Read the PDF, I didn’t begin to touch what he talks about.
You may still disagree, but it’ll be about something other than the masks. 
Hopefully there are airlocks along the way?
We’ve all ridden airliners with the doors, though. Buying a set for every vehicle would be a matter of picking up a phone and wiring the money to whatever company makes them. So in a practical sense, it’s a solved problem.
I’m actually thinking about just how bad the gull wing door idea really is. If a door were to open up while the car was traveling at 800 mph, not ONLY would it instantly depressurize the entire cabin and kill everyone onboard, it would probably slam into the side of the tunnel and possible cause the car to break apart…
It’s not quite this bad. If there’s a small hole in one of the joints, you’d send someone to the concrete column in question, have them hook their safety harness to permanent cables attached to the column and climb it, then they’d squirt a can of sealant into the hole. 14 psi isn’t that large of a pressure difference, and there’s huge pumping stations up and down the entire tubeway.
To fix it permanently, you’d shut down the whole system for some time. (well, you might shut-down only the northbound or southbound side, and you’d have to close pressure doors that are located every few kilometers. You’d repressurize and fix that section. There would have to be a way to easily install a maintenance platform on top of each concrete column you’re working on, then a way to quickly get tools and supplies to the work crew on top, and so on.
To some extent this sounds like a problem that could be dealt with at the planning stages, there could be pre-installed anchor points at the top of each support column, pre-installed cables, some kind of system for winching up supplies, and so on.
I’m actually wondering how you’d design pressure doors that could isolate a tunnel section in a way that absolutely guaranteed that a terrorist couldn’t get to one of the doors and crank it closed…
Does the system recover energy during braking? I didn’t notice that.
As for gull wing doors, I think each passenger was supposed to have his own door (of sorts). You may have noticed that there is not enough space for an isle between seats. My suggestion would be to make the capsule wider to fit an isle and then have a single door at each end.
That would make the capsule wider, which would increase system costs. However, I suppose you could make the capsule longer, then just have an aisle and a single row of seats. Or, an even clever method would require fold-down seats and the passengers to get off in a specific order…
Unsafe beyond all reasonable measure.
In the event of a catastrophe where a train crashes somehow and comes to a near immediate stop, there’d initially be 28 casualties.
The train following it, assuming there is 2 minutes apart at 800mph, will be about 26 miles behind it when it receives the braking signal from the pressure sensors along the tube. A “comfortable” (read: survivable) deceleration is about 4g, maybe slightly more or less. At this speed, and ignoring the influence of a potential influx of air, the following train should be able to stop within roughly a mile (coincidentally the distance a freight train takes to stop), so they’d be safe. They’d be far off the crashed train too, so they could safely decelerate more slowly and reduce the risk of injury/unconsciousness from the high G force. The influx of air could potentially increase the deceleration but probably not to a dangerous level.
The worst case scenario for some kind of crash is probably just the unfortunate death of the 28 passengers on the particular train that crashed, with a few other trains following it trapped for however long it takes to rescue them. I’m not sure if it’s detailed in the pdf, but I imagine there’s probably some way of reversing the capsules to the previous station in the event of a major problem up the line.
That is detailed in the PDF. The massive battery pack that runs the jet turbine would be instead used to run an electric motor attached to a mechanical wheel, which would slowly move the car along the tube (at perhaps 45 mph?) back to the station.
Technically there’s enough energy to do this. However, several members have suggested emergency exits. And, for that matter, if there’s a possibility of getting stuck in that tube car for hours, it definitely needs a bathroom.
And if the worst case scenario is only losing 28 passengers, I think security screens are pointless. Any person with a gun (and some hi-capacity magazines…) can hold 28 people hostage and/or kill them in a mass shooting. 28+ people are in groups at a near infinite number of places at any time, in public places all over the U.S.
So what’s the point in screening the passengers? A terrorist could go virtually anywhere else and kill the same number of people, including in the very security line the passengers are waiting in.
In airliners, there’s a realistic possibility of a passenger turning the airplane into a guided missile that kills 20,000 people. (that’s how many would have died, PER TOWER, if the terrorists had timed their attack differently and aimed at a lower floor on 9/11)
I suspect you could make them more reliable than that, but that’s speculation on my part.
I was thinking it might be really nice to have 3 tubes instead of two. You could have two in service at any time, and the third dedicated to inspection/maintenance. It’s 50% higher cost for the tubes, but you gain a huge amount of flexibility. The pylons need to support more weight, but the number of cars passing at any one time isn’t any higher, so you don’t need more lateral strength or anything. The tubes could be arranged as a triangle.
