Horn Brain

It's quite impractical to try and do this all with one vehicle. Even with nuke engines, if I need to carry a sizeable lander beyond Duna I bring an interplanetary cruise stage, the lander itself, and a refueling tanker. The cruise stage carries the lander and uses up the vast majority of its fuel getting to Jool and then getting into orbit. Then the lander goes and lands. In the meantime, the refueling tanker (flown separately from the other two) arrives with about half of its fuel or more still remaining and docks with the cruise vehicle in orbit, transfers all of its fuel and then deorbits with RCS. The lander returns to the now partially-fueled cruise stage, the crew transfers and the lander is ditched before burning back for Kerbin. By doing this I have plenty of fuel left upon my return to Kerbin to circularize my orbit and rendezvous and dock with my space station so I can reuse the interplanetary stage. Here is a breakdown of my standard deep-space mission design so that you can see what kind of infrastructure you're going to need to bring: Cruise stage: 1 big orange tank, four nuke engines (mounted radially on some structural parts), four 1-m RCS tanks and RCS thrusters for docking, one hab module, one 2-kerbal lander can, and senior docking ports on the front and back (plus extras like solar panels, lights, batteries, antennae, etc. for utility/show). I have a second version which has landing legs and can land on the Mun pretty easily, so it has decent thrust leading to not intolerably long burns. Lander: Varies, I estimate the landing dV beforehand and then give myself a margin, then I make the lightest vehicle that I can with the requisite dV and TWR. My Vall lander had two nuke engines, but the Dres version only used 2 LV-909s, it turns out that the extra fuel required by the chemical engines weighed less than the difference in mass between the nukes and the chems. It's important that you minimize the mass of this because it will cost you loads of fuel to get it out to your target! Refueling tanker: One big orange tank, two nuke engines mounted radially on structural parts, two 1m RCS tanks (for replenishing supplies on the cruise stage, if necessary), one of the big probe cores (it's not the lightest but it looks better) and some small solar panels tacked to the sides. There is a single senior docking port on the front. This is a flying fuel can and should be able to get anywhere in the kerbolar system with plenty of fuel left to get your cruise stage home, especially if you aerobrake at Kerbin.

Hey I just landed there last night! How do you like my little surface excursion landers (I call them "Wild Things" for obvious reasons)? They still need a little tweaking, but they work alright for now.

I'm currently on the return leg of a kerballed mission to Dres, but the things I haven't done are: The I-just-haven't-gotten-around-to-it category: Land anything on Tylo. Manned mission to Eeloo and back (I do have a rover, though. That was fun!). Proper manned mission to Laythe and back, using a small spaceplane as a lander (previous mission was before docking, carried one kerbal, and had only the small pod). The Return-trip-to-Eve-difficulty category: Manned mission to Tylo and back. Do a low-atmosphere sample of Jool robotically using the airship mod. This is just a crazy plan, not sure if it's ever been done. It will be difficult to get the required dV with the low mass-fraction of fuel available in kerbal tanks. Definitely a robotic mission.

That would be a great tool for airless bodies! For atmospheric reentry, I think the best thing to do now is (only because with the way the game calculates drag almost everything has the same ballistic coefficient, i.e. drag will have the same effect on most things that don't have lots of aerospikes or parachutes) to launch a series of probes into orbit around your target and experiment. My procedure for targeted reentry (worked on Eve, would also work on Laythe, Kerbin, and Duna, but on Duna you should expect much less precision due to the terrain elevation): 1.) Launch a probe into a low circular orbit that you will use to begin all targeted descents (Obviously remember the altitude. Mine for Eve is 135km) 2.) Choose a target location beneath your orbit and quicksave (I chose a patch near the equator with about 4.5km hills, but you may also like landing a probe there to judge distance more accurately). 3.) When you reach the exact opposite point in your orbit (other side of planet), do a de-orbit burn and put your periapsis directly over where your target will be when you get to the other side. Remember the Pe altitude. 4.) See how close you get to your target. If you land before your target, raise your Pe next time (do a quickload and try again), if you land past it, lower it next time. 5.) Repeat steps 3+4 until you are satisfied with your accuracy. 6.) Remember the orbit altitude and the Pe altitude and use them for targeted deorbits from now on! On Eve, a 135km x ~68km orbit will land you directly beneath your periapsis (search around that Pe, don't take my word for it. I had lots of parachutes and aerospikes because I was doing an Eve return). Also, if you have lots of wings and fins, you may be able to adjust your trajectory significantly to account for errors. Good luck!

