Wikipedia:Reference desk/Archives/Science/2023 June 13

= June 13 =

Edgware Road
So, let's say an outer rail service is just coming in from Kensington and will terminate at Edgware Road -- will it (normally) stop on Track 1 or Track 3? If on Track 1, then how will it reverse without disrupting other outer rail trains coming in from Hammersmith (especially since, according to a (possibly outdated) schedule, there's one coming in from Hammersmith and going on toward King's Cross just 1 minute after the terminating service from Kensington)? Also, once the train reverses and goes around the inner rail, after it comes in from King's Cross, will it (normally) stop on Track 3 or Track 4 before continuing to Hammersmith? And last but not least, if I remember the layout of the Praed Street junction rightly, trains departing for Hammersmith from Track 4 must cross over both the Kensington-bound track of the Circle Line (which becomes Track 3 in the station) and the District Line track (Track 2) -- how do they avoid delays due to this? 2601:646:9882:46E0:6998:C31A:9514:BAA4 (talk) 02:53, 13 June 2023 (UTC)


 * Courtesy link: Edgware Road tube station (Circle, District and Hammersmith & City lines). --142.112.221.43 (talk) 03:23, 13 June 2023 (UTC)


 * This is a big problem for the tube trains, you can't have two going in different directions along the same tube - they have to be disassembled at the end of the line and transported back by lorry ;-) Sorry having looked at it I don't know. NadVolum (talk) 08:54, 13 June 2023 (UTC)
 * Surely you can just run it up an escalator? ;-) Alansplodge (talk) 15:18, 13 June 2023 (UTC)
 * I'm not locally familiar, but I found this map. Apparently, trains leaving westbound from track 1 can't move to the left-hand track. All London Underground trains run on the left (a few exceptions involve grade-separated crossings), so trains terminating at Edgeware road and returning to Kensington must use tracks 2 and 3. I assume one is for District line trains, the other for Circle line trains. A train leaving for Kensington from track 2 or 3 doesn't interfere with trains coming in from Hammersmith on track 1.
 * Assuming tracks 2 and 3 are for trains for Kensington, it makes most sense if trains for Hammersmith use track 4. The line from Edgeware Road to Praed Street junction appears to be simply double track, so all trains leaving westbound have to be merged onto one track. A bigger problem is that trains for Hammersmith have to weave through those from Kensington.
 * It looks like the track layout at Edgeware Road was designed such that any one of the tracks can be taken out of service without blocking any of the routes served, although capacity would be reduced. PiusImpavidus (talk) 09:09, 13 June 2023 (UTC)
 * I was on Edgware Road station on Sunday, when there was no service between Edgware Road and Hammersmith.  Trains coming from the east would cross to platform 1 where they would dwell before setting off eastwards again.   The signage is "platform 1 eastbound".   I've never known trains going east leave from any other platform - it's difficult to see how they could because platforms 2 and 3 are for Circle and District line trains completing their journey having passed through Paddington (that's the Praed Street station - there's another one near Bishop's Bridge Road for trains going to and from Hammersmith).   When these trains leave they go back the way they came.   If you're going Paddington (Bishop's Bridge Road) and stations to Hammersmith platform 4 is the place for you. 2A00:23C6:2417:3101:CD7D:9961:DB38:71A (talk) 11:28, 13 June 2023 (UTC)


 * Track layout map here. Alansplodge (talk) 15:18, 13 June 2023 (UTC)

I forwarded the original query to Clive Feather, who answered as follows. (I haven't compared his answers to the ones above.)


 * So, let's say an outer rail service is just coming in from Kensington and will terminate at Edgware Road -- will it (normally) stop on Track 1 or Track 3?
 * It will use platform 2, though it could in principle use platform 3.
 * The normal platform use is:
 * Trains towards Baker Street
 * Circle trains from Kensington terminating and returning to Kensington
 * District trains from Kensington terminating and returning to Kensington
 * Trains directly towards Hammersmith
 * If on Track 1, then how will it reverse without disrupting other outer rail trains coming in from Hammersmith...
 * There is no practical way for it to reverse in platform 1.
 * In principle a train could shunt forward into the running tunnels, then back into platform 2 or 3, but that would be incredibly disruptive.
 * Also, once the train reverses and goes around the inner rail, after it comes in from King's Cross, will it (normally) stop on Track 3 or Track 4 before continuing to Hammersmith?
 * It will normally use platform 4, though it could use platform 3 if there is disruption.
 * ...trains departing for Hammersmith from Track 4 must cross over both the Kensington-bound track of the Circle Line (which becomes Track 3 in the station) and the District Line track (Track 2)
 * No, that's not how it works.
 * The four platform tracks converge into two tracks just west of the station. The outer rail connects to platforms 1 to 3; the inner rail to platforms 2 to 4. This is done by a scissors crossover between the platform 2 and 3 tracks immediately west of the station, then 1 & 2 merging and 3 & 4 merging west of that.
 * Then, a short distance further west, there is a simple double junction and westbound Hammersmith trains will cross over the outer rail from Kensington.
 * So the platform 4 track doesn't "cross" either the platform 3 or 2 tracks. Rather, the three converge into one before splitting into Hammersmith and Kensington options. Similarly, eastbound, the two lines merge into one before splitting into platform 1, 2, and 3 tracks.
 * how do they avoid delays due to this?
 * Careful timetabling.

