Elon Musk and the Hyperloop
jacquesmattheij.com
jacquesmattheij.com
- If you can call your car, it can park farther away. Parking is one of the biggest problems with cars in urban areas.
- Instead of calling of your car, you could just call a car. A self driving car and a self driving taxi are pretty similar, but one can run 24hrs and reduce the parking problem more.
- Computers can do things people can't do. Once enough auto-automobiles are out there, there can be autoauto only "features". The same square footage of tarmac might be able to move cars much quicker. Maybe autoautos can handle 200km speeds. Maybe they can cooperate to make traffic smoother. Maybe they can link together like trains to overcome congestion.
There is nothing quite as good as having a car to take you exactly to and from where you want to go. If self driving cars can really mix with human drivers everywhere they may have a nice smooth path to innovate on gradually. Big vision plans for revolutionizing transport are so centralized, so premeditated.
edit: one more thing. self driving cars interact in an interesting way with public/mass trasport, especially if people dont own their own. It may reduce the demand by competing more directly on one hand. OTOH, it will compliment by providing the last-mile component.
It's terrible to think that people invent cars, then make them driverless, then make them come pick you up from anywhere when you whistle (or press a button on your phone) but the bottleneck that we can't solve is inconsiderate assholes leaving fish and chip rubbish in the back seat.
- reputation systems
- computer vision / cameras / contract law
- price discrimination based on the required state of the vehicle (i.e., premium pricing for a guarantee that a vehicle has been cleaned)
But yeah, I do think we agree :)
In almost two years of being a ZipCar client, I've never had any issues. Yes, the occasional coke can or candy wrapper left behind, but never a mess that I thought to be unreasonable.
If it turns out to be a bigger problem, install a camera that takes a shot when you take possession of the car, and when you relinquish it. Charge the renters credit card for messiness.
Such a system would be good to have anyway as part of a dashcam. One facing forward, and a fish-eye facing the rear window and also capturing the interior of the car. I am hoping that car insurance companies will start offering discounts for cars with installed dashcams.
A tangent of a tangent now, this would be great (though creepy) for public restrooms. Given that there's a physical method, like positioning on the door, that prohibits it from seeing anything while the stall door is closed or locked, I think it would cut down on the seriously depraved messes people leave behind when they know they can get away with it.
In the US we accept 10000 fatalities annually from drunk drivers alone. Vietnam killed nearly 60000 and wounded countless others over the course of 20 years. In just the last 20 we've accepted more than 200,000 alcohol related deaths just getting from point A to point B.
So when people are afraid of a glitch in the system causing a pile up it is worth remembering such tragedies are possible, but it would still be something that could be corrected.
IMO the best application of self-driving cars in an urban landscape is last-mile demand for people who live outside the city, where it's not economical to run buses or trains. But inside a city proper, or even on a busy freeway, having one car per passenger takes up a lot of space that could be better used elsewhere, both while moving and while parked.
(Also, self-driving technology could be applied to buses and streetcars to reduce operating costs and expand frequency and range of service.)
Minibuses don't need the same versatility as consumer cars have. They "only" need to be able to handle anything they can encounter on a predefined route. The route itself could even be optimized for them (eg pedestrian barriers).
And:
- On a long journey, you can read, watch a film, surf the net or sleep. Overnighters become an option.
- Too drunk to drive yourself home? No problem. Too young? Likewise.
- No driver's licence? No problem. A driver's licence may eventually become a niche thing to have, like a pilot's licence.
You can do more with normal cars today, already. (In fact, I've done so.) The economics of those speeds just don't add up for cars.
God, I cannot wait for either hyper loop or self driving cars (I'm sure we'll see the latter first). I live in Northern California and have family in Southern California. We're starting to flip the holidays this year (wife's family for thanksgiving, mine for Xmas). Just went down to LA for a friends wedding and the drive was horrendous.
A lot of factors come into play:
1, as much as I've hated them in the past, a big suv (especially with a kid) is the way to go. We have a VW Passat (wife's) and it was so crammed it was ridiculous. Very uncomfortable and it's not even a compact. 2, I'm over 40 now and it just physically hurts to drive for 7 hours. I was totally sore the next day. Frickin sucked. 3, Directions. I totally need navigation now. Even with the turn by turn on my phone, I still made wrong turns. 4, I hate driving. Hate it. I see it as such a waste of time. I can't imagine needing to drive to make a living. In my late teens and early 20's when I lived in Southern California, I hade to do just that. After I moved to the city I vowed I never would again. 5, Other drivers. I was pretty appalled by some of the drivers on the road. I kept thinking, I can't wait for computers to handle this so that guy driving down the emergency lane wouldn't be able to.
If I could just hop in a driverless car for that drive, it would be great. I could nap, read, compute, converse, etc. Things I can't do now. I'm sure there's going to be some resistance at first by those with a 'god given right' to drive. But after a while, I see common sense prevailing.
I listened to a great talk at The Long Now Foundation by Tim O'Reilly. He was talking about driverless cars and how the DARPA car challenge when it started could only do around 7 miles. It wasn't the algorithms that got that much better (although I'm sure they did) for the Google cars, it was the data set. Google Maps was a huge advantage and the 'killer app' for the latest generation of their cars.
So a lot to look forward to in this arena. I think my kid will grow up with the internal combustion car for him being like the 8-track for me.
Issues that we deal with when driving an automobile are amplified 100x when in the cockpit of an aircraft. Traffic is nearly impossible to spot during the day time without someone like an air traffic controller (ATC) telling you about it. Managing a plane in inclement weather is an effort in bravery and extraordinary situational decision making. Crashes, though statistically unlikely to be fatal, can have extreme consequences for both passengers an innocent by-standers. All-in-all, the experience of flying is just plain difficult. There are dozens of controls to handle at any given moment, and even the distractions are amplified (ever tried writing something down while driving? It happens in planes all the time)
So how did we overcome these issues? Regulations and technology. Just looking at the traffic issue, there are rules for which alititude to fly at for which direction and type of flight you are, rules for entering high traffic areas by verifying you have communicated intentions first, rules for what equipment your plane has to have before entering specific high traffic areas, and strict expectations that pilots will be able to control their aircraft to the best of their ability. There is plenty of redundancy in the system as well.
