London Subway Extended by Tunneling (1926)

From nycsubway.org

Electric Railway Journal · Vol. 68, No. 22 · November 27, 1926 · pp. 959-962.


[Left] Digging the Tunnel by Means of the Greathead Shield. This View Shows the Work South of Camden Town. [Right] Rotary Excavator in Use When the Tube Was Pushed Through Soft Material.

London Subway Extended by Tunneling. Five-Mile Addition to Underground System Built in Ten Sections -- Accurate Surveys Required to Insure Proper Alignment -- In Hard Ground Shields Were Driven Forward by Hydraulic Rams, While in Soft Ground Rotary Excavators Were Used.

Construction of the extension of the London Underground from Clapham to Morden, which was begun in 1923, was completed during the present year. Practically all of the new line is tube construction. In hard ground the tunnel was advanced by means of the Greathead shield, while in softer ground a rotary excavator was used. Methods adopted for the layout and execution of this work are typical of the latest British practice, and differ in many respects from those usually followed in this country. The story of the work as told by the London Underground is therefore of particular interest.

From Clapham to Morden is a distance of 5 miles. Except for the last half mile, where the extension runs in the open, the line is of standard tube design, with tunnels of 11 ft. 8-1/2 in. diameter, located at an average depth of 40 ft. below the surface. Between the stations the tunnels are an average distance of 5 ft. apart. At the stations, where the tunnels are of 21 ft. 2-1/2 in. diameter and the lower portion of an escalator shaft comes between them, they are considerably farther apart.

On leaving a station each tunnel has a declivity of 1 in 30 for about 300 ft., while on approaching a station it has a rise of 1 in 60 for about 600 ft. Thus the tunnels are at the same level for a considerable distance between the stations, but they rise and fall at different angles in the sections immediately adjoining them. This switchback principle, applied wherever practicable, promotes acceleration of the trains as they depart from the stations and retards speed as they approach stations, making for economy and energy consumption and reducing wear and tear on the brakes. Preliminary to construction a route survey was made. First, the course to be followed was laid down approximately on large-scale maps. A detailed survey of the route was then carried out. The chief consideration was to select an easy route, making the curves as flat as possible, and to insure that the tunnels would not encroach on private property except where this was unavoidable. The levels of the highway were taken and all available data collected with respect to underground works, sewers, gas and water mains, etc.

Except for one short stretch, it was found practicable to lay out curves of such long radius that the tunnels could be of the standard 11 ft. 8-1/2 in. diameter throughout. At one point, however, the northbound tunnel had to be made of 12 ft. diameter. In general the curves of the Underground tunnels previously built are of larger diameter than are the tangents.

Arbitrary lines from which the center lines of the tunnels could be calculated were laid out on the street. This process is shown in an accompanying illustration. All places where the tunnels approached close to the building line were carefully checked with a theodolite, to make sure that there would be no encroachment on private property.


Having arrived at a satisfactory layout, the next step was to decide the general scheme of construction. Boring the tube from one point only would have been an extremely protracted undertaking. It is estimated that under such circumstances the Morden extension would have taken about fifteen years to complete. Work was arranged, therefore, in ten sections, five of which represent the distances between stations and another the section from the last station to the point where the line rises to the surface. Two of the sections between stations were subdivided to facilitate operations where the work had to be carried out under compressed air, the ground being waterlogged. In addition, there were concrete tunnels 550 yd. in length, built by the cut and cover process, where the line rose to the surface, and also an above-ground section linking the Morden station with the car sheds beyond.

Certain sections of tunnel were bored from south to north, but the majority from north to south. The average length of the longer sections was about 3,000 ft. Intervening among these sections six stations had to be constructed, each with escalator shafts, passageways, etc.

The next step in the work was the sinking of service shafts at the various sites from which tunneling was to proceed. In all, eighteen shafts were sunk. While in the construction of the older tube railways of London these shafts were from 16 ft. to 30 ft. in diameter, on the Morden extension they were only 10 ft. 6 in. in diameter. In previous construction it was customary to install elevators and emergency staircases for the stations in the shafts after the completion of tunneling. On the Morden line, however, the stations have escalators and the temporary shafts were made of small diameter for reasons of economy.

