> The pressure inside and outside the lunar landing module was 1,000,000,000,000,000x. 100 kPa to 10^-10 Pa
We didn't pump air into the lunar module from the Moon. We carried it from Earth pre-pressurized. Unless you're proposing we build and seal the Hyperloop tubes in space before bringing them down, the analogy isn't appropriate.
Also, consider a soda can. It's stronger when pressurized from the inside. Modern nuclear attack submarines collapse around 730m [1]. Since for "every 33 feet (10.06 meters) you go down, the pressure increases by 14.5 psi," we're talking about water pressure of about 1,000 psi or about 72 atm.
Both are pumped down to very high vacuum, not required for the Musk Hyperloop, (low pressure electric jet in a tube) but probably needed for the Hyperloop One. (maglev vactrain) And, obviously, both chambers were built terrestrially, and evacuated using conventional vacuum pumps.
> I am not sure what your objection is here. Are you saying vacuum chambers can't be built on Earth?
No, I specifically linked to the Space Power Facility in my original comment! The "materials science" problem I call out is in the tube's lateral thermal expansion.
The best solution I've come up with for that is to take a regenerative rocket engine [1] and make it a tube. Pumping fluid in spirals along kilometers of a vacuum tube isn't easy, but it isn't as hard as trying to invent a material that won't deform when the top gets hotter than the bottom, or the east side gets hotter than the west.
>The "materials science" problem I call out is in the tube's lateral thermal expansion.
I admire your method of argumentation.
The comment I replied to said nothing at all about thermal expansion. Your original comment, which I wasn't replying to, was mostly strange references to pressure ratios, and a single line about thermal problems, with no numbers cited. In that comment you did link to the Space Power Facility... as an argument against Hyperloop One! You did something similar downthread: https://news.ycombinator.com/item?id=15459777 But by asserting that you addressed the problem in your original comment, you make the person you're arguing with look like a bully, without having to actually address their points. Very efficient.
>The best solution I've come up with for that is to take a regenerative rocket engine [1] and make it a tube. Pumping fluid in spirals along kilometers of a vacuum tube isn't easy, but it isn't as hard as trying to invent a material that won't deform when the top gets hotter than the bottom, or the east side gets hotter than the west.
Using the phrase "regenerative cooling" in this context is another headscratcher. Regenerative cooling in rocketry is running propellant through channels in the nozzle, then either dumping it overboard, using it to power a gas generator, or burning it in the rocket. It's a great way to get rid of megawatts of heat.
None of these things would be useful for a hyperloop tube? You don't want to run kerosene/liquid oxygen/hydrazine/whatever through a cooling jacket then dump it on the ground, you don't need a gas generator for anything, and there's no way to feed the heated propellant to the actual hyperloop car. And if you could, you wouldn't want to, since if you combusted it in the car it would just dump the exhaust in the tube, killing the vacuum.
Presumably active cooling of a hyperloop tube would use a closed refrigerant cycle, which has little to do with regenerative cooling, besides the idea of cooling channels.
Talking about regenerative cooling in this context isn't wrong, exactly, it just betrays a rather shallow understanding of the problem at hand, as seen in your first comment.
> Regenerative cooling in rocketry is running propellant through channels in the nozzle, then either dumping it overboard, using it to power a gas generator, or burning it in the rocket
"Regenerative cooling" is a rocket term. It came to mind because I used to be an aerospace engineer. There's no requirement in the definition of regenerative cooling for the coolant to be dumped.
> Presumably active cooling of a hyperloop tube would use a closed refrigerant cycle
I don't think one can just presume that. You've already got lots of pumps for pumping air. Given (a) the seal on the tube will be periodically broken (for entry, exit and maintenance) and (b) a safety factor, you'll have more pump capacity than you'll need. Filtering and then compressing atmosphere, running it through a heat exchanger, and then letting it expand through the cooling channels before dumping it doesn't seem obviously worse than having kilometers of refrigerant running around.
> which has little to do with regenerative cooling, besides the idea of cooling channels
See above. Also, I assumed if you did this you'd use it to boost SpaceX's nozzles' economies of scale.
>"Regenerative cooling" is a rocket term. It came to mind because I used to be an aerospace engineer. There's no requirement in the definition of regenerative cooling for the coolant to be dumped.
?
Regenerative cooling in rocketry uses propellant. Propellant is always dumped overboard, because it's propellant.
Please show me a rocket that uses closed loop cooling of the rocket nozzle.
