• Our booking engine at tickets.railforums.co.uk (powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Loop Lengths

Status
Not open for further replies.

Legolash2o

Member
Joined
27 Sep 2018
Messages
601
How do they measure the maximum length of a train that can go into a loop?

If a loop is physically 400m in length from point to point, would the max train length be 350-380m? How is it calculated?
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

30907

Veteran Member
Joined
30 Sep 2012
Messages
17,867
Location
Airedale
How do they measure the maximum length of a train that can go into a loop?

If a loop is physically 400m in length from point to point, would the max train length be 350-380m? How is it calculated?
Essentially, the train has to stop at the signal protecting the exit from the loop AND to be sufficiently clear at the entrance to allow a train to be signalled past on the main line. The exact relationship between that and the physical length of the loop will vary according to the signalling, but you are right in saying that there's a difference.

PS my 1960 Sectional Appendices show loop lengths as "N wagons plus (engine and) brake van" not in yards.
 

CW2

Established Member
Joined
7 May 2020
Messages
1,922
Location
Crewe
How do they measure the maximum length of a train that can go into a loop?

If a loop is physically 400m in length from point to point, would the max train length be 350-380m? How is it calculated?
There are several factors to consider, including:
- What is the line speed of the loop?
- Where is the signal controlling loop exit located?
- Where are the track circuits located (at both entry and exit ends of the loop)?
- What gradients are involved?
- Is the loop unidirectional or bidirectional?
- Are there any trap points involved?

In recent years, defensive driving policies have resulted in trains crawling slowly along loops before coming to a stand well short of the signal. It's safer that way, but it does eat into capacity (both line capacity and train length).

A point-to-point length of 400m would often produce a usable length of less than 350m.
 

Dr Hoo

Established Member
Joined
10 Nov 2015
Messages
3,912
Location
Hope Valley
How do they measure the maximum length of a train that can go into a loop?

If a loop is physically 400m in length from point to point, would the max train length be 350-380m? How is it calculated?
There is a big difference between 'how long could a train be and still (just) tuck its tail in beyond the fouling point so that the entry points can eventually be re-set for the main line' and 'how long would a loop be if you wanted to be able to get a maximum length train to enter briskly but safely so that a following service can overtake as soon as possible'.

So ideally you would want something like 75mph entry and exit points, a full overlap beyond the exit signal, a couple of hundred yards/metres to spare to allow for a 'slow'/professional driving run-up and a reasonable 'stand off' for good signal viewing.

Something like 775m (train with two locomotives) + 200m overlap + 200m slow approach + 25m stand off = 1,200m would do nicely, thank you very much. You then discover that there are level crossings, station platforms, viaducts, tunnels, overhead line neutral sections, awkward gradients, curvature, National Parks, Areas of Outstanding Natural Beauty, Sites of Special Scientific Interest, National Trust land rare bats, newts, orchids, etc. in the way but you get the general idea.
 

FGW_DID

Established Member
Joined
23 Jun 2011
Messages
2,724
Location
81E
Each loop or siding will have a maximum SLU - (Standard Length Unit) it can accommodate.

1 SLU = 21 feet

As long as you know the SLU of the train, you’ll know if it can fit.
 

The Planner

Veteran Member
Joined
15 Apr 2008
Messages
15,838
There is a big difference between 'how long could a train be and still (just) tuck its tail in beyond the fouling point so that the entry points can eventually be re-set for the main line' and 'how long would a loop be if you wanted to be able to get a maximum length train to enter briskly but safely so that a following service can overtake as soon as possible'.

So ideally you would want something like 75mph entry and exit points, a full overlap beyond the exit signal, a couple of hundred yards/metres to spare to allow for a 'slow'/professional driving run-up and a reasonable 'stand off' for good signal viewing.

