What could/should have been done to boost capacity on the MML?

[Trainman40083]

Yes, that's why end coaches are A and F, because plan was to buy additional intermediate coaches as business grew (but Virgin never did, then over time DfT got more involved with decisions and expansion strategy was killed). The poor layout was because they also wanted 3 class trains, each with own large toilet.

Yes, but due to DfT dithering, the extra coach had an increasing shorter lifespan if it was expected to be scrapped same time as original voyagers, so its cost-benefit ratio got worse over time.

Back to the 810s, and the complete rehash of a relatively space efficient 26m 80x fleet, altering about 70% to squeeze it into 24m has in redesign, recertification, and delays, cost (along with unplanned extension of 222s) hundreds of millions of pounds. I would not be surprised if it has ended up lot more than adding 20m platform extensions to a few stations.

Whats more nobody else likes the bodged 24m design of the 810s, open access operators prefer the more space efficient and known design 80x 26m version. Even DfT controlled operators have never ordered more 810 style trains, so it will remain a non standard oddity until 2060s

mods note - split from this thread

Class 810 for East Midlands Railway Construction/Introduction Updates

Looks like the new Hitachi AT300's based on the IET platforn for EMR are going to be called the Class 804. These are going to be bi-mode trains and should be capable of running at up to 140mph, even if 125 is all they see in service. We're looking at 33, 5-car long sets, with 23m long coaches...

www.railforums.co.uk

Maybe it would have been easier and cheaper to just extend St Pancras platforms

 

[Failed Unit]

Maybe it would have been easier and cheaper to just extend St Pancras platforms

If we look at the loading on the MML now (granted we didn't have crystal ball). Perhaps they should have ordered the standard 9 car 26m version. Until we see the diagrams we don't know how many will split / join during the day or how many all day 10 car diagrams will exist.

If you take the Lincoln extension, LNER already operate 9 car, Newark Castle would not really work, but they probably could kill the direct to London services off now. They only run Southbound now and EMR have 3 car trains now so they might not be needed now anyway.

 

[Merle Haggard]

Maybe it would have been easier and cheaper to just extend St Pancras platforms

What puzzles me is the comparison between the MML platforms (short, out on a limb, badly lit, draughty) with the ones for the services to East Kent. Travelling to Ashford recently I was amazed by the opulence of the platforms and the luxury of the trains (though I had arrived on a 360 from Wellingborough)

Presumably some planner predicted that there would be a massive demand for travel from London to Ashford, Canterbury Margate and Ramsgate - even though there was already another service - but thought that the demand from London to Leicester/Derby/Nottingham/Sheffield (let alone the smaller towns serve - most still larger than those in East Kent) one be tiny. So platforms for 12 car trains were built for the former and short ones for the latter.

 

[londonmidland]

Maybe it suggests a serious lack of capacity which stops late trains recovering their schedules.

Around Sheffield, it's congested at the best of times, however one small delay can cause a ripple effect to other services, as mentioned.

You've got multiple services departing minutes between each other, all utilising the same tracks. You have the XX:56 Southbound XC, the XX:00 EMR St. Pancras, XX:06 NT Nottingham, XX:11 TPE Liverpool and finally the XX:14 NT Manchester. The XX:00 EMR will usually have to wait for the XX:56 XC, if its running a few minutes late.

 

[ChilliSauce]

And further

What puzzles me is the comparison between the MML platforms (short, out on a limb, badly lit, draughty) with the ones for the services to East Kent. Travelling to Ashford recently I was amazed by the opulence of the platforms and the luxury of the trains (though I had arrived on a 360 from Wellingborough)

Presumably some planner predicted that there would be a massive demand for travel from London to Ashford, Canterbury Margate and Ramsgate - even though there was already another service - but thought that the demand from London to Leicester/Derby/Nottingham/Sheffield (let alone the smaller towns serve - most still larger than those in East Kent) one be tiny. So platforms for 12 car trains were built for the former and short ones for the latter

And further more the "old St Pancras" pre Euro-Star was a 6 platform station: 2 - 7 ...with the shorter platform 1 after having had slab track fitted as part of the Midland Suburban Electrification and associated resignalling never getting "connected up", saved 1 point end!

