Hyperloop : pie in the sky or is it time?

[Habeed]

The Concorde itself had several severe problems that made it not worth it.

The nastiest was the fuel consumption. Per wikipedia, it gets 14 mile-per-gallon per passenger, versus 91 mpg for a Boeing 747-400. (and the 787 dreamliner is about 30% better than THAT)

That’s the problem. Fuel costs are about 30% of the cost of an airline ticket. If you just look at fuel costs, that’s a 6.5 fold difference. Then there’s the fact that a supersonic aircraft is much more expensive to build, can carry fewer passengers, more expensive to maintain, and creates nasty sonic booms so it cannot fly just anywhere.

It’s also a higher performance design, which means pushing the envelope on safety - all other factors held equal, an SST is going to be less safe because there’s design pressure to have fewer redundant systems to save weight, and there’s enormously more aerodynamic load on the airframe.

In any case, a round trip ticket to europe costs about a grand today. If the fuel prices are 6.5 times higher for an SST, and fuel costs are about 30% of that $1000 ticket, then at a minimum the airline MUST charge another $2000 for a ticket on a supersonic transport, or they go out of business. ’

Simple physical reality, obviously, the real ticket prices will be higher still.

Most people, saving 4.5 hours is not worth paying $2000. In fact, very few people are “worth” $444 an hour, even neurosurgeons aren’t paid that much. (some lawyers “bill” for that much, perhaps)

[Sam_Stone]

I’ve read the paper, and I’d call it an “interesting thought experiment”, not a real blueprint for a new travel mode.

Musk’s a smart guy. When this was first announced all the tech wags assumed it was a tube with air being pushed at hundreds of miles per hour and the capsules being pushed along by the air so they experienced no drag. Anyone familiar with fluid dynamics should have known that that was a complete non-starter, so I’m glad that Musk’s concept doesn’t do that. He’s not as dumb as most tech journalists, at least.

Also, in concept and at a high level the engineering of it is fairly straightforward and uses known materials and methods. There are already maglev trains pulling themselves along a track. This is basically an enclosed maglev in a partially-evacuated tube, allowing high subsonic speeds with less drag, at the expense of having to build an enclosed tube around the entire thing.

I think Musk is smart enough to have realized partway into this that it wasn’t something that was anywhere near ready to build, so rather than get bogged down in a huge project like this he just released it open source to let people use as they see fit. Very cool of him to do that.

Now, the problems…

First, the major safety issue seems to me to be the partial vacuum inside the tube. Can it be maintained? How much energy is required to pump out the air and keep it sustained? What about pressure waves when capsules are loaded an unloaded? Do they go through an airlock? Even an airlock will spill out air into the tube.

I think the biggest safety concern is a catastrophic failure of a tube section. Even if no pods are near it, any major break of the tube is going to cause a violent inrush of air (the energy differential is huge between the low-pressure tube and the outside environment). And if the tube’s pressurization fails, what happens to those 760mph pods? They’ll be slamming into a lot of air mass very quickly. You could probably fix this with some kind of pressure blow-out valves in the tube section so a pod hitting a wall of air would cause the air to expel, cushioning the slowdown.

Another safety issue is the big impeller fan that would be at the front sucking air from the front of the vehicle and pushing it out the back. This is necessary to keep the air from pressurizing in front of the car and slowing it down. What if that fan suffers a failure? Airplanes are always designed with multiple redundancy, but this is a single point of failure. Not a big deal if the failure just means rapid but safe deceleration, but I’m not sure.

Then there are a million little engineering details that would have to be worked out. For example, pumping out that much air is going to cause a temperature drop inside the tube, which might cause frost and ice issues. The loop itself will expand and contract with temperature fluctuations, requiring airtight expansion joints for each section. The loop will also have to handle topology changes due to tremors, sinking of pylons, plate shifting if it’s going over fault lines, etc.