(Clive's answers posted with his permission.) --142.112.221.43 (talk) 16:06, 13 June 2023 (UTC)


 * Which all matches with the track layout map linked above. PiusImpavidus (talk) 18:01, 13 June 2023 (UTC)
 * Just a note that the map was linked twice, at 09:09 and 15:18 yesterday.  The second link has a zoom button but the first link is already enlarged. 2A00:23C6:2417:3101:DC52:6F3C:C31E:314A (talk) 14:02, 14 June 2023 (UTC)


 * Thanks, everyone! This pretty much answers most of my questions, except the last one -- the thing is, the circle portion of the line in either direction is scheduled for exactly 59 minutes (although, assuming no delays, it might take as little as 56, but then the train has to stay in the station until scheduled departure time) -- so, under the best circumstances, you'll have an outbound inner rail service departing for Kensington just 1 minute behind another one departing for Hammersmith (and going the other way, you'll likewise have a Baker Street-bound outer rail service coming in from Hammersmith just 1 minute behind a terminating service coming in from Kensington!  And there's also the H&C shuttle service to consider (I'm not quite sure of the times, but it would have to reverse on the same Track 4 which will also have to be used by the inner rail service coming in from Baker Street and heading for Hammersmith), as well as the District Line trains -- so, with all this in mind, I still have to wonder how the dispatcher can juggle all these without them becoming a royal mess (and how much aspirin does the dispatcher have to consume to deal with the resulting headaches)? 2601:646:9882:46E0:A55D:ACD1:411:3D8F (talk) 11:07, 17 June 2023 (UTC)
 * Just rechecked the (possibly outdated) timetable I have, and gasped in disbelief -- all these movements are (or at least were) scheduled to happen simultaneously, within 2 minutes! For a snapshot, let's take 8:29 AM on a weekday -- at that time (and not even considering the District Line trains), the following movements are (or at least were) supposed to happen: inner rail service arrive from King's Cross St. Pancras/depart toward Hammersmith (8:28); inner rail service depart toward Kensington; terminating outer rail service arrive from Kensington; H&C shuttle depart toward Hammersmith; terminating H&C shuttle arrive from Hammersmith; and outer rail service arrive from Hammersmith/depart toward Kings Cross St. Pancras (8:30).  So we have six (or even eight, depending on how you count) movements on three tracks within no more than two minutes -- how is it even possible? 2601:646:9882:46E0:FDCB:9371:5931:CCD4 (talk) 03:49, 18 June 2023 (UTC)
 * At all junctions controlled by signals it's a matter of waiting your turn.  At Leytonstone the two eastbound lines diverge and the two westbound lines converge.   Controllers make it as easy as possible for passengers.   All trains dwell for a minute or so and the driver of an eastbound service always announces the name of the next station.   The signal turns green, the driver closes the doors, and off she goes.   The westbound indicator board has an arrow pointing to the platform number of the next train out, and there's a similar facility at Hammersmith.   At any station inbound trains may have to wait outside for their platform to become free.   There may be a potential conflicting movement at Edgware Road every two minutes or less, but for any track layout points and signalling are electronically controlled to make it impossible to set up a conflicting movement.   In earlier times projections on the levers served the same purpose mechanically.   This is all very laid back compared to a busy road junction where there may be potentially conflicting movements every two seconds.   Traffic lights also handle filtering, late start, early cutoff and an "all-red" phase to allow pedestrians to cross.   Many years ago at the busy Balls Pond Road/Kingsland Road junction in Dalston a diagonal crossing was introduced, which may still be operational.   Sensors embedded in the road surface send information to computers which control the traffic light sequence to maximise traffic flow, often over a very wide area.   There are numerous level crossings - the biggest danger there is people zigzagging through the barriers.
 * Of what can be controlled by the transportation authority, timetablers have to deal with fast trains, slow trains, freight trains and single-track sections (the only example of single-tracking in London is the Romford-Upminster line).  The line to and from Liverpool Street is very busy - trains ex Liverpool Street bound for Barking have to cross the line between Forest Gate and Manor Park.   When the "jazz train" service was introduced on the West Anglia lines in the 1920s it was so popular that minimising the turnaround time at Liverpool Street became a fine art.   Freight trains can be as long as half the distance to the next station.   At Canonbury at six o'clock on Friday morning four freight trains passed through within five minutes, along with numerous empty passenger trains positioning themselves to enter service.
 * Railway planning is a ponderous process.  The new East London line, which eventually saw the reopening of Dalston Junction and a new station at Dalston Kingsland (thanks to the efforts of traders at Ridley Road Market - the original plan was to demolish part of the market to build the new station) was delayed for years because a trader at Petticoat Lane market had noticed that if construction does not begin within the time allowed by the grant of planning permission the whole planning process has to be begun again from scratch.   There is also the political aspect - Margaret Thatcher had to choose between the east-west crossrail (Elizabeth Line) and the north-south route (Hackney to Chelsea line).   She chose the former because "nobody I know lives in Hackney."     I don't recognise the "H&C shuttle".   Platform 4 is for westbound trains - running a train from Hammersmith in there would be suicidal. 2A00:23C3:FB81:A501:906F:D525:FEB8:5529 (talk) 13:38, 18 June 2023 (UTC)
 * Don't some trains from Hammersmith short-turn at Edgware Road and go back the way they came, instead of continuing on the outer rail of the Circle Line proper? If so, that's your "H&C shuttle" for you -- sorry for the non-standard terminology (that's what New Yorkers would have called such a service, at any rate)!  So, this one also uses Track 3 to reverse, just like the full Circle Line trains? And once again, given the fact that reversing a subway train probably takes more than just 1 minute, how do they handle the various terminating services at this station so they don't delay each other or the through services (or vice versa)?  2601:646:9882:46E0:7D3B:73DF:F255:8C8D (talk) 07:47, 19 June 2023 (UTC)

