The $10⁵ Delta: Hyperloop in 2019

Samarth Sandeep
oneloop
11 min readJul 23, 2019

--

Edited by Kirk McGregor and Delahny Devendran.

July 21st, 2019 was a remarkable day for the future of transportation. It was the 2019 SpaceX Hyperloop Competition, a day where the most hardworking, most daring, and least inhibited engineering students from universities around the world came together to see who can make the fastest pod that runs in SpaceX’s mile-long vacuum chamber.

This competition has been running for the past four years, first conceived through Elon Musk whitepaper detailing the creation of a new mode of transport for distances between 100 to 1000 miles and consisting of pods fitting anywhere from 5 to 80 people running in a resistanceless vacuum chamber. Thinking bullet train speeds (200 mph)? Think more like super sonic speeds (~700 mph).

For the 3rd year in a row, the winner was TU Munich’s team, TUM Hyperloop, with a top speed of 288 mph. To achieve this speed and its reactionary deceleration in less than 1 mile is beyond impressive. For reference, the fastest production car in the world, the Koenigsegg Agera, can hit 278 miles per hour, but needs about 1.4 miles to fully accelerate then decelerate, and took almost two decades of development to get to its final form. The fact that a student team could make something even faster in a year within a completely new paradigm shows the type of fire that the competition has brought out in some of the top brains in the world. More intensely, another 20 teams were brought to the SpaceX headquarters with similar goals in mind. But has the competition become more accessible? Is it now cheaper than ever to build a Hyperloop pod? What happened in the last few years to make such pods possible?

In some ways, the competition has become much simpler since more information has become publicly available about the different drivetrains used by each team, the sponsors they work with, and the backgrounds of each team member. What started out as a completely new paradigm is now one that has a few definite methods that each team is employing, such as DC motors, linear induction motors, and cold gas propulsion.

However, to be able to gain the equipment and space needed to make such a pod has only gotten more difficult. I think this point was summarized brilliantly by Darrell Etherington of TechCrunch when he said:

“Many of the finalists have deep-pocketed corporate backers, including Airbus, while some of the smaller schools have next to no funding — resulting in a cost delta of hundreds of thousands of dollars when it comes to the total bill for the test pods built.”

This delta is no joke: it’s a gargantuan Nile Delta of government contracts, university expenditure, and corporate sponsorships that separates top performing teams in the Hyperloop competition from smaller universities; it is ultimately Musk’s and SpaceX’s brand of Darwinism that will build a Hyperloop in their neck of the woods or not, and this delta fjord is only going to deepen as the competition is pushed towards tunneling mechanisms as well.

But are we any closer to making Hyperloop a reality?

Bridging the gap: the different models of Hyperloop adventure

With Hyperloop consultancy firms, large Hyperloop startups, globally-sourced Hyperloop multinational startups, multi-university hyperloop partnerships, Hyperloop government coalitions, full-time Hyperloop teams, and Hyperloop club teams, yes, we are closer to making a Hyperloop a reality, but the model that will bring about a finished track is still unclear.

Chronologically speaking, the history of Hyperloop development begins with the Hyperloop alpha paper in 2013. According to Hyperloop One’s timeline, this paper was released after their lead investor and founder, Shervin Pishevar, and Elon Musk were both on a humanitarian trip to Cuba talking about transportation in California, and Pishevar pushed Musk to share his ideas. Months later, Pishevar presented the whitepaper to government officials and recruited a team to make this possible. This would eventually become Hyperloop Technologies, and then Hyperloop One. This is currently the largest individual company working on Hyperloop development.

Musk and friends on humanitarian effort in Cuba

Around the same time as when Pishevar and Josh Giegel of Hyperloop One were working on their initial design and business model, Hyperloop Transportation Technologies got its start as a campaign on the JumpstartFund platform, a crowdsourcing platform for engineers and engineering projects. Due to the interest seen in their project, Dirk Ahlborn and other supporters decided to incorporate the company Hyperloop Transportation Technologies. However, rather than running a traditional corporate structure, the team distributed around the world, with contributing members getting anything from perks to shares to payment for their work, creating a globally sourced startup.