I’m not convinced of his energy estimates. I think it might take a hell of a lot of power to keep that tube partially evacuated. The fan in front of the pods will basically be a jet engine which will require power. You also have to power the maglev track.

Still, it’s a good attempt at thinking outside the box for transportation, and by making it open source maybe it will spur a private movement to improve it and test it out. I’d love to build a scale model of this using parts from one of those pneumatic tube systems. It’d be the coolest model train ever and you could get some useful data out of it and help contribute to the project.

[Habeed]

@Sigene

one technical question that I don’t understand. propulsion of the cars requires higher pressure on the backside of it vs. on the front side. Is this correct?

you need a relative vaccum on the forward end compared to the rear end. How do you get multiple cars going through a tube. Won’t the high pressure side (back) of the first car have to be the lower pressure side (front) of the second car? Wouldn’t the back of each car have to be higher pressure the the back of the car in front of it? This seems like an unsustainable series of increasing pressures.

Any help on explaining that to me?

No, the propulsion comes from linear electric motors in the track. It’s quite neat, really : read the PDF in the OP. It’s the same electric motor in a Tesla, “unrolled”, to where it’s stretched out in a line. There’s a blade sticking out the bottom of the transit car, and a “slot” the blade must slide in to when it reaches the part of the track that has the linear motor.

This is a weak point in the design. If the car fails to precisely align itself with the slot (once in the slot, electromagnetic forces will help to keep it centered, it just has to enter it correctly) then bad things will happen, and for part of the journey it is going 800mph.

Most of the thin air in the tunnel will flow past the car, but some will get sucked into a turbine in front. Some of that air will be sent right out the back (about 60% will be “bypassed”) and the rest goes to the air nozzles keeping the car from touching the sides, and to providing air to the passenger cabin.

[Sam_Stone]

@Sigene

Yes perhaps some of the technology is here now, but …how do you know that a tube of 0.8 inch steel is enough? Once you get into the nuts and bolts of it, you might need 3 inch steel; or 5 inch steel incased in 2 feet of concrete. These are the details that are just guessed on in the current business case, that very likely (again my opinion) are going to be found to be insufficient. Things like this will cause severe cost overruns.

Musk is talking about evacuating the tube to about 1/1000 of sea level air pressure. That means the tube walls are going to have a force of nearly 14.7 pounds per square inch applied to them. Still, when it comes to steel pressure vessels, that’s not very much. A water tower has a lot more force on its lower walls than that. A SCUBA cylinder has an internal pressure of 3,000 PSI. However, that’s putting the steel cylinder in tension since the pressure is pushing out, not pushing in.

Still, 14.7 PSI should be manageable. I think the required strength of the steel walls will have other limits - environmental, hurricane resistance, resistance from rifle bullets, that sort of thing.

[Sigene]

Thanks for the propulsion summary:)

[Sam_Stone]

@Sigene

one technical question that I don’t understand. propulsion of the cars requires higher pressure on the backside of it vs. on the front side. Is this correct?

you need a relative vaccum on the forward end compared to the rear end. How do you get multiple cars going through a tube. Won’t the high pressure side (back) of the first car have to be the lower pressure side (front) of the second car? Wouldn’t the back of each car have to be higher pressure the the back of the car in front of it? This seems like an unsustainable series of increasing pressures.

Any help on explaining that to me?

The whole point to the fan in front of the pod is to keep it from pressurizing the air in front of it and leaving a low-pressure area behind it. In theory, if a fan can blow the air around the pod at the same speed the pod is going, then the air mass would be relatively undisturbed. Without the fans, a pod would feel the air mass in front of it like a big spring, slowly building up pressure and slowing the pod down.

Still, the pods will leave a wake behind them. I guess that’s one of the reasons he wants to space them out by 26 miles. You need time for the air to settle back down.

The biggest doubt I have is the feasibility of maintaining a near-vacuum in an exposed tube hundreds of miles long.

