wumpus

Members
  • Content Count

    2,674
  • Joined

  • Last visited

Community Reputation

943 Excellent

1 Follower

About wumpus

  • Rank
    Spacecraft Engineer

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. Used to be the rocket-building interface. The struts still seem to be irrepairably broken, but between autostrut and rigid parts (which I think are stock, I haven't updated those mods in multiple revisions) struts no longer seem critical. Mostly now it is just the way career mode encourages grind. The rocket-building interface (especially if struts are required) has been really *bad* over the years. I felt betrayed when Havester "rewrote the UI" and it turned out to be the inflight UI and not the rocket building one. Not that the inflight improvement wasn't needed and helpful, but it wasn't the UI that I felt was really needed.
  2. Wiki claims the US Navy's Ford Class Carrier program costs $37G to design and $13G per carrier, although it relies on a military industrial complex assuming that such carriers are as critical to homeland defense [snip]. I strongly suspect you could build an Orion with a budget like that, and an Orion would qualify as a "battlestar" by any 21st century definition (even if it is a mid-20th century design). I also think an Orion class "battlestar" could project force significantly better than any aircraft carrier or fleets of carriers (although it wouldn't be able to power stricken cities with electricity the way carriers can). Getting it up to orbit would obviously have political and diplomatic issues, to put it mildly.
  3. On the other hand, the Dawn probe had 10km/s delta-v in its ion engine for use *after* reaching escape velocity. If you have the patience, ion thrusters are absolutely amazing (but I'd expect that you still want to use fancy gravity tricks: ions will simply get you in position for them faster). If your patience depends on the cargo, you can practically break the rocket equation by using ions to move propellant to a depot in the right orbit, and then simply move your chemical-powered crewed rocket from ion-fed depot to ion-fed depot: LEO to escape velocity is ~3000m/s, while going to Mars is another ~1000m/s (and presumably send any return propellant not ISU-generated on the surface via ions as well). Electric propulsion with Isp>1000s is a thing (generally speaking Isp isn't the limiting factor with ion propulsion, if you want more you can probably get it at a price). Remember Robert Heinlein's statement that "LEO is halfway to anywhere": with ions (and especially ion-fed depots), LEO is nearly all the way there (no known way to use electric propulsion to orbit, not to say Escape Dynamics didn't try).
  4. Second hit via duckduckgo: https://atsushisakai.github.io/PythonRobotics/ A quick look at the descriptions of the sample code indicates that robotics is hard, and even python has a hard time making it easy. Of course, you probably don't want most of the hard issues of robotics (determining the local neighborhood visually and navigating around it, or where that car door is that you need to weld) and just have to deal with things that are mostly seen by mechjeb. Accessing the I/O functions are going to be specific to the hardware and "robotics" probably won't be a good search term. Just go through "python", [insert platforms of choice here], and maybe "gpio" (gpio your user controlled output pins). Also look into whatever PID routines you need (developing the parameters is probably much more complicated than writing the code, so don't worry too much about python modules here). As far as Python/C++/C/Java goes, don't be too surprised if you eventually need to drop all the way down to assembler (C *should* work, but you basically need to understand assembler and C to do it) to say this pin needs to be that voltage. I suspect that is what you are asking for, and it will be specific to each hardware platform (arduino, raspberry pi, gumstick, etc.). Good hunting.
  5. looks like the tried and true method is os.system("a.out") while the new hotness is subprocess.run("a.out", "-parameters"). If you are having trouble with python, I'd think twice before trying to tackle C++. It seems that C++ started with C, then added every possible feature they could to the language. Expect to need to know a lot more about the language to do relatively simple things than you ever needed to know about python. I was trying to do exactly that: take an existing C++ library and get it to run in Python. Of course to do this I had to finagle the wrapper functions to deal with the data the way python expects, so it took just a little C++. One of the gotchas was that none of the "learn C++ books" I found at the local library mentioned that the method I was using to create objects (simply typed from "Learn C++ in FIXNUM days") was putting the object (but not the pointer) on the stack (where it would quickly be destroyed), but I needed to put it on the heap instead (I should have known this from simple C programming, but my C was pretty rusty by then). Even this little thing took about a week of debugging (since I really didn't understand the issue).
  6. If "all" you want is an Earth-based SSTO, you don't need unobtanium nozzles, anything that worked with a NTR should get similar efficiency with laser propulsion. That said, Escape Dynamics (the company that tried for 5 years to do this and then went bankrupt) only managed about 500s. You'll need hydrogen reaction mass, but the Isp should get high enough for SSTO. Escape Dynamics' method of "laser propulsion" didn't involve "riding the laser", it instead heated the reaction mass, so propulsive landing (and hovering) should be possible (assuming your valve can withstand maximum temperature for hovering). The lasers are the tricky part, and can be expected to be hid behind a security wall as the US Navy develops them.