I would venture to say that landing a plane is much more demanding than driving a car, even in the most severe driving conditions, yet we trust autopilots to set the plane down in even the most severe 0 visibility scenarios (http://www.youtube.com/watch?v=7BUA3EwKpVM). The technology in vehicles is beginning to make sense to implement in the same way we have it present in airplanes right now, but I would love to see the regulatory structure begin to step up and build a reasonable, redundant, and safe system for driving an driver-less car. Then I can see the world you are describing.
And, of course, the skin friction on a 2.5 meter tunnel would be immense. Using a duct friction loss calculator I get 285 megawatts of loss over the entire tube. You need two tubes. At 120,000 passengers/day (HSR estimate), it would take 114 kWh per trip. That's worst than the Model S, hardly an system in which "the fundamental energy cost is so much lower" than a car.
No, the "theoretically fastest way" to go from Point A to Point B is a great-circle vacuum train connecting them. A launch loop does essentially that, but exploits the vacuum above our heads instead. It just fits better.
Any hyperloop theory needs to deal with supersonic speeds – LA-SF as the crow flies in 30 minutes is just under the wire for subsonic speeds. Elon said "under 30 minutes." There are mountains in the way.
Launch loops are interesting. They seem to be even more difficult than tether-based space elevators, which have themselves been unable to work due to material strength:weight constraints on the tether. I'd expect the belt supporting a launch loop to have similar constraints, and the dynamic bending of the belt would raise fatigue concerns.
Agree with your speed estimations and that supersonic or near- speeds have very high drag costs and complicated analysis.
The reason inertial schemes like the space fountain and the launch loop were created is to overcome those material limitations. Aside from the scale, the engineering and material science of a launch loop is relatively straight forward.
The seminal paper, which discusses rotor fatigue and much more: http://launchloop.com/LaunchLoop%3Faction=AttachFile%26do=ge...
It could be a giant hack of course, it could be Elon saying he knows how to do it when he doesn't and getting all these great ideas to pour out. Or he could die and it would be his 'last company' legacy like Fermat's last theorem, sort of "Wow, really elegant way to do that, won't quite fit in the margin here though."
Anyway, what ever it turns out to be I'm sure it will be cool.
[1] My Dad is an engraver with an FFL and gets all sorts of interesting things in the mail. Usually to engrave, but the promotional literature is cool too.
Google "Evacuated Tube Transport" (ETT) or "Evacuated Tube Transport Technologies" (ET3) (trade marks of ET3 Global Alliance, and et3.com Inc.) on google and youtube -- watch the videos to learn of more advantages. Read the first US patent (5,595,543) to learn exactly how all of what Musk claims (and much more) can be accomplished. NOTE: ET3 (tm) is the product of over 200 experts, companies, and institutions in 18 countries. We hope musk joins our efforts to implement ET3. ET3 is optimized to accommodate BOTH cargo and passengers in car sized vehicles. Initial (local) routes will be at 300-400mph, and ET3 will accomplish over 50 times more transportation (ton-miles and passenger-miles) per KWh than the most efficient electric car or train. ET3 cost is less than a tenth as much as HSR (high speed rail), mostly due to much less vehicle mass It takes 1/30th as much concrete and steel to support 1200lb car sized capsules (400lb vehicles that haul the same 800lb payload as a typical automobile) than to support 100 ton locomotives on elevated structure. ET3 has over 10 times more capacity than HSR at 350mph design speed, and 0ver 100 times more at 4000mph. Refs: search “ doi: 10.3969/j.issn.2095-087X.2011.01.007 “ for a peer reviewed paper published in Journal of Modern Transportation; US patent 5,595,543 (many additional patents in the ET3 portfolio)
Now why do I say almost? Let's make one small modification. Let's put a lot of one way flaps in the tube so that it is easy for a puff of air to blow out, but not so easy for air to come back in. There would be leakage, but that is going to be OK.
As each car comes by, it piles up air in front of it that blows out of the flaps. Then the flaps fall back, and maintains a partial vacuum. The partial vacuum is no problem for people because there is a pile of air in front of their car that can be tapped for breathing air, that can then be released backwards, where it circulates through the tube (and probably out the flap).
This makes his evacuated tube comment even more of a teasing joke. No, the tube is not evacuated. Nor did you pump air out of it. But it winds up almost evacuated. However it is still fine for breathing.
Now the point is that reduced air pressure inside of the tunnel significantly reduces drag caused by the air dragging on the edges of the tunnel and being pushed by the cars. This does a lot to make the whole thing massively more efficient. Elon's claim is then that it is efficient enough that it can be powered by solar panels placed on the tube.
My big question is how hot it will be. There may be very little gas in the tube, but that gas will be very, very hot. Over time the cars will heat up as well. So you'd need to have the cars regularly coming in and out of the system so that they would have time to cool down.
The practical difficulties in building this are immense. But I do not see any physical reason why it is impossible.
As for the flaps, you're right, I didn't think that bit through, I'm not quite sure what the best solution would be there. But if that's the only thing wrong with what I sketched I'll be pretty happy.
But I'm sure Elon has thought about that aspect and it is solvable.
During the acceleration phase this would definitely be an issue, the gas in front of the carriage would be compressed because the carriage is moving much faster than that body of gas. But that (hot) gas could be dumped just prior to insertion of the carriage into the tube.
Maybe the outside of the 'sheath' could be made from a heatsink like profile? I haven't given the thermodynamics of the thing much thought, I was basically just trying to figure out a solution that would not require some far-out technology.
