4 car train

[aslefshrugged]

To the best of my knowledge the gapping problem was coming out of and going into Hainault depot. Coded Manual at 20kph maximum

[d7666]

Apologies, is red text verboten hier ? if so, sorry, I didn't know.

[rincew1nd]

Oct 20, 2021 13:24:03 GMT d7666 said:

Apologies, is red text verboten hier ? if so, sorry, I didn't know.

Generally yes, as explained in this thread staff use coloured ink when making interventions in their capacity as a member of forum staff; you'll notice my minor tweak to your post before has an explanatory note attached in red, because I as a forum admin made this edit - whereas my general posts are just written in black.

[Tom]

Oct 19, 2021 21:30:52 GMT d7666 said:

It is a lot more complex that that. It is an ATO railway. Your driver responding to a lineside warning board is not how the Central Line works. To ensure enough momentum through a gap requires trackside signalling changes to ensure the "go-codes" enable the correct level of motoring AND at the same time maintain train protection; the on train run data firmware will need updating too. Coded manual driving won't fix this either, CM gets the same go-codes as ATO.

Not sure about on-train data but ATO route and run data (which is housed in Signalling Equipment Rooms) would need to be modified to alter stopping positions if particular signals or Block Marker Boards present a particular gapping issue for a short train. There are only about four people in the organisation with the skills to alter it (and most of us are pretty rusty!), it also wouldn't be possible to do a change just for a four car train.

[d7666]

Oct 20, 2021 14:43:03 GMT Tom said:

Oct 19, 2021 21:30:52 GMT d7666 said:

It is a lot more complex that that. It is an ATO railway. Your driver responding to a lineside warning board is not how the Central Line works. To ensure enough momentum through a gap requires trackside signalling changes to ensure the "go-codes" enable the correct level of motoring AND at the same time maintain train protection; the on train run data firmware will need updating too. Coded manual driving won't fix this either, CM gets the same go-codes as ATO.

Not sure about on-train data but ATO route and run data (which is housed in Signalling Equipment Rooms) would need to be modified to alter stopping positions if particular signals or Block Marker Boards present a particular gapping issue for a short train. There are only about four people in the organisation with the skills to alter it (and most of us are pretty rusty!), it also wouldn't be possible to do a change just for a four car train.

I was going on what one very rolling stock orientated ATC person stated (almost certainly you know who) when in my latter days in Wood Lane a gapping issue also around Hainault arose. Pretty sure that was sorted out by the on train firmware only, but it was not the same circumstances, and a few years back too, it is mostly outside my domain, and I probably got it wrong, as by this stage I was a spectator. ((Interestingly the suggestion of a lineside warning board had earlier come up with that one as well.))

I wondered about 4car v. 8car too, but decided since I'm not involved with Central anyway now, and certainly was never in that depth, not to concern my grey matter with it, and hoped someone else would mention it, which you did

[Deleted]

Oct 20, 2021 14:43:03 GMT Tom said:

Oct 19, 2021 21:30:52 GMT d7666 said:

It is a lot more complex that that. It is an ATO railway. Your driver responding to a lineside warning board is not how the Central Line works. To ensure enough momentum through a gap requires trackside signalling changes to ensure the "go-codes" enable the correct level of motoring AND at the same time maintain train protection; the on train run data firmware will need updating too. Coded manual driving won't fix this either, CM gets the same go-codes as ATO.

Not sure about on-train data but ATO route and run data (which is housed in Signalling Equipment Rooms) would need to be modified to alter stopping positions if particular signals or Block Marker Boards present a particular gapping issue for a short train. There are only about four people in the organisation with the skills to alter it (and most of us are pretty rusty!), it also wouldn't be possible to do a change just for a four car train.

How exactly is the ATO code changed - what happens with closed stations? Do drivers go in Coded Manual for closed stations?

[rincew1nd]

When a train is at a station it receives a packet of data about the run to the next station. This includes a provision to skip the next station. However there is only provision to skip one station.

[d7666]

Oct 20, 2021 20:06:41 GMT rincew1nd said:

When a train is at a station it receives a packet of data about the run to the next station. This includes a provision to skip the next station. However there is only provision to skip one station.

