1992 stock Refurbishment?

[100andthirty]

Part of the problem with the Central line is that the DC traction motors are very unreliable. Several suppliers offered to provide new traction motors and update the DC chopper as necessary - a much simpler piece of work. Bot no one could guarantee that replacement DC motors would be any better than the existing ones as no one really understood the root cause of the unreliability. So the DC chopper electronics had to go and a complete new AC traction system provided.

1996 tube stock is different. There's nothing wrong with the motors and it's perfectly possible to replace tired electronics like for like within the same cases or, upgrade to IGBT power modules. All this is within the OEM's capability and has been discussed from time to time. I don't know the current status.

 

[trebor79]

Part of the problem with the Central line is that the DC traction motors are very unreliable. Several suppliers offered to provide new traction motors and update the DC chopper as necessary - a much simpler piece of work. Bot no one could guarantee that replacement DC motors would be any better than the existing ones as no one really understood the root cause of the unreliability. So the DC chopper electronics had to go and a complete new AC traction system provided.

1996 tube stock is different. There's nothing wrong with the motors and it's perfectly possible to replace tired electronics like for like within the same cases or, upgrade to IGBT power modules. All this is within the OEM's capability and has been discussed from time to time. I don't know the current status.

Indeed. An AC motor is just an AC motor at the end of the day and modern power electronics can be programmed to give whatever characteristics are wanted.

 

[GFE]

Perhaps TFL are reminded of the complication/timescales for this type of project and the specialist resources required (similar with other asset types such as signalling)

 

[Recessio]

Part of the problem with the Central line is that the DC traction motors are very unreliable. Several suppliers offered to provide new traction motors and update the DC chopper as necessary - a much simpler piece of work. Bot no one could guarantee that replacement DC motors would be any better than the existing ones as no one really understood the root cause of the unreliability. So the DC chopper electronics had to go and a complete new AC traction system provided.

Is this the flashover problem, or more general failures?

 

[trebor79]

Is this the flashover problem, or more general failures?

Upthread someone has described arcing/pitting on the commutators which nobody has been able to solve. Described the issue being managed by regularly skimming the commentators.

 

[xtmw]

Is this the flashover problem, or more general failures?

Also worth noting that trains do get taken out of service for door problems and what not

But yes, the motors flashing over is [mainly] the thorn is everyones side at this moment in time

 

[Central]

When I worked at Hainault in the 70s & 80s motor flashovers were a regular thing on 62 TS,most times it was down to worn brushes.They never got to the stage they’ve got to with 92TS.

 

[Busman]

From a post by burkitt on the District Dave forum, sounds like things are ramping up:

"I was on the visit to the CLIP project at Acton organised by TechSoc last week. Testing of the first completed train is continuing. This is actually known as Train 3, with Train 1 having had only the replacement traction system, and Train 2 only the new passenger info systems, as trial runs. Train 3 is the first with everything fitted.
They were working on Train 6 during the visit, and aiming for an eventual 50 day turnaround time."


And a post from xmw:

"Further info for anyone interested
Train 1 & 2 were prototypes, so were stripped down and being built into the 'proper thing'
Train 3 is heading for the South Ealing test track (Currently stabled in Northfields depot)
Train 4 should be ready soon
No word on train 5 yet
Train 6 as you said, currently in the works"


So if they're aiming for a 50 day turnaround time for each train, and I presume there will be 2 that can be worked on at once, presumably we'll be seeing a new train every 25 days. That sounds fairly swift but on that trajectory it will take 7 years to get through all 85 trains. That's a long time.

So some of these 92 stock have got a long way to go yet until they get refurbished.

Have to laugh at the first bit here, "sounds like things are ramping up ..." followed by "testing of the first train is continuing, known as train 3".

Train 3 has been testing for about a year now hasn't it? For the rest, it's the same update from 2 months ago, the only difference being train 5 has now been mentioned and that comes with ... "no word" so effectively, no change.