So it was infact Euro-Stars arrival that reduced platform capacity at Pancras for MML ..
Somehow I think the MML was "short changed" on the deal ... If lesser platforms was the offering then the platforms should of been longer.

== Doublepost prevention - post automatically merged: 26 Feb 2026 ==

Around Sheffield, it's congested at the best of times, however one small delay can cause a ripple effect to other services, as mentioned.

You've got multiple services departing minutes between each other, all utilising the same tracks. You have the XX:56 Southbound XC, the XX:00 EMR St. Pancras, XX:06 NT Nottingham, XX:11 TPE Liverpool and finally the XX:14 NT Manchester. The XX:00 EMR will usually have to wait for the XX:56 XC, if its running a few minutes late.

Sounds to me like a justification to reinstate the 4 track: Sheffield to Dore

 

[Trainman40083]

And further

And further more the "old St Pancras" pre Euro-Star was a 6 platform station: 2 - 7 ...with the shorter platform 1 after having had slab track fitted as part of the Midland Suburban Electrification and associated resignalling never getting "connected up", saved 1 point end!

So it was infact Euro-Stars arrival that reduced platform capacity at Pancras for MML ..
Somehow I think the MML was "short changed" on the deal ... If lesser platforms was the offering then the platforms should of been longer.

== Doublepost prevention - post automatically merged: 26 Feb 2026 ==


Sounds to me like a justification to reinstate the 4 track: Sheffield to Dore

Thought they had been talking about 4 tracking Sheffield to Dore for years. Finally got 3 tracks through Dore. I have no idea how they expected to get HS2 trains through Sheffield Midland. There was the campaign to get HS2 into Derby, Nottingham, Chesterfield and Sheffield, rather that East Midlands Parkway and Meadowhall.

 

[Snow1964]

Thought they had been talking about 4 tracking Sheffield to Dore for years. Finally got 3 tracks through Dore. I have no idea how they expected to get HS2 trains through Sheffield Midland. There was the campaign to get HS2 into Derby, Nottingham, Chesterfield and Sheffield, rather that East Midlands Parkway and Meadowhall.

The whole section used to be 4 tracks, a few bits have been sold off, and a couple of underbridges have lost spans, but could probably get a third track back, on former trackbed, at least in parts.

It is a common way of widening in Europe, add a track, make centre one reversible with crossovers at intervals, and use the third track as a sort of overtaking lane. Much cheaper than widening a formation to 4 tracks. It is actually so common in Europe there are probably off the shelf ECTS modules for signalling the 3 track with centre reversible arrangement.

 

[dosxuk]

What puzzles me is the comparison between the MML platforms (short, out on a limb, badly lit, draughty) with the ones for the services to East Kent. Travelling to Ashford recently I was amazed by the opulence of the platforms and the luxury of the trains (though I had arrived on a 360 from Wellingborough)

Presumably some planner predicted that there would be a massive demand for travel from London to Ashford, Canterbury Margate and Ramsgate - even though there was already another service - but thought that the demand from London to Leicester/Derby/Nottingham/Sheffield (let alone the smaller towns serve - most still larger than those in East Kent) one be tiny. So platforms for 12 car trains were built for the former and short ones for the latter.

The "short" platforms are designed to the standards that were in place at the time of design (and construction). Sufficient length for 10x 23m or 12x 20m carriages, with enough extra to allow for splitting and combining.

The IEP project decided to create it's own new 26m standard, in order to take advantage of any extra platform length at the main ECML/GWML termini. That decision meant they wouldn't fit 2x 5 car sets at the majority of London termini - basically anywhere that has been refurbished and provided with standard length platforms.

The 395's on the Kent services would fit into the MML platforms, with space to split and join.

 

[Merle Haggard]

The "short" platforms are designed to the standards that were in place at the time of design (and construction). Sufficient length for 10x 23m or 12x 20m carriages, with enough extra to allow for splitting and combining.