Something like 775m (train with two locomotives) + 200m overlap + 200m slow approach + 25m stand off = 1,200m would do nicely, thank you very much. You then discover that there are level crossings, station platforms, viaducts, tunnels, overhead line neutral sections, awkward gradients, curvature, National Parks, Areas of Outstanding Natural Beauty, Sites of Special Scientific Interest, National Trust land rare bats, newts, orchids, etc. in the way but you get the general idea.
You would need a lot longer than 1200m for a 75mph entry and exit! 40mph entry with flashing yellows and 40mph exit is the target speed currently in most loop upgrades.
 

pdeaves

Established Member
Joined
14 Sep 2014
Messages
5,632
Location
Gateway to the South West
Signalling design takes account of the clearing/clearance point. The train has to clear the relevant track circuit to be known to be inside the loop. The available length can then be considered to be where that track circuit ends (to all practical purposes).
 

GB

Established Member
Joined
16 Nov 2008
Messages
6,457
Location
Somewhere
There is a big difference between 'how long could a train be and still (just) tuck its tail in beyond the fouling point so that the entry points can eventually be re-set for the main line' and 'how long would a loop be if you wanted to be able to get a maximum length train to enter briskly but safely so that a following service can overtake as soon as possible'.

So ideally you would want something like 75mph entry and exit points, a full overlap beyond the exit signal, a couple of hundred yards/metres to spare to allow for a 'slow'/professional driving run-up and a reasonable 'stand off' for good signal viewing.

Something like 775m (train with two locomotives) + 200m overlap + 200m slow approach + 25m stand off = 1,200m would do nicely, thank you very much. You then discover that there are level crossings, station platforms, viaducts, tunnels, overhead line neutral sections, awkward gradients, curvature, National Parks, Areas of Outstanding Natural Beauty, Sites of Special Scientific Interest, National Trust land rare bats, newts, orchids, etc. in the way but you get the general idea.

You are looking at least two miles loop length for that!
 

The Planner

Veteran Member
Joined
15 Apr 2008
Messages
15,838
Signalling design takes account of the clearing/clearance point. The train has to clear the relevant track circuit to be known to be inside the loop. The available length can then be considered to be where that track circuit ends (to all practical purposes).
Surely made easier with axle counters?
 

Legolash2o

Member
Joined
27 Sep 2018
Messages
601
Wow, thanks for all the answers! I was thinking they'll be a rule of placing a signal 20m from the exit point and have the train within 20m passed the entry point to ensure the train was fully in. Essentially 400m minus 40m, making an effective max train length of 360m). Didn't even consider the speed of entry and everything else mentioned.
 

The Planner

Veteran Member
Joined
15 Apr 2008
Messages
15,838
Why 40mph? Is that the best trade-off between cost and "usefulness"?
As above, make it quicker and you need a substantial increase to the length of the loop. There will be an element of cost too due to the size of the S&C. You aren't stopping 1600t of Intermodal in a hurry from 75mph...
 

Dr Hoo

Established Member
Joined
10 Nov 2015
Messages
3,912
Location
Hope Valley
It is worth noting that to maximise the usefulness and flexibility of a loop it is sensible to allow for 'run through' at reasonable speed. Quite a few times I have been on trains that have had a problem and the driver has judiciously coasted to a halt alongside a loop so that following services can still pass. (I appreciate that with radio this may be less critical than it was in some cases.)

The other thing that can be useful is to anticipate some passenger and ancillary use too. Empty stock, railhead treatment trains, light locomotives and so on can all be 'popped in' and may well have rather different operational and performance characteristics than a 775m, 2,000+ tonnes intermodal.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,232
Location
Torbay
It is worth noting that to maximise the usefulness and flexibility of a loop it is sensible to allow for 'run through' at reasonable speed. Quite a few times I have been on trains that have had a problem and the driver has judiciously coasted to a halt alongside a loop so that following services can still pass. (I appreciate that with radio this may be less critical than it was in some cases.)

The other thing that can be useful is to anticipate some passenger and ancillary use too. Empty stock, railhead treatment trains, light locomotives and so on can all be 'popped in' and may well have rather different operational and performance characteristics than a 775m, 2,000+ tonnes intermodal.
Many former goods-only loops have been upgraded to passenger status in recent years. For new schemes, unless engineers and operators can make a good case otherwise, signalling on freight lines should provide all the same safeguards as that on passenger lines by default. In the past, freight lines could legitimately get away without overlaps and facing point locks.

