Should the HS2-HS1 tunnel under London be uncancelled?

[yoyothehobo]

I Think the best way to.make this work is to terminate HS2 trains at Euston.

Fully segregated track work to reduce delays, i can just imagine the reliability issues you would have trying to intertwine HS2, javelin and eurostar services would be disastrous running through St Pancras.

My honest suggestion to help would be to give up on this idea. There is no physical way to make any of this happen in any way that would be beneficial.

This idea is deficient in structual design, track engineering, signal engineering, timetable planning, passenger demand forecast, train speed planning.

It seems to be based upon this wild notion that there is a massive untapped demand of people travelling from Kent to the North West.

 

[John Jefkins]

I Think the best way to.make this work is to terminate HS2 trains at Euston.

Fully segregated track work to reduce delays, i can just imagine the reliability issues you would have trying to intertwine HS2, javelin and eurostar services would be disastrous running through St Pancras.

My honest suggestion to help would be to give up on this idea. There is no physical way to make any of this happen in any way that would be beneficial.

This idea is deficient in structual design, track engineering, signal engineering, timetable planning, passenger demand forecast, train speed planning.

It seems to be based upon this wild notion that there is a massive untapped demand of people travelling from Kent to the North West.

Except there isn't enough government money to put the 10 or 11 platforms we'd need at Euston. A consultant feeding frenzy of overspecification made HS2 cost 10 times what anybody else would pay. Then Putin and Trump have wrecked it more.

To get HS2 trains to actually reach central London we have to stop being so purist. The route from O.O.C. to London is a 70mph railway that lots of other trains could share.

I simply do not understand why different trains running east of them all stopping at O.O.C. cannot be intertwined when that happens all the time with other services on say Thameslink coming into London Bridge from all directions.
Its the same running west as well. We do it elsewhere. Why is do HS2 tracks east of O.O.C. need to be kept purely for HS2 trains. That's a slow speed (70mph) standard railway in a tunnel to Primrose Hill and following NLL tracks to St Pancras.

Why do we need to p**s £6,000,000 against the wall rebuilding Euston decades before the rest of that station needs a rebuild - when we've already got loads of spare platforms along HS1 to use for trains coming from HS2.

Kent to the North west is just part of the demand. Two Eurostars could easily be accommodated too from OOC.
Most of the demand is from South London and North London via the Thameslink network and from all the other trains into Kings Cross or St Pancras wanting to connect to HS2 quicker than they could reach Euston.
Most of the interchange penalty uplift is actually coming from the better connections for people in the London area of these maps.
Most of the HS2 trains (eg 4 tph) could be terminated to turnback in a 15 minute slot at St Pancras with just 4 per hour reversed to extend east to terminate Stratford or Ebbsfleet and to Ashford too at rush hours.

My suggestion is to terminate HS2 at Euston, and make the interchange as smooth as possible. Through trains to the continent from beyond London will only make sense if we're part of Schengen or equivalent frictionless travel.

I'd absolutely get the ORR to grant access rights to a London-Continent competitor as well, most likely Virgin's proposal at the moment but maybe DB if they're interested.

You cannot make an interchange smoother if it doesn't physically connect to Thameslink, Eurostar, Javelin and all of the other trains coming into Kings X or from the Midland Mainline.
And Javelins coming from Kent and Stratford carry 18 million passengers a year now - yet another not insignificant other market.

So extending 4 Javelins per hour west to O.O.C would free up those 3 platforms to allow around 8 HS2 trains per hour to reach St Pancras (with half of them continuing to terminate at Stratford or in Kent).
That postpones the need to get a smaller and cheaper Euston open and allows it to become a much more integrated future terminus for today's trains as well as HS2 trains sharing adjacent and integrated platforms.

We get HS2 trains into central London earlier - because its quicker to upgrade the NLL and St Pancras than build a completely new station at Euston.
We boost HS2 revenue - because we remove a 1 hour interchange penalty for about 70% of the HS2 market to add about 15 million extra HS2 passengers /yr ( £1 billion / yr extra revenue by 2040 ).
We also add a competitor to the Elizabeth Line by adding express services (with loos on them) from Ashford, Ebbsfleet and Stratford to St Pancras and Old Oak (plus possibly Heathrow and Woking.to re

Totally agree with getting another Eurostar competitor. But to reach the 2 million passengers/yr still flying London-Paris or London-Brussels and to win over more of the Heathrow-Germany or Amsterdam passengers we need Eurostar to O.O.C.
That also frees St Pancras platforms of some of the demand as 25% of existing passengers would get on at O.O.C along with about 5 million newly won passengers (making most trains full enough to bypass St Pancras & stop at Ashford).



What's not to like

 

[Benjwri]

Totally agree with getting another Eurostar competitor. But to reach the 2 million passengers/yr still flying London-Paris or London-Brussels and to win over more of the Heathrow-Germany or Amsterdam passengers we need Eurostar to O.O.C.
That also frees St Pancras platforms of some of the demand as 25% of existing passengers would get on at O.O.C along with about 5 million newly won passengers (making most trains full enough to bypass St Pancras & stop at Ashford).

This bit I absolutely disagree with. Where is your source St Pancras doesn’t have capacity? Because there have been numerous studies that have found capacity can be more than doubled in the long term.

On market reasons I would also love a source. I used to live in the areas that you mention (Oxfordshire/Berkshire) are ‘losing out’, and I have never heard someone say the London crossing is too difficult, certainly not post Elizabeth Line. The three main reasons I have heard are:
- The Eurostar is too expensive - OOC will make this worse as infrastructure will have to be duplicated, and there will be the same HS1 track access charges, but also access charges on your link and to HS2.
- Difficulty getting to a station in the first place - Usually around parking, high levels of crime in car parks etc, obviously OOC does nothing for this.
- Fear GWR’s constant delays will mean they miss their Eurostar - This is an imagine problem, I suspect Eurostar isn’t fixing it because they don’t need to as they are almost always near or at capacity. If needed this could be fixed with a good ad campaign, the agreement already exists and is used. It would certainly be cheaper than an entire new terminus.

