complex than it needs to be. I hope trying to make yourself look smart is worth my wasted time cleaning up the mess you're causing.

Yes, it's possible. It's also a massive pain to engineer that kind of vessel, and it takes more mass, more fuel, more tech, and more effort to design. The player is clearly new at the game, or at least this concept. Why bring it up and confuse him or her. Instead of designing a vessel to go all the way from the surface of Kerbin to the surface of Laythe, and back, a much more efficient strategy would be to take the SSTO into LKO, then use a transfer stage to get the SSTO to Laythe, then ditch the transfer stage (or keep it use it as a tug, w/e). But now we have to have this stupid discuss because you wanted to nit pick and muddy the issue with Laythe.

So how is what you've described with your "Do-all-spaceplane" more efficient than using specialized stages!? THAT's my point!

And this discussion and explanation would have been completely unnecessary if you had just let the statement lie.

On my Duna missions I typically launch 2 SSTOs to my orbital station. 1 carries a small transfer stage, the other carries the Duna lander. The lander and transfer dock in orbit, go to Duna, lander does it's thing and rendezvous back with the transfer stage. Both go back to Kerbin, and the SSTOs can land them again. All reusable, but the smaller craft are definitely designed specifically for the mission.

If you were to read the wiki link I provided describing the original STS, you'll realize that your Duna mission description is exactly what I'm talking about: You use SSTO's to deliver cargo to LKO (in your case, one payload is a transfer stage, one payload is the lander).

That's what SSTO's are good for: delivering cargo to-and-from LKO on the cheap.

You can use them for the same thing on Laythe. If you know anything about Laythe, you don't need this explained...

Finally, the term "efficiency" in my previous post refers to cost, mass, and typical engineering difficulty. Having specialized vessels for each job achieves all of these criteria, and using the same vessel for everything is nearly as inefficient as you can get.

Edited December 15, 2014 by LethalDose

[LethalDose]

Actually, this is not true. In KSP, jet engines lose thrust with altitude.

Actually, what I said is true. For the exact reasons that RIC pointed out.

TJs and all the other air-breathing engines in KSP work just like all the others: ISP varies by altitude, thrust is constant. If you are give a TJ all the air it wants TJ at 20 km altitude, it produces exactly same thrust as it would at full thrust at sea level. And this is possible with air-intake spamming.

Engine thrust is independent of altitude.

@RIC: THanks, tried to give rep for that post, but apparently I like what you post too much.

[The_Rocketeer]

The optimal launch mass for a rocket powered by a single turbojet is about 10.9 tonnes.

This is not a rocket in the conventional sense - a turbojet is not a rocket engine. This is confusing the fundamental question - most players will assume a SP is at least partly powered by jets and a rocket is mainly powered by rockets engines. You're also using air-hogging, which frankly I frown upon and never features in my SP designs - if it did I daresay I'd considerably improve my efficiency.

Also, in your landing precision demo I have no idea what you were trying to land on/at. I can put a fully reusable SP on the last centreline stripe at the western end of the KSP runway with 0 fuel from reentry every time (or anywhere else, but I'm trying to specify somewhere you can find). I dare you to match it 5 times in 10.

Edited December 15, 2014 by The_Rocketeer

[Pecan]

This isn't the most detailed or in any way conclusive list, but from the top of my head:

1: Most importantly, variety. (Not everybody gets their kicks from building phallic monstrosities.)

2: SPs are more fuel efficient than rockets per payload mass.

3: SPs are safer and cheaper to land on bodies with atmospheres than rockets.

4: SPs are reusable - rockets tend to come apart on the way up/down as they drop stages.

5: SPs look cool

6: One SP design can efficiently meet a very wide variety of mission profiles, whereas rockets are usually one-trick ponies.

*Sigh*

1. Absolutely.

2. WRONG - Jets are more fuel-efficient than rocket engines.

3. WRONG - (Well, unless you're really bad at landing rockets). Parachutes are cheap and safe, if you need them!

4. WRONG - "A spaceplane" may well stage. A SSTO spaceplane is no more reusable than a SSTO rocket.

5. Absolutely.

6. WRONG - and I can't even imagine how someone could suggest that!

The good reasons to make spaceplanes are all about personal preference. The bad ones are about pretending they are 'better'.

That said, I agree with you that the 'personal preference' reasons ARE good ones!

[Red Iron Crown]

I'm not entirely convinced the payload fraction is better on a jet-powered spaceplane compared to a jet-powered "rocket". Essentially, a plane trades engine mass for wing mass, it's probably pretty close to a wash.

I will say that a jet plane is easier to land precisely than a jet "rocket", the slow throttle response of the jet engine is more of a penalty when doing a powered landing.