IIRC [too many years since I did this] each station sends the run to the next two consecutive stations. If the stations are A B C D E, I think it goes something like
A will send the AB and BC runs,
B will send BC and CD,
C CD and DE;
so each run gets sent twice this is for resilience to allow for a failed transmission at one station,
and,
it allows for station skip, because, obviously, if you skip, you don't stop, so you don't get the run data;
if B is skipped it can't send BC and CD,
but A will have already sent BC,
and C will give it CD.

Or something along those lines. I'm trawling grey cells from 15 years now so the above is possibly flawed. If what I described is not the mechanism, then there is something else that ensure it gets sent twice, part of which is to work around a skipped station. TBH I know more about the Local Site Computers and the central processor kit at Wood Lane than all this trackside bobbins. It probably shows ;o)

[Tom]

The data allows for two stations to be non-stopped, this goes back to the days when there were stations closed on Sundays like Chancery Lane.

So, for example, at Hanger Lane, a train gets enough data to be able to run to White City.

There is, however, a catch if one of the stations being skipped is the last one before leaving the tunnel or the first after entering one - as that's where the train gets it's new brake rate from (the brake rates being different underground to allow for the better (more constant) adhesion conditions. When Shepherd's Bush closed for refurbishment we had to do a mod to the data at Holland Park to enable the surface brake rate to start there and make sure a train didn't arrive in the wet at White City with a tunnel brake rate.

[aslefshrugged]

Oct 20, 2021 21:55:26 GMT Tom said:

There is, however, a catch if one of the stations being skipped is the last one before leaving the tunnel or the first after entering one - as that's where the train gets it's new brake rate from (the brake rates being different underground to allow for the better (more constant) adhesion conditions. When Shepherd's Bush closed for refurbishment we had to do a mod to the data at Holland Park to enable the surface brake rate to start there and make sure a train didn't arrive in the wet at White City with a tunnel brake rate.

ATO won't do "station skip" at Mile End on the eastbound, every time it shuts we have to do Bethnal Green to Stratford in Coded Manual.

Catches out a lot of the new recruits, for some reason that particular quirk of the Central Line is not covered in the training.  Last time Mile End closed (flooding?) I had to tell Wood Lane as they didn't know about it either and we were repeatedly being held while drivers went through the procedures at Bethnal Green.

[Tom]

Yes - that's the same as at Holland Park WB with Shepherd's Bush skipped.

There's no data uploaded because the train needs to stop at the next station to get the new brake rate data, similarly both platforms at Redbridge (for brake rate changes at Wanstead and Gants Hill respectively).

[d7666]

I hope the readers - especially a certain joiner - are taking note of how complexicated ATO is, and why all the systems and software and hardware needs exhaustive testing. Just look at everything that has come up: 4car train, gaps, station skip, closed stations, brake rates, ATO driving, CM driving, adhesion, etc etc. And this is just one small aspect of it.

Which is the underlying reason why the 4LM SMA5 dates given in another thread are what they are. OK, 4LM is off topic for this thread here now, we are concerned with the Central line, but this is what the Central's upgrade project had to go through 20+ years ago. More so now today as there is far more software involved these days.

Much as plonking a line side warning board for 4car trains at gaps may seem obvious to the layman, I'm afraid that's not the way 21C integrated systems work. I'm sure there is a huge philosophical debate here about complexity versus simplicity, about persons in cabins with mechanical levers and flags and oil lit lamps versus central vehicle control computers, never mind people in train cabs or not.

OK, end of rant, it's just a spin off from me linking what's been said in this thread with the 4LM one.

[Colin D]

Speaking of trains getting gapped, I’ve been reading this months issue of Underground News and seems seven and eight car trains are susceptible to gapping as well. In August a couple of Bakerloo’s stalled at Harrow & Wealdstone and one at West Ruislip did the same thing. Doesn’t it also occur at Epping on occasion as well?

[spsmiler]

Hopefully as technology advances the signalling control systems will become more flexible and able to cope with more possibilities, such as different train lengths, brake rate data, without requiring so many limitations. For instance, brake rate data changes could be made available to the train when it is in motion (along with the signal location where to make the change) or even be stored on the train at all times.

