Boring Company

[RDoc]

They have been working on software platooning on normal roads for years with large vehicles using ICE powertrains.
Loop is lower mass vehicles with electric drivetrains on a closed course with better dimensional accuracy along with design restricted lateral positioning.
If they have a bumper with 12 inches of travel and a 12"x24" profile, they can even go mostly sensorless. Concider, by having the approaching pod go 0.6 MPH faster than the front pod. Speed delta of 0.88 ft/sec with a 1 foot compression zone (half of bumper pair) and linear deceleration rate yields >2 second joining event. Acceleration of 0.44 ft/sec/sec or 1/70th of a g.
And that assumes no contact sensors on either pod to change the motor setting. For instance, on contact, the front pod could go light regen (or a power level less than its own drag) to maintain coupling. Bumper compression sensors would also allow for continued monitored contact during the journey.

This is already done via bump drafting in NASCAR.

But is there any need for the pods to touch or even be closer than a few meters other than minimizing air resistance? For that matter, I'm not at all sure how air resistance would work in a Loop tunnel.

Assuming 7 m pods, max acceleration 1 m/s2, line length 1609 m (1 mile), 20 pods per platoon, 100 msec between pods, and a launch spacing of 30 sec per platoon.

That gives 11 seconds between platoons and a maximum of 4 m between pods. The line capacity is over 38000 passengers per hour assuming full pods with 16 people per pod, and a travel time of 80 seconds, max speed of 40 m/s (90 mph).

Unfortunately this also points out a pretty serious problem with at least my understanding of how the Loop system would work. For the example above, the pods would never reach full speed because they can't accelerate enough. They'd spend half the time accelerating and half decelerating, which means they couldn't merge into a full speed bore. At 1 m/s2 it takes about 2250 m (1.4 miles) and 67 sec to get to 67 m/sec (150 mph) and the same to stop.

I think there would have to be a set of different speed bores with interconnects so multiple lines could share the same "trunks". With a max speed of 22 m/sec (50 mph), about what current subways use, it would take 250 m to accelerate (22 sec), so the minimum length would be 500 m (.3 miles). The total time for the above example would still be only 94 sec and the max pod to pod spacing would be just over 2 m.

 

[AWDtsla]

But is there any need for the pods to touch or even be closer than a few meters other than minimizing air resistance? For that matter, I'm not at all sure how air resistance would work in a Loop tunnel.

Assuming 7 m pods, max acceleration 1 m/s2, line length 1609 m (1 mile), 20 pods per platoon, 100 msec between pods, and a launch spacing of 30 sec per platoon.

That gives 11 seconds between platoons and a maximum of 4 m between pods. The line capacity is over 38000 passengers per hour assuming full pods with 16 people per pod, and a travel time of 80 seconds, max speed of 40 m/s (90 mph).

Unfortunately this also points out a pretty serious problem with at least my understanding of how the Loop system would work. For the example above, the pods would never reach full speed because they can't accelerate enough. They'd spend half the time accelerating and half decelerating, which means they couldn't merge into a full speed bore. At 1 m/s2 it takes about 2250 m (1.4 miles) and 67 sec to get to 67 m/sec (150 mph) and the same to stop.

I think there would have to be a set of different speed bores with interconnects so multiple lines could share the same "trunks". With a max speed of 22 m/sec (50 mph), about what current subways use, it would take 250 m to accelerate (22 sec), so the minimum length would be 500 m (.3 miles). The total time for the above example would still be only 94 sec and the max pod to pod spacing would be just over 2 m.

No reason you can't go to 0.5G, cut acceleration time down to 15 seconds. Given this is a Tesla project, go to 1G at around 7 seconds. Probably allow another 2-3 seconds to limit jerk at that acceleration. No reason for acceleration to be limited by friction of steel on steel.

 

[BioSehnsucht]

If you're planning on standing room packing of passengers, 1G and perhaps even 0.5G are going to be a bit much for acceleration. From some googling it sounds like many public transport systems currently peak out at 0.16G or so.

 

[AWDtsla]

If you're planning on standing room packing of passengers, 1G and perhaps even 0.5G are going to be a bit much for acceleration. From some googling it sounds like many public transport systems currently peak out at 0.16G or so.

Depends on how many beers your train conductor has had and how <insert your local sports team> did last night. I'm quite certain I've exceeded .16G on a subway before.