[Sam_Stone]

@Habeed

This is a weak point in the design. If the car fails to precisely align itself with the slot (once in the slot, electromagnetic forces will help to keep it centered, it just has to enter it correctly) then bad things will happen, and for part of the journey it is going 800mph.

I’m not sure I understand this point. The inverse-square law would cause the force between the ‘ski’ and the slot to rise dramatically if it moved too close - that’s why maglev trains can flot above their track with no risk of hitting it. I think you said the same thing when you say the electromagnetic force will help keep it centered. So where is the alignment trouble? When the pod is first lowered to the track? It won’t be moving then.

[Blakeyrat]

My opinion: far too ambitious for something far too new.

What he should have proposed is building this as an attraction for an amusement park-- imagine one these circling Disney World or Disney Land the way their monorail does now, for example. That gives them a chance to building a smaller scale prototype, resolve some of the issues, it’s not a big deal if it’s shut down for maintenance, etc.

The concept seems workable, but frankly I’d prefer a system with a transparent tube-- people are going to feel really, really cooped-up in this tiny coffin-like vehicle. Airliners have windows for a good reason.

[Habeed]

@Sam_Stone

First, the major safety issue seems to me to be the partial vacuum inside the tube. Can it be maintained? How much energy is required to pump out the air and keep it sustained? What about pressure waves when capsules are loaded an unloaded? Do they go through an airlock? Even an airlock will spill out air into the tube.

I think the biggest safety concern is a catastrophic failure of a tube section. Even if no pods are near it, any major break of the tube is going to cause a violent inrush of air (the energy differential is huge between the low-pressure tube and the outside environment). And if the tube’s pressurization fails, what happens to those 760mph pods? They’ll be slamming into a lot of air mass very quickly. You could probably fix this with some kind of pressure blow-out valves in the tube section so a pod hitting a wall of air would cause the air to expel, cushioning the slowdown.

Another safety issue is the big impeller fan that would be at the front sucking air from the front of the vehicle and pushing it out the back. This is necessary to keep the air from pressurizing in front of the car and slowing it down. What if that fan suffers a failure? Airplanes are always designed with multiple redundancy, but this is a single point of failure. Not a big deal if the failure just means rapid but safe deceleration, but I’m not sure.

Then there are a million little engineering details that would have to be worked out. For example, pumping out that much air is going to cause a temperature drop inside the tube, which might cause frost and ice issues. The loop itself will expand and contract with temperature fluctuations, requiring airtight expansion joints for each section. The loop will also have to handle topology changes due to tremors, sinking of pylons, plate shifting if it’s going over fault lines, etc.

I’m not convinced of his energy estimates. I think it might take a hell of a lot of power to keep that tube partially evacuated. The fan in front of the pods will basically be a jet engine which will require power. You also have to power the maglev track.

Still, it’s a good attempt at thinking outside the box for transportation, and by making it open source maybe it will spur a private movement to improve it and test it out. I’d love to build a scale model of this using parts from one of those pneumatic tube systems. It’d be the coolest model train ever and you could get some useful data out of it and help contribute to the project.

The paper you didn’t fully read calls for airlocks. Also, amusingly, the system requires there be a small amount of air in the tube, and it to be fresh air. This is because some of the air from the turbine in front goes to pressurize the cabin like on an airliner, so that you don’t need a closed loop life support system. So, you can’t allow CO2 to buildup in the tunnels.

Every section is supposed to be cleanly seam-welded to every other section, so it’s supposed to be a continuous mass of steel tube from one end to the other. Hopefully nowhere for it to leak except at the airlocks, pumps, etc.

