Track Circuits and Mansell Wheels

DerekC

Established Member
Joined
26 Oct 2015
Messages
1,193
Location
Hampshire (nearly a Hog)
Britain was a very slow adopter of the track circuit. The first fail-safe track circuits were developed in the USA in the 1870s, but it wasn't until after WWI that railways in Britain started to use them widely. One reason I have seen given for this is the use of the Mansell wheel, which had a steel axle and tyre but a wooden wheel centre - so wouldn't have operated a track circuit. Does anyone know which railways used these wheels and for how long?
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

Cowley

Established Member
Associate Staff
General Discussion
Railtours & Preservation
Modelling & Games
Joined
15 Apr 2016
Messages
8,600
Location
Devon
I must admit that I’d never heard of these wheels before. They look like a lovely bit of engineering.
There’s a (very brief) overview on Wikipedia here with a good picture that’s worth a look. It says that copper bonding was used to overcome the problem, but doesn’t say much else.



The Mansell Wheel is a railway wheelpatented by Richard Mansell, the Carriage and Wagon superintendent of the South Eastern Railway in the UK.[1] The design was created in the 1840s and was eventually widely used on passenger railway stock in the UK. This is an interesting example of a composite wooden wheel, using the same principle as an artillery wheel but with a solid wooden centre instead of spokes. The drawing (right) is from an old railway design book[2] from the early 20th Century.
Edit - Have a look at this post on rmweb:

It seems to show a van with those wheels still running in the 1970s.
There were plenty of Big Four parcels vans kicking about in that era, especially on the SR.
 
Last edited:

dubscottie

Member
Joined
4 Apr 2010
Messages
813
I read about a later version that used bonded paper. It was common on UK pullman and LNER stock. IIRC the parts were bolted together but I cannot find the name for them now. Engines didn't use them so a train would still be detected.
 

big all

On Moderation
Joined
23 Sep 2018
Messages
826
Location
redhill
I was a railwayman on the platform and in the signal box until the early 70s and then a second man and driver from 1972 onwards and yes there were all sorts of parcel vans at Redhill but no van ever stood out, other than a six wheeled Midland, and never for having wooden centres and I was super observant on technical points. Not saying it never happened but if it didn't look like a disk wheel I would have noticed.
 

Taunton

Established Member
Joined
1 Aug 2013
Messages
5,006
You could consider the same problem with Resilient Wheels, those with a rubber ring inside the steel tyre, which seem to have been developed in the 1930s for USA PCC streetcars, and then spread to various mainstream rail vehicles. The German ICE Eschede major accident arose from a resilient wheel breaking up.

Most of the "big four" vehicles left by the 1960s-70s were actually built postwar, and indeed well into the BR era, just to the former company designs.
 
Last edited:

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
19,170
Location
Nottingham
I read about a later version that used bonded paper. It was common on UK pullman and LNER stock. IIRC the parts were bolted together but I cannot find the name for them now. Engines didn't use them so a train would still be detected.
The important thing for the signaling is to detect the rear of the train. Just detecting the engine might be OK in the early years of track circuits when they were seen as supplementing visual observation by the signalman, but not later on when they came to be the sole means of detection.

I assume by that time any surviving Mansell wheels would have had some electrical connection between the tyre and the axle.
 

John Webb

Established Member
Joined
5 Jun 2010
Messages
1,884
Location
St Albans
Britain was a very slow adopter of the track circuit. The first fail-safe track circuits were developed in the USA in the 1870s, but it wasn't until after WWI that railways in Britain started to use them widely. One reason I have seen given for this is the use of the Mansell wheel, which had a steel axle and tyre but a wooden wheel centre - so wouldn't have operated a track circuit. Does anyone know which railways used these wheels and for how long?
According to Simmons and Biddle in "The Oxford Companion to British Railway History" (OUP, 1997) the Mansell wheel, first patented in 1848, consisted of 16 teak segments forming the disc built up around a metal boss. The disc was forced onto the bevelled inner face of the metal tyre, which had two grooves in the tyre which took metal rings and everything was bolted together. This type of wheel stopped being produced around the start of WW1. The entry does not say which companies used this wheel design.
 

matchmaker

Member
Joined
8 Mar 2009
Messages
1,038
Location
Central Scotland
According to Simmons and Biddle in "The Oxford Companion to British Railway History" (OUP, 1997) the Mansell wheel, first patented in 1848, consisted of 16 teak segments forming the disc built up around a metal boss. The disc was forced onto the bevelled inner face of the metal tyre, which had two grooves in the tyre which took metal rings and everything was bolted together. This type of wheel stopped being produced around the start of WW1. The entry does not say which companies used this wheel design.
Don't think the second last sentence can be correct - the LMS used Mansell wheels on many of their sleeping cars, and the last sleeping cars to an LMS design were built in 1951/52. The 12 wheeled 1st class sleepers certainly had Mansell wheels.
 