>Filtering and then compressing atmosphere, running it through a heat exchanger, and then letting it expand through the cooling channels before dumping it doesn't seem obviously worse than having kilometers of refrigerant running around.
My turn to be pedantic about definitions: this sure sounds like closed loop cooling to me! The working fluid is air, you draw it from a big reservoir, (the atmosphere) cool something with it, then return it to the reservoir. (Something you can't do with open-loop regenerative cooling of a rocket nozzle, since the coolant gets burned at the end of the cycle)
However, maintaining the air dryers and replacing the filters sounds like it wouldn't be any cheaper than conventional phase-change cooling, and having to build custom vacuum/compressor pumps for the hyperloop project is going to be more expensive than buying COTS vacuum pumps.
> Please show me a rocket that uses closed loop cooling of the rocket nozzle
The most famous one is the under-development SABRE [1], which includes a closed-loop helium cycle. For cryogenic rockets, closed-loop cooling of the nozzle has been explored [2] to avoid hydrogen embrittlement and oxidation of the nozzle channels, as well as to simplify plumbing.
In any case, we've devolved into arguing semantics.
> My turn to be pedantic about definitions: this sure sounds like closed loop cooling to me!
I was figuring on dumping the air once done versus worrying about a reservoir. That said, I haven't done any math on the benefits of saving the return piping (and reservoir cost and maintenance) versus using something traditional.
> maintaining the air dryers and replacing the filters sounds like it wouldn't be any cheaper than conventional phase-change cooling, and having to build custom vacuum/compressor pumps for the hyperloop project is going to be more expensive than buying COTS vacuum pumps
Fair enough. As you observe, it's a problem I haven't seen a suitable solution to (apart from burying, which trades the thermal problem for, in my view, the better water management problem and the scarier land-use problem.)
>The most famous one is the under-development SABRE [1], which includes a closed-loop helium cycle.
From the linked article:
>>The 'hot' helium from the air precooler is recycled by cooling it in a heat exchanger with the liquid hydrogen fuel.
It's open-loop regenerative cooling, with helium as an intermediate coolant. The heat still ends up in the propellant, which gets dumped overboard. Not closed-loop.
That patent also specifies a heat exchanger to the propellant tank.
>I was figuring on dumping the air once done versus worrying about a reservoir.
A poorly telegraphed joke. The atmosphere, here, is the reservoir. I have edited my comment.
>However, maintaining the air dryers and replacing the filters sounds like it wouldn't be any cheaper than conventional phase-change cooling, and having to build custom vacuum/compressor pumps for the hyperloop project is going to be more expensive than buying COTS vacuum pumps.
Just want to point out a facet of such a project that you probably haven't considered. The EPA puts strict requirements on the operators of refrigerant systems, the way they categorize operator class is based on pounds of refrigerant. Thus even though a COTS phase change refrigerant system seems like the obvious choice for Hyperloop One, because of regulatory burden it would almost certainly unfeasible. A refrigeration system big enough to cool a hyperloop would have to have every inch of refrigerant line spray tested every 3 months or an insanely expensive monitoring system.
How about wrapping the tube in insulating foam and using peltier modules to regulate temperature. Or even using double walled tubes. It would not be challenging to regulate the temperature, but nobody talks about it because it doesn't even seem necessary.
Because foam is cheaper? A nice K.I.S.S. solution.
Having seemingly solved this thermal expansion problem "in the same category of problems separating us from a space elevator," I'll be expecting The Fountains of Paradise to come true soon. ;)
> We seem to have solved this thermal expansion problem
I guess I'll have to telegraph my buddies in Hawthorne and at NASA :).
Joking aside, no, foam doesn't solve the problem. You'll still have flexing. This is one of the limits on how long you can have a launch tank on the pad. Imperceptibly small flexing, but of the kind that weakens every metal we know.
>Joking aside, no, foam doesn't solve the problem. You'll still have flexing.
Of course the flexing isn't zero, but by adjusting the thickness of the foam you can drive the diurnal/solar heating thermal gradient arbitrarily low without compromising strength by making the steel walls thinner. That was your material science objection,[1] right?
>but of the kind that weakens every metal we know.
That's true, but misleading. It's true that every metal gets weaker, but in steel and titanium this weakening levels off (all other metals continue to get weaker until failure). Once these two metals are at the fatigue limit they stop weakening and have unlimited flexing cycles.[2] So you can set that as your final material strength and size the tube thickness accordingly.
So repeated flexing is fine, and flexing can be made arbitrarily small with insulation. Foam solves the problem.
