Home | Membership | News | Diary | Walks | Calvocoressi Fund | Books | Journals | Links | Database | e-papers | About us

London Hydraulic Power pumping station, Wapping

Address: Glamis Road, Wapping

LHP, Glamis Road, Wapping. © Robert Mason, 28.3.12 LHP, Glamis Road, Wapping. © Robert Mason, 28.3.12

LHP station opened 1892. Coal delivered to adjacent Shadwell Basin and water extracted from this same dock. Six boilers provided steam for six inverted vertical triple-expansion pumping engines; hydraulic energy stored in two accumulators. Two electric turbo pumps added 1923, plant radically modernised 1950s. Steam plant replaced by electric pumps which continued until closure summer 1977. World's last working hydraulic pumping station on public supply system. Much plant removed, but now plans for orchestral rehearsal studio. Tanks to be removed from roofs at the back. Adjacent is underground water storage reservoir, 354805, which held 420,000 gallons. Roof supported on cast-iron columns with beam and jack-arch spanning system.

Text taken from the GLIAS Database

Relevant GLIAS articles:

See also:

The Wapping Project - defunct



1 Location
2 Introduction
3 Historical Background
  3.1 Hydraulic Power
  3.2 The London Hydraulic Power Company
  3.3 The LHP Network
  3.4 LHP Customers' Machinery
4 History of Wapping Pumping Station
5 General Description
6 The Engine House
7 The Engine House Basement
8 The Turbine House
9 The Turbine House Basement
10 The Electrically-driven Pumps
11 Electric Motors and Motor Control
12 The Fire Extinguisher System
13 The Control Room
14 The Power Supply
15 The Accumulators
16 The Pipework
17 The Boiler House
18 The Coal Store
19 The Filter Houses
20 The Workshop and Store
21 The Mess Room
22 The Yard
23 The Significance of Wapping Pumping Station
24 Sources and Bibliography
25 Acknowledgements

Appendix 1 The Steam Pumping Engines
Appendix 2 The Steam Turbine
Appendix 3 Pumps
Appendix 4 Basic Data
Appendix 5 Engine Data
Appendix 6 Electric Pumps Data
Appendix 7 Remote Control Facilities*

Author: Tim R Smith
30 Gaveston Drive
Berkhamsted, Herts HP4 1JF

©Tim R Smith 1999
* ©Dave Gunning 1998


Glamis Road, Wapping Wall
Tower Hamlets
London E

NGR: TQ 354806

The entrance to the pumping station is at the junction of Wapping Wall and Glamis Road. The site is bounded by Shadwell Basin to the north, former warehouses on Wapping Wall to the south, Monza Street to the west and Glamis Road to the east. One of the warehouses on Wapping Wall was demolished during the summer of 1998.


The London Hydraulic Power Company (LHP) was established by Act of Parliament in 1884, though, the public supply of hydraulic power began in September 1883 with the opening of Falcon Wharf pumping station on Bankside by the parent company, the General Hydraulic Power Company. Wapping pumping station was built 1889-92, and was the third large pumping station of LHP. The station was enlarged in the 1920s when the original boilers were replaced and the turbine house was added. During the mid-1950s the steam plant was removed and electric pumps substituted. The station closed in 1977. Since then it has lain empty except for occasional use for art installations, etc. Ownership passed to the London Docklands Development Corporation which made numerous attempts to find a new use for the buildings. The pumping station is currently owned by the Women's Playhouse Trust who are converting it for their own purposes.

Members of the Greater London Industrial Archaeology Society (GLIAS) photographed the station shortly before its closure and again about 1980. The Royal Commission for Historical Monuments, England (RCHME) photographed the buildings in 1987. In 1998, prior to work starting on the conversion GLIAS carried out detailed recording of the engine house and turbine house, producing an inventory, further photographs and some measured drawings. This report is the result of the 1998 GLIAS survey.

Nine site visits, totalling 40 hours, were undertaken. In all, twelve people were involved, giving a total of 111 man hours on site. Most areas were looked at except for the roof-top tanks, the undercroft of the coal store and boiler house, and the underground reservoirs. The boiler house, coal store, filter houses and mess room were not recorded in any great detail. Effort was concentrated on the turbine house, the engine house and basements.


3.1 Hydraulic Power

Hydraulic machines used a static head of water for their power. The larger the head of water the less water was needed. Hydraulic power systems provided artificial heads of water, at pressures of 600 to 800 lbs per square inch (40 to 55 Bar) or more, distributed to often large numbers of machines through networks of pipes. The London Hydraulic Power Company ran the largest hydraulic system. At its peak in the 1930s, it had 187 miles of pipes across London, supplying power to over 8000 machines. [The main archival sources for LHP are the records held by the Metropolitan Archives]

The first machines to utilise a static head of water were the water-pressure engines used in mining. Developed in the mid-C18th they were described by John Smeaton in a paper to the Society of Arts in 1765 after he saw examples at work in mines in the Northern Pennines. [For a general account of hydraulic power see McNeil]

In 1795 Joseph Bramah (1749-1814) patented the hydraulic press, a hand-pumped device which found many uses throughout industry. In 1812 Bramah took out a patent on water supply. In it he envisaged the supply of power through high pressure water mains and anticipated the accumulator. He had already produced what was, in effect, the fore-runner of the hydraulic accumulator in his beer engine. Dr Ure alluded to accumulators working in a Glasgow factory in an article in the Glasgow Mechanics Magazine in 1824. Unfortunately he did not describe what form they took.

In 1846 William Armstrong (1810-1900) demonstrated his hydraulic crane on the Newcastle Town Quay, using the local water mains as his source of supply. Four years later, he built his first weight-loaded hydraulic accumulator and paved the way for the introduction of the first hydraulic power networks. In London the first networks were those at Poplar Dock (1851), Regent's Canal Dock (1853) and London Dock (1853) together with those at Paddington goods station and King’s Cross Goods Depot. By 1858 all the docks on the north side of the river were using hydraulic power. [See Smith, T]

3.2 The London Hydraulic Power Company

Economics dictated that hydraulic power systems should have several machines to be viable. Although riverside wharfingers were keen to install powered cranes the cost of small hydraulic systems was prohibitive. Nevertheless when Beal's Wharf was rebuilt after a fire in 1856 it was fitted with hydraulic cranes with its own pumping plant in the basement. The solution seemed to be to set up a public company to supply power from a central station. In 1860 the London Hydraulic Power Company Ltd was registered and an Act of Parliament obtained for the necessary powers. Unfortunately the scheme was blocked by strong opposition from the water companies who saw the new enterprise as a threat to their monopolies. In 1871 a new company was formed, but the Wharves & Warehouses, Steam Power & Hydraulic Pressure Company faired no better than the first, though for different reasons.

Edward Bayzand Ellington (1845-1914) was engineer and general manager of the Hydraulic Engineering Company of Chester which manufactured hydraulic machinery. He became involved in a scheme to start a public hydraulic power company in Hull and this began operations in 1877, becoming the test bed for future schemes. In 1882 the General Hydraulic Power Company was formed with E B Ellington and Corbet Woodall as its engineers. It acquired the Wharves & Warehouse, Steam Power & Hydraulic Pressure Company and, using powers under the 1871 Act, set up a pumping station at Falcon Wharf, Bankside. Pumping began in September 1883. As the network expanded other pumping stations followed: Millbank in 1888, Wapping in 1891, City Road Basin in 1894 and Rotherhithe in 1903. The Millbank station was compulsorily purchased by the LCC and demolished, along with other buildings, to make way for the Victoria Tower Gardens to the south of the Houses of Parliament. A replacement pumping station was built on Grosvenor Road, opening in 1910. In addition there were two small pumping stations at Philip Lane and Kensington Court, the former serving new warehouses in the Barbican, the latter providing power to a prodigious housing development where hydraulic lifts replaced the back stairs. Both these stations closed soon after the mains reached these properties, but both were retained as remote accumulator stations.

The mainstay of the company's activities was provision of power to hydraulic lifts. Lifts accounted for over half of the machines connected to the company's mains. The production of efficient electric lifts in the early years of the C20th had a great impact on LHP, though it was not felt for some years. There were many private owners of hydraulic networks in London, particularly the railway companies and dock companies. As their pumping plant, particularly boilers, wore out they tended to switch to LHP rather than replace their own equipment. For this reason use of LHP rose to a peak in 1928 and continued at a high level until World War II, whereas overall use of hydraulic power in London probably started to decline well before this. During the war bombing did a great deal of damage to pipework particularly in Dockland. Customers who lost power or machines through war damage tended to buy electrically powered replacements.

In the post-war years LHP electrified its pumping stations. By this time Falcon Wharf had already closed but, in the early 1960s, LHP took over operation of the East India Dock pumping station in Poplar. This had already been electrified in the 1920s. Around the same time the LHP mains were extended to serve the railway goods yards around King's Cross and St Pancras. But the writing was on the wall. One by one the pumping stations were closed. Wapping was the last to close on 1st July 1977.

3.3 The LHP Network

The Act of Parliament giving powers to the Wharves & Warehouses Steam Power & Hydraulic Pressure Company restricted the area of its activities to a narrow strip of riparian properties in Southwark and the City. The site of the first pumping station at Falcon Wharf, Bankside, was chosen to serve that area. The 1884 Act, which renamed the company to London Hydraulic Power Company, extended the range over which the company could operate. Even so it took many years before the network reached its greatest extent of some 187 miles of mains.

In order to serve some new warehouses outside the range of the mains a temporary pumping station was established at Philip Lane, in the City. Once mains had laid up Wood Street these pumps became the stand-by for the whole network. When the Millbank pumping station opened the Philip Lane pumps were removed, but the accumulator was kept. A similar situation arose at Kensington Court where a prestigious development required hydraulic power for the lifts which took the place of the back stairs. A small pumping station was built at the south end of the site. Interestingly, at the north end of the site an electric power station was built so that the properties could have electric lighting. Both buildings survive, much altered. Again, once the mains reached Kensington the small pumping station was taken out of service but its accumulators were retained.

The building of Wapping pumping station opened up a whole new district of London for LHP, anticipating future developments. The new pumping station was linked to existing network via Lower East Smithfield. Two or three years later LHP purchased the Tower Subway and established new links across the river. Earlier links had been made across Blackfriars Bridge. The Tower Subway was the first tunnel to be constructed using the Greathead Shield. After a short spell with a rope-worked railway running in the tunnel it became a foot-tunnel and continued so until the opening of Tower Bridge in 1894 made it redundant. The tunnel is still used by utilities, especially Thames Water. The northern entrance, a circular brick structure, can be seen on Tower Hill, still bearing the name ‘London Hydraulic Power Company’.

Two more large pumping stations were built, City Road Basin and Rotherhithe. City Road Basin pumping station, on Wharf Road, took its water from the Regent's Canal. The Rotherhithe pumping station buildings still survive. In the early years of the Twentieth Century the London County Council (LCC) decided to do something about the riverside just south of the houses of Parliament. They planned to demolish all the buildings north of Lambeth Bridge to make way for the gardens which are there today. One of those buildings was the LHP Millbank pumping station which the LCC compulsorily purchased. A site was found along Grosvenor Road for a replacement which was opened in 1910. Extensions to the pipe network continued but these were mainly infilling areas within the overall network.