We don't need to make this standing room only. Especially if we decrease acceleration time by a factor of 10.

 

[BioSehnsucht]

I do wonder what the crossover point would be for lowering per-pod capacity to seating only, and being able to increase acceleration rate and thus top speed and run more pods...

 

[bxr140]

The Tube also uses curved walls with full standing height only in the center of the carriage. I don't see what the point is to the Boring Co. design with straight walls and wheels on the outside of the carriage.

Only the lines that use the ancient tubes (most opened well over 100 years ago...) have carriages with curved walls that maximize capacity in those small bores. The sub-surface lines aren't tiny circular bores and as such the carriages are much bigger, have ~vertical walls, and easily facilitate wall to wall standing.

There's a lot of value in mostly straight walls, but more importantly there's really no value in having Tube-like curved walls with the big bend/knuckle unless you have a really tiny bore diameter like the Tube. There's simply no practical geometry/configuration of bore+track+carriage that benefits from anything more than a slightly angled/curved wall (like the image floating upthread). Layering in the higher speeds of whatever pod thing runs through these new tunnels certainly makes aerodynamics more of a thing, but big picture the delta between a relatively straight/angled vs curved/bubbled cross section is going to be pretty small overall.

I do wonder what the crossover point would be for lowering per-pod capacity to seating only, and being able to increase acceleration rate and thus top speed and run more pods...

There could very well be such a crossover...but...there's certainly an upper limit to public transportation acceleration that is already well understood relative to the bell curve of human tolerance, and if I had to guess I'd say that limit isn't too much different for seated vs standing passengers. I also don't think top speed changes with seated vs standing, though of course getting to top speed sooner will gets you there sooner.

A couple other important concepts in your equation:
-The additional time required to seat/unseat passengers
-Near as I understand, the concept of these pods results in more direct/point-to-point travel, which means fewer starts/stops and thus less opportunity for saving time during accel/decel

Mandatory seating would also present an interesting cultural shift--many people are used to, and in fact prefer to stand on public transportation. Given how difficult it seems to be to get everyone on an airplane to take their seats where everyone knows that's the drill, there's an opportunity for a lot of inefficiencies. Unfortuantely, anything but a full mandatory, air-travel like seating process means the pod performance has to accommodate standing passengers.

 

[RDoc]

Only the lines that use the ancient tubes (most opened well over 100 years ago...) have carriages with curved walls that maximize capacity in those small bores. The sub-surface lines aren't tiny circular bores and as such the carriages are much bigger, have ~vertical walls, and easily facilitate wall to wall standing.

There's a lot of value in mostly straight walls, but more importantly there's really no value in having Tube-like curved walls with the big bend/knuckle unless you have a really tiny bore diameter like the Tube. There's simply no practical geometry/configuration of bore+track+carriage that benefits from anything more than a slightly angled/curved wall (like the image floating upthread). Layering in the higher speeds of whatever pod thing runs through these new tunnels certainly makes aerodynamics more of a thing, but big picture the delta between a relatively straight/angled vs curved/bubbled cross section is going to be pretty small overall.

The Jubilee Line was completed in 1979, diameter 12.5' and the cars have somewhat curved walls and a very curved roof. Having full height out to the edge of the car when running in a bore wastes a lot of the diameter which could be used by having low ceilings over seats on the sides while still having full headroom in the center for standing passengers.

The other thing about the published design is putting the wheels outside the carriage. This seems very space wasteful.

There could very well be such a crossover...but...there's certainly an upper limit to public transportation acceleration that is already well understood relative to the bell curve of human tolerance, and if I had to guess I'd say that limit isn't too much different for seated vs standing passengers. I also don't think top speed changes with seated vs standing, though of course getting to top speed sooner will gets you there sooner.

For subways with standing passengers, the limit seems to be somewhere around .1 g, although that's exceeded at times, but is uncomfortable. Airliners with seated passengers regularly use .25g. A reasonably fast car with seated passengers can pull well over .5g and some Teslas can get very close to 1g.

You can't perceive velocity, so, apart from rough roadbeds, there's no difference caused by how fast you're moving.