In before "rockets are jets" pedantry.

[The_Rocketeer]

*Sigh*

1. Absolutely.

2. WRONG - Jets are more fuel-efficient than rocket engines.

3. WRONG - (Well, unless you're really bad at landing rockets). Parachutes are cheap and safe, if you need them!

4. WRONG - "A spaceplane" may well stage. A SSTO spaceplane is no more reusable than a SSTO rocket.

5. Absolutely.

6. WRONG - and I can't even imagine how someone could suggest that!

The good reasons to make spaceplanes are all about personal preference. The bad ones are about pretending they are 'better'.

Wow. Pecan, if you disagree maybe you could illustrate how exactly you think I'm wrong.

2. A SP of equivalent mass can reach orbit on less fuel than a rocket. Prove me wrong.

3. A SP can glide from orbit to land literally anywhere without using any fuel at all, without using parachutes. A rocket can't.

4. An SSTO rocket is MUCH LESS EFFICIENT than an SSTO SP. Lets stick to comparing specimen a (a rocket) with specimen b (a SP) shall we?

6. Really? Why? Show me a single complete rocket design that can compete with a single complete SP design (of comparable delta-V) for versatility of mission profile.

This isn't a question about which is better, it's a question about why people would bother making SPs. I also couched my answer by saying it wasn't particularly detailed. If you like, I will specify that when I say SP I mean 'craft that launches horizontally, has wings and uses jet engines for primary thrust' and by rocket I mean 'craft that launches vertically and uses rocket engines for primary thrust'.

As for whether a SP will be better, that COMPLETELY depends on the mission. There's no reason to suggest anyone's 'pretending' they're better - for a lot of things, they literally ARE better.

Edited December 15, 2014 by The_Rocketeer
too argumentative - emotive language removed

[peachoftree]

Jet engines: Jets in KSP have silly, huge, ludicrous ISP ratings.

I think the word you're looking for is plaid.

[sal_vager]

Guys, this isn't worth arguing over.

[Jouni]

This is not a rocket in the conventional sense - a turbojet is not a rocket engine. This is confusing the fundamental question - most players will assume a SP is at least partly powered by jets and a rocket is mainly powered by rockets engines.

Every beginner should get rid of such false assumptions as soon as possible. To a large extent, KSP is a game about engineering. To be a good engineer, you should think like an engineer. It doesn't matter how things are conventionally done. It doesn't matter what the parts are intended for. The only thing that matters is what the parts can be used for.

In KSP, a jet engine is a ridiculously efficient rocket engine that takes the oxidizer from the atmosphere, while a rocket engine is a ridiculously inefficient jet engine that takes the intake air from the ship itself. A spaceplane is a rocket with wings and weak engines, while a rocket is a spaceplane without wings and with powerful engines. A space station is a lander that remains in orbit, while a lander is a space station that can descend to the surface and return back to orbit.

An engineer should always prioritize processes over concepts, as concepts only confuse the people who don't understand the processes.

Almost all the advantages conventional spaceplanes have over conventional rockets come from reusability and from using jet engines instead of rocket engines. The plane in itself is just a harder way to build a reusable jet-powered SSTO.

You're also using air-hogging, which frankly I frown upon and never features in my SP designs - if it did I daresay I'd considerably improve my efficiency.

Airhogging is the only way to go, when we're talking about efficiency without qualifiers. Personally I don't use the jet engines for any real missions, because the fuel usage bug that makes them use 16x less fuel than they should is far more ridiculous than any amount of airhogging.

Also, in your landing precision demo I have no idea what you were trying to land on/at. I can put a fully reusable SP on the last centreline stripe at the western end of the KSP runway with 0 fuel from reentry every time (or anywhere else, but I'm trying to specify somewhere you can find). I dare you to match it 5 times in 10.

I wasn't probably aiming at any particular spot. (That picture is from March, so I don't really remember that much about the flight.) My point was that with a bit of practice, it's easy to land within a short walking distance of any particular target (or within the full recovery zone).

[Pecan]

Wow. Pecan, if you disagree maybe you could illustrate how exactly you think I'm wrong instead of just slamming my answer. Apart from anything else, it's pretty offensive to have any potential misunderstandings on my part torn apart.

2. A SP of equivalent mass can reach orbit on less fuel than a rocket. Prove me wrong.

3. A SP can glide from orbit to land literally anywhere without using any fuel at all, without using parachutes. A rocket can't.

4. An SSTO rocket is MUCH LESS EFFICIENT than an SSTO SP. If you're going to shout me down, at least do so consistently. If you want to change example per example, this is going to get tedious.

6. Really? Why? Show me a single complete rocket design that can compete with a single complete SP design for versatility of mission profile.