During World War II much of the network of pipes was damaged in air raids. After the war there were further contractions until the 1950s. With the LHP pumping stations being electrified two major extensions came about. British Railways decided to close their pumping stations at Somers Town and Kings Cross goods depots and to buy their hydraulic power from LHP. The LHP mains had to be extended to do this. Then in 1960 LHP took over the Port of London Authority's pumping station at East India Dock, and the supply to the West India and Millwall docks as well. The East India Dock pumping station dated from 1857 but had been electrified in 1926.

3.4 LHP Customers Machinery

At its peak LHP served around 8000 customers' machines of which about half were lifts, both passenger and goods. Many of these were installed in the offices and warehouses of the City. People still remember these lifts and the ropes which were used to open the valves to operate them. Some had push button controls which were introduced in the early 1900s. Hotels and department stores were early users of passenger lifts. The first domestic passenger lifts were to be found in Kensington Court but soon other blocks of high-class flats in the West End had hydraulic lifts.

On riverside wharves and in the docks cranes were the most important machines. Elsewhere in the docks were hydraulic capstans, swing bridges and lock-gate mechanisms. Railway goods depots also used hydraulic cranes and shunting capstans. Many of those in London were approached on viaduct and were able to utilise space underneath by installing hydraulic wagon hoists. At Waterloo a hydraulic lift took carriages to and from the Waterloo & City line.

Besides these mundane, bread-and-butter machines there were a number of more exotic uses for hydraulic power. In the theatre, safety curtains and stage mechanisms such as the revolving stage at the Coliseum, were powered by LHP. Car showrooms used vehicle hoists. At Earls Court Exhibition Hall the floor can be lowered on hydraulic jacks to form an Olympic sized swimming pool. This was originally supplied from the LHP network. In industry, at Grey & Martin's City Lead Works, in Southwark Bridge Road, lead pipes were extruded by hydraulic presses made by the Hydraulic Engineering Company and supplied with pressure water by LHP. The hydraulic injector fire hydrant found widespread use, especially in London's art galleries whilst several churches, including St Paul's Cathedral, had high pressure organ blowers supplied from LHP mains.

Perhaps the most unusual use of hydraulic power was in the hydraulic vacuum cleaner. This device used a pipe ring around the rooms of an office or hotel. Into this pipe high pressure water was injected from the mains causing a partial vacuum in the pipe. Hoses from the cleaners could be coupled to the pipe creating a vacuum cleaner. Dirt sucked into the cleaner passed straight into the pipe and was washed away down the drains — no bags to empty, no loss of suction!

Tower Bridge had its own hydraulic system with steam pumping engines under the arches of the southern approach viaduct. During the conversion of the bridge to its current electrical operation the steam pumping engines were taken out of service and LHP supplied pressure water.

All customers had one or more meters which registered the amount of pressure water used by them. They were charged a rent for the meter and so much per thousand gallons, depending on the total number of gallons consumed. After use water from machines passed into the drainage system thus fulfilling the requirement in the company's Act of Parliament that all water taken from the Thames must be returned to the river.


In March 1888 LHP were approached by wharfingers in Wapping who wanted to use hydraulic power. Later in the same year the London & St Katharine Docks Co began negotiations for a supply from LHP. At this time the LHP mains came nowhere near Wapping. Agreement was reached in 1889 with the London & India Docks Joint Committee, which had succeeded the London & St Katharine Docks Co, and land was purchased from them for the pumping station at Wapping. A contract was let with John Mowlem & Co to build the pumping station and with the Hydraulic Engineering Co of Chester for four sets of compound inverted vertical steam pumping engines at £2510 each. In November 1889 the Hydraulic Engineering Co offered to change the order to triple expansion engines for an additional £50 each. This was accepted. In April 1890 another two steam pumping engines were ordered.

The first four engines were brought into service in the second half of 1891, the other two began work the following year. Six Fairbairn-Beeley boilers were installed by Thomas Beeley of Hyde Junction Iron Works, Manchester. Filters, for filtering water before it was pumped to high pressure, were supplied by the Pulsometer Engineering Company. A 7-inch (0.2m) main was laid along Wapping Wall, Wapping High Street and Lower East Smithfield, eventually joining up to the rest of the LHP network. A 4-inch (0.1m) main was laid up Milk Yard to connect with the dock system at New Gravel Lane. Various cottages adjoining the site were purchased by the company, including Nos 21, 22, 23, 24 & 25 Star Street, Nos 1, 2, 3 & 4 Sconces Alley and three houses in Wapping Wall. In 1891 work started on building an inspector's house and office at the entrance to the pumping station. It was built by James Smith & Sons of Wapping for £459-7s. An article describing the new pumping station appeared in The Engineer in 1893 and in 1897 members of the Institution of Civil Engineers were invited to visit Wapping, on Tuesday 25th May. Ellington was instructed to provide refreshments. Thereafter Wapping became the ‘show’ pumping station of the London Hydraulic Power Company.

The original plant consisted of six inverted vertical, triple-expansion steam pumping engines supplied with steam from the six Fairbairn Beeley boilers. Water was supplied from a well in the grounds of the pumping station, supplemented by water fed from Shadwell Basin via a syphon pipe into the well. Hydraulic pumps in the well raised water to roof-top tanks over the boiler house and coal store. The water passed through Torrent filters into underground storage reservoirs. Before use it was pumped back up to other roof-top tanks, via the surface condensers of the steam engines. The high-pressure force pumps were fed from these ‘clean water’ tanks through the main suction pipe. A high pressure ring main connected all the pumps to the two accumulators. The mains going out to the street were also connected to the ring main.

In the early 1920s LHP purchased two Mather & Platt electric turbine pump sets which were installed in the centre aisle of the engine house. They were brought into service in the first quarter of 1923. This allowed the boilers to be shut down on 22 March 1923. The boilers were removed and the rooftop tanks over the central section of the boiler house were raised to their present height to enable three Babcock & Wilcox boilers to be installed. The new boilers were brought into service 23 July 1923, just four months later. During this period all pumping at Wapping was done by the electric pumps. About this time the turbine house was built on the south side of the engine house and a Parsons' steam turbine was installed. The steam turbine was brought into use in the week ending 22 April 1926. By 1930 the steam turbine was supplying about a third of the high pressure water from Wapping.

After the terrible smogs of the early 1950s and the passing of the Clean Air Acts the steam pumping plant at Wapping was removed and electrically driven three-throw ram pumps were substituted. The Mather & Platt pumps had not been used for some time, the cost of electricity being much higher than the cost of coal. To facilitate the changeover these pumps were refurbished and put back into service during the conversion. Pumping by steam ceased in September 1956. The steam turbine had been stopped in May 1952. The costs of coal and electricity were little different, electricity being slightly more expensive. But savings were made on wage bills and maintenance costs making the overall cost of pumping by electricity less than the overall cost of pumping by steam. When pumping ceased in 1977 only thirteen men worked at the station. Closure of the docks and railway goods stations in the late 60s and early 70s reduced demand for hydraulic power to an uneconomical level. For about a year Wapping was kept in operation to supply Tower Bridge during its conversion from hydraulic to electric operation. But the end came at 8am of 1st July 1977 when the pumps at Wapping were finally stopped. It was the last LHP pumping station and, indeed, the last public hydraulic pumping station anywhere. The closure of Wapping brought about the end of an era. Hydraulic power is still used in a few places, notably on the Manchester Ship Canal where a few swing bridges, including the famous Barton Swing Aqueduct, still retain their water hydraulics. But these are the exceptions. Where machinery is hydraulically operated today oil is used as the working fluid and each machine has its own electrically powered pump.


The pumping station stood in its own grounds behind high walls. The main entrance, in Glamis Road, was through large double steel doors. There was a small door for pedestrians to the north. The former superintendent’s house stood on the north side of the entrance. There was a second vehicular entrance in Monza Street giving access to a lower yard with a loading bank up to which carts could back to load ash from the boilers. At the higher level of Glamis Road, the main yard extended around the south side of the buildings to the vacant land to the west. In front of the pumping station the yard covered two underground reservoirs. There was a narrow strip between the main buildings and the northern boundary wall. This had been the location of a hydraulic grab crane used to bring coal over the wall from Shadwell Basin. There had also been a toilet block which had been demolished.

At the extreme east of the pumping station there was the unmistakably tall accumulator tower. Adjoining it to the west was the engine house, with the turbine house on its south side. A connecting corridor led from the engine house to the boiler house. The corridor had a dog-leg because the engine house and boiler house were built at an angle to each other. On the north side of the corridor was the workshop and a store. Following electrification the workshop moved into the old boiler house and the store was enlarged. On the south side of the corridor was the mess room. A door from the corridor led to the narrow north yard.

The boiler house floor was below ground level and steps led down from the corridor and from the yard through a double door. Originally the boiler house roof was level but rebuilding in the 1920s to accommodate new boilers had left the western two-thirds higher. At the western end of the boiler house was the coal store. The rooftop tanks covered both boiler house and coal store. Built onto the south side of both was the filter house, divided by the entrance from the yard to the boiler house. The filter house looked as though it had been added on as an afterthought, though it was part of the original buildings.


The engine house was a tall building in red brick with a stone plaque high on the front gable reading ‘LONDON HYDRAULIC POWER COMPANY 1890’. The roof, which had recently been renovated by the LDDC, was supported by Polonceau type timber and iron trusses [The Buildings of England, London Docklands, by Elizabeth Williamson and Nickolaus Pevsner, 1998, page 224] The internal piers supported the track of a hand-operated double-girder overhead travelling crane. The floor was of York flag except where there were machine bases. Rectangular in plan, the engine house was about 66 feet (20m) by 40 feet (12m).

The building was completed in 1890, and by 1892 the steam pumping engines had been installed, three down each side of a wide central aisle. Each engine was capable of delivering up to 250 gallons per minute into the high-pressure mains. They were in use until 1956 but little of them now survives in the engine house. Brackets along the walls probably supported the main steam pipes. Two pipes passing through the engine house from its basement to the roof-top tanks, against the west wall, were part of the circulating delivery mains. Water from the underground reservoirs passed through the surface condensers of the steam engines and up through these two pipes to the roof-top tanks. The surviving pipes were disconnected after electrification. The main suction pipes were retained after electrification and could be seen on either side of the west door to the engine house. The two air vessels on either side of the main door at the east end of the building were also connected to the main suction pipes. This pipework is discussed later.

In the 1920s two Mather & Platt pumps were installed in the central aisle, each capable of delivering 500 gallons per minute. The sets were mounted on concrete blocks supported by rolled steel beams spanning the gap between the steam-engine beds above the high-pressure mains. The concrete beds were still visible in the engine house as were circular holes, infilled with concrete, through which the connections to the main suction pipe had been made. The connections to the high-pressure mains had been through rectangular holes, similarly infilled with concrete, between the two concrete beds. At each end of the concrete beds there were circular marks on the York flag floor probably where control pedestals and starters had been positioned. Each pumping set had a central electric motor with a turbine pump at each end. The three-phase induction motors used a 6000 volt supply and ran at between 1480 and 1500 rpm. By the 1950s when the steam plant was replaced, the Mather & Platt pumps had been out-of-use for several years. As part of the electrification they were refurbished and provided with new manual starters, in order to provide service for the duration of the refurbishment work. Afterwards their new starters were used for the new Pumps 1 and 2, and the Mather & Platt sets were removed, though the date of their removal is not known.

