Reducing friction was crucial for the success of Harrison’s early timekeepers. The oils available at the time were the curse of clockwork, causing mechanisms to run inconsistently and break down. From his understanding of the properties of woods, however, Harrison realized that the tropical hardwood lignum vitae, which contains a natural lubricant, could be used for the bearings and would allow him to dispense with oil entirely.
Success at Brocklesby Park encouraged Harrison to design precision pendulum clocks, of which he built three in the 1720s (Fig. 7). All incorporated his friction-reducing ideas and eventually also included a device that dealt with temperature variation. For a pendulum-driven clock, the time of the pendulum swing depends on its length. If the pendulum rod is made of metal, it lengthens as the metal expands in warmer conditions, causing the clock to beat slower and lose time. Harrison got round this by using a pendulum consisting of a series of rods of two different metals, brass and steel, connected in such a way that the expansion of the metals cancelled each other out and the pendulum’s effective length remained constant. This is known as a gridiron pendulum. Once adjusted, Harrison’s clocks ran with unprecedented accuracy and reliability.
Harrison later wrote that it was in 1726 that he learned about the longitude rewards and turned his attention to designing a timekeeper that could work properly on a ship. This would dominate the rest of his life, with his attempts focused on the challenges that had so taxed Huygens, Sully and other clockmakers before him: reducing friction; compensating for the effects of temperature; and isochronism – ensuring that each beat of a timekeeper took the same time.
Armed with his ideas for sea-clocks, Harrison came to London in about 1727–28, looking for support and the promised rewards. As Harrison tells it, he began with Edmond Halley, Flamsteed’s successor as Astronomer Royal. Halley received him warmly at Greenwich but felt unable to judge Harrison’s work and so sent him to George Graham. After an unpromising start, Harrison eventually impressed the London clockmaker, probably with his elegant method for temperature compensation, to the extent that Graham even offered a loan to help him develop his ideas. From then on, Graham would be Harrison’s main supporter in London, giving advice and access to books and contacts. These would be crucial as his work progressed.
For the next few years Harrison worked in Barrow on a marine timekeeper, now known as H1 (Fig. 8), probably helped by his brother James. It was an extraordinarily complex machine with over 1440 parts – over 5400 if you count each chain segment separately. In essence, it was a portable version of his pendulum clocks and, like those clocks, used a grasshopper escapement, lignum vitae bearings and a gridiron pendulum. It also had several further refinements to counteract the effects of motion at sea: it used a spring rather than a hanging weight to power the clock and the bar balances (which look like dumbbells) were connected in such a way that the effects of motion on one would be counteracted by the other.
After testing the clock on the river Humber, Harrison proudly brought it to London in 1735 and installed it in Graham’s workshop to be shown to London’s scientific community. Backed by a trusted member of that community, the clock and its maker quickly impressed all who came to see it: ‘The sweetness of its motion’, wrote the antiquarian William Stukeley, ‘cannot be sufficiently admired’.20 Graham’s close contacts with the Royal Society helped as well, since within a short time the Society issued a certificate praising the clock’s potential. At last, it seemed, here was a timekeeper that might be used to determine longitude at sea. A trial was called for.
In May 1736, Harrison and H1 were therefore taken aboard the Centurion (Fig. 9), about to set sail for Lisbon. As the First Lord of the Admiralty told her captain,
The Instrument ... has been approved by all the Mathematicians in Town that have seen it, (and few have not) to be the Best that has been made for measuring Time; how it will succeed at Sea, you will partly be a Judge.21
Fig. 8 – Marine timekeeper H1, by John Harrison, completed in 1735
{National Maritime Museum, Greenwich, London, Ministry of Defence Art Collection}
Fig. 9 – Model of the Centurion, by Benjamin Slade, made in 1747 for George Anson
{National Maritime Museum, Greenwich, London}
Stern view of the Centurion
Captain Proctor was cautious: he feared Harrison had ‘attempted Impossibilities’.22 Indeed, the voyage began poorly. Proctor’s log entries showed that the clock soon diverged from the officers’ reckonings (Fig. 10), although the passage was too rough for proper testing and Harrison was seasick. But he seems to have overcome these early problems and had his machine going more reliably by the time they reached Lisbon, where it was transferred to the Orford for the return, with much better results. As they neared England, Harrison announced – correctly – that a headland the officers thought was the Start was in fact the Lizard: they were sixty miles off course and in danger.