A quick scan through various Sectional Appendices reveals Table A track schematics are is always annotated with lengths of goods loops, in both metres and either feet or SLUs
 

CEN60

Member
Joined
17 Dec 2018
Messages
266
To put it in context - for an 50mph turnout from straight track - from the points / switch toe to the location where the tracks would be back to a standard 1970 sixfoot is around 110m, but as mentioned above - the flashy light brigade have weird rules that also contribute to the usable length" of a loop.

Pway . Track observation - "Put 2 signaling engineers in a room and you get 3 opinions (sometimes 4)"
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,232
Location
Torbay
Pway . Track observation - "Put 2 signaling engineers in a room and you get 3 opinions (sometimes 4)"
But one thing both would probably agree on is that the initially submitted pway design needs to change in some way to meet the operational requirements!
 

Merle Haggard

Established Member
Joined
20 Oct 2019
Messages
1,979
Location
Northampton
On a single line, a train doesn't have to fit in a loop to be passed by a train in the opposite direction, providing that only one of the two trains is longer.

Looking back to around 25 - 30 years ago, so my memory may not be entirely correct; but Automotive car-carrying trains were 100 s.l.u. or slightly more (if 10 Cartics + 1 Autic). I have a distinct recollection that one of the services (to Southampton?) was routed Coventry - Leamington and would use Kenilworth loop to pass trains in the opposite direction. They were too long for the loop, but this was not normally a problem because the car train would come to a stand in the loop, the other, shorter, would train draw up to the starter - obviously, with the tail of the car train in front, turning on to the loop. The tail of this train would be clear of the exit points for the car train, and, he line forward having then been cleared, the car train could then move off, and when its tail had cleared the entry points, points changed and starter cleared for the other train to set off, too.

Both loaded and return empties services followed this route, and the process remains clear in my memory because of the chaos that was caused when the loaded service passed Coventry and was sent forward at about the same time as the (very late running) returning empty service presented itself at Leamington and was, without thinking things through, also sent forward. We tried to keep a low profile while Railtrack worked out a solution - I never did know what it was, but it couldn't have been easy.

As an aside, the Harwich P.Q. trains were of similar length, and exceeded the lengths of most loops between there and Stratford, so they always had a clear run!
 

Oxfordblues

Member
Joined
22 Dec 2013
Messages
651
I remember attending a meeting in 1993 to discuss options for running 775m-long Channel Tunnel freight trains. We were told by the operators that the maximum length of a train was determined by the length of the shortest loop. One example given was Oubeck DGL, on the WCML south of Lancaster. A quick check of the WTT revealed that no down trains were booked to be looped there, so one of the commercial people suggested that the simplest answer was to take it out of use. The operators were insistent that it be retained "for strategic reasons" (without specifying what these were). The only solution was that any train longer than Oubeck DGL would have to be classified as exceptional with an "X" headcode to remind the signallers. In the event the traffic never materialised but it was a measure of what we were up against!
 

DelW

Established Member
Joined
15 Jan 2015
Messages
3,834
Both loaded and return empties services followed this route, and the process remains clear in my memory because of the chaos that was caused when the loaded service passed Coventry and was sent forward at about the same time as the (very late running) returning empty service presented itself at Leamington and was, without thinking things through, also sent forward. We tried to keep a low profile while Railtrack worked out a solution - I never did know what it was, but it couldn't have been easy.
It is feasible to cross two trains at a loop, even when both are longer than the loop, but I doubt the manoeuvre complies with UK regulations. It needs one train to be divided, and needs several uncouplings and recouplings and reverse moves, including one with the loco buried in the middle of a consist of the whole of one train and part of the other.
I believe it's sometimes done in unsignalled areas of the US and maybe elsewhere though.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,232
Location
Torbay
It is feasible to cross two trains at a loop, even when both are longer than the loop, but I doubt the manoeuvre complies with UK regulations. It needs one train to be divided, and needs several uncouplings and recouplings and reverse moves, including one with the loco buried in the middle of a consist of the whole of one train and part of the other.
I believe it's sometimes done in unsignalled areas of the US and maybe elsewhere though.
It's like one of those seemingly impossible shunting puzzles. One train splits on approach leaving a rear cut on the single line then its loco draws the forward portion fully into the loop. The second train now passed that and couples up with the cut left behind. The forward portion of the first train is now released and must proceed out of the loop a distance sufficient to allow the second train to shunt behind it. The second train now reverses back through the loop hauling the rear part of the first train which it uncouples and leaves in the loop. It can now pass this and continue on its forward journey. The first train then reverses back into the loop to pick up its rear cut. This is all somewhat time consuming!
 

