Ely platform 3 too short for 2 trains? 1st June 2024

[Talyna]

I was at Ely around 11:50 on Saturday 1st June, waiting on platform 1 to go to Kings Lynn.

While waiting, I saw an EMR service approach platform 3 comprised of 2 x 2 car 170s. Taking up the rest of platform 3 was a GN train waiting to depart for London.

Despite the EMR train getting what looked uncomfortably close to the other train, it couldn't fit entirely on the platform including the last set of doors. I saw the guard disembark from the rear vehicle with a degree of haste, presumably to discuss with the driver / platform staff what do do.

My train then arrived a few mins later and I left, though I think passengers on the rear train were eventually let out through one open door.

As a more casual rail user, I've never seen anything quite like this happen before. I can see both the EMR and GN trains were both somewhat delayed in departing on RTT

I thought this would be interesting to share, and I'm curious to know what leads to something like this happening, and what the resolution is when it does?

 

[55018ballymoss]

Could be down to a number of factors, either the EMR or GN train was over formed (More coaches than planned), or both. It might have been accommodated in the wrong platform, or maybe the GN train has not stopped in the correct location to give enough space for the EMR train to be accommodated.

That is assuming that this was a timetabled move and not just an ad hoc one.

 

[greenline712]

That's interesting ... my 1022 Peterborough-Liverpool train on the same day was also 4x158s.
Normally, Norwich-Nottingham is 2x158, with an extra 2x158 being added/detatched at Nottingham.

A new timetable cannot be the reason, as this was the last day of the winter timetable.
I didn't think that EMR had enough units to be able to do this anyway.

Some occasional re-platforming must be necessary at Ely ... the GN terminators lay over at Ely for some time, and there's nowhere else for them to go!

 

[liamf656]

I'm not massively familiar with Ely but I believe that EMR/GN can only share platforms if the EMR service is 2/3 coaches, hence the issues caused

 

[louis97]

I was at Ely around 11:50 on Saturday 1st June, waiting on platform 1 to go to Kings Lynn.

While waiting, I saw an EMR service approach platform 3 comprised of 2 x 2 car 170s. Taking up the rest of platform 3 was a GN train waiting to depart for London.

Despite the EMR train getting what looked uncomfortably close to the other train, it couldn't fit entirely on the platform including the last set of doors. I saw the guard disembark from the rear vehicle with a degree of haste, presumably to discuss with the driver / platform staff what do do.

My train then arrived a few mins later and I left, though I think passengers on the rear train were eventually let out through one open door.

As a more casual rail user, I've never seen anything quite like this happen before. I can see both the EMR and GN trains were both somewhat delayed in departing on RTT

I thought this would be interesting to share, and I'm curious to know what leads to something like this happening, and what the resolution is when it does?

The stop markers are pretty much back to back for an 8 car 387 and a 2/3 DMU. There are some trains booked 4 car 158s, these are booked other platforms or not to share. This train I suspect was either booked a 4 car and an oversight on the platform or it was an overform on the day.

 

[Parallel]

I’ve been on a service at Castle Cary before which was formed of a 4 car 158 which was put into platform 3, and was too long for the platform. The guard released the platforms doors externally one at a time. Not sure if that’s standard protocol or if it varies between TOCs.

 

[InkyScrolls]

If a driver is signalled into a platform where they know they will not fit, they won't move and instead will contact the signaller for explanation. If they only discover after passing the protecting signal that the train won't fit, they must continue as far as they can, give the guard 2-2 (do not open doors) on the buzzer if applicable, and after coming to a stand contact the guard to clarify which doors can be opened. The guard would then open those doors manually, using the emergency releases if there is no other way of doing so, or by using ASDO.

 

[LAX54]

Appears to have been a planning error by EMR

 

[LowLevel]

The platform sharing at Peterborough (Joint line) and Ely (Liverpool/Norwich) is based on EMR running 2 car trains. You can fit 11 coaches in at Ely platform 3 and 14 coaches at Peterborough platform 1. Ad hoc overforms or short term planning changes (like yesterday to account for the heavier EMR traffic between Nottingham and Peterborough for the MML closure) can cause this not to work and if it isn't picked up at the signalbox/station level this is the result - by which I mean it should have been sorted by then by planning or control.

