Progress on Avanti West Coast's 805/807s Hitachi AT300 sets

[Railperf]

I could've sworn I was told it was, but fair enough. A lot of articles certainly state so in error if it isn't the case.

It was mentioned in a TV documentary that there was a risk that it could happen with APT - not Pendolino.

 

[meld3]

How is this an Avanti fault?

Surely DfT micro-manage all aspects of train design & procurement these days?

 

[Merle Haggard]

Also not the case. The pendolinos could, in theory, be on maximum tilt anywhere and in either direction and not hit any infrastructure or a train coming the other way. Thats the reason for their cross sectional profile.

I could do with enlightenment here!

I thought that 'tllt' was a slight misnomer in that the bodies revolve (along a longitudinal axis) - possibly around the centre of gravity. The rotation is arranged virtually - there's no actual pivot rod. OK, that produces the perception of a tilt to the passenger, but it's not like a cyclist leaning over around corners (unless they were very acrobatic!). Am I correct?

 

[Railperf]

How is this an Avanti fault?

Surely DfT micro-manage all aspects of train design & procurement these days?

I'm not aware of Avanti being blamed for anything. This was a simple request to Avanti/Hitachi to fit a speed supervision device to the 80x, similar to what is fitted to other Avanti rolling stock.
This hasn't been required of TPE for similar stock running north of Wigan on the WCML. But TPE drivers are not asked to remember two different speed limit profiles depending on the train they drive.

 

[Rail Quest]

I'm not aware of Avanti being blamed for anything. This was a simple request to Avanti/Hitachi to fit a speed supervision device to the 80x, similar to what is fitted to other Avanti rolling stock.

In all fairness, the original article linked regarding this issue above was the TSSA directly blaming Avanti for incompetence in not fitting the system in the first place. They'll also no doubt get blamed for all services that will be delayed by the Everos being restricted to 110mph (so no new speed limits) in the meantime whilst the equipment is fitted.

Whether its right that Avanti themselves are blamed for that probably depends on how on earth themselves and the ORR got into this position. Clearly a step was missed somewhere along the procurement process by any involved company that should have spotted this safety concern would be the flagged by the ORR.

 

[notadriver]

The temporary fix for this is to implement a 110mph limiter on the 805s and 807s. TASS doesn't supervise until beyond 110mph so this would be an equivalent level of safety to the pendolinos.

There’s already a driver adjustable speed limiter on the units.

 

[LNW-GW Joint]

TASS on-board physical equipment is probably available from the Voyager class 221 fleet now not working in tilt mode on the WCML.
But interfacing it to Hitachi's AT300 TMS doesn't sound a trivial task, especially as Alstom has declared that the product is "dead" in development terms.
A speed limiter to 110mph doesn't solve the problem of enabling AT300/HS2 stock to run at up 125mph on some sections of the WCML.

 

[poffle]

TASS on-board physical equipment is probably available from the Voyager class 221 fleet now not working in tilt mode on the WCML.
But interfacing it to Hitachi's AT300 TMS doesn't sound a trivial task, especially as Alstom has declared that the product is "dead" in development terms.
A speed limiter to 110mph doesn't solve the problem of enabling AT300/HS2 stock to run at up 125mph on some sections of the WCML.

There are 23 class 80x trains so I there wouldn't be enough TASS equipment in the legacy Avanti 221s even if it were practical to swap it over.

Thee would be major costs in modifying the software and hardware on trains which are in service to add functionality they were never designed to have.

Maybe some of the camera based automatic speed limit technology used in cars could be adapted to give a warning as a standalone system.

It maybe that Avanti will have to have two separate driver pools with a set of drivers who only drive 80x trains between Euston and Liverpool. ( I'm assuming that this issue doesn't arise Crewe - Holyhead.)

I can imagine that ORR will be feeling a bit of heat from multiple parties here in raising this issue at this late stage. Introduction of non tilting 80x trains was part of the Avanti WCML franchise agreement which was awarded in 2019.

We've already had the issue over power supply north of Crewe impacting on the second train per hour to Liverpool plan.

 

[LNW-GW Joint]

There are 23 class 80x trains so I there wouldn't be enough TASS equipment in the legacy Avanti 221s even if it were practical to swap it over.

There were 44 (?) class 221s originally, all fitted with tilt and TASS for use on both Virgin and XC routes.
Arriva XC deactivated the tilt on its 20-odd retained 221s, and presumably will do so on those coming from Avanti.
So the physical equipment for 44 trains/88 cabs must be around somewhere (Central Rivers?).

