Wing Patterns and Stability / Hypersonic Performance

[Raideur Ng]

BLUF: Can X pattern and / or otherwise unconventionally mounted wings provide everything required of wings during hypersonic SSTO ascent? 

I ask this because theres only so many ways you can mount enough gosh darn Rapiers without just ramping up your ballistic coefficient with forward area. It makes the most sense to mount them on wings since that keeps overall profile low. However, flat, straight wings can only mount so many before you have them reheating each other or just run out of room.

 

Any ideas?

I'm going for efficiency as opposed to brute size.

[GoSlash27]

In KSP, wings are wings. They're not particular about how they are mounted or what your wing loading is at high speed. Just maintain the proper longitudinal relationship between the CoM and CoL, and build it so your nose is pointed at your prograde vector. If possible, take advantage of those BigS deltas. They are an aerodynamically cheap way to carry your fuel. Much lower drag than fuselage tanks. Also high temperature tolerance.

If you do it right, you need very little engine to get yourself hypersonic at high altitude. Low drag is more important than high thrust.

*edit*

Use wing incidence to knife your way through Mach 1 instead of spamming engines/ intakes. If you can keep that needle- nose pointed straight ahead when drag is highest instead of pitched up, you'll top out the speed without needing lots of engines. Less engines means more payload.

 Use a few large panels instead of a bunch of small ones. Big wings don't need struts and struts are the devil.

Best,
-Slashy

 

 

Edited August 27, 2016 by GoSlash27

[Raideur Ng]

KSP does seem to care though. Dihedral and anhedral seem to cause some surprizing issues, much like canted tails. I ask because I do indeed find myself spamming engines just to break even when lifting super heavy payloads to make 0.60 TWR.

 

 

I am aware the Mk3 Command Pod is heavy and draggy, but designing vehicles like yours means it ALWAYS tail heavy and requires trimming on the fly. This also has 60 tons of ore in it.

[GoSlash27]

24 minutes ago, Raideur Ng said:

KSP does seem to care though. Dihedral and anhedral seem to cause some surprizing issues, much like canted tails. I ask because I do indeed find myself spamming engines just to break even when lifting super heavy payloads to make 0.60 TWR.

 

 

I am aware the Mk3 Command Pod is heavy and draggy, but designing vehicles like yours means it ALWAYS tail heavy and requires trimming on the fly. This also has 60 tons of ore in it.

Not sure what you mean. My designs are not tail- heavy and the CoM always stays the same whether fully- loaded or empty. I suppose that's a trick too; build it so that the CoM doesn't shift as the fuel drains.

Best,
-Slashy

[Raideur Ng]

The CoM and CoeM (empty) should not only be at roughly the same spot, but you want them centered in your cargo hold, so when you release cargo, it does not move. Getting them to do that requires forward weight, especially with heavy lifter payloads. What you displayed doesn't appear to carry any cargo. Vall's designs, however, are so large the differences probably average out. I suppose I am trying to find a middleground.

[GoSlash27]

In the design I posted, the fuel in orbit is the cargo. 1 orange tank minus the tank.

Of course... if you wish to carry something else instead of fuel, you would put it in a cargo bay or whatever, but the same design philosophy would apply.

Also, this design (like all of my designs) has the CoM stable. The forward fuel tank is locked to provide ballast. This design would work just as well with (say) a Mk3 cockpit.

 

All of this is getting away from my point: You don't need to spam wings because you don't need to spam engines. If your design is aerodynamically clean and you use the proper amount of wing incidence, you can do the job with very little thrust.

Best,
-Slashy

 

[dire]

Raideur what is the difference between where your nose is pointed -.- vs your prograde velocity -o- on the navball when you are hypersonic? If you angle your wings (should be shift-w in the editor) you can get your nose much more closely aligned to prograde and if you do that your drag will go way down because you want to be nose-on to your velocity and not belly-on. 

Slashy I notice you said your craft payload was a full orange tank and you have 2777 fuel showing but your resources readout says you actually have 3762 LFO. Does that ballast tank stay locked and return to KSC full?

[GoSlash27]

dire,

That ship is just an example, not a full- on working design. It was just meant to show that you can get useful payloads to orbit without needing to spam engines.
 Most of the extra fuel is actually jet fuel in the wings, but yeah... fuel used as ballast in the nose stays locked there for the whole trip.

Best,
-Slashy

[Raideur Ng]

I use ore as the boilerplate mass during most test missions. Ore is jettisoned in orbit for a lightweight return (Learned that the hard way...) Also, I ask Slashy, what is the point of ballast fuel if you're not stable after using it?