Richard Scott

Member
Joined
13 Dec 2018
Messages
1,100
I read about a later version that used bonded paper. It was common on UK pullman and LNER stock. IIRC the parts were bolted together but I cannot find the name for them now. Engines didn't use them so a train would still be detected.
What about a divided train, the one part may 'disappear'?
 

MarkyT

Established Member
Joined
20 May 2012
Messages
4,627
Location
Torbay
Britain was a very slow adopter of the track circuit. The first fail-safe track circuits were developed in the USA in the 1870s, but it wasn't until after WWI that railways in Britain started to use them widely. One reason I have seen given for this is the use of the Mansell wheel, which had a steel axle and tyre but a wooden wheel centre - so wouldn't have operated a track circuit. Does anyone know which railways used these wheels and for how long?
That is an interesting claim from the Wikipedia page, but I don't think it stands up to much scrutiny. As others have said, it was easy to arrange metal bond straps to maintain electrical continuity across the wooden components where track circuits were in use. I think the slow adoption in UK, and elsewhere, was mainly because they simply weren't needed with the typically closely-spaced boxes, where visual observation of passing trains sufficed. Most boxes had at least some points that couldn't be operated remotely from elsewhere at the time, and which could be secured by locking bars where required to prevent movement under trains. More widespread application of electrical techniques, including track circuits, only started after WW1, when improvements to block circuit security were developed and signalbox control areas began to expand beyond the range of direct sight, allowing many smaller block posts to be closed, and large multi-box station installations to be consolidated. There were some early niche applications before this such as underground railways.
 

John Webb

Established Member
Joined
5 Jun 2010
Messages
1,884
Location
St Albans
Don't think the second last sentence can be correct - the LMS used Mansell wheels on many of their sleeping cars, and the last sleeping cars to an LMS design were built in 1951/52. The 12 wheeled 1st class sleepers certainly had Mansell wheels.
I'm no expert on coaching stock. Is it possible old Mansell wheels were used? And while the last sleeping cars to an LMS design were built 1951/2, might have they been put onto more modern bogies? (I'm bearing in mind the greater complexity of the Mansell wheel, the cost of teak etc.)
 

DerekC

Established Member
Joined
26 Oct 2015
Messages
1,193
Location
Hampshire (nearly a Hog)
That is an interesting claim from the Wikipedia page, but I don't think it stands up to much scrutiny. As others have said, it was easy to arrange metal bond straps to maintain electrical continuity across the wooden components where track circuits were in use. I think the slow adoption in UK, and elsewhere, was mainly because they simply weren't needed with the typically closely-spaced boxes, where visual observation of passing trains sufficed. Most boxes had at least some points that couldn't be operated remotely from elsewhere at the time, and which could be secured by locking bars where required to prevent movement under trains. More widespread application of electrical techniques, including track circuits, only started after WW1, when improvements to block circuit security were developed and signalbox control areas began to expand beyond the range of direct sight, allowing many smaller block posts to be closed, and large multi-box station installations to be consolidated. There were some early niche applications before this such as underground railways.
That's a very coherent explanation which I am sure was the view of railway management at the time, but sadly the history of accidents suggests that visual observation often didn't suffice. Hawes Junction (1910), Pontypridd (1911), Quinitinshill (1915), St Bede's (1915), Hull Paragon (1927), Winwick Junction (1934) and others too would all have almost certainly have been prevented by track circuits. Cost must have come into it alongside unwillingness to change and I wonder if also there was an element of "not invented here". I certainly agree that Mansell wheels are a pretty poor excuse!
 

