mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Dec 1, 2009 12:27:23 GMT
I *think* [1] you can only achieve counter-conditional locking electrically - this is by virtute of electric interlocking being non-reciprocal - every single locking function has to be provided individually for electrical locking, the converse is not automatically there as it is with mechanical locking.
(I know you and Tom understand what I'm going on about; but for the the benefit of the OP and anyone else that might be interested):
'Dead' Locking
locking which is always required: in mechanical locking dead locking is always reciprocal: if 1 locks 2, 2 will lock 1.
Conditional interlocking
Locking that is dependent on on the position of other levers if mechanical or the state ofother functions if electrical.
Counter-conditional interlocking
Needed when the condition (or function that creates the condition) would otherwise be free *after* the locking is effective.
Using the 'Green Textbook' to illustrate reciprocal and non reciprocal interlocking:
Reciprocal - double line trailing crossover:
Signal 10
-------|--------------------/102/ points-----------------
---------------------------/102/ points-------------------
Signal 10 requires 102 points normal; for 102 points to move normal to reverse Signal 10 is required normal. Mechanically this would be achieved with a straight dead lock; electrically this would be achieved with TWO separate lock coils: one active when lever/switch 102 is normal and locking 10 normal; the other active when lever/switch 10 is normal and locking 102 normal.
Non-reciprocal
Signal 10 Signal 12
------|-----|---------------/102/ points-----------
---------------------------/102/ points------------
Same as above, trailing crossover between two lines, but with two signals in this case (due, perhaps to the braking rates of trains approaching the crossover and that Signal 12 is protecting the crossover directly) the designer wants to achieve points 102 moving from normal to reverse requiring 10 normal *but* in order to clear signal 10 102 points are not required to be normal - this cannot achieved mechanically [2] but can be achieved electrically with lever/switch 10 normal there is a lock coil holding lever/switch 102 normal.
[1] I'm sure this is the case - even with exotic stuff like rotation locking; sequential locking and Scott-Ross check locking. (essentially mechanical means of making sure signals are pulled and replaced in the correct order before they can be pulled again)
[2] I'm *not* mentioning sprung dogs here; as this is about UndergrounD practice - but I do have a copy of the dog chart and locking table for the 60 lever frame at Rochdale Goods Yard if anyone is really interested.
|
|
|
|
Post by programmes1 on Dec 1, 2009 16:27:58 GMT
I remember the old frame at Queens Park it was possible to normalise the levers controlling entry into the North sheds before a train was fully berthed, the signalman called it sectional release.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 1, 2009 17:41:39 GMT
sounds like early blacklock release to me but without looking at the site plan and a the lock circuit print you will never know
|
|
mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Dec 1, 2009 17:52:49 GMT
I remember the old frame at Queens Park it was possible to normalise the levers controlling entry into the North sheds before a train was fully berthed, the signalman called it sectional release. Yes; I've come across that before - (from memory) it was a form of Train Operated Route Release (TORR); which in effect is early backlock release - but you'd never know unless ....... *lightbulb switches on* - scurries off into library.... It's mentioned in First Supplement to TC 36/1953 - I'll key in the relevant bits from it later as I'm just about to go out with AI. ;D |
|
|
|
Post by tubeprune on Dec 1, 2009 20:39:14 GMT
I worked at Queens Park. I remember the early release. I always thought it was so you could set up quickly for the next train.