The reality is Heathrow is also a pain to get to from the west. Very few people take the train to it from there, and once you add the change at Paddington, time to Heathrow and the longer time you need to leave before your flight, it is more of a hassle getting there than St Pancras.

 

[The Planner]

We get HS2 trains into central London earlier - because its quicker to upgrade the NLL and St Pancras than build a completely new station at Euston.

You are suggesting upgrading lines already in use, how is that going to be quicker based on the disruptive possessions required as well as the revenue loss and compensation?

 

[MarkyT]

I Think the best way to.make this work is to terminate HS2 trains at Euston.

Fully segregated track work to reduce delays, i can just imagine the reliability issues you would have trying to intertwine HS2, javelin and eurostar services would be disastrous running through St Pancras.

My honest suggestion to help would be to give up on this idea. There is no physical way to make any of this happen in any way that would be beneficial.

This idea is deficient in structual design, track engineering, signal engineering, timetable planning, passenger demand forecast, train speed planning.

It seems to be based upon this wild notion that there is a massive untapped demand of people travelling from Kent to the North West.

Fully agree. Long trains to and from many far-flung places interworking, sharing track, snaking across slow flat junctions in tight margins across Camden, interacting with NLL freight, is a gourmet recipe for a performance meltdown should anything go wrong. Much better to keep them operationally separate and dramatically improve the pedestrian link between the termini, connecting to a new east side Euston entrance. That would also better connect Euston to the Kings Cross goods yard developments for all its passengers. It would cost a fraction of any additional rail works. It's almost certainly not possible to collapse Eurostar to using fewer platforms. While there'll be an element of wanting to keep all the rails shiny, they manage to use all of the International platforms over the course of a day, often with long layovers including sets parked overnight. They no doubt use this time to clean and service the trains. Their operation is restricted significantly due to border concerns. Certain arrival and departure combinations can't be scheduled alongside each other on the same island as that would lead to mixing of passengers after clearing. To leave any margin for growth, whether by Eurostar or a new competitor, the International station can't afford to lose any platform capacity.

 

[bspahh]

Fully agree. Long trains to and from many far-flung places interworking, sharing track, snaking across slow flat junctions in tight margins across Camden, interacting with NLL freight, is a gourmet recipe for a performance meltdown should anything go wrong. Much better to keep them operationally separate and dramatically improve the pedestrian link between the termini, connecting to a new east side Euston entrance. That would also better connect Euston to the Kings Cross goods yard developments for all its passengers. It would cost a fraction of any additional rail works. It's almost certainly not possible to collapse Eurostar to using fewer platforms. While there'll be an element of wanting to keep all the rails shiny, they manage to use all of the International platforms over the course of a day, often with long layovers including sets parked overnight. They no doubt use this time to clean and service the trains. Their operation is restricted significantly due to border concerns. Certain arrival and departure combinations can't be scheduled alongside each other on the same island as that would lead to mixing of passengers after clearing. To leave any margin for growth, whether by Eurostar or a new competitor, the International station can't afford to lose any platform capacity.

Request for Comment documents are used setting standards for computer networking. Most of these are technical. However, RFC 1925 from 1996 has this list of fundamental truths. They were written about computer networking, but they seem appropriate for a rail network too.

The Fundamental Truths

(1) It Has To Work.

(2) No matter how hard you push and no matter what the priority, you can't increase the speed of light.

(2a) (corollary). No matter how hard you try, you can't make a baby in much less than 9 months. Trying to speed this up *might* make it slower, but it won't make it happen any quicker.

(3) With sufficient thrust, pigs fly just fine. However, this is not necessarily a good idea. It is hard to be sure where they are going to land, and it could be dangerous sitting under them as they fly overhead.

(4) Some things in life can never be fully appreciated nor understood unless experienced firsthand. Some things in networking can never be fully understood by someone who neither builds commercial networking equipment nor runs an operational network.

(5) It is always possible to agglutinate multiple separate problems into a single complex interdependent solution. In most cases this is a bad idea.

(6) It is easier to move a problem around (for example, by moving the problem to a different part of the overall network architecture) than it is to solve it.

(6a) (corollary). It is always possible to add another level of indirection.

(7) It is always something.

(7a) (corollary). Good, Fast, Cheap: Pick any two (you can't have all three).

(8) It is more complicated than you think.

(9) For all resources, whatever it is, you need more.

(9a) (corollary) Every networking problem always takes longer to solve than it seems like it should.

(10) One size never fits all.

(11) Every old idea will be proposed again with a different name and a different presentation, regardless of whether it works.

(11a) (corollary). See rule 6a.

(12) In protocol design, perfection has been reached not when there is nothing left to add, but when there is nothing left to take away.

 

[87 027]

While I do not claim the engineering expertise of expert forum members who have provided comments on the proposal, as a finance professional, having read the challenges and the responses put forward to address those challenges, if I were in the position of being an investment committee member judging the merits of the proposals based on the various extracts of presentational material as submitted to this forum, I might have lingering doubts over the:

a) robustness of the forecast benefits and ability to quantify the cashable elements
b) technical feasibility of the proposed solution given the conflicting information
c) completeness of the costs associated with the proposed technical solution
d) timescales associated with the proposed solution

I slightly worry that in commenting "let's work out a way to make this work. What would you suggest to help rather than just finding ways to hinder please" (post 239) and "What's not to like?" (post 242), there is an assumption that the proposed solution is workable with minor tweaks and has skated over deeper concerns such as in posts 192, 196 and 211.

 

[Bald Rick]

Buckle up everyone!

For ease, I have repeated my questions, and added in the answers received:

Q1: What is your cost allowance for the complete replacement of all the track, signalling, overhead line and other assets in the St Pancras approaches? Include the compensation to Eurostar and Southeastern for loss of revenue.

Q2: What cost allowance have you made for the complete rebuilding of track, signalling, overhead line and structures between the Silo curve and Primrose Hill, including a new station for Camden Road, and continued provision for freight to operate via Primrose Hill?

The answer to both questions (in total) appears the be £250m. Hmm.

Q3: What are the results of the assessments you have made of the 80+ different technical and operational factors that have to be conducted when investigating the raising of linespeed? Horizontal alignment is just one of them.