I'm not one for acronyms to describe my thoughts but 'lol'.

 

[Dstock7080]

Paths will be available for CLIP Train 3 to re-enter passenger service on the Central from 9 September, running one late-morning and one afternoon trip from Ruislip depot to Hainault and return.
no guarantee the train will operate each day.

 

[GFE]

The central line is currently in bad away - still regularly got shortage of trains even with the reduced timetable and the Customers need this project to deliver ASAP.
Can't help but think that the project is taking very small steps/tiptoes forward.
Seems like its not got any particular Senior Manager pushing it (or got their performance bonus based on it ;-} )

 

[RacsoMoquette]

Am I right in thinking that the whole 1980s and 90s Chopper traction systems on trains have universally been notoriously unreliable? Untried and overcomplicated technology I guess? Because the ABB built RATP MF88 on Paris Metro line 7bis has been particularly troublesome over the years, due to its traction control systems (with other unrelated issues) A similar story to the 1992TS. Again we have the notorious MA200 cars in Rome which had to be temporarily withdrawn in the early 2000s when the cars were not even five years old! These are just two other noticeable examples of trains let down by their GTO traction systems. Though the 1992TS have worked extremely hard over the past 30 years and haven't been withdrawn for prolonged periods. (Chancery Lane incident was not prolonged) so their GTOs fared quite well. (apart from recent major reliability problems due to the Stock being neglected for possibly the last fifteen years) Maybe we would not find ourselves in the chaos the Central Line is in today!

 

[Wolfie]

Suits me, the 92 stock are my favourite tube trains. Were best before the 2010s "refurbishment" anyway. I grew up playing with the "Juniors" 92 stock toy tube train.

The windows are aesthetically appealing but practically disastrous. The 92s are like a mobile greenhouse.

 

[boiledbeans2]

Am I right in thinking that the whole 1980s and 90s Chopper traction systems on trains have universally been notoriously unreliable?

See my post #317

1992 stock Refurbishment?

It looks like they're going for a darker moquette judging from the Twitter pic but could be wrong. It appears to be a car without any seats in it.

www.railforums.co.uk

 

[Sad Sprinter]

Be fair to 1992 stock, the current service demand, and hence mileages, and knock on impact on fatigue, now is way higher than envisaged when they were specified.

Yes, ATO specified from start, and, high performance, and depending which source you choose to read or believe, 30 or 33 or 34 TPH - but that was all around one hour high peaks only, the rest of the day was barely 15 TPH. Now we have 3 hour full one hour peaks, with much much higher load factors, and 24TPH off peak all day every weekday, and night tube on top of that. All this has eaten into component life. 92TS fatigue design life - nominally 40 years - was exceeded in the late 2010s and a specially instrumented weighted test unit was run around to ascertain work needed.

Sure 92TS had its design shortcomings, but please do be fair to them, they have already done much more than they were ever designed to do.

Hear hear. Were always my favourite trains ever since they were the most modern on the network as a child. I don’t suppose you worked in the design or manufacture of the trains?

 

[Mikey C]

The windows are aesthetically appealing but practically disastrous. The 92s are like a mobile greenhouse.

Plus the seats are low and small. I'd hate to travel in from West Ruislip or Epping every day.

 

[Wolfie]

Plus the seats are low and small. I'd hate to travel in from West Ruislip or Epping every day.

I'd forgotten the damn seats. Spot on.

 

[Lockwood]

There are seats on the Central Line?!

I thought it was just a square foot or standing space squashed against a door or pole

 

[Nym]

There are seats on the Central Line?!

I thought it was just a square foot or standing space squashed against a door or pole

A whole square foot of standing space?

 

[starlight73]

Am I right in thinking that the whole 1980s and 90s Chopper traction systems on trains have universally been notoriously unreliable? Untried and overcomplicated technology I guess?