Are the most recent artifactual definitions of kg and meter still made of the best kg/meterstick substance known to man?
Well best to be the one national kilograms and metersticks aim for. Sagittarian Milky Way (talk) 03:55, 13 June 2023 (UTC)


 * The current definitions are not artifactual. The most recent artifactual definition of the kilogram is embodied in Le Grand K, manufactured in 1879. It was fashioned of a 90% platinum and 10% iridium alloy. The same alloy was used for the most recent artifactual definition of the metre, the International Prototype Metre, manufactured between 1886 and 1899. I don't know how to compare it with potential other substances (92% adamantium and 8% unobtanium?) for goodness. A search for "meterstick substance" did not turn up promising results. --Lambiam 06:09, 13 June 2023 (UTC)
 * I know the kg and meter are no longer the mass and length of physical objects by definition. I was wondering if they'd still use platinum with about 10% iridium if the objects were being built today or if they've realized something better is possible some time in the last 124+ years. Maybe some other platinum-group metal alloy? Or a 1 kilogram diamond? (it would have to be artificial as the largest uncut natural diamond ever was only a pound and a third). Would the kilo be spherical instead of cylinderical to minimize surface area? ~ Sagittarian Milky Way (talk) 17:04, 13 June 2023 (UTC)
 * If you define what the properties are you want to optimize for, the question might, perhaps, become answerable or at least researchable. (: Obviously, clay is better, as it is a hell of a lot cheaper. :) --Lambiam 20:39, 13 June 2023 (UTC)
 * Not clay but close to it ;-) See Sphere Made to Redefine Kilogram Has Purest Silicon Ever Created. Personally I'd prefer a kilogram sphere of pure carbon with a cubic crystal structure! NadVolum (talk) 10:05, 16 June 2023 (UTC)
 * The purpose of manufacturing that sphere, however, was never to have an improved prototype (to serve as an artifactual definition), but to have an object of a pure material whose mass conformed as precisely as could be measured to the then (and last) International Prototype, constructed in such a way that scientists could calculate the number of atoms. The point of the effort was that this would make it possible to get a more precise experimental determination of the value of the Planck constant to be used in its upcoming redefinition as a fixed constant. The choice of material served the aim of being able to calculate the the number of atoms. If the purpose of manufacturing a new object had been for it to serve as an improved IPK, there is no reason to think silicon would have been the preferred choice. See also Alternative approaches to redefining the kilogram. --Lambiam 08:55, 17 June 2023 (UTC)

Double Slit Experiment.
As we know, multiple single photons can be fired through 2 slits to form interference patterns as can electrons and small particles.

I believe there is no explanation for this - it is merely observed.