A couple other important concepts in your equation:
-The additional time required to seat/unseat passengers
-Near as I understand, the concept of these pods results in more direct/point-to-point travel, which means fewer starts/stops and thus less opportunity for saving time during accel/decel

Mandatory seating would also present an interesting cultural shift--many people are used to, and in fact prefer to stand on public transportation. Given how difficult it seems to be to get everyone on an airplane to take their seats where everyone knows that's the drill, there's an opportunity for a lot of inefficiencies. Unfortuantely, anything but a full mandatory, air-travel like seating process means the pod performance has to accommodate standing passengers.

The main issue I see with acceleration is how long the entrance/exit tunnels are. This can also be addressed by limiting the top speed.

 

[RDoc]

I do wonder what the crossover point would be for lowering per-pod capacity to seating only, and being able to increase acceleration rate and thus top speed and run more pods...

Yes, I wonder about this as well, but as has been pointed out, loading time is an issue, perhaps to be dealt with by full falcon wing doors? There are also problems with things like bicycles, baby carriages, wheel chairs, etc. in a general purpose mass transit system.

My take is that there will be local and long distance bores with different speeds. I don't have a good solution to the long acceleration tunnels needed to get into the high speed bores though.

 

[BioSehnsucht]

Passenger loading times if using seating only (no standing) might be solved by simply having lots of separate pods loading separately and leaving when ready to leave (i.e., instead of a typical subway loading area something more like the branching loading area of the renders for the Chicago plans).

Though what was given for Dugout so far doesn't seem conducive to this, more like loading up passengers for Space Mountain (in-order dispatch of pods) than taxi service (capable of out of order dispatch).

 

[RDoc]

Passenger loading times if using seating only (no standing) might be solved by simply having lots of separate pods loading separately and leaving when ready to leave (i.e., instead of a typical subway loading area something more like the branching loading area of the renders for the Chicago plans).

Though what was given for Dugout so far doesn't seem conducive to this, more like loading up passengers for Space Mountain (in-order dispatch of pods) than taxi service (capable of out of order dispatch).

Yes, I was wondering about something like that as well. Perhaps in conjunction with some way to automatically reconfigure the car to accept wheelchairs and bicycles. The taxi model seems much more like what Loop would be than any kind of subway or Space Mountain so the notion of organized platoons of pods probably doesn't make sense. There would be lots of loading platforms for individual vehicles and they would leave in any order as they were filled and time themselves to maintain a minimum spacing in the tunnels.

The loading process is a bit interesting as well. For a taxi, there's the driver to help and to decide when the cab is loaded. I'd think Loop would want to be largely automated because of the high volumes so perhaps some kind of AI vision system. This would be a major issue with loading the pods at the individual elevator stations scattered over the city since obviously there wouldn't be any attendants there.

 

[BioSehnsucht]

I think we'd see platooning (on the eventual real Loop not necessarily a single tunnel Dugout), but it would just be of whatever pods happen to end up near each other for a while, with pods leaving and entering the platoon as their routing dictates.

 

[RDoc]

Some more numbers:

If there is no platooning, the pods just load and go in any order and departing at an interval of 1.5 seconds average, would be a capacity of over 38000 per hour, typical for a subway line. That would require loading a total of 40 pods per minute. For comparison, a busy taxi stand in NYC supports about 2 cabs per minute. I'm not sure what this implies, but it's tempting to think that there would have to be about 20 pods loading at a time, but obviously there would be wide variations. So, maybe space for 30 or so simultaneously?

Acceleration distances to speed:
.1G - 67 mps (150 mph) - 2294 m (1.4 mi)
.4G - 67 mps (150 mph) - 573 m (.36 mi)

.1G - 22 mps (50 mph) - 255 m (.16 mi)
.4G - 22 mps (50 mph) - 64 m (.04 mi)

 

[RDoc]

I think we'd see platooning (on the eventual real Loop not necessarily a single tunnel Dugout), but it would just be of whatever pods happen to end up near each other for a while, with pods leaving and entering the platoon as their routing dictates.

I don't see what the advantage of platooning would be, although I'm sure they would be communicating with nearby pods to manage spacing and changes in velocity due to exits, curves, etc., as well as emergencies.

On highways, there is a significant aerodynamic advantage, but in a pretty tightly fitting tunnel I don't think that's true. The aerodynamics are very different with the train acting more like a piston than pushing the air aside. This might be different in the Loop design because of the large open area under the pods, but I have no idea how to estimate that.

