28 BULLETIN 194, U. S. DEPARTMENT OF AGKICULTUKE. 



DESCRIPTION OF CHANNELS. 



The descriptions in the following pages are to be considered as 

 supplementary to the table beginning on page 19, which gives all 

 the information necessary to a clear understanding of the hydraulic 

 conditions holding at various tests with the exception of a detailed 

 description of the channel, which would have made the table too 

 cumbersome. The descriptions of the channels follow the same 

 order and are numbered hke those in the table, "v^'ith the exception 

 that data obtained from other sources than the work of this division 

 are abstracted in the Appendix commencing on page 62. 



CONCRETE LININGS. 



No. 1, Expt. S-26, New York Canal, Payette-Boise project, U. S. Reclamation 

 Service, Idaho. A large canal, in concrete, rough as an orange, with plastered expan- 

 sion joints. Experiment rated as class C, because water issues from an open check 

 in canal about 600 feet above station and the canal passes into an earth section 

 about 100 feet below station 6. (See PI. I, fig. 1.) Cross sections developed from 

 office notes of United States Reclamation Service through courtesy of W. G. Steward. 

 Impossible to determine condition of bottom with water in canal. There is a slight 

 curve in the reach which disturbs the filaments of current. For additional data on 

 this canal see Nos. 2 and 3, Table I. Coefficient n=0.0101. 



No. 11, Expt. B-4, Ridenbaugh Canal, Nampa-Meridian Irrigation District, Idaho. 

 This test was made two years before and covered about the same reach as in test No. 

 13 below. Mr. Bark found a slightly lower value for slope than later experimenters, 

 which accounts for the lower value of n found. The preponderence of evidence 

 would indicate that a value of about 0.0125 is right for this nearly perfect piece of 

 concrete, to include both tangents and curves. Coefficient n=0.0110. 



No. 12, Expt. S-24. This lining in the same canal as No. 11 is a very smooth, 

 hand-troweled, cement wash on a base of concrete 3J inches thick. The reach is on 

 tangent with about a 6° curve beginning at station 9. The lining was placed in slabs 

 16 feet long with iron dowel pins and strips of tarred paper between slabs. After 

 the forms were removed the joints were poured with a neat cement. As a rule the 

 joints are as smooth to the hand as any other part of the lining (PL I, fig. 2), though slight 

 cracks are opened during cold parts of the day. This is an exceptionally well-made 

 Lining, and this, coupled with the fact that the curves are spu'aled into the tangents, 

 accounts for the very low value of n as found by all experimenters. For additional 

 experience on this canal see Nos. 11 to 15, Table I. Coefficient n=0.0121. 



No. 13, Expt. S-24a, was made on the same canal as S-24, but the reach included 

 not only the 901 feet of tangent as above, but also the above-mentioned curve, which 

 was about 600 feet long, and a short reach of tangent below the curve, making the 

 total reach 1,819 feet long. As is to be expected, the value of n is a little higher 

 than on tangent. The slabs on curves were but 12 feet long. Coefficient n=0.0129. 



No. 14, Expt. F-3. This experiment was made on approximately the same reach 

 of canal as S-24, but was 1,020.6 feet long, with one slight curve in the reach. The 

 slopes of the surface in this and experiments 26 and 28 were found by a line of levels 

 run between the ends of reaches as usual, but the water surface was found by means of 

 a gauge constructed on the piezometric principle. The hydraulic grade as given in 

 the table is the mean of 23 tests. An instrument of this form should give the surface 

 at the time of reading very closely, but the experimenter must be sure that this slope 

 is the same as the one held at the time of the discharge measurement. Coefficient 

 71=0.0124. 



