Bunsen

The trouble with using thrusters for fine attitude control is that every little tiny twitch expends propellant. When you're keeping something pointed with a precision that's right down near the attitude sensor noise level, you inevitably do a lot of back-and-forth and no-wait-back-the-other-way actuation, and that bleeds off propellant rather quickly. Reaction wheels and CMGs only burn electricity to do that. But you always need thrusters (unless you can get away with magnetorquers, but we're talking about beyond-LEO missions here) to handle net torques, and there's a definite advantage to using finely throttleable thrusters so that you can burn propellant as a backup in case the wheels fail. That backup plan won't buy a ton of time, though, because those thrusters or their propellant supply always have a limited lifespan. I'm not sure how much benefit you actually get from having that contingency option, nor how the cost balance plays out between hydrazine and electric thrusters. As for LISA Pathfinder, that requires fine thruster control because it's compensating for disturbance forces, not just torques. I'm guessing that attitude control imposes a minor increase in the delta-v requirement compared to canceling out radiation pressure, so it was simpler to do attitude control with the thrusters. It's also possible that a wheel-based system would have created too much vibration for that mission.

Look at a map. I usually use the sky chart generator on heavens-above.com (tell it where you are and what time you want the chart to show, and it gives you something like this), but I'm sure there are a hundred other websites or apps that would accomplish the same thing. You'll get to know your way around the sky after a while.

It's not just that the propellants themselves are cheap -- propellant is such a small fraction of the total launch cost that they could spend ten times what they do and hardly affect the price of a launch (with the pricing I could most easily find, a $5*107 launch burns about $2*105 worth of propellant, maybe $4*105 now with petroleum and energy price increases). Using old, established technology means most of the expensive, time consuming mistakes have already been made, and you can focus your engineering effort on making things light, reliable, and easy to manufacture instead of trying to figure out how to make it all work in the first place. For a young company selling its first product line, that's a much smarter thing to do than to gamble on breaking new technological ground. Conjuring new technology out of thin air is awesome work for the scientists and engineers, but it's absolute hell on budgets and schedules -- and companies live and die by budgets and schedules. Once they build a steady revenue stream, then you can expect to see more effort go toward higher-risk projects (like that methane-oxygen engine). Fuzzy Dunlop's points are also quite important. Production lines are expensive to establish and support. The benefits of a high-energy upper stage have to be weighed against not just the mass of larger, insulated tanks and the inconveniences inherent in working with liquid hydrogen, but also the costs of cultivating expertise in another technology, designing a new piece of hardware from scratch, and running a separate manufacturing line.

Okay, who's going to add the "Bareass Device" to the mission ribbon generators? I'm thinking a little (( symbol at the extreme edge.

I mostly do that, but I see a difference between "boring hard" and "interesting hard." Using mods to dodge the boring parts sometimes makes new, interesting challenges accessible without excessive tedium. And, of course, some mods make almost everything harder, but do it in interesting ways -- F.A.R. being my favorite example.

GPS works in low Earth orbit. Further out than that, I believe standard practice is to use a combination of celestial observations (planets and such) and Earth-based measurements with radar or radio transponders (sorta like radar, but the signal that goes back to Earth is amplified by electronics on the spacecraft, not just a passive reflection). Radar on board the spacecraft is normally only used for short-range work like rendezvous, because long-range radar takes rather a lot of electrical power.

The problem I have with that approach is the "Waitaminnit, did I remember to flip that port?" moment that comes after I've built a monstrosity of a booster under the thing with a bunch of struts and action groups that tend to get screwed up if you detach parts, which means that double-checking the port alignment is very much nontrivial. I know half the posters in this thread will simply respond "Well it's your fault for not looking the first time" (basically "absolutely everything is the user's fault," also known as "the Linux ethos") but there's this idea known as playability -- one which we can tell Squad cares about, because if they didn't then KSP would be Orbiter. An immersion-preserving texture change wouldn't be too hard, and KSP's asthetic leaves plenty of room for this sort of thing. If NASA can do this: then KSP can get away with a little "↑ THIS SIDE TOWARD SPACE ↑" note on the edge of the big port.

The first one is Kerbal Engineer, which I know and love despite the fact that it can get confused about staging once in a while. The second is apparently called the Subassembly Manager, which I have not used yet (though I probably should).