I know that lots of people like to link space with Elon Musk but I think he's much more practical than that, space is a means for him to get to mars and Nasa a convenient way to finance it.
Other than that Tesla doesn't have anything to do with space and is ground transportation, I don't see any reason to go into low earth orbit with all the associated risks if you could do the whole thing on the ground and be:
- safer
- as fast
- not have the energy expenditure of having to work
against gravity
- easier to maintain
- shorter path
That's mostly a surface area increase problem, the air over a highway is notably warmer than the air around it simply because all those cars dump the combined energy of burning all their fuel into the environment. If you can reduce the energy input into the system then you'll reduce the amount of energy you have to get rid of afterwards. So I can see that reducing that input energy is really important and that changing the speed of the cars is brutal in this respect because it causes the air behind it to pile up against it (heating it up) when decelerating (but the air in front would cool down due to expansion) and vice versa.
It's an interesting problem and if I wouldn't have a ton of other stuff to do I'd happily cook up a simulator for this.
I'll do a blog post later today with it. But now I have to get my daughter to school.
One issue with a tunnel is that even if the air and carriages are moving at the same speed, there is still drag or friction between the air and the wall of the tunnel by Poissiulle's law. At the proposed speeds (~300km/h) and distances (600km), this becomes a lot, and higher pressures (if my quick calculations are right) lead to higher energy requirements through higher air density.
In a vacuum, this friction would not exist. I can post my calcs if there is interest - I used a Moody chart and Darcy's friction equation, and ended up with an energy requirement that there would have to be ~80 million carriages going each way to be as efficient as a Telsa roadster, and neglecting any other losses.
The limiting case is when the whole tube is full of carriages, say they're 4 meters long that would mean there are a maximum of 150,000 carriages in that 600 Km long tube at absolute maximum capacity, something that you probably should not want to get close to.
More realistically, a regular roadway has a spacing of about 30 meters per vehicle at 60 Mph, you might be able to pack them in that tight in a tube like this but that's still pretty tight and leaves very little room for error when shifting carriages in and out of the loop.
At that spacing you could stuff (600,000 / 34) = 17650 carriages in (none of those assembled into impromptu trains, which of course would increase the density).
If we assume train like assemblies of 10 cars with 30 meter spacing you'd be looking at a 40 meter train + spacing is 70 meters, or 8500 trains, so 85000 carriages.
Those would then be pushing a column of of air 30 meters long ahead of them.
I'm not sure what the figure should be but this seems very high.
Consider if each carriage is pulling 30m of air a long distance at great speed. Consider if instead of the air moving against a stationary tunnel, a 30 meter section of tunnel was moving at the same distance and speed through stationary air. The drag on this would be high, much higher than a car at highway speeds.
Consider a tub of water moving at the desired speed...
If that works, you could then have "station" sections where you could place/remove "inner" carriages to/from the carriages running inside the loop and thus you'd never have to disconnect the main carriage loop.
Hype-er loop.
The big issue would be air traffic control at launch and landing. 600 m/s (Mach 2) at 0.5g acceleration would require 36km of launch rail which is kind of a lot. But you could loop the track and reduce the acceleration to make up for centripetal forces — a 200m radius loop might be about right, and once you use a loop you can go a lot faster (and this explains the name: hypersonic loop).
I suppose it might skip short distances requiring pylons or something for speed top ups and travel at lower speed.
Digging a tunnel takes years and cost billions (see London underground new tunnel http://www.bbc.co.uk/news/business-16320945)
Making the tunnel fit the carrage = no room for emergency
Security (terrorists + bomb = nightmare at speed of sound)
Windows? What would you look at?
If he thinking of just goods (not people) then some of these problems are much simpler - the tunnel could be less than 1m in radius.
My wild idea is that he is going to us a rail gun to fire drones up a couple of KM into the air, these then glide down to mini airports. Replace freight railroads.
By this logic Planes are a bad idea too (no room for emergency, high speed, tiny windows with nothing to look at)...
They don't like power-lines in the country side.
They certainly wouldn't like a 2.5M+ diameter tube running around. In London, 56% (IIRC) of the "Underground" is above ground, much of that above road level.
There is a farmer who has 2 turbines up, he lives on the far side of a hill about a km away - you can still hear those whoosh-whoosh-whoosh all day long. Painful.
Powerlines also run over a lot of campsites near my house, you can hear them hummmmmm and crackle. In the damp/rain/mist/fog you can hear a lot of crackling and humming from them.
I can understand why people get angry with them. In urban areas where there is a lot of noise, I guess you wouldn't notice it so much.
Off topic, but my sister worked in a place where they put the magnets into wind turbines. There was a loud crash one day, someone had a magnet on a pallet, lifted it with a forklift and drove it out to a van. Half way the magnet got sucked by another magnet (in one of the factories) and burst through the wall and the two clamped together. They never got them apart and had to take the remaining bits of the turbine apart from around them.
What city was this in? Which manufacturer?
0: http://techcrunch.com/2012/11/19/elon-musk-with-jobs-gone-go...
Now I'm really curious...
Smaller landing zones, no air pollution, much less noise...
It's also a good combination of Telsa (rail gun) and SpaceX (rockets can be seen as motorized slugs).
Only limit I see with this system is that I'm not sure anyone could withstand the acceleration.
Centripetal acceleration is a = v^2 / r so for 300 m/s (about mach 1) speeds, for 1 gee centripetal acceleration you'd need 9 km radius.
Would be interesting if some of them could be resurrected somehow.
Another idea for improving transport: with modern technology better routing should be possible. Instead of all people boarding the same train that stops at every station, why not only board a carriage that goes directly to your destination? That could save a lot of time, I think.