Answer: Haven’t done it.

Q4: What is your proposed solution if (when) it turns out that the previous studies (done by teams of professional engineers) that concluded rebuilding both the NLL tracks for GC gauge was impractical, is correct, and you can’t do the work without completely rebuilding the whole viaduct? What cost allowance have you made for this risk?

Answer: possibility not even considered.

Q5: On what basis have you decided that a repeated 7.5 minute (or 6 minute) departure to departure turnround is in any way practical or physically possible for 200m / 400m high speed rolling stock with intercity door arrangements and long distance passengers potentially with luggage etc? Note how long it takes any long distance train at Euston or Kings Cross to be empty of passengers from the moment it receives a proceed aspect on the home signal, and how long it takes to reload, depart, and then clear the fouling point for the platfrom line on departure. (I have studied this, at length, several times during my career).

Answer: It’s done in Frankfurt, occasionally.

Q6: What cost allowance have you made for gaining the necessary consents to build and operate the new lines, including the legal processes, land purchase, and compensation to nearby affected property owners, businesses and residents who will be affected by the work and the end result?

Answer: no costs allowed for.

Q7: What is your proposed solution for turning trains at Stratford to/from the west, and what cost allowance have you made for this work to be done? If such an engineering solution is not reasonably practical, where do you intend to send the trains and what work is necessary for that to happen?

Answer - some crayons on a picture, and no costs allowed for.

Q8: What is your proposed solution for the signalling of Javelin services on HS2, and HS2 services into St Pancras and onto HS1, and what cost allowance have you made for this? Currently, it is three different signalling systems (HS2 is ETCS L2 with ATO, HS1 from the London tunnels eastwards is TVM430, St Pancras area is KVB.)

Answer: someone else‘s problem.

Q9: What conclusion have you reached when assessing the capacity of the power supply on HS1 to accommodate the extended HS2 services, in particular the capacity of the local distribution around St Pancras to accommodate trains arriving and departing every minute or so. What cost allowance have you made for this?

Answer: someone else’s problem.

Q10: What cost allowances have you made for
a) the link between OOC and the GWML
b) extending the platforms at Heathrow to accept trains longer than 190 metres
c) Providing international border control and security at OOC.

Answer: someone else’s problem.


Now to a few other points:

Track replacement costs are about £150,000 / km

If it were, then the railway would be in a much healthier state than it is today. It is more than 10 times that now, and that is an average, including easy sites with easy access, which this is not (not least as there’s bridges in the way, as you keep saying). You have also missed that when moving tracks, you have to move other elements - Overhead line for example, signals that will need re-siting (and re-sighting), pointwork that has different clearance points due to the wider gauge, trackside equipment that is now foul to gauge of the lower wide profile of GC (including viaduct parapets).

So I’m afraid we have to dismiss your competence as an estimator of railway engineering works.

When TfL trains are not limited to 15 mph, why are you saying that these trains cannot run at around 40 mph? This isn't rocket science. Other trains run this route at 40mph.

You are absolutely right, it is not rocket science. TfL trains are limited to 20mph through Camden Road station through all routes. Any train (TfL or otherwise) heading to or from the Kings Cross Incline or St Pancras is limited to 15mph. Any train heading to or from Primrose Hill is limited to 15mph from a point about 150 metres to the west of Camden Road Junction. This is all as per the Sectional Appendix (in the public domain on NRs website), and signed on the ground, to which all train drivers are obliged to comply. If you are aware of any trains exceeding thse speed limits, please let me know so I can take it up with the appropriate operations director.

I’m afraid we have to dismiss your competence as a Railway Operations expert.

Arup's drawings clearly state that they are using the existing viaduct but relaying tracks and replacing steelwork over the bridges. Look at that drawing.

They also clearly show doing this for only one line at GC gauge, not two. That is, to get GC gauge in, you have to move both tracks (and signals, and OLE, and any other equipment in the way), to get just one track available at GC gauge.

Look at where those Javelins go. Two per hour at rush hour really do only go to Ashford.

They really don’t.

Here is tomorrow’s evening peak Southeastern High Speed departures from St Pancras for the two hours from 1707.



Please can you point out which of these 13 trains “only go to Ashford”.

I’m afraid we have to dismiss your competence as a reader of timetables.

As for train turnaround times, look at TfL trains which turn in 2 minutes. These are only 200 metre long classic trains where the driver needs 2 minutes to walk to the end.

Aside from the Elizabeth Line Class 345s (205m) there are no TfL services that are 200 metres long, and none that have planned turnrounds in 2 minutes (not even the Waterloo and City Line, with 65 metre trains.) The Elizabeth line’s shortest planned turnrounds are (IIRC, please correct me if I’m wrong) 8 minutes, with a 10 minute departure to departure cycle off the same terminus platform, in the peak only.

And on the basis that Metro trains do this in this country all of the time. Why is 7.5 minutes too short a time to just reverse a 200 metre long train? (none are 400m ones).

1) there will be 400m trains. HS2 Birmingham services will be 400m from the off, as will services to Scotland when they start. Most peak Javelin trains are 240m long also. And then you mention Eurostars going to OOC. They are 390m.

2) Your 7.5minutes is the time between successive departures on the same platform. That is:

a) train 1 departs
b) train 1 clears the junction for a subsequent train to arrive
c) signalling system resets the route from departure to arrival
d) train 2 approaches, having been no closer than the ‘home’ signal, then arrives
e) passengers alight
f) any train servicing requirements (restock, cleaning, running repairs)
g) staff changeover / handover
h) passengers board
i) signalling system resets the route for a departure
j) proceed aspect displayed
k) dispatch procedure initiated
l) dispatch prociedure completed, train departs.

All in 7.5 minutes (correct me if I have this wrong).

Given that the minimum junction margin at St Pancras is 3 minutes for conflicting moves anywhere on the international part of the layout, and the minimum reoccupation time is three minutes for the domestic platforms and 4 minutes for the international platforms (as per the Timetable Planning Rules), that means 6 or 7 minutes of your 7.5 minute turnrounds are taken up by a) to d). That leaves 30-90 seconds for e) to l). Bear in mind these are for departures to from HS1, and not for your proposed lower speed route to HS2.