ITT: None of the names mean anything in Spanish. Amazing.

Since this is KSP, I say go for it and show us your pictures. In real life, however, this wouldn't make sense unless you could detach the lander pretty close to (a few days before) your encounter with Moho. Unless you give the thing a huge kick when you do that, though, you'll only have a few hours on the surface, tops. If you're going to expect someone to survive for several days, then you need to bring along something about the size of the Apollo capsule at least. I would say the best you could do (thinking Mercury, here) is have a small crew capsule undock from the large interplanetary habitat a few days (maybe a couple weeks, tops) before encounter, give itself a significant dV kick to give at least a day on the surface in head start time on the mothership, then perform a direct descent and landing to the surface. The ascent vehicle would be very small (think like some lawn chairs on top of a fuel tank), and would have to perfectly time the launch and boost to intercept the mothership, which would then be responsible to maneuver in close enough for either capture with a robotic arm or EVA. The savings for this system would be in not having to slow down and reboost the interplanetary habitat, as well as whatever dV could be saved by using the flyby as a gravity assist maneuver to get an encounter with Earth on the way back out. The costs would come mostly from the extra kick you'd have to give the lander in order to get a lead on the habitat, as well as the extra costs of slowing down from the non-optimized encounter by the lander. Having thought about it, it doesn't sound so crazy after all in terms of the savings/cost analysis, and I could imagine seeing some potentially large mass savings for an actual mission, depending on lots of other parameters, like the actual mass of the required interplanetary habitat for shielding and other purposes, etc. The real problem with this plan is that there is very little time for anything to be fixed if even a single thing goes wrong once the lander leaves the habitat, and there is absolutely no backup if there is a significant delay in the ascent phase. Even in Apollo, they could have fiddled with the engines on the surface and waited for the next orbit of the CM, and the CM could come get them if they got into a bad orbit due to engine trouble after liftoff, but with this plan, there would be exactly one chance for the engines to work, and after that it would likely be impossible to reach the habitat. Then again, I don't know the exact mechanics of the flyby. Maybe you could arrange it so that you had some extra dV in the ascent stage so that you could catch up to the habitat after a little bit of a cruise. All these things add weight, though, and cut into the savings in the first place. The fact that there are this many considerations tells me the plan isn't totally crazy, and it may be worth a look just to get an idea of the feasibility of it. Good thinking!

Fuel is free compared to hardware. If the cost of your shuttle program is anything like the cost of your expendable rocket program, then you're doing your shuttle wrong. Assuming the fuel is something like RP-1, you can use aviation kerosene as a ballpark estimate. That costs $925 per metric ton right now according to the IATA (http://www.iata.org/publications/economics/fuel-monitor/Pages/price-analysis.aspx), so call it $1000 per ton. Liquid oxygen is cheaper than that, so I would just use $1000 per ton of fuel/oxidizer. Those are real dollars, not kerbal dollars, which I have no idea how to convert, but the point is that even large rockets running on RP-1 aren't going to have even $1,000,000 in fuel costs, while the cost of a large rocket launch is often in the hundreds of millions of dollars range. Ignore fuel cost.

If it said 25km to the ground and you hit the ground, you either read the altimeter wrong or you found a bug. It probably said 2.5km to the ground. The Mun's elevation is usually around that high, up to about 4km near the poles.

