EzinX

Use a droplet radiator. Instead of losing coolant permanently, recollect it. http://www.5596.org/cgi-bin/dropletradiator.php VFX wise, a droplet radiator is a foggy mist in a 2d plane. It's perfectly 2d - the electrostatic guidance of the droplets at either end would force it into such a plane. The droplets are too small to be seen, so too small for a particle effect. There's absolutely no reason to vent coolant permanently because the droplet radiator can have vast surface area basically for free - the systems would be about the same mass whether the droplet area were 100 meters long or multiple kilometers long. (A really long droplet system would probably need intermediate electrostatic booms that stick out and smooth the path of the droplets as they travel, but to first approximations it's the same) At 1000 kelvin, Tin coolant does have a vapor pressure, so you are losing coolant over time. For an interstellar spacecraft, you would probably need some kind of exotic system that perhaps uses solid ceramic nanoscale spheres or something instead of droplets so that no vapor is lost. Also, at high temperatures, you get enormously more performance - black body radiation is proportional to T^4. Once you are talking 1000 kelvin or more, raising the temperature to the fourth power gets you crazy high heat rejection rates. For instance, using that calculator, and 1000 Kelvin for the coolant temperature, a 100 x 1 kilometer droplet radiator radiates 155 gigawatts. You'd have a pair of them, for about 310 gigawatts of capacity. Even with crazy high performance gas core reactors...how much power did you have in mind? The thing is, your problem is that if you are doing nuclear thermal, the velocity of the exiting particles is determined by the temperature that the physical materials your rocket motor nozzle can tolerate. If you are doing nuclear electric, the mass of your radiator is only one component. You also have the mass of the generators, the mass of the heat engine, the mass of the core, the control systems, and the mass for the electric engine. This is why when you sum it all up, you get disappointing levels of thrust and you have to run the engine for the entire journey and still take years.

Yeah. The Rift and comparable devices do have a couple advantages : 1. True stereoscopic display with no ghosting (bleed from left to right, which you get with every other method for 3d) 2. Perfect head tracking. This is important because one of the reasons those flight simulators require many computers to drive them is that you have to render every perspective from every window on the aircraft. If you know what the user is actually looking at, you only have to update what the user can actually see. We could develop a form of virtual desktop that would work the same way. Instead of hiding windows behind other windows the way we do it now, you would have everything visible, all the time, in a 360 degree virtual space around the user. In the event that a user has so many tabs and windows open that there isn't enough room to show them all, you would make the less important tabs smaller. You only update the windows inside the field of view of the VR headset user. But, if you're wearing a headset, how do you see your hands and the switches? The way to make this work is a headset that projects the image right onto your retina, overlaid on the real world. As you move your head around, there would be black construction paper or something in the cockpit windows, and the headset would display the stereoscopic image of what is visible outside the cockpit beamed right onto your retina. You would track the user's head position with fixed cameras inside the cockpit aimed at the trainees, that would be looking at marker LEDs on the outside of the headset. The actual Rift could be used for the monitoring stations - the instructors who watch the pilots would wear rifts which would display the feed from cameras inside the cockpit (so you can see what the trainee hands are doing) and the stereoscopic view of the outside world.

Ok, as I understand it, black hole time dilation is only relative to external observers. If you are riding a spacecraft into a black hole, from your perspective you are approaching a supermassive object and getting ripped to shreds in a few minutes or hours (depending on your approach speed) From the perspective of external observers, it takes years for you to get shredded to pieces. If the black hole fails to shred you because godlike entities have tampered with it...well...the same argument still applies, no matter how unlikely the existence of said entities may be.

I read that straight LNG is fine for rockets. If you read Ignition, you'll see when he gets to describing the "standards" for RP-1 : they are some of the laxest standards for fuel imaginable. They simple restrict certain hydrocarbon chains that tend to foul up the rocket motors. The exact fuel composition is up to the manufacturer and can vary massively, otherwise. Using straight LNG with a few dissolved gas impurities is not a big deal.