A shaft having been sunk to the requisite depth, a temporary heading was driven out from it by hand under, and at right angles to, the roadway to the site of the running tunnel. A theodolite was then set up on the surface on the center line of the tunnel as defined on the street, and a point ranged in line opposite the center of the shaft. The instrument was placed over this point and a right angle from the tunnel line was turned off. A steel piano wire was then suspended from the far side of the shaft, being kept taut by a 28-lb. weight immersed in a tank of water, great care being taken to insure that the wire hung perfectly clear. This wire was then adjusted across the field of view until it exactly bisected the cross-hairs of the telescope. A similar wire, hung from an adjusting screw, was suspended from the near side of the shaft, the two wires being as far apart as the shaft would allow. By means of the adjusting screw the second wire was then traversed across the field of view until it coincided with the first wire. The two wires were thus on a true line at right angles to the center line of the tunnel. The distance from the instrument to the nearest wire was next measured.

To transfer the center line to the heading, the theodolite was set up in the heading at the same distance from the near wire as was the case on the surface. The wires were again sighted and the instrument moved on its sliding plates until the two wires appeared as one. When this had been done the instrument was plumb under the point at which it had been set on the street above. With this as a base, the center line of the tunnel was determined by turning off a right angle. The operation was repeated from time to time as the tunnel increased in length until the engineer was assured that the center line had been established with sufficient accuracy to be used to the end of the drive.


The base line below ground having been established, tunneling began. A chamber 15 ft. in diameter was cut at right angles to the heading. The shield was then built up into the chamber, and the correctness of its position and direction having been verified by repeated checkings with the base line data, the task of excavating the tube began in earnest.

The shield was driven forward by means of hydraulic rams around the periphery, pressing against the iron lining segments of the tunnel already constructed. By regulating the force exerted by the various rams, a straight tunnel, a curving tunnel, a downward slope or an upward slope were made as occasion demanded. An exigency that had to be taken constantly into consideration was that owing to the weight of the shield there was a tendency for the nose to go downward when uniform pressure was exerted by all the rams.

Operations began on one tunnel only, work on the other not being undertaken until the first tunnel had been driven some little distance ahead. Two shields, therefore, were never working alongside each other.

The skin was carried forward as the shield went ahead and the space that it occupied was then grouted through holes in the lining segments. At the end of a drive, when the shield could be projected no farther, the skin was necessarily left in place between the segments and the earth.

Each cast-iron ring with which the tunnel is lined consists of six segments and a key piece, the latter being inserted in the top of the ring. Where the diameter of the tunnel is increased to 21 ft. 2-1/2 in. at the station platform, twelve segments and a key piece are used. The segments were placed in position and bolted together at the flanges after each advance of the shield, the forward movement being 1 ft. 8 in. for the standard tunnel and 1 ft. 6 in. for the large diameter tubes.

Two types of shields were used, the Greathead and the rotary excavator. In the case of the Greathead shield, men work inside the shield and dig out the earth at the face, as it is broken down by the forward movement. This method was used where the ground was firm. With the rotary excavator the digging is performed by rotating cutters and the earth is thrown back mechanically. The rotary excavator was used only for tunneling through clay when expeditious work was necessary.


Trial bores were made of the ground through which the tunnel was to be driven. While such borings afforded a general idea of the nature of the strata in which operations were to proceed, they did not, of course, indicate every peculiarity of the lower levels throughout the course of the line. Borings were made from the surface and also horizontally ahead of the shields. This was done with a long-handled carpenter's augur with sufficient length to penetrate for several feet through the earth to be excavated. When work was proceeding under normal conditions, this was done as a precaution against suddenly encountering waterbearing or other bad ground. When operations were being carried out in compressed air, it was necessary also as a precaution against running unexpectedly into some outlet for the air. The wisdom of these precautionary measures was demonstrated several times during the course of the construction.