I just had another thought, although maybe a bit ridiculous: the main issue with "public cars" to me seems to me hygienic, how to keep them clean. What if in the future instead of owning whole cars, you only own a capsule with seats that can be picked up by transporters? I'd still prefer a world where parked car things are completely gone, but such capsules might still save a lot of space in the meantime.
An evacuated tube has the advantage of being much more stable at high speeds and avoids the issue of excess heat from repeatedly compressing and expanding the air in the tube.
In any case, great read. I almost wonder if Musk has thrown this idea out half-baked just to get more people to start thinking outside the box about transportation...
Using Halbach arrays would do it, without any active control.
waiting for passenger
SF [||| <> - - - - - - - - - - - - - - - ] LA
pressurized --> pretty spaced out "stuff"
SF [-| - |- |- |- <>-| - |- |- -| - |- |- ] LA
in transit..
SF [ - - - - - - - - - - - - - - - <> ||| ] LA
<- presurized for returnI always imagined Hyperloop as carriages pushed by a plasma armature, in a rail gun loop. just like a plasma-rail gun.
I suspect that a fully evacuated system using magnetic propulsion and energy recovery would be more efficient - the turbulent nature of air and the low-quality nature of heat energy would probably work against you to make a pneumatic spring fairly inefficient.
Specifically, due to a specific statement that it is not an evacuated tube.
Yacht designers refer to the "wetted area" component of drag, and an entire tunnel wall is a heck of a lot of wetted area.
This is a long story, but I'll try to keep it very short. Anyone interested contact me off-list for far more details and an unfinished paper with some of my research.
A couple of years ago my son and I were watching a documentary on the subject of concrete. It was very interesting. They covered a wide range of topics but one of them really started to trigger my curiosity: The Panama Canal.
I don't know why, but I became very interested in the financial metrics surrounding the canal. How much does it cost to cross it? How long does it take? How many ships cross it per year? Where do they come from? Where do they go?
The more I learned the more the reality of the Canal horrified me. Why?
The bulk of the commercial traffic through the Canal are container ships. And these ships burn something very nasty: Bunker Fuel. This is, by almost any measure, the dirtiest fuel you could burn. it's horrible stuff.
When I started to do the math I started to realize the magnitude of the problem. These ships move at about 20 miles per hour. They could go faster but there's a balance between the high cost of hydrodynamic drag and fuel costs. A trip from Shanghai to Long Beach takes about 18 days and will burn somewhere in the range of 3,600 to 7,200 metric tons of fuel. For those not comfortable visualizing units in the metric system, that's from 7,936,560 to 15,873,120 pounds. Yes, fifteen million pounds of the nastiest crap you could burn is used to bring your iPhones (conjecture) and other stuff from Shanghai to Long Beach.
If my research is correct, the fleet of about 100,000 cargo ships (Yes, 100,000!!!) burns over a million metric tons of bunker fuel PER DAY.
400 million metric tons of bunker fuel per year, which is equivalent to 120 billion gallons.
Can't relate to that number?
Here's an interesting comparison:
To get a better sense of how large this number is we can try to relate it to how many cars one could fill-up with fuel and for how long. 120 billion gallons would provide enough fuel to supply 100,000 cars (assuming a 20 gallon tank) with a full tank of gas every week...for over 1,000 years.
100,000 cars.
20 gallons per week.
For a THOUSAND years.
The evil, when it comes to pollution and energy dependence, isn't the much-abused light bulb; it's the elephant in the room: Ocean-going cargo ships.
While our mass media chooses to focus its attention on an oil spill (because it is sensational and it serves political purposes), what is really killing our planet slowly is the transportation of iPhones, Blackberries, TV's, blenders, washers, cars, widgets and gadgets on inefficient and highly-polluting ocean-going vessels. Even the latest Gulf spill is insignificant in terms of environmental impact when compared to what 100,000 ships are doing to our environment each and every year.
It is estimated that the fleet of nearly 100,000 cargo ships in the world produces over 20 million tons of Sulfur Oxides (SOx) per year. For comparison, the entire fleet of automobiles in the world (about 800 million cars) produces about 80,000 tons of the same contaminant.
How about the Canal?
A container ship traveling from Los Angeles to NYC through the Canal will burn about 4,500 metric tons of buker-C fuel. This amount of fuel costs approximately US $1.8 million. Canal fees would run somewhere around $300K. The trip from L.A. to NYC through the Canal runs well over two million dollars, without including handling, insurance, crew costs, amortization, maintenance, etc. That's quite a chunk of change, however, when divided by the thousands of containers a ship can move it becomes a few hundred dollars per container.
How many ships go through the Canal per year?
Approximately 15,000.
I'll leave you to do the math. I have far more detail in my notes. What these ships are doing to our environment is simply horrific. The pollution doesn't stop at the act of burning fuel.
Cargo ships are also the source of an unusual form of pollution. Ships use huge ballast tanks to stabilize themselves. These ballast tanks are filled and emptied of sea water during loading and unloading operations at port. It is through this mechanism that cargo ships are responsible for transporting harmful organisms across the world into ecosystems that cannot handle them. The introduction of non-native species into a new ecosystem can have devastating consequences.
And so, from watching a simple documentary I came to the realization that, for some strange reason, we have been ignoring the most significant source of environmental pollution on our planet. And, beyond that, one of the largest --if not the largest-- consumer of petroleum products.
I didn't stop at just identifying the problem. I also wanted to take a stab at a solution. I came up with something I called "The American High Speed Cargo System" (AHSCS) as a loose proposal. This would be a cargo-only, electric powered, high speed rail system. It would connect --at the very least-- both coasts and, ideally, other major US ports. The idea would be to move cargo over land from port to port at 200 miles per hour. High speed passenger trains in the US are a waste of money and that's particularly true in California (don't get me started there). Not so for high-speed cargo.
