Lumo overspeed incident at Peterborough (17/04/2022)

[paul1609]

As I understand it, it wouldn't have received any flashing yellows as the diverging route wasnt set as it was following the EMR service ahead which was then crossing over at 15mph to platforms 4/5/6/7. Once the EMR had cleared the junction, the route was then set for the Lumo service which was by then approaching P468 at red, subsequently changing to green with a route indication for the up slow/platform 1.

As before, happy to be corrected.

To me it does really ask the question as to the adequacy of the infrastructure on a railway that has been electrified for many years now. I'm not really familiar with the Peterborough area but why on earth is the crossover that the EMR train took limited to 15 mph? There appears to be plenty of space. I'd have though that 40 mph would be the minimum. Similarly the turnout from the down fast to P1 & 2 I'd have thought a minimum of 40 mph for trains stopping there, where a train is passing through on to the slow lines as in this case 60/75mph. The current situation where a train routed through platform 1 from the downfast is limited to 30 mph through the first turnout and 25 mph through the second seems quite bizarre to me.

 

[172007]

A signal with a theatre box can have a theatre box fail showing blank but the main aspect showing a proceed aspect. I have seen it a number of times at a location where both routes for that signal are 25mph and it does not matter which route is taken as it's a station with equal platform lengths unless position lights are used.

I immagine where a theatre Box is used and different speeds and or possibility of wrong routing could occur would be designed to fail safe I.e. no theatre Box illumination the signal stays red. I would imagine if the RAIB had found the theatre Box had failed to illuminate and given a proceed aspect in this case they would have mentioned it as something they were investigating.

Anyone know of any signals with theatre boxes that will fail and the main aspect still illuminate with different speeds of routes? It's interesting that the one I know even when the signaller knows the theatre box has failed just advises the driver that it won't illuminate then pulls the signal off and they don't treat it as a failed signal.

 

[Stampy]

Just to say, I was down Peterborough station on this day, I arrived AFTER the incident and caught a Thameslink train from P2 at 1124..

The Lumo train was half-way out of Platform 1 when I got in the FRONT of my train (I'd seen the back of the Lumo when I got on)

We pulled out onto the Up Fast, and the Lumo followed us out on the Up Slow.

 

[30907]

To me it does really ask the question as to the adequacy of the infrastructure on a railway that has been electrified for many years now. I'm not really familiar with the Peterborough area but why on earth is the crossover that the EMR train took limited to 15 mph? There appears to be plenty of space. I'd have though that 40 mph would be the minimum. Similarly the turnout from the down fast to P1 & 2 I'd have thought a minimum of 40 mph for trains stopping there, where a train is passing through on to the slow lines as in this case 60/75mph. The current situation where a train routed through platform 1 from the downfast is limited to 30 mph through the first turnout and 25 mph through the second seems quite bizarre to me.

Do you mean Up Fast?
I don't think 30/25 are unusual limits for turnouts into loop platforms. It would be very rare for any of the moves you mention to be made by a non-stopping passenger train (though this thread is of course about one such!).

 

[JonathanH]

So could it be that driver hadn’t been routed down that specific track for long time, and simply forgot (or thought it was different having mixed it up with another location) and thought crossover was nearer 75mph than 25mph, only realising at last moment that it was slower

The driver had potentially never been signalled into platform 1 from the up fast as there is no obvious reason for Lumo trains to be routed there.

 

[Jimini]

The driver had potentially never been signalled into platform 1 from the up fast as there is no obvious reason for Lumo trains to be routed there.

...but (s)he would have been route trained to follow the signals via platform 1 (or indeed 2) when routed that way, I'm guessing? "No obvious reason" doesn't sound like much of a defense, IMO.

 

[TheBigD]

To me it does really ask the question as to the adequacy of the infrastructure on a railway that has been electrified for many years now. I'm not really familiar with the Peterborough area but why on earth is the crossover that the EMR train took limited to 15 mph? There appears to be plenty of space. I'd have though that 40 mph would be the minimum. Similarly the turnout from the down fast to P1 & 2 I'd have thought a minimum of 40 mph for trains stopping there, where a train is passing through on to the slow lines as in this case 60/75mph. The current situation where a train routed through platform 1 from the downfast is limited to 30 mph through the first turnout and 25 mph through the second seems quite bizarre to me.