The IEP project decided to create it's own new 26m standard, in order to take advantage of any extra platform length at the main ECML/GWML termini. That decision meant they wouldn't fit 2x 5 car sets at the majority of London termini - basically anywhere that has been refurbished and provided with standard length platforms.

The 395's on the Kent services would fit into the MML platforms, with space to split and join.

Thanks for the explanation; I hadn't appreciated that the 12 car 395s so much shorter than 2 x 5 car standard 8xx.

 

[WesternLancer]

Thanks for the explanation; I hadn't appreciated that the 12 car 395s so much shorter than 2 x 5 car standard 8xx.

Yeah javelin carriages are noticeably shorter. Presumably to fit twisty bits of classic line they run on eg curved platforms at places like Strood I’ve caught one at.

 

[Killingworth]

Sounds to me like a justification to reinstate the 4 track: Sheffield to Dore

That track quadrupled in 1904, reduced to 2 by 1980. It would now be very difficult to make it even 3 due to removal of bridges across the River Sheaf and roads. That's a big reason why the financial bullet has been avoided.

The recently completed work at Dore Station Junction put in a new bridge across the Sheaf. 4 tracking would need that removing and 2 put in its place. Alongside that bridge one of the two 1904 spans is to be replaced next weekend, the other not being in as bad condition.

From there the track into Sheffield has been slewed with two active loops in place. Extending them should be possible with bi-directional working on a central track. But much has been developed on the lineside even when the track bed appears fairly clear. I could take stills from drivers views videos but this photo of mine illustrates one of many obstacles that will need to be resolved to add back 3rd, let alone a 4th track. The up line out of Sheffield at Abbey Lane, the site of former Beauchief and Abbey Dale station. It's relatively easy to resolve compared with getting across Little London Road and London Road where the bridges were removed for the lifted tracks. The wires are a small part of the potential total cost of adding more capacity.


Which is why I'm a very reluctant supporter of reopening old branch lines. We need to be restoring capacity on very busy reduced tracks like here, and station platforms too. A far higher priority.

 

[AndrewJM70]

Maybe it would have been easier and cheaper to just extend St Pancras platforms

Or just order a mix of 7 and 9 car trains. Plenty of room for passenger growth, fewer staff and fewer sets.

 

[WesternLancer]

Or just order a mix of 7 and 9 car trains. Plenty of room for passenger growth, fewer staff and fewer sets.

hard to argue with this from the average passengers point of view

 

[Yew]

It is worth remembering that the DfT's obsession with 26m vehicles (and hence 260m trains, due to certain mandarin wanting 5-car 'half-units') resulted in eye watering sums of money being spent to move signals at at least one London terminal to accommodate them.

I've probably said it before in these threads, but I don't feel it's possible to make a 26m unit that will run on our network with acceptable ride-quality.

 

[Trainman40083]

Or just order a mix of 7 and 9 car trains. Plenty of room for passenger growth, fewer staff and fewer sets.

Well, if you get a failure, you lose the whole train, whereas with two 5 cars coupled together (one day) you might only lose half the train. If you had 7 and 9 cars , you'd not be able to platform share between two services at St Pancras. You would probably need more siding/depot space that isn't available at Derby. . I think a shorter 7 car Class 222 fitted, but not sure how much more space. So you'd need a brand new depot somewhere else, maybe Brighton near Sheffield. But I concede, far more space for passengers and growth

 

[WesternLancer]

Well, if you get a failure, you lose the whole train, whereas with two 5 cars coupled together (one day) you might only lose half the train. If you had 7 and 9 cars , you'd not be able to platform share between two services at St Pancras. You would probably need more siding/depot space that isn't available at Derby. . I think a shorter 7 car Class 222 fitted, but not sure how much more space. So you'd need a brand new depot somewhere else, maybe Brighton near Sheffield. But I concede, far more space for passengers and growth

Of course all valid points - but actually not too hard to overcome in most cases and potentially cheaper to resolve than the long term legacy nuisance and costs created by the decisions that have been taken.