Onboard each capsule are some compressed air tanks that feed air through regulator valves to the individual skis. The jet turbine is to reduce drag and keep these air tanks full. If the turbine fails, the car is going to slow to a stop gradually - most air will bypass the car because it isn’t “tight” against the tube walls like a syringe, about 40%-60% of the frontal area is empty space. (I wasn’t able to determine which, maybe you can)

The “emergency mechanical braking system” would be some conventional ceramic brake pads that stick out from the tube car on a hydraulic ram and scrap along a piece of the tube that the skis never pass over. (maybe somewhere along the top or bottom). That way, any “roughness” in the tube surface from cars emergency braking won’t affect operation.

In the other thread on this, someone did the math on the thermal capacity of the thousands of meters of steel a high speed tube car would pass over - conventional brakes will do it. Amusingly, it’s just like the linear motors - it’s the drum brakes in your car, “unrolled”.

If a tube section catastrophically fails in that someone blows it up, people will probably be killed. But that failure mode is present in the alternative methods of transportation.

There are some tricky bits totally unaddressed in the paper. For instance, how do you build “junctions” in the system, to where there are 2 tubes that a onrushing car could go down…how do you force the car to go down one tube or the other? (and never under any circumstances slam into the dividing wall)

I think you could have 2 closely spaced “slots” for the same linear motors used earlier, and the two seperate motors would gradually diverge from one another, forcing the car down one tunnel or the other.

Still, junctions like this sound both expensive and possibly dangerous.

[Amateur_Barbarian]

This idea is a perennial, going back at least to the NYC Pneumatic Subway of the 19th century. I’ve seen variations of it in the pop-sci press for decades.

It’s a Wankel engine, a Moller air car, or any of many other great solutions looking for a problem to solve. It’s “different” - and not much else.

[Habeed]

@Sam_Stone

I’m not sure I understand this point. The inverse-square law would cause the force between the ‘ski’ and the slot to rise dramatically if it moved too close - that’s why maglev trains can flot above their track with no risk of hitting it. I think you said the same thing when you say the electromagnetic force will help keep it centered. So where is the alignment trouble? When the pod is first lowered to the track? It won’t be moving then.

There’s a huge aluminum blade sticking out the bottom of the tube car. However, in order to handle turns, the car has to be able to bank so that the G force of the turns is oriented downwards relative to the passengers. This is an essential element of the design, apparently, high speed trains have big problems with curves because passengers don’t like side-ways Gs.

The banking is done either by using a huge control moment gyroscope in the tube car that causes the entire car to tilt sideways, or reaction thrusters that send a constant stream of compressed air out a rocket nozzle to achieve the same effect. (the original paper says it’s one or the other, but he didn’t figure out which method would be better)
Anyways, since it can bank, there’s a risk that it could be in a “banked” configuration when it reaches a part of the track that has one of those linear motors. The massive aluminum blade would slam into the iron cores of the stator coils embedded in the track, at possibly 800 mph.

That sounds bad.

[aNewLeaf]

Switching junctions just don’t seem feasible.
But the system is designed for single hops- you don’t have to branch off at Dallas to head toward Houston.
Instead you get in the pod that was headed from Denver or wherever to Houston. When it gets to Dallas, it brakes and switches, the re-accelerates down the new tube.

[Habeed]

Not feasible at all? You could decelerate to “merely” 300mph or so, then there would be electromagnetic coils that force the blade of your tube car into one configuration or another. They would not be abrupt, 90 degree linkages. Instead, one direction would continue on in a straight line, while the other would have a very gradual curve that slowly redirects your car into another track over a few kilometers.

This is the part that sounds expensive, because it would require a section of tubeway where the overall tube diameter is much, much larger, and special guide pieces inside it, all made to very precise specifications.

[Inigo_Montoya]

Any problems with building a coast to coast prototype–NYC to Seattle–and just pushing freight? Do that for a couple years, to work out some bugs, add a couple switches to divert to SanDiego and New Orleans and work up a safety record before sticking people on it. That could reasonably take a lot of semis off the interstates and free up traditional rail systems for passengers or freight going to more site-specific destinations. In other words, engineer it to be useful first, then try sticking people in it if it seems like a good idea. It’d be a massive gamble, but there are individuals in this country with the cash to make it happen.