I’m afraid we are going to have to dismiss your ability as a timetable planner.

This is an "upgrade of an existing railway" - so the bridges only need Camden Council's planning permission.

It isn’t. It is new railway, with change of land use as well. You will need primary consent. I assume that you consider this to be a strategic proposal in the national interest, in other words a Nationally Significant Infrastructure Project. That automatically requires a Development Consent Order, and given what has already happened with this project I wouldn’t bet against it being called in for Parliamentary debate, ie requiring an Act.

This calls into question your expertise in town planning matters.

I keep telling you that only one property needs any land purchase.

Yes you do. And I’m afraid you are wrong.

One small example - To get the extra tracks round the corner towards Gospel Oak from Camden Road needs the permanent acquisition of part of the playground of Hawley Primary School to widen the viaduct. How do you get access to that land to pile the foundations of the new viaduct? Just outside school hours? Or perhaps only during lessons (excluding outdoor PE?)

You also mentioned earlier that to get GC gauge on the existing viaducts you could demolish the parapet walls and cantilever walkways off the side. Leaving aside the practicality of that, that is oversailing others’ property.

I’m afraid this means we can discount your understanding of the land required for major railway constructuon projects.

As this is an existing railway, these electric trains would make less noise than diesel freight trains.

The diesel freight trains would still be there, along with an extra 16tph HS2 trains (by your estimate) and 8tph Javelins trundling through Camden. That will be more noise, and lots of it. There is also the noise during construction. I have lived and worked next to major railway construction sites, indeed worked on them and had to deal with neighbours - have you?

What "compensation" when residents chose to live by this existing railway?

They didn’t chose to live beside a new, wider railway with an extra 24 trains per hour (minimum) running past their back doors, nor a major construction site. The compensation for the Agar Grove bridge superstructure renewal in Camden ran to significant sums (that I can’t quote here). The compensation bill for HS2 is billions.

Road traffic would be the main concern for the 6 week periods needed to replace bridge superstructures

Agar Grove in Camden was (and still is) closed for 9 months for a deck replacement. Other durations of closure are of course possible. But six weeks on site is fundamentally impractical.

I’m afraid we are going to have to dismiss your competence in construction planning.

HS1 are thus already in the process of changing to ETCS according to recent publicity.

They are in the process of thinking about it. They have no confirmed plans to implement ETCS - and no need to as the current TVM430 is barely 20 years old. The (different) KVB system at St Pancras turns 18 this year. The interface between the KVB and Conventional NR Computer Based interlocking signalling on the North London Line at Camden (controlled from Upminster) has never been commissioned.

We are going to have to dismiss your understanding of railway signalling systems.

An S curve reversable single track ramp through the station box to OOC GWR platforms doesn't need new tunnels (except for a short cut & cover at the top).
It almost all fits within the station box.

“Almost”.

I’m sure you know that when designing a new railway ”it does, or does not; there is no almost”. (with apologies to Yoda)

Crudely drawn, here is how the existing route to the Temple Mills S curve could have two extra tracks added to connect to the central platforms to allow trains to reach or reverse from the opposite platform.

You appear to have missed the central supporting columns down the middle of that covered way. Also the connections to Temple Mills and the height difference of that line.

We can therefore dismiss your competence as a track designer, as this is basic stuff.

A solution - It would be possible to provide the crossover, at the cost of building a new depot for Eurostar somewhere else (Temple Mills cost £400m 20 years ago), and rebuilding the covered way, with signifcant closures and temproary propping, and closing the DLR and freight lines above while you did it. Allow £2bn.

Concrete Seriously there is a concrete factory right there on site.

Indeed, but you propose building a viaduct right through it, which would render it unusable, at least during the construction period and probably permanently (if they have to move out for months, they may as well go permanently). You will need to find another site for the business, and it needs to be in the vicinity as concrete has a limited time from being mixed at the batching plant to being used on site (which is why the site is there now).

To get HS2 trains to actually reach central London we have to stop being so purist. The route from O.O.C. to London is a 70mph railway

The HS2 route from OOC to Euston is almost entirely a 100mph+ railway; a third of it is 140mph+. All of the section from Primrose Hill to OOC in 100mph+. This is shown by the linespeed profiles on the HS2 sectional drawings, in the public domain on the DfT website.

Please check things like this before stating them as fact. What you are writing is not true.

Most of the demand is from South London and North London via the Thameslink network

As written, you are suggesting that more than half the demand for London to the West Midlands / North West and Scotland origninates on the Thameslink network. I don’t believe this is true, but have not got to hand the data to prove otherwise.

Please can you provide a source for this; as above you do have form for stating as fact things that are not supported by evidence and are not true.

Totally agree with getting another Eurostar competitor. But to reach the 2 million passengers/yr still flying London-Paris or London-Brussels and to win over more of the Heathrow-Germany or Amsterdam passengers we need Eurostar to O.O.C.

No, you don’t. In my opinion you need more trains from London - Paris / Brussels / other destinations, perhaps coupled with bigger incentives to travel by rail vs air, ie higher Air Passenger Duty. All of which comes at no additional cost in terms of capital expenditure on infrastructure in the UK.


A final point.

Why would HS1 want to go to the expense of remodelling, resignalling and reconfiguring their infrastrucutre at St Pancras, Stratford, Ebbslefft and the like with no guarantees of a return on their investment?

Please don’t reply with copies of your presentation again, we’ve all seen enough of it, and know what your proposal is.

 

[MotCO]

Game, set and match?

 

[Nym]

Game, set and match?

With the level of fixation, I doubt it.

 

[John Jefkins]

This bit I absolutely disagree with. Where is your source St Pancras doesn’t have capacity? Because there have been numerous studies that have found capacity can be more than doubled in the long term.

