|
Post by Dstock7080 on Aug 18, 2021 12:26:59 GMT
Temporary changes TTN94/21 12 September 2021-3 October 2021:
|
|
|
Post by capitalomnibus on Aug 18, 2021 23:25:02 GMT
What is a rusty rail movement?
|
|
rincew1nd
Administrator
Junior Under-wizzard of quiz
Posts: 10,222
|
Post by rincew1nd on Aug 18, 2021 23:33:22 GMT
If trains don't regular run over a section of track the surface of the rails oxidise and become coated in rust as a result.
To make sure that all is well should these sections of track be needed in anger, there are scheduled moves over these rusty rails.
|
|
Tom
Administrator
Signalfel?
Posts: 4,083
|
Post by Tom on Aug 19, 2021 9:14:21 GMT
To further expand on rincew1nd's post: If the rail surfaces become rusty, there is a chance that track circuits may not operate correctly when a train runs over them (as rust is a poor conductor of electricity), which risks the train becoming invisible to the signalling system. 'Rusty Rail' moves are also written in to the timetable to ensure that Train Operators maintain their familiarity with the less common moves on the line, the theory being that if the Train Operator follows their roster, they will cover all the duties at their depot, and over the course of the roster they will do every move.
|
|
|
Post by superteacher on Aug 19, 2021 10:03:48 GMT
Please don’t tell me that running times are being increased further - the line is already ridiculously slow.
|
|
|
Post by nig on Aug 19, 2021 19:37:09 GMT
Please don’t tell me that running times are being increased further - the line is already ridiculously slow. timings have stayed the same but only for 3 weeks then autumn (leaf fall ) timetable comes in on October the 4th and that does increase running times
|
|
|
Post by ruislip on Jan 16, 2022 22:08:10 GMT
When did the Picc stop using set numbers starting with 200? Two years ago I remember riding on SN 203 on Valentines evening between Eastcote and South Harrow with my wife(BTW her first trip outside the USA). It was ex-Uxbridge terminating at Arnos Grove.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Jan 17, 2022 15:31:36 GMT
The Picc still use 2xx numbers - 220-256, excluding '8' and '9' (plus, of course, the 3xx series).
|
|
|
Post by ruislip on Jan 17, 2022 18:30:07 GMT
I was referring to set numbers in the 20x range, unless I got mixed up and actually travelled on SN 230.
|
|
towerman
My status is now now widower
Posts: 2,860
|
Post by towerman on Jan 17, 2022 19:10:45 GMT
Back in the day the numbering system on The Picc was 251-277 Rayners Lane/Uxbridge & 301-367 Hounslow then later Heathrow Central as was.
|
|
|
Post by nig on Jan 17, 2022 20:27:35 GMT
don't think they have used 200 series as running numbers although there is a car number matching that .. numbers that have been around for a while is 220 - 256 Northfields Depot starters. 300 - 335 Cockfosters Depot starters. 340 - 344 Arnos Grove Sidings starters. 350 - 357 Arnos Grove - Northfields. 360 - 367 Uxbridge and south Harrow stabler's/Starters
|
|
|
Post by MoreToJack on Jan 17, 2022 22:25:30 GMT
Given the use of the 20x/21x series by Circle line services, these would have been avoided by the Picc due to the shared signalling control at Earl’s Court. Indeed, trains 200, 211 and 212 (SuX) and 207 and 215 (SuO) all start or finish at Ealing Common depot, immediately precluding their use within Piccadilly line numbering (although this is a relatively recent development).
I do recall, however, that there was some alteration of train numbers to avoid duplication with the Piccadilly, although this may have been related to the H&C 26x/27x series, which also has some trains starting or finishing at Ealing. Interestingly (without getting too off topic), the relatively crowded 2xx series also led to the revival of the former East London line numbering, 17x, for two change-over trains during the day from Ealing (and Upminster), the now-withdrawn 171/172 (the former will always fill me with dread).
|
|
Tom
Administrator
Signalfel?