[GoSlash27]

Raideur Ng,

 I think you're hung up on the wrong thing. The ballast is not used, so that's not an issue. More importantly the ballast isn't the point. The point is that if you build an aerodynamically clean space plane, you don't need a lot of engines and therefore don't need a lot of wings.

Best,
-Slashy

[foamyesque]

Reducing drag's all well and good, but it doesn't really answer his question.

[Signo]

If you want efficiency you should follow slashy advice - a clean aero with proper wings AoA will allow for less wings and engines, therefore less overall drag and reducing your placement issue.

On the other hand KSP does not care so much about "IRL aero", so you can fly bricks if you wish so and you give them enough power overhead.

So you can have Batman

 

 

Or Aliens

 

Or an "IRL wannabe"

 

If are you looking for ideas on how to build "huge wing nacelles" housing clusters of 6/8 engines there are a few building tricks - I would suggest to avoid Mk2 decouplers due to their drag, but offsetted Mk1 parts built around an Mk3 nacelle can do the trick.

Edited August 29, 2016 by Signo

[GoSlash27]

3 hours ago, foamyesque said:

Reducing drag's all well and good, but it doesn't really answer his question.

Well.. it certainly wouldn't be the first time I've misunderstood a question.

The way I read his question, tho'... it actually does, especially if he's looking for efficiency.

Best,

-Slashy 

[Val]

On 27/08/2016 at 4:43 AM, Raideur Ng said:

KSP does seem to care though. Dihedral and anhedral seem to cause some surprizing issues, much like canted tails.

What surprising issues have you seen from dihedral and anhedral?

If you do an X-wing configuration then the dihedral and anhedral should cancel each other out and act as horizontal wings.

[Raideur Ng]

If this isn't optimized for drag, I don't know what is:

(Precooler clipped into the tail) It does have an excess of power, even on a single Rapier.

The question is when you start having to mount engines out beyond the body on pylons. For lifting VERY heavy payloads, you need a lot of engines to get at LEAST 0.5 TWR, otherwise take-offs generally don't occur how you'd like. For the same area, it would seem best to mount them on the wings, rather than outrigged on the body, significantly increasing cross-section and drag. However, when you REALLY start stacking engines, you run out of wing space before you are burning aft parts with exhaust (I prefer to use one forward set of wings and one aft and no horizontal stabilizer, since they hold no fuel). I realize engine power isn't required for hypersonic, but rather the initial takeoff, when Rapier performance is very low.

Thus, a better summary would be: It is worthwhile to have two sets of wings, front and back, one giant one, or an exotic X-wing layout for max lift and the most engine mounting for the least drag.

 

Additionally, concerning the X-wing layout, I did indeed try a heavy style version, with properly rotated wings for max lift, and then canted. It does work, but requires additional rear stabilization and it makes mounting engines on the wings themselves really nightmarish.

[Signo]

As I wrote above there are many ways to build very good engine clusters for your needs w/o resorting to any exotic configuration.

Take a look at this - Last craft depicted has got 36 rapiers clustered in 4 big Mk3 wing nacelles. My 4 rapier Mk3 cargo can lift an orange tank to LKO so I must assume that a 36 rapier one can do the job with a 400t payload.

 

[Val]

6 hours ago, Raideur Ng said:

If this isn't optimized for drag, I don't know what is:

(Precooler clipped into the tail) It does have an excess of power, even on a single Rapier.

(Nitpicking) If the wings are not mounted with Angle of Incidence, then you can still reduce drag by doing that. Giving the wings built-in Angle of Attack means the craft no longer needs to fly with a constant nose-up attitude relative to prograde in atmosphere.

 

6 hours ago, Raideur Ng said:

The question is when you start having to mount engines out beyond the body on pylons. For lifting VERY heavy payloads, you need a lot of engines to get at LEAST 0.5 TWR, otherwise take-offs generally don't occur how you'd like. For the same area, it would seem best to mount them on the wings, rather than outrigged on the body, significantly increasing cross-section and drag. However, when you REALLY start stacking engines, you run out of wing space before you are burning aft parts with exhaust (I prefer to use one forward set of wings and one aft and no horizontal stabilizer, since they hold no fuel).

From a drag perspective, KSP aerodynamics doesn't really care how you attach the engines. If the engine is not in the same stack as the fuselage (like the little spaceplane in your picture above) then it's drag is the same no matter how you mount it. The craft will have the same frontal area and drag, no matter whether you clip the engine into the wing, the fuselage, or don't clip it at all.

Moving the engines to the wings also has it's own set of issues when you get up in the heavy class.

