Kerbal Space Program, Revisited.

[Simon_Jester]

The tug is performing less well than planned for reasons best attributed to "lack of designer experience," but will hopefully still be able to get the tour bus home...

[Jub]

The preview version of 1.2 is up for those who want to opt in. Just remember the usual disclaimers of bugs and lack of mod support before you make the plunge.

http://wiki.kerbalspaceprogram.com/wiki/1.2

[Napoleon the Clown]

There's a fanfic in this.

Done.

[Simon_Jester]

I have learned with some satisfaction that while I cannot make a high-altitude supersonic jet that flies stably in the sense of maintaining constant speed and altitude even with SAS, I can make them fly in a stable equilibrium state. That is, with no control input other than SAS, they will automatically slowly ascend to (say) 17000 meters, then they lose a bit of lift/power and start descending to lower altitude on their own, bottoming out at (say) 12000 meters, where they regain enough lift/thrust to start ascending again. This makes those round the world flights for survey missions a lot less annoying.

Does this square with anyone's experiences? Does anyone know how to build a plane that really WILL fly level?

[Imperial528]

Making high-altitude jets fly level, or even low-altitude ones, is rather difficult in KSP. Personally I think it has to do with the engine being optimized for space flight first, so planes tend to behave more closely to atmospheric orbiters than true airplanes.

I've had little success in making level aircraft at high altitude, and the rare successes I did have generally were flukes, and even those varied by one or two km over time.

Though they do seem to do better if you have a large degree of pitch control/response, and enough lift that you can stay in flight at high altitude at less than full throttle.

[Adam Reynolds]

I think the best solution is to look at mods that give better aerodynamics in the game. The default is somewhat limited in terms of how it models areodynamics, more than any issue with lack of parts. I have heard good things about the Ferram Aerospace mod, but have never got around to trying it myself. You might want to check that out if you are interesting in playing with aircraft in Kerbal.

[Dominus Atheos]

I have to disagree with both of you. Making jets fly level is rather difficult in real life. The physics are adequate, but designing an airplane is supposed to be nearly impossible.

[Napoleon the Clown]

Designing a decent plane isn't terribly hard in KSP. Just make sure the center of thrust is about in line with the center of mass and the center of lift is a bit behind the center of mass. Keeping a relatively constant elevation is more difficult, but as far as a plane that doesn't wobble itself a part you just need to avoid too many joints.

KSP is actually pretty forgiving about flight characteristics between subsonic and supersonic. Stresses increase with speed, but stable at 200 m/s will be fairly stable at 400 m/s, so long as it's structurally sound.

[Simon_Jester]

Yeah. I was able to figure out a plane that would fly (yes, land no) relatively easily. And once I had an adequate engine, making high altitude supercruise planes that could circle the planet wasn't that hard either. The main limiting factor seems to be the granularity of the controls.

Making high-altitude jets fly level, or even low-altitude ones, is rather difficult in KSP. Personally I think it has to do with the engine being optimized for space flight first, so planes tend to behave more closely to atmospheric orbiters than true airplanes.

I've had little success in making level aircraft at high altitude, and the rare successes I did have generally were flukes, and even those varied by one or two km over time.

Though they do seem to do better if you have a large degree of pitch control/response, and enough lift that you can stay in flight at high altitude at less than full throttle.

Well, I think the real problem is that there is some exact combination of speed and pitch setting that enable level flight at a given altitude. The precision of the controls isn't enough to allow me to set things that accurately on a reliable basis.

[Napoleon the Clown]

Hit capslock for fine control. Alternately, it supports joysticks as I understand. Those are a bit more precise.

[Simon_Jester]

Yes, but then if I want to switch away for anything, I have to worry about caps lock being on.

[Simon_Jester]

So, Jebediah and Bill Kerman have heroically stowed away boarded a prototype crewed Mun rover that was not ready for prime time a valiant attempt to correct the errors that led to the tragic loss of the first prototype.

At this time I have no provision for their safe return to Kerbin, though I do have a functional lander design that HAS been prototyped and tested, so all is not lost. Still, only in the Glorious People's Democratic Republic of North Kerbalistan could the first journey to the surface of the Mun begin with "Ah, fuck it, we're doing it live!"

[Simon_Jester]

"That's... uh-oh... uh, phew! He's okay. That's... one little faceplant for a Kerbal... one giant WOO-HOO! YEEAAAAAHH! for Kerbalkind!"
-Jebediah Kerman, observing Bill Kerman descend the ladder to the Munar surface on the hastily rechristened one-way descent vehicle Mun or Bust!

[Zeropoint]

In KSP, "bust" is always a reasonably likely outcome.

Nice lander/rover, by the way.

[Simon_Jester]

Thanks. It'd be a great design if the wheels actually spun. I'll figure out the bugs eventually, but for now I'm concentrating on how to get them back from the Mun.

Also...

[TimothyC]

Because no else mentioned it, KSP 1.2 is out, and so is BDB 1.0.1 For all* of your stock/porkjet-alike US rocketry needs.

*Not all US rockets included yet, but the rockets (not shuttles) that actually lifted people are included.