  7. Hyperfission? Basically we already split the low-hanging fruit of the nuclear reactor: splitting any more isn't clear it will add anymore power. You might also want to look up "plutonium poisoning", I think that is why current reactors pull the fuel rods. Breeder reactors were designed to allow more of the fuel to be used, that is probably what you are looking for (since the amount of nuclear fuel available never turned out to be the limiting factor, breeder reactors simply haven't been used). Earth based or space based? <RANT> Earth-based nuclear [fission] power seems unlikely in the West (especially USA and EU). Most of the reasons are political, although political/economic factors such as the nuclear power plant construction industry grew up with time&materials contracts and now appears completely incapable of developing a power plant remotely on time or schedule. Remember, these things are essentially pre-paid electricity. If you are paying interest on a debt while watching your nuclear power plant slowly being built while running into delay after delay, it is even more painful to look at (also prepaid electricity) solar farms being put up on time and schedule, and also watching the profits on such panels being pumped back into R&D making the next solar panels (which will compete with your electricity, assuming you ever go on line) much more efficient. I doubt any nuclear reactor constructed +/- 10 years will ever be profitable. *** NOTE *** This isn't suggesting Germany's abandonment of nuclear power remotely makes sense. Construction of the nuclear plant is a sunk cost and decommisioning the reactor isn't going to bring anything back. And nuclear power (especially after the sunk cost) is always going to be "more green" than any replacement over the expected life of the reactor. </RANT> Note that after sufficient decades pass it might make sense to have another go at Earth-based nuclear power (especially if using designs made off world), but the whole infrastructure behind nuclear power is hopeless. Nevermind what can be done in the lab, your technology is your infrastructure (see anything by James Burke for a detailed analysis). Spacebased ordinary nuclear power is likely sufficient, although cooling is even more of an issue and big, cheap, heatsinks (like gravity fed watertowers above the reactor) simply aren't an option off planet (maybe some liquid on Mars? Sounds difficult, but possible). Solar power (within Mars orbit or so) is going to be wildly more effective than anything on the ground, so I'd assume that engineers will go for the tried and true (solar) than the difficult and risky (non-RTG nuclear power). And don't forget that there has been one fusion-powered SSTO developed and manufacturable with 1960s tech: the Orion (not the current use of that legendary name, the real Orion.) Unfortunately early calculations ignored the magnetosphere and Van Allen belts, so most of the "fallout" would return to Earth. This could still be avoided by manufacture and launch at Antarctica, but you might still face threat of a nuclear response from India once they realize that they would have the lion's share of deaths due to a planetwide barrage of nuclear pollution (and the Chinese might as well back them on this).
  8. If the engineering solution is a parabolic nozzle, the "ideal nozzle" for vacuum is infinitely long. Thus any real vacuum nozzle is a truncated parabola. What you probably want is something that looks a lot like an additionally truncated parabola for atmospheric use and then clamp down the "only mildly truncated" rest of the nozzle for vacuum use. This would certainly not be optimal for atmospheric use, but then again no nozzle is ideal for the whole range of the atmosphere anyway. I expect my "quick, dirty, and expensive to implement" solution still isn't as efficient as the RS-25 nozzle (across all pressure levels), although maximizing vacuum Isp is absolutely critical for this type of thing. While you Ve equation looks nasty, there aren't a whole lot of parameters that define a 3-d parabola: I think at most 2, not including absolute size (note that the RS-25 doesn't use a true parabola).
  9. GAH. This could be a great physics problem. How many wrong assumptions does this question have? A(n ordinary c-traveling) photon has an energy of hv (that "v" should be a greek "nu" but I'm too lazy to look up how to type it). But that is apparently derived from the Planck constant times c (presumably changed to match the new, improved speed) divided by the wavelength. So the energy would presumably increase for a "superphoton" of the same wavelength. Of course, once it hits your eye (or camera), a curious effect happens. Your eye isn't calibrated for non-c photons. It is calibrated for standard ones, so it assumes that the wavelength (the color) of the light is inversely proportional to the energy of the photon (probably. Or perhaps your retina uses filters. I'd still expect the non-c photons to pass filters based on energy, not "real wavelength"). "Brightness" is based on number of photons, or probably more likely the total energy of all the photons. Each photon of light can only be one "brightness" dependent on the frequency of that photon: this is one of the reasons "quantum" physics is "quantized". Don't bother with "superluminal light": it makes for bad sci-fi. And of course anyone looking at said light would have "second sight" as they are looking into the future.
