Ford introduces more fuel-efficient Focus 1.0L EcoBoost; 99 gCO2/km
Tongji team estimates vehicle emission factors in Shanghai with MOVES and GPS data

Elon Musk unveils Hyperloop preliminary design study

Linear accelerator concept for capsule acceleration and deceleration between 300 and 760 mph (480 and 1,220 km/h). Click to enlarge.

Tesla Motors and SpaceX CEO Elon Musk released the preliminary design study for what he calls “The Hyperloop”—a new high-speed electric transportation system targeted for the the specific case of high-traffic city pairs (e.g., San Francisco and Los Angeles) that are less than about 1500 km or 900 miles apart. (For longer distances, Musk suggests, quiet supersonic air travel would be faster and cheaper.)

Hyperloop—which is an open-source concept, user feedback is welcome—consists of paired partially-evacuated tubes (0.015 psi, 100 Pa), with passenger capsules (or pods) that are transported at both low and high speeds throughout the length of the tube. The capsules are supported on a cushion of air, featuring pressurized air and aerodynamic lift. The capsules are accelerated via a magnetic linear accelerator affixed at various stations on the low pressure tube. Stators are located on the capsules to transfer momentum to the capsules via the linear accelerators.

Hyperloop concept sketch. Click to enlarge.

To overcome the limiting problem of air building up in front of the traveling pod, Hyperloop proposes mounting an electric compressor fan on the nose of the pod that actively transfers high pressure air from the front to the rear of the vessel. The mechanism would also create an air cushion that would create a low-friction suspension system.

Passengers may enter and exit Hyperloop at stations located either at the ends of the tube, or branches along the tube length. The capsules may be passenger-only, or, for more flexibility (and at a higher system cost), passenger and automobile combinations.

“When the California “high speed” rail was approved, I was quite disappointed, as I know many others were too. How could it be that the home of Silicon Valley and JPL...would build a bullet train that is both one of the most expensive per mile and one of the slowest in the world?...The underlying motive for a statewide mass transit system is a good one. It would be great to have an alternative to flying or driving, but obviously only if it is actually better than flying or driving.”
—Elon Musk

For the study, the Hyperloop design team used a system servicing San Fransicso and Los Angeles. Total trip time is projected to be approximately half an hour, with capsules departing as often as every 30 seconds from each terminal and carrying 28 people each, thereby supporting a total of 7.4 million people each way per year. The total cost of Hyperloop in this analysis is under US$6 billion.

The two steel tubes would be welded together in a side-by-side configuration to allow the capsules to travel both directions. Pylons placed every 100 ft (30 m) support the tube pair. Solar arrays will cover the top of the tubes in order to provide power to the system. For the design study, the team proposed that the majority of the route would follow the I-5 and that the tube would be constructed in the median.

The compressor. Air processing in the Hyperloop passenger capsule is proposed as follows:

  • An axial compressor compresses tube air with a compression ratio of 20:1. Up to 60% of this air is bypassed, travelling via a narrow tube near the bottom of the capsule to the tail. A nozzle at the tail then expands the flow, generating thrust to mitigate some of the aerodynamic and bearing drag.

  • Up to 0.44 lb/s (0.2 kg/s) of air is cooled and compressed an additional 5.2:1 for the passenger version with additional cooling afterward. This air is stored in onboard composite overwrap pressure vessels. The stored air is eventually consumed by the air bearings to maintain distance between the capsule and tube walls.

  • An onboard water tank is used for cooling of the air. Water is pumped at 0.30 lb/s (0.14 kg/s) through two intercoolers (639 lb or 290 kg total mass of coolant). The steam is stored onboard until reaching the station. Water and steam tanks are changed automatically at each stop.

  • The compressor is powered by a 436 hp (325 kW) onboard electric motor, with an estimated mass of 372 lb (169 kg), which includes power electronics. An estimated 3,400 lb (1,500 kg) of batteries provides 45 minutes of onboard compressor power, which is more than sufficient for the travel time with added reserve backup power.

Propulsion. The propulsion system must:

  1. Accelerate the capsule from 0 to 300 mph (480 km/h) for relatively low speed travel in urban areas.

  2. Maintain the capsule at 300 mph (480 km/h) as necessary, including during ascents over the mountains surrounding Los Angeles and San Francisco.

  3. To accelerate the capsule from 300 to 760 mph (480 to 1,220 km/h) at 1g at the beginning of the long coasting section along the I-5 corridor.

