Steel

I have a feeling this is a little sensationalised. The researchers most likely shut the project down because they determined that they wanted the end result to be in English. Since the AI had deviated a way from that, they had to shut it down, re-evaluate their rewards system and then start again. I don't honestly imagine that they shut it down because they were scared of it or were worried it would become out of control and start sending terminators back through time.

Ok, if it's just you or a small company behind this then it is quite difficult, the power required to ensure that any measurable trace of the signal (a video signal is very complex) would exist 3000 ly away could power a city (if not a small country) The Aricebo Message [1] is the closest thing humanity has done to this before AFAIK, and that took 1 MW of power just to send a simple binary message encoded onto a radio wave to a star to a galaxy 25,000 ly away, and it still only transmitted at 10 bits per second. So either you transmit like this and it would take several years to transmit a video (and thus several years to receive it), or transmit at a higher speed, but likely use far more power in the process. [1] https://en.wikipedia.org/wiki/Arecibo_message

I think an ablative chamber would be the way to go, by far and away the simplest option. SpaceX used it for their Kestrel 2nd stage engine on Falcon 1 (maybe not the most ringing endorsement, but shows it can be done). The AJ-10 used a chamber coated in Tungsten-Carbide from what I have read. TR-201 used on the second stage of Delta launches 1972-1988 also used an ablative chamber. Probably not possible (especially if it's hypergolic), was just spit-balling to come up with other positives. I think the same principle might apply. When releasing a high pressure gas on top of a viscous gel, its usually more energetically favourable for gas to force it's way through a tunnel in the gel than for it to force the gel downwards I guess.

Im not sure, I'd still go with HTP because it (A) is not cryogenic and (B) can still be decomposed and used as monoprop for the RCS and potentially to pressurise the tanks (not 100% whether that can be done in practice). Switching to LOX would mean having to design plumbing and tanking to deal with cryogenic temperatures as well as having to carry an additional tank of nitrogen or helium to pressurise the tanks.

Sorry to interject more of my thoughts, but I'm going to interject another of my thoughts: Since the rocket engine proposed here is quite complex, with complex burn face geometries, head pressures that need to be monitored and controlled as well as a complete lack historical data and experience to suggest whether this design would actually work or not, is it not actually easier just to design a liquid bi-prop, pressure-fed, HTP/Kerosene engine? That way you have 60 years worth of design experience, existing guides on how to go about designing all aspects of the engine, all the technical literature you could ever want and, perhaps most importantly, data and models that can give reasonably accurate performance numbers before construction and test firing?

I think this question is tough because of all the symantics involved. "Dirt" is a bit of an Americanism and (as @kerbiloid points out above) means you can end up confusing "dirt" (soil) and "dirt" (as in dirty, something unclean) which are two different concepts. "dirt" (i.e dirty) means unclean - and so has nothing to do with what causes the unclean-ness - where as "dirt" (soil) is the organic material that covers most of the ground. Most people (at least where I'm from) in the UK call it "soil", in order to avoid the confusion. With regards to the OP's question, dirt/soil (as many others above have pointed out) does not exist on Mars due to lack of organics. The surface material is called "regolith" (again, thanks, to those above) much like the moon. In regards to what you could call soil/dirt if you took it to Mars, IMO you would probably still call it "earth" because it's name is not actually a reference to where it is found, it's just an artifact of the complex evolution of the english language. You'd probably want to stick to soil/dirt to avoid confusion though.

To be completely honest, the aerodynamics of rocket like this are not really that important. All rockets are pretty similar in design, so at most it will be a difference of a few hundred ms-1 of dV. If the calculations budget in a "worst case scenario" number that's based on rockets of a similar design with a +10% safety factor, then you can't go too far wrong.

Which, unfortunately, won't cover much of the flight for a small vehicle like this. Likelihood is that it'll be supersonic in well under 30 seconds or so from liftoff ( @sevenperforce what do your initial calculations say about how quickly it might break the sound barrier?)