For those who didn't learn chemistry in school : methane is not really any more polluting than using hydrogen. You can produce methane purely synthetically - you electrolyze water to get hydrogen, but you don't try to actually distribute the hydrogen. It takes up way too much volume in a tank, and it leaks through virtually any seal. You follow up electrolysis with the Sabatier reaction to get methane. The CO2 could have come from the atmosphere, giving you zero net CO2 emission. Or, the CO2 could have been captured from the smokestack of a coal power plant. Using raw hydrogen has never, ever been a good idea. I guess the people pushing it do it because the general public doesn't realize this? Given that this method is much less energy efficient than batteries, it would only be appropriate for vehicles that cannot run on batteries - such as airplanes and possibly long haul trucks. The other half of the solution is better batteries. Lithium-iron chemistry batteries use raw materials that are far cheaper than the assembled batteries cost, so their prices could be brought down by a factor of 5-10, probably. (the lithium-cobalt batteries we use in laptops, etc today are only a few pennies more than the cost of the cobalt, so they won't get any cheaper). Right now, it's about $30,000 for the lithium iron batteries equivalent to what's in a Tesla, bought as an individual. A 5 fold cost reduction would make electric cars affordable for everyone.

Lasers. Cheap and practical. There's a paper somewhere with the proposal - you don't push stuff into a higher orbit, you want to adjust the orbit the stuff is on so that part of it scrapes the atmosphere and friction does the rest. The lasers would be fired from the ground, and would not actually be all that big or expensive.

Basically no one here wants to mention the obvious. Production of compressed hydrogen gas is not even 50% efficient, starting with electricity. And then the fuel cells that burn it are only about 50% efficient as well. So no more than a quarter of the energy becomes electricity that can be used to drive the car. With batteries, the electric grid itself is about 90% efficient, and so are batteries. About 80% of the electricity is available to drive the car. (the rest of the drivetrain should be about the same efficiency whether the car is battery or fuel cell) Fuel cell is just an expensive version of a fuel driven generator. Well, you can get hydrogen from hydrocarbons. Once it's gas or liquid hydrogen, it takes an insane amount of volume in the tanks, and it is extremely leak-prone. Or, you can convert longer chained hydrocarbons to methane. CH4 burns in fuel cells, regular engines, diesel engines, turbines...everything. Liquid methane has good volumetric density and better energy per kilogram than gasoline and diesel. We could switch airliners to using liquid methane, as well as trains, trucks, etc.

Also, your simple analysis is missing crucial details that would require research. Read the relevant chapter in Ignition! . I can dig it out, but simply put, you need to know the actual ratio of products produced in a real rocket engine, specific to the conditions actually encountered. You don't simply get H20 with a LH2/LOX rocket, nor do you only get CO2 and H20 with a methane/LOX rocket. It's easier to just use the results of numbers others have already worked, like just saying "the ISP is 80% of the ISP with H2/LOX"

Pretty sure your numbers are wrong. ISP takes into account all of this, and with liquid methane, it's not that much worse than liquid hydrogen. The reason why would take a whole reacting species analysis, I don't have a way to disprove your stoichiometric ratios off the top of my head. In short, though, that extra mass of oxygen becomes a combustion product heading out the nozzle at high speed. It contributes to thrust. The average velocity of the exhaust products is 3816 for LOX/H2 and 3034 for CH4/H2. So the methane is 80% as good as H2.

http://www.ibtimes.co.uk/carbon3d-amazing-new-3d-printing-technique-100-times-faster-using-light-oxygen-1492315 To summarize : instead of moving a print head over the entire surface one layer at a time, they use a UV sensitive plastic. The membrane they are firing the UV through is oxygen permeable, and the plastic doesn't cure where oxygen is present, create sharp boundaries between the printed regions. This is brilliant. It eliminates most of the flaws I have seen with plastic printed parts - they have these clearly visible seams between layers, and a rough surface. Parts out of this machine are continuous, mathematically smooth contours. It's much, much faster - printing with a 2d image means it takes a lot less time. And it's much stronger - a continuous process can be optimized for maximum mechanical strength. A "layered" process can't because there is variance in the layers. So, how can this be done with metal? You could use several scanning beam projectors to create a 2d laser image on the metal surface. You use several so the beams overlap and create a smoother, higher resolution image. With quality optics, there would be enough beam intensity to melt metal powder. However, how do you get new pieces of metal powder to the surface in a continuous manner? You can't have the metal in a liquid form because heating it would only make it more molten. If you try to use liquid as a carrier, you are contaminating the metal with whatever the liquid is. The only thing that comes to mind is if the pieces of metal are extremely tiny - smaller than the wavelength of laser light - the light would diffract around them. So, in a vacuum chamber, you would be spraying this continuous mist of metal pieces towards a surface. The laser beams would shine right through the mist to the surface, continually adding features. Anyone have a better idea?

