Double-Deck trains in UK

[Shrop]

Which of those stops would you remove and how do you replace/solve the detriment to those who lose a service?

It’s not sensible to try to provide a meaningful answer to this question because any evidence would be skewed by the existing service pattern, and in any event I don’t want to stray too far off the topic of double deck trains, which is why I’ve not responded before now.

The subject of fewer stops has been discussed previously, but I do think it’s a worthwhile subject for further discussion so I may start a new thread when time permits and when I feel suitably motivated

 

[Technologist]

This comment raises an interesting point about the manner in which trains in the UK operate. I remember well the days when you could travel from London to Glasgow stopping only at Preston and Carlisle, or from Birmingham to Plymouth stopping only at Bristol TM and Exeter (and there are many more examples). These days, modern thinking is that trains should stop much more frequently because they can do so while losing less time thanks to better acceleration, and also because running them long distances non-stop simply means they're more likely to catch up other trains.
What this means is that the disadvantages such as taking longer to get on and off, stairs etc, are exaggerated as a direct result of more frequent stops, and thus the chances of ever getting double deck trains considered seriously in the UK are even less.
For what it's worth though, I recently travelled XC from Southampton to Birmingham, and at every one of the NINE intermediate stops in the journey of less than 150 miles, there was a huge amount of shuffling of passengers, changing of seats as the reserved sections refreshed and seated passengers were turfed out by those claiming the seats, luggage being moved in an out of the storage areas to retrieve trapped cases. There really is merit in having fewer stops on some services if only the rail planners could see it!

Eventually operators will realise that battery electric locomotives can be designed accelerate more in line with battery electric cars than current assumption around how fast a train can accelerate. You could potentially operate high speed lines like metros and still have long distance journey times better than we have today for express trains.

 

[The Planner]

Eventually operators will realise that battery electric locomotives can be designed accelerate more in line with battery electric cars than current assumption around how fast a train can accelerate. You could potentially operate high speed lines like metros and still have long distance journey times better than we have today for express trains.

There is still a limit you would place on acceleration for comfort purposes.

 

[SynthD]

Eventually operators will realise that battery electric locomotives can be designed accelerate more in line with battery electric cars than current assumption around how fast a train can accelerate.

Internal combustion engine cars are limited by a narrow torque range, while electric cars aren’t. But the tyres and weight still matter. Diesel-electric trains have the same wide torque range as electric cars, I believe. Diesel alone cannot accelerate like diesel-electric or electric (or diesel-hydraulic?), but that isn’t the comparison being made.

Battery life concerns may remain while the new generation of batteries try to show this is outdated.

 

[zwk500]

Eventually operators will realise that battery electric locomotives can be designed accelerate more in line with battery electric cars than current assumption around how fast a train can accelerate. You could potentially operate high speed lines like metros and still have long distance journey times better than we have today for express trains.

What is wrong about the current experience and assumptions around how fast a modern electric train can accelerate? The 80X's don't hang around on electric power.

 

[Technologist]

There is still a limit you would place on acceleration for comfort purposes.

In my modelling I've used 1.3m/s/s which is what a tube train accelerates at. A high speed train with 5-7% battery mass will have about 3x the power of a conventional EMU and will manage to accelerate at that rate past 100mph.

This basically means that you can get up to 125mph in about 1500m. Comfort standards in the case of acceleration and braking are for the guidance of the wise and the adherence of an industry that wants to be destroyed. Buses aren't limited to the same degree, passengers still stand on them and move around even though buses accelerate and brake at up to 2.5m/s/s in normal operation and 4m/s/s in a hard stop.

In practice passengers will adapt to higher accelerations especially if they result in faster trains, cupholders and similar are relatively cheap.

Internal combustion engine cars are limited by a narrow torque range, while electric cars aren’t. But the tyres and weight still matter. Diesel-electric trains have the same wide torque range as electric cars, I believe. Diesel alone cannot accelerate like diesel-electric or electric (or diesel-hydraulic?), but that isn’t the comparison being made.

Battery life concerns may remain while the new generation of batteries try to show this is outdated.

EMUs are ultimately limited by the OHL infrastructure, battery EMUs are limited by the discharge rate of the battery. Once you get up to 5% of the trains mass being in the battery the discharge rate that the battery can manage is massively higher than what we would expect from an EMU of that size.

The second big advancement due to BEVs is that traction motors are now commodities, any of the larger car markers now builds more traction motors in a few months than the entire rail sector has ever used and these things have very much higher power density than anything being used on rail. This means that tripling the power output of the next generation of BEMUs will be pretty trivial from a cost/mass perspective.

What is wrong about the current experience and assumptions around how fast a modern electric train can accelerate? The 80X's don't hang around on electric power.

