Tide analysis and prediction machines

Contents of this page:
Kelvin's Tide Predicter
USCGS Tide Predicting Machine No. 1
USCGS Tide Predicting Machine No. 2

Two harmonic analyzers are shown on the Fourier analysis page. In addition, besides the picture on the Tides base page this page is keyed to, machines are mentioned in connection with Beauchamp Tower's suggestion.

Kelvin's Tide Predicter
Click for larger image.

This machine was exhibited at the South Kensington Museum, where it currently resides, in 1876. It is designed to sum the M2, S2, N2, K1, O1, K2, L2, P1, M4 and MS4 constituents, in that order reading from the left. (See Kelvin, p. 289). ``The machine may be turned so rapidly as to run off a year's tides for any port in about four hours.''

The U. S. Coast and Geodetic Survey Tide Predicting Machine No. 1
This machine, built by William Ferrel in 1885, was used for prediction through 1911. It summed 19 constituents. This is presumably the machine on display in the Smithsonian.

Return to top of page.

The U. S. Coast and Geodetic Survey Tide Predicting Machine No. 2
Click for larger image.

This machine, built by R. A. Harris and E. G. Fischer, was completed in 1910. According to Schureman it is about 11 feet long, 2 feet wide, 6 feet high, and weighs approximately 2,500 pounds. It is designed to sum 37 constituents. This machine was used from 1912 until 1965 to generate tide tables for U. S. ports. It sits at present in the lobby of the NOAA in Silver Spring, MD.

There are 2 complete sets of pulleys, one on each side of the machine, running in tandem. Those on the left side sum the heights of the constituents. Those on the right, or ``time'' side, sum their derivatives, so a high or low tide (a local maximum or minimum) is registered by the index on the time chain passing zero (see below).

Click for larger image.

This side view shows some of the pulleys and the cams designed to convert rotational motion of the gears into vertical motion of the pulleys. The total length of the chain linking the pulleys on this side is 27.6 feet. Schureman's book gives complete instructions for how to set and run this machine. For example, on p. 142:

``If the predicted high and low waters for the year are desired, the operating crank is turned forward until the machine is automatically stopped by the brake at a high or low water. To avoid the strain on the machine due to sudden stops, the operator should watch the small index on the time chain, and as this approaches the fixed index in the center of the opening on the face of the machine, turn the crank more slowly until the machine is stopped as the indexes come in contact with each other. The time and height may then be read directly from the dials on the face of the machine. The movement of the height pointer before the stopping of the machine and also the tide curve will clearly indicate whether the tide is a high or low water. After the tide has been recorded an inward pressure on the crank handle will release the brake and the machine can be turned forward to the next tide, the process being repeated intil all the tides of the year have been predicted and recorded.''

The crank was replaced by an electric motor for the last 5 years of the machine's operation (1960-65).

Back to top of page.
Back to Main Tide page.
Back to Tony's Home Page.

Tony Phillips
Math Dept SUNY Stony Brook
February 17, 1997