Provided, of course, that you can find a supersonic wind tunnel (which are somewhat difficult to come by!). Realistically, for an amateur project CFD is the best you're going to get before flight testing.

That's a good point, traditional hybrid engines use thin channels to make sure no oxidiser is too far away from the fuel grain and also to take advantage of turbulent boundary layers to improve mixing. With a chamber of the size and geometry you're suggesting a lot of the oxidiser will just being injected in and fly straight out of the nozzle without getting anywhere near a wall.

But if it has the consistency of rubber it won't flow, so the whole pressure feed concept won't work. Still interesting to know though!

@sevenperforce I've been thinking about your engine concept, and I have a few questions. 1. With this design, will you not have to very accurately control the "head pressure" so that it matches the chamber pressure? Since you have a liquid rather than solid fuel, if your chamber pressure is less than the head pressure the propellant will be forced into the combustion chamber area, reducing the size of this area and potentially stopping combustion altogether. On the ther hand, if the chamber pressure is higher than the head pressure then the propellant will be forced back up the motor, increasing the size of the chamber and the burn area, which sounds to me like it's potentially explosive. This problem will only get more challenging with the addition of throttling, where the head pressure and chamber pressure would have to change in sync to avoid any problems. 2. Can you guarantee that the burn surface of the propellant will remain in the shape you intend? With the napalm being a liquid (albeit a highly viscous one) will it necessarily flow in a way that maintains the shape of the combustion chamber or will it just flow via the path of least resistance - which may or may not be the desired shape. Needless to say, a combustion chamber that changes geometry during flight is not something you want.

The air should be breathable, there's no reason for that to change. I'm assuming you mean a solar flare? There would be no fallout as there is no radioactive component to a solar flare. It would cause an increased exposure to high energy particles (protons and neutrons) and high energy radiation for the duration of the flare, but the Earth's magnetosphere shields us reasonably well and there are no lasting effects once the flare is over AFAIK.

I would imagine anything to do with the engines would have to be obtained experimentally, the design is a pretty novel one - hybrid with a non-solid fuel as well as a pretty unique pressure feeding mechanism. I doubt there will be any exisitng models for this kind of thing, unlike for a simple biprop liquid engine.

Do you mean CFD (Computational Fluid Dynamics) software? You could run some CFD on a model, you'd only be able to get the drag forces (and so a drag coefficient) at certain velocities and atmospheric conditions, you wouldn't easily (not to say it's impossible, just enormously time consuming) be able to compute an entire ascent. Also with CFD there's always the issue that you can never be 100% sure it's given you the correct answer until you validate it in a windtunnel, for instance: Still, CFD is really the only way to get aerodynamic data in the design stages.

Well Carl Sagan, his co-author friends and their model disagree with you on that one. Firestorms are very different to a small fire. The firestorms created by nuclear strikes (i.e. the firestorm after the bombing of Hiroshima), firebombings (i.e. the firebombing of Dresden) or indeed large scale bushfires or wildfires are well known to pull soot up into the troposphere and stratosphere.

EDIT: Having re-read this, I'll preface it with an "POSSIBLY OFF-TOPIC" warning. The radiation is not so much of a problem as the contaminants and the fires started by the blasts. One of the most cited papers on the topic of nuclear winters [1] (including a certain Carl Sagan as an author) projects that just 100 megatons worth of mid-yeild (~100 kT) air blast bombs targeting cities and nearby industrial areas could ignite 50% of urban flammable material, or a "conventional nuclear exchange" targeting military and infrastructure targets could ignite 25% - 75%. The resulting soot in the atmosphere could cause, according to their model, "midsummer land temperature decreases that average 10 to 20°C in northern mid-latitudes, with local cooling as large as 35°C, and subfreezing summer temperatures in some regions" along with "disruption of monsoon precipitation and severe depletion of the stratospheric ozone layer in the Northern Hemisphere", which is pretty apocalyptic if you ask me. The temperature drop alone would basically end outdoor agriculture. [1] R. P. Turco et.al., Climate and Smoke: An Appraisal of Nuclear Winter, Science, 247, 166–176, 1990

And that's ignoring the nuclear winter that would wipe out most of the plant life that hadn't already died from the radiation!

