Harrison H1 Chronometer

This is my attempt at replicating John Harrison's first sea clock in wood. It is 3/4 scale. Many of his innovative features are included but some are not functional because they just don't work in wood. Think  temperature compensation with a wooden gridiron ?

A video of the clock in operation can be seen below.

But first a little history.

In the 18th century navigation at sea was difficult and inaccurate. Latitude was easily measured using the height of the sun at midday and some simple tables. Longitude was much more difficult and relied on dead reckoning if out of sight of land for a long time. Many, many ships and lives were lost by thinking they were somewhere safe when they were not.

 In 1714 the British government threw down the gauntlet to solve the greatest scientific challenge of the century by offering a range of prizes for methods of determining longitude at sea. The value of the prizes depended on the accuracy achieved by the method with £20,000 the highest on offer.

Two competing methods were seriously pursued. An accurate mechanical clock or using nature's celestial clock, the stars and moon.   An accurate clock on a ship was thought unattainable and the heavens were insufficiently mapped to the degree of accuracy required,

Enter John Harrison, carpenter and self taught clockmaker from the north of England who believed he could make such a clock.

After 5 years work he produced a clock that was sufficiently promising to trial on a warship voyaging to Lisbon and back in 1735 where it acquitted itself well. This clock has come to be known as H1. This was its only voyage.

Harrison was not satisfied though and developed his ideas over the next 30 years through a series of clocks, H2, H3 and finally H4 which proved to be more than accurate enough to satisfy the Board of Longitude and hence gain for its inventor the £20,000 prize.

H1 which is the subject of my interest is a large clock. Some 4 feet in all dimensions and weighing 75lb before it was fitted with its gimballed, locked case. It has a number of innovative features invented by Harrison and was the wonder of the world when displayed in London.

My model is approximately 3/4 scale and largely made of wood which is an immediate difference from the original.

The oscillating balances are interlinked by partial gears, but in the original are linked by metal tapes. 

The oscillating masses at the top and bottom of each balance are wooden balls which have been hollowed out and lead filled to give a weight of just under 1Kg each. The springs are made from piano wire and are at both the top and bottom of the balances.

Just as in the original the balance arbors are supported by a small part of the circumference of large anti friction rollers. A single counter balanced spoke supports the portion of rim.

Temperature compensation is two identical gridirons with a complex linkages to adjust the top and bottom springs between the balances. Clearly this cannot work in wood so this part of the clock is just for show.

The escapement is known as a grasshopper escapement, invented by John Harrison it is beautifully suited to wood because there is no sliding friction and no lubrication required.

The arbors of most of the going train are supported on antifriction rollers designed  to reduce friction and the need for lubrication.

The major difference I have employed is to make the clock weight driven with automatic rewind. A mercury tilt switch triggers the small electric motor which is at the other end of the bar and drives the third wheel.

The original has two  spring barrels working onto a single fusee with Harrison's maintaining power. I am not comfortable with springs in a wooden movement and certainly can't make a double fusee in wood let alone a single one.

The going train has wooden gears and lantern pinions just as in the original.

The dials are laser cut and laser etched.

The plates are made from 5mm plywood and the spacer rods from 12mm dowels. The multitude of finials are 3d printed in plastic. All wood is painted with a gold acrylic.

This is a logging trace of the time of each oscillation over a period of nearly 12 hours. With a scale top to bottom of 0.1 seconds it can be seen that there is an almost regular periodicity with most of the oscillations within +- .01 seconds in the 2 second period, but with no steady deviation. The results would suggest that over a 24 hour period the time error would be of the order of nearly a minute. Not as nearly as good as John Harrison's clock but perhaps could be adjusted to make it better. It should be noted that he determined the rate of his clock before its voyage and used this to adjust each daily reading to get precise time.

This was a complicated and frustrating model to make. There are no detailed drawings available so I worked from photographs of H1. It took about two years of work and was put on ice several times when lack of knowledge and skill frustrated me.  I salute John Harrison, a self trained man, who was able to design and construct such a wonderful machine.