 

[Boodiggy]

Unlike other European networks , the UK authorities specify that any tilt enable rolling stock needs the TASS (Tilt Activation and Speed Supervision) system to be fitted. This system determines on what sections of track the tilt system activates and deactivates. In Italy - trains with tilt are simply authorised to run at higher speeds on certain bits of the network, there being no TASS type system there.

Class 397's and Class 802's which run on the WCML are not tilting trains and so the drivers have to observe a 110mph maximum speed limit for non-tilt trains. There is no speed limiters fitted. Simply drivrs have to obsrve the speed limits in force for those trains. But as we said before, TPE drivers have route learned the northern section of WCML at PS speeds and have been driving Class 350's before the 397's arrived. Even though the 397's arrived as 125mph capable, it appears TPE and Network Rail have not come to any agreement as to locations where they can run at higher speeds, so for now the speed limit remains at 110mph maximum - which TPE drivers are used to! I'm sure TPE drivers are reminded that speeding is a serious offence enforced by OTMR downloads.


So my understanding is that when a unit experiences a tilt fault, TASS tends to be Isolated too, which prevents any overspeed warnings being displayed.
Does anyone know what the procedure was in Virgin days when a Class 221 ventured across to the ECML. I'm guessing tilt and TASS having to be isolated to allow 125mph non-tilt running? And as soon as the unit was reused back on the WCML at EPS speeds it had to have tilt and TASS switched back on manually?

Tilt will be isolated. The Speed is still supervised if the unit is running non tilt.
The train needs to pick up two balises on the WCML. When running on other routes the 221s would just been driven to the line speed like all of the other trains.

When slowing on curves for red signals of ESRs the train will not tilt below 40mph

The issue is not so much ensuring the 125mph is not passed, as important as that is.
It is sites like Atherstone where on the Down TV Fast there is a potential roll over from a train taking the curve to fast. If memory serves me right it is 134mph.
Other sites with lower speed profiles be it PS MU or EPS do not have turnover speeds as low on the mainline. That is why there is a TSR at Atherstone that ironically only delays EPS trains…

 

[poffle]

I wonder if this may be drawing a line for the HS2 trains ?

Originally there were HS2 trains that would do most of their journey on the HS2 network and would then travel a bit over the conventional network at end.

As HS2 has evolved ( dissolved ?) it's now proposed that trains should go a little bit on HS2 and then do most of their journey on the existing WCML.

Maybe the safety people are trying to lay down a line that these trains will need to have a properly designed speed supervision system to operate on the WCML. If the new trains are Hitachi then it also be possible to retrofit some of that functionality to class 805/7s.

Very speculative thinking not having really seen what ORR have actually said to Avanti.

 

[Bald Rick]

I could do with enlightenment here!

I thought that 'tllt' was a slight misnomer in that the bodies revolve (along a longitudinal axis) - possibly around the centre of gravity. The rotation is arranged virtually - there's no actual pivot rod. OK, that produces the perception of a tilt to the passenger, but it's not like a cyclist leaning over around corners (unless they were very acrobatic!). Am I correct?

I genuinely have no idea what you mean, sorry. The Pendolinos very definitley physically tilt, and that is done theough actuators moving the coach body relative to the bogies.

In terms of gauging, the trains must stay within the structure gauge at all times, regardless of the state of the tilt system ie active, inactive, failed locked vertical, or failed ‘wrong way’. This is why the trains have a reduced cross sectional profile, with the coach bodies angled inwards towards the roof at the same degree to the vertical that the train can tilt.

A train at full tilt one way or the other will always be within the structure gauge, and therefore can not come into contact with fixed structures alongside the track. Similarly, a train with tilt failure the ‘wrong’ way around a tight curve will not come into contact with any other train on a adjacent track, even if it is also a Pendolino and tilting the correct way.

 

[Nicholas Lewis]

ORR is responsible for certifying new rolling stock albeit through companies using a third party certification body before these trains could be used on the network yet this issue never came up then? Perhaps ORR need to be added to the list of bodies no longer fit for purpose when the reorganisation of the railways is finally defined.

 

[800001]

ORR is responsible for certifying new rolling stock albeit through companies using a third party certification body before these trains could be used on the network yet this issue never came up then? Perhaps ORR need to be added to the list of bodies no longer fit for purpose when the reorganisation of the railways is finally defined.

Surely it is the train company responsible for speccing a train not ORR, the blame for this should lay with the company who ordered and specified the train?

