The woes of building a space-plane.

[Vanamonde]

Thanks to the suggestions you folks gave me yesterday, I just landed on the runway for the first time. In fact, it was easy.

As for controls, how do you guys have yours configured? Since I\'m mostly interested in rockets, I use the stick for yaw and pitch but buttons for roll. I imagine the classic pitch/roll stick would be easier for spaceplanes, but I don\'t want to have to go the settings screen and screw around every time I go from rocket to plane. It would be nice if there was a mode switch.

[Khrissetti]

This is the first plane I\'ve ever built which has actually managed to stay aloft -- the Wright

It glides fairly well, the only problem is controllability, I have to take SAS off, enter a direction, then hit the SAS back on again to control it.

Still, made it half way to KSC2 with 2 fuel tanks left on the first go.

[Disfunctional_Rockets_INC]

What I have noticed is that large planes that have wings made up of many pieces are much easier to make than little planes. When you balance a little plane you have to do it extremely carefully because you can\'t adjust the wing lift values of each wing, only the position and angle. When you balance a large plane you can just kind haphazardly add or remove wing parts to change the balance. It also seems that big planes with lots of wings and only a few fuel tanks can fly extremely flat while holding an altitude (no pointing the nose upward for level flight).

Take a look at this low quality video of one of my better aircraft.

I think until we can adjust the amount of fuel in each fuel tank or the lift each wing has we won\'t be able to make little planes that fly well very easily (it can still be done, it just seems to take more effort or luck), for now just try making some big planes (remember to use some struts on the wings or they get all bendy).

Here are some reference screenshots of the aircraft.

http://imgur.com/a/yik1Q#0

[Zephram Kerman]

None of my planes work very well at all. But I\'ve noticed some rules that seem to make a difference:

• remember the KSP drag model is weird (a full fuel tank has more drag than an empty one);

• center of mass should slightly forward from center of lift;

• center of drag should be slightly above center of jet thrust (for pitch trim convenience)

• (except outside the atmosphere, then the center of mass should be in-line with rocket thrust);

• for tricycles, main landing gear should be just behind center of mass

• (for tail-draggers, main landing gear should be just forward of center of mass);

• do not use rudders to turn... instead, first bank, then pitch up towards your desired heading;

• put fuel tanks near the center of mass, so that they do not affect trim as they empty out;

• canards give lots of control authority, but cause a tumble if Angle-Of-Attack is too large;

• put control surfaces aft... if they\'re near the middle they don\'t help;

• it takes at least 2 Gs of lift to 'un-stick' from the runway.

I hope these tips help you build better planes than I do!

[Vanamonde]

On many radically different designs I find that I have the same problem: I struggle to get the nose up while taking off, then struggle to keep the nose down if I can get into the air. Every change I make to ease one problem makes the other worse. And yet often a plane that can\'t translate while on the runway will spring up as soon as I run off the end of the runway. Does anybody understand this?

[TheGripenSaidWhat?]

Van, that\'s not so much a bug as it is a downfall of the simplified atmospheric engine we have in place right now. Wings do not generate lift at a 0 degree angle of attack, so even though it may seem like your plane doesnt have enough wing surface area, it\'s really that until your plane pitches off the runway and you\'re able to pitch the wings a little, the only things that are actively lifting the plane are the canards/control surfaces. I often use tail-draggers to help mitigate that problem,and my favorite atmospheric plane as of late is the Howler Monkey, a relatively large Biplane that will bring it\'s tail up shortly after thrust is applied at around 35 m/s. [WARNING: Uses Damned Aerospace engines]

[Vanamonde]

Wings do not generate lift at a 0 degree angle of attack

Thank you. That\'s good to know. I\'ve been deliberately avoiding other people\'s hints and designs because I want to do it myself, so I only come back to this kindly thread when I\'m desperate. But it seems I\'ve thereby failed to learn something that\'s, oh, kind of FLIPPING CRITICAL! :

But it\'s not the whole story because most of my designs are tail-draggers that sit on a bit of an angle and some of them sit on the tarmac like Gibraltar, and yet a couple of planes that I built with level postures can takeoff. I keep putting canards up front to force the stupid nose up, and that works, but then most planes with those go into an uncontrollable pitch-up, and I take the canards right back off again. I tweak and un-tweak and tweak again, and that\'s how a kind of pretty design (if I do say so myself) like #24 here evolves into the monstrosity of #39. #39 seems to be like a black hole: it has reached a size where no known force can prevent its collapse.

[HarvesteR]

When I was little, I used to build model balsa-wood airplanes, and I learned some useful rules of thumb that might be worth sharing:

One, if it looks like a plane, it will probably fly like a plane. Try a very conventional design before you go for the successor to the SR-71.

The KRJ-100. Looks very much like a plane. It will glide without input.

Two, the center of mass should be at the first third of the wings. Mentally divide your wing into three sections. The plane should be balanced at the first division threshold. (landing gear can be just aft of that, to provide a nice fulcrum)

The KRJ-400 JetLiner, it\'s as heavy as it looks like.

Three, vertical stabilizers should actually be pushing down, not up. Use the rotation feature in the SPH to pitch your elevators down a notch or two before placing. The wings alone will make the plane pitch down, while the stabilizers keep the nose up. Planes are stable when those forces even out.

The Annihilator Mk1, the V tail acts as both rudder and elevator, and they\'re pitched down 10 degrees.

Four, canards are like pushing on a rope. See rule 1.

The Ravenspear Mk1 is much trickier to fly than the Annihilator. Pitch too aggressively, and it will tumble out of control.