|
|
rincew1nd
Administrator
Junior Under-wizzard of quiz
Posts: 10,286 
|
Post by rincew1nd on Dec 1, 2009 21:12:17 GMT
...the designer wants to achieve points 102 moving from normal to reverse requiring 10 normal *but* in order to clear signal 10 102 points are not required to be normal Why though? As I read this if a train can pass 10 (clear) and then SPAD 12 it could run through 102 set against it, however if it were to SPAD 10 and 12 then 102 could be moving under the train. ***clunk*** Is this because the overlap of 12 is over the cross over, but 10 isn't, thus you could stop a train at 10 and have another train crossing over (if it was moving towards 12 (authorised) then the route would have to be set over the cross-over). However, it allows the signaller to change the points to normal after a SPAD at 10, so that in the event of a second SPAD at 12 the points can be thrown over to minimise damage? Rambly I know, let me know if I'm confusing people. |
|
mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Dec 1, 2009 23:47:09 GMT
Rambly I know, let me know if I'm confusing people. You've read my training paper on Reg. 4 et al [0] overlaps, haven't you? Look through the bits at the end for FpEx LEs - there is a casual, throwaway [1] comment about 'swinging overlaps' part of this (in the textbook) is looking at a swung overlap but not from the point of view of the movement for which the overlap is swung but a movement into which an overlap could be swung - in this particular textbook example no spatial concepts are given - it is dimensionless! Not through any desire to confuse, more to get people thinking of the principles. The other part of the answer is 'Yes, your penny might be spinning, but it has dropped in the correct orientation' [0] *not 5* or 4A, for that matter - 8, possibly 9 with a hint of 15. [1] ahem, <clears froat>  ;D |
|
mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Dec 2, 2009 1:06:25 GMT
I worked at Queens Park. I remember the early release. I always thought it was so you could set up quickly for the next train. *returns to usual chair* Yes - that's certainly the impression I've got from reading the peril - the germane bit of the text reads: 8 - TRACK CIRCUIT CONTROL OF SIDING INLET SIGNALS
(a) Nos 22 and 23 sidings [1] will each be divided into two separate track circuits. Shunt Signals BB 4B and BB 7 [2], authorising movements into the sidings will be controlled by short track circuits at the south ends of the sidings of the sidings, which will extend to a point 150 feet north of the siding outlet signals [3]
(b) Nos 25 and 26 sidings [4] will each have one track circuit. Shunt Signals BB 21A and BB 25A [5], authorising movements into these sidings, will be controlled by these track circuits. Two push buttons will be provided in the lever plate of signal BB 21A and two in the lever plate of signal BB 25A for use (by the operation of the appropriate button) when signalling a train into either No 25 or 26 road when that road is already partially occupied. On each occasion when a push-button is operated it must be held in until the train proceeding to siding has passed the shunt signal concerned, and that signal has returned to Danger.[1] North Shed, dead end sidings [2] Right road and Bang road discs into North Shed dead ends [3] Tracks GD and GJ respectively [4] South Shed [5] Bang road and Right road discs into South shed Given the distance between the first set of blades (platform side) to the appropriate fouling point/track circuit boundary - the time taken for just over half-a-train-length is saved by having the points behind released by the track in front (d'you get my drift?) I'm pretty sure that the early train operated release only affected the point levers to/from the New Lines. Given the fact that a similar (ish!) sort of release is provided for in different ways between the North Shed and the South Shed - both are releases for point levers (one is from normal to reverse, the other is from reverse to normal), I wonder if this was an experimental installation, rather than using ∆s which were proved in service by 1939. There is also the possibility that ∆s were not quite accurate enough! |
|
|
|
Post by railtechnician on Dec 2, 2009 10:37:33 GMT
I wonder if this was an experimental installation, rather than using ∆s which were proved in service by 1939. There is also the possibility that ∆s were not quite accurate enough! Presumably this was the older delta circuit where a relay was picked up directly by a shunt on the isolated running rail section rather than the 10kC delta circuit that we all know and love. Such older deltas were still around when I began my career at LT. |
|
mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Dec 2, 2009 12:57:26 GMT
Presumably this was the older delta circuit where a relay was picked up directly by a shunt on the isolated running rail section rather than the 10kC delta circuit that we all know and love. Such older deltas were still around when I began my career at LT. I asked a similar question about the First ∆s here back in 2007. |
|
|
|
Post by railtechnician on Dec 2, 2009 14:01:21 GMT
Presumably this was the older delta circuit where a relay was picked up directly by a shunt on the isolated running rail section rather than the 10kC delta circuit that we all know and love. Such older deltas were still around when I began my career at LT. I asked a similar question about the First ∆s here back in 2007. Thank You, that answered my question nicely. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 2, 2009 19:42:13 GMT