MarkyT

Established Member
Joined
20 May 2012
Messages
4,627
Location
Torbay
That's a very coherent explanation which I am sure was the view of railway management at the time, but sadly the history of accidents suggests that visual observation often didn't suffice. Hawes Junction (1910), Pontypridd (1911), Quinitinshill (1915), St Bede's (1915), Hull Paragon (1927), Winwick Junction (1934) and others too would all have almost certainly have been prevented by track circuits. Cost must have come into it alongside unwillingness to change and I wonder if also there was an element of "not invented here". I certainly agree that Mansell wheels are a pretty poor excuse!
It is a sad fact that, historically, management in many industries have resisted the expense of new technical developments for as long as they can, while pushing existing safeguards to and often far beyond their capability, occasionally with catastrophic results. The Quintinshill disaster in particular, with its extreme traffic conditions during wartime, is often claimed to have been pivotal in changing attitudes towards the subsequent wider adoption of track circuits. What I'm not clear of is how widespread the use of track circuits was elsewhere in the world and whether the UK really was any slower to adopt than other nations. Much of the US, the original home of the technology, was decidedly 'dark' territory at the time, with few signalling controls recognisable to British eyes and written order procedures used widely instead of block instruments and tokens, but there were some high traffic areas on urban railways such as city elevated and subway lines where automatic block using track circuits had an early application by the turn of the century. The niche use on early tube railways in London, from 1903, was probably influenced by the American entrepreneurial involvement with some of these schemes.
 

John Webb

Established Member
Joined
5 Jun 2010
Messages
1,884
Location
St Albans
That's a very coherent explanation which I am sure was the view of railway management at the time, but sadly the history of accidents suggests that visual observation often didn't suffice. Hawes Junction (1910), Pontypridd (1911), Quinitinshill (1915), St Bede's (1915), Hull Paragon (1927), Winwick Junction (1934) and others too would all have almost certainly have been prevented by track circuits. Cost must have come into it alongside unwillingness to change and I wonder if also there was an element of "not invented here". I certainly agree that Mansell wheels are a pretty poor excuse!
Unfortunately I've lost the reference I found a while ago, but the Midland Railway after the Hawes Junction disaster drew up a list of some 2000 locations where track circuits were thought to be needed. They weren't quick to install them - At St Albans City the track circuits were installed in December 1915 and are numbered between 400 and the low 500s as part of the MR's national TC numbering system. But we're deviating from the 'wheel' subject!
 

matchmaker

Member
Joined
8 Mar 2009
Messages
1,038
Location
Central Scotland
I'm no expert on coaching stock. Is it possible old Mansell wheels were used? And while the last sleeping cars to an LMS design were built 1951/2, might have they been put onto more modern bogies? (I'm bearing in mind the greater complexity of the Mansell wheel, the cost of teak etc.)
I suppose it is possible that old wheels were used, but the bogies were certainly standard LMS.
 

Merle Haggard

Member
Joined
20 Oct 2019
Messages
642
Location
Northampton
Unfortunately I've lost the reference I found a while ago, but the Midland Railway after the Hawes Junction disaster drew up a list of some 2000 locations where track circuits were thought to be needed. They weren't quick to install them - At St Albans City the track circuits were installed in December 1915 and are numbered between 400 and the low 500s as part of the MR's national TC numbering system. But we're deviating from the 'wheel' subject!
There's a few things that puzzle me. To understand the situation at the time, I'm trying to place myself in the early part of the 20th century when electricity was an innovation that was not widely available outside large towns, rather than the present day when its universal availability is taken for granted.

Where did the power for track circuits come from? In the case of St. Albans, was it from the local electricity power station, or from accumulators, and , if the latter, how were they recharged?

In the case of Quintinshill (which would perhaps not have been the highest priority for track circuiting without the foreknowledge of the disaster) probably many miles from the nearest power station how would power supply be arranged?

With electric (colour light) signals it's obviously easy to use track circuits to operate signals, but I can't visualise how, in 1915, the St Albans' track circuits could have prevented signals being cleared, only how they could provide a reminder to the signalman of the presence of a train if he looked at the indicator.

If track circuits only indicated to a signalmen that a line was occupied it would not prevent the signalman from nevertheless clearing signals to create a collision.

Perhaps Quintinshill is not a good argument for track circuits to remind signalmen of the presence of a train; did not one of the signalmen involved actually alight from the train that was forgotten? In the case of Hawes Junction, the light engines were not far from the 'box, and other safeguards (e.g. Rule 55, or use of the engine whistle) were not used to remind the signalman of their presence, for reasons justified in the accident report. In such circumstances of forgetfulness, would the opportunity for the signalman to look at a track circuit indicator really have made a difference? He also had the opportunity to look out of his window!

I'm making the observation that we are using our present knowledge, not to mention hindsight, to find fault with policies of a century earlier. Hawes Junction could also have been avoided if the Midland Railway had had engines powerful enough to work their premier route without assistance (the signalman was distracted by the movements of assisting engines, including the ones involved in the collision), which seems to me a more reasonable criticism of policies of the day and lesson to be learned.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
4,627
Location
Torbay
There's a few things that puzzle me. To understand the situation at the time, I'm trying to place myself in the early part of the 20th century when electricity was an innovation that was not widely available outside large towns, rather than the present day when its universal availability is taken for granted.