I'm not certain what's going wrong, but start by placing a node as close as you can get to the apoapsis marker. Drag the progade burn thingy out until you start to see the apo/periapsis markers on the forecast orbit rotate -- that means the planned orbit is getting close to circular. The closer you placed the node to the apoapsis marker, the faster the apo/peri markers will swing around and the closer to circular the resulting orbit will be. Don't go crazy trying to get it perfect. If you haven't dropped any stages since your last burn, the maneuver node's burn time will be fairly close to accurate, so use that to time your burn roughly 50% before and 50% after the exact time. The calculated burn time does assume you'll run at full throttle, but you don't have to do that 100% of the time -- it's easier to get the delta-v right if you tail off gently and carefully, and if you did 95% of the burn at full throttle then being a few seconds late with the last 5% won't throw things off by much. This may help with the "apoapsis goes crazy" syndrome, since most of the change there happens near the end of the burn. Mess around with this a few times, not necessarily right at launch launch -- if you get a 300km apoapsis, try circularizing at 300km just for the practice. You'll find that changes from one orbit to another generally require much shorter burns than it takes to get into orbit in the first place, and that can make it easier to see what's going on and figure out how it all works.

"Honey, did you know you can buy a complete LR-101 engine for about $3000?" "No, of course I didn't! I didn't even spend a tenth of that."

I clicked on this thread expecting to end up ranting about whippersnappers who don't know from nothin', and how we had proper games back in my day with none of this newfangled Steam business, why you had to dig out the floppy disk or PRESS PLAY ON TAPE, and that the important thing is that I had an onion on my belt which was the style at the time. While there are a few of you who definitely need to get off my lawn, I'm pleasantly surprised to see X-COM so well represented. And SimAnt, damn, I haven't seen that in ages. But I know I can't be the only one here who's played the greatest game ever (shut up yes it is):

In equitorial orbits, it makes for stupidly easy alignment -- point at the north or south pole on the nav ball, turn on SAS, and translate. Also, if you like to take things slowly, you never have to correct for the changing relationship between the vessel axes and nav ball. If you're still getting used to translation controls and the nav ball symbols, that little bit of simplification can really help.

And here when I saw the thread, I thought somebody else had just heard about Norton Sales. Which is an actual junkyard and spaceship parts company and a place I must visit before I die.

Like almost every true/false question, the answer is "definitely maybe." Long burns are sometimes inherently inefficient, whether due to gravity losses (as Gus pointed out) or the Oberth effect (where your optimal efficiency is only available at periapsis, but much of your burn happens at higher altitide because it takes so long -- this happens a lot with interplanetary transfer burns). And there's always human impatience, which is why we don't do all our orbital maneuvering with ion engines. As for the uselessness of the Poodle, you should take a look at what happens when you put some more payload mass on top of that tank. When the engine accounts for much of the stage's dry mass, engine mass has a huge influence on delta-v and the light engines look great (that's why the 24-77s look so good in your list -- they're light as hell, despite having crap Isp). But when there's a bunch of payload also, the engine mass has a smaller impact and the Poodle's high Isp often wins out over the T30's low mass -- and the 24-77s will look utterly terrible.

This is a worthwhile feature request, I think. I'd love to be able to mess with the in-game playlists and the events that trigger changes, like starting space music when you cross 70km. Cue the Strauss when entering docking mode, for example, or the other Strauss just before sunrise. I have a certain fondness for the Star Control II hyperspace music during high time warp, too.

Practiced proximity operations a little as soon as RCS existed. Didn't actually get much good at it. Persistence came into the game, and I decided to start catching and deorbiting debris. Spent a while with a calculator, paper, and the Wikipedia page on orbital mechanics to figure out rendezvous. Skimmed a Ph.D. thesis by some MIT student named Edwin Aldrin, which didn't turn out to be very useful but was kinda cool. Can't tell you how happy I was once the nonrotating reference frame view of ellipses and the rotating frame view of centrifugal/Coriolis forces finally clicked with each other. Got rendezvous figured out. Then they added maneuver nodes and target indicators on the nav ball, and it got easy. Practical proximity operations and docking were another matter. I think my trial and error perioid caused a worldwide shortage of monopropellant, but I got it down eventually. The target indicator on the nav ball also helped this immensely. Learned the importance of RCS balancing and lights.

Yeah. It's a pain in the ass to start it right, and the standard KSP aerodynamics make it not work all that well unless you're headed to a high orbit. It's much more useful if you're running FAR, where you don't have to be quite so scared of reaching high speeds at low altitudes.

The pressure drop across the tube is going to disappear under the inaccuracies of ideal gas approximations when your working fluid is stored as a liquid and gets expelled as a solid/gas mixture. I think energy balancing is the easier way to get a decent approximation, but taking the phase changes (one of them incomplete) into account.

http://www.youtube.com/watch?v=5l3oUUUzOxE

Cool video, but that's also electrostatics, not magnets. Some similar physics applies, particularly since the attractive force is mostly due to the water droplet being polarized by the rod's electric field, which gives you 1/r>=3 behavior for the force (depending on whether it's near a uniformly-charged section of rod, or closer to the end or some other nonuniformity) and unstable orbits. You don't get to see the alignment dependence that magnets would give you, though. It would have been cool if they had also done it with a strongly charged droplet (which you could get just by squirting it out of a grounded conductive tube near the charged rod). You'd be able to get stable orbits that way, and see lots of strange orbital mechanics that we don't usually see because all our central bodies are more or less spherical.

