Can someone explain each air intake and when it is best used?

[openatheclose]

I've been trying to build SSTO's for a long time, but I just can't get it. The only tip I can ever seem to find is "make sure center of lift is behind center of mass!" but I still can't get it. I know air intakes are important, but with so many different kinds and no explanations of them, it's difficult to understand how to engineer my ships. Can anyone explain each air intake and its pros and cons?

[Anquietas314]

I think this thread should cover your needs as far as intakes go, although bear in mind the structural ones are great for aesthetics and as the description suggests the idea is to put a lot of them on in a small area.

As for building an SSTO, this thread might be helpful to you. You might be better starting off with something a bit smaller though. A plane made of Mk2 parts isn't necessarily hard to design, but you can make much simpler ones using the Mk1 parts, especially since you can make them fly pretty easily with just the one jet engine (Mk2 ones tend to be a bit on the heavy side for that).

[Pecan]

Okto, fl-t200, 48-7s = ssto.

[UmbralRaptor]

If you can fit them, ram intakes or shock cones. (As best I'm aware, the extra intake area of the shock cone usually offsets the disproportionately greater mass) If you're adverse to the design choices (eg: cubic octagonal struts) necessary for large numbers of them, radial intakes. (Worse area, but easier to mount, and the mass penalty is mostly gone) Avoid the structural intakes (terrible area) and nacelles.

Okto, fl-t200, 48-7s = ssto.

This is an important point, though you'll get significantly more payload if you add an FL-T100. And with a bit of playing, you can get useful SSTOs out of most of the engines. (eg: LV-T30 with 1600-2400 L of LFO, aerospike with 1600 L, Skipper with 6400 L...)

[numerobis]

As best I'm aware, the extra intake area of the shock cone usually offsets the disproportionately greater mass

Nope; the shock cone has 20% more area than the ram intake but 2.5x the mass. It does, however, have the most area per part of any intake, so it's your best choice if you are limited by part count.

Avoid the structural intakes (terrible area) and nacelles.

The structural intake has not much area, but that makes it the best intake for a low-altitude subsonic jet: by being low-area, it has low drag.

The nacelle is terrible because it's heavy, which is bad on its own, but in KSP it also means it has high drag. At 0.3t versus 0.01t for most of the other intakes, it suffers from drag *thirty times* as bad as the others.

[UmbralRaptor]

Nope; the shock cone has 20% more area than the ram intake but 2.5x the mass. It does, however, have the most area per part of any intake, so it's your best choice if you are limited by part count.

I would expect so, but given that both represent a tiny part of the craft's overall mass (barring excessive intake spam), the advantage seems minor.

The structural intake has not much area, but that makes it the best intake for a low-altitude subsonic jet: by being low-area, it has low drag.

Point taken. At what speeds does it max out its Cd?

[numerobis]

I would expect so, but given that both represent a tiny part of the craft's overall mass (barring excessive intake spam), the advantage seems minor.

It would be minor except for the role of mass in the drag calculations. Multiply that tiny mass by 11.5 to compare its drag at high speed to that of most rocket parts (or by 23 to compare to the Mk2 parts): a shock cone produces slightly more drag than an empty long Mk2 fuel tank, whereas a ram intake produces slightly less drag than an empty *short* Mk2 fuel tank.

The radial scoop maxes out its drag at 600 m/s; at that speed, the structural intake has half the drag coefficient. If the radial scoop produces more air than you need (i.e. you're sticking to low altitude), then you will see a slight benefit from swapping it out (you also save 2 kg).

[Foxster]

This is something that still confuses me. Wonder if someone could summarise when to use each one? Like, use one type for low altitude and another for low mass, etc.

[GoSlash27]

Foxster,

For stock turbojet or Rapier powered SSTOs, this is the simplest, most straightforward advice I can give you:

Use 4 XM-G50 intakes per engine minimum and put them as far back on the plane as possible. You can use more if you wish and it'll keep your jet lit at higher altitudes, but it gets spammy if you use too many. 4 is adequate if you fly it right.

If you can't find room for 4 XM-G50 intakes, use 2 XM-G50 intakes and a single shock cone intake per engine.