You don't have to do it that way. You could extend the walls of the inner tank to the bottom, and actually use the inner tank as the primary load bearing structure. There could be all kinds of internal structural pieces, as well. I suspect it could be made lighter than 2 tanks still because of this sharing.

Got a source for that? I am skeptical : with tank in tank, you have shorter plumbing runs (because the bottom of both tanks are at the bottom of the stage near the turbopumps), less stress across the walls of the inner oxygen tank (because it is being pressed inward from pressure in the methane tank - you would tune the pressures so the methane is at a higher pressure), and so on. Pretty sure fundamental physics make tank-in-tank smaller if the design is optimized to theoretical limits for available materials.

No. The thermocouple would be dual purpose, both structural and to reverse the normal direction of heat flow.

That's a clever solution. Lighter than my idea of stirring rods, also simpler. The tank wall itself would be acting like the thermocouple. You'd probably be able to stop running it at liftoff - all the vibration would probably prevent any significant amount of methane freezing. Or, if you had sufficient power once in orbit, you could keep running it, as well as a cryocooler on the outer tank to keep the methane cold.

SpaceX can probably just order pure methane in bottles - NASA got it somehow. More expensive, but probably still a drop in the bucket compare to the other costs. You can use methane in a fuel cell just as easily as hydrogen. Possibly more easily. - - - Updated - - - Correct. However, you could insulate the inner oxygen tank a little bit, and keep the methane in the outer tank always moving with something to stir it so it doesn't freeze against the sides.

Since you have to have a LOX tank anyway, couldn't you do the same thing they do for storing liquid helium, and make the lox tank inside a larger tank that holds the liquid methane? This would let you make the lox tank walls much thinner, as the cold methane layer would act to keep the lox cold. You'd have recirculation or something to prevent any of the methane from actually freezing. A bigger cryogenic tank that essentially stores both components to fuel in the same assembly would require a lot less insulation.

So, ok, there is this chart on wikipedia. If you look, liquid natural gas (liquid methane) has better MJ/kilogram than kerosene but it is not as good as liquid hydrogen. It's in between. Similarly, it has better MJ/liter than hydrogen, but worse than kerosene. Similarly, it's harder to store than kerosene but easier than hydrogen. I'm unable to find specific numbers on what dry mass ratios look like, but I would expect intermediate numbers. So, how come, way back when, they did this : http://www.reactionengines.co.uk/images/saturnv/Satirn-V-1024-cut.jpg Instead of 2 fuels, why not use an intermediate fuel for the upper and lower stages, getting almost as good a performance, but being able to use the same kind of tanks and the same engines for both rocket stages. I'm guessing I'm missing a subtlety here, and that it must in fact be worse than either fuel, and the only reason SpaceX wants to use it is because of Mars ISRU, and they want to reuse lower stages easily, and methane doesn't leave residue behind.

How much does this really help? You need equipment to do experiments, water and food and oxygen for the astronauts, rocket fuel : all this stuff has the same mass independent of internal volume. Did NASA launch entire modules for the ISS solely for the purpose of boosting interior space?

Well, I just wanted to know why no one yet does it that way. Because if this idea does work, I've got some contacts at NASA.... Amusingly, SpaceX actually intends a form of Asparagus staging for the Falcon Heavy. Some of the methods that work in KSP work IRL.

So...why does this even work? What is KSP doing wrong in their simulation that makes this a viable strategy? Let's see. Fuel tanks are too heavy empty compared to IRL tanks. Pumps that let you pump that much fuel vertically as fast as you burn it use 0 energy and add 0 weight. Gyroscopes provide too much torque, allowing this monster structure to be controlled. Structural stresses are probably unrealistically low. Any major reason I missed?