On market reasons I would also love a source. I used to live in the areas that you mention (Oxfordshire/Berkshire) are ‘losing out’, and I have never heard someone say the London crossing is too difficult, certainly not post Elizabeth Line. The three main reasons I have heard are:
- The Eurostar is too expensive - OOC will make this worse as infrastructure will have to be duplicated, and there will be the same HS1 track access charges, but also access charges on your link and to HS2.
- Difficulty getting to a station in the first place - Usually around parking, high levels of crime in car parks etc, obviously OOC does nothing for this.
- Fear GWR’s constant delays will mean they miss their Eurostar - This is an imagine problem, I suspect Eurostar isn’t fixing it because they don’t need to as they are almost always near or at capacity. If needed this could be fixed with a good ad campaign, the agreement already exists and is used. It would certainly be cheaper than an entire new terminus.

The reality is Heathrow is also a pain to get to from the west. Very few people take the train to it from there, and once you add the change at Paddington, time to Heathrow and the longer time you need to leave before your flight, it is more of a hassle getting there than St Pancras.

Sources are on that graphic (and capacity issues for the border controls are evidenced by the long queues at times at St Pancras plus worries about the new ETIAS scheme).

I agree that HS1 are indeed finding ways to improve capacity, but my doubt is that they could double it for Eurostar passengers. Meanwhile the 1st floor has plenty of capacity for more domestic passengers.
The main reasons for a cross London link is to boost the domestic markets from the home counties (Kent, Essex, Suffolk, Herts to Surrey, Berks, Oxfordshire) as well as provide better east-west cross London capacity.

I agree Eurostar is expensive, which is why we need competitors to force the cost down.
Eurostar would though only be around 2 of the potentially 12 to 14 domestic trains per hour using the NLL link each way to HS1.

 

[Class15]

Game, set and match?

With the level of fixation, I doubt it.

It should have been ‘game, set and match’ in page 1 of this thread. We’ve had 250 posts on this thread - I don’t think we’ve heard the last of this!

 

[John Jefkins]

Game, set and match?

Far from it - as he is not comparing like with like.

Agar Road replacement was on a working railway.

The replacement route for the NLL could be created before any rail closures - so compensation would be limited to the switch over period.
And compensation wouldn't cost £6 billion would it...

My costs of about £250m won't be far out.

Buckle up everyone!

For ease, I have repeated my questions, and added in the answers received:





The answer to both questions (in total) appears the be £250m. Hmm.



Answer: Haven’t done it.




Answer: possibility not even considered.




Answer: It’s done in Frankfurt, occasionally.




Answer: no costs allowed for.



Answer - some crayons on a picture, and no costs allowed for.




Answer: someone else‘s problem.




Answer: someone else’s problem.



Answer: someone else’s problem.


Now to a few other points:




If it were, then the railway would be in a much healthier state than it is today. It is more than 10 times that now, and that is an average, including easy sites with easy access, which this is not (not least as there’s bridges in the way, as you keep saying). You have also missed that when moving tracks, you have to move other elements - Overhead line for example, signals that will need re-siting (and re-sighting), pointwork that has different clearance points due to the wider gauge, trackside equipment that is now foul to gauge of the lower wide profile of GC (including viaduct parapets).

So I’m afraid we have to dismiss your competence as an estimator of railway engineering works.




You are absolutely right, it is not rocket science. TfL trains are limited to 20mph through Camden Road station through all routes. Any train (TfL or otherwise) heading to or from the Kings Cross Incline or St Pancras is limited to 15mph. Any train heading to or from Primrose Hill is limited to 15mph from a point about 150 metres to the west of Camden Road Junction. This is all as per the Sectional Appendix (in the public domain on NRs website), and signed on the ground, to which all train drivers are obliged to comply. If you are aware of any trains exceeding thse speed limits, please let me know so I can take it up with the appropriate operations director.

I’m afraid we have to dismiss your competence as a Railway Operstions expert.




They also clearly show doing this for only one line at GC gauge, not two. That is, to get GC gauge in, you have to move both tracks (and signals, and OLE, and any other equipment in the way), to get just one track available at GC gauge.




They really don’t.

Here is tomorrow’s evening peak Southeastern High Speed departures from St Pancras for the two hours from 1707.

View attachment 177471

Please can you point out which of these 13 trains “only go to Ashford”.

I’m afraid we have to dismiss your competence as a reader of timetables.




Aside from the Elizabeth Line Class 345s (205m) there are no TfL services that are 200 metres long, and none that have planned turnrounds in 2 minutes (not even the Waterloo and City Line, with 65 metre trains.) The Elizabeth line’s shortest planned turnrounds are (IIRC, please correct me if I’m wrong) 8 minutes, with a 10 minute departure to departure cycle off the same terminus platform, in the peak only.



1) there will be 400m trains. HS2 Birmingham services will be 400m from the off, as will services to Scotland when they start. Most peak Javelin trains are 240m long also. And then you mention Eurostars going to OOC. They are 390m.

2) Your 7.5minutes is the time between successive departures on the same platform. That is:

a) train 1 departs
b) train 1 clears the junction for a subsequent train to arrive
c) signalling system resets the route from departure to arrival
d) train 2 approaches, having been no closer than the ‘home’ signal, then arrives
e) passengers alight
f) any train servicing requirements (restock, cleaning, running repairs)
g) staff changeover / handover
h) passengers board
i) signalling system resets the route for a departure
j) proceed aspect displayed
k) dispatch procedure initiated
l) dispatch prociedure completed, train departs.

All in 7.5 minutes (correct me if I have this wrong).

Given that the minimum junction margin at St Pancras is 3 minutes for conflicting moves anywhere on the international part of the layout, and the minimum reoccupation time is three minutes for the domestic platforms and 4 minutes for the international platforms (as per the Timetable Planning Rules), that means 6 or 7 minutes of your 7.5 minute turnrounds are taken up by a) to d). That leaves 30-90 seconds for e) to l). Bear in mind these are for departures to from HS1, and not for your proposed lower speed route to HS2.

I’m afraid we are going to have to dismiss your ability as a timetable planner.





It isn’t. It is new railway, with change of land use as well. You will need primary consent. I assume that you consider this to be a strategic proposal in the national interest, in other words a Nationally Significant Infrastructure Project. That automatically requires a Development Consent Order, and given what has already happened with this project I wouldn’t bet against it being called in for Parliamentary debate, ie requiring an Act.

This calls into question your expertise in town planning matters.




Yes you do. And I’m afraid you are wrong.