 

[Rail Quest]

Surely it is the train company responsible for speccing a train not ORR, the blame for this should lay with the company who ordered and specified the train?

Personally, I think this is the likely source of the problem but it may not be quite that simple. Firstly, given the highly complex regulatory/safety-critical environment the railway operate in, I would assume Avanti get consultants in to inform them of the safety/regulatory requirements the train order will need to meet so perhaps there was a missed process there (if said consultants do exist). Secondly, from reading what I've seen about this, the ORR aren't actually enforcing an explicit regulation as such - it seems they're objecting to the current specification as it is technically a downgrade of safety when compared to previous units, rather than a violation of a written rule. If this is correct, then Avanti/consultants may have reasonable grounds to not foresee an issue.

 

[poffle]

Surely it is the train company responsible for speccing a train not ORR, the blame for this should lay with the company who ordered and specified the train?

New trains have to through an approval process by ORR.

Here is a paper I found describing the approval process for Crossrail trains.

The Role of Rolling Stock Approval and Assurance within Major Railway Programmes - Crossrail Learning Legacy

1 Introduction and Industry Context In the UK, rolling stock cannot be placed into passenger service without following robust, legally mandated processes to attain the appropriate level of approvals and authorisations. To attain an Approval to place into Service (APIS) the trains have to be...

learninglegacy.crossrail.co.uk

 

[800001]

New trains have to through an approval process by ORR.

Here is a paper I found describing the approval process for Crossrail trains.

The Role of Rolling Stock Approval and Assurance within Major Railway Programmes - Crossrail Learning Legacy

1 Introduction and Industry Context In the UK, rolling stock cannot be placed into passenger service without following robust, legally mandated processes to attain the appropriate level of approvals and authorisations. To attain an Approval to place into Service (APIS) the trains have to be...

learninglegacy.crossrail.co.uk

Yes they do, not disputing that . However, who ever was responsible for ordering the Avanti fleet should/would not what is required to operate on the wcml.

ORR is there to enforce and ensure that, that is the case?

 

[poffle]

Yes they do, not disputing that . However, who ever was responsible for ordering the Avanti fleet should/would not what is required to operate on the wcml.

ORR is there to enforce and ensure that, that is the case?

As the 805s are in passenger service then we assume the design must have been approved by ORR. I would doubt that Hitachi would manufacture a design that hadn't been approved. I wouldn't think that Rock Rail would pay Hitachi for trains that hadn't passed approval.

There are a lot of things we don't know because we've only got very sketchy reports of what is going on.

Are the issues that ORR now have to do with the design of the trains or with how Avanti propose to operate the trains ?

When did ORR raise concerns about the trains with Hitachi/Avanti ?

Given that the trains have already been manufactured and are in service I don't that making major modifications to the design is going to be a realistic option.

At this stage there are going to be lawyers crawling all over the paperwork of what was agreed at different stages in the design process.

It would seem there have gaps in communication between and or within the parties involved.

The issue over whether there was a need for a new speed supervision system for the class 80xs to operate at over 110 mph on the WCML should have been resolved years ago at the tender stage.

 

[Railperf]

There's clearly a case os double standards here as far as the rolling stock is concerned. TPE have not been required to fit speed supervision on its Class 802's or Class 397's which are 125mph capable trains being used on the WCML. So why should Avanti suddenly have to do this on their 80x series? And was it a requirement to do so
TASS only has to be fitted on a tilting train where the operator wants to run it at EPS speeds. Otherwise there is no need to fit it.
So when did the rules change? And why aren't TPE being required to fit speed supervision too? Why is network rail allowed to run the NMT without speed supervision on the WCML? And while we are at it, why isn't every train fitted with speed supervision?

 

[LNW-GW Joint]

Then there's the case of the 10-mile route through Penkridge which is signed 90/125MU/125EPS.
Avanti's 390s and 221s (with tilt) run at 125EPS.
XC's non-tilt 22x run at 125MU (as did HST's before them); 350s run at 110MU.
Some of the XC 22x will be former Virgin 221s with tilt isolated (the line is almost dead-straight).
The 90PS was the original line speed for all trains before the WCRM upgrade, though it took several years before electric trains could use higher speeds*.

So the XC non-tilt 221s are running at 125mph on a WCML route without TASS, a similar situation to the 80x proposals, using the MU speed profile.
While the sponsor for the 80x programme is Avanti, Network Rail as system operator is also deeply involved and appears to support the MU speed profile.

* there was a curious period before EPS was enabled and the catenary upgraded when HSTs/22x could run at 125MU, while electric trains (390/350) were still limited to 90PS.