Installation of the electrically driven pumps was undertaken in two stages. Firstly the two 500 gallons per minute pumps, Nos 1 & 2, were installed together with one of the variable speed pumps. Then the rest were installed. Pumps Nos 1 to 5 were in the engine house. They are described in detail later. The pumping sets were mounted on plinths built up on the old engine beds. The sixth plinth in the engine house, the middle plinth on the south side, was used to carry switchgear, including monitoring equipment for Pumps 1 & 2 and 8 & 9 (thought to be the Mather & Platt pumps), and control equipment, including circuit breakers and Buckholz alarms, for the two 800kVA transformers out in the yard. Other electrical equipment, including fuse boxes and switch boxes, was mounted on the piers dividing the engine house from the turbine house. Starter cabinets and Ward Leonard equipment for the variable speed sets were placed alongside the sets themselves, against the north wall. The manual starter pedestals for Pumps 1 & 2 were placed against the west wall, on opposite sides of the door leading to the boiler house. All starter resistors and rheostats were placed in the basement.

Initially the two large weight-loaded accumulators controlled the starting and stopping of the smaller constant-speed pumps. A flexibility box, which was still in situ, allowed the three constant speed pumps to be switched to be controlled by either accumulator. Later the large accumulators were disconnected. On a visit in 1976 they were seen to be partially broken up. The three pumps were thereafter controlled remotely, by hand, from the control room.

A small pilot accumulator was placed in the north-east corner of the engine house to control the two variable speed pump units, via Ward Leonard equipment. This accumulator is described later. Next to the accumulator, on the north wall, was fire-extinguishing equipment similar to that in the turbine house, described later.

In the south-west corner a battery of twenty-four 2-volt cells provided back-up power for telecommunications equipment. On one of the piers on the south side, Evershed No-Flote equipment was mounted together with selection controls for the tanks and tank pumps.

The overhead travelling crane was of 5 tons SWL. The main girders were of the fish-belly type made up of riveted wrought-iron sections. Along each side were wooden walkways. Access to the walkways was originally by ladder at the south end. Installation of electrical switchgear cabinets precluded its continued use and it was cut to a very short section of a few rungs. An access platform was installed on one of the piers, making use of a redundant wall bracket which had once supported the main steam pipe.

On the east wall there were two tall indicator boards, calibrated from 0 to 23 feet (0 to 7m), which were used to show the positions of the two accumulators. Another indicator protruded above the pilot accumulator cabinet to show that accumulator's position. On the west wall two more large indicator boards, calibrated from 0 to 9 feet (0 to 2.7m), showed the depth of water in the clean water roof-top tanks. Below these were a couple of electric bells whose purpose is not known, but they might have been low-water alarms.

Along the south wall two of the four arched openings between the engine house and the turbine house were filled by automatic starting equipment cabinets for the constant speed pumps and the tank pumps.


A ladder through a hole just east of Pump No 1 gave access to the basement. Much of the floor space of the basement was taken up by the six engine beds. The original engine beds consisted of bases of a coarse mass concrete surmounted by large blocks of sandstone, probably from a Yorkshire quarry. The tops of the sandstone blocks were about 13 inches (0.33m) below the floor level of the engine house. It seems likely that any gap between the steam engine frames and the floor were spanned by chequer plate. When the electric pumps were installed use was made of the steam engine beds which had to be raised to engine house floor level and extended to receive the concrete engine bases. This was done by using a fine mass concrete. The vertical shuttering had left its mark on the blocks, as the horizontal shuttering had left marks on the earlier coarse mass concrete. Crow holes used to give access to holding down blots in steam days had been roughly infilled with concrete. The Mather & Platt turbine pumps were supported on concrete blocks, strengthened with rolled steel beams, over the centre aisle. These blocks were mounted on rolled steel beams with their ends embedded in concrete piers built onto the sides of the concrete bases of the steam engine beds. The piers were of a fine concrete with horizontal shuttering marks.

Down the centre of the basement, between the two rows of engine beds, ran the two high-pressure mains. They were at floor level with branch pipes from each of the pumping sets above. These appeared to use the same junctions as had the steam pumps, but in two cases the pipework had had to be extended to a new position for the electric pump. On the middle south block there was no longer a pump so the branch had been blanked off. Similarly branch pipes from the Mather & Platt turbine pumps had been partially removed and blanked off. Each junction was protected by manually operated valves. In addition there were a number of pressure relief valves which would open if the pressure in the mains increased above a pre-determined level. Two small pipes branched from each of the two mains and then joined to form a single feed to the pilot accumulator. The two mains looped round through the turbine house basement and out under the yard coming back past the accumulator tower. There were branches to each accumulator. As they looped through the yard they crossed so as to form a single pipe. East of the turbine house two mains branched off towards the street (Wapping Wall). A third branched from the west end passing through an extension to the basement under the corridor connecting the engine house to the boiler house. A 4 inch (0.1m) branch pipe ran parallel to it. These two pipes were uncovered out in the west yard, near the gate into Milk Yard, during excavation work for the foundations of a new building. They almost certainly carried on up Milk Yard to the junction with Garnet Road, where the London Dock Company once had a hydraulic pumping station. Here they connected onto the dock system.

The two branches of the main suction pipe ran at shoulder height down each side of the central aisle of the basement. They connected at a gate valve at the east end. On each side of the gate valve vertical branches connected to the two air vessels in the engine house. Beneath the air vessels were brick supporting piers. There were connections from the main suction pipes to each of the pumping sets.

Along each side wall there were ventilated boxes containing resistor banks, the starter resistors for the motors. Most were in wall-mounted boxes but a few rested on the floor on legs or, in two cases, on a frame. Against the west wall there were four large boxes with semi-circular lids. These were the liquid starter resistors for the large pumps, Nos 1 & 2. On a few of the piers there were brackets which presumably once supported pipes. On two of the piers of the north wall there were small pulleys, the purpose of which has not been determined.

There were cable runs along most of the passageways between the engine blocks. These have not been recorded except photographically. A number of holes through the engine house floor and in the wall between the engine house basement and the turbine house basement took cables, conduit, trunking and wiring. A large hole through the east wall, south of the pipe tunnel, just below the engine house floor, was for the main cables and control cables from the transformers out in the yard.

In the middle of the east wall of the basement a segmental arched opening led to the pipe chamber at the side of the accumulator tower. The floor of the pipe chamber was higher than the floor of the basement. The two high pressure mains rose into the chamber, becoming closer together at the same time. Inside the chamber, and under the arch, the pipes were carried on brick piers. The south pipe was carried higher than the north pipe so that a branch from it to Accumulator No 2 could pass over the north pipe. The pipe chamber was originally covered with iron plates.

At the west end of the engine house basement a tunnel ran under the corridor from engine house to boiler house. Stairs led up into the corridor but were no longer accessible from above. To the right of the stairs there was a short narrow tunnel, blank at the end. The high pressure pipes to Milk Yard ran along it and through the end wall. This area had been used as a store. Some shelving survived.

There were two doorways through the dividing wall between engine house and turbine house basements, one at the east end and one at the west end. The floor level in the engine house basement was about 17 inches (0.44m) higher than that of the turbine house basement. Just east of the door at the east end there was a circular pipe hole through the dividing wall. It was blocked on the engine house side by a section of wall from the east, which somehow looked, but could not be, original. Before electrification either the circulating suction main or the circulating delivery pipe probably passed through this hole. It seems more likely that it would have been the former as this pipe would have originally emerged from the underground reservoir at this point, before the reservoir was shortened by the building of the turbine house.

At the west end, just west of the west doorway, there was a pipe arch through which a number of pipes passed. These are discussed later.


In the 1920s an extension was built on the south side of the engine house. This was the turbine house. It was partly built over the eastern underground reservoir which was consequently shortened, a new wall being needed between the reservoir and the turbine house basement.

The turbine house was rectangular in plan, about 63 feet (19m) long by 22 feet (7m) wide, slightly shorter and about half the width of the engine house. It was not as tall as the engine house. In the east gable a wide triple door through which equipment could be brought was provided. There was a single door in the west gable. The dividing wall between the engine house and the turbine house was opened out with intermediate piers supporting longitudinal arches. Unlike the engine house, the turbine house had a dado of green glazed bricks. The roof trusses were of angled section steel. The roof had a raised longitudinal lantern. The red floor tiles appeared to date from the 1950s electrification.

The turbine house was equipped with a hand-operated single-girder overhead travelling crane with two blocks of different capacities. The safe working loads of these were not displayed. The crane tracks were supported on rails on horizontal rolled-steel beams. The beam on the south side rested on brick piers, corbelled out at the top. The beam on the north rested on a series of rolled steel stanchions which passed through the floor to basement level. The corner supports, both north and south, were of concrete or brick encased with concrete, square in plan, and rested on cantilevered concrete blocks in the walls some 6 feet (1.8m) above basement level.

The turbine house, as its name suggests, was built to house the Parsons multi-stage steam turbine. This stood on a girder platform supported on a framework of concrete piers and arches in the basement, with several transverse beams, some deep flanged rolled steel beams, others re-enforced concrete. There were also some rolled steel beams with smaller flanges. The turbine was arranged longitudinally down the centre of the house. A bracket in the west gable suggested that steam was supplied via an overhead pipe, a branch from the main steam pipe in the engine house. On the other hand there was evidence of a pipe at basement level from the boiler house. The condenser was almost certainly in the basement, but its exact location is not known. All other pipework was at basement level.

The eleven stage steam turbine ran at 8000rpm and drove a nine stage high lift turbine pump running at 1550rpm through single reduction gearing. The pump was a Mather & Platt Plurovane type. The steam turbine was use from April 1926 until May 1952, though there were periods during World War II when it was idle. It was removed to make way for two constant speed electrically driven three-throw ram pumps. These are described later. The specification drawing shows the two pumps closer to the west end of the turbine house than they were actually installed. [Metropolitan Archive B/GH/LH/06/30] The two 25kVA transformers are shown at the east end of the turbine house, whereas they were actually installed behind the 800kVA transformers out in the yard. The specification drawing does not show the control room at the west end.

The pumps were mounted on low concrete plinths resting on the old steam turbine supports. All pipework was in the basement but valve covers in the floor allowed control of the valves from the turbine house. Starter cabinets for all three, 200 g.p.m. constant speed pumps plus the three tank pumps were located in the centre arches of the dividing wall between turbine house and engine house.

In the south-east corner there was a cabinet containing monitoring equipment by Evershed. Next to it four metal floor plates could be raised to give access to the top of the Rotoklene strainer unit in the basement. An iron staircase to the west of this gave access to the basement.

Further along the south wall was the fire extinguishing equipment for the turbine house, which is described later. On two of the piers were depth indicators for the underground reservoirs. A float mechanism in the reservoir was attached to a wire which passed through the wall of the turbine house basement and round pulleys to the top of a wall-mounted box. The wire passed over a pulley and into the box where it was attached to a sliding contact. The contact could engage with fixed contacts which were electrically connected to the indicators in the turbine house. Each indicator had a column of lights against which a depth was written. Thus the light which was on gave the depth of water in the reservoir. The top and bottom lights were interconnected with electric bells which presumably sounded different notes to show that the reservoir was either full or nearly empty.


Access from the turbine house to its basement was via an iron staircase next to the south wall towards the east end. A wooden partition which protected the stairwell in the turbine house appeared to be a later addition.