Where did the power for track circuits come from? In the case of St. Albans, was it from the local electricity power station, or from accumulators, and , if the latter, how were they recharged?

In the case of Quintinshill (which would perhaps not have been the highest priority for track circuiting without the foreknowledge of the disaster) probably many miles from the nearest power station how would power supply be arranged?

With electric (colour light) signals it's obviously easy to use track circuits to operate signals, but I can't visualise how, in 1915, the St Albans' track circuits could have prevented signals being cleared, only how they could provide a reminder to the signalman of the presence of a train if he looked at the indicator.

If track circuits only indicated to a signalmen that a line was occupied it would not prevent the signalman from nevertheless clearing signals to create a collision.

Perhaps Quintinshill is not a good argument for track circuits to remind signalmen of the presence of a train; did not one of the signalmen involved actually alight from the train that was forgotten? In the case of Hawes Junction, the light engines were not far from the 'box, and other safeguards (e.g. Rule 55, or use of the engine whistle) were not used to remind the signalman of their presence, for reasons justified in the accident report. In such circumstances of forgetfulness, would the opportunity for the signalman to look at a track circuit indicator really have made a difference? He also had the opportunity to look out of his window!

I'm making the observation that we are using our present knowledge, not to mention hindsight, to find fault with policies of a century earlier. Hawes Junction could also have been avoided if the Midland Railway had had engines powerful enough to work their premier route without assistance (the signalman was distracted by the movements of assisting engines, including the ones involved in the collision), which seems to me a more reasonable criticism of policies of the day and lesson to be learned.
Early electrical signalling apparatus was often powered by primary cells that had to be changed periodically. When I was an S&T trainee in the the early 1980s, just before resignalling, much of the external equipment around the Exeter area was still powered this way, although the supplies at the signal boxes were all mains connected by this time. The mains was usually a low quality spur from the nearest station supply rather than a generator secured dual supply typical of later colour light schemes, so rechargeable battery backup was used to cover inevitable outages and equipment remained low voltage DC. Early equipment was designed to be ultra low consumption, so primary cell operation was practical. Solenoid lever locks were used to interlock track circuits and block releases with manually operated semaphores.
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
19,170
Location
Nottingham
Early electrical signalling apparatus was often powered by primary cells that had to be changed periodically. When I was an S&T trainee in the the early 1980s, just before resignalling, much of the external equipment around the Exeter area was still powered this way, although the supplies at the signal boxes were all mains connected by this time. The mains was usually a low quality spur from the nearest station supply rather than a generator secured dual supply typical of later colour light schemes, so rechargeable battery backup was used to cover inevitable outages and equipment remained low voltage DC. Early equipment was designed to be ultra low consumption, so primary cell operation was practical. Solenoid lever locks were used to interlock track circuits and block releases with manually operated semaphores.
How did they charge these with no power supply on site? Would they swap them with charged ones and take them away (on a train?) to be charged elsewhere.

I may be biased by the Peco point motor but I got the impression solenoids were quite power-hungry, and for failsafe they would have to be energized when the track circuit was unoccupied, which would be most of the time. Or was there a switch contact on the lever to energise the solenoid (subject to track circuit unoccupied) when the signalman started to pull it?
 

MarkyT

Established Member
Joined
20 May 2012
Messages
4,627
Location
Torbay
How did they charge these with no power supply on site? Would they swap them with charged ones and take them away (on a train?) to be charged elsewhere.

I may be biased by the Peco point motor but I got the impression solenoids were quite power-hungry, and for failsafe they would have to be energized when the track circuit was unoccupied, which would be most of the time. Or was there a switch contact on the lever to energise the solenoid (subject to track circuit unoccupied) when the signalman started to pull it?
Chemical batteries of various types were used. I think most were leclanche derived, which is an ancestor of today's alkaline dry cell. 'Economiser' contacts of some kind were always employed to only energise the lock when required. Some operated from the catch handle, some with a foot treadle. The GWR and BR(W) used a hand plunger on the instrument shelf. For reverse locks such as FPLs, a time delay circuit can be arranged to hold the lock energised for a few seconds after plunging so the signaller can get properly behind the lever to push it back normal.
 