 

[LAX54]

The platform sharing at Peterborough (Joint line) and Ely (Liverpool/Norwich) is based on EMR running 2 car trains. You can fit 11 coaches in at Ely platform 3 and 14 coaches at Peterborough platform 1. Ad hoc overforms or short term planning changes (like yesterday to account for the heavier EMR traffic between Nottingham and Peterborough for the MML closure) can cause this not to work and if it isn't picked up at the signalbox/station level this is the result - by which I mean it should have been sorted by then by planning or control.

Signaller/Signalbox always seem to be the last people to find out !

 

[ac6000cw]

Basically all the platforms at Ely are long enough for 12 x 20m trains (there are 12 car stop markers at the north ends of them) so 8 x 20m (387 or 700) plus 4 x 23m won't fit. The EMR train in question was listed as 2 x 2-car 170 on RTT, so I assume had been planned to be 4 x 23m on the day.

Looking at RTT around the time the EMR train arrived at 11:48, platform 2 was empty between about 11:35 and 12:10, so it could have used that one instead (which is the opposite face of the platform 2+3 island, so ideal as an alternative). It was timetabled to arr. at 11:45 and dep. at 11:50 on platform 3.

Some occasional re-platforming must be necessary at Ely ... the GN terminators lay over at Ely for some time, and there's nowhere else for them to go!

It can and does happen if trains aren't running to time (including the considerable number of freight trains), but it helps a lot that the all the running lines are fully bi-directional between the south end of the station and Ely North Junction. Sometimes the issue is that e.g. a westbound/northbound freight is already occupying the long loop on the east side (which serves both directions) and an eastbound/southbound one needs to run through the station to 'unblock' the routes a bit further north. Sitting on the station and looking at Traksy etc. while watching the action can be quite interesting at times, in terms of how the signaller keeps things moving.

 

[LAX54]

Basically all the platforms at Ely are long enough for 12 x 20m trains (there are 12 car stop markers at the north ends of them) so 8 x 20m (387 or 700) plus 4 x 23m won't fit. The EMR train in question was listed as 2 x 2-car 170 on RTT, so I assume had been planned to be 4 x 23m on the day.

Looking at RTT around the time the EMR train arrived at 11:48, platform 2 was empty between about 11:35 and 12:10, so it could have used that one instead (which is the opposite face of the platform 2+3 island, so ideal as an alternative). It was timetabled to arr. at 11:45 and dep. at 11:50 on platform 3.

Of course if the Signaller is unaware that its not the normal formation, he will route it to the booked platform.

 

[ac6000cw]

Of course if the Signaller is unaware that its not the normal formation, he will route it to the booked platform.

Yes, I understand that - but isn't (pretty important) train length info available to signallers once the stock allocations are done (e.g. RTT gets it's info from industry systems, AFAIK) or is the railway still in the dark ages, IT wise, with a variety of not-very-interconnected data systems?

Maybe that train (and some others on the same day) should have been STP/VAR flagged or similar, just to highlight to the signallers that they are not the 'normal' trains for that day of the week?

 

[800301]

Yes, I understand that - but isn't (pretty important) train length info available to signallers once the stock allocations are done (e.g. RTT gets it's info from industry systems, AFAIK) or is the railway still in the dark ages, IT wise, with a variety of not-very-interconnected data systems?

Maybe that train (and some others on the same day) should have been STP/VAR flagged or similar, just to highlight to the signallers that they are not the 'normal' trains for that day of the week?

Not always, and even then there is the human element. Most of the time if there is a formation change that would result in this the signaller is notified by control but this doesn’t always happen, along with platform restrictions EG Paddington, you can’t couple an 800 in P4/5/6 but you can with a 387 but that doesn’t stop them trying to route you in there

 

[LAX54]

Not always, and even then there is the human element. Most of the time if there is a formation change that would result in this the signaller is notified by control but this doesn’t always happen, along with platform restrictions EG Paddington, you can’t couple an 800 in P4/5/6 but you can with a 387 but that doesn’t stop them trying to route you in there

Indeed, the TOC have to tell NR Control, if they don't, then of course the Signaller won't get to know, the Siggie does have access to TOPS, but he cannot check every train that is booked to platform share, you have to take it as trains are running as booked. (Times of disruption is a different story of course)

 

[Wilts Wanderer]

I’ve been on a service at Castle Cary before which was formed of a 4 car 158 which was put into platform 3, and was too long for the platform. The guard released the platforms doors externally one at a time. Not sure if that’s standard protocol or if it varies between TOCs.

It is permitted to plan overlength trains (longer than 3 cars) into Castle Cary platform 3 - the preference is to use platform 2 but where this is not possible, the ‘local door’ principle applies. If this is a planned instance it is noted on the traincrew schedule cards. In some cases the dwell is long enough that the guard can open multiple doors using the outside ’butterflies’ which makes it easier for bikes to board/alight for instance.