This isn't a question about which is better, it's a question about why people would bother making SPs. One reason is, in some respects, they are simply a better design for the job.

@ALL - I think Rocketeer's second post clarifies a lot which I had not read when I posted above, and I have given him rep for it. However ...

2. *Challenge accepted* - what payload or total vehicle mass do you require? You build the spaceplane, I'll beat it (probably using jets). Where we're going we don't need wings and landing-gear!

3. You said safer and cheaper. While you could make a 'tail-sitter' spaceplane or use parachutes your wings are then just useless mass and have only cost you more to have and to haul. For horizontal-landing you are restricting yourself to a clear, longish and flatish area. How 'safe' would it be to land a spaceplane in a Mun canyon?

4. It is more efficient to SSTO with jets than rockets. That has nothing to do with reusability. In any case, even with rocket engines, the fuel cost is insignificant. How much fuel does your 100t-payload to orbit spaceplane use?

6. I refer you to the Jool-5 challenge thread.

I must agree that spaceplanes are the best way to land stuff at KSC. How much needs to land though? Returning crews, science, empty fuel-tanks (maybe).

[The_Rocketeer]

Every beginner should get rid of such false assumptions as soon as possible. To a large extent, KSP is a game about engineering. To be a good engineer, you should think like an engineer. It doesn't matter how things are conventionally done. It doesn't matter what the parts are intended for. The only thing that matters is what the parts can be used for...

I completely disagree with this post, not for the factual detail contained but because personally I think it saps all the fun from the game. I derive no pleasure from engineering craft that demonstrate the absolute limits of the game's mathematical formulae, I get a real kick out of chucking rockets and spaceplanes around and blowing them up when they fail or let me down somehow. I think this is true for the majority of players - the real challenge is to succeed in the face of uncertainty. I accept that there is a hardcore group that want to find the absolute limits for a particular set of rules, but I don't think that grabs the casual gamer and almost certainly isn't what the OP was asking for.

2. *Challenge accepted* - what payload or total vehicle mass do you require? You build the spaceplane, I'll beat it (probably using jets). Where we're going we don't need wings and landing-gear!

3. You said safer and cheaper. While you could make a 'tail-sitter' spaceplane or use parachutes your wings are then just useless mass and have only cost you more to have and to haul. For horizontal-landing you are restricting yourself to a clear, longish and flatish area. How 'safe' would it be to land a spaceplane in a Mun canyon?

4. It is more efficient to SSTO with jets than rockets. That has nothing to do with reusability. In any case, even with rocket engines, the fuel cost is insignificant. How much fuel does your 100t-payload to orbit spaceplane use?

6. I refer you to the Jool-5 challenge thread.

I was fairly specific about the need for an atmosphere for most of these SP advantages to become relevant - that pretty much accounts for the Jool 5. Atmospheric SP descents are safer because they're slower and don't depend on thrust or parachutes, and cheaper because they require no extra parts or fuel. Short-roll landings aren't that hard for someone with my (virtual) piloting experience even on relatively steep slopes, and parachutes or VTOL remain an option if absolute precision is required - you'll still need less than a relaunch-capable rocket of equivalent dV. I am prepared to be proven wrong, but I am convinced that SSTO with a lift-generating craft is more efficient than SSTO with a thrust-only craft. This is based largely on experience of ion SPs with TWR of < 1.0, which simply wouldn't fly as rockets.

Perhaps we should simply say that atmospheric missions will gain a number of advantages using a SP, but bodies without atmospheres are typically better off with a rocket.

In any case, we are getting off topic by debating which is a superior mission platform. This is just a question of 'why bother'. We can add the assumption (correct or otherwise) that "a SP will be better for this mission" as such a reason.

Edited December 16, 2014 by The_Rocketeer

[Jouni]

I completely disagree with this post, not for the factual detail contained but because personally I think it saps all the fun from the game. I derive no pleasure from engineering craft that demonstrate the absolute limits of the game's mathematical formulae,

That's not what engineering is all about. Engineering is a creative task, where you solve problems, think outside the box, and discover new ways of doing things. Of course you have to know how things work in the game, if you want to think outside the box, but learning stuff you want to learn is usually fun.

I built most of my spaceplanes in 0.23. At some point, I realized that planes with less wings flew almost as well as planes with more wings, while behaving better when the center of mass shifted during the flight. I took that to the logical extreme and removed all wings, and the craft still flew quite well.

The reason why jet-boosted rockets / wingless spaceplanes work so well is tied to one of the pecularities of the game. In stock aerodynamics, lift is proportional to air density, while drag is proportional to the square of the density. As a result, wings help you to take off and fly in the lowest 10-15 km with less thrust. Once the spaceplane starts accelerating, it must climb higher, where lift becomes negligible compared to drag, and the plane starts behaving more and more like a rocket.

