Wanderfound

Landing in the dark, with mods: use Scansat to pick a flat spot of known altitude. Landing in the dark, without mods: check your radar altimeter frequently and touch down lightly enough that you can immediately lift off again if you find yourself on an excessive slope. Use the map view to ensure that your trajectory isn't clipping a mountain.

Okeydoke; see http://s1378.photobucket.com/user/craigmotbey/Kerbal/Challenges/SSTYC/story for an annotated album. Didn't quite make it to Laythe, but I did drop a satellite around Jool. Could've made Laythe if I'd been less careless with my fuel.

LV-N's make interplanetary easier, but they aren't compulsory. Set up some orbital refuelling stations and you can do it with conventional engines. I'd highly recommend trying the Fine Print mod. It's due to be integrated into stock in the next update anyway, and it adds a lot more variety to the contracts, especially around Kerbin. Launching some satellites or space stations would give you the science you're after. If you must stay stock, launch a lab/lander/fuel combo to the Mun. Leave the lab and fuel in orbit, and use the lander to pop down to every biome, refuelling and cleaning your science gear at the lab in between each hop.

Well, my (small, three engined) spaceplane just hit orbit with over 6,000ÃŽâ€V in the tank. Where to go? I think Laythe.

Orbital refuelling, yay or nay?

...which is why there is substantially balancing to come in .90 and beyond.

Takeoff with the view swung around to watch the landing gear. It could be a few things: * Wingflex disrupting the steering geometry. If this is the case, the plane will veer while the rear wheels are solidly on the ground. You may be able to see the rear gear wobble. * Rear gear lifting before the front gear, due to the static pitch design. If this is the case, the rear gear will start to lift off the ground first, leaving the plane running as a front wheel unicycle (or assymetric bicycle, if one wheel lifts first), which will then wobble and crash. You could avoid the problem entirely by placing the rear gear on the lateral tanks, just behind CoM. Use a small hardpoint to raise the front gear if the plane is pitched down.

By FAR analysis tools and test flying. If you're going to get these pitch-up problems, you'll have bad pitch derivative numbers and a misplaced Cm line. Just the AoA sweep that you posted alone is enough to say that it was going to go heavily unstable at moderate AoA; you can get away with a hint of Cm inflection if you've got sufficient pitch authority, but once it starts shooting up like that it's game over. Have you had a poke at the "How to use FAR graphs" part of the Kerbodyne build tutorial? "Too much" vs "too little" is always tricky. There does seem to be a fair tendency for KSP builders to err on the side of "too much", though. Too little wing means high AoA, high takeoff speed, slow turning. Too much wing is excess weight, excess drag, excess fragility and possible aerodynamic complications. As a general rule, the more conventional the airframe, the easier it is to get the aero right. There's a reason they built them that way. The benefit of a delta is that it's very good at maximising lift while minimising drag; the tradeoff is that deltas have crappy stability. You didn't see a lot of successful deltas in the pre fly-by-wire days. You can make most things work with enough fiddling, but when in doubt, use the design clichés. It also helps to think about the problem in terms of basic forces. If it's pitching up, there's too much lift at the nose and not enough at the tail. So, either remove lift from the front, add lift to the back, or extend the nose or tail to alter the position of the lift relative to CoM (thereby increasing or reducing the leverage). Similarly, if it's yawing, you need more tailfin drag at the back. Roll instability calls for wingtip extensions or dihedral.

As is this one: You can design around the problem, sure. But when they go wrong, they often go wrong in this particular way, and it's due to the lack of tailplane and rear-biased CoM tending to give inadequate high AoA pitch stability. Which is part of why the ones that work tend to be not too excessive in wing area, and taper down the wing towards the nose.

Actually, that might be an interesting idea for a contract: take the same spacecraft to X,Y and Z. Mini Grand Tours.

Actually, you may have a point there. Doesn't matter; they still make good Scansat and lander carriers. Actually, that messes up my other plan, too; I was going to send up a stash of satellites into orbit so that the spaceplanes could reload without having to come back down. Poot.

I've been making the most of the fact that stations and don't have to stay in place. Instead of a one-off station for each moon, I'm making big nuke-driven research ships and sending them along to each space station contract as it comes up.