  4. To decelerate the capsule back to 300 mph (480 km/h) at the end of the I-5 corridor.

The designers project that the Hyperloop as a whole will consume an average of 28,000 hp (21 MW). This includes the power needed to make up for propulsion motor efficiency (including elevation changes); aerodynamic drag; charging the batteries to power on-board compressors; and vacuum pumps to keep the tube evacuated. A solar array covering the entire Hyperloop is large enough to provide an annual average of 76,000 hp (57 MW), significantly more than the Hyperloop requires, according to the plan.

The power architecture includes a battery array at each accelerator, allowing the solar array to provide only the average power needed to run the system. Power from the grid is needed when solar power is not available.

Cross-section of rotor and stator. Click to enlarge.

Linear induction motor. The Hyperloop uses a linear induction motor to accelerate and decelerate the capsule. The moving motor element (rotor) is located on the vehicle for weight savings and power requirements, while the tube will incorporate the stationary motor element (stator).

Each linear accelerator has two 65 MVA inverters, one to accelerate the outgoing capsule, and one to capture the energy from the incoming capsule. Inexpensive semiconductor switches allow the central inverters to energize only the section of track occupied by a capsule, improving the power factor seen by the inverters.

The rotor is an aluminum blade 49 ft (15 m) long, 1.5 ft (0.45 m) tall, and 2 in. (50 mm) thick. Current flows mainly in the outer 0.4 in. (10 mm) of this blade, allowing it to be hollow to decrease weight and cost. The gap between the rotor and the stator is 0.8 in. (20 mm) on each side. A combination of the capsule control system and electromagnetic centering forces allows the capsule to safely enter, stay within, and exit such a precise gap.

The stator is mounted to the bottom of the tube over the entire 2.5 miles (4.0 km) it takes to accelerate and decelerate between 300 and 760 mph (480 and 1,220 km). It is approximately 1.6 ft (0.5 m) wide (including the air gap) and 4.0 in. (10 cm) tall, and weighs 530 lb/ft (800 kg/m).

Laid out symmetrically on each side of the rotor, its electrical configuration is 3-phase, 1 slot per pole per phase, with a variable linear pitch (1.3 ft or 0.4 m maximum). The number of turns per slot also varies along the length of the stator, allowing the inverter to operate at nearly constant phase voltage, which simplifies the power electronics design, according to the team. The two halves of the stator require bracing to resist the magnetic forces of 20 lbf/ft (30 N/m) that try to bring them together.

Energy storage allows the linear accelerator only to draw its average power of 8,000 hp (6 MW) (rather than the peak power of 70,000 hp or 52 MW) from its solar array. The storage element is proposed to be built out of the same lithium ion cells available in the Tesla Model S. With proper construction and controls, the battery could be directly connected to the HVDC bus, eliminating the need for an additional DC/DC converter to connect it to the propulsion system, the team suggests.

A high speed transportation system known as Hyperloop has been developed in this document. The work has detailed two version of the Hyperloop: a passenger only version and a passenger plus vehicle version. Hyperloop could transport people, vehicles, and freight between Los Angeles and San Francisco in 35 minutes. Transporting 7.4 million people each way and amortizing the cost of $6 billion over 20 years gives a ticket price of $20 for a one-way trip for the passenger version of Hyperloop. The passenger plus vehicle version of the Hyperloop is less than 9% of the cost of the proposed passenger only high speed rail system between Los Angeles and San Francisco.

An additional passenger plus transport version of the Hyperloop has been created that is only 25% higher in cost than the passenger only version. This version would be capable of transport passengers, vehicles, freight, etc. The passenger plus vehicle version of the Hyperloop is less than 11% of the cost of the proposed passenger only high speed rail system between Los Angeles and San Francisco. Additional technological developments and further optimization could likely reduce this price.

—“Hyperloop Alpha”

Future work. Among the recognized additional required work on the design is:

  • More expansion on the control mechanism for Hyperloop capsules, including attitude thruster or control moment gyros.

  • Detailed station designs with loading and unloading of both passenger and passenger plus vehicle versions of the Hyperloop capsules.

  • Trades comparing the costs and benefits of Hyperloop with more conventional magnetic levitation systems.

  • Sub-scale testing based on a further optimized design to demonstrate the physics of Hyperloop.



The pdf linked via Tesla is corrupt, I believe.

The one via Space X works:


This looks good on paper but in reality, it will have to be ADA (Americans With Disabilities Act) compliant, have systems in place to accomodate children & the elderly, and also be able to accomodate morbidly obese Americans. In addition, restroom facilities will also be needed since someone is undoubtedly going to have to go #1 or #2 even if the trip is 35 minutes.


Yeah, that is why every car on the freeway where no stopping is allowed has a rest room installed inside it.


