TfW class 398 Stadler Citylink tram trains

[Cardiff123]

The delays are down to multiple reasons.

1) COVID - they lost nearly a year to that alone for various reasons including staff from Stadler not being able to travel to the UK.

2) Funding - the Welsh Government can't borrow money for the project so any spending has to come out of the yearly budget, so accelerating delivery would haven taken money away from other things like health or education

3) Rampant inflation - this has added significantly to the funding pressure in point 2.

4) Supply chain issues - according to TFW some of the electrification parts got stuck near the Suez Canal when the Ever Given got stuck and even though this was only a few weeks the whole supply chain would have been disrupted for a while afterwards.

5) Thefts - When they have been able to complete the electrification work cables have been stolen before they went live so the work had to be redone.

6) Testing issues - a standard part of bringing in a new set of trains, but certainly hasn't helped them catch up.

7) Training staff - again a standard part of the process, but if staff are sick / on holiday then it needs to be delayed as they have a railway to run first.

OK, so where's the honesty from TfW with the 398s? Instead, every few years, they've just re-announced that the 398s are starting testing, presumably hoping that it's forgotten that the year before, the exact same announcement was made.

Given that so much of TfW's final fleet plans depend on the 398s entering service, are there contingency plans in place if the 398s continue to be delayed and plagued by problems? The 150s won't be around for much longer unless either TfW or the ROSCO pays to put the 150s through overhauls.

 

[StripeyNick]

Anybody know roughly when 398030 arrived onto the scene please?

I saw that one on 1st September and from what I remember, it was quite new then so possibly sometime in August?

 

[The_Train]

I saw that one on 1st September and from what I remember, it was quite new then so possibly sometime in August?

That will do for me, thanks very much

 

[DelW]

Given that so much of TfW's final fleet plans depend on the 398s entering service, are there contingency plans in place if the 398s continue to be delayed and plagued by problems? The 150s won't be around for much longer unless either TfW or the ROSCO pays to put the 150s through overhauls.

That's something I've been wondering about as well. TfW has been relying heavily on the 150s (and possibly now resuscitated 153s as well) to cover shortfalls on a variety of routes, utilising their go-anywhere ability. (Okay, they struggle on the Marches at 75mph, but they can and have regularly done it). There's no other traction that can do that, but the 150s must surely be running down the clock on both hours and miles by now. Even if the funding is found (by whom?), major overhauls will take units out of the fleet for months, so if the stock cascade continues to be deferred, what will cover the shortfalls then?

 

[Cardiff123]

That's something I've been wondering about as well. TfW has been relying heavily on the 150s (and possibly now resuscitated 153s as well) to cover shortfalls on a variety of routes, utilising their go-anywhere ability. (Okay, they struggle on the Marches at 75mph, but they can and have regularly done it). There's no other traction that can do that, but the 150s must surely be running down the clock on both hours and miles by now. Even if the funding is found (by whom?), major overhauls will take units out of the fleet for months, so if the stock cascade continues to be deferred, what will cover the shortfalls then?

Presumably the only option TfW have if the 398s are indefinitely delayed, is bringing all of their 153s out of storage to cover mainline services in South Wales, and possibly Bridgend to Barry to Cardiff services, if all of the 150s are gone and the 398s are still sat at Taffs Well.

 

[Caaardiff]

Or keep the 158s and move them to non-cambrian routes.

 

[Peter Sarf]

Or keep the 158s and move them to non-cambrian routes.

Contract for 158s was extended by 2 years recently - iirc ?.
(or am I thinking of some other fleet)

 

[VR494]

Now aiming to get three car sets out only to replace 756s on Cardiff to Treherbert. Unable to run in multiples, they don't like it. So only 3 cars will run.

Issues found on 756 pantographs raising at the same locations and speeds, creating cuts in carbon strips, wires trapped in grooves and creating excessive wear and tear.

The 398 batteries are not good enough to the provide extended range they want.

Almost half the fleet needs a rewire following the thermal event inspections on the units.

More feeder stations are needed for the 4 an hour uplift, there is not enough power in the system, and the Bay line OLE is not complete; it went bang on switch on!

Treherbert depot facilities are not ready to stable them; more work is needed in the sidings.

