1759 - Harrison's Masterpiece : H4 Chronometer

After twenty-three years of tries since his first clock meant to precisely metering time at sea, John Harrison's work has come to its highlight this year with the completion of the H4 chronometer, as we all name it, or of the watch number one, as the old clockmaker names it. This is effectively the fourth chronometer attempted by John Harrison, 66 years old, to achieve the goal specified by the Longitude Act forty-five years ago, that is a mean of keeping time at sea so precisely that the calculation of the longitude can be done with a minimal error.

¤ It is well known that latitude can be metered with accuracy since it is merely based on the metering at noon of the altitude of the sun above the horizon, i.e. it only depends on the quality of the octant and of the dexterity of the person handling it. As for longitude, time is the essence : one observes culmination of the sun at local noon at an absolute time which can directly be translated in longitude, as one hour of time difference with Greenwich means an arc of longitude of 15°, West if later, East if earlier. The issue consists in keeping time when time elapses - something that has seemed out of reach until Henry Sully's montre de la mer demonstrated in 1816 and Harrison's H1 in 1836. Sully and Harrison achieved to make precise clocks that do not change too much their rates of going with temperature but faced the question of mechanical shocks at high seas. As for precision science clocks, the master was Horologist George Graham, inventor of the cylinder escapement in 1725 and discoverer of the diurnal variation of the magnetic field in 1722, who died in 1751 after having given a hand to Harrison in the form of money, of network and of knowledge.

¤ The Longitude Act ediced in 1714 promises a £20,000 reward to the inventor of a method permitting a transatlantic voyage to the West Indies with an error of longitude not exceeding half a degree, a £15,000 for less than two-thirds of a degree, a £10,000 reward for less than one degree.

¤ It obviously depends on which point of landfall. Bridgetown (Barbados) lies at 59°W from Greenwich, Saint-John (Antigua) at 61°W, and Kingston (Jamaica) at 76°W, for instance. Error is sought to be less than 0.67 or 0.85% in terms of distance, but it is not the key point which is the drift of the chronometer with time elapsing during the journey. To traverse 59 or 76 degrees of longitude, the ship has to travel over 4,000 to 5,000 nautical miles at a speed of four or five knots for the good sailers, six knots for the finest frigates under the best conditions. It means some 30 to 50 days under sail to reach out Bridgetown. Time must be known with a 80-second accuracy to reach out an error of 20 miles after a journey of 4,000 miles. In forty days, it requires a drift under 2 seconds a day.

Harrison spent twenty-eight years in building precision clocks without reaching out his ideal before turning suddenly to a watch. H4 looks like a pocket watch from a distance and has the structure of a watch. It is much bigger than a pocket watch since it is five inches in diameter and three inches in thickness. It weighs three pounds and a fifth.

Why did Harrison stop working on his precision clocks and move to a marine watch? It seems that a sharp stimulus was given by Thomas Mudge's feat that convinced him that a better path was possible.

¤ Thomas Mudge, 44, was George Graham's apprentice twenty years ago. He already had a reputation when, four years ago, he invented the detached lever escapement, first for a clock, then for a few pocket watches. Several years earlier, in the 1740s, Mudge had improved cylinder escapement and produced watches that were said to be as precise as the existing precise clocks, say a few seconds drift in a day, perhaps thanks to the higher elasticity of crucible steel he could use. Lately, when Mudge came under the light, Harrison became aware of his works and of the possibilities of a marine watch.

¤ However Harrison seems not to know that Pierre Le Roy, 42, invented eleven years ago the detached escapement, or a pivoted detent escapement. He has been working since then on isochronous balances and produced a first chronometer three years ago.

Harrison was not a beginner with watches. He designed and supervised the construction in 1752-53 by John Jefferys, who died the following year, of a precision watch for which he improved the old rest escapement, and introduced thermal compensation as well as a new invention, the going train fusee to allow winding while running. Design and construction of H4 took less than two years. Harrison is completely sure of his success, which explains that

____________

LINK WITH PREVIOUS CHRONICLES

1714 - The Longitude Act

1736 - The First Sea Chronometer by John Harrison

1746 - Accurate Lunar Tables

____________

IN RETROSPECT FROM TODAY

Note - on the trials of H4

The H4 travels to Kingston from London aboard the ship-of-the-line Deptford in 1761-62. The voyage lasts 81 days. The error at the landfall is 1½ mile, which is much better than required by the Navigation Act, and well beyond any expectation. It means that H4 has drifted by five seconds only.

It is considered a fluke by astronomer Nevil Maskelyne, followed by the Board of Longitude. They suspect that a sort of compensation has occured between, for instance, a rate at low temperature and another at high temperature, or simply by irregularities in going rates. Harrison is teribly angry and frustrated from the reward. He declines a governmental £5,000 proposition for his design.

A new trial abord the corvette Tartar is performed. This time, the watch displays a 39-second time difference, which results in a 10-mile error at landfall. During the same voyage, Maskelyne obtains a 30-mile error by using the lunar distance method. Once again, the result is suspected of being a fluke and Harrison, once again, is frustrated from the reward. He declines a new governemental £10,000 offer plus an extra £10,000 for supplying data to other British watchmakers so as to duplicate H4. For him, it is now a question of pride.

Maskelyne proceeds with new trials on land in 1765 and observes that the going rate of H4 displays irregularities, which is certain. It happens that the two trials were lucky as Maskelyne suspected. Yet, if Maskelyne is right on a theoretical basis, he is wrong on a practical one. Maskelyne is generally depicted as the villain who has engineered a plot against Harrison to make the lunar distance method and himself triumph. H4 does the job, what the Parliament has understood when it offers money for the design.

Meanwhile Harrison carries on the construction of H5, or marine watch number two. Trialed on land in 1772 by King George III himself, it is found to have a going rate of one third of a second - a measurement devoid of real significance but politically correct which liberates the payment of a £8,750 by the Parliament, in addition of the £10,000 down payment settled in 1765, declined though, and in addition to the £4,315 paid by the Board of Longitude throughout the years.

The marine watches are incredibly expensive and their number remains limited until the 1820s. The lunar distance method wins the race.

____________

SUGGESTED BIBLIOGRAPHY

Eva Germaine Rimington Taylor - The Haven-Finding Art : A History of Navigation from Odysseus to Captain Cook - London, 1956

Dava Sobel - The Illustrated Longitude : The True story of a Lone Genius who Solved the Greatest Scientific Problem of His Time - New York, 1998