Actually, you don’t even need that. I just didn’t think that part through. If you need to replace an expansion joint, all you’d have to do is slide in a steel plate on each end to seal both ends, then replace the joint, then unseal it.
Or, you could use an iris type sealer. That would allow you to deploy them automatically to retain vacuum in the case of a breach. Any failed section of the track would automatically trigger the closures (iris, sliding panel, whatever) on each side of the breach, maintaining the vacuum in the rest of the tunnel to prevent the shockwave from propagating and destroying the whole thing. Of course, you’d have to make sure that no pods were close to the panels before you did that, so you might have to wait a few minutes before deploying them in the worst case.
But for routine maintenance, that sliding doors or irises built into the interface between the tube and the expansion joint would allow you to remove any of them without losing vacuum to the tunnel.
The point remains, however, that a hyperloop with 5,000 expansion joints is a very different thing than one with a single welded-steel tube. You need to start feasibility and economic analysis from scratch. The stresses will be different, the pylon designs will be different, and the cost estimates are totally different.
Why 5,000? The max expansion is about 50" per mile. So do one big one every mile. A sliding 50" joint with a bellows covers 10 feet wide or so so that the bend min/max in the bellows will be less. That’s a few hundred joints (plus some kind of bearings on all pilons to accommodate the sliding of the pipe due to expansion/contraction).
No safety-critical system like this is going to be allowed to operate by fixing a leak with spray sealant. In fact, no safety-critical system is going to be allowed to remain operational to the point where it starts springing leaks at all. You need to develop a safety control plan that establishes safe limits of operation, an inspection regime to make sure everything stays within those limits, and a control plan to ensure that the procedures continue to be followed.
In the aviation business, for example, aircraft are inspected at fixed intervals - 50 hours, 100 hours, etc. That inspection is carefully designed around the fatigue life and failure modes of the components being inspected, and repairs or replacements are made before they start to fail. Those inspections can be extremely expensive depending on what has to be checked, and how.
These sliding expansion joints are not simple devices. You can’t just put a pipe inside another pipe and a gasket around them. At the weights and forces we’re talking about, everything would be moving on bearings. The seals would be constantly flexed, which leads to fatigue. If there’s rubber involved, there are environmental concerns like temperature and UV degradation. There may be springs, hydraulic dampers, etc.
Part of the engineering of such a device involves coming up with an inspection regime for it, and getting that certified by putting the device through load tests, cycle tests, vibration tests, temperature extremes, etc. This would take a long time and cost a lot of money. Part of the engineering of the device would have to include ways to gain inspection access to the moving parts without having to disassemble the entire thing. If those parts are inside the vacuum area, you would probably have to seal both ends, re-pressurize it, then open inspection ports or something.
Part of the feasibility analysis of the system would have to include these costs, including the effect on downtime of the system.
For example, let’s say there’s 5,000 of these things. Each one has to be inspected once per year. An inspection takes 10 hours. Right there we’re talking about 50,000 man-hours for inspections. At $200/hr, that’s $10 million per year just for the labor. And if you only have manpower to inspect 100 at a time, that’s 500 hours of downtime for the train per year - almost 20 days of downtime.
I don’t know if those numbers are even remotely in the ballpark - the point is is that you have to know them before you can figure out the cost of the system and the ROI of the project. You also have to factor in the cost of replacing the things - how do you get them up and down off the towers, some of which are going to be in remote locations? You might even have to build a separate crane track to run with the tubes so you can move heavy equipment and people along the track for doing this kind of maintenance. But now you’re adding the cost of that track, plus the hardware to use it, plus the maintenance of all that hardware and the equipment track itself.
This is the kind of stuff that blows out the budget of engineering projects. Everything looks simple from 10,000 ft. The devils are always lurking in the details. Some of those details can be discovered and dealt with by careful design and planning, but others are uinknowable until you run into them. Your careful design of an expansion joint works great on paper, until you discover that large birds are nesting in them and their feces are corroding the moving parts, or careful analysis of wind stresses didn’t take into account turbulence unique to the terrain, or whatever.
This is especially true of brand new systems like this one. The first railroads suffered all sorts of calamities and cost overruns, but now we have over 100 years of experience in building and running railroads and most of the unknowns are known. Same with airplanes. We know to put oval windows in pressurized planes because the first jet designed with square windows kept crashing until engineers figured out what was going wrong. We know what the time-between-overhauls should be for a piston aircraft engine because we have millions of hours of real-world experience to draw from.