[Lighthawk]

To add another use for (space)planes, which granted doesn't apply to the OP, but anyway...I find them useful for completing those 'test part x at altitude y and speed z' contracts. It's much easier to hit the height and speed requirements in a jet than a rocket.

[The_Rocketeer]

Sorry Jouni, but without wings that generate net lift, the vehicle is not a plane. A missile would probably be most accurate.

I know what engineering is - my dad worked for Rolls Royce testing helicopter engines most of his life. But I disagree that KSP is at heart an engineering game. That implies that you start with an objective (specification) and design your craft to achieve it. I think that all newcomers, and many long-time players, tend or prefer to build a rocket that looks probable and then see what it can do. For these gamers, including myself 90% of the time, the engineering process is simply a case of refining what you consider probable, and not a case of mathematical engineering a vehicle that I know is capable on paper. The stock game encourages this style of play by deliberately limiting the amount of information you can harvest from it at any given stage of the construction or simulation process. It's only because mods have become available that a truly engineering approach can be viably made.

Your second paragraph - realising that a plane performs as well with smaller wings - is an example of the engineering process I've just described. However, I think you have an unusual concept of what constitutes things like 'planes' and 'flight'. There is a massive difference between aerodynamic flight and the motion of a ballistic object under propulsion. In game terms, it's easiest to describe this in terms of whether the craft has a "direction of lift".

And yes, wings are most beneficial at low altitude - precisely where rocket propulsion is least efficient. Also, the total drag force on wings is far lower than the lift force they generate, so this tends to increase the altitude at which lift gains are outweighed by drag losses, and basically means you need less dV overall.

Anyway, as I've said before, I think we're straying from topic. Let's stick to the discussion of reasons to use SPs. If you want to start a new thread to debate what constitutes a SP or rocket, I'd be happy to weigh in.

Edited December 16, 2014 by The_Rocketeer

[Jouni]

And yes, wings are most beneficial at low altitude - precisely where rocket propulsion is least efficient. Also, the total drag force on wings is far lower than the lift force they generate, so this tends to increase the altitude at which lift gains are outweighed by drag losses, and basically means you need less dV overall.

It's not a matter of lift vs. drag. It's about planes behaving like rockets at high speeds, requiring a lot of engine power as a result.

Let's assume that a plane powered by turbojets is capable of level flight at a certain altitude at 500 m/s. A turbojet generates 75% of the nominal thrust at both 500 m/s and 1500 m/s. However, the drag is 9x higher at 1500 m/s, so the plane has to fly 11 km higher, where the air density is 9x less. Because the plane is flying only 3x faster, the lift is 1/3 of what is needed for level flight (or actually a bit more due to weaker gravity and the curvature of the planet). For the rest, the plane has to use engines.

In practice, the plane needs to have TWR > 1 at that point. This means one turbojet for every 15 tonnes or so, making engine mass around 8% of the total. (This corresponds quite well with the common spaceplane design guidelines.) As a comparison, a typical jet-boosted rocket might have one turbojet for every 9 tonnes (a bit more power than the optimal, as optimal jet-boosted rockets crawl the first kilometers very slowly), making 13.3% of the total mass engines. A spaceplane saves 53 kg of engine mass out of every tonne, but it needs to use some of that for wings. Common guidelines aim for around 1 unit of lift for every tonne, which uses 50 kg out of that 53 kg. Essentially, spaceplanes just trade 1:1 engine mass to wing mass.

Anyway, as I've said before, I think we're straying from topic. Let's stick to the discussion of reasons to use SPs. If you want to start a new thread to debate what constitutes a SP or rocket, I'd be happy to weigh in.

This is still a discussion of reasons to use spaceplanes. Or, more accurately, about why many of the reasons other people have given are not reasons for using spaceplanes, but reasons for using any kind of reusable jet-powered craft.

[Ziff]

Okay, fine, then if you didn't miss my point, then you're trying to make the explanation way more complex than it needs to be. I hope trying to make yourself look smart is worth my wasted time cleaning up the mess you're causing.

Take a chill pill dude. Go relax before you have an aneurysm. I'm sorry I missed the post that said the OP only wanted the most simple of explanations and that you were the keeper of the law around here.

[Specialist290]

Alright, everyone, let's keep the personal attacks and the passive-aggressiveness to a minimum. Like sal_vager said on the last page, a simple question like this isn't something to get upset over.

EDIT: After further consultation, we've decided to close this thread. People have provided perfectly good answers to OP's questions, and most of the rest of the discussion past that point seems to be mostly arguments that have gone off on a completely different tangent than the topic.

Edited December 16, 2014 by Specialist290