It only makes sense to do this if you're mining kethane from the Mun and storing it in munar orbit. If you're getting your fuel from Kerbin, then you're just wasting time and fuel moving it up to the Mun. I did some analysis on this, but the forum monster ate it. :-(

If you're trying to get a satellite there, it's actually quite easy if you just use ion engines. Your probe should be the lightest probe core, enough solar panels to run the ion engine full throttle (if you don't use it until you have escaped Kerbin, you can get away with just two or three near Moho), one xenon tank, one battery, and the engine. I had one slightly heavier than that and I got to Moho with one big SRB with fins, one medium 1-meter fuel tank and an LV-909, and the probe. I still had about half my Xenon left, too. Ion engines are just not useful for big vehicles. Keep it light. Here is what I got into a polar orbit around Moho: The bare-bones vehicle I described will be even easier to fly there.

Jason's got it. Put a thrust-vectoring engine on there and you should be fine. Here's the reason for your problem: Once you get to high altitudes, tail fins and winglets don't have as much authority and all that matters is drag. The parachutes have a higher drag than the other parts, and (with the way the game calculates drag) that means that any vehicle with a parachute on the front is going to have the center of drag ahead of the center of mass, which means it's aerodynamically unstable (it flips over).

Put a light on the side of your vehicle pointing down and turn it on when you're landing. It makes judging the distance to the ground super easy during the last two hundred meters or so. Don't forget to provide power, though.

That thing needs about 12 more orange tanks with mainsails/skippers on them, all asparagus staged, if you even want to smell space. I can lift an orange tank and a little extra payload nearly fully fueled to orbit with 6 orange tanks of asparagus boosters and a core stage not too different from the ones you've got beneath each of your intended payloads, so that's probably a good ballpark figure of how much you're lacking. You might try running a fuel line from the payload to the boosters, then dumping them (transfer forward all remaining fuel first) just before you reach orbit and adding a few of the tiny radial engines to the payload to perform the circularization. Then you can send up more reasonable tankers to refuel your station. Launching two orange tanks into orbit is approaching the limits of what the game can handle on a regular machine.

I think it could be fun for someone who has studied ancient mythologies much better than I have to construct a mythology involving the names of the planets and moons in the Kerbolar system.

Your heavy launcher's first stage should not be using RCS for steering or stability. You should be using gimbals or (if you are using fixed engines) winglets. You'll save a lot of weight by getting rid of all that RCS hardware and fuel. Usually all you need is one mainsail on the center stack with the gimbal on and that will be plenty of steering authority. If it's wobbly you need struts, not more wobbly bits.

There would be significant tides from Vall and Tylo. Tides from Jool are fixed and would not cause variation.

It's 1.7 metric tons of lead. Kerbface is right. Much easier to bury the base under the sand or use water-filled walls or place the base underwater. It wouldn't have to be very deep. The underwater plan is probably best, since you can send submarines to go explore the entire oceans of Laythe without having to expose Kerbals to radiation at the surface. Speaking of that radiation. Holy crap! Wiki lists the dose on Io's surface at 36 Sieverts PER DAY! That means you could only stay on the surface unprotected for 40 minutes before you'd get acute radiation sickness and maybe die, not to mention the risk of cancer. That's like standing next to the Chernobyl reactor after it exploded for over half an hour! So how do we deal with this? The options I can think of are: Robotics - Don't deal with it. Put a station or base on Vall or Tylo or Bop or Pol and remote control assets on Laythe's surface Separate crew from construction - build the base robotically, then when it's ready, simply land a small, heavily shielded descent vehicle and get under the waves as quickly as possible. Laythe has a thick enough atmosphere that landing shouldn't require much thrust.

When doing it this way (which is the very best way, btw), just hit tab until your camera is focused on your target body and look at which end of the hyperbolic encounter orbit has the "Jool/Duna/Eve/Whatever escape" icon. Your craft will come from the other end and travel to this end as it passes through, so that will show you the direction of your orbit. Prograde orbits are counterclockwise when viewed from above. Use radial burns to switch which side of the body you're arriving on! Good luck!

No, you're actually spot on. I made a small error in the Hill Sphere radius calculations of my post (it shouldn't affect the final numbers because of the tolerances I used), but if I hadn't made that error I would have noticed that this orbit passes through the L1 and L2 points roughly, so you wouldn't even expect to see a single orbit completed in that configuration.