Later history would add some irony to the choice of the Centurion for H1’s first proper sea trial. In 1740, shortly after the outbreak of the War of the Austrian Succession, the Centurion was the flagship of Commodore George Anson’s squadron, sent to the Pacific with orders to damage Spanish interests. Anson returned a hero but the voyage was a series of arduous trials: most of the ships’ crews died and only the Centurion completed the voyage. Their navigational problems began in March 1741 as they began the much-feared leg around Cape Horn to enter the Pacific. Severe storms separated the ships, with some wrecked in the icy seas. A month later, the sailing masters estimated that they were safely west of Tierra del Fuego and could head north. Days later, however, they sighted Cape Noir, over 300 miles east of their estimated position (Fig. 11). It was all very reminiscent of William Dampier’s problems in the same waters (see Chapter 1).
Back in London, after the safe return of the Orford, the results of the Lisbon trial suggested that Harrison might qualify for a reward under the Longitude Act and the Admiralty requested a formal meeting of the Commissioners. Accordingly, eight of them assembled on 30 June 1737 to discuss Harrison’s ‘curious Instrument’.23 Among the evidence was a certificate from the Master of the Orford, who described the incident near the Lizard, which the ships’ officers had kept quiet. The Commissioners quickly agreed a payment of £500, £250 to be paid up front, to allow Harrison to build an improved clock, which he promised to do within two years. This marked a new phase in the story, with the Commissioners and Harrison meeting on several occasions over the next thirty years to discuss progress and agree further payments. As the recipient of ongoing government funding to develop his clocks, Harrison found himself in a unique position. By 1746, he even pleaded that he was so committed to the work that he was ‘quite incapable of following any gainfull employment for the support of himself & family’.24
Fig. 10 – Log of the Centurion, by Captain John Proctor, 1736; the second column from the right gives longitude estimates from dead reckoning and from Harrison’s H1 (detail)
{National Maritime Museum, Greenwich, London}
Fig. 11 – The estimated and actual tracks of the Centurion around Cape Horn in 1741, from ‘A Chart of the Southern Part of South America with the Track of the Centurion’, by Richard Seale, 1748 (detail)
{National Maritime Museum, Greenwich, London}
Fig. 12 –Marine timekeeper H2, by John Harrison, 1737–39
{National Maritime Museum, Greenwich, London, Ministry of Defence Art Collection}
Fig. 13 – Marine timekeeper H3, by John Harrison, 1740–59
{National Maritime Museum, Greenwich, London, Ministry of Defence Art Collection}
Harrison moved to London soon after the Lisbon trial, and within the promised two years he finished his second sea-clock, H2 (Fig. 12). This reflects its London manufacture, since Harrison drew on the capital’s skilled clock- and instrument-making workforce for its brass plate, steel springs, engraving and basic finishing. The new clock looked quite different from its predecessor, but was similar in con
ception, with extensive anti-friction work, a grasshopper escapement and gridiron temperature compensation. It did, however, incorporate some new features, most notably a remontoire, which is a secondary winding mechanism that eliminated variations in the driving force supplied to the escapement and greatly improved the clock’s accuracy. Yet H2 never went to trial because Harrison discovered a fundamental flaw that made it susceptible to external motions, and the clock had to be abandoned. A third timekeeper, he assured the Commissioners, would perform better.