The numbers are there to support it: A cargo ship spends over two million dollars to get from L.A. to NYC. Probably closer to three. Those same containers could be moved far more efficiently over land, at similar or lower costs and pollute far, far less. You are exchanging aerodynamic drag for hydrodynamic drag. Huge difference.
In terms of energy costs (just the electricity), I came up with numbers in the order of $10K for a trip from L.A. to NYC. I further estimated that the system would require around 700MW of power, let's call it 1,000MW. We have 53 nuclear plants that can source 1GW each. This is a case where nuclear power might be a really good option.
However, the scope of the project needs to be realized. Developing and building such a systems has the potential to generate hundreds of thousands of jobs, if not millions. It should be revenue neutral if not positive (sorry Panama). It would allow for the installation of upgraded communications and power backbones that would be synergistic to the process of building the rail system. It would also allow for the potential to install huge solar and wind-power farms to fully or partially power the system.
I have not explored every angle but would like to think that, if my numbers and assumptions are right, this could be the most important project this nation could embark on. You have to think in terms of a hundred or two-hundred year scale. These ships are not going to go away unless something very significant changes. Of course, the same concept ought to be replicated across the planet. Again, if I am right, we should strive to eliminate most, if not all, container ships traversing our oceans. We are making an absolute mess out of our planet.
http://news.discovery.com/tech/shipping-network-map.html
Like I said, there's more. If interested email me off list and I can send you a copy of my notes so far. It'd be interesting to have someone go over my notes and verify my assumptions and calculations. I tried to raise the issue with politicians but, what can I say, I only have so much time to deal with morons.
NOTE:
I thank you for your comments. I have to ask that you do me a favor. Please refrain from making categorical statements about the relative efficiency of ships vs. a proposed high-speed electric train without having done the math yourself. Please drop me an email and I'll be more than happy to provide you with a copy of my calcs, an unfinished paper as well as links, PDF's and references. Then we can talk about the merits of the concept. I am actually very interested in having the concept, calculations and assumptions criticized. Arguing outside of a common frame of reference is rather difficult.
https://en.wikipedia.org/wiki/Panama_Canal#Third_set_of_lock...
http://www.siteselection.com/issues/2011/nov/eastern-seaboar...
Here's a back-of-the-envelope calculation for moving a 40-foot shipping container over rail from NYC to LA: 3 cents per ton-mile * 30 tons * 2800 miles = ~$2500. I couldn't find a quote for that route by sea, but from NYC to London I got a $1750 quote, and I imagine the costs are comparable.
I believe you that container ships are huge polluters. It sounds, though, that it'd be easier to fix this by taxing those externalities, rather than making two arguments simultaneously: that container ships are bad and that we should build a giant high-speed domestic freight rail network.
Take the numbers above with a huge grain of salt, but here are my sources for them: 3 cents per ton-mile: http://www.kehoe.org/owen/portfolio/truck_vs_rail.pdf a 40' container weighs 29kg loaded: http://hamburg-sud.com/hsdg/en/hsdg/servicesproducts/contain...
One Panamax size ship moves 4,400 TEU (container units). Larger ships can do up to 15,000 TEU. If you even tried to move that many containers using conventional trucks or diesel trains you'd probably create more pollution than the ships, not to mention the overloading of the roads and rail.
One of the reasons for AHSCS having to be a high-speed system is that you need to bring speed into the equation in order to deal with the sheer magnitude of the cargo moving about.
The other reason is that you need to be able to offer something competitive in order for shipping companies to choose to go over land with AHSCS versus through the Canal.
The third reason is both economic and geopolitical in nature. By traversing the American continent quickly (~3,000 miles in a day) the system would become competitive for cargo that goes from, say, Shanghai to London/Europe through the Suez Canal.
If, again, my numbers are correct, it would be cheaper and less risky to send Europe-bound cargo from Shanghai to the US West Coast for transportation across the continent using AHSCS and then by ship to Europe. You'd be able to get the cargo there in about the same time or faster, at a lower cost and with far less pollution because you'd eliminate days of bunker-C burning transportation from the process.
In the process you'd also significantly modify the economics of the petroleum economy in the Arab world and isolate the rest of the world from the perils and political issues surrounding the Suez Canal.
As I said in my note. If something like this makes sense from an ecological and business perspective there ramifications go far, far deeper than the simple idea of moving containers by land on an electric high-speed train. It could be a game changer for the next couple of hundred years.
http://www.inference.phy.cam.ac.uk/withouthotair/c15/page_95...
http://www.inference.phy.cam.ac.uk/withouthotair/c20/page_13...
POWER to overcome drag is proportional to the CUBE of the velocity. This is the main reason container ships have been slowing down from about 30 miles per hour to 20 in direct response to the cost of fuel.
The power to overcome drag is also directly proportional to the density of the fluid.
Air density: 1.2041 kg/m^3
Water density: 1,000.00 kg/m^3 (pure water)
Here's a practical example: Let's move a Toyota Camry through air and water at 20 miles per hour and compare. We'll ignore friction.
http://www.toyota.com/camry/specs.html
Frontal area: 57.9in x 71.7in = 2.68 m^2
Coefficient of drag: 0.28
Power = 1/2 * density * velocity ^ 3 * area * coefficient of drag
Power in air =
1/2 * 1.2041 Kg/m^3 * (8.9 m/s ^ 3) * 2.68 m^2 * 0.28
= 0.43 horsepower
Power in water =
1/2 * 1,000 Kg/m^3 * (8.9 m/s ^ 3) * 2.68 m^2 * 0.28
= 353.02 horsepower
How about now moving up to 40 miles per hour?
Power in air = 3.4 HP
Power in water = 2,824.14
If we moved the Toyota Camry through air at 188 miles per hour we would need nearly exactly the same power required to move the same car through water at 20 miles per hour.
Bottom line, moving anything on water is very expensive if you have to burn fuel to do it. Unless I am missing something very fundamental there simply is no comparison.