The basement was largely open space with concentrations of pipework at each end and along the north and south walls. The central section had re-enforced concrete stanchions supporting a central concrete arch and transverse beams, some of concrete others of rolled steel. This structural concrete was built to support the Parsons' steam turbine and later supported the two electrically driven pumps. The south wall of the basement was a false wall, being built out from the main wall to support the various beams. In the south-west corner there was an alcove through which one of the circulating suction pipes passed. The by-pass pipe which connected the two reservoirs via the circulating suction pipe passed through the south wall at another alcove in the false wall. This alcove did not extend the full height of the false wall but had a low, flattened arch.

The circulating suction pipe ran along the west wall and through into the engine house basement. A branch from it ran along the south wall to the east end of the basement. Both branches were supported on concrete piers and short iron pillars, 2 feet 6inches (0.78m) high. Broken sections of concrete piers suggested that alterations had been made to the pipe layout here.

In the south-east corner stood a Rotoklene strainer unit built by Ashworth & Parker of Bury. By 1926 the Torrent filters, in the filter house, had been taken out-of-use. Heavier solids were deposited in the Shadwell Basin. Passing the water through gratings and small mesh screens was found to be sufficient to remove any sediment. Just when the Rotoklene strainer unit was installed is not known. Provision was made for the Rotoklene strainer to be sluiced out with clean water from the roof-top tanks, via the main suction pipe.

Two tank pumps, numbered 1 and 3, were located near the east end of the basement. They were both small centrifugal pumps driven by electric motors and stood on two small concrete plinths. They were made by Mather & Platt. The location of the third tank pump is not known.

Two capacitor cabinets stood by the east wall on a low plinth. These were provided to even out supply from the two 800kVA transformers.


The seven three-throw ram pumps were made by the Hydraulic Engineering Company of Chester with electrical equipment by Metropolitan Vickers, Trafford Park, Manchester. Pumps 1 to 5 were in the engine house and were mounted on concrete plinths built onto the original steam engine beds. Pumps 6 and 7 were in the turbine house, on plinths which utilised the concrete base for the Parsons steam turbine.

Pumps 1 & 2 were constant speed pumps each delivering 500 gallons per minute. They were driven by three-phase ac, slip-ring induction motors, with manual starters. The starting resistors were immersed in oil baths. These large machines probably took the base load of the pumping station. Four large resistor banks for these machines stood in the basement of the engine house, against the west wall. They were housed in rectangular steel boxes with semi-circular lids. The purpose of the oil was to quench any sparking. The starter pedestals were mounted against the west wall of the engine house, close to the pumps they served.

Pumps 5, 6 & 7 were also constant speed pumps each delivering 200 gallons per minute. Like Pumps 1 and 2, they were driven by three-phase ac, slip ring induction motors. But they had automatic starters, housed in cabinets in the arches between the turbine house and the engine house. The air-cooled resistor banks were placed against the south wall of the engine house basement, some being wall-mounted.

Pumps 3 & 4 were variable speed pumps, each delivering up to 200 gallons per minute. They were driven by shunt wound dc motors. The dc current was derived from motor generator sets mounted on the same bed plate. The pumps were mounted on plinths built up on former steam engine beds. These pumps had Ward Leonard equipment controlled by a pilot accumulator in the north-east corner of the building. Automatic control cabinets were placed alongside the pumps, against the north wall of the engine house. The accumulator turned a small shaft which was connected to a cabinet containing sliding contacts arranged in rings around discs which were turned by the shaft from the accumulator. Air cooled starting resistors were wall mounted in the engine house basement, on the north wall beneath the pump controls.

Pumps 5, 6 & 7 were identical. Each was mounted on a concrete plinth, roughly rectangular in plan, but shaped to accommodate the machine bedplate. The cast iron bedplate was made up of three sections, one carrying the motor and its shaft, one carrying the gearing, and the third carrying the three-throw pumps. The motor was mounted at one corner of the plinth with the sliprings projecting beyond the plinth. Pumps 6 & 7 were still fitted with Pyrene automatic fire-extinguishing apparatus over the sliprings. This is described later.

The motor shaft was extended across the plinth via a coupling box to a pinion, and was carried on SKF roller bearings. The pinion, 9 inches (0.23m) in diameter, drove a spur wheel, 6 feet 3 inches (1.9m) in diameter, through double helical gear, giving a reduction of 8.33 to 1. The motor speed was 730rpm, the spur wheel speed 87.6rpm. Measurements here are approximate because of the difficulties of access to the gearing.

The spur wheel was at one end of the three-throw crank shaft which drove the pump plungers via crossheads. Gearing, crankshaft and crossheads were all fully enclosed with moveable covers. A lubricating oil reservoir was mounted on the crankshaft cover with lubricating lines drip feeding oil to each crosshead. The pump cylinders were contained in a rectangular block at the opposite end of the plinth to the motor. The main suction pipe was brought up through the floor from the basement and under a valve box mounted on the end of the pump block. The delivery pipe was connected to a second valve box mounted on top of the pump block, and led down to the high pressure mains in the basement. A spring loaded pressure relief valve was fitted at the end of the high pressure valve box and acted as a safety valve, should the pressure rise above a pre-determined value. A small pipe leading from it took its exhaust water to a drain.


In the direct current (dc) motor the stationary part of the motor comprises an electro-magnet to provide the magnetic field. The armature forms the moving part of the motor. In order to turn the dc motor current has to be introduced to the rotor windings and has to be reversed every half-turn. This is done using a commutator and carbon brushes. The dc motor has the great advantage of providing a high torque at any speed. It is, therefore, the ideal variable speed motor. Its weak point is the commutator.

In the alternating current (ac) induction motor the magnetic field is provided by the rotor (moving) windings in which the current is induced from the stator (stationary) windings. Thus it is the magnet that is made to turn, not the wire carrying the current. Because current in the field windings is induced the problem of passing current to the moving part of the machine is overcome. The three-phase ac induction motor runs at a constant speed.

In order to start a dc motor the field and armature windings are put in series with a large resistance. The resistance is gradually reduced and the windings are switched into parallel mode, with the resistance reinstated and again gradually reduced. The final mode is to reduce the current to the field windings, known as ‘weak field’. In this mode the motor runs at its highest speeds. In the dc motors of Pumps 3 & 4 it was this process that was controlled by the pilot accumulator and the Ward Leonard gear.

With the ac induction motor, resistance is put onto the field windings in order to start the motor. This is done via a slip ring, which, like the commutator of the dc motor, uses carbon brushes to form the contacts between the moving rotor and the stationary resistors. The resistance is gradually reduced as the motor approaches synchronous speed. With Pumps 1 & 2 this was done manually from the control pedestals. With Pumps 5, 6 & 7 it was done automatically using relays which incorporated clockwork timing devices.

Pumps 1 and 2 had manual starters. The starter pedestals stood near each pump, against the west gable wall. They were about 3 feet (0.9m) high with a sloping top face and a handle to turn on the side.

Two equipment cabinets on the central plinth in the engine house were labelled ‘Pump No 1’ and ‘Pump No 2’. They were built in three sections. Each lower section, larger than the other two, contained an oil bath circuit breaker to deal with the 6600 volts supply to the pump motor. The supply was taken directly from the LEB sub-station, independent of the supply to the transformers. The smaller section above included relay controls to adjust load-factors etc, and instrumentation. The top section, which did not have an opening door, had one instrument. The section probably housed bus bars.

The starting resistors for Pumps 1 & 2 stood in the basement, against the west wall, immediately below the starter pedestals. The resistors were immersed in oil baths in orange coloured rectangular steel boxes which stood on the floor in the basement. Each box had a curved lid. One lid had been removed and leaned against an engine bed, the others were in situ. Each pump had two resistor boxes.

Pumps 3 & 4 each had a pair of associated equipment cabinets, or bays. One pair stood against the north wall of the engine house near Pump 3, the other pair stood in a similar position near Pump No 4. Also associated with these pumps was another cabinet which stood next to the pilot accumulator, to which it was physically attached by means of a shaft which revolved. The shaft was turned by the rising and falling of the accumulator by means of a chain attached to the weight table. In the cabinet the shaft turned a disc which had copper contacts arranged around its edge. As the disc turned these passed or stopped on fixed contacts. In this way the accumulator controlled the speed of Pumps 3 & 4, and hence the amount of water they delivered into the high-pressure mains.

One bay of each pair housed automatic starting equipment for the ac motors which drove the dc generators. The other bay housed Ward Leonard equipment which controlled the pump speed from signals from the accumulator cabinet. Starting resistors were housed in boxes affixed to the north wall of the engine house basement, just below the pairs of cabinets. They were about 4 feet (1.2m) above floor level and were air-cooled.

The automatic starter cabinets for Pumps 5,6 & 7 were placed in the two centre arches between the engine house and the turbine house. There were six cabinets, one each for Pumps 5,6 & 7 and one for each Tank Pump. The starters for Pumps 5, 6 & 7 were all identical. To start one of these pumps all that had to be done was the pressing of a button on the front of the starter cabinet. The starting sequence was then controlled by relays within the cabinet. Provision was made to switch control from ‘local’ to ‘remote’. This enabled the pumps to be started by pressing buttons on a panel in the control room instead of the starter cabinets. The starting resistor banks were housed in boxes in the engine house basement. Some were attached to the south wall, about 4 feet (1.2m) above ground, others stood on low racks, and at least one stood on its own legs on the floor. All were air-cooled. It was not possible to distinguish between those resistor banks used for the main pumps and those used for the tank pumps.

The Tank Pump automatic starters were simplified versions of the main pump starters. They too could be switched to remote control, in the control room.


Pumps 6 & 7 were fitted with a Pyrene fire extinguisher system. The Pyrene Company was of American origin. A catalogue dated 1910 gives their address as 34 West Thirty Third Street, New York. By the 1930s they had established offices and works on the Great West Road at Brentford. Pyrene was a trade name for bottled compressed carbon dioxide and was eminently suitable for an electrical environment.

Two bottles of Pyrene stood against the south wall of the turbine house. They were fitted with valves which could be opened by means of falling weights. The weights were kept in position by wires which ran in pipes to the tops of the pump motors. Over the motors the wires ran in a square above the ventilation grilles in the top covers and were fitted with fusible links. A second set of pipes was designed to feed carbon dioxide to pairs of nozzles which were directed at the rotor windings of each motor. Thus, if the temperature of the motor became too high the fusible links would break. This would release the weights, opening the valves on the gas bottles, killing any fire. In addition the valves could be opened manually.

A similar system was used in the engine room. Two carbon dioxide bottles stood against the north wall of the engine house, apparently for two of the pumping sets only. Presumably there was another system somewhere for the other pumps.

There were a number of fire points but any equipment kept at them had been removed.


A wood and glass partition in the west bay of the turbine house formed a control room from which the pumps and Wapping could be controlled. A remote control supervisory system also allowed the other LHP pumping stations, at Rotherhithe, Grosvenor Road, City Road and East London (formerly East India Docks), to be remotely controlled from Wapping via GPO lines. The system was installed in the 1960s. It is described in detail in Appendix 6.


Agreement was made with the Borough of Stepney in January 1921 for a supply of electricity for the Mather & Platt turbine pumps. This presumably came from the local power station in Limehouse. Later the London Electricity Board (LEB) established a sub-station just inside the yard, opposite the Superintendent's house. Cables were taken from the LEB sub-station through a tunnel which entered the engine house basement, just below the ceiling, to the south of the hydraulic main tunnel. The cables were slung under the ceiling and terminated on isolation switches housed in cabinets in the centre of the engine house.