One small example - To get the extra tracks round the corner towards Gospel Oak from Camden Road needs the permanent acquisition of part of the playground of Hawley Primary School to widen the viaduct. How do you get access to that land to pile the foundations of the new viaduct? Just outside school hours? Or perhaps only during lessons (excluding outdoor PE?)

You also mentioned earlier that to get GC gauge on the existing viaducts you could demolish the parapet walls and cantilever walkways off the side. Leaving aside the practicality of that, that is oversailing others’ property.

I’m afraid this means we can discount your understanding of the land required for major railway constructuon projects.




The diesel freight trains would still be there, along with an extra 16tph HS2 trains (by your estimate) and 8tph Javelins trundling through Camden. That will be more noise, and lots of it. There is also the noise during construction. I have lived and worked next to major railway construction sites, indeed worked on them and had to deal with neighbours - have you?




They didn’t chose to live beside a new, wider railway with an extra 24 trains per hour (minimum) running past their back doors, nor a major construction site. The compensation for the Agar Grove bridge superstructure renewal in Camden ran to significant sums (that I can’t quote here). The compensation bill for HS2 is billions.




Agar Grove in Camden was (and still is) closed for 9 months for a deck replacement. Other durations of closure are of course possible. But six weeks on site is fundamentally impractical.

I’m afraid we are going to have to dismiss your competence in construction planning.




They are in the process of thinking about it. They have no confirmed plans to implement ETCS - and no need to as the current TVM430 is barely 20 years old. The (different) KVB system at St Pancras turns 18 this year. The interface between the KVB and Conventional NR Computer Based interlocking signalling on the North London Line at Camden (controlled from Upminster) has never been commissioned.

We are going to have to dismiss your understanding of railway signalling systems.




“Almost”.

I’m sure you know that when designing a new railway ”it does, or does not; there is no almost”. (with apologies to Yoda)



You appear to have missed the central supporting columns down the middle of that covered way. Also the connections to Temple Mills and the height difference of that line.

We can therefore dismiss your competence as a track designer, as this is basic stuff.

A solution - It would be possible to provide the crossover, at the cost of building a new depot for Eurostar somewhere else (Temple Mills cost £400m 20 years ago), and rebuilding the covered way, with signifcant closures and temproary propping, and closing the DLR and freight lines above while you did it. Allow £2bn.




Indeed, but you propose building a viaduct right through it, which would render it unusable, at least during the constructuon period and probably permanently (if they have to move out for months, they may as well go permanently). You will need to find another site for the business, and it needs to be in the vicinity as concrete has a limited time from being mixed at the batching plant to being used on site (which is why the site is there now).




The HS2 route from OOC to Euston is almost entirely a 100mph+ railway; a third of it is 140mph+. All of the section from Primrose Hill to OOC in 100mph+. This is shown by the linespeed profiles on the HS2 sectional drawings, in the public domain on the DfT website.

Please check things like this before stating them as fact. What you are writing is not true.




As written, you are suggesting that more than half the demand for London to the West Midlands / North West and Scotland origninates on the Thameslink network. I don’t believe this is true, but have not got to hand the data to prove otherwise.

Please can you provide a source for this; as above you do have form for stating as fact things that are not supported by evidence and are not true.




No, you don’t. In my opinion you need more trains from London - Paris / Brussels / other destinations, perhaps coupled with bigger incentives to travel by rail vs air, ie higher Air Passenger Duty. All of which comes at no additional cost in terms of capital expenditure on infrastructure in the UK.


A final point.

Why would HS1 want to go to the expense of remodelling, resignalling and reconfiguring their infrastrucutre at St Pancras, Stratford, Ebbslefft and the like with no guarantees of a return on their investment?

Please don’t reply with copies of your presentation again, we’ve all seen enough of it, and know what your proposal is

It's mother's day. I'm off out. Your reply isn't accurate and you really need to think it through. What alternative would you propose to reduce the crazy cost of Euston?

Its done at Charing Cross every rush hour by the way. 44 movements on a flat junction.

You can wait a few hours for a proper reply as I'll answer each of your points patiently yet again this evening.....
Watch this space.

 

[Class15]

that means 6 or 7 minutes of your 7.5 minute turnrounds are taken up by a) to d).

There is absolutely no way that one train leaving and the next one arriving is taking anywhere near 6 minutes. Loads of termini do it in 2 or 3 minutes.

 

[Bald Rick]

There is absolutely no way that one train leaving and the next one arriving is taking anywhere near 6 minutes. Loads of termini do it in 2 or 3 minutes.

The planning rules for the international platforms at St Pancras international require it.

Its done at Charing Cross every rush hour by the way. 44 movements on a flat junction.

49 actually. albeit across 3 flat juctions, one of which is well over a mile out, which splits it into 2 seperate 3 platform stations each with 2 approach tracks (3 approach tracks for one half for the last section).

But, crucially, never more less than 12 11 minutes between departures at any platform, and more usually 13-14. Not 7.5 minutes.
(edited to correct mself)

Game, set and match?

Don’t be silly!

 

[Class15]

The planning rules for the international platforms at St Pancras international require it.

Out of interest, why? It seems like a weird requirement to me.

 

[MotCO]

But, crucially, never more than 12 minutes between departures at any platform. Not 7.5 minutes.

Less?

 

[yoyothehobo]

Far from it - as he is not comparing like with like.

Agar Road replacement was on a working railway.

The replacement route for the NLL could be created before any rail closures - so compensation would be limited to the switch over period.
And compensation wouldn't cost £6 billion would it...

My costs of about £250m won't be far out.



It's mother's day. I'm off out. Your reply isn't accurate and you really need to think it through. What alternative would you propose to reduce the crazy cost of Euston?

Its done at Charing Cross every rush hour by the way. 44 movements on a flat junction.

You can wait a few hours for a proper reply as I'll answer each of your points patiently yet again this evening.....
Watch this space.

You are delusional i am afraid and i fear you are living in an echo chamber and cannot face the fact you are wrong.

I really cannot believe @Bald Rick has gone to such the time and effort with the reply which to be honest this idea scarcely deserves.