The return Weymouth Wizard HST was nominally planned to use platform 3 with SDO (releasing front 2-3 coaches only) but every time I travelled on it, the service was running late enough that it could access platform 2 and do a full train door-release. This does show how guards need to be on the ball, not only thinking about which side the platform is, but how much of the train will be platformed and where they need to be positioned to release the doors.

 

[ac6000cw]

Indeed, the TOC have to tell NR Control, if they don't, then of course the Signaller won't get to know,

This process feels a bit 1924 vintage rather than 2024...

 

[800301]

This process feels a bit 1924 vintage rather than 2024...

Welcome to railway, there are many more things that are archaic as well

 

[RGM654]

It is common enough in other places that one set of doors or even one or two coaches don't fit on a platform, so selective door opening is used, with appropriate PA announcements. Even if that instance at Ely was unexpected, why did it present a problem?

 

[LowLevel]

It is common enough in other places that one set of doors or even one or two coaches don't fit on a platform, so selective door opening is used, with appropriate PA announcements. Even if that instance at Ely was unexpected, why did it present a problem?

Selective door opening on class 170s is not designed for reversals as it works using the position of the drivers key in the desk which has to be removed to change ends. Amongst other things once the driver has changed ends it's not possible to re-release the doors on the train during dispatch if the need arises.

The train might also have been beyond the platform starting signal.

 

[Horizon22]

This process feels a bit 1924 vintage rather than 2024...

Lots of systems interacting with each other separately which still require some element of manually adding it in. As has been alluded to this appears to have been ad-hoc strengthening and then not all systems downstream were updated. You'd be surprised how disconnected some systems are and how a lot of the railway runs on a system originally (and significantly changed but the base remains similar) built in the 1960s!

Not always, and even then there is the human element. Most of the time if there is a formation change that would result in this the signaller is notified by control but this doesn’t always happen, along with platform restrictions EG Paddington, you can’t couple an 800 in P4/5/6 but you can with a 387 but that doesn’t stop them trying to route you in there

And even when the signaller is aware, things still can get mis-routed! This is where drivers can challenge routes, although that's a last resort and not ideal waiting for signal timeouts etc.

 

[Sunset route]

Indeed, the TOC have to tell NR Control, if they don't, then of course the Signaller won't get to know, the Siggie does have access to TOPS, but he cannot check every train that is booked to platform share, you have to take it as trains are running as booked. (Times of disruption is a different story of course)

At our ASC the signallers don’t have access to TOPS/TRUST only the SSMs do and they have to watch over 10 panels and they would meed to know that there is a stock alteration. Plus only one of our major stations has weekly stock working sheets for planned formations only. So not all of NR is equally well informed.

Yes, I understand that - but isn't (pretty important) train length info available to signallers once the stock allocations are done (e.g. RTT gets it's info from industry systems, AFAIK) or is the railway still in the dark ages, IT wise, with a variety of not-very-interconnected data systems?

Maybe that train (and some others on the same day) should have been STP/VAR flagged or similar, just to highlight to the signallers that they are not the 'normal' trains for that day of the week?

You will be surprised by how lacking in information some very large signalling centres have and length and type of stock is one of them.

 

[transportphoto]

Selective door opening on class 170s is not designed for reversals as it works using the position of the drivers key in the desk which has to be removed to change ends. Amongst other things once the driver has changed ends it's not possible to re-release the doors on the train during dispatch if the need arises.

Off topic, but I wonder how Anglia managed this with three car 170s at Sheringham back in the day. They must have experienced this in the same manner (albeit planned reversal!)

 

[LowLevel]

Off topic, but I wonder how Anglia managed this with three car 170s at Sheringham back in the day. They must have experienced this in the same manner (albeit planned reversal!)

It is possible but not ideal - either it was a situation they accepted and planned for, or they used the local door instead of the SDO. The 3 car 170s also originally had a different SDO system which was changed some years later to the standard one. @dk1 Can probably advise.

 

[Trainman40083]

Many years ago, in the days of train operator Central Trains, the train I was travelling on was a 3 car Class 170. On the approach to Derby , it was routed into the former bay platform 5, yet it was too long. Driver queried the move, and we went in platform 5.. Doors were normally opened onto the platform 6 side, but in this case opened onto the platform 4 side. Train blocked platform 6 as it was beyond the signal at the end of platform 5.