Harrison began work on H3 (Fig. 13) in 1740. It was running and being tested within five years, but it was clear from the start that the clock would struggle to keep time to the accuracy desired, forcing him to make changes and adjustments. A drawing completed in the early 1740s, for example, shows a part of the mechanism that Harrison subsequently redesigned (Fig. 14). H3 retained many of the elements of the first timekeepers but also included further innovations. One was a bimetallic strip for temperature compensation, which drew on the principles behind the gridiron pendulum. It consisted of a brass strip and a steel strip riveted together. With one end fixed, the movement of the other adjusted the balance spring as the temperature changed. This idea would come into its own in the twentieth century for switches in kettles, toasters and thermostats. A second innovation was the caged roller-bearing, another anti-friction element, which can be seen as the predecessor of the caged ball-bearing, ubiquitous in complex machinery today.
Despite its novel features, H3 never ran to Harrison’s satisfaction, although he laboured on it for nineteen years. As an extraordinary and complex mechanism, however, it attracted attention and praise, gaining Harrison the Royal Society’s highest honour, the Copley Medal, in 1749. It features prominently in Thomas King’s portrait of Harrison (Fig. 15), where it sits in a gimballed case designed to steady it when on a ship (H1 and H2 originally had similar cases). The clock also impressed visitors to Harrison’s workshop in Red Lion Square: William Hogarth thought it ‘one of the most exquisite movements ever made’,25 while Benjamin Franklin happily paid ten shillings and sixpence for a viewing.
Fig. 14 – Drawing of part of H3, by John Harrison, c.1740
{National Maritime Museum, Greenwich, London}
Harrison was already exploring new approaches by the time Franklin visited in 1757, however, and was thinking about watches rather than clocks. After experimenting with various ideas, in about 1751–52 he commissioned John Jefferys to make a watch (which Harrison holds in Fig. 15), with a radically new type of balance. It worked well, so Harrison incorporated it into his fourth longitude timekeeper, H4 (Fig. 16). While it looked like a large pocket watch, H4 was quite different (Fig. 17). The secret can be heard in its rapid ticking, five times a second, since its large balance beats more quickly and with larger oscillations than a typical watch. This contradicted horological orthodoxy, which favoured a small, light balance with minimal oscillations. Harrison’s thinking was that its rapid pulse would not be affected by a ship’s much slower motions and would beat reliably. The only compromise was that H4 needed oil, since Harrison could not miniaturize the anti-friction devices, although he used perfectly shaped jewelled bearings to minimize friction.
Fig. 15 – John Harrison, by Thomas King, c.1765–66
{Science Museum / Science & Society Picture Library}
Fig. 16 – Construction drawings for the mechanism of H4, by John Harrison, c.1756
{The Trustees of the Clockmakers’ Museum}
Fig. 17 –Marine timekeeper H4, by John and William Harrison, 1755–59
{National Maritime Museum, Greenwich, London, Ministry of Defence Art Collection}
Fig. 18 – John Hadley, attributed to Bartholomew Dandridge, early 1730s
{National Maritime Museum, Greenwich, London}
The sea-watch was complete by 1759 and, the following July, Harrison suggested that it might accompany H3 on a sea trial. In the end, the continuation of the Seven Years War meant that it was 1761 before the Commissioners gave permission for John’s son, William, to prepare for a voyage to Jamaica. Moreover, by the time the Deptford sailed with William and H4 on board, John Harrison had decided that H3 would remain in London. The watch’s trial seemed to go well. On the way out, William used it to predict an earlier landfall at Madeira than the crew were expecting, so impressing the captain that he asked to buy their next timekeeper. Returning on the Merlin, however, William found himself cradling the precious watch in blankets to protect it from tempestuous seas.
It was back in England that trouble began. The Commissioners decided that the test had not, after all, been sufficient. They could not confirm that H4 had succeeded, they said, because the longitude of Port Royal in Jamaica was not properly known. On top of that, they were concerned that its rate, the amount of time it gained or lost each day, had not been agreed beforehand. Despite the Harrisons’ protests, the trial was not to be decisive. The Commissioners did concede, however, that H4 was ‘an Invention of considerable Utility to the Public’ and awarded Harrison £2500, of which £1000 was to be paid after another successful trial.26 John and William were not mollified.