In fact the reply is probably best pinned as an example to what is required when planning railway routes when faced with such ludicrous ideas.

 

[growse]

I think we should consider terminating the HS2 trains at the old Eurostar platforms at Waterloo. Think of the cross connectivity benefits for the good people of Surrey and the southwest (aren't they always complaining how HS2 doesn't benefit them?). There's an easy connection from OOC to the WLL (just need to use our imaginations).

If we re-enable the ebbsfleet and stewart's lane connections, we can also run more Eurostar services into Waterloo to connect and grow the continental market! Think of the possibilities!

All done for half a billion I reckon.

Agar Road replacement was on a working railway.

The NLL and St Pancras throat, meanwhile are so barely used that they might as well be disused. Easy!

 

[Class15]

The NLL and St Pancras throat, meanwhile are so barely used that they might as well be disused. Easy!

Yeah exactly, not like a travel on a packed train every morning on the NLL (admittedly not through Camden Road, I travel Gospel Oak - West Hampstead) is it. Camden Road is one of the busiest stations on the line too, you can’t just close it for an extended period.

 

[yoyothehobo]

I think we should consider terminating the HS2 trains at the old Eurostar platforms at Waterloo. Think of the cross connectivity benefits for the good people of Surrey and the southwest (aren't they always complaining how HS2 doesn't benefit them?). There's an easy connection from OOC to the WLL (just need to use our imaginations).

If we re-enable the ebbsfleet and stewart's lane connections, we can also run more Eurostar services into Waterloo to connect and grow the continental market! Think of the possibilities!

All done for half a billion I reckon.


The NLL and St Pancras throat, meanwhile are so barely used that they might as well be disused. Easy!

to be honest i think we missed a trick not running HS2 services through the Elizabeth line and out the east side of London somewhere back on to HS1 somewhere. I am sure a 400m HS2 train can fit through no problem with no extta work, i reckon it could be done for about 100 million.

Come to think of it are there any major captial cities in the world that run through services through their major terminii or reverse continous services through them?

Paris doesnt, Tokyo doesnt.

 

[JonathanH]

I agree Eurostar is expensive, which is why we need competitors to force the cost down.

Are Eurostar making supernormal profits or do they just have a high cost of operation? Is another organisation in the supply chain making supernormal profits? How does competition help in this particular market?

I could see that it might allow fixed costs to be spread better, but is there a genuine prospect that competition in this market could actually deliver lower end user prices?

 

[Bald Rick]

Out of interest, why? It seems like a weird requirement to me.

It’s simple, it takes that long for a Eurostar to get out, clear of the conflicting junctions, and for the next arrival to be signalled in (on non-restrictive ’aspects’) and then arrive.

It is quicker at some other termini as the ’home’ signals, and relevant junctions, will be closer to the platform ends, and the trains shorter of course and therefore able to clear the fouling points more quickly.

Less?

Oops, sorry, I was wrong, also wrong on the minimum departure to departure time, albeit 12 minutes is the minimum if the second train is longer than 8 cars. I have corrected.

Come to think of it are there any major captial cities in the world that run through services through their major terminii or reverse continous services through them?

Berlin at HBf for one. It’s not particularly unusual.
However it is absolutely not done off 3 or 4 platforms with 24+ departures per hour.

Your reply isn't accurate and you really need to think it through.

I have thought my reply through - for two days.

If it is not accurate, then I will be happy to be corrected, as indeed I have been. I note that where you have been inaccurate, you have not acknowledged that, or dismissed it, or diverted the discussion.

What alternative would you propose to reduce the crazy cost of Euston?

Build Euston more quickly. It will be cheaper, and we get the benefits earlier.

Given that a decent proporton of the money for Euston has been spent or committed already, your proposal does not have consent, has no design, no assessment of risks, is not operable, in some cases is not physically possible, and has benefits that are grossly overstated (and not explained, despite repeated requests from various posters), I think I know what most people reading this thread will think.

Of course I might be wrong, and will happily stand corrected. Will you?

 

[Class15]

It’s simple, it takes that long for a Eurostar to get out, clear of the conflicting junctions, and for the next arrival to be signalled in (on non-restrictive ’aspects’) and then arrive.

It is quicker at some other termini as the ’home’ signals, and relevant junctions, will be closer to the platform ends, and the trains shorter of course and therefore sble to clear the fouling points more quickly.

Thanks @Bald Rick , that makes sense, seems a bit silly to build the junctions so far out but that’s perfectly understandable.

 

[Stephen42]

There is absolutely no way that one train leaving and the next one arriving is taking anywhere near 6 minutes. Loads of termini do it in 2 or 3 minutes.

I agree I think that is an incorrect interpretation of the rules or an accidental double count. St Pancras has 3 minute platform reoccupation booked on the domestic platforms, Eurostar would be longer with the longer trains. That would leave 4.5 minutes on the domestic for a departure every 7.5 minutes if a single platform is in use.

However that is considering each platform in isolation not the overall layout. Looking at either St Pancras HS1 domestic or half of Charing Cross that's three platforms with layout such that parallel moves in and out can take place provided the in and out don't directly cross. Taking 3 minutes between conflicting moves (which is the figure in train planning rules for platforms 1-3 of Charing Cross) needs 4 conflicting moves as either arrival or departure on platforms at each end blocks all other moves. That cycle takes 12 minutes which is the limiting factor and limits the three platforms to 15 departures per hour.

The general point still stands, to get to 24 from three platforms seems entirely unrealistic as would need to get it down to under 2 minutes between conflicting moves. London underground achieve it but with much shorter trains and faster speeds from overrun tunnels not comparable to St Pancras. Also these situations aren't with two different destinations with each train needing to take a specific track to get to the right place. The layout at St Pancras would need to be extensively looked at to achieve this and it is a very constrained site.

 

[Bald Rick]

Thanks @Bald Rick , that makes sense, seems a bit silly to build the junctions so far out but that’s perfectly understandable.

The junctions have to be that far out, as they are relatively high speed for a terminus, and therefore quite long. Also, the length of the train makes a material difference.

