Grand Western Canal boat lifts

Remains of Nynehead Boat Lift

The Grand Western Canal boat lifts were a set of seven boat lifts and one inclined plane installed in the early nineteenth century in the Grand Western Canal in Devon and Somerset. A remarkable engineering endeavour, they were ultimately however a failure through poor design or execution.

Origin

The first stretch of the Grand Junction Canal was opened in 1801, intended to link Tiverton in Devon with Taunton in Somerset, across a challenging terrain. In 1829, the chief engineer James Green turned his attention to the link to Taunton, proposing first a system of inclined planes to haul boats up and down levels, as he had used on the Bude Canal. However after the Bridgwater and Taunton Canal opened in 1827, making navigation from Taunton to Bridgwater easier than on the River Tone, Green changed the system and in a report of 1830 he proposed using vertical lifts instead, and estimated the cost of the canal at £65,000.^[1]

Work started in 1831, and progressed quickly, including the construction of a bridge at Bradford on Tone^[2] and Harpford Bridge at Langford Budville,^[3] where a new warehouse was also built.^[4] In addition to the boat lift in Nynehead^[5]^[6] two aqueducts were required within the village.^[7]^[8]

The route in Taunton was altered to connect to the Bridgwater and Taunton Canal directly, rather than using a short stretch of the River Tone. Seven lifts and one inclined plane were to be built, and it was these features that caused the subsequent delay in the completion of the canal. Teething problems with the design of the lifts gradually came to light, and although the cost of rectifying these was borne by Green, the canal could not be opened. There were also problems with the inclined plane. The canal was partially opened to Wellington, in 1835. In January 1836 however, Green was dismissed.^[1]

The engineer W A Provis was asked to survey the works, and to report on the lifts and the causes of the failure of the inclined plane. His clear report is an important source of information on the canal at the time of its construction. Some remedial work was instigated, following the report, including the provision of a steam engine to power the Wellisford incline. The 14-mile extension was fully opened on 28 June 1838, at a cost of about £80,000.^[1]

Design

Diagram showing the arrangement of the boat lifts

Green's use of boat lifts was innovative. The idea was not new, as he acknowledged in an article describing the lifts which he published in Transactions^[9] in 1838, having been suggested in principle by a Dr James Anderson of Edinburgh in 1796. Robert Weldon had tried to build one on the Somersetshire Coal Canal in 1798, which was replaced by an inclined plane after persistent failures. One was built at Ruabon on the Ellesmere Canal in 1796, but was replaced as it was not robust enough for regular use. James Fussell built one on the abortive Dorset and Somerset Canal, but the works were abandoned before it was ever used regularly. The lift at Tardebigge on the Worcester and Birmingham Canal was replaced by locks in 1815 as it was "too complex and delicate", according to Rennie. Finally, another lift at Camden Town on the Regent's Canal was replaced by locks in 1815 because it could not be made to work.^[1]

Principles

With no working prototypes, Green set about building seven lifts. The principle was simple. Two caissons were suspended from three carrying wheels of 16 feet in diameter, by wrought iron chains. The caisson at the bottom was jacked against the front wall of the lift to seal it, and a door or gate was opened to allow the boat to float in. The caisson at the top was jacked against the back wall in a similar manner. When both boats were in, the doors on the caissons and lift were closed, and the jacks released. Because a boat displaces its own weight in water, the system should be balanced, and a small amount of energy is required to start the boats moving. When the top caisson reaches the bottom, the jacks are applied, the doors are opened, and the boats can continue.^[1]

In order to maintain the balance, a second chain was fixed to the bottom of the caisson, so that the total length of chain on each side of the lift remained the same. As a caisson descended, the chain coiled up at the bottom of the lift. The small amount of energy was created by ensuring that the ascending caisson was a little too low by the time the descending one reached the bottom. Thus it would hold a greater depth of water by the time it was ready to descend again. In practice, about two inches of water, weighing about a ton, was found to be sufficient.^[1]

Problems

The difficulty was that Anderson had suggested that the water in the caisson chambers should be at a lower level than that in the canal. This had not been implemented, and so the lower caisson would not sink deep enough for either boat to be floated out. Attempts to fit gates and to let the water in the chamber drain to waste had proved ineffective. Ultimately, lock chambers were built at the foot of the each lift. These were filled with water from the upper level, so that the boat could float out, and it then descended the final three feet as it would in a conventional lock. Only the lift at Greenham used Anderson's principle, and included a proper drain to lower the level in the chambers.^[1]

Diagram showing the arrangement of the Wellisford inclined plane

With the inclined plane, the problems stemmed from a miscalculation. Power was provided by a large tank, filled with water, which descended in a shaft, raising one boat on a trolley and allowing another to descend as it did so. When the tank reached the bottom of the shaft, the water discharged, and the second tank was used to reverse the process. Green had been the engineer on the Bude Canal, where this design had been used successfully, but the size of the tanks at Wellisford were much too small. On the Bude Canal, a tank holding 15 tons of water was required to raise a boat weighing six tons. The Grand Western Canal however used boats of eight tons displacement: the tanks only held ten tons of water, but tests indicated that about 25 tons would be needed. It was for this reason that the steam engine was obtained to supply the power.^[1]