The isolation switches were at each end of a group of switchgear cabinets on the north side of the central plinth in the engine house. The top section of these cabinets probably housed 6600 volt bus bars and included some instrumentation including Buckholz alarms and trips for the 800kVA transformers. The lower sections were divided into two. The lowest section of each of the six middle cabinets contained a circuit breaker. They all appeared to be identical. The section above housed more instrumentation and relays, some having wheels and pointers which allowed settings to be altered. The two middle cabinets were devoted to the two 800kVA transformers. The two cabinets to the left were for Pumps 1 & 2, those to the right Pumps 8 & 9. These latter are thought to have been the two Mather & Platt turbine pumps. Pumps 1 & 2 took current directly from this 6600 volt supply. Presumably the Mather & Platt pumps also used a 6600 volt supply.

The other side of the circuit breakers in the 800kVA transformer cabinets were connected to the high tension (HT) side of the 800KVA transformers by cables which were taken back through the tunnel. The tunnel was enlarged into a cable chamber behind the transformers. From the low tension (LT) side of the transformers cables ran back through the tunnel, terminating on a second set of circuit breakers on the south side of the central plinth in the engine house.

The supply to each 800KVA transformer was a three-phase, delta connected, 6600 volts, 70 amperes supply, carried on a single cable connected to the east side of the transformer. Five-position tap changers were fitted. Output, at 440 volts, 1050 amperes (full load), three-phase, star connected, was carried on three separate cables terminating on the west side of each transformer. The transformers were oil-cooled and were fitted with Buckholz devices, the Buckholz alarms being, as noted above, on the HT side bus cabinets in the engine house.

The LT circuit breakers were in cabinets on the south side of the central plinth in the engine house. The lower sections of these cabinets housed the circuit breakers themselves, which were immersed in oil filled pots. The oil served to quench any arcing. The upper sections housed the bus from which the 440 volt supply to Pumps 3,4,5,6 & 7 was taken. Access to the bus was controlled by Castell keys to ensure that the circuit breakers were in the ‘off’ position before anyone could open the door. In an oil circuit breaker the contacts are in special pots in an oil bath. When the contacts are broken there is a huge spark which is quenched by the oil. The shape of the pot causes the spark to draw in the oil, which puts it out.


The weight loaded hydraulic accumulator was central to the success of any hydraulic power system. It performed three functions. Firstly it regulated the pressure in the system. Secondly it provided some storage of energy. And thirdly it could be arranged to control the pumping engines. It also provided a fourth function by taking up any back-pressure from the network, thus protecting the pumps, though this was also be done by the provision of momentum valves on the mains.

Two accumulators were installed in the tall accumulator tower at Wapping. That nearest the pumping station was designated Accumulator No 1, the other was designated Accumulator No 2. They were both connected to the high pressure ring main via manually operated valves so that either could be isolated for maintenance purposes.

Each accumulator had a 20 inch (0.5m) diameter ram of 28 feet (8.5m) stroke, and was weighted to 850 psi (57 Bar). [The Engineer article of 1893 quotes the pressure to be 800 psi whereas in The Engineer article of 1926 the pressure is given as 850 psi]. The weight cases were hung from the crossheads at the tops of the rams, and were filled with a cheap but heavy, ballast, said to be slag but judging from the fill in the bottom of the tower it might have been Thames ballast. Arcs have been cut into the lower sections of the end walls to accommodate the weight cases.

The crossheads ran on guides set diagonally in the tower, an arrangement often used by the Hydraulic Engineering Company of Chester, in contrast to the Armstrong arrangement where the guides were set in such a way that the crosshead is at right angles to the tower walls. The guides have clearly been altered at some stage. The present guides consisted of channel section steel rails welded to ties bolted to the guide timbers. Holes in the guide timbers showed that the original guides were probably of the more conventional angle iron and plate variety.

The accumulator tower showed signs of having been altered on the south side. A section around the door appeared to have been removed and replaced at some time. This section was probably left incomplete while the accumulators were installed, and was then bricked up to form the doorway. At a later date the accumulators appeared to have been replaced with new ones, hence the new guides. The doorway was enlarged to allow this to happen, and then rebuilt to its present form. When the accumulators were removed in the 1970s they were cut up inside the tower and removed in small pieces. The alterations could have taken place in the 1920s when the weights were altered to give a pressure of 850 psi. This would have involved increasing the weight in the weight-case by 7 tons.

The foot of each accumulator was below ground level, about the level of the pipe chamber on the south side of the tower. Inside the tower a ladder ran up the centre of the north wall giving access to a loft. At the top of the tower heavy timber buffers were provided to prevent over-run of the accumulators.

When the electric pumps were installed it was realised that energy storage was unnecessary since the variable speed pumps could react quickly to changes in demand. Pressure regulation could be achieved using a small pilot accumulator which could be linked to the pump motors via Ward Leonard gear to control pump speed.

The small pilot accumulator was installed in the north-east corner of the engine house, protected by a large metal cabinet. The accumulator cylinder rested on small plinth in the corner of the engine room basement. The moving part of the accumulator was above, on the ground floor. The guide supports, of channel section rolled steel, ran up from the basement plinth, through a square hole in the floor, to the top of the cabinet. The stainless steel guides were round in section and were attached to the front of the guide supports. The ram was 3 inches (0.08m) diameter by 10 feet (3m) stroke. On top of it there was a crosshead which supported a square frame on which the weight, consisting of blocks of cast iron, rests. The total weight on the column of water was about 2.68 tons, giving a water pressure of 850 psi (57 Bar).

Attached to the side of the accumulator, a sprocket and chain mechanism turned a shaft as the accumulator rises and falls. The shaft was connected to an adjacent electrical equipment cabinet and controls Pumps 3 & 4 via Ward Leonard gear. In order to bring a degree of fine control to the system chain was lifted out of two barrels on the basement plinth as the accumulator rises. The size of chain increased as the accumulator nears the top of its stroke.

The pilot accumulator was connected to the two parallel sections high pressure main in the engine house basement by a small bore pipe which bifurcates. There was a manual valve on this pipe at the base of the accumulator. Above the cabinet on the ground floor an indicator showed the position of the accumulator.


The water supply for the pumping station was taken from two sources. In the yard, to the west and south of the boiler house, a well was sunk to the London Clay, yielding a good supply of water from the gravels. In addition, a syphon pipe took water from the Shadwell Basin of the London Docks to the well. In December 1906 a second 12 inches (0.3m) cast iron suction pipe was ordered. Two hydraulically powered pumps were placed in the top of the well to pump water from there into two of the roof-top tanks over the boiler house. The hydraulic supply for these pumps came from the company's main rather than from Wapping itself. This meant that when the Wapping pumping station was shut down overnight, these pumps could continue to operate. The well-head was still visible, and clearance work on the site to the west revealed some of the associated pipework.

From the ‘unfiltered water’ tanks, the water passed down through filters, in the filter house, south of the boiler house, into two underground storage reservoirs, under the yard. In the yard there were two large manholes giving access to the reservoirs. There were also a number of stop valve covers which were clearly marked. It was possible to close these valves so that the filters could be washed through with filtered water. One of the underground reservoirs was shortened when the turbine house was built.

In the turbine house, a pipe, 15.5 inches (0.39m) external diameter, entered the basement in the south-west corner, crossed the building and entered the engine house basement through the dividing wall. Branching off this another pipe ran along the south wall, supported on iron posts. About half-way along a branch pipe went back through the wall. There was a gate valve on the branch. Near the iron staircase into the basement, the pipe turned sharply to the north and then to the east to pass round the Rotoklene strainer. A second branch, with gate valve, was connected to the top of the Rotoklene strainer. Between the Rotoklene strainer and the east wall a second pipe entered the basement and joined the first. There was another branch from here into the top of the Rotoklene strainer. A third connection to these pipes came from lower down the Rotoklene strainer. Gate valves were positioned to allow control of water through these pipes.

The pipe looped back and ended about a third of the way along the north wall. Here there was a connecting pipe, with gate valve, to another pipe at a higher level. There were also connections to two Tank Pumps and a third blanked-off branch. This pipe was the circulating suction pipe, which brought water from the underground reservoirs to the Tank Pumps, for pumping up to the roof-top ‘filtered water’ tanks. The blanked off branch might have been the connection to the third Tank Pump, though there was no other evidence of it. There was no concrete plinth on which it could have stood.

The higher level pipe was the main suction pipe, taking water from the roof-top tanks to the main pumps. The connection to the circulating suction pipe probably enabled water to be diverted from the roof-top tanks to the Rotoklene strainer, in order to flush it out occasionally.

The two Tank Pumps were small, electrically-driven centrifugal pumps, by Mather & Platt. They sat on small concrete plinths on the floor of the turbine house basement. They were served by branches from the circulating suction main, which entered the bottom of the pump. The delivery pipe came out of the top of the pump on the opposite side and went up to the circulating delivery pipe which ran down the centre of the basement, at high level, supported on tall iron posts. The circulating delivery pipe was 11.5 inches (0.29m) external diameter. In steam days, water from the underground storage reservoirs passed through the surface condensers of the steam engines. The steam turbine also had a condenser which, presumably, also used water circulating between the reservoirs and the roof-top tanks. On electrification, the condensers were removed and the pipework altered accordingly. At the west end of the turbine house basement, the circulating delivery pipe passed through the wall and ran underground to a point near the stump of the chimney. Here it could be seen on the outside of the building as a vertical pipe running up the wall from below ground to the roof-top tanks. Back in the basement, a branch from the circulating delivery pipe into the engine house basement, where it joined the pipes running up through the engine house, had been blanked off. The pipe to which it connected was in situ through the wall, in the north-west corner of the basement. Against the west gable wall in the engine house two pipes entered from the basement, one towards the north end of the wall, the other towards the south end. Both pipes ran up the wall to a high level where they turned west through small arched openings the roof-top tanks. In the basement these two pipes had been disconnected. They were the circulating delivery pipes from the surface condensers.

On each side of the west door to the engine house there was a pipe which went down into the basement. The one on the north side had a flow-meter attached to it. The one on the south had something similar but there was a cover screwed onto it which might conceal another flow-meter. These were two arms of the main suction pipe. They ran along the ceiling of the connecting corridor from the boiler house to the engine house. In the boiler house they were difficult to see in the gloom of the roof, but they crossed the boiler house and had branch connections to the roof-top tanks. The connections had been altered when the distribution of filtered and unfiltered water in the tanks was changed.

In the basement of the engine house the two pipes were carried along each side of the central aisle to the east end, at about 6 feet (1.8m) above floor level. At the east end they each joined a connecting pipe at a tee section on top of a brick pier built against the end wall. About mid-way along the connecting pipe there was a gate valve which could be closed from the engine house using a key, through gearing. The outer ends of this connecting pipe turned upwards into the air vessels which stood on each side of the east door of the engine house. The air vessels were 8 feet 8 inches (2.64m) tall and 25½ inches (0.65m) in diameter. Each had a relief valve on the top and, at the side, a glass pipe with try-cocks to enable checks to be made on the water level in the air vessel.

By each engine bed there was a branch pipe from the main suction pipe to the pumps. These had been lengthened to take account of the different positions of the inlets to the electric pumps compared to the steam pumps they replaced. There were blanked off connections to the Mather & Platt turbine pumps. A branch from the main suction pipe ran between the engine bed of Pump No 5 and the central plinth and through into the turbine house basement to serve Pumps 6 & 7. It ran along the north wall of the turbine house basement with connections to the two turbine house pumps. There was also a vertical connection down to the circulating suction pipe (see above). This was protected by a gate valve. One use of this connection has already been postulated. It could also have allowed the roof-top tanks to be emptied, the water flowing back into the underground reservoirs. At its west end this branch of the main suction pipe dropped to near ground level, where there was a blanked off branch. The pipe then zig-zagged behind the north-south circulating suction pipe to join it at a gate valve. There were, therefore, two connections between the main suction pipe and the circulating suction pipe.