In water-logged ground below the surface compressed air was used. An airtight compartment was built in the tunnel between the shaft and the tunnel face. In some cases it was a section of the tunnel itself, from 10 ft. to 20 ft. or more in length, closed by stout bulkhead walls of brick or concrete, each wall having an airtight steel door for the passage of men and materials. In other cases the air lock was a horizontal steel cylinder with a bulkhead wall and steel door at each end. Sometimes the pressure used was as low as 5 lb. per square inch. In exceptional cases it was as high as 25 lb., or even higher. The average was approximately 10 lb. per square inch. When work was proceeding under a river the pressure had to be adjusted in conformity with the ebb and flow of the tide, which caused a fluctuation in the weight on the clay above the tunnel.


During the process of tunneling, checks on the position of the shield and the direction of drive were taken periodically. As the tunnel was bored iron dogs were fixed in the roof about 35 ft. apart. From them plumb lines were suspended. These dogs were set by an engineer and the plumb lines were placed in exact line with the theodolite. When necessary to check the position of a shield, a bar called a fiddle was placed across the horizontal diameter of the shield. The top of the fiddle was at axis level and had a notch in the middle denoting the center of the shield. A small piece of wood having a slit illuminated by a candle was then placed on the fiddle and shifted back and forth until the slit coincided with the plumb lines. It was then noted how much the notch on the fiddle was to the right or left of the center line of the tunnel and the error in position determined.

For testing the level of the shield, hangers were bolted to the roof of the tunnel on which adjustable pins had been set by the engineer to an exact level. From these pins T-shaped rods were suspended, the crosshead representing the true axis level of the tunnel. The piece of wood was then placed against the face of the fiddle with the slit horizontal and raised or lowered until it coincided with the line of sight of the crossheads. The position of the top of the fiddle was then noted in relation to the slit and any necessary adjustment of the shield was made by variation in the force exerted by the rams.

For further guidance, square lines were set out at right angles to the center line of the tunnel, and trailing rods divided into feet and inches were attached to each side of the shield. If the shield was proceeding square with the center line, readings were equal on the scales of the trailing rods. When the readings were unequal it indicated that the shield would soon be out of line.

For boring curves the engineer furnished the foreman with a list showing for every foot the shield advanced how much to the right or left of the center line it should be. When sighting on the fiddle the foreman could therefore ascertain his position by referring to the list. Trailing rods with different scales were provided for curves. While the rod for the inside of the curve was divided into feet and inches, on the outer rod the divisions, although marked as feet and inches, were "stretched" to correspond with the lengths on the inside. Vertical lines also were established by the engineer and it was thus possible to ascertain from the readings on the trailing rods whether or not the nose of the shield was being kept to the correct line of the curve. The maximum deviation from level or center allowed in a shield was 14 in.

When the construction of the Morden extension was at its height, eighteen shields were in operation simultaneously on various parts of the line. Twelve were at work on the running tunnels and six of extra large size on the platform tunnels of the stations.

Measuring One of the Main Surface Survey Lines, from Which the Center Lines of the Tunnels Were Calculated. Metal Flags Were Fixed In the Pavement to Mark the Ends of the Survey Lines. This Work Was Carried Out at Night.
Transferring the Center Line Below Ground. No. 1. Engineer getting the plumb lines in the service shaft in line with the point over which his theodolite was set in the street. The man in the shaft is adjusting the near wire. No. 2. In the service shaft, showing the two plumb lines and, in background in the street, the engineer with his theodolite. No. 3. Engineer in the heading getting his theodolite in line with the two plumb lines at the base of the shaft. No. 4. At the base of the service shaft, showing the nearer plumb line, with weight in tank of water, and, in distance, the engineer in the heading.
[Left] Checking Horizontal Location of the Shield by Means of the Plumb Line. [Right] Checking the Level of the Shield by Sighting Over Suspended T-Rods.


Electric Railway Journal, McGraw Hill Company, Digitized by Microsoft, Americana Collection, archive.org.

nycsubway.org is not affiliated with any transit agency or provider.