 

[Topological]

Well, if you get a failure, you lose the whole train, whereas with two 5 cars coupled together (one day) you might only lose half the train. If you had 7 and 9 cars , you'd not be able to platform share between two services at St Pancras. You would probably need more siding/depot space that isn't available at Derby. . I think a shorter 7 car Class 222 fitted, but not sure how much more space. So you'd need a brand new depot somewhere else, maybe Brighton near Sheffield. But I concede, far more space for passengers and growth

Agreed.

I am a strong advocate for trains that are capable of splitting. Some train is better than no train.

However, 2x5 26m would have been better as then a proven unit was available. So the answer to this thread is longer platforms at St Pancras.

 

[HST43257]

I would like to see crossovers added/moved between Luton and London. It’s crazy to me at Thameslink Bedford services can stop at West Hampstead and St Albans on the fasts. So rather than crossing the alignment between Harpenden and St Albans, it should be between St Albans and Radlett. And rather than crossing the alignment between West Hampstead and Kentish Town, it should be between Cricklewood and West Hampstead. The current situation is adding useless delay minutes to the TL core and the rest of the MML and it’s far too frequent in my experience.

It would also allow same platform connections for passengers from north of Luton going to Hendon etc, as an additional win

They’ve got it right on the ECML, a where the TL fasts don’t stop at all on the fasts but they instead get a high frequency opportunity to floor it to Stevenage at minimal disruption

 

[GilfachFargoed]

Agreed.

I am a strong advocate for trains that are capable of splitting. Some train is better than no train.

However, 2x5 26m would have been better as then a proven unit was available. So the answer to this thread is longer platforms at St Pancras.

In an ideal world St Pancras MML would have had longer platforms, but the cost of achieving that post the station rebuild would have be exorbitant - either requiring substantial internal alterations to allow the buffer stops to be pushed back c.30m, or at the very least expensive modifications to the signalling (as at Paddington), platforms and possibly track to provide additional length at the country end.

I too see a benefit in trains that can be split, although mainly to allow multiple destinations to be served by one path out of a major origin, or to allow train length to be reduced later in a train's journey on routes where passenger volumes decline as the journey progresses. However I am not a fan of 'pint sized' units with no end gangways, where customers / train managers / catering facilities etc. end up stuck in one part of an overall train for potentially long periods of time.

 

[Wyrleybart]

Agreed.

I am a strong advocate for trains that are capable of splitting. Some train is better than no train.

However, 2x5 26m would have been better as then a proven unit was available. So the answer to this thread is longer platforms at St Pancras.

If you have 33 x 5 car trains that is 66 cabs full of equipment but if you were to have 17 x 9 car trains that is 34 cabs to kit out - a fundamental decrease in costs of equipment, maintaining it and replacing it when it goes wrong.

It is swings and roundabouts

 

[Topological]

If you have 33 x 5 car trains that is 66 cabs full of equipment but if you were to have 17 x 9 car trains that is 34 cabs to kit out - a fundamental decrease in costs of equipment, maintaining it and replacing it when it goes wrong.

It is swings and roundabouts

There are many arguments.

33 - 5 cabs going wrong is 28 trains available
17 - 5 cabs going wrong is 12 trains available

Whilst the cost of purchasing driving vehicles is much higher than intermediate vehicles, the maths still supports the 2x5 approach from what I can see.

I am sure the people catching 1C17 from Sheffield most mornings wish it was run as 2x810 so that there was a chance one of them would turn up.

 

[InTheEastMids]

33 - 5 cabs going wrong is 28 trains available
17 - 5 cabs going wrong is 12 trains available

You've effectively doubled the failure rate in the longer trains in this example. Why?

Back

Of the 16 EMR IC weekday departures from St Pancras between 4pm and 8pm, at least 12 will be 5-car trains with up to 4 as 10-car.
Inbound in the morning, the picture is similar - around 3-4 services will be 10-car and the rest 5-car.