One thing to point out: You could not have the super-Earth moon tidally locked to both the grandmoon and the grandplanet (that's what I'm calling them because it's adorable ). Setting the orbital periods of the grandmoon around the moon and the moon around the grandplanet equal, you can see that the semi-major axis of the grandmoon's orbit is equal to the Hill-Sphere radius of the moon, so it would not be stable no matter how far away the moon is from the grandplanet. The moon can be locked to the grandmoon (likely), the grandplanet (very unlikely), or neither (unlikely), but never both (impossible). I think the most likely situation is that you have a very small grandmoon relatively close to the moon, which is way the hell away from the grandplanet but still comfortably in the sphere of influence. That means that we would like a huge star to orbit around, but you wouldn't want to be orbiting a blue supergiant, either, because those last only millions of years. Actually you must have a star very close to the Sun's mass or smaller, because the lifetime of a star drops off as the -5/2 power of the mass, and the Sun is probably only about twice as long-lived as it needs to be to develop complex life. That means tidal effects from the star will be hard to avoid if we have to stay in the habitable zone, unfortunately. Now we're talking about something approaching a binary star system, with the Sun and a big brown dwarf for the near-partner of the star at roughly Mars' orbit (that's about as far out as you can go and still keep warm with the greenhouse effect). Take an object on the border between Super-Jupiters and brown dwarfs (13 Jupiter masses) and put it at Mars' orbit, and you get a Hill Sphere of about 0.35 AU, or a SoI of about 0.26 AU. Put a 5 Earth mass moon at 0.15 AU and it will have an SoI of something like 4 times the Earth-Moon distance. You could have a Mars-sized grandmoon hiding in there, I think, but just barely, because the Roche limit (distance below which gravitational tide will disrupt the grandmoon) for a 5 Earth mass moon is about 1/4 the Earth-Moon distance. It may seem like a lot of room but it isn't. The fact that these numbers are in the same ballpark means we'll probably have to deal with tidal heating and the surface will be very... interesting. My potentially feasible (I think?) system: 1 solar mass star - call it Lol 13 Jupiter mass grandplanet/brown dwarf at Mars orbital radius - call it Jupiderp 5 Earth mass moon at 0.15 AU from the planet - call it Herpth Mars-sized grandmoon at the Earth-Moon distance - call it Goldilocks I would imagine that Goldilocks and Herpth would be mutually tidally locked in nearly circular orbit, just because if they weren't, Goldilocks' porridge would be too hot (tidal heating = lotsa volcanoes), and also the drag on the rotation of the bodies would affect the orbital radius like our Earth's fast rotation is pushing the Moon farther away from us. Since we have a very narrow window for a stable orbit, that can't be going on very much for very long. This gives an orbital period of about 12 to 13 of our Earth days, which would also be the day/night cycle on Goldilocks. You would want a slightly elliptic orbit in order to encourage some tidal heating to keep Goldilocks' porridge hot enough to maintain plate tectonics and an atmosphere thick enough to keep the surface habitable. All of these bodies are going to be pretty unique. There won't be a system of large, Herpth-sized moons around Jupiderp, because they would perturb Goldilocks. Similarly, Jupiderp will be the only large planet anywhere near its orbit, since Herpth and Goldilocks are quite far out in the SoI. How did this form? I think a not-ludicrous story would be that Herpth and Jupiderp formed around Lol, Herpth was captured by Jupiderp and smashed into a large moon in the process of clearing out most of the other moons of Jupiderp. Think Triton's arrival at Neptune. The impact carved off Goldilocks into the perfect region for stability and also happened to keep Herpth's rotation slow enough that Goldilock didn't slowly spiral out of its SoI before tidal locking was mutual. Well that was fun! I apologize if this isn't the most well-organized thing I've ever written, but it's late and I'm kind of riffing. I hope this helps! There are other configurations you could use, but this is kind of a middle-of the road approach. For example, you could use a much bigger star and therefore give yourself a lot more room to play with if you don't need life to have evolved on this planet independently. If it's a colony of an advanced race, the the few million years that it'll be around would be plenty of time to establish a permanent base provided something could be done about the wild volcanic activity on the young grandmoon.