Peterborough was remodelled in the mid 1970s (1977 i think?). The infrastructure reflects the standards and traffic of the time.
Whilst a new platform face has been built on the up fast, and a new platform 6/7 built on the former flyash line, Peterborough track layout remains pretty much unchanged.

 

[ComUtoR]

...but (s)he would have been route trained to follow the signals via platform 1 (or indeed 2) when routed that way, I'm guessing? "No obvious reason" doesn't sound like much of a defense, IMO.

I sign a busy metro area. Which breaks down to around 9 independent route, a couple of hundred signals, 5 main signal boxs, with aroud 15 different panels. Around 70 (ish) stations that spiderweb between them. 3 different traction families all that can be worked in a single diagram.

Miles of track, a few level crossings, foot crossings, a handful of significant tunnels, more bridges that I can count, some major junctions, minor crossovers, and a metric ton of pointwork. I suppose that gradients are also probably worth a mention too.

Without writing them down, I would struggle to give an estimate of how many changes of speeds there are.

I have a couple of mainline routes that might need to be chucked in there but I wasn't sure if they were important enough to mention.

All in all, a plethora of information squeezed in my tiny little brain; which at 0400 in the morning is still telling me off for getting up so early.

Sometimes.. I forget stuff. Probably not the best defense but I can't help being human; I was born that way.

 

[Taunton]

I really am surprised that on a line with a 105mph limit that a train being diverted by a 25mph crossover from that is given a green signal rather than a single yellow, regardless of route indicator. 99% of trains taking that route would be stopping in the platform anyway. And I'm even more surprised that the initial RAIB report does not mention they will be looking at that.

 

[bramling]

I really am surprised that on a line with a 105mph limit that a train being diverted by a 25mph crossover from that is given a green signal rather than a single yellow, regardless of route indicator. 99% of trains taking that route would be stopping in the platform anyway. And I'm even more surprised that the initial RAIB report does not mention they will be looking at that.

Pretty standard practice. You’d find numerous examples all the way down the 4-track section of the ECML.

That said, given that this is the second serious near-miss involving the up slow at Peterborough, there certainly seems to be scope for a look at the setup here.

 

[niggill617@gma]

The

I’m not going to comment on the exact circumstances of the incident (those of you who know me and know of my connection to Lumo will know why) but nobody here knows how long before, if at all, the brake was put in emergency before the train hit the points or the de-acceleration effect afterwards.

For those of you that haven’t experienced an emergency brake application from a relatively high speed on an 800 series unit, it’s completely different to that experienced on an HST or modern EMU. If you are standing up or not sitting securely in your seat, you will be at risk. Luggage will move too, depending on size and how it is stowed.

That is all before you hit any points and potentially get displaced by the sideways movement and then you have those remaining de-acceleration effects to consider.

People should be very careful about throwing accusations of lying if they don’t know the full facts to disprove them. The reasons for LUMO saying what they have said have been stated above, by calmer posters. It’s the job of the RAIB to determine the full facts here.

Braking performance of 2 plus 8 HST is 1.1 m/s/s with about 3 to 5 seconds freewheel, class 800 is 1.2 m/s/s with around 2 seconds freewheel. The influence of inertia will be similar.

 

[Snow1964]

Pretty standard practice. You’d find numerous examples all the way down the 4-track section of the ECML.

That said, given that this is the second serious near-miss involving the up slow at Peterborough, there certainly seems to be scope for a look at the setup here.

Does seem to be very low speed crossover between a fast and slow line. Presumably even trains using the slow line platform have to go much slower due to it than if just braking for a station stop.

Restrictive aspects are normally used where curves or complicated layouts preclude installing medium speed crossovers, it’s not clear to me why a 25mph crossover was specified at this location.

 

[bramling]

Does seem to be very low speed crossover between a fast and slow line. Presumably even trains using the slow line platform have to go much slower due to it than if just braking for a station stop.