[Simplicio]

@Habeed

Not feasible at all? You could decelerate to “merely” 300mph or so, then there would be electromagnetic coils that force the blade of your tube car into one configuration or another. They would not be abrupt, 90 degree linkages. Instead, one direction would continue on in a straight line, while the other would have a very gradual curve that slowly redirects your car into another track over a few kilometers.

This is the part that sounds expensive, because it would require a section of tubeway where the overall tube diameter is much, much larger, and special guide pieces inside it, all made to very precise specifications.

The strength of the system is that you can accelerate up to a speed and then coast. A lot of breaking and re-accelerting kind of kills the point . I like the idea, but its probably limited to fairly strict point A to point B type routes, instead of a lot of intermediate stops or branchings (though at either end you could probably move the car over to a more conventional system and route it to one of several stations in the destination city).

[Habeed]

@Simplicio

The strength of the system is that you can accelerate up to a speed and then coast. A lot of breaking and re-accelerting kind of kills the point . I like the idea, but its probably limited to fairly strict point A to point B type routes, instead of a lot of intermediate stops or branchings (though at either end you could probably move the car over to a more conventional system and route it to one of several stations in the destination city).

Sorry, didn’t mention that part. If your specific tube car is not going to a specific branch, the coils in the track stay de-energized and you buzz right through that special section at full speed.

That would be the ideal solution, at least. It means you’d have to have a routing algorithm that made sure that tube cars that are turning have more space behind them.

But yeah, a junction that could do this might be incredibly expensive.

[Terr]

@Sigene

one technical question that I don’t understand. propulsion of the cars requires higher pressure on the backside of it vs. on the front side. Is this correct?

you need a relative vaccum on the forward end compared to the rear end. How do you get multiple cars going through a tube. Won’t the high pressure side (back) of the first car have to be the lower pressure side (front) of the second car? Wouldn’t the back of each car have to be higher pressure the the back of the car in front of it? This seems like an unsustainable series of increasing pressures.

Any help on explaining that to me?

2 min distance between cars. By the time the car behind you gets to where you were, the (low) pressure would have stabilized. As for the individual car’s propulsion, the turbine fan draws the air from the front (lowering the pressure there) and expels it out of the back/sides (increasing the pressure behind, as well as providing “air bearings”).

[aNewLeaf]

The car isn’t propelled by the compressor fan, there are linear motor elements at regular spacing.

[Kenm]

@Sigene

It won’t happen and here’s why (my opinion)

Too costly to build. He underestimates the true cost
Overselling the capabilities. It won’t go that fast, you won’t be able to get from point A to Point B that quickly.
Because it won’t live up to the hype, there will not be sustainable use to make it profitable.
It will take up a LOT more land than he is suggesting.
Too dangerous. Very little needs to go wrong at those speeds to have catastrophic deadly consequences.
It will be under repairs for a large part of the time, shutting down the whole system. One stuck car, means the whole tube is backed up behind it. This will happen often.

Basically, not enough need, interest (cash flow coming in) to justify the cost of researching, constructing and building.

All of these objections were raised against sewers, trains, aircraft, cars, paved highways, spacecraft, the moon landings, Mars landings, on and on and on.

[Exapno_Mapcase]

@Kenm

All of these objections were raised against sewers, trains, aircraft, cars, paved highways, spacecraft, the moon landings, Mars landings, on and on and on.

All of these objections were also raised toward a zillion proposals that never got built because they truly were too expensive, unworkable, and dangerous. In the real world, a hundred stupid proposals are laughed out of existence for every one that winds up working. And every technology that has ever worked started out small and inefficient and costly and evolved over decades or centuries to get to the point where you can talk about them.

“They laughed at Columbus” is the single dumbest point that can ever be made in favor of something.