I don’t have experience with electronics apart from AA batteries … but one thing that comes to mind is that older trains were electrically much simpler, so their spare parts are larger and relatively easier to improvise? (Somebody described a class 313 as being like a ‘bicycle’ somewhere on a forum — but I’m sure that’s not the whole story.)

the 92TS has technology that’s old enough to fail, but at the same time, it’s modern enough to have complicated circuitry and micro-processors. This makes replacement much harder. (Whether that is specifically the traction control system, I don’t know.)

 

[Nym]

One of the problems is of course...

Try and have a university graduate understand the old RPA (Rotary Pneumatic Accelerators) equipment like a Notching Relay...

However, we're now in the days of where contemporary graduates don't understand anything but MOSFETs (Metal Oxide Semiconductor Field Effect Transistor), IGBTs (Insulated Gate Bipolar Transistor) and maybe SiC (Silicon Carbide) equipment, so the likes of a GTOs is just not understood any more. And this is even 5 to 10 years ago, nowadays, I doubt anyone would truly understand a GTO unless they worked in a spesific trade such as Marine Propulsion (even there, the likes of IEP (Integrated Electric Propulsion) ships now mean even this niche is no longer understanding them.

 

[Dstock7080]

In a further attempt to prolong the life of the existing trains, from 6 September 2024 Auto-coast was finally implemented from Wood Lane Control room in ATO mode “this means that the motors get a chance to rest rather than constantly propelling the train”.

 

[Blindtraveler]

Has this always been an available feature in ATO operation or has a software modification become possible? Just curious as if it's long-term being on offer you can't help wondering why they haven't made use of it. Not least because it's presumably better in terms of energy requirements etc. As well as probably being better for the motor or then again perhaps it will be more strain constantly having to shut off and then regain any lost momentum

 

[100andthirty]

In a further attempt to prolong the life of the existing trains, from 6 September 2024 Auto-coast was finally implemented from Wood Lane Control room in ATO mode “this means that the motors get a chance to rest rather than constantly propelling the train”.

Have they increased inter station run times to compensate for coasting?

 

[Dstock7080]

Has this always been an available feature in ATO operation or has a software modification become possible?

I don't have the answer for the Central.
TBTC and CBTC do have this feature.

Have they increased inter station run times to compensate for coasting?

The timetable hasn't been adjusted.

 

[D365]

However, we're now in the days of where contemporary graduates don't understand anything but MOSFETs (Metal Oxide Semiconductor Field Effect Transistor), IGBTs (Insulated Gate Bipolar Transistor) and maybe SiC (Silicon Carbide) equipment, so the likes of a GTOs is just not understood any more. And this is even 5 to 10 years ago, nowadays, I doubt anyone would truly understand a GTO unless they worked in a spesific trade such as Marine Propulsion (even there, the likes of IEP (Integrated Electric Propulsion) ships now mean even this niche is no longer understanding them.

During my Autumn 2018 semester I attended a series of lecture covering the rapid development of thyristor technology through the 80s, 90s, and 00s. Out of the 10 or so lectures, I was lucky to have [maybe] half a lecture covering GTOs, and as you would expect that was only the fundementals.

There's just as few people left that understand the GEC and Brush microprocessor systems.

 

[xtmw]

Has this always been an available feature in ATO operation or has a software modification become possible? Just curious as if it's long-term being on offer you can't help wondering why they haven't made use of it. Not least because it's presumably better in terms of energy requirements etc. As well as probably being better for the motor or then again perhaps it will be more strain constantly having to shut off and then regain any lost momentum

There has always the option to turn coasting on for the Central Line, it's been discussed for many years, although it was finally done a few days ago.

I believe it gets it's coasting instruction from the PAC, happy to be corrected.

