Hammersmith Eastbound Departure Board

Change at Barons Court-cross platform

You've obviously tried this and it works!

From what? the phantom District Line train that didn't turn up at Hammersmith? The OP's point was that if he had known there was no imminent District Line train approaching Hammersmith he would have taken the Piccadilly to South Ken in the hope of picking up a Circle or ex-Wimbledon train there to continue eastwards. But because he trusted the automatic lying machine at Hammersmith, he is still at Hammersmith.

There are quite a few useless indicators all over the SSR, could they not be linked in with Trackernet or some proper describing system until resignalling is done?

LT/LU has always been at the forefront of new technology, however, it has also had an appalling record of introducing it network wide, thus new technology constantly has to interwork with 100+ year old technology and many of the technologies partially introduced or trialled in the meantime. In many ways the entire network has long been treated as an experiment with various trial installations at many sites, some standing the test of time, some then resulting in further installations whilst others have remained in service as one offs and many have been withdrawn.

Train description is one area where remnants of old technologies remain in service although various initiatives are very slowly replacing them. Traditionally the District used the drum system and the Picc the ribbon storage system, much of the drum system was replaced and ribbon storage had AFAIR all gone by the early 1980s. However, these are not the whole picture as various scanning systems and programme machines were also involved in the transmission of TD information and different types of combinator equipment to display the descriptions on different types of TD systems and signage from around 1960. The late 1970s brought the electronic TD which replaced ribbon storage and drum storage and the 1980s brought computers to replace traditional TD equipment at the east and west ends of the Picc. Similar computer technology was rolled out with the 1980s resignalling of the Met and Jube and at the end of that decade on the Bakerloo with its resignalling. The Central line was a complete departure from tradition with its resignalling in the early 1990s and the Northern sat on the technology that I first worked on in the 1970s until its recent upgrade to TBTC.

The introduction of dot matrix gave a 'modern' image to platform train description from the late 1980s onward but it was simply a front end for the existing technologies already in service, indeed DMI has also been through several generations. My understanding is that programme machines on the Picc were replaced relatively recently with computer electronics and that train following systems were introduced but AFAIK newer technology simply highlights much more visibly the shortcomings of the old technology that remains in service passing data to the new signs. I expect that until Picc and SSR are fully upgraded these shortcomings will remain.

There are many things in signalling and communications that should have been addressed network wide decades ago and kept up to date since but quite simply LUL is less a network than the corporate image appears to present and more of an association of separates. The lack of investment in the system by successive governments led very much to 'line projects' and 'special projects' where network integration and the use of compatible technologies would've saved fortunes in terms of procurement, maintenance and upgrade. Outsourcing of engineering, devolution and the PPP have all cost the taxpayer fortunes where proper investment would have paid real dividends in terms of standardisation of rolling stock and signalling components if the engineers that once ran the system had been given the funds to do so. The problem these days, and particularly for the last 25 years or so, is that accountants run the system rather than engineers so 'engineering' decisions are not necessarily the best that could be taken and of course the purse strings for capital projects are held firmly by the Treasury and Transport Ministers over decades have not been too clever at prising the necessary funds from its clutches without plenty of mileage in political argument for political gain. Thus like all publicly owned utilities short termism, shortsighted and politically motivated though it is and always has been, is the real reason that the Underground remains a long way from the fully integrated network that it could be and also why train descriptions at many sites are not meaningful to the average commuter and probably won't be for a few years yet!

I went for vaguely humorous but, apart from not being very helpful to anyone wanting to know what the thread is about, it's also a bit unfair considering the millions of people LU transport every day and the vast number of things they do have to organise and get right in order to do so.

The light boxes each had several bulbs, so even if one bulb failed, it could be read.

I did notice there used to be a problem at Earls Court: because the light box is double-sided, when the sun was very low it could shine right through the box and light up all the arrows at once!

On a related note upon boarding the District Line at Ealing Broadway today the light boxes were wrong, showing the first train out as on platform 7, where there was no train waiting and the next train was on platform 9. Ended up going forward to the driver to ask when he was due out and got a green signal as he told me everything was out of sync due to the delays earlier.

"This picture illustrates the Train Describer on platform 1 - that on platform 2 is identical. The ‘black box’ visible between the illumination lamps illuminates is used to indicate which train is to depart first and is used most frequently when trains in both platforms have the same destination." (www.districtdave.co.uk/html/earls_court.html). It's my understanding that this is mainly intended to help, say, people wanting to go to Upminster to decide which Upminster train to get on. If you want to go to High Street Ken, it's junk information to know that the Upminster on platform 2 is gonna be departing first. Of course, if you've only got to get to Plaistow and there's one for Barking and one for Upminster, it would, naturally, be handy to know which one was out first. I suppose they were just more concerned with conveying the destination, since Earl's Court is before a diverging junction?

I agree! So I've added it to the OP of the thread.

