I feel certain that you could build a machine that works in lava.
But you’re going about it all wrong by imagining it would be some room temperature-workable device fitted out to work at really high temperatures.
good engineering design is something that takes advantage of the ambient circumstances and works in harmony with them, rather than trying to work in spite of them. Don’t view the high-temperature , “sticky” lava as an obstacle, but as a design element.Take advantage of things you only have in an under-lava situation.
I’m not suree what those would be =-- there hasn’t been a lot of study of materials at that temperature, under those circumstances. I could certainly build a lever out of refractory materials – that qualifies as a “machine” by the most basic definition. But I’ll bet there are devices that could be found that work well under such extreme circumstances that wouldn’t work at all under normal temperatures.
You can pump “heat” in any direction. It only becomes a problem when the environment you are pumping “into” is so hot your AC components melt. The hotter the dumping environment and the cooler you want the cool environment, the harder it becomes, but lava isnt so hot IMO that its beyond the realm of “doable”.
If the machine required interior parts that had to work in much more reasonable temps, it would probably have to be pretty large, given that volume (energy storage) goes up as sized cubed, while surface area (heat “leaking” in) goes as size squared. The bigger you make it, the better your volume to surface area ratio gets. You need the energy storage to keep pumping the heat out. With lava outside you are going to be doing some serious heat pumping.
I suspect if it had to work for any significant length of time, the thing would have to be nuclear powered.
Finally, here is that stupid paper!
Precisely. Can lava itself be a lubricant? What is its properties? Is it corrosive?
There are different types of lava. Could you channel low-melting-point lava in a high-melting-point rock? Foam the high-melting-point rock as an insulator for your testbed lava container?
The Orbiter of the Space Shuttle is subject to 3000F temperatures during reentry, quite a bit hotter than lava, and not only do its systems remain functional, its human occupants are well protected. But then again, the Orbiter isn’t subject to those temperatures for very long, so I don’t know how relevant this is.
Simply amazing. Is the idea sound? Has anyone else come forward to say it would or would not work? How do you melt a million tons of iron at the same time?
How would that work? The only way your AC components are going to be able to cool the system is if they have someplace to dissipate the heat energy they’ve absorbed from the system. Nothing in the machine is going to be hotter than the lava, how can you use the lava to cool it down?
The upper end of the scale is 11 million tons of iron and I have no idea how you could melt that much iron. Not only that, but I have no idea how long the crack in the earth’s crust after the nuclear detonation lasts. So you have to worry about transporting the molten iron to the blast site post haste.
Nonetheless I think it’s a really friggin cool idea. It’s made more interesting by the recent theory that the earth’s magnetic field could be magnetohydrodynamic - i.e., created by the oceans and not a molten core.
Granted, lava is hot, but it isn’t that hot. Wikipedia suggests typical temperatures between 900-1600 degrees Celsius. Warm, but I suspect not completely beyond the realms of our materials sciences abilities. Certainly it isn’t beyond the realm of theoretical possibility. For instance, tungsten has a melting temperature of around 3410 °C. Carbon’s about 4000 °C.
Extremely simple AC theory answer here. You can make “parts of the machine” as hot as you want till they melt. Other answers here have shown some materials melt at temps significantly hotter than “lava”. You make something extemely hot, use the lava to cool it down to just lava hot and viola! you’ve dumped heat INTO the lava, which was the whole purpose. If it was a gas, in tubing say, then you let that gas expand in volume and drop in pressure and it cools way down. Now you have a cool gas to cool things down. I’ve done a lot of handwaving here, but thats the gist of it.
LOL
that has got to set some sort of record for cool idea / crappy diagram ratio.
Remember, you’ll have to deal not only with heat leaking in from outside, but waste heat from whatever you’re using to power your probe. And if your power supply technology is in any way based on a heat engine (as most are, including nuclear reactors) then the hot end of your engine (your reactor core, in this case) is going to have to be significantly hotter than the magma outside. I think it’d be much more practical to design the thing with a consumable supply of some high-heat-capacity material (a big block of ice, in the simplest design) and just plan your missions to be short enough that you don’t use it up.
A million tonnes of molten iron, eh? I wonder what the probe would be made of. And how would it sense passing chemical compositions if it was in the midst of all that iron? And what would power it?
Fascinating. I’d be keeping an eye out for unexpected volcanic eruptions near the launch site though.
Heat transfer can be achieved with superheated steam which can easily be increased to well above any material melting point but let’s say 100C above the immediate lava temp outside transferred through say tungsten pipes and outer skin so below the working temp of the final heat pump stages.
Such a heat pump would require several stages and a lot of power, so nuclear reactor for source sounds about right and the later heat transfer stages would require some very exotic pumps. Clever insulation would direct the heat efficiently, maybe foam ceramics etc so the living/control area could be kept to 25C.
The sub would need several skins for the temp and pressure but could be pressurised to reduce the need for massive hull strength, but if a robotic submarine this could be made much cheaper and smaller, with internal temp at say 100C but notably the pressure could made very high using say a water filled sub to counter the vast pressures as you go deeper. Controlling the pressure to match outside would be done by hydraulics and electronic control.
As the pressure mounts however this would be more problematic than the control of temperature, as you are looking at pressures of 1.5 million atmospheres which will crush any component with a flaw in it. At the very least a few thousand atmospheres would be a starting point.
The molten lava will also be very difficult to travel through, so the tractive effort of tank tracks, an elaborate drill or the like would also be required. The way a mole digs through soil would be a start with material from the front being passed to the back.
This sort of robo lava sub is no doubt possible but would be a highly integrated machine with several inter dependent systems working to the limits of design. The project would be similar to the problems encountered by space travel and significantly more difficult than deep sea technology.
Only the US, China and Russia would have the money and tech to pull this off but I guess a lot depends on what you’d use this device for?
Where are you dumping heat to? You have to have a cold temperature reservoir into which to reject excess heat.
Stranger
You dump the heat into the lava. You just have to pump the heat into the cooling coils really really fast. The cooling coils would be hotter than the lava.
So they’re made of unobtainium?
As for pumping the incoming heat back out to the same reservoir, permit me to introduce you to my favorite physical law, the second law of thermodynamics, as typified by this diagram. Even if you are doing negative work (putting energy into the system) you still have to have a lower temperature reservoir to dump heat to, and in fact in doing the work in a closed system (pumping and/or compressing the fluid) you are adding more heat to the system, which is self-defeating. You could of course carry your heat sink around with you–say, a supercooled block of lead–but that means that your reservoir is a very finite resource, while the energy flux from even “cool” molten lava is going to be enormous. It just isn’t feasible to cool the hull of a “lava submarine” (or rather “subterran”) via convective cooling.
Stranger
I didn’t say I could engineer the damned thing, but it doesn’t violate the second law of thermodynamics any more than your air conditioner that takes heat from your house and dumps it into the hotter outdoors.
Does this require five nuclear reactors a million miles of coil and nonexistent materials? Probably, but the second law simply means that it will take a lot of energy to move that heat, not that it can’t be done.