Restrictive aspects are normally used where curves or complicated layouts preclude installing medium speed crossovers, it’s not clear to me why a 25mph crossover was specified at this location.

Seems to be fairly typical at Peterborough, there’s fairly slow speed limits on other crossovers, for example from the down fast towards platforms 4 and 5 (used by most down services calling at Peterborough), and of course the points at the centre of the previous overspeed incident at the London end of the up slow.

 

[snowball]

Will ETCS help in a situation like this, or would that require a higher level than the level 2 that is coming soon on the south ECML?

 

[AlexNL]

That said, given that this is the second serious near-miss involving the up slow at Peterborough, there certainly seems to be scope for a look at the setup here.

The roll-out of ETCS along the southernmost part of the ECML is already underway, meaning that this particular safety gap will be closed in the not too distant future. ETCS will enforce a braking curve to 25 mph.

I do expect an overall recommendation to look at similar signalling setups throughout the network.

 

[Stampy]

Seems to be fairly typical at Peterborough, there’s fairly slow speed limits on other crossovers, for example from the down fast towards platforms 4 and 5 (used by most down services calling at Peterborough), and of course the points at the centre of the previous overspeed incident at the London end of the up slow.

Not 100% sure, but I think the turnout from the Down Fast towards Platform's 4&5 are 30mph..

The other turnout (If that's the one you are on about) from the Up Slow to the Up Fast at Fletton Junction is 25mph.

 

[dk1]

I really am surprised that on a line with a 105mph limit that a train being diverted by a 25mph crossover from that is given a green signal rather than a single yellow, regardless of route indicator. 99% of trains taking that route would be stopping in the platform anyway. And I'm even more surprised that the initial RAIB report does not mention they will be looking at that.

It has worked fine since resignalling in the mid-1970s.

 

[Efini92]

Can someone explain the sequence of signals that would (should) have been displayed before the cross-over, please? Especially, whether the turnout is signalled by feathers or a route indicator board?

I am struck by similarities with the derailment at Bletchley in 2012, which may or may not have parallels with this case: https://www.gov.uk/raib-reports/derailment-at-bletchley-junction-bletchley

I don’t think it’s a coincidence the MD of Lumo stepped down just before this happened.

 

[LCC106]

The

Braking performance of 2 plus 8 HST is 1.1 m/s/s with about 3 to 5 seconds freewheel, class 800 is 1.2 m/s/s with around 2 seconds freewheel. The influence of inertia will be similar.

Please could you explain what m/s/s means? Thanks.

 

[hexagon789]

Please could you explain what m/s/s means? Thanks.

Metres per second per second

 

[Davester50]

Meters per second per second.

 

[Signal_Box]

Can we remember there is a driver out there with a potentially career ending investigation over their head right now.

His, her, their whole life is at stake because of this mistake, maybe we can remember that for a moment…?

 

[357]

Can we remember there is a driver out there with a potentially career ending investigation over their head right now.

His, her, their whole life is at stake because of this mistake, maybe we can remember that for a moment…?

And the manager (s) who passed them as competent too will be feeling similar, waiting for the outcome of the investigation. Especially if training gets picked up as an issue.

 

[Nottingham59]

because of this mistake

We don't actually know if there has been a mistake yet, or on who's part.

 

[Signal_Box]

We don't actually know if there has been a mistake yet, or on who's part.

Either way there is someone’s life / future on the line.

 

[Snow1964]

The

Braking performance of 2 plus 8 HST is 1.1 m/s/s with about 3 to 5 seconds freewheel, class 800 is 1.2 m/s/s with around 2 seconds freewheel. The influence of inertia will be similar.

I have dug out my Two miles a minute book by OS Nock which handily has the braking graph. It must be remembered that at time of introduction of HSTs in 1976 there was plenty of tread braked passenger stock. The graph (W curve) shows speed vs distance to stop at different braking rates

Basically main lines were signalled with braking at 7% of gravity, HSTs could brake at 9%g so were able to stop from 125mph within same 100mph signal spacing. Since then higher braking rates have become possible (I think west coast pendolino can do 135mph on a 100mph signalled railway).

