linear scale affects appreciably both the cost and the accuracy of test 

 results, and should be selected, to the extent possible, so that viscous 

 and surface tension forces are negligible compared with gravity forces. 

 Economy dictates that the models be as small as possible; however, if a 

 model is too small it may be difficult to adjust the test data analyti- 

 cally so that the results are of sufficient accuracy. 



(1) Rubble-Mound Breakwaters, Jetties, Wave Absorbers, and 

 Seawalls . Rubble-mound stability models are designed and operated based 

 on Froude's law, and the scale effects are minimized as much as possible 

 by selecting the largest practical size of model, relative to the proto- 

 type size. Within the limits prescribed by the need for accurate results, 

 selection of scale is based on such practical considerations as the size 

 of model armor unit available (compared with the estimated size of proto- 

 type armor unit required for stability), depths of the available wave 

 flumes (con5)ared with the prototype depths), and the capability of avail- 

 able wave generators (compared with the prototype wave dimensions) . In- 

 sofar as possible, it is better to err on the conservative side. The 

 linear scales that have been used with considerable success in the past, 

 and can be used as a guide, range from about 1:5 to 1:70, model-to- 

 prototype, depending primarily on the prototype water depths and wave 

 dimensions relative to the dimensions of available wave flumes and the 

 wave generator capacity. With the large wave flume (15 feet wide, 20 

 feet deep, and 635 feet long) at CERC, which has a wave generator capa- 

 ble of generating waves up to 6 feet in height, it is possible to conduct 

 stability tests on rubble-moimd structures using linear scales from about 

 1:5 to 1:10, depending on the prototype water depths and wave dimensions. 

 These tests are expensive, however, and this facility is used mostly for 

 basic research and for the solution of special problems. Tests in the 

 large wave flume have also been useful in establishing the bases for 

 determining correction coefficients for scale effects (obtained for 

 similar models of smaller size). A large number of rubble-moimd sta- 

 bility tests and model studies have been conducted at WES during the 

 past 30 years, most of which were conducted in a wave flume 5 feet wide, 



4 feet deep, and 119 feet long with a wave generator capable of generat- 

 ing waves up to nearly 1 foot in height. The linear scales of the model 

 studies ranged from 1:30 to 1:68, with most of the studies conducted at 

 scales from 1:40 to 1:50. 



(2) Vertical-Wall and Composite Structures . Model studies are 

 seldom necessary for vertical-wall breakwaters or jetties that will be 

 situated in water depths sufficient to prevent the breaking of waves on 

 the structure face, if the waves approach with an angle of incidence of 

 90°, or nearly so, and if there is no appreciable overtopping of the 

 structure. For these conditions the theories of Sainflou (1928) (also 

 Hudson, 1953; and Kamel, 1968a) and Miche (1944a) (also Kamel, 1968a) -are 

 adequate for determination of wave forces. In conditions where breaking 

 waves occur, where the face of the structure is irregular, or where appre- 

 ciable overtopping occurs, model studies usually determine the disposition 

 and intensity of pressure over the face of the structure and the quantity 

 of overtopping water. 



336 



