WRDS Library [Home] Digital Library Publications Videos Card Catalog

### Chapter V Sumamry and Conclusions

A constant flowrate to farmers when they need it and accurate flow measurement are important f actors in making efficient distribution of limited supplies of water in agriculture. The long crested weir has proven successful in providing minimal fluctuations in flowrate through canal turnouts so that farmers can maintain a fairly constant flowrate. This research used data from experiments with 67 different long crested weir models to develop equations which will permit long crested weirs to accurately measure the flowrate in canals.

Three types of equations were tested to see which gave the best prediction of the coefficient of discharge. The first type of equation was a multivariable linear regression. The second type of equation was a second-order polynomial, and the final type of equation was the power function. Of these three, the second-order polynomial provided the best prediction of the discharge coefficient, as shown by the r2 values of fitting the data to the Cd equation and the graphs showing the actual flowrate versus the predicted flowrate.

By applying certain limitations to the data used to develop the equations to predict Cd for all of the weirs tested, the bulk of the values of predicted flowrates fell within plus or minus 5 percent of the actual flowrates measured in the Laboratory. Equations (29) and (30) are general equations to predict Cd for all of the weirs tested, and both equations have r2 values greater than 90 percent. These can be used for suppressed long crested weirs as well as the full range of the contracted weirs that were tested.

If the long crested weir to be designed is a contracted weir, equations (32) and (33) are recommended to calculate the Cd. They both have r2 values greater than 90 percent and have improved r2 values from the equations used for all of the weirs. Because these equations deal specifically with the contracted weirs, they are able to give a better prediction of the flowrate than the more general equations.

Some standard weirs can measure the flowrate to within plus or minus 2 percent, but these long crested weir equations sacrifice some accuracy to provide a general application to many weir shapes. The Cd equations can give a flowrate prediction within plus or minus 5 percent of the actual for the weir models tested in the Laboratory. However, this research did not involve any tests on full-scale field weirs, so the applicability of the equations for full-scale weirs remains to be determined. The application of the equations is limited to use with the duckbill type of long crested weirs since they were the only type of weir that was tested; labyrinth and diagonal weirs have different flow characteristics than the duckbill weirs, so the equations should not be applied to them. The 67 laboratory models all had sharp crests, so the equations apply for sharp crested weirs. Crests that are rounded or square have different flow characteristics, and the sharp crested Cd equations would not apply directly to other types of weir crests.