Everything I've ever heard or read says the Russians spend 1/10 the cost to develop almost the same capabilities. The reasons for it are myriad 1. Russians work for less 2. Everyone working for the Russian space agencies is generally working directly for the government 3. NASA subcontractors have an economic incentive to charge as much money as they possibly can, because they are guaranteed 10% of the tab or so as profit. One way to charge more money is to subcontract the work to another subcontractor, because they are going to charge more than simply sending the receipts for your materials and labor to NASA. 4. Congress has an incentive to encourage #3 5. If NASA gets someone killed, they face losing everything. Thus, they are enormous incentives to not lose crew. 6. The only way to lose a government job is to do something that makes your or the agency look bad. See #5. Spending extra money is ok. 7. You can get a 90% as good solution to virtually any problem spending a lot less money. 8. Sometimes volume makes up for inefficiency - see the various Soyuz launches. And so on and so forth. I fully believe this. Heck, SpaceX could get a station in orbit for 4.4 billion, with money to spare. That's what, 4 falcon heavy launches, where the falcon heavy is just 3 of their existing rocket cores bolted together? So they need to develop a station to launch. I bet Bigelow aerospace could give them something that would work, cheap.

This is what would kill my ships. I would need to do a long burn of 20-40 IRL minutes. Trouble is, especially with physical timewarp running, little pertubations would get started in one section of the ship. The oscillations would be amplified by the docking port to the other piece. SAS tends to try to correct for this in a way that actually exacerbates the problem, making the ship oscillate even more wildly. Docking ports are treated like springs by the physics solver, so energy that never existed is being injected into the system, making it even worse. Mechjeb's autopilot is also more aggressive, making this problem worse - and in my playthrough, I consider it completely legit to use mechjeb to calculate a correct transfer trajectory and to execute planned burns. Anyways, I found that if I cut back the thrust and was willing to wait 2 hours or something to do a burn at 1x, sure, I could do an interplanetary burn. It was just not worth it to me to wait that long, I quit my campaign in the hopes that future patches would fix this. I feel the implementation of physics in this game is wrong. Pieces of ships should be tied together by rigid body transforms, and the game should only compute stress analytically and not attempt to apply it using springs using inaccurate integrations. At the force summing stage of each timestep, the game should compute all of the uncontrollable forces on a ship, check the ship's intent (basically there would be a number of possible flight modes that reflect what the ship is trying to do), and then cancel all unwanted forces using the ship's engines and control surfaces, limited of course by their maximum available torques and thrusts at this timestep. You simulate the fact that rocket engines can't increase in thrust instantly and control surfaces can't deflect instantly by making their maximums reflect the average availability over a delta T.

Ok. If I put a brick wall in an air stream, such that the normal from the wall surface is aligned with the incident air direction and then stick my winglets behind that brick wall, the amount of air that reaches the winglets is negligible. Mentally, I see winglets as really small occluding walls that block air to things behind them.

You're saying they will work even if the winglet in front completely occludes the one behind?

So, Ferram. Or anybody, really. You've got a big chunk of source code that does all the work. However, what isn't there is any documentation as to what any of it is for. Skimming the code, I saw a lot of methods that looked quite familiar from books on various 3d math, but what I don't know is, what method or technique is tying it all together. What's the big idea here? Does a book or papers exist that describe the modeling technique you're using? Thing is, you can get Autodesk or other tools to tell you quite accurate numbers for parts in isolation. But how do you modify these calculated drags to take into account adjacent parts? That, to me, seems to be the fundamental problem that your code is solving over many thousands of lines - but how does the basic algorithm even work? Surely you're not approximating navier stokes in realtime... Also, instabilities. How's it done? Moreover, you've got a problem that the physics model of this game does not update as quickly as the control fins on, say, a fly by wire aircraft (200 hz or so). This means that a user may get a large torque applied to their craft due to instabilities that a real aircraft could correct for - but the game's physics model doesn't even inform them until it's already integrated that torque over a timestep.