I wonder if this was an experimental installation, rather than using ∆s which were proved in service by 1939. There is also the possibility that ∆s were not quite accurate enough! Presumably this was the older delta circuit where a relay was picked up directly by a shunt on the isolated running rail section rather than the 10kC delta circuit that we all know and love. Such older deltas were still around when I began my career at LT. Whitechapel still has them but on the diagram they are marked up as a delta but its just a track about 30Ft in length |
|
rincew1nd
Administrator
Junior Under-wizzard of quiz
Posts: 10,286
|
Post by rincew1nd on Dec 2, 2009 20:22:35 GMT
You've read my training paper on Reg. 4 et al [0] overlaps, haven't you? Naturally, I still have a copy in my bag for when I have to train people. Remember it was primarily your tuition that gained me the piece of card. Look through the bits at the end for FpEx LEs - there is a casual, throwaway [1] comment about 'swinging overlaps' part of this (in the textbook) is looking at a swung overlap but not from the point of view of the movement for which the overlap is swung but a movement into which an overlap could be swung - in this particular textbook example no spatial concepts are given - it is dimensionless! Not through any desire to confuse, more to get people thinking of the principles. So the signaller can reverse the points, only after warning anyone working in the West Shed? The other part of the answer is 'Yes, your penny might be spinning, but it has dropped in the correct orientation' I didn't know this was a forum where "orientation" was mattered! "Big winking smiley" ;D |
|
mrfs42
71E25683904T 172E6538094T
Big Hair Day
Posts: 5,922
|
Post by mrfs42 on Dec 2, 2009 20:23:41 GMT
Whitechapel still has them but on the diagram they are marked up as a delta but its just a track about 30Ft in length Still just about on-topic a little historical interlude about ∆s! Like these: ∆ 43/44 ∆ 100 and ∆ 5 There's also ∆ 360 and ∆ 350 sculling around somewhere. Now; am I right in thinking that these are just straight pick up/drop tracks - just like a normal TC and active throughout the whole 30' of the rail between the block joints? Unlike a 10kHz ∆ which will pick up 'within a zone'. Yes? In the case of ∆ 360 it is part-way along BB track - in fact I'm pretty sure that all the ∆s at Whitechapel are fitted into TCs (it's the same one either side of the delta); is it just as simple as that there will never be a train run with less than a 30' wheelbase so it won't vanish in the middle of the track? |
|
|
|
Post by railtechnician on Dec 3, 2009 9:13:37 GMT
Whitechapel still has them but on the diagram they are marked up as a delta but its just a track about 30Ft in length Still just about on-topic a little historical interlude about ∆s! Like these: ∆ 43/44 ∆ 100 and ∆ 5 There's also ∆ 360 and ∆ 350 sculling around somewhere. Now; am I right in thinking that these are just straight pick up/drop tracks - just like a normal TC and active throughout the whole 30' of the rail between the block joints? Unlike a 10kHz ∆ which will pick up 'within a zone'. Yes? In the case of ∆ 360 it is part-way along BB track - in fact I'm pretty sure that all the ∆s at Whitechapel are fitted into TCs (it's the same one either side of the delta); is it just as simple as that there will never be a train run with less than a 30' wheelbase so it won't vanish in the middle of the track? Yes, straight relay pick up (via low voltage high current circuit, the shunt connects the secondary of the feed transformer to the primary of the relay transformer) when shunted. The isolated rail should be 11' 6" long, I think I still have the official original delta example circuit diagram somewhere as issued in the little black books that all linemen once had as a pocket reference. The isolated rail would be some 30' in rear of a signal where used to approach control the signal. I recall that this could cause problems when drivers drew up very slowly and stopped short waiting for the signal to clear! More than once I remember someone having to shout at drivers to 'pull right up'. Just like the 10kC circuit it is a track circuit within a track circuit, hence the way it is shown as the isolated section. The use of transformers rather than direct pickup was to mitigate any possible interraction of track circuit and delta rail circuit in the event of failed blockjoint(s). Obviously the 10kC circuit more accurately detects the front of a train, usually to within 5-10 feet depending upon the type (ISTR there are six types of 10kC circuit in use across the various lines using different transmitter/receiver/relay combinations). I worked mostly with type 'A' and type 'D' deltas and the odd 'C' type. Standard delta tracks IIRC are minimum 40 feet long and the signal resonance occurs at the zero feet mark (the point where the designation arrow is placed on the scale plan) i.e. the signal increasing to resonance as the train approaches. I recall one such delta which is not even 20 foot long as it is at the very beginning of a track, 406E, approaching Hatton Cross from Crane Bank. I recall there being little adjustment in it, it either worked or it didn't but otherwise did not quite fit the standard resonance profile. |
|
Tom
Administrator
Signalfel?