John Webb

Established Member
Joined
5 Jun 2010
Messages
1,884
Location
St Albans
There's a few things that puzzle me. To understand the situation at the time, I'm trying to place myself in the early part of the 20th century when electricity was an innovation that was not widely available outside large towns, rather than the present day when its universal availability is taken for granted.

Where did the power for track circuits come from? In the case of St. Albans, was it from the local electricity power station, or from accumulators, and , if the latter, how were they recharged?

In the case of Quintinshill (which would perhaps not have been the highest priority for track circuiting without the foreknowledge of the disaster) probably many miles from the nearest power station how would power supply be arranged?

With electric (colour light) signals it's obviously easy to use track circuits to operate signals, but I can't visualise how, in 1915, the St Albans' track circuits could have prevented signals being cleared, only how they could provide a reminder to the signalman of the presence of a train if he looked at the indicator.

If track circuits only indicated to a signalmen that a line was occupied it would not prevent the signalman from nevertheless clearing signals to create a collision.

Perhaps Quintinshill is not a good argument for track circuits to remind signalmen of the presence of a train; did not one of the signalmen involved actually alight from the train that was forgotten? In the case of Hawes Junction, the light engines were not far from the 'box, and other safeguards (e.g. Rule 55, or use of the engine whistle) were not used to remind the signalman of their presence, for reasons justified in the accident report. In such circumstances of forgetfulness, would the opportunity for the signalman to look at a track circuit indicator really have made a difference? He also had the opportunity to look out of his window!

I'm making the observation that we are using our present knowledge, not to mention hindsight, to find fault with policies of a century earlier. Hawes Junction could also have been avoided if the Midland Railway had had engines powerful enough to work their premier route without assistance (the signalman was distracted by the movements of assisting engines, including the ones involved in the collision), which seems to me a more reasonable criticism of policies of the day and lesson to be learned.
I'm not certain when St Albans got mains power, but I expect the original source would have primary cells, quite likely of the Leclanche 'wet cell' type; these had a capacity of around 25 to 50 Watt-hours. These cells needed maintenance in keeping the 'Sal Ammoniac' liquid topped up and replacing the zinc and other components as they were consumed. 'Dry' cells were rapidly introduced for use in offices and other places where the fumes were not liked, and these became standard for S&T use. Construction and capacity varied depending on use - larger cells for signal workings and smaller cells for telephone/telegraph usage. Capacity could be up to 250 Watt-hours.
As MarkyT says, equipment was designed to work on low power and could operate locks on levers, and with relays (well-developed by then) these locks could allow for interlocking with both track circuits and the block instruments to give 'Line Clear' to the section signal worked by the box in rear only when the correct conditions existed. I'm not certain exactly how much locking with track circuits was introduced at St Albans in 1915, however.

Once mains power became available, secondary (rechargeable) batteries could be used as the back-up at least within the signal box. I never saw it for myself, but when the Trust took over St Albans South there was a cupboard full of 'Nife' cells in the locking room and the council declared it to be a 'hazardous waste' area. Veolia came to our rescue and removed them all for disposal at no expense to us!

Track circuits use quite low voltages and currents - the relay in a simple DC track circuit may have a resistance of around 10 ohms, pick up (ie operate) at around 0.4 volt and drop-out (release) at around 0.25 volt. The subject is discussed at length in "Railway Signalling and Communications", a collection of papers put together by the LNER for their S&T staff, and reprinted by Peter Kay (ISBN 1 899890 24 6).
 

John Webb

Established Member
Joined
5 Jun 2010
Messages
1,884
Location
St Albans
Chemical batteries of various types were used. I think most were leclanche derived, which is an ancestor of today's alkaline dry cell. 'Economiser' contacts of some kind were always employed to only energise the lock when required. Some operated from the catch handle, some with a foot treadle. The GWR and BR(W) used a hand plunger on the instrument shelf. For reverse locks such as FPLs, a time delay circuit can be arranged to hold the lock energised for a few seconds after plunging so the signaller can get properly behind the lever to push it back normal.
Strictly speaking the 'zinc-carbon' drycell is the successor to the Leclanche cell - the alkaline dry cell is a different chemistry.

Plungers on the block shelf were far more widespread across BR - we've plenty of them at St Albans South:
1st floor, from S end MOD.jpg
Some are the large all-brass plungers, others bear a remarkable resemblance to the AWS cancel button used on early BR DMUs!

Some locks worked directly on the locking bars of the 'tumbler' frame; later locks were connected to the downrods as there were limits on the number that could be fitted to the frame. All the electric locks (and most of the plungers) were stripped out for use elsewhere or for spares, and we have no such locks fitted now.
 