Fig. 19 – Hadley quadrant or octant, c.1744; although the instrument itself is not signed, there is a handwritten label, now barely legible, signed by George Hadley
{Het Scheepvaartmuseum, Amsterdam}
This was the point when relations between the Harrisons and the Commissioners began to deteriorate. Harrison’s friends and supporters began a propaganda campaign of newspaper articles, broadsheets and pamphlets. A petition to Parliament also suggested that he might disclose the watch’s principles in return for certain guarantees. A group of ‘Commissioners for the Discovery of Mr Harrison’s Watch’ was duly created but the process broke down and it was decided by all concerned that a second sea trial was the only solution. But other methods had now come to fruition. John Harrison’s twenty years as the only serious contender had come to an end. By the 1760s, two rival schemes had emerged to challenge his claim: the measurement of lunar distances (or ‘lunars’) and observations of the eclipses of Jupiter’s satellites. Both would soon be put to test with H4, with all three longitude methods benefiting from the introduction of a new navigational instrument for measuring latitude and local time.
On reflection: the development of the octant
Most longitude methods aimed to establish two local times simultaneously, at the ship and at a reference point of known longitude. Even with a mechanical timekeeper carrying the reference time, local time at the ship had to be found from the Sun or stars, so instruments and techniques for doing this accurately were essential.
The most common method was that of equal altitudes, which involved measuring the altitude of the Sun at some time between three and five hours before midday, and then timing how long it was before the Sun reached the same altitude later in the day. The halfway point between these observations would be local noon. An alternative was to use a single altitude observation of the Sun or a star, although the calculations were more complex. In principle, an instrument for determining latitude could be used for these observations, although greater accuracy than the cross-staff or backstaff could achieve was desirable. For finding longitude by lunar distance, which involved additional observations that were impossible with a backstaff, accuracy was even more critical.
Fig. 20 – Thomas Godfrey’s proposal for a double-reflection instrument
{The Royal Society}
In the 1730s, two men came up with ideas for a more accurate and versatile instrument. In England, it was John Hadley (1682–1744, Fig. 18), who was Vice-President of the Royal Society by the time he addressed a meeting in May 1731. Hadley suggested that optical theory could improve observations by applying the principle of double reflection (that is, using two mirrors in sequence). There was precedent for his ideas; Robert Hooke, Edmond Halley and Isaac Newton had previously devised instruments in which an observer rotated a mirror in order to bring the images of two different objects toge
ther – one being reflected in the mirror, the other viewed directly – to measure the angle between them. Edmond Halley had used Newton’s instrument on his first Atlantic voyage and was allegedly able to determine longitude ‘better then the Seamen by other methods’.27 Like Newton, John Hadley proposed to use double reflection for indirectly viewing one of the objects; the other being viewed directly. As he demonstrated, this meant that the observations were not greatly affected by the ship’s motion – a distinct advantage over other instruments.
As a leading member of the Royal Society who could present his ideas in a solidly theoretical context, Hadley had little difficulty in having his proposal taken seriously and a sea trial in the Thames estuary was organized the following year. It was an august group: Edmond Halley, James Bradley (1693–1762), Professor of Astronomy at Oxford, John Hadley and his brothers Henry and George, who had helped develop the instrument (Fig. 19). Although the observers were ‘Persons quite unaccustomed to the Motion of a Ship at Sea’ and the Admiralty yacht Chatham was small and lively, Hadley reported that good observations were quite possible.28
Hadley’s claim to priority of invention was nevertheless soon challenged, when the Astronomer Royal received a letter from James Logan, Chief Justice of Pennsylvania in America. Logan described an instrument (Fig. 20) devised by Thomas Godfrey, the son of a maltster, who had trained as a plumber and glazier after being orphaned. Later developing an interest in astronomy and mingling with Philadelphia’s shipmasters, Godfrey began working to improve navigational instruments. His double-reflecting instrument was ready in 1730 and tested on the sloop Trueman. Having seen it himself, Logan believed it might merit a longitude reward.
Finding Longitude Page 7