In the engine house basement three small pipes branched from the main suction pipes. Two, which went to the ends of the turbine pump engine beds, might have been used to prime these pumps. The other ran up the tunnel at the east end, past the accumulators. As it had been cut off just inside the tunnel its destination and use is not known. Gate valves were provided at intervals, especially on branches, on all the low-pressure pipes.

The high-pressure hydraulic main comprised a double ring around the engine house and turbine house basements with branches to the accumulators and to the street mains. Two pipes ran along the floor of the central aisle of the engine house basement, turned south at the west end, and into the turbine house basement. Here they turned east and ran along the floor near the dividing wall then went through the east wall under the yard. At the east end of the engine house basement they went up into the tunnel and stopped at the far (east) end of it. There were connecting pipes from there to the other ends of the pipes which came out of the turbine house basement. These pipes were connected in such a way as to form one continuous pipe which passed through the basements twice in a double loop. There were two connecting pipes to the street mains from under the yard. Another connection went west, under the yard, from the north-west corner of the turbine house basement. Two more branch pipes, one 10 inch (0.25m) diameter, the other 4 inch (0.1m) diameter, left the engine house basement at the west end, through the tunnel under the corridor between engine house and boiler house, before disappearing underground. They were picked up again on the building site to the west of the pumping station, having been uncovered by removal of spoil. They left the site by the west gate heading up Milk Yard. These probably connected to the London Dock pipe network at the top of Milk Yard, the site of a former hydraulic pumping station belonging to the London & St Katharine Docks Company.

In the basements there were connections from each of the pumps. Two branches which ran down the centre aisle of the engine house basement formerly served the two turbine pumps. All connections were provided with stop valves, as was the main itself, so that any section could be isolated. The high-pressure main was protected by weighted relief valves at intervals along its length. Two small-bore branch pipes taken from each side of the main in the centre aisle of the engine house basement joined together to form the connection to the pilot accumulator. There was a stop valve near the accumulator for maintenance purposes.

In the east tunnel there was a connection, in situ, to the accumulator nearest to the building, with a stop valve on the branch. The connection to the second accumulator had been removed, but it too had had a stop valve on its branch. In the yard there were a number of stop valve covers of the usual LHP pattern, some with separate plates identifying the pipes they were on.

Over the years there have been many modifications to and renewals of pipework and some of this was reflected in the blanked branches to both high- and low-pressure pipes, particularly in the basement areas. All pipes were of cast iron in varying lengths. High-pressure pipes rarely exceeded 9 feet (2.74m) in length. Joints of low-pressure pipes were usually flanged. High-pressure pipes always had flanged joints, recognisable because they were oval in shape; low-pressure flanged joints were invariably round. High-pressure joints had gutta-percha or leather seals. Suppliers of pipes often embossed their names and a date on their pipes. Some high-pressure pipework in the basements bore the name Staveley. Their source was the Staveley Coal & Iron Co Ltd of Staveley Works, near Chesterfield, Derbyshire who specialised in the manufacture of pipes of all kinds.

Some of the gate valves also bore the names of manufacturers, these included Hamilton Woods & Co Ltd of Manchester. The high-pressure relief valves, whilst not bearing makers' names, were almost certainly supplied by the Hydraulic Engineering Company Ltd, of Chester. Similar valves appear in their catalogues.

The first main to be laid from Wapping Pumping Station was a 7-inch (0.18m) pipe along Wapping Wall and Wapping High Street to connect with the St Katharine Docks main in Nightingale Lane (now Thomas More Street). A 4-inch (0.1m) main was laid along Milk Yard to connect with the London Docks system at one of the Dock's pumping stations. Soon a network developed around Wapping, connecting at the west end to the existing LHP mains, at various times extensions and alternative connections were added. In early 1925, for instance, a new 8-inch (0.2m) main was laid from the pumping station, along Milk Yard and south down New Gravel Lane to connect with the existing main in Prussom Street. It cost £2400 and might have been the pipe which left the building from the west end of the engine house, running parallel to the 4-inch (0.1m) main to the docks.


The boiler house initially contained six Single Fairbairn-Beeley boilers built by Thomas Beeley of Hyde Junction Iron Works, Manchester. They had two barrels, one mounted on top of the other, with a steam receiver on the top. They were fitted with Vicars mechanical stokers. Coal was lifted, in a wheeled tub, by hydraulic hoist onto a gantry, and pushed by hand to the boilers where it was tipped into hoppers which fed the stokers. Behind the boilers were the economisers. Flue gases passing through the economisers heated the boiler feed water.

In 1923 the Fairbairn-Beeley boilers were removed and three Babcock & Wilcox water tube boilers were installed in preparation for the installation of the Parsons steam turbine. The western two thirds of the boiler house, and the roof-top tanks above, had to be raised to accommodate the new boilers. The contract for additional ironwork, including the columns, went to the Horseley Bridge & Engineering Company, in the Black Country. Two columns were increased in height whilst the other two columns were completely new.

The second set of boilers was removed on completion of the electrification of the station in 1956 and the boiler house was fitted out as a workshop. The former workshop became a store. After closure of the pumping station some of the equipment and machine bases survived for a while but all have now been removed.

In 1998 the boiler house formed a huge open space. In the roof various pipes could be seen, including the main suction pipes leading from the roof-top tanks to the engine house. There was also recently installed pipework for drainage of the tanks. The walls showed much evidence of earlier phases of the boiler house. Parts of the walls were covered with white glazed brick. There were several burned off rolled steel beams of various sizes and at various levels in the walls. A blocked doorway once gave access to the filter house. There was another blocked aperture in the south wall where the flue had gone through to the chimney. In the north-west corner there was once a doorway leading to a hydraulic lift shaft used to raise ash to ground level for tipping into carts for removal. The east wall had a number of blocked holes for pipework.

A number of bricks were recovered from the tunnel under the boiler house. These included firebricks marked Star Work Glenboig, Douglas and Best Stourbridge. The Stourbridge area of the West Midlands produced good quality firebrick which was used for boiler settings among other things. The Glenboig area of Lanarkshire was equally renowned, producing even better quality firebrick than Stourbridge. The Star Fireclay Works in Glenboig (NS 724687) operated from c1876 until c1982. The Douglas Fireclay Works of Dalry, Ayrshire (NS 294478) was in business from 1914 until c1983. [Brick, Tile & Fireclay Industries in Scotland, by Graham Douglas & Miles Oglethorpe, RCAHMS, 1993]

The Stourbridge bricks found under the boiler house might be from the original Fairbairn-Beeley boilers. The Scottish bricks were from the 1920s Babcock Wilcox boilers. Babcock Wilcox was, and indeed, still is, based at Renfrew in Strathclyde.


Coal was lifted over the wall from lighters in Shadwell Basin by hydraulic crane. A rectangular coal store was formed under the roof-top tank at the west end of the boiler house. There was a doorway into it from the boiler house. In the south-west corner was a concrete cover which protected a stairway leading to an undercroft. Entry to the stairway, which was blocked with breeze blocks, was in the western section of the filter houses. Further storage of coal was found under the open shed in the yard to the south of the boiler house.


The buildings adjoining the boiler house to the south were the filter houses. They were divided by the entrance stairs to the boiler house. Originally it was thought necessary to filter all water before use and Torrent filters were purchased from the Pulsometer Company in Reading. Water was pumped from the well into two of the roof-top tanks. From there it passed down through the filters and into the two underground storage reservoirs. By the 1920s the filters had gone out of use as it was realised that settlement of larger particles was taking place in the Shadwell Basin and that any other material in suspension could be removed by screening. After electrification, however, an Ashworth & Parker Rotoklene strainer was installed.

The east filter house had a blocked opening in the east wall, above floor level, apparently leading to the base of the chimney. The west filter house had a large blocked opening in the north wall. Again this was well above floor level, leading into the coal store. There is no evidence to suggest what its use was. To the west of this blocked opening, just below the ceiling, there was a circular pipe hole. There were various pipes in the two filter houses.

Water from the underground reservoirs was pumped back into the other two roof-top tanks before flowing down the main suction pipe to the pumps. The ram pumps were kept primed by the head of water which also prevented damage to the pumps in the event of discontinuity of the flow.

There were four roof-top tanks, each 9 feet (2.75m) deep and 21 feet 4 inches (6.5m) wide. They were lettered A to D from the west end. Originally tanks A and B held unfiltered water whilst tanks C and D held filtered water. After the middle two tanks were raised to allow installation of the Babcock boilers the arrangement was altered. Tanks B and C, the higher tanks, held unfiltered water and tanks A and D held filtered water. Thus the suction head at the pumps was kept the same as it had been before the alterations.


The original workshop was in a room of irregular plan, opposite the rest room with a store to the east. The wall between store and workshop was removed at some time. The workshop was moved to the boiler house in 1950s after the boilers had been removed and the original workshop became a store. Bearing boxes still in situ high on north and south walls indicated the position of line-shafting across the workshop. Marks on north wall appeared to indicate position of a motor, probably a hydraulic motor, with belt drive to a large pulley from which there would have been a belt drive across to the line-shafting. Individual machine tools would have been belt driven, with fast and loose pulleys, from the line-shafting. Various marks on walls could indicate positions of machines and benches. The floor had been re-screeded.


The mess room was located on the south side of the corridor linking the engine house with the boiler house. All fixtures and fittings had been removed. A photograph, taken by RCHME in 1987, shows wash-basins, cooker and a fridge in this room.


The large yard stretched from Monza Street to the west to Glamis Road to the east, round the south side of the pumping station. There was a narrow strip of land between the pumping station and the northern boundary wall. The western end of the yard, west of the well, was disused. The well was located in the yard just west and south of the boiler house. The top was capped with an octagonal chequer plate top on concrete walls. When excavations for the foundations of new buildings to the west of the pumping station were dug pipework around the top of the well was revealed.

South of the boiler house an open shed in the yard was used originally as a coal store. After electrification it was used as a pipe store. The two underground storage reservoirs were located under the yard. There were large two perforated cast iron grilles for ventilation of the reservoirs. A hydraulic travelling crane ran across the top of the reservoirs; the crane tracks were in situ. A number of valve box covers marked LHP could be found around the yard, many having secondary plates giving their usage. The yard was paved with setts.

To the east of the engine house the yard contained the transformer bays. On the south side, two bays, unroofed and open-fronted, divided by a wall, housed the two 800kVA transformers. The floor of each bay was raised above ground level. Along the back of the bays there was a cable chamber and behind that were two smaller bays in which the 25kVA transformers stood. At the east end of the yard was the LEB sub-station.


Wapping Pumping Station was the third large pumping station built for the London Hydraulic Power Company. After its enlargement in 1926 it was said to be the largest hydraulic pumping station in Britain (and probably the world), though Rotherhithe (also LHP) probably equalled it in size after the installation of a Parsons steam turbine there in 1927.

LHP was the largest and longest lived of the public hydraulic power companies. The others are listed in Table 1. Of the LHP pumping station buildings, three survive, Wapping, Rotherhithe and Kensington Court. The latter has been converted to domestic use. Rotherhithe appears to be empty and disused, though there have been schemes for its conversion to flats. Only Wapping still contains pumping machinery.