ABB125

Established Member
Joined
23 Jul 2016
Messages
3,746
Location
University of Birmingham
It's like one of those seemingly impossible shunting puzzles. One train splits on approach leaving a rear cut on the single line then its loco draws the forward portion fully into the loop. The second train now passed that and couples up with the cut left behind. The forward portion of the first train is now released and must proceed out of the loop a distance sufficient to allow the second train to shunt behind it. The second train now reverses back through the loop hauling the rear part of the first train which it uncouples and leaves in the loop. It can now pass this and continue on its forward journey. The first train then reverses back into the loop to pick up its rear cut. This is all somewhat time consuming!
Sounds fun! But I really can't see it happening in Britain...
 

DelW

Established Member
Joined
15 Jan 2015
Messages
3,834
It's like one of those seemingly impossible shunting puzzles. One train splits on approach leaving a rear cut on the single line then its loco draws the forward portion fully into the loop. The second train now passed that and couples up with the cut left behind. The forward portion of the first train is now released and must proceed out of the loop a distance sufficient to allow the second train to shunt behind it. The second train now reverses back through the loop hauling the rear part of the first train which it uncouples and leaves in the loop. It can now pass this and continue on its forward journey. The first train then reverses back into the loop to pick up its rear cut. This is all somewhat time consuming!
That's a more concise explanation than I'd have managed, but it's the sequence I'd envisaged too. As ABB125 said, I can't imagine it being allowed here outside very exceptional circumstances.
 

alexl92

Established Member
Joined
12 Oct 2014
Messages
2,268
How long can a loop be before it's just considered as double track line instead?
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
24,793
Location
Nottingham
It's like one of those seemingly impossible shunting puzzles. One train splits on approach leaving a rear cut on the single line then its loco draws the forward portion fully into the loop. The second train now passed that and couples up with the cut left behind. The forward portion of the first train is now released and must proceed out of the loop a distance sufficient to allow the second train to shunt behind it. The second train now reverses back through the loop hauling the rear part of the first train which it uncouples and leaves in the loop. It can now pass this and continue on its forward journey. The first train then reverses back into the loop to pick up its rear cut. This is all somewhat time consuming!
Google for "Railroad double saw by".
 

CEN60

Member
Joined
17 Dec 2018
Messages
266
But one thing both would probably agree on is that the initially submitted pway design needs to change in some way to meet the operational requirements!
Nope - never heard 2 of the railway traffic light designers agree on anything - except that they disagree :)
 

The Planner

Veteran Member
Joined
15 Apr 2008
Messages
15,838
On a single line, a train doesn't have to fit in a loop to be passed by a train in the opposite direction, providing that only one of the two trains is longer.

Looking back to around 25 - 30 years ago, so my memory may not be entirely correct; but Automotive car-carrying trains were 100 s.l.u. or slightly more (if 10 Cartics + 1 Autic). I have a distinct recollection that one of the services (to Southampton?) was routed Coventry - Leamington and would use Kenilworth loop to pass trains in the opposite direction. They were too long for the loop, but this was not normally a problem because the car train would come to a stand in the loop, the other, shorter, would train draw up to the starter - obviously, with the tail of the car train in front, turning on to the loop. The tail of this train would be clear of the exit points for the car train, and, he line forward having then been cleared, the car train could then move off, and when its tail had cleared the entry points, points changed and starter cleared for the other train to set off, too.

Both loaded and return empties services followed this route, and the process remains clear in my memory because of the chaos that was caused when the loaded service passed Coventry and was sent forward at about the same time as the (very late running) returning empty service presented itself at Leamington and was, without thinking things through, also sent forward. We tried to keep a low profile while Railtrack worked out a solution - I never did know what it was, but it couldn't have been easy.

As an aside, the Harwich P.Q. trains were of similar length, and exceeded the lengths of most loops between there and Stratford, so they always had a clear run!