 

[dk1]

It is possible but not ideal - either it was a situation they accepted and planned for, or they used the local door instead of the SDO. The 3 car 170s also originally had a different SDO system which was changed some years later to the standard one. @dk1 Can probably advise.

Our 3-cars 201-208 made that possible as guards de-selected the rear coach on arrival at Sheringham then we changed ends and it remained OOU.

In the Ely scenario however the train would be ahead of the signal to return so that adds to the complications. Route shouldn't be locked so just a case of following signallers instructions.

 

[bahnause]

This process feels a bit 1924 vintage rather than 2024...

Building interfaces for data transmission is not difficult. As soon as the transmitted data is security-relevant, it becomes very difficult. Especially if a confirmation of correct transmission or acknowledgement of the information recieved is required. An old-fashioned process may then be the better solution.

 

[LAX54]

At our ASC the signallers don’t have access to TOPS/TRUST only the SSMs do and they have to watch over 10 panels and they would meed to know that there is a stock alteration. Plus only one of our major stations has weekly stock working sheets for planned formations only. So not all of NR is equally well informed.


You will be surprised by how lacking in information some very large signalling centres have and length and type of stock is one of them.

Where I worked, it took a bit of time and persuasion, but in the end all the panels ended up with TOPS and TRUST access, however this is used for WTT / Platforms / Regulating etc, The Signallers do use it for Freight lengths sometimes if they need to put a train 'away'
However they will only check whats on a passenger train if they suspect something, or there is something out of the ordinary

 

[The Puddock]

Yes, I understand that - but isn't (pretty important) train length info available to signallers once the stock allocations are done (e.g. RTT gets it's info from industry systems, AFAIK) or is the railway still in the dark ages, IT wise, with a variety of not-very-interconnected data systems?

Lengths of loco hauled trains are quite easy for signallers to look up on the signalbox computer but multiple unit trains are, surprisingly, very hard. To find out the length of a multiple unit train on the IT systems available to a signaller, you’d need to search for a train on Web Gemini (TRUST will tell you vehicle running numbers but nothing more than that) to find out what unit/units is/are on it then look up the details for each individual vehicle in the formation to find the length of each and add them all together. That all takes quite a bit of time and assumes you have a log in for Web Gemini in the first place. That’s why many signalboxes have a locally produced table or spreadsheet handy showing the lengths of the most common different formations. Similarly, there is no easily accessible information about platform lengths available on the various bits of software used by signallers, so again locally produced tables/spreadsheets/diagrams are used. Nothing about it is easy or simple.

A contributory factor is that signallers are given no formal training at all about the different trains and traction types. None. Nothing. You just have to work it out yourself.

 

[LAX54]

Lengths of loco hauled trains are quite easy for signallers to look up on the signalbox computer but multiple unit trains are, surprisingly, very hard. To find out the length of a multiple unit train on the IT systems available to a signaller, you’d need to search for a train on Web Gemini (TRUST will tell you vehicle running numbers but nothing more than that) to find out what unit/units is/are on it then look up the details for each individual vehicle in the formation to find the length of each and add them all together. That all takes quite a bit of time and assumes you have a log in for Web Gemini in the first place. That’s why many signalboxes have a locally produced table or spreadsheet handy showing the lengths of the most common different formations. Similarly, there is no easily accessible information about platform lengths available on the various bits of software used by signallers, so again locally produced tables/spreadsheets/diagrams are used. Nothing about it is easy or simple.

A contributory factor is that signallers are given no formal training at all about the different trains and traction types. None. Nothing. You just have to work it out yourself.

When I was at Colchester PSB, we had one GEMINI login, and that was for the SSM, the Signallers ( 2 x NX and 6 x MCS) have no access, and virtually impossible trying to get it.
I used to find it useful on a weekend when there was engineering work, and trains terminated short, the Panel simplifier and TRUST to a certain extent was of no use, so I printed off a 'line up' for locations to aid the Signaller, this gave them Next working / Unit Numbers etc, but they were not always easy to read. did in the end make up an easy read list of DMU/EMU lengths for the panels 1 x755, 2 x 755 (or 720's etc)
As 'The Puddock' says there is no formal training on units/traction types, or TOPS/TRUST, it's all passed on Signaller to Signaller during panel training, although we also had an 'Easy guide to...' whaich gave the basics, and was quite useful.
TOPS/TRUST goes back to the 1970's and has not changed that much in that time, but it is a rare case of if its not broken don't try and fix it !