The first train indicators at Earls Court are very complicated to work on let alone explain. But they are controlled by the programme machines for the 2 starters which in fact are non mechanical and are purely electronic cards. Of course the signal op can overide them but when that happens they usually stay off until the signal op decides which one is going first.

The DMI's at Hammersmith work on the TD's when they leave Turnham Green for the District and Acton Town for the Piccadilly. And on the other road when a Piccadilly leaves Barons Court it shows the standard 1min but can get more info from the reciver that works Hyde Park to Barons Court but that is actually housed at Hammersmith. The District DMI again gets it info from when the train leaves Barons Court but it can be wrong so when the train comes into Hammersmith the programme machine will correct the TD and thus the sign at the last minute.

For the last 2 years the TD system on the whole of the Piccadilly upto South Harrow has been overlayed with a train following system and also the District from West Kensington going west. Like all addon systems it can and often goes wrong with little or no warning and it happens so quick the signal ops sometimes dont notice it until its too late.

My recollection of Earls Court eastbound first train indicators is that quite often the first train indicator would be pointing to the second train out! It was not an error, the timing of the illumination of the first train arrow is what is seldom understood by passengers. Quite often with two trains routed towards Gloucester Road one would already have a green but had not yet closed doors or departed and quite correctly the next train would be indicated as the other train still sitting at a red. The idea is simple enough, once the signal is cleared the driver is expected to be able to close the doors and depart without passengers attempting to board as s/he does so.
Back in 1979 when I was based at Earls Court I had plenty of time to experience the 'which train will depart first' game every afternoon as I travelled to Victoria on my way home. In those days the drivers would sometimes be having a chin wag for a minute or two after one had been given a green and as several of us travelled the same route home via the Victoria line we'd usually man a door on each train while watching for the first train sign to illuminate and then change as one starter was cleared. There was always enough time for those of us on the 'wrong' train to transfer to the other as the doors closed unless the signal operator changed his/her mind about which train was going first and gave us all another minute or two to think about it!
Camden Town on the Northern was another place where passengers often complained about the next train indicator being wrong, two chances of avoiding a wait there, a dog's chance and no chance as it was a race up and down stairs between platforms rather than a few steps across an island platform as at Earls Court. Of course difficult for anyone with no knowledge of signalling to 'see' what was happening there.

[/color]

If they could just remove this phantom train you would be left with a system that told you, seemingly with reasonable accuracy: "Next xxx train within 5 mins" which would be quite useful.[/quote]


Generally speaking if a train is indicated on a DMI then it is somewhere between that DMI and the point at which its description is transmitted forward. The minute countdown is not based upon real time but the dropping of tracks as the train proceeds between the two points. The indicated time of 1 minute usually indicates that the point at which the description was transmitted ahead is relatively near (e.g. platform 5 DMI at Baker St H&C shows 1 min as soon as a train departs Edgware Road). If the train sits between the two points for several minutes the countdown stops (hence 10 minute countdowns on the Northern might often be timed at 15 minutes or more).
When the train does not arrive at all and subsequently disappears it is usually because the train is no longer on the route, in the case of Hammersmith I suspect a train has entered Acton Town sidings and thus disappeared.
No doubt you next question would be another 'why?' and that is difficult to answer without observing exactly what occurs at Hammersmith IMR and then doing the same at Turnham Green and Acton Town East IMRs. There are a number of possibilities including basic incompatibility of old and new technologies but more likely a minor fault. Eastbound trains arriving at Acton Town and terminating there should have such a description transmitted ahead from Ealing Broadway (Picc & District) or Northfields (Picc) but if trains are running out of turn it is quite possible that a train terminating at Acton Town will be indicated as an Arnos Grove or Cockfosters rather than Terminates Here or Special etc. Descriptions are transmitted ahead when a train passes the starting signal and replaces it to danger and so it is possible for wrong descriptions to be transmitted ahead although it shouldn't happen. Once a TD receiver has an approaching train 'stored' it will want to display it via the local sign combinator (the piece of TD kit that used to light all the appropriate bulbs in the old illuminated TD signs) and will also be waiting to retransmit the description to the next site.
I suspect the signal operator corrects the phantom train as the next actual Upminster train is stored and that may be why no time is then indicated. You may wonder why wrong descriptions get sent ahead and that is another tin of worms which in my days on the system was often the result of programme machine problems but could also be the result of an open leg in a cable changing the ABCDE code from the correct code to an incorrect one by 'dropping a letter' or a short circuit contact fault 'adding a letter'.
I never got too deeply involved with the interface between DMI an the older TD technologies as most of my maintenance area still used the old technology and I know nothing of the circuitry and interfacing of train following systems so i cannot speculate further. The TD cable between Acton Town East and Turnham Green used to give trouble but I understand that it has been replaced since I retired, of course programnme machines have also been replaced in that time too.
Daily I know more and more about less and less and will no doubt shortly know absolutely nothing about everything!