I'm not advocating for computers which are sentient (we really don't want a real-life HAL as per 2001 Space Odyssey) but right now it does seem that the computers are taking over in ways that are more restrictive on train operations than clockwork technology was decades ago - when very frequent off-peak short length trains were 'the norm' and the Morden-Edgware Northern line even ran some extra-long rush hour trains on part of the line!

[jimbo]

Oct 21, 2021 20:10:09 GMT Colin D said:

Speaking of trains getting gapped, I’ve been reading this months issue of Underground News and seems seven and eight car trains are susceptible to gapping as well. In August a couple of Bakerloo’s stalled at Harrow & Wealdstone and one at West Ruislip did the same thing. Doesn’t it also occur at Epping on occasion as well?

Of course, full length trains can cover most gaps but there are some long ones, usually placed where a train is unlikely to come to a stand. Otherwise current rails can become out of gauge. Sometimes a high rail can knock pickup shoes from a train, causing them to gap somewhere they would normally be OK. Other times a current rail may be a little low, so that a pickup shoe unexpectedly does not contact it, causing a gapped train although with momentum a train would normally cross such a spot OK.

[Tom]

Oct 21, 2021 20:10:09 GMT Colin D said:

Speaking of trains getting gapped, I’ve been reading this months issue of Underground News and seems seven and eight car trains are susceptible to gapping as well. In August a couple of Bakerloo’s stalled at Harrow & Wealdstone and one at West Ruislip did the same thing. Doesn’t it also occur at Epping on occasion as well?

Yes, it's not unheard of for trains to become gapped at both West Ruislip and Epping.

[superteacher]

We have drifted significantly away from the thread topic. Whilst gapping may be relevant, the vagaries of Central line ATO should be discussed elsewhere.

[spsmiler]

One way to solve the (short) train gapping issue is to have trains with traction battery (or capacitor) energy storage.

Even if the onboard energy storage only provides enough power for a short distance it could also help a train reach the next station if there is a power supply failure* and reduce significant arcing where trains which are motoring have just a few cars gapped at a time because they need to traverse some pointwork and hence there is a gap in the power rails (eg: Ealing Common eastbound).

In this way the main topic for this thread could be resolved and indeed by eliminating the gapping issue it could help reduce track costs at many locations where is pointwork.

*Much nicer than being stuck in a tunnel - from personal experience!

Onboard energy storage with regenerative braking can also help reduce total energy use. I favour capacitors because they will last the life of the train and are less prone to fires than batteries.

[Tom]

As it happens, I was at a (virtual) talk the other day from a former TfL Professional Head of Vehicles where the speaker mentioned the idea of batteries providing limited traction power was being developed for new rolling stock. In the same talk he also talked about the issues of having to do 'one around' replacement of capacitors on a whole fleet because they were no longer adequately performing.

[jimbo]

The new Picc trains are promised to have such a feature to reduce the number of times trains are stuck between stations because of power failure. That could simplify depot track layouts once the current fleet is withdrawn. But would it be worth modifying layouts that are already there to save maintenance?

[aslefshrugged]

Traction failures are extremely rare, mostly we're stuck in tunnels due to signals, faulty trains, passenger related incidents.

1992 stock has batteries for emergency lighting, radio, PA, etc. when the juice is off, obviously they're not powerful enough to move the train otherwise there wouldn't have been a gapping problem at Hainault and I doubt very much if it would be economical viable to retro-fit them with more powerful batteries.

I can only remember a handful of gapping incidents in the last 19 years.

[zbang]

Batteries to solve a gapping problem sounds like the solution in search of that problem. OTOH batteries to provide both starting surge power and to capture regenerative breaking might make a lot of sense, that they could also move a train a couple hundred meters would simply be a bonus.

Whether any of that matters to running 4-car trains is questionable- what problem would 4-car address? Granted that at times 4-cars might be quite enough for the passenger load and would reduce the mileage on each car, that's at the expense of making up different trains and the effects on signaling/planning/etc.

FWIW, San Francisco BART pretty much made the decision long ago that it's not worth the trouble to cut 8- and 10-car trains down for very-off-peak service. OTOH BART also isn't know for it's forward planning.

[jimbo]

The previous version of Crossrail planned to halve train lengths in the evenings in order to start overnight maintenance earlier. But then it was a different world back then!