[walthamstow]

Of all the things TFL could do to improve the tube, cell service would not even be close to top of my list. We've had WiFi on the platforms for about 10 years which is enough to send/receive whatsapp messages while the train is stopped.

I'm not sure it's fair to dig out TFL for being backwards. They introduced the Oyster in 2003, something NYC didn't get until 2019.

[Nextgrid]

That Wi-Fi is unusable. Not only does the association & DHCP negotiation take a large part of the time the train spends stopped at the station, but there's a stupid captive portal that also wastes whatever valuable time & bandwidth you manage to get despite the other issues.

I've been travelling to places where conventional mobile service is available in the subway and it's such an improvement. Being able to look at maps, message or browse the net is a godsend especially as a tourist in a foreign country.

[cjrp]

Just having accurate next-train-arrival boards on each platform is a huge win, and not something every other subway system seems to have.

[londons_explore]

Reliable comms everywhere allows fully centralized train control, with no employees on each train. You can use cameras to check no passengers are stuck in doors, on the track, or fighting in the carriages or platforms.

Now you can redeploy the staff away from repetitive tasks of driving trains and supervising stations, and towards engineering a better service. Get the typical journey speed up from 12 mph up to 45 mph. Make one direction of all tracks an express route that only stops at one station in 10 and drives 60 mph through all the others. Modify the old tunnels with an extra rail on a wall or ceiling for places where the tunnel alignment can't do 60 mph.

[walthamstow]

Are you a railway engineer or in some other related field?

Some of these suggestions sound a bit bizarre to a layman, especially considering your idea that the only thing standing in our way is 4G signal and reshuffling some staff.

> Get the typical journey speed up from 12 mph up to 45 mph.

So instead of 30 minutes from Walthamstow to Brixton it'll be 8 minutes? How is that possible?

There are 16 stations, or 14 stops excluding the terminii. If you stop for 30secs at each, that's 7 minutes in itself.

> Make one direction of all tracks an express route that only stops at one station in 10 and drives 60 mph through all the others

So we would have an express lane in one direction and a stopping lane in the other? Why is that a good thing?

To me an express service seems /slower/ than the status quo, because I would have to wait longer for the train that stops where I want it to.

[londons_explore]

> So instead of 30 minutes from Walthamstow to Brixton it'll be 8 minutes? How is that possible?

> There are 16 stations, or 14 stops excluding the terminii. If you stop for 30secs at each, that's 7 minutes in itself.

Or... you could stop at only 6 of the stops (express service). You could stop for 15 seconds door-open-time (as long as the crowd waiting for the train entirely fits on the train, thats very achievable - and crowd control can be managed by controlling platform ingress). Busses often achieve 3 seconds of door open time for comparison.

At a top speed of 60 mph, max acceleration of 0.2g, max jerk of 3 m/s^3 and travel distance of 2 miles per stop, the whole travel time becomes 13 minutes. Plus the 1.5 minutes for stops. =14.5 mins = 49 mph.

Lets schedule in one 30 second slowdown (ie. drunk guy holds door for mate for 30 seconds), and the average speed hits the 45 mph target.

I suspect the victoria line, being more modern in its track alignment, could easily be made to go 100 mph too...

[nsteel]

How do the express trains get around the slower stopping trains? Or are you going to entirely remove two thirds of the stops?! I don't think you'll have any risk of crowded platforms, it'll be far too useless a service to attract many passengers.

You cannot compare bus stopping times with tube stopping times. It's meaningless. But if you must, the average dwell time for a London bus is closer to 15 seconds. Which is actually quite remarkable when you consider you've got to pay as you get on. And there's often a buggy or two trying to negotiate the driver having parked too far from the curb. The peak and off-peak must be wildly different and make this single measurement pretty useless.

https://clondoner92.blogspot.com/2022/07/tfl-publishes-avera...

[walthamstow]

Ok, so your big plan for quadrupling the speed of the tube is for trains to breeze through most stations without stopping. It's like amputation as a weight-loss tactic.

[londons_explore]

Simulate it and see... The vast majority of journeys become quicker, even though many trains don't stop at many stations.

The optimal strategy is probably to have multiple express routes - ie. Train 1 stops at station A, C, E, and train 2 stops at B, D, and F. A better signalling system becomes necessary for that though - moving block signalling isn't sufficient - you need one that can take into account velocity-acceleration-jerk of both trains to get trains close enough while still ensuring passenger safety. That in turn requires Comms systems not just train presence information.

[londons_explore]

> So we would have an express lane in one direction and a stopping lane in the other? Why is that a good thing?

Simulate it and see...

Turns out that despite most users needing to make more changes, they will complete their journey far faster. The few journeys that are not completed faster (eg. taking the tube one stop) tend to either be walking distance or have an equivalent bus.

Combine it with the fact there are multiple routes from A to B and there are many people who can take the fast train in both directions.

For the railway operator, the trains are a big capital and operational cost. If you can make the trains run faster, you can get better utilisation of the seats, and therefore extra capacity/revenue. That more than outweighs the extra maintenance and electricity costs of going faster.