4 ms·

> Issues glossed over: Soil variability. … The back end. Good points. And beyond the technical aspects, there's a lot that the FAQ doesn't say about the servic

> Issues glossed over: Soil variability. … The back end. Good points. And beyond the technical aspects, there's a lot that the FAQ doesn't say about the service: - How do you tell your sled where to go? Does it always go to the same place? If so, do you merge lanes? If you merge lanes, wouldn't there be jams there? If instead you build many parallel tunnels, don't you remove the gains you had from building smaller tunnels, by having many more tunnels? - How is it meant to compete on price or human bandwidth with subways? Surely at 200 km/h you need to maintain space between sleds in case one fails. Plus, the sleds are not large, which limits human bandwidth. Finally, if you have lots of tunnels (at least one for each origin/destination, instead of subways with many stops), you will need to maintain and repair that infrastructure, which will require service tunnels and backup tunnels, which adds to the cost. Another thought: they don't mention how they'll design the tunnel network yet. My guess is that it will be backboned by double circular tracks (one for clockwise driving, the other for anticlockwise), from which you have regular exit/entrance ramps at each stop. The circularity would avoid having a terminus where everyone slows down. Still, any design would have a maximum capacity corresponding to how fast vehicles can exit in the worst case, if they all suddenly decided to leave at the exact same stop.

hwillis 9y ago
TBMs can turn, albeit slowly. Bigger tunnels are not necessarily more efficient than parallel tunnels. Consider changing a 1 lane to a 2 lane tunnel. You can dig two adjacent tunnels, or you can use a TBM that is twice as wide, and therefore has to remove 4x as much material. Even if you can only build 2 or 3 TBMs per bigger TBM, it's a win. The tunnels are still pretty big, and the sleds could be 10' wide and 10' vertical if the tunnel ID is 14'. That's pretty comparable to a subway or bus, and the increased distance between the sleds is made up for by the increased speed.
elif 9y ago
>How do you tell it where to go? the giant ipad on the dashboard of your tesla, or your phone otherwise. >If you merge lanes, wouldn't there be jams there? If instead you build many parallel tunnels [etc] because if the increased speed, one lane is the equivalent of about 2.5 highway lanes in terms of throughput per hour. Regarding safety distance, I think you can actually safely achieve smaller spacing with a roller-coaster style mechanism than with tires on asphalt. I would say that as well is a 2x throughput multiplier. So parallelization seems largely unnecessary. >How is it meant to compete on price or human bandwidth with subways? ok so take an 8 car subway train with 40 people per car on a 10 minute interval. 8 * 40 * 6 = 1920 people per hour travel from A to B. then take sleds each containing a car with 2 people, paced 300ft apart traveling at 125mph. that is 2 people * (125mph / 0.0568182mi) = 4400 people per hour
- espadrine 9y ago
  > take sleds each containing a car with 2 people, paced 300ft apart traveling at 125mph. that is 2 people (125mph / 0.0568182mi) = 4400 people per hour* I think 300 ft is for 75 mph. Using https://en.wikipedia.org/wiki/Braking_distance https://en.wikipedia.org/wiki/Braking_distance, the distance at 125 mph is 55 m/s * 55 m/s / (2 * 1 * 10 m/s2) = 150 m = 500 ft, resulting in 2 people * 125 mph / 0.09 mi = 2.8 khumans/h. (On the demo video, their vehicles look like they are tire-on-asphalt.) Also, subways have a typical capacity closer to 30 khumans/h in dense urban areas (https://www.thoughtco.com/passenger-capacity-of-transit-2798765 https://www.thoughtco.com/passenger-capacity-of-transit-2798...).
  - Robotbeat 9y ago
    The vehicles use a monorail, not asphalt. Both in the concept video and the actual prototype of the electric sled in a tube video. Also, the concept video shows multi-user pods. If you pack them as full as a rush hour subway, you could probably equal or exceed a subway's throughput.
- ACow_Adonis 9y ago
  Using actual numbers: Melbourne's light rail E-class trams have a carrying capacity of 64 seated: 146 standing. They can regularly mix with pedestrians. The E class is the "big" one at about 40 seconds: https://www.youtube.com/watch?v=kmBlZ4xpS5Q https://www.youtube.com/watch?v=kmBlZ4xpS5Q Melbournes older, 6 car trains have a crush capacity of ~1526. Seatwise they have ~500-600, with a standard target capacity of about 800 (above that the extra variance/numbers people start to complain about overcrowding). Seen here: https://www.youtube.com/watch?v=4AfbBVXGJzY https://www.youtube.com/watch?v=4AfbBVXGJzY Now, those old trains aren't the newest, and they're single-decker and not the biggest or double-decker like sydney's, and they're not the bestest like some of the asian fully-auto metros. But, my point being is that by the numbers before we start getting too crowded, our quite average older 6 car trains have point in time comfortable capacity about 2-3 times more than your 8 car subway estimate. Hell, three of our light rail vehicles regularly move the numbers of your 8-car subway. But wait... We've not actually accounted for the trips accurately, because if a train runs the full length of its track only carrying its estimated point in time comfortable capacity, the carriages would be quite empty: the total passengers using a train aren't its point-in-time-carrying capacity, but how many get on and off between all the stops along its path. That goes for light rail too. So to find the actual number of people moved by train or light rail, you generally have to multiply the carrying capacity by some additional multiple. On the other hand, private vehicles generally, in practice have a person per vehicle estimate UNDER 2 (barring geography, time, purpose variances), meaning your estimate is overly optimistic for the cars, and out by about an order of magnitude for the rail vehicles. I was not being facetious when I made another comment in this thread when I said I was shocked by how inefficient private vehicles were when I actual went down to the intersection and started counting how long it took them to move say 100 people (equivalent of 1 of our trams), nor am I being facetious when I say not only that this is likely a problem with an already known solution, but a problem where The Boring Companies apparent solution is starting with a rough order-of-magnitude efficiency penalty. I'm honestly a bit...feel like I'm taking crazy pills...that people are talking about the proposal of putting private individual vehicles on rails...in tunnels.