Last edited:

MarkyT

Established Member
Joined
20 May 2012
Messages
4,627
Location
Torbay
Track circuits use quite low voltages and currents - the relay in a simple DC track circuit may have a resistance of around 10 ohms, pick up (ie operate) at around 0.4 volt and drop-out (release) at around 0.25 volt.
Also, TCs draw most power when occupied, when the rails are shorted out by an axle, which for most outlying running line examples such as a 'home approach' track, or for an IBS, is typically a small proportion of the time, so primary cell power is practical. Where trains are more likely to occupy TCs for longer periods, when laying over in larger station platforms for example, TCs at these locations would more likely be mains powered from the signal box AC supply - The feed sets for this configuration typically have a single dedicated LA secondary cell for backup/smoothing purposes.
 

DerekC

Established Member
Joined
26 Oct 2015
Messages
1,193
Location
Hampshire (nearly a Hog)
There's a few things that puzzle me. To understand the situation at the time, I'm trying to place myself in the early part of the 20th century when electricity was an innovation that was not widely available outside large towns, rather than the present day when its universal availability is taken for granted.

Where did the power for track circuits come from? In the case of St. Albans, was it from the local electricity power station, or from accumulators, and , if the latter, how were they recharged?

In the case of Quintinshill (which would perhaps not have been the highest priority for track circuiting without the foreknowledge of the disaster) probably many miles from the nearest power station how would power supply be arranged?

With electric (colour light) signals it's obviously easy to use track circuits to operate signals, but I can't visualise how, in 1915, the St Albans' track circuits could have prevented signals being cleared, only how they could provide a reminder to the signalman of the presence of a train if he looked at the indicator.

If track circuits only indicated to a signalmen that a line was occupied it would not prevent the signalman from nevertheless clearing signals to create a collision.

This prompted me to do a bit of digging in Peter Woodbridge's A Chronology of UK Railway Signalling, with interesting results.

1835 - the electric relay invented by Davy in the UK and (separately) Henry in the USA
1837 - the Daniell cell (zinc-copper wet cell) was invented. It was the first practical battery, extensively used in the early development of the electric telegraph.
1854 - Tyer & Co propose a complete train movement control system including a treadle-operated train presence detector but it proves too complex and sophisticated so he went on to develop his train signallng telegraph
1859 - Gaston Plante invents the lead-acid rechargeable battery
1866/7 - the first practical dynamos (DC generators) invented separately by three inventors (Wheatstone, Siemens and Varley)
1866 - Leclanche invents the zinc/manganese dioxide (later zinc/carbon) dry cell
1872 - William Robinson invents the "Closed Track Circuit" in New York
1875 - Charles Spagnoletti conducts first experiments with power operated signals (on the Circle Line)
1879 - Edison (US) and Swan (UK) invent a practical incandescent lamp
1879 - First electric train demonstrated by Siemens in Berlin
1881 - First DC generating station coupled to a distribution system, in New York
1886 - First AC generating station coupled to a distribution system, in Pittsburgh
1889 - First electric point motor (Baltimore & Ohio)
1894 - First UK implementation of track circuits (in Gasworks Tunnel, Kings Cross) to indicate line occupied/clear to signaller
1901 - First UK implementation of track circuit block, using electropneumatic signals (by the LSWR at Andover Junction)
1902 - Taylor System of all-electric power signalling installed at Grand Central Station, New York - including approach locking. (The system was powered by a 100V battery which was charged by running a petrol-electric generator once per week)
1903 - Hall automatic semaphore signalling installed by the NER between Alne and Thirsk
1903 - AC track circuits adopted by the Metropolitan District Railway
1904 - First colour light signal in East Boston Tunnel, USA

It's pretty clear that by 1910 when the Hawes Junction accident happened, all the technologies were in place to support track circuits interlocked with the signalling. So the fact that it took such a long time to roll out afterwards is not because of technology issues.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
4,627
Location
Torbay
This prompted me to do a bit of digging in Peter Woodbridge's A Chronology of UK Railway Signalling, with interesting results.