1. Assuming the same speeds. Ships go slower than trains (say 40kmph compared to 80kmph). 2. The frontal area is not the same as only part of the ship is underwater. Further, for a long train, the frontal area drag is likely less than the skin friction drag. 3. The coefficient of drag is not the same as the CD of a train is different than for a boat. 4. The amount of cargo on each is not the same, and would need to be normalized for a fair comparison.
I'd be happy to take a quick look at your materials off-line, please send to jd2planks at hotmail
With regards to the rest of your comments, I've done some of those calculations. Again, there is no comparison.
I need to repeat that the idea here is NOT to in ANY WAY use existing truck or cargo rail systems. They just don't perform. Here in the US you are lucky if you can get cargo trains moving on average at 20 to 30 miles per hour. In other words, the move just as slowly as cargo ships, except that they can't pull the same number of containers.
The idea is that a new purpose-built system is the solution.
Also, please keep in mind that the idea here isn't necessarily to search for the most efficient mode of cargo transportation in complete isolation of the realities of where they derive their energy from.
Here's a very interesting paper on the aerodynamics of high-speed trains:
http://ompldr.org/vMXlsOA/Aerodynamics_HSR02.pdf
Anyhow, I sent you all my files via a shared Dropbox folder. I'd be interested in what you have to say once you have a chance to look at it all.
To be sure, there are huge issues with the idea that have absolutely nothing whatsoever to do with science and technology. I could not imagine the problems one might encounter in terms of the various special-interest groups that would get in the way of building such a thing, even if it was so good that it would save the world. I think the phrase "can't get out of our own way to save our lives" rings true in more than one way, at least in the US.
I'd be happy to look at your calculations. My knowledge of hydrodynamics is somewhat limited, but I do have a fairly good understanding of aerodynamics and a lot of that stuff is transferable. I can't decipher your email riddle though - is [4:5] saying the 4th and 5th characters? Mine is leorampen (at) gmail.
The camry example was just an illustration of the power relationship of something moving through water. Nothing more than that.
The comparison of a container ship with 5,000 TEU (2,500 40 foot containers) vs. a train is, of course, different. And, of course, it isn't as simple as looking at cross-sectional area. That would be a gross oversimplification. Sometimes you have to assume that a cow is a uniform sphere of milk just to start to have some numbers to talk about.
Imagine a train design to carry containers at 200 miles per hour. Single file. So, now, the cross sectional area is that of a single container, which is very small. If the train is designed to minimize parasitic drag and all of the frictional components of the rolling system it could be incredibly efficient when compared to a ship pushing some 4000 square feet (400 square meters) of cross sectional area of water at 20 miles per hour.
And, as I have said multiple times, this isn't really solely about pure energy efficiency as much as a forward looking idea to that hopes to deal with the grotesque level of environmental pollution these ships produce as well as what might happen over the next 100 or 200 years with fossil fuels. One really has to start looking in at least a 50 to 100 year time scale to understand that what we are doing today is not sustainable in any way at all. To some extent, arguing the finer points of how much one displaces vs. the other and whether one solution is a few percent more efficient than the other is to miss the point.
I'll go ahead and send you a link to the data.
Unless it's a submarine, a ship doesn't move through water, it floats on top of it. The wetted surface is going to be a tiny fraction of the total frontal area. 90% or more of the hydrodynamic drag of a ship (by area) is in the air, not in the water.
For example, here is a nice little 21-foot 1200-lb motor boat that can reach 16 knots with a 25 hp motor, while carrying four people: http://www.southportislandmarine.com/ordereze/1011/page.aspx
The nice little boat you are linking to can do what it does largely because at a certain speed it will start to plane. These are very, very different conditions.
Two of the components of drag on a boat are form and parasitic or frictional drag. The first can be thought of as simply the drag created to push water out of the way. Here the shape of the boat is important. The second is the drag created by water "rubbing" against the portion of the hull in contact with it. Think of sandpaper against wood. More wood and more sandpaper means more resistance to motion.
As speed increases both friction and form drag increase with the square of the speed.
On a planing hull --one that can generate lift and get the boat up and partially out of the water-- form drag decreases significantly once you are "on a plane" (off the water) and friction drag increases. As it turns out, each boat has a sweet spot where it operates best. And, of course, as you get on a plane the power required in order to reach or sustain a given speed decreases significantly when compared to a displacement (non-planing) hull.
A Panamax ship carrying 4,400 TEU does not plane. It is, as far as a know, a pure displacement vessel. So, yeah, you need a tremendous amount of power to counteract significant components of both form and parasitic drag.
A Panamax container ship can have up to nearly 400 square meters of cross-sectional area under water. That's about 4,000 square feet. So, while it isn't exactly the same thing, only for the purpose of visualization, imagine a 100 foot long x 40 foot rigid piece of plywood fully submerged in water. Not try to push it forward at 20 miles per hour. Right. That's why they have to burn nearly a million and a half dollars of bunker-C fuel to get from Los Angeles to NYC.
Physics can be really inconvenient sometimes.
Also according to the US Transportation Energy Book, waterbourne transport more than a order of magnitude more efficient than road freight, and about 50% more efficient than rail.
http://en.wikipedia.org/wiki/Fuel_efficiency_in_transportati...
On the second point. Our cargo rail system is crap. Nowhere in my write-up am I proposing that we unload 100,000 cargo ships per year and load these containers onto diesel trucks or the current train system. That would be incredibly dumb, inefficient, more harmful to the environment and, well, actually impossible.
No, the idea is that a new highly-efficient, all-electric, highly-automated and cargo-only transportation system be built with the express purpose of moving container-based cargo from coast to coast (and port to port) at the highest possible speed and lowest possible power consumption. The idea further suggests that the required energy can come from nuclear, wind and solar power.