Try driving a decent BEV, see above points on the quality and cost of the parts/engineering. Rail is going to be buffeted by the electrification of road transport, self driving, AI and electric flight, we need to think bigger or become irrelevant. A BEMU service with rapid acceleration could allow us to turn much of the UKs rail network into high speed metros, no need to run mixed traffic.

 

[zwk500]

In my modelling I've used 1.3m/s/s which is what a tube train accelerates at ... even though buses accelerate and brake at up to 2.5m/s/s in normal operation and 4m/s/s in a hard stop.

I've never been on a bus, even a hydrogen/EV bus that accelerates twice as fast as a tube train.

Try driving a decent BEV, see above points on the quality and cost of the parts/engineering. Rail is going to be buffeted by the electrification of road transport, self driving, AI and electric flight, we need to think bigger or become irrelevant. A BEMU service with rapid acceleration could allow us to turn much of the UKs rail network into high speed metros, no need to run mixed traffic.

I think you've missed my point. I'm asking why BEMUs need to completely rethink acceleration rates when existing trains already achieve extremely good performance that results in competitive journeys. See reports abound of 80Xs running into problems overspeeding because the acceleration is massively more than legacy stock. Why do we need to completely redraw the curves when we could simply design BEMUs to achieve the 80x or 700 power outputs?

I'm a big advocate of BEMUs being a fundamental part of the long-term railway low-carbon solution. I'm not advocating against BEMUs at all.

 

[Backroom_boy]

I could see doubledecker trains on the Eurostar services. Isn't it already cleared gauging wise for the route?

 

[stuu]

I could see doubledecker trains on the Eurostar services. Isn't it already cleared gauging wise for the route?

Yes, HS1 is designed to the same standards as the French LGV network, so not a problem

 

[zwk500]

I could see doubledecker trains on the Eurostar services. Isn't it already cleared gauging wise for the route?

The issue for Eurostar running double deck trains would be meeting the evacuation requirements in the tunnel. It's certainly not insurmountable, but it is another barrier to clear.

 

[AM9]

Is that really true? I am sure there must be routes that are not accessible to 26m long IET stock due to platform curvature clearances.

I believe that the clearance for the 26m long class 8xx stock is the same as for MK111 23m stock. This is achieved by insetting the bogie centres to:

maintain a similar inswing at the centre of the cars. the additional outthrow at the ends is reduced by the cars having a tapered profile​

the overall width of the cars is slightly reduced​

By these means, the routes permitted for the class 8xx stock is very similar that for the MK111 profile.

 

[Haywain]

I believe that the clearance for the 26m long class 8xx stock is the same as for MK111 23m stock. This is achieved by insetting the bogie centres to:

maintain a similar inswing at the centre of the cars. the additional outthrow at the ends is reduced by the cars having a tapered profile​

the overall width of the cars is slightly reduced​

By these means, the routes permitted for the class 8xx stock is very similar that for the MK111 profile.

That's not the same as what was suggested, which was that new stock must be able to access 101% of the network.

Because all stock, irresp' of propulsion, must be capable of being hauled on 101% of UK's infrastructure inc' non-electrified track with its neat fitting arch bridges & tunnels.

 

[Technologist]

I think you've missed my point. I'm asking why BEMUs need to completely rethink acceleration rates when existing trains already achieve extremely good performance that results in competitive journeys. See reports abound of 80Xs running into problems overspeeding because the acceleration is massively more than legacy stock. Why do we need to completely redraw the curves when we could simply design BEMUs to achieve the 80x or 700 power outputs?

I'm a big advocate of BEMUs being a fundamental part of the long-term railway low-carbon solution. I'm not advocating against BEMUs at all.

Industries die when they decide that something is good enough and when backwards compatibility stops new products operating to their full potential. The performance of a class 800 might be acceptable when compared to an older train, however if it's compared to a self driving car or an eVTOL much less so.

If all trains can accelerate very rapidly stopping trains cease to be obstructions and thus we can run far more lines like metros vastly improving user experience. Not running mixed traffic also means that line speeds can be increased and rapid acceleration means that you can hit those line speeds.

 

[Haywain]

If all trains can accelerate very rapidly stopping trains cease to be obstructions

Are they not creating any obstruction while they decelerate and during the time they are stopped?

 

[stuu]

If all trains can accelerate very rapidly stopping trains cease to be obstructions and thus we can run far more lines like metros vastly improving user experience. Not running mixed traffic also means that line speeds can be increased and rapid acceleration means that you can hit those line speeds.