 

[Merle Haggard]

I genuinely have no idea what you mean, sorry. The Pendolinos very definitley physically tilt, and that is done theough actuators moving the coach body relative to the bogies.

In terms of gauging, the trains must stay within the structure gauge at all times, regardless of the state of the tilt system ie active, inactive, failed locked vertical, or failed ‘wrong way’. This is why the trains have a reduced cross sectional profile, with the coach bodies angled inwards towards the roof at the same degree to the vertical that the train can tilt.

A train at full tilt one way or the other will always be within the structure gauge, and therefore can not come into contact with fixed structures alongside the track. Similarly, a train with tilt failure the ‘wrong’ way around a tight curve will not come into contact with any other train on a adjacent track, even if it is also a Pendolino and tilting the correct way.

Thanks for the detailed reply.

What I was getting at (and it's hard to describe in words so this might not be any clearer ) was whether (i) the body simply tilted about an axis at the level of the top of the under frame (such as it is, i.e. above to top of the bogies) which would move the centre of gravity to one side (when viewed longitudinally) or (ii) the body tilted (by clever systems) about a virtual point about half way up the body so that the centre of gravity remained at the same point viewed longitudinally.

If the objective was only passenger comfort then (ii) would achieve it, but not increase the speed at which a train could take a curve of particular radius without overturning (i) would also allow the train to travel faster around a curve before overturning becomes a risk - just as a cyclist leans into a curve.

The reason I'm puzzled is that I have seen on this forum comments along the lines of 'tilt is only for passenger comfort, not speed', made forcefully, implying the system is not as simple as (i) but something else and (ii) is just my own idea of a tilt system that would achieve passenger comfort without increasing turnover speed (retaining the position of the CoG unchanged being implicit). If it is indeed as straightforward as (i) then those experts seem to be misinformed. There's a few of them.

 

[Railperf]

Thanks for the detailed reply.

What I was getting at (and it's hard to describe in words so this might not be any clearer ) was whether (i) the body simply tilted about an axis at the level of the top of the under frame (such as it is, i.e. above to top of the bogies) which would move the centre of gravity to one side (when viewed longitudinally) or (ii) the body tilted (by clever systems) about a virtual point about half way up the body so that the centre of gravity remained at the same point viewed longitudinally.

If the objective was only passenger comfort then (ii) would achieve it, but not increase the speed at which a train could take a curve of particular radius without overturning (i) would also allow the train to travel faster around a curve before overturning becomes a risk - just as a cyclist leans into a curve.

The reason I'm puzzled is that I have seen on this forum comments along the lines of 'tilt is only for passenger comfort, not speed', made forcefully, implying the system is not as simple as (i) but something else and (ii) is just my own idea of a tilt system that would achieve passenger comfort without increasing turnover speed (retaining the position of the CoG unchanged being implicit). If it is indeed as straightforward as (i) then those experts seem to be misinformed. There's a few of them.

My understanding is that the Class 390 and 221 can safely negotiate curves at EPS speeds without overturning whether rhe body tilts or not.
But when the body does not tilt to counteract the higher G forces, then passengers trying to walk through the train would likely find it harder to walk without handrails, luggage will more likely fall out of overhead shelves and items on tables will slide off.
The centre of gravity appears to be very low as the train bogies ate quite heavy and most of the heavy components are hung under the body.

THis video demonstrates the body tilt movement while standing still.

 

[Bald Rick]

Thanks for the detailed reply.

What I was getting at (and it's hard to describe in words so this might not be any clearer ) was whether (i) the body simply tilted about an axis at the level of the top of the under frame (such as it is, i.e. above to top of the bogies) which would move the centre of gravity to one side (when viewed longitudinally) or (ii) the body tilted (by clever systems) about a virtual point about half way up the body so that the centre of gravity remained at the same point viewed longitudinally.

If the objective was only passenger comfort then (ii) would achieve it, but not increase the speed at which a train could take a curve of particular radius without overturning (i) would also allow the train to travel faster around a curve before overturning becomes a risk - just as a cyclist leans into a curve.

The reason I'm puzzled is that I have seen on this forum comments along the lines of 'tilt is only for passenger comfort, not speed', made forcefully, implying the system is not as simple as (i) but something else and (ii) is just my own idea of a tilt system that would achieve passenger comfort without increasing turnover speed (retaining the position of the CoG unchanged being implicit). If it is indeed as straightforward as (i) then those experts seem to be misinformed. There's a few of them.