When you come up with an entirely new system, you’re operating with a certain amount of ignorance about real-world performance and problems. So you have to build in extra safety margins and tighter inspection schedules and all that until you develop enough operational data that you can expand the envelope. That’s one reason why the first systems always cost way more than later ones.
The potential for terrorism opens another whole can of worms. We’re talking about 350 miles of very exposed tubing with support towers in critical locations. A fertilizer bomb might just bring down a tower, shutting the whole system down for months. A simple bomb attached to an RC aircraft could breach the tube at any point. Are you going to defend every mile of this thing? If so, better add that to the budget as well…
Expansion joint maintenance would be handled constantly, during up-time, using a cylindrical tube crawler that moves along the outside of the (above-ground sections of the) tube: when it centers over a joint, the crawler cylinder forms an airtight seal over the joint, allowing a worker in a pressure suit to perform inspection and repairs or upgrades without shutting down the system. Two minutes between cars should be enough to do a couple things on the joint, though I should think four or five minutes would be more realistic.
Two possibilities for doors: one would be a sliding canopy, wherein the passengers get in like they would a canoe; the other might be having each seat attached to the sidewall and the passenger just sits and pivots it into position (they ought to do this with 2-seat automobiles and pickups).
Maybe. The tradeoff would undoubtedly be the ease of making a joint that can expand long distances vs more joints that have small expansion. You might be right that only a few hundred are needed - we’d have to wait and see the designs.
In any event, even a few hundred expansion joints significantly changes the entire design. You need to have a way of getting these multi-ton devices up on the towers and back down, cranes for hoisting them into place, etc. You need a way for workers to get to them as well.
I suspect this system will need a maintenance track like a railroad track or a road that can carry workers between pylons and moving crane to replace expansion joints.
I-5?
Sam Stone, your perspective on this is invaluable. I agree completely with your general point.
With regards to a fertilizer bomb : wouldn’t the blast wave of bomb set off at the base of a tower reflect off the reinforced concrete of the tower footing and mostly damage softer structures nearby? As I understand it, explosions take the easiest path. And inch thick steel isn’t easy to damage, either.
Technically, big suspension bridges like the Golden Gate bridge all depend on two ridiculously thick cables. If the terrorists have the kind of shaped charge expertise to blow a concrete footing or break a tube, they could probably take down a major suspension bridge or do similarly catastrophic damage.
Point is, this tube system merely needs to be as secure against attack as other commonly used methods of transportation.
Big portions of the tubeway are supposed to go through farmer’s fields and the like, where the concrete footings provide minimal disruption in land use. If you have to add a roadway, and maybe some fences with razor wire on top, and maybe some security sensors, and perhaps an SUV loaded with armed guards ever so often…
The cost rises very very quickly.
Yanno, active thermal management doesn’t sound all that bad now that I think about it a bit more.
Try this:
The outside of the tube is insulated with a fairly cheap and effective material; some kind of foam, say. The tube itself is heated to 120 degrees or whatever you choose as the upper limit. This should require very little energy, since the inside is close to a vacuum and the outside is well insulated.
Because it’s heated, you can use very cheap and reliable resistive heaters (you might even get away with using the steel itself as the resistive element!). Cooling equipment is very expensive in comparison. It would be largely solar powered. At night, you just use batteries. The raw thermal mass might be enough to keep it going for a while, actually. There would also be a grid backup.
To account for failures, tube segments would have breakaway sections. The inside would only have a tiny seam, and the segments would be joined in a way such that they split if the forces rose above some value. There wouldn’t be any cascading failure from local damage.
You would still have bearings, so that a few point failures would have no effect–the tube could slide around a bit. You could even have nearby segments increase their temperature to avoid any stresses at all.
The whole system is still pretty stiff, and so you might still want a handful of sliding joints, but they could be pretty far apart and wouldn’t need a huge amount of slack.
The pods collect hot water already, use that to heat the pipes. 
Re: loading the capsule.
ISTR that Gerard K O’Neill (I think) solved that problem in his book 2081.
The capsules in his tube train had an outer cylindrical shell with a large door, more a lid really, at one end. There was an inner cart on wheels. On arrival at the station, the capsule left the mainline, backed to an airtight station door at the end of the branch tube, and sealed itself to the door.
Then the station door and the capsule doors opened simultaneously, and the inner cart–bearing the seats, passengers, luggage compartment, etc–rolled out into the station. The cart floor was level with the station floor, and there were no longer any enclosing walls around it, so people could just saunter on and off with their luggage.
I can imagine that there could be a second safety door inside the tube, so that when the capsule was docked to the station door, the whole segment surrounding it could be temporarily pressurized. This might decrease chances of leaks into the tube from the station.