For low altitude farting around, the structural intake will serve the need.

As for when to use the other types... don't. *

Best,

-Slashy

*Yeah, I know that isn't 100% true... but this is highly simplified advice and it'll keep him out of trouble.

Edited January 3, 2015 by GoSlash27

[UmbralRaptor]

This is something that still confuses me. Wonder if someone could summarise[sic] when to use each one? Like, use one type for low altitude and another for low mass, etc.

If the craft primarily flies low and slow (most likely using basic jets): structural (but not very many)

If the craft primarily flies high and fast (long suborbital hops or goes into orbit): ram intakes, with radials if there are issues using lots of them.

[Dispatcher]

GoSlash27: I'd love to see a comparison of all the air intake parts; similar to the neat comparison you did with wings/ control surfaces, lift, drag, etc. IIRC. Is that something that you'd be willing to do?

[GoSlash27]

GoSlash27: I'd love to see a comparison of all the air intake parts; similar to the neat comparison you did with wings/ control surfaces, lift, drag, etc. IIRC. Is that something that you'd be willing to do?

I wouldn't mind at all, but honestly Numerobis is better qualified for the job IMO.

Best,

-Slashy

[NeatCrown]

I saw someone do this a month or so ago. I'll edit this post if I find of again.

[Dispatcher]

@ GoSlash27 & NeatCrown: it would be cool if Numerobis wouldn't mind doing that.

[numerobis]

Sorry, I was working on another little project -- figuring out how to maximize fuel efficiency given a particular plane.

I doubt I'll have time for this for a while. The holidays are over. But maybe someone else can grab the KSP-scripts and generate tables of whatever is interesting.

[GoSlash27]

Alright. I'll knock something together...

Best,

-Slashy

[GoSlash27]

Okay,

Results are in, and they're kinda surprising.

So here's what I did:

I assumed enough surface area to adequately keep a turbojet lit at 43km altitude and 2.3 km/ sec velocity, where terminal velocity and aircraft velocity both exceed orbital velocity at 70km altitude.

I assumed an aircraft built to the standard rules of thumb; 13 tonnes of aircraft per engine and 1.0 lift coefficient per tonne of aircraft.

Finally, I assumed that the intakes themselves were granular. I computed the number of intakes required, the mass, the drag coefficient when running at high altitude and 2.3 km/ sec, and the cost.

Here's what I found:

Stock KSP .90 Intake Comparo for SSTO Turbojets

[TABLE=width: 500]

[TR]

[TD]Type[/TD]

[TD]Number per engine[/TD]

[TD]Drag coefficient[/TD]

[TD]Mass[/TD]

[TD]Cost[/TD]

[/TR]

[TR]

[TD]Mk1 Fus intake[/TD]

[TD]2.67[/TD]

[TD]7.06[/TD]

[TD].320[/TD]

[TD]$1,922[/TD]

[/TR]

[TR]

[TD]Radial[/TD]

[TD]3.20[/TD]

[TD]10.6[/TD]

[TD].480

[/TD]

[TD]$5,280[/TD]

[/TR]

[TR]

[TD]Nacelle[/TD]

[TD]3.20[/TD]

[TD]10.6[/TD]

[TD].480[/TD]

[TD]$4,960[/TD]

[/TR]

[TR]

[TD]Ram air intake[/TD]

[TD]1.60[/TD]

[TD].354[/TD]

[TD].0160[/TD]

[TD]$4,288[/TD]

[/TR]

[TR]

[TD]Circular intake[/TD]

[TD]2.00[/TD]

[TD].442[/TD]

[TD].0200[/TD]

[TD]$1,360[/TD]

[/TR]

[TR]

[TD]XM-G50[/TD]

[TD]2.67[/TD]

[TD].589[/TD]

[TD].0267[/TD]

[TD]$2,670[/TD]

[/TR]

[TR]

[TD]Structural[/TD]

[TD]6.40[/TD]

[TD]1.13[/TD]

[TD].0512[/TD]

[TD]$5,760[/TD]

[/TR]

[TR]

[TD]Shock Cone[/TD]

[TD]1.33[/TD]

[TD].736[/TD]

[TD].0333[/TD]

[TD]$4,057[/TD]

[/TR]

[/TABLE]

The single most important consideration is drag coefficient. All of these intakes will allow the engine to make the same amount of thrust, but their drag will impact the top speed of the aircraft. Clearly, the fuselage intake, radial, and nacelle are unsuitable for spaceplane SSTOs.