Posts: 4,083
|
Post by Tom on Jan 18, 2022 0:04:26 GMT
Many years (15 or more now) there were some 17x train numbers used by the District, though I believe they began where the East London line numbers finished. I think 176 was the lowest, possibly 175?
For many years the Piccadilly line number series was 250-367, with 250 being allocated to the Aldwych shuttle until 1994.
|
|
|
Post by d7666 on Jan 18, 2022 0:41:36 GMT
There is also another reason I have heard of where train numbers are allocated in different ranges to avoid confusion.
Connect radio "cells"***
Similar in principle to one signalling control centre controlling two lines with radios is where two or more lines intersect at one location in a common radio cell, practice should be no duplication.
Before people dive in, and list train numbers, it is not directly obvious which locations do share cells. If there are stations where two lines respective train numbers overlap, then you can be pretty sure there is no common radio cell issue at that location.
In normal radio operation should there not be any confusion. But it is one of these things that should be avoided i.e. ALARP risk to avoid situations with a wrong channel like train 123 on the Bakerdilly emergency call confused with train 123 Pickerloo at same station but does not have an emergency.
Somewhere I have a diagram of these cells including intersecting stations; not that I could upload it anywhere pretty sure that's not for the public domain; probably out of date anyway now (these things are not static behind the scenes).
*** I am using cell here loosely for want of another term.
|
|
towerman
My status is now now widower
Posts: 2,860
|
Post by towerman on Jan 18, 2022 17:15:45 GMT
Back in the day the numbering system on The Picc was 251-277 Rayners Lane/Uxbridge & 301-367 Hounslow then later Heathrow Central as was. As an aside,when The Victoria Line opened trains were numbered in the3** series,they only changed to 2** series when the line opened to Victoria.
|
|
Tom
Administrator
Signalfel?
Posts: 4,083
|
Post by Tom on Jan 18, 2022 22:00:53 GMT
That's interesting as they would have clashed with the Picc Line numbering sequence (though as there were no Programme Machines at Finsbury Park and no train radio it probably wouldn't have been a problem). Did the Vic therefore take the old Northern City numbering sequence which was 301-306, and by then had moved to 271-274?
|
|
towerman
My status is now now widower
Posts: 2,860
|
Post by towerman on Jan 18, 2022 23:50:14 GMT
Trains from N/Park were numbered 307 onwards,Walthamstow sdgs starters were 305 & 306.
|
|
|
Post by sealingmark on Jan 28, 2022 16:46:54 GMT
wtt-59, which is itself a PDF says: "A digital version of this Timetable can be found on the Establishment Planning Sharepoint site at transportforlondon.sharepoint.com/sites/EPTimetables/" What do they mean by digital version? Could there be a spreadsheet available, pretty please?! I've wasted hours and days trying to write an app to pick apart the pdf. Not possible :-( Presume there must also be temporary revisions (sorry; airline-speak) issued to wtts between revisions? Surely a timetable can't stay valid for a year or more?
|
|
|
Post by Dstock7080 on Jan 28, 2022 16:50:45 GMT
What do they mean by digital version? Could there be a spreadsheet available, pretty please?! The print and PDF versions are the same, so anyone looking at the print version are directed onto the internal intranet PDF version. Presume there must also be temporary revisions (sorry; airline-speak) issued to wtts between revisions? Surely a timetable can't stay valid for a year or more? Timetables can remain unchanged for sometime (when I first started the District were still using a WTT from 1974!) they can be updated by using TTN Temporary Timetable Notices, as is currently happening on the Northern Line and on the Piccadilly because of the overrunning work at South Harrow:
|
|
|
Post by sealingmark on Jan 29, 2022 15:41:21 GMT
Useful, thanks. I don't think TTN are available to the public, unlike wtts, which are.