 

[hwl]

Has this always been an available feature in ATO operation or has a software modification become possible? Just curious as if it's long-term being on offer you can't help wondering why they haven't made use of it. Not least because it's presumably better in terms of energy requirements etc. As well as probably being better for the motor or then again perhaps it will be more strain constantly having to shut off and then regain any lost momentum

The biggest problem with coasting (not just on Central line but some other lines e.g. Northern Line where it isn't enabled) with sophisticated speed profile setting signalling systems (inc. ETCS L3) is that you have to understand the Davis equation coefficients for the train very very well and then be able to integrate this behind the scenes in the decision logic.

 

[Silent]

We simply have to replace the 72 stock on the baker Lou first and then I suspect there will be three fleets that will be in competition for replacement or major upgrade in a short space of time, the 95 and 96 stocks which have had reliability problems of their own in recent times, as well as the 92. At a time of fairly obvious managed decline on the underground, the level of investment needed to keep it moving. Looks like it might be the death or certainly limited functionality of huge parts of it

What type of reliability issues does the 95 and 96 stock have.

I don’t know if it’s me just being used to using the 96 stock a lot but today I used an unfurnished 92 stock and though operational the motors sounded worn out to me. But I often don’t use Central Line and am not an expert on motors. I understand the washing machine analogy now as well. When I use the 96 stock it still sounds healthy to me but then again I’m not an expert.

 

[Central]

95 TS had problems with damaged shoe beams,96TS are getting fractures.

 

[boiledbeans2]

CLIP in the Evening Standard recently:

Central line unlikely to improve until December after delays to refurbished trains

Only one train has re-entered service after being modernised under the £500 million Central line improvement programme (Clip), though a second is due within weeks

www.standard.co.uk

Passengers on the Central line are unlikely to see improvements until the end of the year after delays to the arrival of refurbished trains.

Only one train has re-entered service after being modernised under the £500 million Central line improvement programme (Clip), though a second is due within weeks. A new timetable that would improve train frequencies — gaps of five minutes or more are commonplace — is anticipated by December.

However, there are no immediate plans to tackle waits of up to 20 minutes on the line’s Hainault “loop” in east London until more trains are refurbished.

Transport for London is a year into its five-year Clip programme, which aims to extend the life of the 30-year-old trains by fitting new motors, seating and lighting, and introducing on-board CCTV.

But the need to train engineers to carry out the 10-week rebuilds and to overcome “snagging and software issues” has meant that no refurbished trains have re-entered service since March. The revamped trains, which can be spotted by their new moquette seating, have to undergo weeks of testing.

TfL commissioner Andy Lord apologised to passengers for almost a year of delays and overcrowding that has resulted from the shortage of trains. Many trains have had to be taken out of service at short notice to replace burned-out motors, in addition to the five trains at a time undergoing renovation.

Mr Lord said: “I’m looking forward to an improved timetable before the end of the year, and then more refurbished trains coming into service during 2025.

“The good news is that the second train will be entering service shortly… it will take us a couple of years to do the whole fleet.”

Mayor Sadiq Khan was briefed by TfL on the Clip programme last week. He admitted that the current service on the line, which is used by 200 million passengers a year, was not acceptable.

TfL decided to refurbish the trains because it could not afford to buy a new fleet, but the plan was delayed by the pandemic. But one expert said: “Sadly, with the benefit of hindsight, if TfL had known then what they know now, they probably wouldn’t have embarked on this and gone straight for new trains.”

The biggest problem with coasting (not just on Central line but some other lines e.g. Northern Line where it isn't enabled) with sophisticated speed profile setting signalling systems (inc. ETCS L3) is that you have to understand the Davis equation coefficients for the train very very well and then be able to integrate this behind the scenes in the decision logic.

I'm not familiar with the Central Line trains/signalling, but I'd imagine the simplest way to do coasting is to use the train speed?
E.g. if the line speed was 85 km/h, coasting would activate at 85 km/h, and would deactivate at 80 km/h, where the train would re-accelerate back to 85 km/h and the cycle repeats.
No need for any complicated mathematical model.