The complete elimination of the burning of the most harmful form of fossil fuel from the transportation system is one of the stated goals of the idea.
Furthermore, the idea is also to look 50, 100 or even 200 years into the future. Pollution aside, fossil fuel cost is absolutely, positively guaranteed to increase over the next 200 years. The current system of burning millions of tons of bunker-c fuel per year to bring widgets to Walmart isn't a viable and scalable long-term solution. Not to mention the fact that it is royally fucking up the environment at many levels. I am not a crazy environmentalist by any measure, but this is stupid.
The US freight rail system is the best in the world.[1] We move 10x more than Europe most years on a tons times distance basis. The running gear may look antiquated, but the fundamental numbers are staggering.
"No, the idea is that a new highly-efficient, all-electric, highly-automated and cargo-only transportation system be built with the express purpose of moving container-based cargo from coast to coast (and port to port) at the highest possible speed and lowest possible power consumption."
There is very little demand for raising the speed of freight rail in the US. Speed generally compounds inefficiencies.
Freight handling is by and large already automated and container based in the US rail network.
Existing electric designs preclude the use of double stacked container flat cars, one of the keys of US freight efficiency. Perhaps you have an innovation here?
Building an entirely new high speed electric rail network for freight is an absurd idea. I don't think you've even begun to model the economic realities, particularly the capex of building a new rail network radius'd and graded for high speed, with all above grade crossings.
What I meant is that it could be much, much better. The article you point out hints at some of the issues: The inter-mingling of passenger and cargo trains ends-up seriously affecting the potential performance of cargo trains.
With regards to it being an absurd idea. Well, of course it is! And so is the case for nearly every idea that sparked a revolution in technology and thinking. The earth NOT at the center of the universe? Absurd! The earth not flat? Absurd? Everyone needs a computer on their desk? Absurd!
Not that I am comparing this to any of those shifts in technology. What I am trying to point out is that being open minded enough to even consider the seemingly absurd is an important part of discovering new ideas and developing new technologies. I always consider the absurd. I have hundreds of ideas catalog that are probably absurd. The alternative is to discount everything as ridiculous without even giving it an honest look. I just don't do that.
For example: Could this high speed train take the form of an underground tunnel connecting both coasts? This is absolutely ridiculous and probably even insane to consider. However, I take the approach of at least doing the research to try to get a sense of proportion in terms of cost and technical feasibility. For example, would there be an important mining element to the project that could offset the costs? I don't know and I am not trying to argue for a 3,000 mile tunnel. All I am saying is that I believe that it is very important to give seemingly ridiculous ideas a chance to be explored and, if need be, discarded based on data and not knee-jerk reaction.
This probably deserves a little more attention.
Having done a lot of sailing as well as ocean-fishing on different size vessels I know exactly what you are talking about. What you might be confusing is that at zero speed there's a huge difference between something floating on water and a truck resting on the ground.
In the case of a large boat floating on water, so long as you are not after making it move quickly you are almost just fighting inertia. So, yes, a fifteen year old kid pulling on a rope could make a large boat move.
When it comes to a truck sitting on the ground you have to contend with rolling friction as well as the static friction (because nothing is moving) and the viscous drag of all the mechanical elements, oil and grease in the drivetrain. It could take hundreds of pounds of force to get it to budge. Once your break through static friction then you have a much reduced dynamic friction and viscous drag (oil and grease in the drivetrain) --again, so long as speed is very, very low-- as well as the rolling friction of the tires.
If, on the other hand, that truck was resting on a frozen lake bed and if the tires are not frozen-glued into the ice, and, if the ice is not deformed under the tires, the same fifteen year old kid could move it just as well as he moved the boat. In fact, it might even be easier because there isn't a significant v^3 component to the power required to make it move.
Once you start to add speed to the problem it is easy to see that moving the same object through water will require at least 830 times more water.
At speed the power required to counter drag comes into play.
Power = 1/2 * fluid density * velocity ^3 * coefficient of drag * area
If it takes one horsepower to move something in air at a given speed you'll need over 800 horsepower to achieve the same speed in water. It's the difference between a model airplane engine and a Formula 1 race car engine.
There's a good article in Wikepedia on drag:
http://en.wikipedia.org/wiki/Drag_(physics)
Get down to the "Power" heading. That's the formula you want to look at.
To get back to your yacht vs. truck example. What you would really have to experience is pulling each of them at 20 or 30 miles per hour with a rope (chain?). Of course, it would take super-human strength and endurance to be able to do that.
The only thing that could make land-based transport less efficient (other than fuel and power-plant issues) is rolling and drivetrain losses. I doubt that those make-up for the over 800-to-1 power discrepancy.
The greater point here isn't a pissing match between individual efficiencies but rather to look at the problem with a long (50 to 200 years) perspective in terms of pollution, fossil fuel consumption, geo-political and economic concerns and more.
One interesting article says that one of the mega container ships pollutes more than 50 million cars see:
http://www.guardian.co.uk/environment/2009/apr/09/shipping-p...
The claim is that just 15 of the worlds biggest ships pollute more than all cars in the world combined.
More recent report from Europe comparing different transport modes for CO2 by kg-km traveled. Shipping is reported as most efficient over rail and road. http://www.eea.europa.eu/data-and-maps/indicators/energy-eff...
If you've not been, I highly recommend a trip to Panama City to checkout the canal - it's simply mind blowing watching a super tanker go through the locks.
Do you have any data on the number of container ships that move through the canal that are simply moving from one US coast to the other?
There's a webcam, but it's not a live stream.
Keep in mind that there's a lot more than container ships that also move across the canal. Tankers being one example.
Direct wind power would seem to beat out most of your proposal for land-based transport, which, even running on renewable energy, would require a double-conversion to electricity and then back to motive force. It certainly beats out nuclear!