I agree I think that is an incorrect interpretation of the rules or an accidental double count. St Pancras has 3 minute platform reoccupation booked on the domestic platforms

I agree it is 3 minute reoccupation on the domestic platforms - obviously for a non conflicting move on the approach. It is 4 minutes for the international platforms.

My assessment is built on two points of variation, one of which you mention in your post:

1) for trains to get between the domestic platforms and the link to the NLL, they will have to make use of the international part of the layout, and therefore the longer time could reasonably be expected to apply, especially considering the lower speed of 25mph onto the Silo curve towards the NLL. Albeit our proposer has ignored the linespeed on that curve and thinks it will somehow, magically, be quicker.

2) because of the quantity of movements - as you say you can’t consider one platform in isolation. 10 HS2 movements per hour and up to 8 Southeastern High Speed movements per hour, each way, to each of HS2 and HS1, plus up to 4 Eurostar / international movements each way to HS1, that is up to 80 trains per hour through the throat of St Pancras international. Even with parallel moves and additional grade seperation, it is inconceivable that these can all be done without regular conflicts in the throat, and with limited space between the junctions the junction margin will also apply. I will agree that it won‘t apply to every move as some will be parallel, but is likely to apply to many.

Of course what we need to see is a timetable of the proposed station workings. But the proposer won’t do that as it will be either someone else’s problem, or possibly after a couple of hours trying will find it is physically impossible and give up.

 

[Class15]

The junctions have to be that far out, as they are relatively high speed for a terminus, and therefore quite long. Also, the length of the train makes a material difference.

Ahhhh, thank you very much for the answer.

 

[The Ham]

We boost HS2 revenue - because we remove a 1 hour interchange penalty for about 70% of the HS2 market to add about 15 million extra HS2 passengers /yr ( £1 billion / yr extra revenue by 2040 ).

You've still not answered which if these the 1 hour interchange penalty relates to:

a) Euston to St Pancras takes an hour (no allowance for an interchange between St Pancras services)

b) Euston to St Pancras takes 75 minutes and an interchange between services at St Pancras is 15 minutes leaving a difference of an hour (but doesn't allow for a long distance interchange penalty for people trying to ensure they've made their connection)

c) Euston to St Pancras takes 95 minutes and an interchange between services at St Pancras is 35 minutes leaving a difference of an hour (does allow for making the connection at Station Pancras to a long distance service)

Sources are on that graphic (and capacity issues for the border controls are evidenced by the long queues at times at St Pancras plus worries about the new ETIAS scheme).

I agree that HS1 are indeed finding ways to improve capacity, but my doubt is that they could double it for Eurostar passengers. Meanwhile the 1st floor has plenty of capacity for more domestic passengers.
The main reasons for a cross London link is to boost the domestic markets from the home counties (Kent, Essex, Suffolk, Herts to Surrey, Berks, Oxfordshire) as well as provide better east-west cross London capacity.

I agree Eurostar is expensive, which is why we need competitors to force the cost down.
Eurostar would though only be around 2 of the potentially 12 to 14 domestic trains per hour using the NLL link each way to HS1.

From Surrey (likewise Hampshire) it's not easy to get to St Pancras or Old Oak Common, so not sure it'll help that much for that county.

My costs of about £250m won't be far out.

I suspect that they will be.

Construction inflation has been mad in the last 5 years.

 

[zwk500]

You've still not answered which if these the 1 hour interchange penalty relates to:

a) Euston to St Pancras takes an hour (no allowance for an interchange between St Pancras services)

b) Euston to St Pancras takes 75 minutes and an interchange between services at St Pancras is 15 minutes leaving a difference of an hour (but doesn't allow for a long distance interchange penalty for people trying to ensure they've made their connection)

c) Euston to St Pancras takes 95 minutes and an interchange between services at St Pancras is 35 minutes leaving a difference of an hour (does allow for making the connection at Station Pancras to a long distance service)

On the PDFH point, I'll point out that the interchange penalties are applied based on the distance of the total journey being made, so scale accordingly. The function of the interchange penalty is to model (important word - not replicate) the function of demand where people will choose a slower but no-change train over a potentially faster route that involves changing, but up to a point. As London-Paris is 212 miles, the Interchange Penalties for anybody changing into Eurostar should start at the 200 miles and up values.

When applying PDFH you calculate the *generalised* journey time for the end-to-end journeys (which includes transfer and waiting time such as check-in) and then add the required interchange penalties. PDFH also requires that a station-to-station transfer be treated as 2 interchange penalties (as mentioned above, I think that in general this is pessimistic and could be factored down to 1.5 and for Eurostar some modification is needed for the check-in process as this time is already an interchange penalty).

Given the length of waiting time and service interval penalties for Eurostar, I'd suggest the interchange time saving is a smaller proportion of the GJT than the OP thinks. Maybe this evening I'll sit down and do a spread of journeys from around London and see what the current proportion of interchange penalty is as % of GJT.

 

[MarkyT]

On the PDFH point, I'll point out that the interchange penalties are applied based on the distance of the total journey being made, so scale accordingly. The function of the interchange penalty is to model (important word - not replicate) the function of demand where people will choose a slower but no-change train over a potentially faster route that involves changing, but up to a point. As London-Paris is 212 miles, the Interchange Penalties for anybody changing into Eurostar should start at the 200 miles and up values.

When applying PDFH you calculate the *generalised* journey time for the end-to-end journeys (which includes transfer and waiting time such as check-in) and then add the required interchange penalties. PDFH also requires that a station-to-station transfer be treated as 2 interchange penalties (as mentioned above, I think that in general this is pessimistic and could be factored down to 1.5 and for Eurostar some modification is needed for the check-in process as this time is already an interchange penalty).

Given the length of waiting time and service interval penalties for Eurostar, I'd suggest the interchange time saving is a smaller proportion of the GJT than the OP thinks. Maybe this evening I'll sit down and do a spread of journeys from around London and see what the current proportion of interchange penalty is as % of GJT.

If stations are right next door like kings cross and st pancras I think adding more than one interchange penalty is too much. According to the model's logic, tearing down all the historic buildings at each and constructing a grand new modern terminal uniting them with a massive shared concourse would improve attractiveness of journeys even though all the platforms would be exactly where they are today and no passenger would have a shorter walk.