That is not really screaming of a railway with a capacity problem that needs immediate infrastructure investment.

From where we are now, the fully deployed 810 fleet should allow a further 5-ish peak inbound/outbound services to be doubled up.
If that's not enough, the simple solution to more capacity could be to top up the 810 order whilst the production line exists.

If you had 7 and 9 cars , you'd not be able to platform share between two services at St Pancras.

I agree on not having microfleets that promote inefficient use of space. However, if more trains is the answer and Hitachi are out of play, then Option 2 is to top up LNER's 897 order. You'd base them at Neville Hill along with the other 897s and the only difference would be the vinyl wrap on the outside. This mitigates some of the microfleet issues, although you still end up with a mix of 5 and 10-car trains delivering the service.

 

[Palmerston]

My previous suggestion for the MML was to re-open the line from Market Harborough to Northampton. Today, this would make no sense as there is no capacity for additional routes to Euston. Once the majority of the fast services are using the HS2 line, however, there should be space on these lines, and more likely to be room at Euston, even if there will be fewer platforms for the traditional lines. I understand it's a not a straightforward reopening, but is feasible (albeit not cheap).

 

[GRALISTAIR]

Full electrification will help with acceleration and thus help a bit with capacity.

 

[Nottingham59]

However, 2x5 26m would have been better as then a proven unit was available. So the answer to this thread is longer platforms at St Pancras.

In an ideal world St Pancras MML would have had longer platforms, but the cost of achieving that post the station rebuild would have be exorbitant - either requiring substantial internal alterations to allow the buffer stops to be pushed back c.30m, or at the very least expensive modifications to the signalling (as at Paddington), platforms and possibly track to provide additional length at the country end.

Not really that much. Surprisingly, Platforms 1-4 at St Pancras already have a usable length of 260m according to the Sectional Appendix. All they need is enough extra length to accommodate splitting and joining.

HS2 Ltd specified 420m platforms to split and join 2x200m units. I'm not convinced that EMR would need that much spare. They could get up to 15m by moving the buffer stops south by around 4 paving slabs (3m), and converting the 12m long platform ramps to actual usable platform, with repeaters for the starting signals so the driver could see them.

See the discussion here:

Could EMR run 260m trains into St Pancras with SDO on the last carriage?

Platforms 1-4 at St Pancras have a working length of 240m, so in its franchise bid EMR commissioned the 810, a new version of the AT300 from Hitachi with 24m coaches just to fit 2x5-car trains. Speculating idly, I was wondering if instead they could have ordered standard 5-car 80x with 26m...

www.railforums.co.uk

 

[The exile]

You've effectively doubled the failure rate in the longer trains in this example. Why?

Doubling may not reflect reality, but it is going to differ. Tree hits front cab of twin unit = 1 cab from 4 as opposed 1 of 2, for example.

 

[Topological]

You've effectively doubled the failure rate in the longer trains in this example. Why?

Back

Of the 16 EMR IC weekday departures from St Pancras between 4pm and 8pm, at least 12 will be 5-car trains with up to 4 as 10-car.
Inbound in the morning, the picture is similar - around 3-4 services will be 10-car and the rest 5-car.

That is not really screaming of a railway with a capacity problem that needs immediate infrastructure investment.

From where we are now, the fully deployed 810 fleet should allow a further 5-ish peak inbound/outbound services to be doubled up.
If that's not enough, the simple solution to more capacity could be to top up the 810 order whilst the production line exists.


I agree on not having microfleets that promote inefficient use of space. However, if more trains is the answer and Hitachi are out of play, then Option 2 is to top up LNER's 897 order. You'd base them at Neville Hill along with the other 897s and the only difference would be the vinyl wrap on the outside. This mitigates some of the microfleet issues, although you still end up with a mix of 5 and 10-car trains delivering the service.

I assume that the probability of a cab failing is independent of the number of cabs. Therefore the cab failing in the 5-car is the same as the 9-car. The reason that you get a difference is because the 5-car previously had 4 cabs and 2 of them are still in a healthy unit. Hence, the train formed of 2x5 is only shortformed.