Also looking likely guards will have to close the doors.

Some of the above is still subject to change. Not much luck on the recent test runs, more work needed.

 

[tfw756rider]

Contract for 158s was extended by 2 years recently - iirc ?.
(or am I thinking of some other fleet)

It was the 158s.

 

[Carlisle]

Also looking likely guards will have to close the doors.

Why, & does a modern team train system anywhere else in the world operate in that way ?

 

[Chris125]

Unable to run in multiples, they don't like it. So only 3 cars will run.

Of course, like any new train design these days. Unless it's essential from day one multiple-unit working will always be a lower priority.

 

[Peter Sarf]

Of course, like any new train design these days. Unless it's essential from day one multiple-unit working will always be a lower priority.

Yep. Same with 810s, testing in multi is still ongoing while some in service as singles. Nothing new.

I think also 777s were also not running in multi at first.

 

[chargesmith]

OK, so where's the honesty from TfW with the 398s? Instead, every few years, they've just re-announced that the 398s are starting testing, presumably hoping that it's forgotten that the year before, the exact same announcement was made.

Given that so much of TfW's final fleet plans depend on the 398s entering service, are there contingency plans in place if the 398s continue to be delayed and plagued by problems? The 150s won't be around for much longer unless either TfW or the ROSCO pays to put the 150s through overhauls.

It's a fair comment - I don't know why TFW have this approach, all of these issues are in the public domain but you never seem to see them in their public communications. Although I suspect most organisations are like this now in terms of their PR.

 

[VR494]

Why, & does a modern team train system anywhere else in the world operate in that way ?

Issues with the cameras on the units, and the time it's taking for the doors to close under the current process. Guard gives 1-2, driver gives 1-2 and closes the doors and drives off, won't meet the metro times they are after.

If the cameras fail or driver cannot see, the alternate method would be for gaurds to dispatch in normal way, which is not in the RMT agreement. The valleys DOO dream is not working out as planned.

If the driver is to open, and gaurd close, it's a much better solution. And would mean flirts would not need mods either to work on VOG.

 

[Caaardiff]

Issues with the cameras on the units, and the time it's taking for the doors to close under the current process. Guard gives 1-2, driver gives 1-2 and closes the doors and drives off, won't meet the metro times they are after.

If the cameras fail or driver cannot see, the alternate method would be for gaurds to dispatch in normal way, which is not in the RMT agreement. The valleys DOO dream is not working out as planned.

If the driver is to open, and gaurd close, it's a much better solution. And would mean flirts would not need mods either to work on VOG.

Wasn't the agreement from TFW that there would always be a guard on board even if not doing dispatch duties? So DOO was never a thing.

 

[TransportHub]

Wasn't the agreement from TFW that there would always be a guard on board even if not doing dispatch duties? So DOO was never a thing.

Now there is level boarding, the GA approach of always having a guard (with dispatch compentancy) on the train but leaving the doors to the driver, seems to be the best idea.

It gets the speedy dwell times; retains a back up if the cameras fail; keeps a second safety critical person on the train; and frees up the guard to do revenue duties without the doors being a distraction.

Where's the down sides....?

 

[VR494]

Wasn't the agreement from TFW that there would always be a guard on board even if not doing dispatch duties? So DOO was never a thing.

A gaurd was guaranteed on every train until the end of the franchise in 2028. Tram extentions will be DOO, james Price stated this in the Green Signals interview.

== Doublepost prevention - post automatically merged: 16 Apr 2026 ==

Now there is level boarding, the GA approach of always having a guard (with dispatch compentancy) on the train but leaving the doors to the driver, seems to be the best idea.

It gets the speedy dwell times; retains a back up if the cameras fail; keeps a second safety critical person on the train; and frees up the guard to do revenue duties without the doors being a distraction.

Where's the down sides....?

They don't even want this on the CVL, and recently removed the need for a gaurd on ECS moves on the TAM too. That's why they have installed all this tap on tap off equipment.

 

[TransportHub]

They don't even want this on the CVL, and recently removed the need for a gaurd on ECS moves on the TAM too. That's why they have installed all this tap on tap off equipment.