Posts: 4,196
|
Post by Tom on Dec 5, 2009 23:33:44 GMT
From memory there are five varieties: A, B, C, D and CD. A, D & CD use an F3 relay whereas B & C use QN6s. IIRC the CD uses a C stlye unit and a D style reciever.
The spread area of a delta can be up to 76' long, depending on the type of circuit (i.e. maximum permitted spread area 38' either side of the actual rail connections), hence the reason they are positioned a train's length plus 40' away from the points (used to be the last set of facing points, now the last set of points).
I would disagree with the 10kc Delta being a more accurate detector of train position - it will pick up if there is a shunt anywhere in the spread which can increase over time The correct operation of a 10k/c Delta places a lot of emphasis on the Technician correctly setting it up.
A set of blockjoints with the old style Delta or a Position Detector is far more accurate in that respect, plus the PDs need less maintenance and cannot affect the track circuit's operation (remember the 50p on the 'Track' terminals in the school?)
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 6, 2009 17:31:55 GMT
try a paperclip on that new one they have
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 6, 2009 17:48:35 GMT
plus you forgot treadles i personally think they are far more realiable
|
|
|
|
Post by railtechnician on Dec 6, 2009 18:03:13 GMT
From memory there are five varieties: A, B, C, D and CD. A, D & CD use an F3 relay whereas B & C use QN6s. IIRC the CD uses a C stlye unit and a D style reciever. The spread area of a delta can be up to 76' long, depending on the type of circuit (i.e. maximum permitted spread area 38' either side of the actual rail connections), hence the reason they are positioned a train's length plus 40' away from the points (used to be the last set of facing points, now the last set of points). I would disagree with the 10kc Delta being a more accurate detector of train position - it will pick up if there is a shunt anywhere in the spread which can increase over time The correct operation of a 10k/c Delta places a lot of emphasis on the Technician correctly setting it up. A set of blockjoints with the old style Delta or a Position Detector is far more accurate in that respect, plus the PDs need less maintenance and cannot affect the track circuit's operation (remember the 50p on the 'Track' terminals in the school?) Tom, I spent a lot of time on delta spreads in my routine maintenance on the Picc and the relays certainly did not all pickup 'anywhere' in the spread. My suggestion that a 10kC delta was more accurate assumed pickup at 5' either side of the zero point i.e. 10' whereas the old style was 11' 6". Of course it all depends upon the type because type A had a must pickup figure of at least 27' whereas type C was anything from 5' having just looked it up! I can't speak about PDs as I never had any involvement with them, we had none while I was on maintenance, ISTR the first units going in at Hanger Lane Junction but I was working in the control room by then. I did get the paperwork on them but as usual no training! I don't know why I thought there were six types, I've just looked up some of my old maintenance returns and the schedules do show just the five! I think senility is setting in, I seem to be having senior moments more often now! |
|
Tom
Administrator
Signalfel?
Posts: 4,196
|
Post by Tom on Dec 6, 2009 20:42:14 GMT
Point taken - in an ideal condition it should pick up with a shunt anywhere in the spread, but as we both know things aren't always ideal. Some were positioned with the intention of having a small spread, especially if used for TETS circuits (Ealing Broadway springs to mind). I've got a feeling that at least one type's performance tails off the closer you get to the rail connections as well. try a paperclip on that new one they have Tin foil on a PD is another of my preferred failures to simulate... ;D Just don't get me started on Treadles - I used to detest changing them and they got so badly hammered by the wheels the average life was about six months. |
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Dec 6, 2009 20:47:59 GMT
treadles are nice and simple bit of kit but have to be set up correctly in first place we have some which are over 10 years old and still work so dont give up on them just yet Tom
|
|
;D