JumpStartFund’s Hyperloop Page

Over the next two years, both organizations would continue to work on their initial pod designs and try to gain funding and attention to aid in these efforts. Then, it all changed in June of 2015, when the Hyperloop competition was announced by SpaceX. This was the first time university students would get involved in this project at such a large scale. The first event was a design weekend at Texas A&M in January of 2016.

The teams who demonstrated sound designs at this event would then be selected to run their pods on the SpaceX test track later in 2016. However, the test track was not complete until January of 2017, when the first real on-track competition began. MIT, University of Washington, University of Wisconsin, TU Munich, and TU Delft attended, and have continued to be mainstays in the competition.

As the competition annualized in August of 2017, late July of 2018, and late July of 2019, three distinct Hyperloop student team models began to develop:

  1. Hyperloop Club Teams were the original of the Hyperloop team models. A team begins as a university campus club that attracts undergraduate engineering students, and works during the school year while undergraduates also take courses. The majority of Hyperloop teams tend to follow this model or at least start this way, as it is the simplest way to bootstrap. Some examples include Oneloop and IIT Madras’ Avishkar.
  2. Multi-university Hyperloop partnerships: In an attempt to better represent a geographic region or to pool resources together between teams that have made it into the Top 20 and teams that have not, club teams form partnerships with other teams to make mega-teams between multiple universities. Examples of this include Midwest Hyperloop, which represents 3 universities in Ohio, Indiana, and Illinois, and Eirloop, which represent several universities in Ireland. The majority of the students in these teams do not work full-time on Hyperloop.
  3. Full-time Hyperloop student teams: Initially Hyperloop club teams, full-time hyperloop teams are organizations that have the funding and university backing to fully take students out of their coursework or to support graduate students, and focus on Hyperloop development to maximize construction hours. Some teams that have followed this model are Delft Hyperloop, TU Munich, and MIT.

From 2015 to today, various Hyperloop companies and Hyperloop university teams have worked with engineering firms to develop feasibility studies around designs. In order to do this, they had to get the public on board with the technology. Hyperloop One and Hyperloop Transportation Technologies did this virtually by setting up web pages that allowed people to identify routes for which Hyperloop would be best suited. Local people in these areas who pushed hard for Hyperloop would go on to found Hyperloop coalitions focused on the technology’s proliferation. This has included influential groups, such as the Missouri Hyperloop Coalition now heralded as a possible model for all of America, Maharashtra’s Hyperloop Coalition as a model for maximizing travel within intercity corridors in India, and others in areas where there was not a strong enough governmental interest in Hyperloop to just spur full on development from one of the startups or teams. As a tactic, this has had mixed success in the past, such as with the Chicago New York All Electric Rail Line in 1909, but can be a great way to mobilize support.

In other areas where such support existed, Delft Hyperloop gained a partnership with the Netherlands government, TU Munich with the Bavarian government, Hyperloop One with the Emiratis of the UAE, and Hyperloop Transportation Technologies with Toulouse’s government.

Silos and possible solutions

While Hyperloop development has seen more international and multi-organizational development than most other new infrastructural development project, most teams and companies are still quite siloed from one another. And while Hyperloop teams are definitely sharing ideas and testing platforms, they are not yet getting testing time and space outside of the competition.

Worse yet, every organization is trying to develop every aspect of its track and pod completely independently. For example, Hyperloop One and Hyperloop Transportation Technologies both developed their own tracks, their own pods, and other infrastructure, although they are ultimately vying for the same land space. Even startup spinoffs from these larger companies and teams, such as Delft’s Hardt Hyperloop, are working with their initial partners to develop a full vision, in which their narrative is the only narrative that exists.

rLoop tried to address this by being the first Reddit-based Hyperloop team. Started in response to Elon Musk’s call to action, the team won an innovation award for its pod design on the first Competition Weekend in January 2017. It consists of engineers from various organizations, similar to Hyperloop Transportation Technologies.

However, rather than forming a formal company and giving key contributors shares and funds, the organization decided to start a cryptocurrency instead focused on providing key contributors with a way not only to make money from the growth of the project, but to also provide key input into the development of the project and of the organization as a whole.