Here's my version of what you're proposing. Care to help out? http://forum.kerbalspaceprogram.com/showthread.php/32194-MetaChallenge-Thread

0.02c Dang.

On Minmus you can literally get to orbit and land again with EVA to spare. On the Mun you need just a tiny bit more fuel to get to orbit. It's so close it's almost like the devs purposefully gave us 95% of the propellant necessary to frustrate us. :-) Let's do a ballpark estimate: Assume instantaneous maneuvers for launch and landing, a flat ground, and constant gravity (I don't expect to be reaching orbital speeds if I have to turn around and come back). This will also obviously ignore surface rotation. From high school physics, the best launch angle is 45 degrees, so let's assume the kerbal uses 25% of his dV to launch himself at a 45 degree angle to the surface. The EVA packs have roughly 550 m/s of dV (correct me if I'm wrong, here), so that means a speed of 137.5 m/s initially. If you do the math, you'll see that means a loft time of almost exactly 2 minutes, which translates to a distance traveled of about 11.6 kilometers. Call it 12. Therefore, in the perfect scenario, the kerbal launches himself at exactly 45 degrees and performs an exactly opposite maneuver (suicide burn) upon landing, then repeats the same process for the return trip. Really, you can't do this because you don't have perfect control over your thrust angle, you have finite burns (Oberth effect!), and you can't judge when to do your landing burn accurately enough to do a real suicide burn (not if you're being even mildly safe). So I wouldn't expect anyone to be able to do a jump and return of much more than 12 km without special circumstances or a very clever exploitation of the extra physics I omitted. Based on what I've read here, that seems pretty close to what people are able to do, so I wouldn't plan on any 50 km return trips. That's why we have rovers. And rockets.

Sometimes you need MOAR thrust. Even in space. Just because you're in space doesn't mean that suddenly you don't care about thrust at all. Having a higher thrust allows you to better approximate impulse burns which are usually more efficient than finite-lenght burns. For example, on an outbound interplanetary transfer, if you have a thirty minute burn, that's about a full Kerbin orbit! Obviously you're going to waste a ton of fuel burning at weird angles that you wouldn't have to waste if you brought more thrust. Therefore, there is still a balance between Isp and acceleration, even in space! The way to use low thrust in space would be to do small burns at the point in your orbit where you want your periapsis to be each time you go around. This will take hours or days real time, and even if you had 10x physical warp, it wouldn't help much because you'd spend most of your time making sure everything was lined up and starting and stopping the non-physical warp at the right times. Additionally, you can't be sure that you won't hit the Mun's SoI by chance and screw up your orbit! This is why even NASA designs large transfer stages to Mars with multiple engines despite the mass penalty. You save fuel on big burns, you don't have to slowly spiral away from Earth and spend weeks in the Van Allen belts, and there is much less chance of something going wrong (imagine losing an engine halfway through this process. Bummer.). So in short, what I'm saying is: That's space, bro.

Consider: If you have a rocket visibly wobbling in KSP, maybe no big deal. If a real life rocket is visibly wobbling, and sometimes even if it's wobbling imperceptibly, you're going to kill everyone. Kerbal ships are designed to withstand the stress of relatively high speed, poorly-aligned docking. The Space Shuttle weighed over 100 tons fully loaded in space and docked at about 0.03m/s, perfectly aligned. If you come screaming in (relatively speaking) to dock at 0.5 m/s in real life and you aren't perfectly aligned, you're going to kill everyone. Kerbal engines are designed to support the weight of medium-to-large-sized rockets on their bells while awaiting launch in case the engineers forget to include docking clamps in the design. If you drop a real life rocket on its engine on the pad, you're going to kill everyone. The "reason" could be that the kerbals design their structures to be much sturdier because they can't trust their pilots to fly the things with the right precision and they can't trust their engineers not to design wobbly, floppy rockets. Obviously the real reason is that if you had parts with real life mass-ratios you could get to space trivially easy, even with Kerbin's "pease porridge" atmospheric drag physics, but this should make you feel a little less swindled in the technological capabilities department.