Any idea why? The company gets the speedy dwells they need and the union gets the guarantee of a second safety critical person. I can't see why people wouldn't want it?

And it's a proven method of work even using the same supplier's train

 

[AdamWW]

Issues with the cameras on the units, and the time it's taking for the doors to close under the current process. Guard gives 1-2, driver gives 1-2 and closes the doors and drives off, won't meet the metro times they are after.

Maybe I'm missing something but is that really slower than having the guard press the button to close the doors, the warning sounds, the doors close, the steps go in - then the entire cycle runs again with the locally opened door?

Issues found on 756 pantographs raising at the same locations and speeds, creating cuts in carbon strips, wires trapped in grooves and creating excessive wear and tear.

Seems unfortunate given that having pantographs raising on the fly is not one of the novel parts of this scheme.

 

[TransportHub]

Maybe I'm missing something but is that really slower than having the guard press the button to close the doors, the warning sounds, the doors close, the steps go in - then the entire cycle runs again with the locally opened door?

For one stop it's not very much, say 20seconds but multiply that by the number of stops and it gets big

Treherbert to Cardiff has 18 stops so at an extra 20seconds per stop that makes the journey 6mins longer end to end. That's were it matters

 

[wickham]

I saw that one on 1st September and from what I remember, it was quite new then so possibly sometime in August?

Agreed, 398030 appears to have arrived in early August (w/c 4th appanently).

 

[VR494]

Maybe I'm missing something but is that really slower than having the guard press the button to close the doors, the warning sounds, the doors close, the steps go in - then the entire cycle runs again with the locally opened door?



Seems unfortunate given that having pantographs raising on the fly is not one of the novel parts of this scheme.

I think it would have been okay if the train was stationary when the pantographs go up, but over time, the speed range of them going up is between 40-60mph, and if the train is constantly doing that same speed at the locations they go up, it's hitting the carbon in the same spots. Over mulitple trips a cut is then being created, wire gets trapped in the cut, ADD or wires tripped out. It's happening all the time. And the amount of strips that are being replaced at the moment in Barry depot is very excessive.

Also the proposed method is driver arrives, checks train is on correct PSM and releases the doors, CSDE will prevent a wrong side release. For dispatch guard gives 1-2 and steps on, driver gives 1-2, closes the doors, completes a PTI and signal check then moves away. Current policy would also not allow the train to depart if anyone is within range of the camera, not just behind the yellow line as per other TOC's.

Test runs were done, doing the above, and then gaurd closing up and giving the ready to start signal, it was much quicker. They want a max of 30 seconds per stop, and to also do the above on the 756 too.

Part of agreeing to the pilot, ASLEF wanted all ASDO beacons replaced on the TAM, they did not work properly in multiple locations, so AIW will need to change them before entry into service.

Also if the train stops short, the cameras do not come on until the doors are released, and even though the train has passed over the beacon all doors would still open.

 

[AdamWW]

For one stop it's not very much, say 20seconds but multiply that by the number of stops and it gets big

I get that but the suggestion seemed to be that guards directly closing the doors would be faster than the guard buzzing the driver to do it.

At the moment when stops are closely spaced they seem to be losing the advantage of better acceleration by the extended dwells due to the whole door/step cycle having to run four times at each stop rather than two.

== Doublepost prevention - post automatically merged: 16 Apr 2026 ==

Test runs were done, doing the above, and then gaurd closing up and giving the ready to start signal, it was much quicker. They want a max of 30 seconds per stop, and to also do the above on the 756 too.

If that's what the tests showed then it's what they showed though it seems somewhat surprising.

I think it would have been okay if the train was stationary when the pantographs go up, but over time, the speed range of them going up is between 40-60mph, and if the train is constantly doing that same speed at the locations they go up, it's hitting the carbon in the same spots. Over mulitple trips a cut is then being created, wire gets trapped in the cut, ADD or wires tripped out. It's happening all the time. And the amount of strips that are being replaced at the moment in Barry depot is very excessive.

Right but although it's new to the UK the principle of raising pantographs on the move certainly isn't novel.
It's entirely routine elsewhere in Europe at changes of voltage and I imagine at higher speeds too.