So a wtt stays mostly valid for quite a while. Therefore it's worth me spending a hour or so manually extracting a list of Northfields depot trains.
Wouldn't fancy doing that every month, but if it's a couple of times a year, and then the results are mostly good, that works for me.
My regular journey is SEA to HSD and it's very nice to get on a near-ermpty train.
Looks like NFD depot starts usually show a CurrentLocation of 'Northfields crossover', but I suspect that if the track is clear, you could blink and miss that as they run straight to a NFD platform.
|
|
Chris M
Global Moderator
Forum Quizmaster
Always happy to receive quiz ideas and pictures by email or PM
Posts: 19,359
|
Post by Chris M on Jan 30, 2022 1:03:39 GMT
I don't know whether it's more useful for your purposes but you can work out which trains start at Northfields Depot from online journey planners - for the time period you are interested in, plan a journey from Boston Manor to South Ealing, noting the arrival times. Then plan a journey from Northfields to South Ealing for the same time of day and again note the arrival times. Any trains on the second list not on the first list started at the depot. It might be possible to get this from the API data, but I wouldn't have a clue how.
|
|
|
Post by Dstock7080 on Jan 30, 2022 7:27:49 GMT
I don't know whether it's more useful for your purposes but you can work out which trains start at Northfields Depot from online journey planners The online Working Timetables already easily show which trains depart from each depot/siding in the Depot Working pages at the front. (Piccadilly pages 9, 10, 11)
|
|
|
Post by sealingmark on Jan 30, 2022 9:10:37 GMT
Chris, that was one of my plans, but foiled by the way the Timetable data works.
The snag is that it doesn't seem to cope where times are something like 08 38 3/4 (meaning 08:38:45). Also is doesn't cope with 09b38 (b meaning arrives 1 minute earlier.
The TT data *always* thinks it takes 4 minutes from BOS to SEA, which it mostly does, but not when there's "arrives x minutes earlier" or a departure time involving odd seconds.
The net result is that the TT says e.g 09 38 from BOS arrives SEA at 09 42. But SEA says it arrives at 09 44 (which it does).
So when I do that comparrison, I see a 09 44 for SEA that doesn't match any departure from BOS. So it's a NFS depot train - WRONG, it's just got an incorrect time listed.
Working on Plan C now
|
|
|
Post by rheostar on Jan 30, 2022 9:47:04 GMT
wtt-59, which is itself a PDF says: "A digital version of this Timetable can be found on the Establishment Planning Sharepoint site at transportforlondon.sharepoint.com/sites/EPTimetables/" What do they mean by digital version? Could there be a spreadsheet available, pretty please?! In Service Control, we'd download and print out the digital PDF version of the timetable, then put each page into a clear plastic folder. The line controllers would use this version to keep track of cancellations, short tripped trains or reforms using a chinagraph pencil. At the end of each traffic day the chinagraph marks made on the timetable are cleaned off. The digital version is in a slightly bigger format to the printed and bound timetable making it easier to write on with a chinagraph pencil. The same was done for TTNs. It's not very high tech, but it worked. To my knowledge, there was never a spreadsheet version of the timetable available.
|
|
|
Post by MoreToJack on Jan 30, 2022 10:40:03 GMT
The same is still very much done across service control centres today: as you say, it works!
No timetables in spreadsheet form; some other machine-readable formats do exist but these are all internal use only for the signalling computers, and generally only have the data for an individual site.
|
|
|
Post by sealingmark on Jan 31, 2022 18:24:47 GMT
The online Working Timetables already easily show which trains depart from each depot/siding in the Depot Working pages at the front. (Piccadilly pages 9, 10, 11) Completely correct.
I phrased my question poorly.