That\'s about it for now, hope this helps.

Cheers

[Vanamonde]

Wow. Advice from the Big Kerbal himself. Thanks!

If a piece has a lift rating and you place it vertically, does it generate any forces?

[TheGripenSaidWhat?]

^Another Reason to love this community

[Vanamonde]

My monster spaceplane is almost working. The last major hurdle is that the whole thing explodes the instant it touches the ground even at 6m/s descent speed. (I am not ignoring your kindly help, HarvesteR. It\'s just that I\'ve invested so much time in the thing already that I can\'t bear the thought of giving up on it. In fact, a couple of your tips in that post are the only reason the thing works as well as it does now. ) It happens all at once, too fast to see, but the F3 autopsy report indicates that it\'s so big that the landing gear crushes, triggering a catastrophic cascade of calamities.

As much as I dislike using mods (for personal neurotic reasons we needn\'t go into here), I think I might be able to make the whole agonizing travail worthwhile if I could slap some heavy duty landing gear on there. Any suggestions? Nothing fancy: just a strong wheel-and-strut kind of a thing. Retractible would be preferable.

I\'d post images or the .craft, but I want to save the reveal for a ludicrous punchline awe-inspiring spectacle.

[ModernArc]

When I was testing a SpacePlane, I tested how high it could go. I stalled and it started shaking, I crashed into the ground. I think everyone had that happen to them. Lol.

[Vanamonde]

My monster spaceplane is almost working.

My monster spaceplanet was almost working for three GD days, but never actually did. I worked and worked and got it to where it could mostly do what I wanted it to do, but it was kind of fragile and needed to be buffed up a bit. I\'d left that to last because I didn\'t want to have to move a bunch of struts each time I tinkered with something. The right tail plane fell off as soon as the screen came up on the runway. It had never given me any trouble before, and only started falling off after I added 8 struts to keep the stupid thing on.

There comes a time when one must accept not just that defeat has become unavoidable, but that it has already happened. Deleted the plane. Deleted all prior iterations of the plane. Deleted all screenshots of the plane. I\'m so angry right now I feel a little sick to my stomach.

But you know as well as I do that I\'m going to come right back and try again tomorrow. What am I, stupid? =P

[nibula]

I love harvesters jet liner!

[Volt]

I\'m glad to have finally had it confirmed that planes don\'t generate lift at 0 alpha. Possibly why my Delta Dart/Lightning replica suddenly became my fastest taking-off aircraft when I twisted the front part of the wing so the point was facing up one notch...

[Vanamonde]

Through our innovative development paradigm of whining to people who are better at this and then stealing their ideas, Yeahletstrythatdyne Aerospacedyne anxiously presents, our first non-embarrassing product line: The F1 Pilum. It takes off, handles reasonably well, and lands, without exploding even a little bit. It may not break new ground in the science of (simulated) aeronautics, but as a wise man once said, 'For me it\'s awesome becouse I made it.'

Three-quarter view, side view, and jet exhaust frying chickens in the barnyard.

Yeahletstrythatdyne Aerospacedyne: if we could make it better, we would have.

[TimeLapseSteve]

One of the things that always bugged me is how the individual engines are far from realistic in the way they distribute weight on an airplane in rel life. In reality, the center of gravity of the engine is near the middle of the cowl - mainly the compressor/fan section. The tailpipe (represented in game, weight-wise, by an entire engine) is actually lightweight compared to the rest of the structure. See this cross section of a RB-211, an engine commonly used on the Boeing 757, to illustrate what I\'m talking about:

I would suggest making an additional engine part by combining three existing individual parts into one - the intake, the cowling, and then the engine (which would then become the tailpipe portion of the new part) - it wouldn\'t be customizable, but it would be realistic. (The alternative would be to make the cowls much heavier while keeping the engines lightweight... but that wouldn\'t make any sense). This would place the combined mass of the pieces in the center of the cowling, which is where it should be attached to the wing or fuselage. It would definitely make things even easier when trying to get the CG of the aircraft right - many people probably have noticed that the planes they build still seem to be tail heavy, even though all of the parts appear to be in the correct location from an aesthetic point of view.

[Vanamonde]

With your kindly help here I\'ve been able to make some small planes I\'m pleased with, but larger models are proving to have problems of their own.

I understand that the violent-veer-on-takeoff is a problem of self-collision, but even so, it\'s stronger on some planes than others. What affects it and how can I reduce it? (I\'ve got a plane right now that slalloms down the field like a competitive skier.)

Do lift-generating parts generate any forces if placed vertically?

Suppose I want to add tanks to side mounts, as in the screenshot, and I want to reinforce the arrangement with small hardpoints. If I place the hardpoints first, as in the second screenshot, the interface has a tendency to try to stick the tanks to them, possibly leaving gaps between the tanks. So my question is, if I attach tank-to-tank and then place the hardpoint between the fuselage and the second tank, as in the third screenshot, is the hardpoint actually connecting to the outer tank, or is that just a visual overlap?

[Nori]

This thread has some really great info, but has fallen down the pages a bit. Could it maybe be moved to the How To forum and possible stickied? Building planes is really hard and having these pages (especially Harvester's post) stickied would be really nice.

[Sherlock21st]

What I've discovered is that triangularly winged aircraft seem to benefit from wings that angle downwards like the tail fins on an F-4 Phantom

Let me give an example

You have a plane, and you have wings.

The first "layer" of wing parts angle downwards, then the rest of the wing parts further from the fuselage are corrected.

That design seems to work with planes whose thrust comes from the back.