 

[VR494]

I get that but the suggestion seemed to be that guards directly closing the doors would be faster than the guard buzzing the driver to do it.

At the moment when stops are closely spaced they seem to be losing the advantage of better acceleration by the extended dwells due to the whole door/step cycle having to run four times at each stop rather than two.

== Doublepost prevention - post automatically merged: 16 Apr 2026 ==



If that's what the tests showed then it's what they showed though it seems somewhat surprising.



Right but although it's new to the UK the principle of raising pantographs on the move certainly isn't novel.
It's entirely routine elsewhere in Europe at changes of voltage and I imagine at higher speeds too.

That's correct, but you have to be whilling to spend on the more robust pantographs.

Having a tram-train is fine, but the current design of these, with what they have to work with is not going to work. I could spend hours, explaining in detail, and showing you evidence to back things up, but I wont be doing that.

Aslef have been rushed into signing off a package that is not complete, and it's all for a certain election in a few weeks time. I feel for the crews who are now going to have to be the guinea pigs on the front line, and when it goes wrong, I doubt there will be any support for them. Those above are making sure they are keeping out of it.

If you like the 398's, do come along and see them. Cardiff to Treherbert is the aim now in the next few weeks.

 

[AdamWW]

If you like the 398's, do come along and see them. Cardiff to Treherbert is the aim now in the next few weeks.

It will be interesting to see how that is spun.

Here are your actual new trians, not the ones that we made a big fuss about being your new trains before.

 

[chargesmith]

That's correct, but you have to be whilling to spend on the more robust pantographs.

Having a tram-train is fine, but the current design of these, with what they have to work with is not going to work. I could spend hours, explaining in detail, and showing you evidence to back things up, but I wont be doing that.

Aslef have been rushed into signing off a package that is not complete, and it's all for a certain election in a few weeks time. I feel for the crews who are now going to have to be the guinea pigs on the front line, and when it goes wrong, I doubt there will be any support for them. Those above are making sure they are keeping out of it.

If you like the 398's, do come along and see them. Cardiff to Treherbert is the aim now in the next few weeks.

I thought you said they didn't have the battery range to do anything other the Bay services?

 

[VR494]

I thought you said they didn't have the battery range to do anything other the Bay services?

They can do around no more than 6.0 miles in reality and that’s it. It’s basically an EMU, with an auxiliary power reserve. They need to have a retrofitted engine of some sort to make it work. The city line loop, even worse. It has to happen to save the depot. Or look at a remodel to get trains in there.

On the Treherbert they will be on the juice for the majority of the way, going in and out of Cardiff Central will be a risk, any time delays and it’s going to put them in critical conditions. It’s the only other option now the Bay is not available for this TT launch. If the wires trip or it goes into ADD, depending on the unit location you could loose your services on the TAM. It’s probably going to need a crane to lift it out.

The biggest problem besides the power is they can’t run for long in multiple. So you’re going to have go from a 4 car 756 on the line to a 3 car with only 106 seats and standing room only. It’s not going to go down well.

If they had the capital to provide the necessary infrastructure, I’m sure they would’ve been out by now in some form. There’s a work around for nearly every single thing on the units.

They cut back on the number of feeder stations to free up capital to finance Taffs Well depot, and because of this anything more than 12 flirts in combined mode causes the system to go into low power. Trains get warnings and systems show up faults. The 4 an hour is out of the question, 36 of these cannot run on the current system.

 

[norbitonflyer]

Right but although it's new to the UK the principle of raising pantographs on the move certainly isn't novel.
It's entirely routine elsewhere in Europe at changes of voltage and I imagine at higher speeds too.

Certainly not new to the UK. happens all the time at Wigston, Scrubs lane, and various points on the Great Western main Line

 

[tfw756rider]

A gaurd was guaranteed on every train until the end of the franchise in 2028.

I thought that the franchise (as was) was going to run for 15 years from 2018 to 2033?

 

[VR494]

I thought that the franchise (as was) was going to run for 15 years from 2018 to 2033?

KeolisAmey Wales yes, they were terminated in 2021, during Covid and it went to operator of last resort. Apologies for the above, but I’m sure the guards agreement was for 10 years. Someone maybe able to clarify that one. But that was for heavy rail traffic only.