 

[AWDtsla]

If there is no platooning, the pods just load and go in any order and departing at an interval of 1.5 seconds average, would be a capacity of over 38000 per hour, typical for a subway line.

The purpose of Boring company is not to just make a new subway. The point is to make tunneling so cheap an unobtrusive, many many many tunnels can be built. Therefore the comparison to typical subway doesn't matter so much. You simply build more tunnels between destinations as required. sled/train/car speeds should be optimized for user experience, such that people actually want to use it.

This is quite a difference from 70+ years and 3+ billion dollars to build 4.3 mile trolley track, unnamed city that should be ashamed. Peter Thiel is right about our stagnation.

 

[RDoc]

The purpose of Boring company is not to just make a new subway. The point is to make tunneling so cheap an unobtrusive, many many many tunnels can be built. Therefore the comparison to typical subway doesn't matter so much. You simply build more tunnels between destinations as required. sled/train/car speeds should be optimized for user experience, such that people actually want to use it.

This is quite a difference from 70+ years and 3+ billion dollars to build 4.3 mile trolley track, unnamed city that should be ashamed. Peter Thiel is right about our stagnation.

Well, I don't think it will be so cheap as to be not a consideration. The goal is to lower the cost by a factor of 10, not 1000.

 

[AWDtsla]

Well, I don't think it will be so cheap as to be not a consideration. The goal is to lower the cost by a factor of 10, not 1000.

I think 10 is just a start only for the tunnel boring. I wouldn't be surprised if it scales to 100 or a 1000 by the time you cut out the corrupt contracting and operating process.

 

[AWDtsla]

Given the TBM is not constrained to be on one plane, building slopes into it would be interesting, although harder to plan. You'd need route planning software to manage all new routes in a city such that new tunnels did not become impossible.

However I bet with proper jerk control the "tolerable" acceleration will go way up, and it's less of an issue than it seems. Less jerk means less change in force, and less surprise.

 

[RDoc]

A possibly crazy solution someone else suggested to the acceleration distance issue might be to have the acceleration/deceleration tunnels at a steep slope with the higher speed tunnels at a lower depth to take advantage of gravity. If the slope was 60 degrees, the passengers would be half their weight which would feel like they were in a steep dive (which they would be). If the pod accelerated forward at .8 g, their weight would be normal but now instead of the pod being subject to 1g, it would be 1.75g.

Going from a 22 m/s (50 mph) tunnel to 67 m/s (150 mph) would require a drop of just over 58 m.

With the same slope to go from 0 to 22 m/s would require a drop of 14.5 m.

 

[RDoc]

Given the TBM is not constrained to be on one plane, building slopes into it would be interesting, although harder to plan. You'd need route planning software to manage all new routes in a city such that new tunnels did not become impossible.

However I bet with proper jerk control the "tolerable" acceleration will go way up, and it's less of an issue than it seems. Less jerk means less change in force, and less surprise.

Yes, the exact downward and back to horizontal curves and the acceleration while on them would have to be closely controlled!

Planning would be somewhat eased because once up to speed, it wouldn't be necessary to stay at any particular depth.

 

[BioSehnsucht]

I don't see what the advantage of platooning would be, although I'm sure they would be communicating with nearby pods to manage spacing and changes in velocity due to exits, curves, etc., as well as emergencies.

On highways, there is a significant aerodynamic advantage, but in a pretty tightly fitting tunnel I don't think that's true. The aerodynamics are very different with the train acting more like a piston than pushing the air aside. This might be different in the Loop design because of the large open area under the pods, but I have no idea how to estimate that.

For Hyperloop, the partial evacuation of the tube partially balances out the higher speeds as far as aerodynamics (you still need to be aerodynamic, but at least you're not trying to push a dense column of air at supersonic speeds).

For Loop, aerodynamics will still matter, since the denser (vs Hyperloop) air must either be pushed out in front or get around the sides / over / under to be displaced (just as it must be on Hyperloop). If you've ever been in a subway you'll probably remember the rush of air that precedes an arriving train. Platooning will probably improve efficiency both electrical (less drag, more time/miles between pods needing recharging) as well as fitting more pods into a given area/time (capacity). Loop won't be traveling at peak speed as it arrives at a station, but it will have speeds far greater than most (all?) existing subways, in theory.

I don't think platooning is necessary for Loop to work, but it will make it work better.