They really don't. Acceleration isn't that big a deal on trains which run medium to long distances, it doesn't save that much time. 1.3 m/s2 (constant) gets you to 125mph about 1.5 minutes quicker than accelerating at half that rate, which is about what 800s manage on electric power. Even XC journeys only stop every 20 minutes so you might save 3 minutes an hour. Game changing

 

[Technologist]

They really don't. Acceleration isn't that big a deal on trains which run medium to long distances, it doesn't save that much time. 1.3 m/s2 (constant) gets you to 125mph about 1.5 minutes quicker than accelerating at half that rate, which is about what 800s manage on electric power. Even XC journeys only stop every 20 minutes so you might save 3 minutes an hour. Game changing

You're missing my point, the rapid acceleration can be applied to every train on the line so that your stopping service accelerates viciously to the line speed (which is then worth investing in raising as more trains can hit it). Combine with designing these trains for tube/commuter dwell times and there isn't a need for express trains as the end to end journey time for a stopping train is at least as good as where the express trains were. Stopping all trains at all stations then means that you can run far more trains.

Now combine with changing planning laws so you can build densely around any station not in an SSI and you have the potential to gives large amounts of the UK a London style commuter rail service that allows long distance walk up and go services. The economic growth related to doing this would be massive, the industry should have this level of ambition.

For full pendry my modelled train was based on class 800s swapping all the diesels and fuel for LFP battery packs, this could conservatively achieve about 3.6 times the power output of a class 801. Obviously you could add more batteries and you could treat them harsher, I designed for a 5C discharge rate, you could go to 10C no problem.

So a class 800 accelerates at 0.7m/s/s until it hits ~57kph, then it tails off, the BEMU goes at 1.3m/s/s until it hits ~112kph. From that point on the BEMU accelerates 3 times faster.

 

[Trainbike46]

While I absolutely see the benefits of faster acceleration, and 1.3 m/s is absolutely realistic on the mainline railway network, because some trains already do it, I wonder what any of this has to do with the, for the UK terrible idea of double-deck trains?

Back to acceleration, the UK FLIRTs have these acceleration speeds provided by Stadler, though I have no clue how these develop through the curve:

On AC power:

• 755/3: 1.3 m/s2 (4.3 ft/s2)
• 755/4: 1.1 m/s2 (3.6 ft/s2)
• 745: 0.9 m/s2 (3.0 ft/s2)
• 756: 1.1 m/s2 (3.6 ft/s2)

On diesel power:

• 755/3: 0.7 m/s2 (2.3 ft/s2)
• 755/4: 0.9 m/s2 (3.0 ft/s2)
• 231: 0.9 m/s2 (3.0 ft/s2)

 

[HSTEd]

They really don't. Acceleration isn't that big a deal on trains which run medium to long distances, it doesn't save that much time. 1.3 m/s2 (constant) gets you to 125mph about 1.5 minutes quicker than accelerating at half that rate, which is about what 800s manage on electric power. Even XC journeys only stop every 20 minutes so you might save 3 minutes an hour. Game changing

You spend the three minutes an hour adding extra stops.

In such an environment you can make stopping trains with minimal journey time disadvantages versus express trains. You can then achieve major "effective" journey time improvements by replacing a mixed stopping/semi fast/fast timetable with a uniform timetablne, putting functional frequencies through the roof.

But I agree this is peripheral to the double deck train argument - but does demonstrate the sort of high capability railway that is buildable with current technology.
Increasing loading gauges will make high capability trains easier to build, although I'd rather focus on extra width if we can get it.

 

[MarkyT]

While I absolutely see the benefits of faster acceleration, and 1.3 m/s is absolutely realistic on the mainline railway network, because some trains already do it, I wonder what any of this has to do with the, for the UK terrible idea of double-deck trains?

Back to acceleration, the UK FLIRTs have these acceleration speeds provided by Stadler, though I have no clue how these develop through the curve:

On AC power:

• 755/3: 1.3 m/s2 (4.3 ft/s2)
• 755/4: 1.1 m/s2 (3.6 ft/s2)
• 745: 0.9 m/s2 (3.0 ft/s2)
• 756: 1.1 m/s2 (3.6 ft/s2)

On diesel power:

• 755/3: 0.7 m/s2 (2.3 ft/s2)
• 755/4: 0.9 m/s2 (3.0 ft/s2)
• 231: 0.9 m/s2 (3.0 ft/s2)

Another modern Stadler train, Merseyrail's cl.777, is far more powerful than the cl.50x units they replace. Car weight is also reduced significantly with the articulated bogie configuration.

Unit Class​	No. of Cars​	Total Power (kw)​	Total Weight (t)​	kw/car​	~t/car​	kw/t​
507/508​	3​	656​	105.00​	218.67​	35.00​	6.25​
777​	4​	2100​	99.00​	525.00​	24.75​	21.21​

 

[HSTEd]

Somewhat associated with this is a question, double decker trains will likely lead to a larger loading gauge with the attendant major structural modifications.

A larger loading gauge would allow trains to load more people into the same length, but in some constrained areas such modifications would force a reduction in the number of tracks.