1835 - the electric relay invented by Davy in the UK and (separately) Henry in the USA
1837 - the Daniell cell (zinc-copper wet cell) was invented. It was the first practical battery, extensively used in the early development of the electric telegraph.
1854 - Tyer & Co propose a complete train movement control system including a treadle-operated train presence detector but it proves too complex and sophisticated so he went on to develop his train signallng telegraph
1859 - Gaston Plante invents the lead-acid rechargeable battery
1866/7 - the first practical dynamos (DC generators) invented separately by three inventors (Wheatstone, Siemens and Varley)
1866 - Leclanche invents the zinc/manganese dioxide (later zinc/carbon) dry cell
1872 - William Robinson invents the "Closed Track Circuit" in New York
1875 - Charles Spagnoletti conducts first experiments with power operated signals (on the Circle Line)
1879 - Edison (US) and Swan (UK) invent a practical incandescent lamp
1879 - First electric train demonstrated by Siemens in Berlin
1881 - First DC generating station coupled to a distribution system, in New York
1886 - First AC generating station coupled to a distribution system, in Pittsburgh
1889 - First electric point motor (Baltimore & Ohio)
1894 - First UK implementation of track circuits (in Gasworks Tunnel, Kings Cross) to indicate line occupied/clear to signaller
1901 - First UK implementation of track circuit block, using electropneumatic signals (by the LSWR at Andover Junction)
1902 - Taylor System of all-electric power signalling installed at Grand Central Station, New York - including approach locking. (The system was powered by a 100V battery which was charged by running a petrol-electric generator once per week)
1903 - Hall automatic semaphore signalling installed by the NER between Alne and Thirsk
1903 - AC track circuits adopted by the Metropolitan District Railway
1904 - First colour light signal in East Boston Tunnel, USA

It's pretty clear that by 1910 when the Hawes Junction accident happened, all the technologies were in place to support track circuits interlocked with the signalling. So the fact that it took such a long time to roll out afterwards is not because of technology issues.
Very interesting, but 10, even 20 years, is quite a short time in railways, with asset life of signalling still typically up to 40 years. Think how long it took for the advanced warning system to be adopted widely in the UK, after the GWR had started using something functionally similar from the early 20th century and French railways had invented the equivalent Crocodile back in the 1870s! The UK network was also very well developed already by 1900, while some other countries were still building out main lines at the time, which clearly, being newer, used state of the art tech from the beginning. It took some major reconstructions in the 1920s and 30s for predominantly electrical signalling to become established in the UK, and then it was justified as primarily an efficiency measure to allow fewer employees to operate larger areas and handle more trains. I'd say WW1 was a rather dead time for technical developments when the country's resources were concentrated fairly and squarely on the war effort rather than implementing railway safety measures. Quintishill, while being a military disaster as well as a railway one, was still small in the context of the whole network and war casualties generally. With the best will in the world, the industry couldn't have set off a swift campaign to retrofit many hundreds, perhaps thousands of track circuits nationwide; the resources and manpower were simply not there to do that. Then the post-war economic climate was not conducive to much investment either while the railway companies, although forced into the grouping of the early 20s, were still primarily profit-motivated privately owned entities.
 

DerekC

Established Member
Joined
26 Oct 2015
Messages
1,193
Location
Hampshire (nearly a Hog)
Very interesting, but 10, even 20 years, is quite a short time in railways, with asset life of signalling still typically up to 40 years. Think how long it took for the advanced warning system to be adopted widely in the UK, after the GWR had started using something functionally similar from the early 20th century and French railways had invented the equivalent Crocodile back in the 1870s! The UK network was also very well developed already by 1900, while some other countries were still building out main lines at the time, which clearly, being newer, used state of the art tech from the beginning. It took some major reconstructions in the 1920s and 30s for predominantly electrical signalling to become established in the UK, and then it was justified as primarily an efficiency measure to allow fewer employees to operate larger areas and handle more trains. I'd say WW1 was a rather dead time for technical developments when the country's resources were concentrated fairly and squarely on the war effort rather than implementing railway safety measures. Quintishill, while being a military disaster as well as a railway one, was still small in the context of the whole network and war casualties generally. With the best will in the world, the industry couldn't have set off a swift campaign to retrofit many hundreds, perhaps thousands of track circuits nationwide; the resources and manpower were simply not there to do that. Then the post-war economic climate was not conducive to much investment either while the railway companies, although forced into the grouping of the early 20s, were still primarily profit-motivated privately owned entities.
I am not sure that we are actually disagreeing with each other. It does indeed take a long time to change signalling assets if you wait until they are life expired - and mechanical signalling with absolute block has an extremely long asset life. The drivers for spending money on changing them early are either increased capacity or efficiency or improved safety - and the private railways weren't really very interested in improved safety. That's really the point where I started from in this thread - that "Mansell Wheels" are not a very good excuse for slow rollout of track circuits and there were clearly a lot of other things holding back improvement. I agree that WWI created a hiatus. However I think the industry could have done better than it did. In the case of track circuits there was quite a lot of "not invented here", I think. Woodbridge quotes a UK railway (unfortunately he doesn't say which one) commenting on the 1902 Grand Central implementation of approach locking "...required only to guard against a sin peculiar to that country (i.e. USA) of a signalman taking the driver's road away and, having been mentioned, need not be further referred to..." . But there was a lot of inertia as well. AWS is a particularly good example of that. Even after the catastrophic accident at Harrow & Wealdstone in 1952 (the pictures still give me the cold shivers) the rollout still took quite a long time to get going.
 