I suspect at 12.5%g the Lumo trains could operate at 140mph on a 100mph signal spacing without a 5th aspect, that is being able to stop within the double yellow distant signal distance. However actually clearly seeing lineside signals above 125mph means need in cab repeaters (which is separate debate to braking rates)

But of course for this incident, the speed passing a signal and speed when reaching a junction which can be few hundred metres later can be significantly different if accelerating or braking with high performance trains, much bigger difference than a HST could do.

I am not sure how far apart the signal and turnout are, but potentially a train could approach the signal at low speed, signal clear, and be doing 70+mph by the junction. The relevance is speed controlling the signal approach doesn’t guarantee train will remain slow after passing the signal.

 

[niggill617@gma]

I have dug out my Two miles a minute book by OS Nock which handily has the braking graph. It must be remembered that at time of introduction of HSTs in 1976 there was plenty of tread braked passenger stock. The graph (W curve) shows speed vs distance to stop at different braking rates

Basically main lines were signalled with braking at 7% of gravity, HSTs could brake at 9%g so were able to stop from 125mph within same 100mph signal spacing. Since then higher braking rates have become possible (I think west coast pendolino can do 135mph on a 100mph signalled railway).

I suspect at 12.5%g the Lumo trains could operate at 140mph on a 100mph signal spacing without a 5th aspect, that is being able to stop within the double yellow distant signal distance. However actually clearly seeing lineside signals above 125mph means need in cab repeaters.

But of course for this incident, the speed passing a signal and speed when reaching a junction which can be few hundred metres later can be significantly different if accelerating or braking with high performance trains, much bigger difference than a HST could do.

My source is from Hitachi data sheet and RAIB investigation graph for 2 plus 8 hst.

 

[Nottingham59]

the speed passing a signal and speed when reaching a junction which can be few hundred metres later can be significantly different if accelerating

As a theoretical question, if a class 803 were standing at signal P468 and it was given a green signal to continue on the up fast through platform 3, how fast would be going by the time it passed the turnouts to platforms 2 and 1? Would it be able to reach 75mph within that distance?

 

[GC class B1]

I have dug out my Two miles a minute book by OS Nock which handily has the braking graph. It must be remembered that at time of introduction of HSTs in 1976 there was plenty of tread braked passenger stock. The graph (W curve) shows speed vs distance to stop at different braking rates

Basically main lines were signalled with braking at 7% of gravity, HSTs could brake at 9%g so were able to stop from 125mph within same 100mph signal spacing. Since then higher braking rates have become possible (I think west coast pendolino can do 135mph on a 100mph signalled railway).

I suspect at 12.5%g the Lumo trains could operate at 140mph on a 100mph signal spacing without a 5th aspect, that is being able to stop within the double yellow distant signal distance. However actually clearly seeing lineside signals above 125mph means need in cab repeaters.

But of course for this incident, the speed passing a signal and speed when reaching a junction which can be few hundred metres later can be significantly different if accelerating or braking with high performance trains, much bigger difference than a HST could do.

The w curve is a historic curve derived from the measured stopping distance from various speeds. If I remember correctly it was derived from measured stopping distances of a Deltic and vacuum braked Mk1 coaches with worn brake rigging and cast iron brake blocks in the 1960s or early 1970s. The W curve has been used as a basis for signal spacing. The W curve is applicable to a Full Service brake application but the retardation rates are not even through the speed range. I think the w curve is equivalent to about 6%g deceleration at 90 - 100 MPH, but because of the characteristics of cast iron brake blocks the deceleration rate at 40 MPH is higher, about 9%g.
For information deceleration (and acceleration) are quoted in either metres/sec/sec or %g. As g = 9.81 10% g is approximately 1 m/s/s.

 

[Taunton]

It has worked fine since resignalling in the mid-1970s.

Well, if we are going back in history, the comparable crossovers approaching Taunton in semaphore days were locked with the distants so they could not give a Clear, equivalent to an mas green, when set to cross over to the Relief platforms. Someone local can tell whether the current mas there does the same.

I honestly thought that TPWS would catch this one.