Two years later, while the rLoop organization exists and is still dedicated to this idea of crowdengineering with projects outside of Hyperloop, the token does not really seem to exist. The last transfer record is 50 million rLoop tokens between one Ethereum wallet to another wallet in February of this year. More curiously, there is no hint of voting functionality within the token; the code only identifies the total supply of tokens at 1000000000000000000000000000, and that all tokens will be part of one wallet.

RLPToken’s Actual Code

But, other less-technically challenging platforms for collaboration have emerged. Delft Hyperloop recently started a platform called Hyperloop Connected that has a map of most of the world’s Hyperloop teams and companies, as well as content on different pod designs and developments. However, the service it currently offers is the same that is offered by InTheHyperloop, the main news source for all things Hyperloop started by one of the founders of the Colorado Hyperloop Coalition.

More Direct Communication and Hyperloop Symposiums

To be able to get to a finished track with real world tests completed, all of the different groups need to work with each other. The traditional pathway for engineering development is for alumni of universities participating in such engineering challenges being selected to work with well funded companies that are more experienced and further along in the development cycle. But, in 2019, this model is archaic, with undergraduates starting companies and career veterans joining student teams. It is no longer a matter of what your CV looks like; it is a matter of what you can provide to the teams and companies in terms of actual work, vision, or connections.

A very, very simple way to allow for more communication between these different organizations is to have each organization create a public communication channel within their favorite communication application, i.e. a Slack channel. This, for example,would allow for students at the University of Maryland to ask engineers at Hyperloop One who are also working on pod designs for input, and allow for engineers with the Department of Transportation to easily ask University of Washington for input on a ruling about Hyperloop without delay. Eventually, this could manifest into its own Slack workspace or chat group, and possibly develop into further university-university club partnerships or even university-company partnerships.

The best way to really channel the greatest minds in an industry in one place is to host a Hyperloop research symposium. This could be very similar in style to the first Hyperloop design weekend, where groups from around the world came together to have their ideas judged by a panel of previously selected judges. While this sounds very similar to the Hyperloop competition already held by SpaceX or the Hyperloop Conference held in Colorado, the competition is restricted to university teams and is only focused on scaled pods for the high speed run. Therefore, it cannot take into account the work of groups like Transpod and Dinclix Groundworks Hyperloop, who are not as large as Hyperloop One and Hyperloop Transportation Technologies to make headline news, not able to use the SpaceX stage, and cannot make it to Colorado.

This symposium could be where ideas like different mechanisms for pylon-based tube development or underground tube development could be discussed further, with construction grants being given to the team with the best design. This would provide more incentive for all of the different groups involved in pod and track development to submit reports more regularly, and reward collaborations.

Before moving to making another engineering competition focused on developing tunneling methods or making the teams run in a larger banked tunnel, a symposium supported by SpaceX would better support the Hyperloop community as a whole (including all of the citizens behind the different coalitions and investors of projects ranging from Hyperloop One’s Series A to rLoop’s token) as an immediate step towards reducing the divisive delta between teams and getting Hyperloop development initiated in more locations.

What Should Happen

A 10⁵ monetary difference, compounded further by the highest funded teams also usually having the most experienced engineers, means that further complexity within the competition will only further the gap between competition leaders who actually enter the tube and teams who barely make it past safety checks to attend the competition in person. On top of this, with companies operating in relatively ironic vacuums, it will become increasingly difficult to make a Hyperloop that sees completion and operation within a decade, while also maintaining a healthy business landscape for future development.

Much of the world’s rail infrastructure is monopolized. The air industry is largely oligopolic. With so many different types of extant Hyperloop entities and exuberance for the project existing in so many people around the world, there should be no reason for the future of transportation to be as bureaucratic and as timid of innovation as the other four orthodox modes of transportation.

We need to erase the delta of cost, and instead employ a delta of growth that puts the Hyperloop ecosystem back on track.

My name is Samarth Sandeep. I am the outgoing Business Team Director for Oneloop, UC Davis’ Hyperloop team. My dream is to make a Hyperloop powered by a decentralized, sustainable grid. The business team and I have been working on how we can make a track at UC Davis, starting by making an incubator eventually focused on Hyperloop development.

--

--