edwin_m

Veteran Member
Joined
21 Apr 2013
Messages
19,170
Location
Nottingham
It's often forgotten how most of the innovations in signaling from Mid-Victorian times onward started out in America. The attitude to safety there was probably just as cavalier as in the UK, if not more, but with longer routes and lower density of trains there was more need for automation to improve productivity. This drove the adoption of Centralised Traffic Control with its attendant technologies such as train detection and multiplexing of controls/indications. Suitably adapted to UK principles it formed the basis of our power signaling systems right up to the advent of solid state in the 1980s.
 

MarkyT

Established Member
Joined
20 May 2012
Messages
4,627
Location
Torbay
It's often forgotten how most of the innovations in signaling from Mid-Victorian times onward started out in America. The attitude to safety there was probably just as cavalier as in the UK, if not more, but with longer routes and lower density of trains there was more need for automation to improve productivity. This drove the adoption of Centralised Traffic Control with its attendant technologies such as train detection and multiplexing of controls/indications. Suitably adapted to UK principles it formed the basis of our power signaling systems right up to the advent of solid state in the 1980s.
CTC didn't really start to make a big impact until the 1930s after the first installations in the late 1920s. Prior to that much of America was run on a timetable and train order basis with more complex junctions and terminals equipped with mechanical interlockings, although there were and still are (from British eyes) worrying numbers of completely non-interlocked point connections in running lines secured by no more than a standard crew key. Even today there are wide areas of 'dark territory' with little or no interlocking although now orders are passed direct to crew via radio rather than via station operators and some forms of PTC can offer additional safeguards today without needing full trackside signalling installations. In general, I think DerekC is correct that UK was slow to adopt TCs, but was not unusually so compared to other nations, and the major exceptions to that trend would have been those flagship insanely busy and enormous cutting edge installations at certain major US terminals like New York's Grand Central that, while impressive and very forward-looking in themselves, really represented only a tiny proportion of the total American network of the time.
 

Merle Haggard

Member
Joined
20 Oct 2019
Messages
642
Location
Northampton
Thank you to all who provided details of electricity supplies in response to my original query.

On the subject of the adoption of track circuits; may I suggest that the philosophy at the time was perhaps different. Reading through early accident reports on Railways Archive (to which, incidentally, thanks must be due to Rupert Dyer) the Inspecting Officers' criticisms of railway companies' failures were directed at, for instance inadequate and non-automatic brakes, but this failure I might describe as 'primary', i.e. it was a fault in itself, not one that was the result of a failure by staff to adhere to Rules and Regulations.
There never seems to be criticism of the lack of track circuiting (or, at least, some form of detection) and possibly this was because it might be thought of as 'secondary' - or of the lack of secondary safety features generally. I am using 'secondary' because it's the second line of defence against 'forgetting' a train/loco's presence; the first is the signalman not doing so, and the fireman carrying out Rule 55, i.e., adhering faultlessly to Rules & Regs..
I would suggest that the railway companies' view (and, probably, the Railway Inspectorate's) was accidents such as Hawes Junction were caused simply by failures of staff to adhere to rules and to be proficient in their duties. Blame always seems to be attributed to staff who do not perform faultlessly, rather than to railway companies for not providing back-up systems for when this happens.
It's perhaps a mistake to think that present-day opinions (for instance, that human beings inevitably make human errors) apply to the situation more than 100 years ago. I'm not trying to defend the old companies, just saying that they might have seen things from a different perspective. After all, staff were called Railway Servants and could be dismissed instantly if they weren't 'perfect' with no difficulty in finding a replacement.

As an aside, were the early track circuits short in length compared to todays'? I ask because, certainly in the 1970s, TCs were prone to failure when it rained, because water provided a lower resistance circuit between the rails, nearer to that of a train. This was, I think intuitively, more likely if the T/C was long but I'm not sure of that...
 

Top