  10. By definition the voltage across the inductor equals the voltage across the capacitor. Likewise, once the switch is open (and not arcing), the current across the inductor equals that of the capacitor. Of course for sufficiently high frequencies, expect the "parts being equal" to effectively move at the speed of "light" (conduction inside the medium).[Edit] I'd also expect that you need to model the resistances of the wires in the classic case as well. I'd expect it to oscillate, but the conventions of schematic drawings don't always allow for the reality of HF effects. Perhaps some HF engineer can help with how the HF world interprets schematics.[/Edit]. If the inductor remains conducting, the inductance shouldn't change (until it melts) and the circuit will act like a boring old classical circuit, only now with ESR. ESR [Equivalent Series Resistance] is a spec commonly listed in capacitors indicating the amount of resistance you can expect across one. I'd assume that inductors have a similar spec, but don't remember having to bother checking one. I don't think I've used an inductor outside a switching power supply and don't really expect to. So the oscillator simply decays in a boring old fashion (assuming it is sufficiently conductive to not explode). In practice, I've heard that most superconductors are encased in copper (or similar conductive material), especially if the superconductor isn't naturally conductive. In such a case, the inductor would be completely shorted (probably to ground) and the whole think would stop dead. The whole "suddenly act classically" is dependent on being sufficiently conductive and superconductive in the same material (otherwise it would likely explode as I2R not only suddenly becomes high, but heating happens extremely *densely*, being an inductor after all).
  11. While I admit that withstanding 1-2 bar isn't in the realm of sub-mm greenhouse film, I still think that 16mm or 7.8mm is still way too big. And width is mostly irrelevant (except for the greater increase of potential weak spots, something presumably examined on the ground), just how deep the fluid is (pressurized air or water in our thought experiment). The reason you need tensile strength is gravity, simply holding the material doesn't require that at all (think more of the sides [but not edges] of a container, they still have to be strong, but not that strong. I'd suspect that a plastic trash bag could handle a lot of water as long as it was sitting on the ground). Still, I'd really hate to have to design something that takes fuel from an inflated bag and then turns it back into cryogenicly cooled fuel inside a fuel tank. Expect that process to be slow and tedious, mostly thanks to using a blackbody as a heatsink (maybe use semi-conductive mylar, and use the fuel as a blackbody? Or at least as the "coolant", although this assumes that you will always leave a significant amount of fuel/oxidizer in the tank). Either way, cooling the propellant down to cryogenic temperature will be slow.
  12. One of the first Falcon 9 flights had an engine failure. It was shut off and the rocket continued on the rest of the rockets. The "SCE to AUX" of legend is a similar thing with electric power. No idea how long a burn would have to be before replacing something in a running rocket would be viable. But replacing something in an ion engine would unlikely to be as exciting as the thread implies (and the only thing I could imagine "burning" long enough to bother replacing hot).
  13. I was guessing that the SLS would be a great candidate for this type of thing (not many rockets are going to have hydrolox engines, much less something like SSMEs designed for going from pad to orbit). Unfortunately, the SLS "first stage" can't lift itself off the pad without the SRBs. So I assumed that you would remove enough fuel to give a T/W ratio of 1.11 (weight 90% of thrust) and started running the numbers. Assuming an Isp of 450s (total flight) you get a delta-v (no payload at all) of 9200m/s. If you assume an Isp of 366s for the first minute (no altitude compensation will give you vacuum thrust, just maximum efficiency vs. backpressure) you get a delta-v of 8700m/s. Don't forget that as a fully hydrolox rocket, it will have significantly higher aero losses than typical, and with the relatively low T/W ratio gravity losses will be higher. I doubt 9200m/s will be enough (and you can't get that). But I'd assume that with even puny single segment SRBs (such as what rings many delta rockets), the thing could lift off with a nearly full fuel tank and take at least some cargo to orbit. Unfortunately, the whole thing is still going to come back down as the SSMEs can't reignite for orbital insertion. You'd need a *tiny* second stage for insertion (see shuttle maneuvering engines), so your SSTO becomes three stages pretty quickly. Then there is always the temptation to make all the stages roughly equal in delta-v... If Falcon9's booster can basically get into orbit, I'd assume that Starship booster could too although I don't think that thing is ever intended for expendable use. It should be designed for delivering more delta-v to the upper stage (than Falcon9 on a reusable flight), and could presumably burn the return fuel to get into orbit. Why you would want to do this is beyond me, as you could easily get more cargo to orbit by using a much cheaper (even expendable) second stage and still reuse the booster. With the SLS there is no worries about reusing the "reusable" RS-25 engines, but it is still a wildly inefficient means of getting to orbit.
  14. How about miniature giant space hamsters?
  15. I suspect that both the Saturn and N-1 could remain on the pad while being topped off for months. Considering the Apollo budget, I'm sure the fuel bill could have come out of petty cash. Worrying about working next to a fully fueled rocket would be significantly worse near a UR-700, as even if you survived the blast from an explosion you would have to deal with the cleanup (the size of the area in danger would be significantly larger).