What specifically in 1962?

@jsisidore For the purposes of a story, a drug similar ex-rad and a healthy dose of genetic engineering would be enough to achieve whatever radiation related plot point it is that you want. Whether or not it is entirely scientifically accurate is another question, but it would be accurate enough that very few people could say for certain whether or not you are wrong. What you say about evolution is true as well. Evolution never stops, but because it is a process that manifests over thousands of generations it's impossible to perceive it happening (and oddly enough, there are many who believe that one of the contributing factors to the genetic mutations that cause evolution over time is DNA damaged by radiation)

The biggest issue with this sort of thing is always getting it to work efficiently. An idea de Laval nozzle has a smoothly varying internal surface. If you have nozzle segments that get added into place on the end, chances are you'll get a gap or a bump which will trip a whole load of shockwaves which will disrupt the flow at the exit. This means unless the extensions are extremely well designed you might not realise the ISP gains that you hope for, or worse you might lower the ISP and still be carrying more weight. EDIT: I didn't know the RL10B-2 did that, thats's quite the feat of engineering, though greatly simplified vs the OP's suggestion by having the nozzle extension in place before ignition.

Unfortunately there's a big difference between a micro-adjustable valve for low pressure garden sprinklers and a computer-controlled micro-adjustable valve for a high pressure HTP system. For the former "micro-adjustable" just means the ability to have a few more positions than just on or off. For the latter it means the ability to make minute adjustments to carefully control the flow. Also because you're working with gas (and high pressure gas at that) the precision of manufacture and the quality of seals e.t.c has to be orders of magnitude higher.

That's also a good point. To have fine throttle control you're going to need to have an entire computer in order to constantly monitor chamber pressures and propellant pressures simultaneously across all (is is 8 first stage engines now?) engines, making adjustments to valves if necessary, as well as the ability to throttle the whole stack independently of the ideal throttle position in case one or more engines run in instability. On this note, you also need some pretty expensive micro-adjustable valves. What might be more realistic is for each engine to have two, three or maybe four pre-determined throttle settings (maybe 100%, 80%, 60% and then minimum, whatever that may be). This way, the valve settings can be pre-determined on a test stand as well as the way to get between each setting controllably.

Also, gamma rays are so highly penetrating (you need tens of centimetres of lead to reduce their intensity to effectively zero, and then (as @SargeRho mentions above) you get potential for secondary radiation) that regardless of whether you make your spacesuit of thin layers of lead, gold, tungsten or spit-n-duct tape, you you don't reduce their intensity by much at all. For some context, the following table is compiled from [1] Shielding required for a 50% attenuation of 1200 keV* gamma rays. Material Thickness / cm Lead 1.0 Iron 1.8 Water 10.8 *1200 keV (or 1.2 MeV) is not particularly high energy, some radioactive decays can produce gamma rays up to about 10 MeV, cosmic rays can be in the GeV or, in extreme cases, TeV range. [1] D. R. McAlister, Gamma Ray Attenuation Properties of Common Shielding Materials, Report for Postgraduate Research Foundation, 2013. Retrieved from http://www.eichrom.com/PDF/gamma-ray-attenuation-white-paper-by-d.m.-rev-4.pdf

Ideally, as long as there's choking at the throat of the nozzle you should get the same exhaust velocity (in practice you'd probably see a small decrease), you're just getting a greatly reduced mass flow rate. Edit: originally in the above I stated "as long as there's no choking at the throat" but in fact choked flow is of course required in a de Laval nozzle.