[Simon_Jester]

When I'm satisfied that I've done everything I care to do with stock rocketry, I'm going to download BDB and go to town. Specifically, the Cape.

Though I've decided to be stubborn about MechJeb. I can calculate my OWN doggone delta-V figures, thankyouverymuch.

In other news, while the first attempt at a rescue launch went horribly awry due to the liquid fuel boosters staying stuck to the TLI stage fuselage and dragging me down, the second attempt has a frighteningly well fueled LanderCan Mk I craft headed in the general direction of Our Heroes' broken rover.

Assembly of the nuclear-powered mothership Garland Belle is underway as well. Would be done, except I had to reload a save after screwing up something due to pure inattention.

[I wouldn't mind losing a Kerbal in a plane crash so much if I'd genuinely done something wrong in the design, but when I get called away to talk and the plane flies out of control in a failure mode I've never seen before in my life, I'm going to class that one as Not My Fault. My second generation stratospheric jet has higher peak performance than the first generation, don't get me wrong... but it's a lot more temperamental]

[TimothyC]

As an example of what you can build in BDB, here is Kāne-Prometheus (Apollo-Titan IIIM). It's a lot of fun, but you should fly a lofted trajectory on it because the core and the second stage have lower TWRs.

[Simon_Jester]

Well, the good news is, the Munar Excursion Module design worked (though I need to work on fuel-efficient landings). The ugly news is that this merely puts Jebediah and Bill in Munar orbit; getting them home is going to be a whole 'nother problem.

The AWESOME news is that the NERVA-powered "space station" is finally assembled (again!) and ready to go for rescue as soon as I get enough of a crew aboard for my planned Munar operations.

Jeb and Bill, back in Munar orbit with several boxes of rocks and some REALLY great pictures, having left in exchange a broken rover and a scattering of fragments from what's left of the rover's descent skycrane...

[Simon_Jester]

So I picked up about five contracts to do temperature readings on Minmus, and launched a satellite bus carrying three tiny little lander probes (unimaginatively I call them "Surveyor"). They make good orbiter satellites too, so I figured what the hey.

Cue me inserting the bus into polar orbit around Minmus. I figure I can get all the individual spots. Slow and steady... So I deploy Surveyor 9 from the top of the stack, nudge it into a slightly different polar orbit.

Then I looked up Minmus' day length and some of the exact altitudes in the contract (wait you want me to go HOW low? Are you sure you didn't mistake me for a tunneling firm?)

Then I did some quick figuring on the Surveyor probe's design. They're built to land on the Mun and have enough fuel in hand to hover for some time flying between closely spaced surface locations. So their total delta-V is several times Minmus's escape velocity.

A much more exciting plan than "spend a week of game time carefully maneuvering a polar orbiter" suggested itself.

The next hour was a frenetic series of maneuver nodes and rapid plane changes as my "polar orbiting" probe bounced around Minmus at an altitude anywhere from three to eight kilometers above reference datum like a rocket-powered hackeysack, jumping and sidestepping about every five minutes into a new ballistic suborbital trajectory to hit another contract location. On the last dregs of fuel, I landed Surveyor 9 near Minmus's south pole, mission so very, very accomplished.

Surveyor Nine in repose, returning an extra 60 science AFTER completing half of a mission intended for three probes, and in 1% the time!

[Simon_Jester]

Well, I figured out a way to build a good lander around a NERV engine, taking advantage of the NERV being able to run off liquid-fuel tanks intended for jets:

Mun Kerbal Surveyor, touched down on the Mun. In retrospect, should have found a flatter landing site.

But the future that we lost is still someplace out there
Orion still rides hellfire toward the blue
And rocketships land proudly on their tailfins
As God and Robert Heinlein meant them to...
-Steve Savitsky

[Zaune]

Given how much fuel you have left, I'm thinking this is one of those things you do just so you can say you've done it.

[Simon_Jester]

No, that was after the craft had already done trans-Munar insertion, and numerous plane changes in Munar orbit. Total delta-V is really high, it's just that I had less than a third of my fuel left BEFORE I decided to land for the heck of it, rather than stopping to top off the tanks before doing so.

[Zaune]

Okay, I'm suitably impressed. Can we see the booster you used to put it in orbit? It must have been impressive; NERVs are pretty much dead weight in the ascent phase.

[Simon_Jester]

Huh? Actually, it was just a two stage Poodle/Skipper arrangement, with some Hammers around the Skipper stage for the beginning of the ascent. Quite a meek and mild booster, at least by the standards of people accustomed to 2.5-meter parts.

I used the nuclear rocket as the third stage when I needed to do a bit of a last burn due to botching my ascent profile.

Remember, this is a single-engine NERV lander. Fully fueled it only weighs 10.5 tons, five tons of which is fuel, and three tons of which is engine.

EDIT: Originally the design was never intended to land at all, just to hop around the Mun taking crew reports at all sorts of different latitudes and altitudes, sort of like the Surveyor 9 mission I mentioned a few posts ago. I realized I could install landing legs after I thought of attaching those small diameter fuel tanks radially like you see.