Can we try focusing on the physics thing here? There's a perfectly appropriate place for useless debates about who should be how grammatically correct in which language, and it's about three forums that way: ↓

Short answer: I don't think so. It's possible for certain mathematically ideal starting conditions (particular alignment of the magnet's field and a perfectly circular orbit), but the one case I know of is unstable and I suspect the other options are worse. The big difference between this and a gravitational or electrostatic orbit is the fact that magnetic dipole fields fall off as 1/r3, not 1/r2, and the force on a dipole is not proportional to the field strength. Two permanent dipoles (think two fridge magnets) will exert a force on each other proportional to 1/r4. A magnetic, but not permanently magnetized, object (think a little iron ball near a permanent magnet) feels a force that falls off even faster -- 1/r7 if you approximate its magnetization as linear in the external field. That force also depends on how the magnetic moment is aligned relative to the external field -- it's attractive if they're parallel, repulsive if they're antiparallel, and zero if they're perpendicular. Even if we take the simplest version of this situation -- two permanent magnets, aligned so as to produce maximum attraction (and zero torque), starting in a perfectly circular orbit so the magnetic force provides the centripetal acceleration -- it falls apart as soon as you breathe on it. The perfectly circular orbit works, but the tiniest deviation from circularity causes the orbit's eccentricity to grow exponentially (it also quits being elliptical, so eccentricity is a crappy term to use, but whatever), until the periapsis goes low enough for a collision. That happens for any central force that falls off as 1/r3 or faster, and we have 1/r4. Here's a reasonable description of why if you're not afraid of calculus. The more complicated versions, where the magnets don't stay at a fixed relative alignment, are probably unstable in more complicated ways. Rotation of the magnets comes into play, because they exert large torques on each other and can exchange spin angular momentum for orbital angular momentum, and now you have more degrees of freedom to worry about. Somebody with too much time on his hands could probably solve for a few example cases, but I'm not that bored right now. That's not true. The force on a magnetic dipole isn't along the field -- it's along the gradient of the field. A magnet that's aligned parallel to an external field will be attracted to regions where that field is stronger. If it's aligned antiparallel to the external field, it'll be repelled from strong fields. If it's perpendicular to the external field, it will feel no net force. Really? I've got some little chunks of neodymium/iron/boron mix on my desk here whose magnetic interactions, when they're close together, are 12 or 13 orders of magnitude stronger than their mutual gravitational interactions. It all depends on the scale. That's electrostatics, not magnetics. It's still (mostly) monopole fields, so it follows very nearly the same math as gravitation. It's interesting to see how far it deviates from ideal 1/r2 behavior due to some fairly small influences, though.

Roscosmos is finally confirming what we've been reading on RussianSpaceWeb's increasingly detailed account of the angular velocity sensor issue. It sounds like installing them upside down was impossible without damaging the units, but somebody managed to flip them over and destructively jam them into place anyway. I'm sure there will be some very unpleasant company seminars and more layers of documentation and verification in the future, but it still means that somebody thought the hammer-and-crowbar approach to assembly was appropriate for rocket avionics, and was so comfortable with it that he never said "Man, I really had to bash that sucker good to make it fit. Is it supposed to be like that?" to anybody. Or, worse yet, did ask that and was told "Yeah, happens all the time with that subcontractor's parts. Go ahead and bash the rest in too" or "It's 4:30 on Friday and I can't go home until you're done. Less talk, more bash." That's got to give potential customers some second thoughts.

And here I thought that the current Kerbals, what with their general disregard for safety, affection for the brute force approach, and tendency to answer any question by adding more engines, were the Russian equivalents.

If you're not into calculus, there's a simpler route. You already know that traveling at a steady speed v for a time t sends you a distance of vt. If you graph speed vs. time, a constant speed for a certain span of time defines a rectangle (shaded). The distance traveled is the area of that rectangle. If you accelerate steadily, then the graph of velocity vs. time is a slanted line, and if you start with v=0 then it describes a triangle. The area of a triangle is 1/2 base*height, and this has a base of t and a height of at. QED. You can extend this to what happens if you start with a nonzero velocity, too. The area under that line can be divided into a rectangle (with an area of initial velocity * t) and a triangle (1/2 at2 again). Calculus becomes necessary if you want to find the areas under more complicated curves.