My understanding is that the Class 390 and 221 can safely negotiate curves at EPS speeds without overturning whether rhe body tilts or not.
But when the body does not tilt to counteract the higher G forces, then passengers trying to walk through the train would likely find it harder to walk without handrails, luggage will more likely fall out of overhead shelves and items on tables will slide off.
The centre of gravity appears to be very low as the train bogies ate quite heavy and most of the heavy components are hung under the body.

THis video demonstrates the body tilt movement while standing still.

Ah, I see. I have never studied the physics of it so cant (spelling pun intended) answer.

What I can say is that the speed required to cause a risk of overturning is well above the EPS line speed even with the tilt off. I have posted a while ago on another thread how in the early days I was in a Pendolino where the tilt failed as we went round Weedon, resulting in a moment where we were tilted the wrong way round one of the curves. Literally everything went off the tables (I caught my laptop, IIRC) but we were in no danger of coming off the tracks.

 

[craigybagel]

If you dig deep enough, it brings into question the whole issue of speed differentials. Yesterday I drove over a particular bridge twice, knowing I was limited to 40. Today I'll be going over that bridge twice again, but signals permitting, I'll be doing 90! Can I be trusted to cope?

Similar situations exist all over the country. Where do we draw the line on what is and isn't acceptable without additional help for the driver?

 

[LNW-GW Joint]

If you dig deep enough, it brings into question the whole issue of speed differentials. Yesterday I drove over a particular bridge twice, knowing I was limited to 40. Today I'll be going over that bridge twice again, but signals permitting, I'll be doing 90! Can I be trusted to cope?
Similar situations exist all over the country. Where do we draw the line on what is and isn't acceptable without additional help for the driver?

I assume that with ETCS, each type of train could have its speed limit at particular locations built in to the route permission/movement authority.
Are there such examples on the Cambrian setup (I think there are some bridges with differential speed limits between class 158 and class 37)?

 

[craigybagel]

I assume that with ETCS, each type of train could have its speed limit at particular locations built in to the route permission/movement authority.
Are there such examples on the Cambrian setup (I think there are some bridges with differential speed limits between class 158 and class 37)?

I would assume that's the case, but I don't sign the Cambrian so can't comment definitively. I'd be somewhat alarmed though if the set up on the 97s didn't take into account any differentials.

 

[Railperf]

In the case of Avanti, their drivers have to currently remember 4 sets of speed differentials depending on the traction they drive on WCML fast lines, plus the slow line limits (PS only). So that is a lot of information to remember.

So the question for Avanti and the wider industry is how to help drivers manage all that information. I would have thought that a form of DAS would be helpful that can use location awareness to help remind a driver understand what the limits are for the diagram they are on. Ideally a form of French KVB is what would do the job, or TOC's need to have a different pool of drivers for different forms of traction. So a dedicated pool of drivers for 80x as an example, who only have to remember the limits for that train.
Back in the APT days there was a proposal to display the permissible speed in the cab and ideally a version of that could be the minimum that is needed. But it did require radio transponders trackside and a receiver unit on the train.

Meanwhile, Avanti claim there have been no overspeeding incidents on 80x to date, and they must be monitoring it closely to say that.

 

[norbitonflyer]

I have received the following from a source within Avanti (Internal email sent to staff) and below are the service's scheduled to be operated by an 805 starting wc 2nd June 2024.

Are there any further diagrams running yet?

 

[DaveyJones]

Are there any further diagrams running yet?

No, but there are starting to be 807s appearing Driver Training

 

[Bletchleyite]

In the case of Avanti, their drivers have to currently remember 4 sets of speed differentials depending on the traction they drive on WCML fast lines, plus the slow line limits (PS only). So that is a lot of information to remember.

So the question for Avanti and the wider industry is how to help drivers manage all that information. I would have thought that a form of DAS would be helpful that can use location awareness to help remind a driver understand what the limits are for the diagram they are on. Ideally a form of French KVB is what would do the job, or TOC's need to have a different pool of drivers for different forms of traction. So a dedicated pool of drivers for 80x as an example, who only have to remember the limits for that train.
Back in the APT days there was a proposal to display the permissible speed in the cab and ideally a version of that could be the minimum that is needed. But it did require radio transponders trackside and a receiver unit on the train.

Meanwhile, Avanti claim there have been no overspeeding incidents on 80x to date, and they must be monitoring it closely to say that.

Replied here: https://www.railforums.co.uk/thread...ed-differentials-of-80x-vs-pendolinos.272669/