Looking at the rest of them...

The ram air intake is the best choice in terms of sheer performance. On the downside, it's fairly expensive and it's clunky shape can make it a bit of a bear to plug into multiengine designs. Overall, an excellent choice.

The circular intake is an outstanding all-around choice. Low drag, low mass, and excellent price. The only drawback is that you have to fit 2 per engine. This is a problem with all stackable intakes, but at least this one has a low-profile shape. Overall, I was shocked to find out how good this intake actually is.

XM-G50: Unremarkable stats. Average in all respects. On the plus side, it attaches radially, allowing it to be placed nearly anywhere you can find room. This is a huge bonus for high-speed stability, as placing the intakes in the back is very helpful. I have had good success with these, and definitely recommend them.

The structural intake is a problem child. Poor performance at an outrageous price. You can make an SSTO spaceplane using these, but nearly any other choice will give you better performance. On the plus side, it does attach radially (see above) and it's low profile design can make for a more compact and pleasant design. Overall, I do not recommend these.

I was surprised to discover how good the shock cone intake isn't. Average stats, medium-high price. It doesn't quite have the surface area to feed an SSTO spaceplane on it's own, so it really holds no practical advantages over the other 2 stackable intakes. Overall, I do not recommend these.

Using the stackable intakes together in concert with XM-G50 radial intakes will provide a neat, compact design in multiengine spaceplanes. The Ram Air or shock cone intakes will feed an engine fine when assisted by a single XM-G50. The circular will need 2 XM-G50s to assist it.

For a single engine design, 3 XM-G50s will do the job, and so will 2 circular intakes.

Best,

-Slashy

Edited January 4, 2015 by GoSlash27

[GoSlash27]

Testing out some of the intake combos on a common test rig.

3 XM-G50 radial intakes. $3,000 and .03t mass. intake area .018.

1 circular intake and 2 XM-G50s. $2,680 and .03t mass. .020 intake area.

2 circular intakes. $1,360 and .02t mass. .016 intake area

1 Ram air intake and 1 XM-G50. $3,680 and .02t mass. .016 intake area.

Best,

-Slashy

Edited January 4, 2015 by GoSlash27

[Anquietas314]

I assumed enough surface area to adequately keep a turbojet lit at 43km altitude and 2.3 km/ sec velocity, where terminal velocity and aircraft velocity both exceed orbital velocity at 70km altitude.

43km seems a bit excessive, especially if you're assuming full throttle; I'm sure I don't need to tell you you can very easily keep accelerating at that altitude with 10% throttle on very badly starved engines. If that's the case, 1.33 shock intakes per engine at 43km is pretty reasonable - that should be roughly 1 intake per engine at more like 35-38km.

Otherwise, good work

[GoSlash27]

43km seems a bit excessive, especially if you're assuming full throttle; I'm sure I don't need to tell you you can very easily keep accelerating at that altitude with 10% throttle on very badly starved engines. If that's the case, 1.33 shock intakes per engine at 43km is pretty reasonable - that should be roughly 1 intake per engine at more like 35-38km.

Otherwise, good work

No, that's assuming throttled back and topped out, so the shock cone just isn't competitive right now.

The single shock cone could be feasible if you're running a lighter SSTO design. Say, 8 or 9 tonnes of spaceplane per engine. I'll have to test that to be sure.

Personally, I think the shock cone needs nerfed to .016 intake area to be competitive with the other offerings.

*edit* I just ran the single shock cone on the test rig. It topped out at 2,413 at 46k. Definitely not enough to haul a full-size SSTO to orbital velocity on it's own, but a half-size would probably be fine.

Best,

-Slashy

Edited January 4, 2015 by GoSlash27