If your position is that the additional cabs mean that the likelihood of failure of 1 of the 4 is twice 1 of the 2, you still do not change the impact.

5 failures = 5 shortforms (if 2 x 5)
5 failures = 5 cancellations (if 1 x 9/10)

Maybe I should edit to say 10 failures = 10 shortforms, but still no cancellations.

== Doublepost prevention - post automatically merged: 27 Feb 2026 ==

Doubling may not reflect reality, but it is going to differ. Tree hits front cab of twin unit = 1 cab from 4 as opposed 1 of 2, for example.

The reality is probably 8 shortforms because, as you say, there are many reasons why the front cab is all that gets taken out.

 

[Peter Sarf]

My previous suggestion for the MML was to re-open the line from Market Harborough to Northampton. Today, this would make no sense as there is no capacity for additional routes to Euston. Once the majority of the fast services are using the HS2 line, however, there should be space on these lines, and more likely to be room at Euston, even if there will be fewer platforms for the traditional lines. I understand it's a not a straightforward reopening, but is feasible (albeit not cheap).

That does lead to the suggestion that eventually the MML will not need any solutions at the Southern end. Give it another 50 or so years however and we might be back to needing more than just Euston.

Options at StPancras would seem to have been :-
1) Order a fleet of 9 x 26m units. + Standard units. Forget about splitting/joining and just keep operations simple, one day demand will mean all trains are full length anyway - it is not a complex network like South of the Thames.
2) Extend platforms to take 10 x 26m standard Hitachi trains whether they be 5car or 10car. + Standard units.
3) Order bespoke Hitachi units that will fit. +Cheapest option. But risky and non-standard - units will not be easy to move and interwork with other 80Xs elsewhere.

Also
I was at StPancras and noticed there is maybe room to widen the MML station by another two platforms. Achieved by building the train deck out over the pavement and the taxi lane, plenty of head room. There is also still plenty of road West of that. I had not looked properly at the country end but iirc it is quite open ?.

 

[JonathanH]

Options at StPancras would seem to have been :-
1) Order a fleet of 9 x 26m units. + Standard units. Forget about splitting/joining and just keep operations simple, one day demand will mean all trains are full length anyway - it is not a complex network like South of the Thames.

Can anyone place a cost on how much a full fleet of 9-car units would cost in terms of additional lease costs and redesign of the depot facilities? Is it genuinely affordable?

2) Extend platforms to take 10 x 26m standard Hitachi trains whether they be 5car or 10car. + Standard units.

Seems impossible given the footprint available.

3) Order bespoke Hitachi units that will fit. +Cheapest option. But risky and non-standard - units will not be easy to move and interwork with other 80Xs elsewhere.

Seems to have been the compromise reached.

I was at StPancras and noticed there is maybe room to widen the MML station by another two platforms. Achieved by building the train deck out over the pavement and the taxi lane, plenty of head room. There is also still plenty of road West of that. I had not looked properly at the country end but iirc it is quite open ?.

St Pancras Old Church and its graveyard seems to prevent that. The approach lines merge very shortly after the platform ends.

 

[amahy]

Well, if you get a failure, you lose the whole train, whereas with two 5 cars coupled together (one day) you might only lose half the train. If you had 7 and 9 cars , you'd not be able to platform share between two services at St Pancras. You would probably need more siding/depot space that isn't available at Derby. . I think a shorter 7 car Class 222 fitted, but not sure how much more space. So you'd need a brand new depot somewhere else, maybe Brighton near Sheffield. But I concede, far more space for passengers and growth

Sigh… Why does everyone in this forum seem to think that long, fixed formation train sets creates cancellations? It is simply not the case. Avanti almost never cancels a train due to the Pendolinos being fixed formation, and the same can be said for Thameslink and the 700s. In reality, there is far more disruption due to short formations and people being left behind, because the DfT refuses to let TOCs have enough trains to properly run their service. Examples of this include Northern, Southern, GWR, SWR - it’s everywhere!