Has anyone ever suggested that a diffraction contour pattern is obviously formed on either side of the slits by the multitude of photons in a given room, all of whom diffract as we see from single photon diffraction? And thus, their energy/momentum can be imparted to electrons and small particles, thus creating the illusion that the particles themselves are diffracting? Byron Forbes (talk) 16:58, 13 June 2023 (UTC)


 * I don't follow what you're saying but the diffraction pattern is formed exactly the same even if the light intensity is so low that only a single photon at most will be in the experiment at a time. SO multitudes has nothing to do with it. NadVolum (talk) 17:07, 13 June 2023 (UTC)
 * Clarification: every single photon hits one and only one point. However, if you test photon by photon the pattern will emerge, as more photons will hit where the wave function superposition is positive, and vice versa.
 * Also, it's not exactly true to say that it is not explained. The Copenhagen interpretation is that the wave function collapses upon measurement (though what consists a measurement is not well defined), and the pilot wave interpretation explains it as the probability the wave function that directs the photon is more likely to lead it to certain areas and less to others.
 * Zarnivop (talk) 17:48, 13 June 2023 (UTC)
 * No physicist has suggested this, because it does not make sense. --Lambiam 20:34, 13 June 2023 (UTC)
 * What part of it makes no sense to you? It is very simple stuff. Byron Forbes (talk) 17:45, 14 June 2023 (UTC)
 * None of it makes sense after the first sentence. --Lambiam 22:23, 14 June 2023 (UTC)
 * LOL - what's so hard to understand about the multitude of photons in a given room, and on either side of a slit, forming a diffraction contour pattern on either side of the slit? Byron Forbes (talk) 11:36, 15 June 2023 (UTC)
 * All of it. You cannot expect random photons within a given space will have their waves interact to form constructive and destructive interference patterns that are orderly as it does in the double slit experiment. The frequencies of the waves will similarly be random, as will their directions, as will their phases. None of this will result in the orderly interference patterns of the double slit experiment. Also, where do you think these other photons in the room are coming from? --OuroborosCobra (talk) 13:02, 15 June 2023 (UTC)
 * Let me just have a laugh at your last sentence because you seemed to understand that from the rest of your post. Are you asking me where photons come from generally? :)
 * Anyway, it has nothing to do with fancy and discrete patterns on a wall - it is to do with the fact that every photon going through a slit will diffract, thus forming a diffraction contour pattern.
 * This pattern is not on a wall - it is a naturally occurring thing in the space on either side of the slit. Byron Forbes (talk) 14:56, 15 June 2023 (UTC)
 * Every photon in the room passing through the slits will continue following a probability distribution set by an interference pattern, but as all those photons have different wavelength and direction, all those interference patterns are different, so the photons as a group don't produce a useful interference pattern. PiusImpavidus (talk) 15:25, 15 June 2023 (UTC)
 * All of you are missing the very simple point here.
 * This is nothing to do with a diffraction "pattern" photographed on a wall.
 * It is the general diffraction contours made by the multitude of diffracted photons as they exit the slit.
 * So, for example, when an electron goes through a slit, if it is a little left of centre of the slit then it would generally be deflected left by the collisions it has with the photons that are travelling left due to diffraction. Byron Forbes (talk) 17:21, 15 June 2023 (UTC)
 * You appear to be here not to learn but to make a point, as you dismiss attempts to make some sense of your incoherent mishmash as evidence of our lack of understanding, and attempts at explanation as being laughably wrong or beside the point. You are wasting our time. Please stop. --Lambiam 19:41, 15 June 2023 (UTC)
 * kettle...............black Byron Forbes (talk) 13:19, 16 June 2023 (UTC)
 * You can form a pattern with one slit but we're talking about the two slit pattern here. And the question which arises there is which slit did the photon go through? Which is unanswerable because if you could find out which slit each photon went through it wouldn't form that diffraction pattern. NadVolum (talk) 17:06, 15 June 2023 (UTC)

Double-slit_experiment explains this. Philvoids (talk) 20:25, 15 June 2023 (UTC)


 * Not even a mention. Byron Forbes (talk) 13:18, 16 June 2023 (UTC)
 * The end of the "Overview" section has multiple examples of single electrons (or other such entities) interfering with themselves in the double-slit experiment. The "explanation" is that these things have a wave-like nature, so it is completely expected that they would have a wave-like behavior and does not require special explanation beyond that. The diffraction experiment with buckyballs specifically detected the buckyballs in the diffraction pattern, disproving your novel proposal that it's the photons or other materials beyond the slit that are responsible for the appearance of diffraction. DMacks (talk) 04:27, 18 June 2023 (UTC)
 * I'm yet to be convinced that any responders here have actually understood what I'm referring to. Byron Forbes (talk) 12:03, 18 June 2023 (UTC)