If we could have trains the size of Superliners or E4 series shinkansen vehicles, would a reduction in tracks be tolerable, especially given the performance that modern multiple units are capable of?

 

[Bletchleyite]

Height is the main constraint. Wider wouldn't be hard by, as you say, removing tracks on 4-track sections. But RIC stock isn't that much wider than UK profile anyway (albeit with the platform protrusion issue) but it is much higher. UK loading gauge is lower than a double decker bus.

 

[Haywain]

Wider wouldn't be hard by, as you say, removing tracks on 4-track sections.

Wider is very hard because of platforms and other lineside structures.

 

[Bletchleyite]

Wider is very hard because of platforms and other lineside structures.

It's much easier than higher if you're willing to lose tracks. It would require platform rebuilding, but this is a lot cheaper than replacing every single OHLE support and bridge/tunnel.

The fundamental issue is that UK loading gauge is too low. If it was two feet higher you'd be able to do something, albeit probably only 2+1 seated downstairs (full width upstairs though - it'd be shaped a bit like the Dutch DD-IRM units with a bulge above platform level). But as it is you don't have the height for two full decks.

 

[HSTEd]

Height is the main constraint. Wider wouldn't be hard by, as you say, removing tracks on 4-track sections. But RIC stock isn't that much wider than UK profile anyway (albeit with the platform protrusion issue) but it is much higher. UK loading gauge is lower than a double decker bus.

Well reduction in number of tracks also aids height in some cases, for example all our arched bridges.

If you have a four track railway with two two-track arch bridge spans, going to four tracks would allow the track to move into the centre of the arch, which will get you a non-negligible amount of height as well as width.

SImilar idea in existing twin track tunnel bores.

 

[Trainbike46]

changing the loading gauge in such a significant way, whether for widening or for heightening, would be really expensive, involve very disruptive infrastructure works, and, as far as I can tell, simply would not be worth it.

Best to just accept that double-deck trains won't happen outside HS1 and maybe HS2, and instead increase frequencies or extend train length.

 

[Mikey C]

The issue for Eurostar running double deck trains would be meeting the evacuation requirements in the tunnel. It's certainly not insurmountable, but it is another barrier to clear.

Also, could STP's passenger lounges cope with even more people on a single service?

 

[MarkyT]

Well reduction in number of tracks also aids height in some cases, for example all our arched bridges.

If you have a four track railway with two two-track arch bridge spans, going to four tracks would allow the track to move into the centre of the arch, which will get you a non-negligible amount of height as well as width.

SImilar idea in existing twin track tunnel bores.

This technique gained space in tunnels of the Tonbridge - Hastings line to overcome especially narrow historic loading gauge due to original substandard construction requiring an extra lining. They only singled the higher speed tunnels though. Either side of Tunbridge Wells, they kept double track with a low speed restriction applied and concrete slab instead of ballast and sleepers was used in tunnels to limit the dynamic envelope (mainly how much cars can swing side to side in motion) and reduce the risk of any track movement.

 

[HSTEd]

changing the loading gauge in such a significant way, whether for widening or for heightening, would be really expensive, involve very disruptive infrastructure works, and, as far as I can tell, simply would not be worth it.

Best to just accept that double-deck trains won't happen outside HS1 and maybe HS2, and instead increase frequencies or extend train length.

What happens when we hit the limits of frequency (set by timetable limitations and physical infrastructure) and train length (set by the same infrastructure).

Trying to boost train lengths on many routes today is going to be just as disruptive as a total reconstruction to a new loading gauge.

 

[swt_passenger]

Well reduction in number of tracks also aids height in some cases, for example all our arched bridges.

If you have a four track railway with two two-track arch bridge spans, going to four tracks would allow the track to move into the centre of the arch, which will get you a non-negligible amount of height as well as width.

You meant to write going to two tracks presumably? Academic though, because I just don’t see anyone seriously proposing converting any of the existing four track railway sections to two track.

 

[Trainbike46]

What happens when we hit the limits of frequency (set by timetable limitations and physical infrastructure) and train length (set by the same infrastructure).

Trying to boost train lengths on many routes today is going to be just as disruptive as a total reconstruction to a new loading gauge.

The few lines where this is likely to occur can be resolved by building new lines, for example HS2, built to a larger loading gauge from the start. The amount of work needed to increase the loading gauge is effectively a complete rebuilt anyway, and if you're doing things like turning 4-track railways into 2-track railways you are then decreasing capacity again.

Another option is what is discussed upthread: increase acceleration, use the time gained to add extra stops (for the faster services), or to reduce the average speed differential between services (by increasing the average speed of slower services), decreasing effective journey time, and increasing capacity for more trains using the lower speed differential between services.

However, on many lines extending train lengths or frequencies is still possible, especially if HS2 gets built to at least Crewe (and ideally beyond there)