I'm not looking to identify which trains start their working day at NFD, but those that start a Trip there (and so will be pleasantly empty when arriving at platform)
In general, the wtt/ttn can answer ALL questions, but i've not found a way to read that PDF file in an app. It's a VERY complex table of table (of tables). Still trying to find a way to parse it.
|
|
|
Post by d7666 on Jan 31, 2022 19:04:14 GMT
Chris, that was one of my plans, but foiled by the way the Timetable data works. The snag is that it doesn't seem to cope where times are something like 08 38 3/4 (meaning 08:38:45). Also is doesn't cope with 09b38 (b meaning arrives 1 minute earlier. The TT data *always* thinks it takes 4 minutes from BOS to SEA, which it mostly does, but not when there's "arrives x minutes earlier" or a departure time involving odd seconds. The net result is that the TT says e.g 09 38 from BOS arrives SEA at 09 42. But SEA says it arrives at 09 44 (which it does). So when I do that comparrison, I see a 09 44 for SEA that doesn't match any departure from BOS. So it's a NFS depot train - WRONG, it's just got an incorrect time listed. Working on Plan C now Internally the signalling control timetable data files are resolved to 1 second; they do not use the hh:mm:ss format but the five digit number format. I know, as I have done it, for work purposes, to import the complete SSL control data files into one single excel file. It is easily human readable if you know how to twiddle the import stage column by column (and surprisingly the resultant file is not that huge). But, elsewhere in thread as Blue Goblin commented, these control data files are for internal use only; can't let you have them, sorry. IIMU resolution to quarter minutes is necessary for the printed versions but AIUI only those stations that appear in the printed WTT have control data times resolved to hh:mm:15 hh:mm:30 hh:mm:45 hh:mmm:00 so they can print 1/4 1/2 3/4 characters or nothing as needed, intermediate stations or other locations not printed can have any value 00 to 59. At least that applies to the timetables I have looked at, it might not apply to all lines or all sites with legacy equipment. Fairly sure these files would give you the solution you seek, but I'm afraid even pretty please we can't give them to you. Maybe you could try a Freedom Of Information request for the timetable control data CSV file(s) for the line(s) you want. I'll predict it would be rejected though.
|
|
|
Post by sealingmark on Feb 1, 2022 18:29:49 GMT
Internally the signalling control timetable data files are resolved to 1 second; they do not use the hh:mm:ss format but the five digit number format. Thanks for yet more 'inside info'; interesting to know what exists, even if I can't see it. I would want to base any 'solution' on publically available data (such as current wtt and anything from the TfL website/API) so as not to impose on good-natured folk here. Not familiar with the five digit number format you mentioned, but thinking about it, I'd speculate it is 'seconds since midnight'. Better than the highly non-standard TfL Timetable notation of hours going up to 24 and 25 (and that's when the night tube isn't running).
I've now got my list of Northfields starters from a lot of head-scratching and 130 lines of python code.
|
|
|
Post by zbang on Feb 1, 2022 19:27:35 GMT
Hours over 23 is not uncommon for scheduling things that slop over the usual day boundary but still need to be managed as part of the original day; I've seen this often for night-time music events and television productions. And a 5-digit seconds counter can go a shade over 27:46:39, so as long as the service ends by 3:46am, it's all good . I may take a crack at parsing the public WTT files just to see if I can.
|
|
|
Post by sealingmark on Feb 3, 2022 16:31:58 GMT
I may take a crack at parsing the public WTT files just to see if I can. I wish you success. I've tried a handful of ways of converting pdf to text or spreadsheet. They all produced differently incorrect results. Could be that the correct tool, or approach, will work, but I haven't found it.
It saved time to extract just 1 pdf page of timetable from the wtt and experiment on that.
I suppose 'success' would be a consistent column of data with a train no at the top and times reliably spaced following (i.e. the Northfields time (or dashes) will always be in row X, following the train no. That would then be trivial to ingest into a database.
The only reason the TfL Timetable data object isn't suitable is that it doesn't include a train no. Live arrival predictions do include train no, but Timetable data doesn't.
|
|