Of the other public companies' pumping stations only those at Machell Street, Hull and Water Street, Manchester survive, the rest have been demolished. The Hull pumping station was being used by a firm of motor mechanics. Its present use is not known. The Manchester building is now the Museum of Labour History. Neither contains any machinery.

There are still a few working water hydraulic systems, all using electrically driven pumps. All are small in scale. Two preserved pumping stations complete with steam pumping engines and accumulators, at Tower Bridge and the Boat Museum, Ellesmere Port are on public display. One disused pumping station, at Birkenhead Docks, is thought to contain its electrically driven three-throw ram pumps but the building was damaged by bombing in World War II. It is also of smaller scale than Wapping. Its future is uncertain. A hundred or so private pumping station buildings survive, most are empty though several are listed. Examples at Swansea, Bristol and Liverpool have been converted to pubs and restaurants. Another in Glasgow was used as an Indian restaurant for several years but is now disused.

Wapping is amongst the more important of the survivors.

Table 1 — Public Hydraulic Power Systems

Town Dates Pressure
Hull 1877 700 psi
London 1883-1977 750 psi
Liverpool 1887-1970 800 psi
Melbourne 1889 750 psi
Birmingham 1891-1932 700 psi
Sydney 1891 750 psi
Antwerp 1894 750 psi
Manchester 1894-1972 1120 psi
Glasgow 1895-1964 1120 psi
Buenos Aires


The main source of material relating to LHP and Wapping is in the Metropolitan Archives, and includes minute books, accounting material, technical material and photographs. The technical material includes the Engine Room Registers and the specification for the electrification of Wapping.

The Tower Hamlets Local History Library holds 178 drawings of Wapping Pumping Station, mainly detailed drawings relating to changes which have occurred through the years.

The National Monuments Record holds a series of photographs taken by the RCHME (now part of English Heritage).

Other sources are:

Acts of Parliament:

Professional Papers:

The National Archives, Kew

Closed Companies Files

  • BT31/465/1802 London Hydraulic Power Co Ltd (1860)

    Books and Pamphlets etc:


    Newspaper Articles:

    Journal Articles:


    Many people have helped in both the recording of Wapping Pumping Station and in the production of this report. Thanks are due to Ian James, Josh Wright, Jules Wright of Wapping Ltd and the Women's Playhouse Trust for allowing us unrestricted access to the site, to various GLIAS members — Colin Bowden, Derek Goddard, Chris Grabham, Dan Hayton, Sue Hayton, John Hinshelwood, Tom Ridge, Andrew Smith, Cherry Smith, Malcolm Tucker — for helping with the recording activities and for reading through the draft report, to Dave Gunning for his report on the remote supervisory control systems, to Terry Evans for explaining the starting arrangements of electric motors, and to the staff of various libraries and archives whose services I have used, particularly the Metropolitan Archives, the Guildhall Library and the Science Reference Library (which has since moved to the British Library site at St Pancras).

    APPENDIX 1 The Steam Pumping Engines

    The first steam engines were the atmospheric beam engines of Thomas Newcomen, the first of which was erected at Dudley in 1712. They were inefficient machines in which cold water was injected into the cylinder to condense steam to produce a vacuum to enable the pressure of the atmosphere to act on the top of the cylinder driving it down for the power stroke. The weight of pump rods on the other end of the beam pulled the piston back up as more steam entered the cylinder. Despite their inefficiency these beam engines were widely used for pumping water from mines.

    In 1769 James Watt patented the separate condenser which allowed the cylinder to be jacketed and kept hot, greatly increasing the efficiency of the beam engine. Watt entered a partnership with the Birmingham entrepreneur Matthew Boulton, and Boulton & Watt engines gained a high reputation. Cornishman Richard Trevithick introduced new improvements in the early 19th Century, notably the use of high pressure steam, or strong steam as it was known. He also built the first horizontal engine in 1802.

    Later in the century marine engineers began using inverted vertical engines in order to drive ship’s propeller shafts. Once compound versions appeared they were also used for water-pumping. Compounding had been tried earlier in the century when both Woolfe and McNaught compounded beam engines but the compound came into its own during the 1870s. A compound engine has two cylinders, a high-pressure cylinder to which steam is admitted first, and a low-pressure cylinder which utilises the exhaust steam from the high-pressure cylinder. The steam is thus used twice. Exhaust steam from the low-pressure cylinder passes to a condenser to create a vacuum.

    Armstrong's early steam pumping engines for hydraulic systems were two cylinder simple horizontal engines. Two of these engines are preserved at the Boat Museum at Ellesmere Port. Horizontal compounds were first used at hydraulic pumping stations in the 1870s. Armstrongs seem always to have used horizontal engines. The engines they supplied to Tower Bridge in 1894 were huge horizontal twin tandem compounds. They are still in situ in the engine rooms under the bridge's southern approach viaduct and form part of the Tower Bridge Experience tourist attraction.

    At Falcon Wharf, E B Ellington installed inverted vertical compound engines and these were also used at Millbank. For Wapping, however, he used an inverted vertical triple-expansion engine, six of which were supplied by the Hydraulic Engineering Company of Chester. The Hydraulic Engineering Company had exhibited a similar engine at the Paris Exhibition of 1889. This type became the standard engine for LHP, replacing some of the earlier engines at Falcon Wharf and Millbank. In the triple expansion engine a third, intermediate-pressure cylinder is placed between the high- and low-pressure cylinders, so the steam is expanded three times. The Wapping engines were fitted with surface condensers through which the circulating delivery water passed.

    The six Wapping triples, installed in 1891-2, ran until September 1956, when they were replaced by electrically driven three-throw ram pumps. They outlived the electric and steam turbine pumps installed in the 1920s. Unfortunately none were preserved.

    The inverted vertical triple expansion steam engine was developed for marine use. It was used in the vessel Propontis in 1880, and more successfully in the Aberdeen of 1881. [Walker page 145]. Its use for water pumping began in 1886 at the Waltham Abbey pumping station of the East London Waterworks Company. Several waterworks examples survive including the two largest at Kempton Park pumping station in west London. They represent the zenith of reciprocating steam power.

    APPENDIX 2 The Steam Turbine

    The generation of electricity required steam engines to run at much higher speeds than had hither to been the case. The conventional reciprocating steam engine was developed into a high-speed machine by Willans and by Bellis & Morecom. Generators were belt driven allowing further increases in speed.

    The Hon. Charles Algernon Parsons patented the steam turbine in 1884. This was a high speed machine which could be directly coupled to a generator. Having several rotors or stages on the same shaft meant it used the same principle as a compound reciprocating engine, but with much higher efficiency. Modern power stations use steam turbines to drive their generators. But the steam turbine had other uses. Parsons demonstrated one of them when he astonished both Royal Navy and onlookers alike at the 1897 Jubilee Naval Review at Spithead by sending his steam turbine-powered launch, Turbinia, through the ranks of naval vessels. Turbinia easily out-paced destroyers sent to intercept her. She now holds pride of place at the Newcastle Discovery Centre. [Victorian Engineering, LTC Rolt, 1970]

    The steam turbine could also be used to drive a rotary pump. The steam turbine installed in the turbine house at Wapping in 1925, was supplied by Messrs Parsons of Heaton Works, Newcastle. It was a more sophisticated machine than the early Parsons' turbines. Other engineers like Rateau in 1896 and Curtis in 1897 had made improvements to the turbine design. The Wapping machine was described as a composite turbine, and was the first example of its type. It had no less than ten stages, the first being a two-row Curtis wheel. This was followed by four Rateau type impulse stages and five reaction stages. [The Development of the Steam Turbine by R H Parsons, 1936, pp 374-6 and Plate LXXXVII] Rated at 817BHP, it drove a Mather & Platt 9-stage turbine pump through gearing and delivered 1000 gallons per minute at 850 psi. The turbine speed was 8000 revs per minute, the pump speed being 1550 revs per minute. The Wapping steam turbine started pumping in April 1926 [Metropolitan Archives B/GH/LH/06/77] and accounted for about a third of gallons pumped. During World War II it was out of use for a couple of years and subsequently found only intermittent use. It was stopped in May 1952. Similar machines were used at Rotherhithe (1927-1959) and Grosvenor Road (1928-1951) pumping stations.

    APPENDIX 3 Pumps

    Three types of pump have been used at Wapping, the ram pump, the centrifugal pump and the turbine pump. The steam pumping engines and the later electric pumps used ram pumps, as did the well pumps. The tank pumps used centrifugal pumps. An electrically driven Drysdale Bon Accord centrifugal pump lay abandoned in the yard for many years, but it is believed to have come from the docks. The 1920s steam turbine and the Mather & Platt sets used multi-stage turbine pumps.

    The ram pump is a type of force pump. In hydraulics the ram was often used instead of the piston because of its simplicity and robustness. The centrifugal pump was developed over a long period of time, having its origins in the work of Denis Papin and others in the 17th century. Both Gwynne and Appold exhibited centrifugal pumps at the Great Exhibition. The Appold pump was later used in Fen drainage schemes.

    The turbine pump is a multi-stage version of the centrifugal pump, the first ones appearing in the last quarter of the 19th century. Mather & Platt, of Park Works, Manchester, developed their version essentially for pumping water from deep coal mines. But the design was eminently suitable for hydraulic systems and many were sold to industrial users and others. The London Midland & Scottish Railway Company and the Southern Railway Company used them to electrify their hydraulic pumping stations. LHP purchased one set for City Road Basin in 1921 and two sets for Wapping two years later. A second set was ordered for City Road in December 1922. Yet another set was installed at Falcon Wharf to pump water from the River Thames. Each set had a central electric motor with turbine pumps at each end.

    APPENDIX 4 Basic Data


    Length Width Source
    feet [metres] feet [metres]
    Engine House 66'8" [20.32m] 40'6" [12.34m] 1
    66'3" [20.21m] 40'6" [12.35m] 3
    Turbine House 63'1½" [19.235m] 22'3" [6.78m] 1
    61'3" [18.7m] 21'2" [6.45m] 3
    Boiler House 63'5" [19.33m] 45'8" [13.93m] 1
    Accumulator Tower 21'7½" [6.59m] 11'1½” [3.39m] 2
    Coal Store 45'8" [13.93m] 19' [5.81m] 1
    Filter House (1)
    Filter House (2)

    1) LDDC booklet
    2) The Engineer 1893
    3) Measured


    Accumulators in tower (set at 11'2" centres)

    Pressure 800 psi (Source: The Engineer 1893)
    850 psi (Source: The Engineer 1926)

    Diameter Stroke Weight
    of ram
    Accumulator 1 20 inches 28 feet 110 tons of slag*
    Accumulator 2 20 inches 28 feet 110 tons of slag*
    * total weight on ram @ 800psi is 112 tons
    @ 850psi is 119 tons

    Small pilot accumulator (in corner of engine house)

    Pilot Accumulator 3 inches 10 feet
    total weight on ram @ 850psi is 2.68 tons

    INSTALLED PLANT (derived from Metropolitan Archives B/GH/LH/06/30)

    Plant Pre-Electrification Plant Post-Electrification
    Type & No GPM Total Type & No GPM Total
    each GPM each GPM

    6 Steam IVTE 250 1500 2 ac const.sp. 500 1000
    1 Steam turbine 1000 1000 3 ac const.sp. 200 600
    2 elect turbine 500 1000 2 dc vari. sp. 200 400
    TOTAL 3500 TOTAL 2000
    2 elect turbine 500 1000
    GRAND TOTAL 3000

    APPENDIX 5 Engine Data

    Steam Engines

    Steam Pumping Engines (set at 19' 6" centres) (source The Engineer 1893)
    Makers: The Hydraulic Engineering Company Ltd, Chester
    Type: Inverted Vertical Triple Expansion, with surface condensers
    Number of pumps: 6
    Diameter Stroke

    hp cylinder 15" 24"
    ip cylinder 22" 24"
    lp cylinder 36" 24"
    pump rams, s/a 5" 24"

    steam pressure 150 psi
    delivery 300 gpm @ 800 psi (Source The Engineer 1893)
    250 gpm @ 850 psi (Source The Engineer 1926)

    Steam Turbine

    Makers: C H Parsons & Co Ltd, Newcastle
    Type: composite turbine
    11 sets of blading: 2 Curtis, 4 Rateau and 5 Parsons reaction type
    Pump: Mather & Platt Plurovane type, 9 stage high lift turbine
    delivery: 1000 gpm @ 850 psi
    steam turbine speed: 8000 rpm
    turbine pump speed: 1550 rpm
    steam pressure: 150 psi superheated to 466deg.F
    condenser vacuum: 28.6" (at 30" of mercury)

    Electric Turbine Pumps

    Mather & Platt Turbine Pumps
    Number of pumps: 2
    pump speed: 1480 to 1500 rpm
    delivery 500gpm @ 850 psi
    electric motors: induction. 6000volt, 3-phase, 50c/s
    rotor voltage 490 volts

    APPENDIX 6 Electric Pumps Data — Constant Speed Pumps

    Information taken from plates affixed to the machines.

