I've been off NGF for quite a while, trying to help Ukrainians win their country back, but a lull in my activities means I've been able to pick up some of my model threads once again and I'm doing some layout planning. Muwahahaha >:D
I want to learn more about the way some UK lines have banking in the curves, designed to improve the balance of trains - particularly fast ones - moving through the curve.
I've got two main questions:
1) What lines are typically superelevated, is it only fast mainlines, or is it actually quite common on your every-day branch line too?
1b) If many different kinds of lines have this sort of camber effect, was this something that might have been phased-in for some lines in one period (such as for the IC-125 routes) and was it then also rolled-out to other sorts of line as time went on?
2) And what sort of cant angles are typical used in typical superelevated sections? (really, I'm just trying to judge what might be good angles to use in my various designs).
If anyone has any knowledge to share on this subject, this Padawan would be most grateful to learn!
Ross.
That's a very good question, Ross. I don't have a lot of information but I'll start off and, hopefully, others more knowledgeable will add more.
Superelevation, or cant, allows for higher line speeds round curves and for increased passenger comfort. In conjunction with the coning of the wheel tyres, it helps prevent flange contact with the inside face of a rail.
(https://www.ngaugeforum.co.uk/SMFN/gallery/131/6222-090423093744.jpeg) (https://www.ngaugeforum.co.uk/SMFN/index.php?action=gallery;sa=view&id=131337)
Beal, Northumberland, on the East Coast Main Line with track engineered for high-speed running
I'm not sure when it was introduced but it was used in the 'Big Four' period. I recall the complex tables in a GWR book regarding radii in chains and superelevation in inches and fractions thereof. The high-speed running which was used (for some especially favoured trains!) pre-War would have been distinctly uncomfortable for passengers and potentially dangerous without it.
The amount of superelevation allowed also depends on the traffic on the route - higher amounts of superelevation can be inconvenient for slow goods trains. Very high speed passenger lines like HS1 are allowed higher maximum cant figures - I think about 7 inches is the maximum. About 6 inches applies on 'normal' UK lines, I believe. (Please, please, someone, correct me if this is wrong - I wish I had listened more attentively to certain conversations in the past!)
The superelevation is introduced gradually in the transition curve, which is one of the reasons why reverse curves are not good for high speeds - the amount of cant and transition curve is limited by the formation.
I understand greater superelevation came in from the 1960s with concrete sleepers, continuous welded rail and deep ballasting. Higher passenger train speeds were being complemented by a reduction in the number of slow freight trains and a lot less switch and crossing work.
What we tend to call 'points' can have modest superelevation on the 'normal' line but tend to have a strict speed restriction on the 'reverse' or diverging line as this introduces what is, effectively, a negative camber to that track. In bullhead rail days, there were many fantastic types of rail chairs available to construct the common crossing.
Cant gives passengers a much more comfortable ride at speed but can be amusing if one's train is stopped at a signal on a superelevated curve - it can be very noticeable. Likewise, where track though a station platform has cant. You can often see the PW engineers' paint markings on the opposite platform face.
With regard to branch lines, I expect it depends on era. I'm not sure there was a lot of superelevation used in steam days but on the remaining passenger branch lines I understand it is in use.
I hope this helps and look forward to reading more erudite contributions.
With best wishes.
John
I can't be sure but would have thought dual running lines could not have one superelevated and one not, as traffic on one line would be canted into the path of oncoming traffic? :worried:
That's a superb answer John, thank you!
So 6 inches maximum cant translates to about 1mm on my track. Subtle, but worth considering during the planning stage. I expect to need quite a lot more ballast to do it right though - given what I saw when Richard did superelevation on Everard Junction (albeit in OO).
You also mention the transition curve: I'm going to have to figure out some good rules for modelling those in N too!
And that's a good point about all parallel lines needing to be canted together, you really wouldn't want to get into any situation where one train clips another accidentally! I guess that if I model a 4-line stretch, with two fast and two slow lines, all four should be canted in the same fashion.
Ross.
It's done on mainlines and all the running lines will be super-elevated. But not enough so that trains fall over when stopped on a curve :doh:
The super elevation is very noticeable, inside the carriages, if a passenger train stops on a curve.
Search for Queensville Curve Stafford, WCML, 4 tracks, with super-elevation.