Davemart - remember this is public transportation, not private.


Here are some other issues - just what happens in the sealed vacuum tube when there is a capsule breakdown? Railroads have switches and siding tracks - the Muskie tube plan makes no mention of maintenance access. Also, what design considerations have been made for seismic activity? Will tremors potentially knock the tubes out of alignment and lead to tragic consequences? Will the tubes be terrorist resistant?


I don't see why there would be an ADA issue. If people with disabilities are helped into and out of their seats... how is it any different than an airplane? The restroom issue: yeah, that's legit.

Issues I see (none are deal breakers):

-Cabin depressurization could be very dangerous. It would be like having a plane depressurize at 150,000'
-There would need to be some mechanism to balance the center of gravity dynamically depending on load
-I'm not real big on storing steam from the intercooler... or "hot swapping" said steam container at the terminal

I'd love to say use aluminum for the tubes but that's just not feasible for a variety of reasons (unless you wanted to build the largest extrusion die/press ever ;)

That's about it. IMO Musk should open-source the whole project especially if he doesn't plan on building it himself. Let the world contribute to and benefit from this idea.


The capsules also need to be large enough for a person to stand, to decrease the feeling of claustrophobia. They could be engineered so the aisle is lower than the floor under the seats.


Is there some sort of a law saying that you have to be able to stand in a transit vehicle? The smaller the pod, the smaller the pipe, the cheaper and more feasible the whole idea is.


Not much chance the cars will derail because of going around a curve too fast. It seems like some air pressure could keep cars from getting too close, or being used to help slow the car. Trains have toilets, so there shouldn't be a problem here. Mechanical emergency brakes could slow the car very quickly because they would push against opposite inside surfaces of the tube.


I did not see any mention of windows. People moving and subject to G forces when they can not see out can cause motion sickness.

I don't think the car carrying idea was necessary. It is a completely different 10 foot diameter tube system, if I am reading the PDF correctly.

Rapid pressurization with a blown tube would cause rapid deceleration, that is not a problem with an airliner.

The electrically driven turbine has a duct between the 2 x 14 seating. That may cause a noise problem.


@SJC: totally agree the car carrying concept is unnecessary. I'd be curious to see numbers on how rapid said deceleration would be. Said duct would lend itself well to noise attenuation IMO.


"..on an array of 28 air bearing skis that are geometrically conformed to the tube walls.." page 20

So he is going to have a 400 hp motor drive a turbine at high speed compressing 1/1000th atmosphere, pushing it through a duct running under the cabin between the seats feeding 28 air hockey "skis"...OK.


"Tucked away inside each pylon, you could place two adjustable lateral(XY) dampers and one vertical (Z) damper..." Page 5

Japan had no major damage to their high speed rail after their 9.0 earthquake.


Go back in time to 1850. Tell an engineer at the time that 160 years hence you'd be to fly from New York to Hong Kong in 16 hours time in a vehicle 30,000' above the ground while be able to access and read nearly any book ever printed on a device just a bit larger than a typical wallet.


This idea is not new at all, it was proposed more than 20 years ago, so Mr Musk is playing the prophet like he doesn't have enough on is plate right now ...we can't even built high speed rail in this country, the infrastructure for this idea would be so monstrously costly that it is not even a remote possibility, just vapor war at the very best, no sorry dreams for the dreamers


He did it as a response to the California High Speed Rail which is projected to cost $68 billion and go 150 MPH. He thought we could do better and we can.

It takes people that have done things to put ideas out there. You or I may not agree totally with what he proposes, but he HAS gotten people thinking about this and that is good.


Apollo 16 on reentry 7.19 g

100 pascals is air density 150,000 feet up, or about 30 miles.
The Apollo capsule might be slowing from 10,000 MPH to maybe 1000 MPH in 20 miles (just a guess). So it is pulling 7 Gs for maybe 20 seconds.

It don't have the calculations, but on of these cars going 760 MPH in 1/1000th atmosphere facing the rapid 1000 fold increase in air density could be more G force. The Apollo capsule does not face that rapid a change.

Roger Pham

Quite a revolutionary idea!
However, the physics of it needs some fine tuning:
--Getting a near-vacuum at 1/1000th atmosphere and maintaining it in a tube hundreds of miles long will be almost impossible to do and at great expense of energy.

--Compressing this thin amount of air even at 1:20 ratio will only raise the density to 20/1000th atm, or 1/50th atm. How can this low density air support the weight of the vehicle?

--A complex variable intake with diffuser must be built at the nose of the vehicle to accommodate supersonic shock wave at the intake lip and to slow down this air, yet, will become supoptimal at subsonic speeds.