BTW: if anyone's curious to see the final version of the ship from the tutorial, here it is: https://www.dropbox.com/s/rp0m3grjm45yqa1/Kerbodyne%20Simplicio.craft?dl=0 Just a few changes after the test flight... Added the forgotten solar to the satellite, and removed the non-stock life support parts. Added some more batteries as well. Extra landing gear. As well as extra toughness and stability, this gives us extra brakes; needed. The RCS was doubled, which should provide better translation balance and more landing retro-thrust. The rearmost normal gear were also shifted slightly rearwards, in order to steepen the pitch angle between the rear gear and the tailstrike guards. This should hopefully ease the takeoff difficulty slightly. As should this. Instead of having all of the airbrakes linked to the landing brakes, they are now split in half, controlled by action groups 9 and 0. This allows the lower half of the airbrakes to be deployed independently as flaps. However, the #1 way to ease takeoff also applies: more speed, more thrust. Just run the LV-N during takeoff and it's fine. See? Easy.

Diversify the tree, split the groups of parts into individual nodes. Make it multi-threaded; an unmanned stream, a manned rocket stream, an aircraft stream, an engines stream, a science stream, a ground vehicles stream. Workable parts in the start nodes of all streams. People can specialise or diversify as much as they want; the same tech tree can do everything from a Gemini start to Goddard and the Wright Brothers. As always, let everyone play the way they want to and you're on a winner.

The lift vector removal was deliberate.

And are you sure you don't have an asymmetrically drained fuel tank or something?

Those full-body delta designs are prone to nasty pitch-up problems; once they start to go, there's a lot of wing area pushing them around and no tailplane to pull them back. That's what your yellow line is telling you; as soon as it inflects up, any pitch deviation will tend to increase rather than damp out. You want that yellow line to stay sloping down until at least 20° and preferably more. Either shift to a conventional long tail design or give it a huge amount of high-speed high-altitude pitch authority (canards and Vernors). Reducing forward wing area may help a bit, but probably not enough.

DCA will definitely help you. There are three keyboard control modes, in increasing order of precision: normal, fine (caps-lock, blue indicators), trim (alt-WASD). Run through the sequence of them as the pressure increases. The other tricks to keyboard flying are to tap rather than hold the keys, and to watch the G-meter as much as the navball. Avoid G-spikes, and back off the controls when the G's climb too much. Wait for them to settle down before you touch the controls again. You'd also likely find Kerbal Flight Data a useful mod. As well as many other useful things, it gives easy and clear stall and pressure monitoring:

Yeah, the weight balance is particularly key on that one, because I used the weight tuning to counteract the thrust asymmetries of the stacked-engine design. And it's designed to make use of its extra-strength wings for low-altitude supersonic nonsense.

Thanks, hopefully fixed now.

Stock wings in FAR used to be modelled on glider wings; very light, very fragile. Now, by default, they're modelled on fighter wings; very tough, very heavy. But they can be adjusted back down to what they were, or up even higher for ridiculousness value. NEAR wings have the weight of old stock wings. Although your ship flew fine in FAR, it would have been carrying a fair bit more weight than it was designed for. This likely explains the thrust/AoA/pitchup issues I had. Most of my stuff is set with the wings halfway between FAR and NEAR, with the tailfins a bit lighter still.

This one wasn't designed as an aerobicist, but given that it held this posture and sustained G-level for several minutes during descent... ...I'm happy enough with how it flies. If I was going for an aerobatics specialist these days, I still wouldn't strut much, I'd instead crank up the mass/strength ratings in FAR: Yeah, either of them could be strengthened further at no mass penalty with some struts. But I just don't like the look of them when exposed, or the hassle of hiding them and dealing with symmetry bugs.

Another part that needs to be stockified. Squad: moar lights, please!

There are captions on every image, and I've fiddled with the setup to try and make them visible by default. They should be okay on the link now, hopefully. The wings are unstrutted because they don't need strutting; they're strong enough as is to handle anything except for actively suicidal piloting. No need to inflate the part count without cause. The RAPIER mode action group is separated from intakes because I don't necessarily always want them to switch together. The RAPIERs are still set to automatic switching (for thrust asymmetry peace of mind), so they may change modes on their own. Having the mode switch and intake toggle on the same action group could result in them going out of sync after an auto-switch. Plus there are times when I want to flick RAPIER modes for just a moment (e.g. takeoff boosters) or toggle the intakes on their own (e.g. fine tuning drag during an engines-off glide).

Okay folks; I've put together a little slideshow tutorial. It goes all the way through the construction of a spaceplane, to taking it to orbit, bringing it through reentry and landing back at KSC. There are just under 200 slides, but they're all JPEGs so it shouldn't be too bandwidth hungry. I deliberately left the mistakes and revisions in the sequence in order to give people an idea of how the process works. See http://s1378.photobucket.com/user/cr...0Landing/story for the slideshow. All pictures have captions.