Regards,
John P
It's in Japanese (there are some English subtitles), but this video shows how one modeller reproduces a cant:
The cant part starts at about 4:40 in.
I did it with some 1mm strips of card under the outside rail. The effect was subtle, but looked good to me. I made no concessions to transitions. Dan from Great Deltan on YouTube has done it very effectively in N. IIRC he used the same thickness.
(https://www.ngaugeforum.co.uk/SMFN/gallery/131/1147-090423214903.jpeg) (https://www.ngaugeforum.co.uk/SMFN/index.php?action=gallery;sa=view&id=131387)
You definitely need to do all lines together. It wouldn't ordinarily happen on branch lines, the speeds simply aren't high enough.
Quote from: njee20 on April 09, 2023, 09:47:47 PM
I did it with some 1mm strips of card under the outside rail. The effect was subtle, but looked good to me. I made no concessions to transitions. Dan from Great Deltan on YouTube has done it very effectively in N. IIRC he used the same thickness.
(https://www.ngaugeforum.co.uk/SMFN/gallery/131/1147-090423214903.jpeg) (https://www.ngaugeforum.co.uk/SMFN/index.php?action=gallery;sa=view&id=131387)
You definitely need to do all lines together. It wouldn't ordinarily happen on branch lines, the speeds simply aren't high enough.
That's exactly the way I was thinking about doing this. Good to know I'm on the right path and in good company!
Unfortunately, the track planning software I've been using, SCARM, can't do transition curves, and it doesn't do bezier curves so I can't even manually implement them, so I'm looking at alternatives. AnyRail can do TC's but a bit of testing this afternoon suggests there are other issues with it, so I'm still looking around.
Happy Easter all!
Ross.
I think you're overthinking it. I meant the transition from canted to flat, but yes you ideally want to ease into curves from straight to curved. Better still simply avoid the 'trainset' approach of straight into an abrupt curve. What are you finding Anyrail doesn't do?
You could try Xtrkcad, which IIRC is a bit more 'technical' and may better suit your needs. I know at various points you've talked about making your own track, in which case Templot is the only program to use.
Quote from: njee20 on April 09, 2023, 09:47:47 PM
. . . It wouldn't ordinarily happen on branch lines, the speeds simply aren't high enough.
It certainly exists on 'slower' lines. The cant may be not as much as the high-speed mains, but it will be there. E.g. on the Severn Beach line:
(https://fosbr.org.uk/wp-content/uploads/2020/10/sea_mills_pacer.jpg)
(Image taken from https://fosbr.org.uk/the-severn-beach-line-a-poem/)
John Armstrong's Track Planning for Realistic Operation: Prototype Railroad Concepts for Your Model Railroad is USA practice, of course, but has quite a lot of technical detail on easements. Its fairly easily obtainable, I believe, and is an interesting read in itself.
This guy (also American, but the principles still apply) used masking tape: https://www.youtube.com/watch?v=Nu9qlIckU7Q (https://www.youtube.com/watch?v=Nu9qlIckU7Q). Easements have to be vertical as well as horizontal.
Quote from: njee20 on April 09, 2023, 09:47:47 PM
I did it with some 1mm strips of card under the outside rail. The effect was subtle, but looked good to me. I made no concessions to transitions. Dan from Great Deltan on YouTube has done it very effectively in N. IIRC he used the same thickness.
(https://www.ngaugeforum.co.uk/SMFN/gallery/131/1147-090423214903.jpeg) (https://www.ngaugeforum.co.uk/SMFN/index.php?action=gallery;sa=view&id=131387)
You definitely need to do all lines together. It wouldn't ordinarily happen on branch lines, the speeds
simply aren't high enough.
1mm is the equivalent of 148mm; I'm not up to date with modern standards but 150mm was about the limit on BR in the Deltic/HST era, e.g. at Hatfield. I experimented with cant on a previous layout and found steam locos weren't happy with 1mm, so I settled on approximately half that using Evergreen strip under the outer rail; it still gives the desired effect.
Kato does it for you?
https://www.youtube.com/watch?v=PVl84OQs820 (https://www.youtube.com/watch?v=PVl84OQs820)
not suitable for everyone...