--3400 lbs of battery is simply too heavy a weight, considering the total payload of 30 passengers will be only around 6000 lbs.

A more realistic solution will have much denser air within the tube at 1/4th atm, similar to the pressure encountered by commercial airliners. The cabin will pressurized to 8,000-foot pressure, or 3/4th atm, thus the cabin air will need compression to only 3:1 ratio, much more practical. The vehicle speed will be reduced down to only high subsonic of 400-500 mph, to the same speed of an airliner at cruise, the physics of all this is very well known. The vehicle will have wheels to sustain low speeds up to 200 mph, then the wheels will be slightly retracted upward and the vehicle will be simply lifted by aerodynamic lift at above 200 mph, similar to that of an airliner. No complex air cushion system that will add weight and cost. Thus, the 3,400-lb battery and all that heavy motor and compressor will be yanked out, replaced by a small motor of a few hp, enough to compress air for cabin consumption. Motive power will of course still come from linear induction motor along the tubeway to give the vehicle a thrust at every 100 feet or so. Induction wires will be used to capture some of this energy to charge up a capacitor to power the modest compressor for cabin air supply. The vehicle will cary no fuel nor any major battery in order to maximize payload and energy efficiency.

Thus, a 30-passenger vehicle of 6,000-lb payload will likely have a gross weight of under 10,000 lbs when made from aerospace components and construction techniques, and travels without any onboard fuel, engines, nor wings nor tail, etc. This will allow it several folds the efficiency of a comparable airliner that must carry heavy weights of fuel, engines, wings, battery, etc...An airliner's gross weight is easily 6 folds the payload, while this ultra-light Hyperloop vehicle without battery nor large compressor motor will have the gross weight only 1.6 folds the payload.

Account Deleted

Travel safety is more important than being able to stand up or go to a toilet during travel. When you travel at 800 mph you need to be able to stop the capsule very fast if there is a problem further ahead in the tube. At that speed everybody dies if the tube malfunctions just a little bit so you need to be able to stop the capsules fast if a problem is detected ahead in the tube. This is why everybody need to be seated and secured with safety belts during the entire journey. You cannot have people flying through the cabin at high speed during an emergency brake.

People who are not sure they can hold their water for 35 minutes must wear a diaper or not use this form of travel. People with poor health that can't handle pulling 2G to 4G for several seconds should not use this kind of travel either. For this kind of travel it may actually be necessary to have a system where only people with a valid health certificate can buy tickets for the travel. You could obtain such a certificate from your doctor certifying that you do not have a peeing issue and is fit enough to pull a few Gs.

Roger Pham

I must hasten to correct my previous posting that cabin pressurization compressor may not be at all necessary. The air around the vehicle will be compressed as the vehicle travels along the tube. Selecting the proper ratio of cross-sectional area of the vehicle vs. the cross-sectional area of the tube will allow proper compression ratio sufficient for cabin pressurization. This very pressure will allow for the vehicle to stay within the tube without touching the wall of the tube, sort of providing a kind of aerodynamic lift or support for the vehicle within the tube. This is much cheaper than MagLev principle that depends on expensive copper stator coils all along the track. This Hyperloop vehicle will need a thrusting segment only once in a while, not lining up densily as would be required of a MagLev vehicle.


I like your ideas.
Shooting for 500 mph rather than 760 mph sounds as though it could greatly reduce engineering complexity and costs.

The lower pressure differential in your suggestion makes me think that savings might be made in tube construction too.

On another blog someone suggested that Solida concrete could greatly reduce the numbers of towers needed also.


See my pdf link above.
Musk is inviting constructive critique.
I suggest you contribute to the debate.

Account Deleted

A key point of this Hyperloop project is that it enables travel on par with commercial air travel. If any changes to the project is made it should allow for even higher speed not less. What would it take to make it supersonic? However, the most fantastic thing is that Musk thinks this project is possible for only one tenth of the cost of ordinary high spread trains that are too slow to be really good substitutes for air travel anyway. If any other than Musk had said this I would not even remotely believe it. When it comes from Musk it just might be possible. If this is indeed possible then it may even be possible one day to link the entire world with such tubes and drop air travel altogether. It runs on electricity so no need for fossils or even bio-fuels either.


Transrapid and swiss metro have been talking about similar things for years, but with maglev rather than aerodynamic lift. I'm very much a fan of both ideas.


Bathroom is not an issue. They don't have one in the entire DC metro system. Not on a train and not in a station. It will be okay to have it at each end of the trip.

The comments to this entry are closed.