    Pump No 1 Pump No 2

    HEC Machine Number 1071 1070
    Electric Motor:
    Type * induction
    Frame * 16848 RN
    BHP * 370
    * PER 50
    voltage * 6600 volts
    * 3 phase
    Speed * 738 RPM
    Current * 30.5 amps
    Rotor * 3
    Rotor voltage * 507 volts
    Rotor current * 328 amps
    Rating * Continuous
    * Dripproof
    Serial No * B414697/1/01
    BSS * 168/1935

    * = plate missing

    Pump No 5 Pump No 6 Pump No 7

    HEC Machine Number 1061 1060 1059
    Electric Motor:
    Type induction induction induction
    Frame 10545 RW 10545 RW 10545 RW
    BHP 150 150 150
    50 PER 50 PER 50 PER
    Voltage 440 volts 440 volts 440 volts
    3 phase 3 phase 3 phse
    Speed 730 RPM 730 RPM 730 RPM
    Current 194 mps 194 amps 194 amps
    Rotor 3 3 3
    PH 610 volts 610 volts 610 volts
    Current 113 amps 113 amps 113 amps
    Rating continuous continuous continuous
    dripproof dripproof dripproof
    Serial Number B414697/2/03 B414697/2/02 B414697/2/01

    APPENDIX 6 Electric Pumps Data — Variable Speed Pumps

    Pump No 3 Pump No 4

    HEC Machine Number 1058 [1057?] *
    dc Electric Motor:
    Type shunt wound shunt wound
    dc motor dc motor
    Excitation Voltage 300 volts 300 volts
    Frame 6643/10 6643/10
    BHP 0/150 0/150
    Speed 0/720 RPM 0/720 RPM
    Voltage 0/500 volts 0/500 volts
    Current 241 amps 241 amps
    Rating continuous continuous
    force vent. force vent.
    Serial No B414697/4/01 B414697/4/02

    Type shunt wound shunt wound
    dc generator dc generator
    Excitation Voltage 300 volts 300 volts
    Frame 6637/10 6637/10
    0-125 MW 0-125 MW
    Speed 980 RPM ###
    Voltage 0/500 volts 0/500 volts
    Current 241 amps 241 amps
    Rating continuous continuous
    dripproof dripproof
    Serial Number B414697### ###/5/01/###
    ### ###

    ac Electric Motor:
    Type induction ###
    BHP 190 ###
    Speed 980 RPM ###
    3 phase ###
    PER 50 ###
    Voltage 440 volts ###
    Current 225 amps ###
    Rating continuous ###
    BSS ###/193# ###
    Type dripproof ###
    Rotor Pol### ###
    Rotor voltage ### ###
    Rotor current ### ###
    Serial Number B414697/6/01 ###
    Frame 0545 PW ###

    ### = illegible * = plate missing

    APPENDIX 7 Remote Control Facilities
    by Dave Gunning

    The equipment described below was located in the Control Room, at the west end of the Turbine House. Just outside the control room, within the Engine House, was a 240V-fed 50V battery charger.

    The equipment in the Control Room was enclosed in five steel cabinets (bays) which were bolted together and these are numbered, for the purposes of this description, 1 to 5 as viewed from the front (east). The rear of each of the bays was provided with twin doors, the left (as viewed from the rear) being held closed by vertical brass bolts and the right was held to the left by two turn latches actuated by chrome handles, one of which was provided with a Yale-type lock. Some equipment was missing.

    The front faces of the bays, from 1 to 4, showed painted representations of pump configurations at four locations, and the fifth showed the local arrangement at Wapping. Bays 1 and 4 showed two pumps each, bay 2 showed four and bay three showed 7 pumps. Also on each of the front faces was a meter which showed water pressure at the location depicted.

    Bay 1 was empty and a substantial rectangle of the front face of the bay had been removed. Only the pressure meter remained and no evidence of internal equipment existed.

    Bays 2, 3 & 4 contained the remains of three remote supervisory control systems as manufactured by the Automatic Telephone & Electric (ATE) Company at their Strowger Works in Edge Lane, Liverpool. These systems utilised the abilities of telephone equipment to perform remote control and indication of distant sites over a single pair of wires. The equipment was of the Strowger type, being of American derivation and which was current in the 1920s and early 1930s before the later ‘3000’ type equipment became the GPO standard. ATE considered that the ‘3000’ type equipment was less reliable for remote, high-security functions and continued to supply Strowger equipment for these applications. This is confirmed by the knowledge that the systems installed at Wapping were commissioned in the 1960s.

    The metering of remote pressure was supplied by Evershed and Vignoles and appeared to be separate from the supervisory systems, operating over a second pair of wires and powered by 240V ac.

    The ATE equipment in each bay was a metal frame on which were mounted a horizontal, mild steel channel shelf and auxiliary items such as 50V fuse panels, 50V supply switches for ac and dc feeds, for lamps and relays respectively, and a screw-terminal panel for external connections. The logical and switching equipment was mounted on mild steel sub-frames (relay groups) which jacked in to the shelf by means of engaging lugs and slots and using up to four sets of 32-point electrical connectors per relay group. These relay units mounted up to four rows of Strowger relays and, depending on the type of group, up to three uniselectors also.

    As a note of explanation, a relay is an electromagnet, the armature of which operates a set of electrical contacts. A uniselector is a rotary step-by-step mechanism which is driven by an electromagnet. This mechanism has a set of up to eight wiper contacts, mounted on its spindle and which sweep over a 175 degree, 25-point contact arc per wiper (25x8). Generally, wipers are double-ended, thus traversing 25 contacts twice per revolution of the shaft. By using, say, four pairs of opposing, single-ended wipers, a 50-point configuration (50x4) can be produced. Other components mounted in relay groups are capacitors, resistors and metal-oxide rectifiers. It was ATE's practice to wire relay contacts in pairs for security, thus having the total contact availability and making the circuit designer's task doubly difficult.

    Each system required a Common Relay Group (CRG), in the left-hand shelf position (as viewed from the rear) labelled ‘Z’, and an Indication Group which was in the centre position, labelled ‘A’. There was a right-hand position, ‘B’?, which was unlabelled.

    The CRGs contained 36 relays and three vertically-mounted uniselectors (two 25x4 and one 25x6). The Indication Groups comprised a variable number of magnetically-latching relays (as required by each system), four common-function relays and one horizontally-mounted uniselector (25x6).

    The CRG handled all inter-location signalling which was by loop impulsing at 10-ips over a single pair of wires at 50V and was polarity-reversible to convey information. It regulated controls such that they were only singly attempted and it code-checked all signals returned from the remote location. It also supplied various flicker rates for indicator lamp lighting to convey selection and change information on the front panels. All signals were total-checked whist controls were additionally checked for single point marking. The Indication Group stored status information regarding alarm conditions and pump operating status by means of magnetically-latching relays which were set by signals which were decoded by the CRG and distributed by the Indication Group's own uniselector.

    The Evershed & Vignoles equipment in each bay comprised a plate mounted on the frame of the bay on which was mounted a 240V supply switch and fuses, an auxiliary relay unit (line interface), a pressure-indicating meter and a Venner time switch. The meter on the front panel was also part of the system and would be a duplicate of the plate-mounted meter and would probably have been connected in series with it.

    We have not yet been able to confirm which pumping stations were controlled by which bay. All groups of Bay 2 equipment were to ATE master drawing RM7437 and the Indication Group contained 14 pairs of magnetically-latching relays. No E&V interface unit was fitted.

    All groups of Bay 3 equipment were to ATE master drawing RM7007 and the Indication Group contained 24 pairs of magnetically-latching relays.

    All groups of Bay 4 equipment were to ATE master drawing RM6991 and the Indication Group contained 10 pairs of magnetically-latching relays.

    The equipment in Bay 5 provided local control of the Wapping installation. It consisted of switches, buttons and indicator lamps which were connected directly to the control contactors and status indicating devices in the power controller cabinets adjacent to the pumps in the Engine House and Turbine House. Investigation did not reveal the operating voltage of the equipment in Bay 5. However, small items of equipment located on the wall of the Engine House adjacent to the high-voltage switchgear bays are marked with 50V references. Whether this 50V was from the charger near the Control Room or from some other source is not known. These items might have been concerned with the low-voltage supplies for the local control from bay 5.

    On the front face, within the diagrammatic representation of the remotely-controlled pumps, each pump had an indicating lamp and a control selection switch (a telephone switchboard lever key, normally referred to as key).

    Beneath the schematic diagram was a centrally located alarm indicator panel. This comprised glass slides, bearing alarm condition transfers, attached behind an opaque glass fascia which was mounted in front of a 6x2 egg box unit with a lamp attached behind each segment. The alarm unit in Bay 4 had a single row (6x1) arrangement. Typical alarms depicted would be feed tank and accumulator extreme levels, pressure limits exceeded, electrical supply transformer alarms and maximum demand excesses.

    Below this unit was a collection of keys and push-buttons to effect controls, acknowledge alarms, perform lamp tests and to instruct rechecking of indications. Lamps were provided to indicate progress of, or errors in execution of, control functions. A key was provided to control a dummy pump. This enabled a complete function test of the control system and would have been a routine performed during each shift.

    To control a pump, its selection key would be operated until acknowledged by the system, possibly by flickering or flashing the pump's indicator lamp. The common START or STOP button would then be pushed and the control would proceed followed by a return of the latest state of all of the indications from the remote location. Any attempted multiple key selection would be indicated by a Key Fault lamp.

    A change of state of any remote device would be signalled by a flashing or flickering lamp appropriate to the device and the sounding of an alarm bell. The condition would be acknowledged by an alarm reset key. The alarm circuits for each system were fused in adjacent bays in order to ensure that any fault in a system would not disable its ability to warn of problems.

  • © GLIAS, 2005