I use the Kato double track with superelevation as my test circuit, and I found that Revolution's Sturgeon wagons would uncouple on the transitions, which I eventually attributed to the fact that they are so long that one was at full tilt while the one behind was still flat, causing too much twist on the couplings. Just something else to chuck into the calculations.
Somebody else mentioned trains coming to a stand on a steeply canted curve and in my experience it can be quite disconcerting if you do not know why one side of the carriage is noticeably higher than the other. Due to regular congestion in the evening peak out of Kings Cross and Moorgate in the 1980s and 1990s, it was not unusual for trains to be brought to a stand just south of Hatfield station on a long curve from south of Welham Green to Hatfield station. The regular travellers understood it was a daily occurrence and nothing for them to be concerned about. Just to complicate things there was a (Kellogg's) private siding which joined the down slow at Welham Green. This was close to the site of the infamous Hatfield derailment in October 2000 which brought to light "gauge corner cracking", resulting in long-lasting blanket speed restrictions being applied across large parts of the Railtrack network.
Xi
Quote from: Yet_Another on April 10, 2023, 02:21:18 PM
I use the Kato double track with superelevation as my test circuit, and I found that Revolution's Sturgeon wagons would uncouple on the transitions, which I eventually attributed to the fact that they are so long that one was at full tilt while the one behind was still flat, causing too much twist on the couplings. Just something else to chuck into the calculations.
Did you use the transition rail piece ( in video) to link flat and superelevated sections?
Quote from: Railwaygun on April 12, 2023, 06:20:53 PM
Did you use the transition rail piece ( in video) to link flat and superelevated sections?
Yes, you pretty much have to, because the canted tracks on the superelevated curves don't connect to straight flat sections at all well.
Interesting that only cant for speed is mentioned but several question slower line speeds. One of the main reasons for cant is to balance the rolling radius on both wheels of the same axle.
With a solid axle both wheels would turn at the same RPM so the 'tyres' are chamfered with the larger diameter adjacent to the flange. The flanges are not fixed at gauge distance apart but slightly under so the axles can 'float' between the rails.
The chamfered tyres in straight line running keep the wheels from hunting side to side together with the rounded rail top.
On a bend the rolling stock is forced outwards and this causes the wheels to move outwards too. The outer wheel now runs on the larger diameter adjacent to the flange whilst the inner wheel runs towards the track centre and contact with the rail is by the smaller diameter of the tyre.
In this way the wheels on the axle have different rolling radii to stop friction and squealing on the curve - mostly!
This diagram is from 'Curving Behaviour' on Science Direct website applicable to a right hand curve.
(https://www.ngaugeforum.co.uk/SMFN/gallery/148/8822-180125152156.jpeg) (https://www.ngaugeforum.co.uk/SMFN/index.php?action=gallery;sa=view&id=148618)
Believe On LU circle line trains used to be turned end for end on the Gloucester Road triangle to balance out the wheel wear.
Quote from: Safety Engineer on January 18, 2025, 04:50:21 PMBelieve On LU circle line trains used to be turned end for end on the Gloucester Road triangle to balance out the wheel wear.
Yep, I also believe that's true
(unless it's an urban myth...!).
Mind you, there's also the Aldgate triangle as well, to turn the units. It was probably not that difficult to diagram into the Working Timetable, given that C69//C77 stock also worked on the H&C and Wimbledon-Edgware Road services. A set of late evening ECS move(s) would surely achieve that.
I would guess that it still happens, even though the Circle Line is not strictly a circle; and now that S7 stock also operates the main District Line. It would also be easier to diagram, as well, I would imagine.
Quote from: Greygreaser on January 18, 2025, 03:25:40 PMIn this way the wheels on the axle have different rolling radii to stop friction and squealing on the curve - mostly!
Yep, true. I distinctly remember the 'squeal' from the wheels on the section between London Bridge and Waterloo East; also Victoria SE - and being pleased that I knew what the noise was (and what was causing it...!).
To get back 'on track' and 'on topic'... :D
...thank you (all) for this thread, as this is something I would seriously like to do, when I get round to building something myself. I have been wondering about some of the pitfalls referred to above, so will bear them in mind. I did once see an N gauge layout which had done this - and felt that they hadn't tilted (or canted) the track enough. But may be they had done it, just enough.
Although Kato track is not really for me, I do like the way they have gone about it, complete with transitional curves. :heart2:
Charles.