> We didn't pump air into the lunar module from the Moon. We carried it from Earth pre-pressurized. Unless you're proposing we build and seal the Hyperloop tubes in space before bringing them down, the analogy isn't appropriate.
It absolutely is appropriate, since the module was emptied and filled each time the astronauts left.
> Modern nuclear attack submarines collapse around 730m. Since for "every 33 feet (10.06 meters) you go down, the pressure increases by 14.5 psi," we're talking about water pressure of about 1,000 psi or about 72 atm.
Yeah- so building a hyperloop is structurally as complicated as building a tube that sits 33 feet under water. Hardly sounds complicated when put like that.
The force exerted by the atmosphere on an evacuated tube is the same as the force exerted on a standard-atmosphere filled submarine diving 10m beneath the surface.
A 100kPa difference.
And since they explicitly do not want a high vacuum - some residual air inside the tube is necessary for the hyperloop concept! - they don't have to deal with advanced tech like turbomolecular pumps. Simple displacement pumps will do.
> Note that downthread they've already moved the goalpost from "these pumps are technologically impossible" to "but they'll need a lot of them!"
I never said vacuum pumps are "technologically impossible". My original comment references the Space Power Facility [1] and categorises the vacuum problem as an engineering problem. A hard one, but engineering nonetheless.
The comment you link to [2] replies to someone claiming the original white paper calls for a 22:1 atmosphere:tube pressure ratio. I pointed out that the figure they're referencing, Figure 11, discusses the capsule and not the tube.
I'm skeptical about the economics of de-pressurising the tube, but that's an engineering problem and I've always held it as such. The materials problem is the thermal expansion of the top of the tube relative to the bottom.
>The comment you link to [2] replies to someone claiming the original white paper calls for a 22:1 atmosphere:tube pressure ratio. I pointed out that the figure they're referencing, Figure 11, discusses the capsule and not the tube.
I happen to be that someone. :) You were talking about state-of-the-art axial compressors (the GEnx-2B67), so I assumed you were talking about the axial compressor on the front of the pod. Mea culpa. But then you drew an analogy to the pressures in the SPS ("To get a sense of the engineering differences between 6:1 and 1,000:1..."), as if the Hyperloop people were trying to make a 1000:1 axial compressor. As I pointed out in my reply,[1] rotary vane compressors can easily maintain those pressures.
>The materials problem is the thermal expansion of the top of the tube relative to the bottom.
If that were really a problem, no pipelines of any kind could be built. Again thermal expansion joints are the solution, since with the abandonment of air-ski levitation the pod walls no longer have a requirement to be ultra-smooth. Tiny leakage on these joints is fine, since it will be made up for by the pumps located along the track.
> If that were really a problem, no pipelines of any kind could be built
The Trans-Alaska Pipeline system, which I believe is the largest at least in the United States, is 1.2m in diameter [1]. We're talking about a pipe almost 3 times wider that needs to hold itself against the atmosphere and keep capsules neatly contained.
Side note: long pipelines zig-zag to allow for thermal expansion and contraction [2]. You can't do that with the Hyperloop. (Bridges handle this with various ingenious methods, most of which will work for the Hyperloop's longitudinal expansion.)
> thermal expansion joints are the solution
Scaling pipe expansion joints where they maintain the near vacuum and deal with the structural stress of a capsule whizzing by will be difficult. By "difficult" I mean these are problems NASA (for the ISS) and Schlumberger (for pipes) have been grappling with for years and with billions of dollars in R&D.
> The materials problem is the thermal expansion of the top of the tube relative to the bottom.
What exactly is the problem here? Is it because the top of the tube is exposed to direct sunlight? I thought it was supposed to be covered in solar panels anyway. Is it still a materials science problem if the tubes are shaded, because providing a structure that shades something with expansion from direct sunlight exposure seems quite a bit easier than to do so while trying to keep vacuum to a particular level.
We didn't pump air into the lunar module from the Moon. We carried it from Earth pre-pressurized. Unless you're proposing we build and seal the Hyperloop tubes in space before bringing them down, the analogy isn't appropriate.
Also, consider a soda can. It's stronger when pressurized from the inside. Modern nuclear attack submarines collapse around 730m [1]. Since for "every 33 feet (10.06 meters) you go down, the pressure increases by 14.5 psi," we're talking about water pressure of about 1,000 psi or about 72 atm.
[1] https://en.wikipedia.org/wiki/Submarine_depth_ratings
[2] https://oceanservice.noaa.gov/facts/pressure.html