I would be amazed if we got away with planning something like that now, I certainly wouldn't try it. Signaller would intervene and not run it like that and delay minutes aplenty would ensue.
I remember attending a meeting in 1993 to discuss options for running 775m-long Channel Tunnel freight trains. We were told by the operators that the maximum length of a train was determined by the length of the shortest loop. One example given was Oubeck DGL, on the WCML south of Lancaster. A quick check of the WTT revealed that no down trains were booked to be looped there, so one of the commercial people suggested that the simplest answer was to take it out of use. The operators were insistent that it be retained "for strategic reasons" (without specifying what these were). The only solution was that any train longer than Oubeck DGL would have to be classified as exceptional with an "X" headcode to remind the signallers. In the event the traffic never materialised but it was a measure of what we were up against!
That hasn't changed, if something doesn't get used for a while there is always a bright spark that wants to rip it out.
 

RSimons

Member
Joined
15 Dec 2016
Messages
62
Location
Alberta
On a single line, a train doesn't have to fit in a loop to be passed by a train in the opposite direction, providing that only one of the two trains is longer.

Looking back to around 25 - 30 years ago, so my memory may not be entirely correct; but Automotive car-carrying trains were 100 s.l.u. or slightly more (if 10 Cartics + 1 Autic). I have a distinct recollection that one of the services (to Southampton?) was routed Coventry - Leamington and would use Kenilworth loop to pass trains in the opposite direction. They were too long for the loop, but this was not normally a problem because the car train would come to a stand in the loop, the other, shorter, would train draw up to the starter - obviously, with the tail of the car train in front, turning on to the loop. The tail of this train would be clear of the exit points for the car train, and, he line forward having then been cleared, the car train could then move off, and when its tail had cleared the entry points, points changed and starter cleared for the other train to set off, too.

Both loaded and return empties services followed this route, and the process remains clear in my memory because of the chaos that was caused when the loaded service passed Coventry and was sent forward at about the same time as the (very late running) returning empty service presented itself at Leamington and was, without thinking things through, also sent forward. We tried to keep a low profile while Railtrack worked out a solution - I never did know what it was, but it couldn't have been easy.

As an aside, the Harwich P.Q. trains were of similar length, and exceeded the lengths of most loops between there and Stratford, so they always had a clear run!
I used to live in a town on the route of the train to Churchill, Manitoba. The train was regularly late because the loops (called sidings here) are often only 3000 feet long or thereabouts while the freight trains can be over 10,000 feet long so the passenger train was regularly forced to wait for the oncoming freight.

Retired railway employees told me that two freights could pass even though they were both longer than the siding by a set of movements called ‘seesawing’.
 

MichaelAMW

Member
Joined
18 Jun 2010
Messages
1,011
Sounds fun! But I really can't see it happening in Britain...
But we did sometimes once put a siding at one or other end, or both ends, of the loop, to allow shunting. This was also done on loops on double-track lines to allow trains to be recessed that were longer that the maximum loop than could be worked from one signal box, e.g. Haddenham.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
6,232
Location
Torbay
But we did sometimes once put a siding at one or other end, or both ends, of the loop, to allow shunting. This was also done on loops on double-track lines to allow trains to be recessed that were longer that the maximum loop than could be worked from one signal box, e.g. Haddenham.
Also potentially useful as a bolthole for a crippled wagon or engineering vehicle.
 

Joseph_Locke

Established Member
Joined
14 Apr 2012
Messages
1,878
Location
Within earshot of trains passing the one and half
"A loop has only one signal section - two signal sections or more is a slow line." - R. Sumner.

As to loop entry and exit speeds, most freight can reach over 40mph (on the flat) in the length of the train, so 50 turnouts are a better choice, IMHO. Therefore the toe to toe length of the loop is the toe-to-CP for the entrance T/O +, 775m (train) + 2x20m (loco and demic) + 5m (inaccurate stopping) + 20m signal standback +a signal overlap (nominally 180m, can be a bit shorter at 50mph) +CP-to-toe for the exit T/O.

The more signalling engineers you put in a room the bigger the bill ...
 
Status
Not open for further replies.

Top