Yeah, I know it's a long-shot, but it fits into my idea of Musk better than a giant mail tube. I'm just seeing pressurized tubes scaling. There's the same problem with the orbital sub-loop, but I'm betting Musk spent a lot of time looking at this idea as part of his Mars dream.
One thing's for sure -- it's going to be a blast seeing how it all turns out!
Its just maglev trains with some novel enclosed design with carriage dissociation?
A bit like this chinese train concept that has drop off/pick up of carriages:*
[0]: It could still have wings but operate under ground effect, though it seems that it would be difficult to keep your leading edge intact travelling just above a hard surface "twice as fast as an airliner". "Twice as fast as an airliner" is Mach 0.85 * 2 = Mach 1.7 at FL360, which translates into Mach 1.5 at sea level. One useful thing we learn from this is that this system will basically have to be propelled by compressed air or electromagnetic force, since props don't work over Mach 1 and we don't have electric turbines.
So maybe it is like a train, one person per, but it runs on cheap road like track, and it's propulsion is provided by fixed stations. If each track was a straight run from station to station you could just ditch the wheels and whatnot and fire the thing between stations along the ground.
There's no way it could move so fast and still use car roads, unfortunately, so it will still have problems buying track space or building tunnels or building elevated paths. Too bad it runs on the ground.
We could look that up. But considering that roads and rails the world over are often build by governments or government owned corporations, I expect Elon Musk to have found a much cheaper way to build his infrastructure.
[1 ]http://en.wikipedia.org/wiki/Swissmetro (the German language article is much more extensive: http://de.wikipedia.org/wiki/Swissmetro)
You'd need on-ramps, where a minimum speed is required before merging with the main line, and of course you'd only build on/off ramps at major hubs. The only issue I see being you're not using tar then, and therefore costs of raw materials would be higher, at least initially, and would likely last longer than tar.
Being cheaper than highspeed rail could fit into this equation because its the vehicle owners paying for the vehicle, and no trains are being build for it - so actual money going into the system, the synergies that would exist, might be greater - though putting it how he does is creating lots of attention. :)
Maybe the hyperloop refers to a an on/off ramping system, where you get accelerated to a certain speed... Fun speculating. And time for tea and breakfast.
The idea is that rather than being a straight line with a sort of conveyor belt, the walkway is made of plates like an escalator, and the ends of the walkway curve off from the main body of it. This means that a user steps on to a slow moving plate, immediately goes around a corner on it, and in doing so accelerates to a much faster speed, with the opposite happening at the destination.
As soon as I heard the word 'Hyperloop', this is what I thought of. It's a looped system, but with an extra dimension in a way as different sections operate at different speeds. I think this could be scaled up to be perhaps a track system that 'cars' are put onto with passengers inside, but I don't know.
Is this a reasonable possibility? Maybe.
✓ Ground based
✓ Weather independent
✓ Like a railgun (if propelled with magnets)
✓ Is not a pressurised tube
✓ Leaves when you arrive
? Could hold solar panels
? Cheap
? Revolutionise the transport industry
✗ No rails - depends on your interpretation of this, one could argue that a pressurised tube is a kind of track or rail for a carriage.
If it's not a tunnel, it's still not an <i>evacuated</i> tunnel.
But if it's "a not evacuated tunnel", then it must be a tunnel.
Edit: Actually, I think with the article "an" between the "not" and the adjective, there is no way in English grammar to interpret the "not" to negate only the adjective instead of the whole adjective+noun "evacuated tunnel". Unfortunately I'm neither a native speaker, nor a linguist, so I cannot back it up with anything but feeling for the language and the fact that this is how I always see it used ;).
Well, if you add another "only" before "interpret" then you'd be correct. Otherwise, it is open to interpretation either way.
!(evacuated && tunnel) == (!evacuated && tunnel)
There will still be resistance though, both from the walls as well as leakage between the carriage and the walls (there can't be a really good seal there or the friction between the carriage and the wall would become too large, and since the wall is stationary and the carriage is moving very fast this friction is likely sizeable).
However, all the most plausible theories I have heard so far, and my own possibility, all rely on some sort of (although not traditional) rails.
How does this prediction meet that requirement? Or am I missing something?
This is the quote I'm referring to:
"It also can’t have a right of way issue, where people have to give up their homes."
http://techcrunch.com/2012/11/19/elon-musk-with-jobs-gone-go...
Unless there is a recycling of roads somehow.
The thing that bugs me about the whole 'space' angle is that if that is the solution it is only usable between endpoints, and roads are usable for everybody (also those people living between the endpoints).
You then get the same issues that you have with airports, you have to actually go to one of very few terminals in order to get on the loop. And ground transportation converging on the loop would be the bottle-neck (think highway LA - SF with only one on-ramp).
So likely we're missing a huge piece of the puzzle still.
(I expound at greater length here: http://franking.tumblr.com/post/36241325898/my-personal-spec...)
expanding this system to long distances would be really interesting, however even more expensive, given the need to construct long tunnels or tubes.
Techwise, I still think the Lofstrom loop theory sounds the most likely: http://en.wikipedia.org/wiki/Launch_loop
The purpose of new innovations is to improve something. It's possible that in the very short term something like wouldn't be the most cost effective thing, however it's not about the very short term, but rather about the potential. In most existing transportation systems, the potential of the system without large changes has probably mostly been realized. The potential of something like this may be much larger, which is why it would be worthwhile.
On open roads, where batteries fall down on range, surely this would make sense? Combine with self drive for easy town to town driving experience.
Very nice article.
That's assuming the mass needs to be on a second track to maintain a schedule. I guess it's probably possible to do something clever at the stations to avoid a second track.
I'm not really sure how it "can't crash", nor how there aren't any "rails" (or things that look suspiciously like rails).
Try "For a while now there have been tantalizing hints that Elon Musk is at it again"
Or perhaps better: "There have been tantalizing hints that Elon Musk is at it again for a while now"