hwillis 9y ago

TBMs can turn, albeit slowly. Bigger tunnels are not necessarily more efficient than parallel tunnels. Consider changing a 1 lane to a 2 lane tunnel. You can dig two adjacent tunnels, or you can use a TBM that is twice as wide, and therefore has to remove 4x as much material. Even if you can only build 2 or 3 TBMs per bigger TBM, it's a win. The tunnels are still pretty big, and the sleds could be 10' wide and 10' vertical if the tunnel ID is 14'. That's pretty comparable to a subway or bus, and the increased distance between the sleds is made up for by the increased speed.

elif 9y ago

>How do you tell it where to go? the giant ipad on the dashboard of your tesla, or your phone otherwise. >If you merge lanes, wouldn't there be jams there? If instead you build many parallel tunnels [etc] because if the increased speed, one lane is the equivalent of about 2.5 highway lanes in terms of throughput per hour. Regarding safety distance, I think you can actually safely achieve smaller spacing with a roller-coaster style mechanism than with tires on asphalt. I would say that as well is a 2x throughput multiplier. So parallelization seems largely unnecessary. >How is it meant to compete on price or human bandwidth with subways? ok so take an 8 car subway train with 40 people per car on a 10 minute interval. 8 * 40 * 6 = 1920 people per hour travel from A to B. then take sleds each containing a car with 2 people, paced 300ft apart traveling at 125mph. that is 2 people * (125mph / 0.0568182mi) = 4400 people per hour

espadrine 9y ago
> take sleds each containing a car with 2 people, paced 300ft apart traveling at 125mph. that is 2 people (125mph / 0.0568182mi) = 4400 people per hour* I think 300 ft is for 75 mph. Using https://en.wikipedia.org/wiki/Braking_distance https://en.wikipedia.org/wiki/Braking_distance, the distance at 125 mph is 55 m/s * 55 m/s / (2 * 1 * 10 m/s2) = 150 m = 500 ft, resulting in 2 people * 125 mph / 0.09 mi = 2.8 khumans/h. (On the demo video, their vehicles look like they are tire-on-asphalt.) Also, subways have a typical capacity closer to 30 khumans/h in dense urban areas (https://www.thoughtco.com/passenger-capacity-of-transit-2798765 https://www.thoughtco.com/passenger-capacity-of-transit-2798...).
- Robotbeat 9y ago
  The vehicles use a monorail, not asphalt. Both in the concept video and the actual prototype of the electric sled in a tube video. Also, the concept video shows multi-user pods. If you pack them as full as a rush hour subway, you could probably equal or exceed a subway's throughput.
ACow_Adonis 9y ago
Using actual numbers: Melbourne's light rail E-class trams have a carrying capacity of 64 seated: 146 standing. They can regularly mix with pedestrians. The E class is the "big" one at about 40 seconds: https://www.youtube.com/watch?v=kmBlZ4xpS5Q https://www.youtube.com/watch?v=kmBlZ4xpS5Q Melbournes older, 6 car trains have a crush capacity of ~1526. Seatwise they have ~500-600, with a standard target capacity of about 800 (above that the extra variance/numbers people start to complain about overcrowding). Seen here: https://www.youtube.com/watch?v=4AfbBVXGJzY https://www.youtube.com/watch?v=4AfbBVXGJzY Now, those old trains aren't the newest, and they're single-decker and not the biggest or double-decker like sydney's, and they're not the bestest like some of the asian fully-auto metros. But, my point being is that by the numbers before we start getting too crowded, our quite average older 6 car trains have point in time comfortable capacity about 2-3 times more than your 8 car subway estimate. Hell, three of our light rail vehicles regularly move the numbers of your 8-car subway. But wait... We've not actually accounted for the trips accurately, because if a train runs the full length of its track only carrying its estimated point in time comfortable capacity, the carriages would be quite empty: the total passengers using a train aren't its point-in-time-carrying capacity, but how many get on and off between all the stops along its path. That goes for light rail too. So to find the actual number of people moved by train or light rail, you generally have to multiply the carrying capacity by some additional multiple. On the other hand, private vehicles generally, in practice have a person per vehicle estimate UNDER 2 (barring geography, time, purpose variances), meaning your estimate is overly optimistic for the cars, and out by about an order of magnitude for the rail vehicles. I was not being facetious when I made another comment in this thread when I said I was shocked by how inefficient private vehicles were when I actual went down to the intersection and started counting how long it took them to move say 100 people (equivalent of 1 of our trams), nor am I being facetious when I say not only that this is likely a problem with an already known solution, but a problem where The Boring Companies apparent solution is starting with a rough order-of-magnitude efficiency penalty. I'm honestly a bit...feel like I'm taking crazy pills...that people are talking about the proposal of putting private individual vehicles on rails...in tunnels.