The pictured craft lifts up to 108 t to 80 km orbit. It's similar to yours with wings and engines in both ends, and very little CoM movement with cargo and fuel changes. (The craft was made in 1.0.5, but still works fine in 1.1.3)

The biggest issue with the design was that the engines warped the wings enormously when they reached maximum thrust. To the point that the craft became uncontrollable until I strutted the engine nacelles. Struts are bad for drag, though.

 

6 hours ago, Raideur Ng said:

I realize engine power isn't required for hypersonic, but rather the initial takeoff, when Rapier performance is very low.

Angle of Incidence and good placement of landing gear can help with takeoff by increasing how much Angle of Attack you have available when you reach the end of the runway.

 

6 hours ago, Raideur Ng said:

Thus, a better summary would be: It is worthwhile to have two sets of wings, front and back, one giant one, or an exotic X-wing layout for max lift and the most engine mounting for the least drag.

Additionally, concerning the X-wing layout, I did indeed try a heavy style version, with properly rotated wings for max lift, and then canted. It does work, but requires additional rear stabilization and it makes mounting engines on the wings themselves really nightmarish.

My recommendation is to use some form of engine nacelle made from bi/tri/quad-couplers, if you want more than a couple engines per wing. Though, I'd prefer mounting the nacelle directly to the fuselage to reduce warping.

[foamyesque]

@Val: I think your example design is running way more intakes than you need. Shouldn't require more than one shock intake per six rapiers; you could swap out the front bicouplers for adapters and put a single intake per cluster.

Edited August 31, 2016 by foamyesque

[Val]

1 minute ago, foamyesque said:

@Val: I think your example design is running way more intakes than you need. Shouldn't require more than one shock intake per six rapiers; you could swap out the front bicouplers for adapters and put a single intake per cluster.

Yeah. It's not been updated since 1.0.5. Also, the drag reduction would be minimal. There's very little difference between the Mk2 bicoupler and Mk2-1.25 adapter.

I'll take a more thorough look when I review my designs for KSP 1.2.

[foamyesque]

Just now, Val said:

Yeah. It's not been updated since 1.0.5. Also, the drag reduction would be minimal. There's very little difference between the Mk2 bicoupler and Mk2-1.25 adapter.

I'll take a more thorough look when I review my designs for KSP 1.2.

 

There is, IME, notable drag reductions in having fewer intakes, though.

[Raideur Ng]

Concerning your nitpicking, they are indeed mounted with proper AoI, otherwise that would never make it into orbit.

Hrm. I was hoping KSP wasn't that simple concerning clipping, but oh well. I avoid mounting -anything- on wings due to warping, including engine stacks or landing gear. It also makes no sense to use anything but the precooler for intakes.

[foamyesque]

14 minutes ago, Raideur Ng said:

Concerning your nitpicking, they are indeed mounted with proper AoI, otherwise that would never make it into orbit.

Hrm. I was hoping KSP wasn't that simple concerning clipping, but oh well. I avoid mounting -anything- on wings due to warping, including engine stacks or landing gear. It also makes no sense to use anything but the precooler for intakes.

 

Precoolers are space-inefficient. If you're horsing truly large things to orbit you're better served by swapping out nosecones on your radial stacks for shock intakes, and using the space that precoolers would take up with ordinary fuel tanks (LF or LFO, as appropriate).

[Val]

1 hour ago, Raideur Ng said:

It also makes no sense to use anything but the precooler for intakes.

Depends. If you're optimizing for drag, then depending on craft design Precoolers may be more efficient.

But, if you're optimizing for weight, a single Shock Cone feeding 5 RAPIERs is much more efficient than 5 Precoolers.

In my experience, there's never a single solution that's best for all cases.

[GoSlash27]

I agree with the stuff the others have said above. If you're mounting engines radially, it doesn't matter whether they're attached to the fuselage, wings, clipped, or not clipped. The drag is the same.

If you're going to mount a large number of engines, it's best to cluster them using adapters with a single shock cone at the front. It's aerodynamically clean, supplies enough air to feed all it's engines, and the engines are nicely occluded by the stack. Save the precoolers for ultra-low t/w designs.

Good luck!

-Slashy

Edited September 1, 2016 by GoSlash27

[Van Disaster]

@Raideur Ng: As usual if people seem interested in aerodynamics I point them at FAR ( Ferram Aerospace Research ) - wing shape most definitely does matter under FAR aero, as does supersonic+ ( technically critical mach+ ) overall craft shape.

( you can still make practically anything fly under FAR because Kerbal material science is centuries ahead of ours, even if the rest of their physics isn't ).

Edited September 2, 2016 by Van Disaster