Page:Willich, A. F. M. - The Domestic Encyclopædia (Vol. 2, 1802).djvu/307

F I L following conclusions, which ap- pear to be - > rll-founded. . Thai tli-- difference of the niveau, or w ater c el, has an es- lential influence on th quantityof the purifi d water thus obtained ; 2. That a prolongation of the stra- tum of sand do snot considerably diminish the product pi the filtre, but remarkably contributes to the purity df the fluid. 3. That if the water be forced to pass through the sand with increased velocity, it will be less pure than by a lowing it a proper time for its passage ; and, -1. That a machine of the di- mensions above described, will fur- nish about three quarts of water in an hour, or eighteen gallons in twenty-four hours. This quantity, however, being too large in pro- portion to die size of the machine, it is advisable, either to lessen the difference of the water-fall ; or, which is still better, to prolong the stratum of sand, in order to reduce the filtration of the water to half die quantity above stated, and to obtain it in greater purity. Thus, a filtering apparatus eighteen inches long from AtoD, two inches thick, and four broad, would afford every hour sis pints of very pure water. If, therefore, so small a machine, containing a very moderate stratum of sand, and requiring only a dif- ference of two or three inches in the height of the water, furnishes a clear and pure fluid, it follows that an apparatus on a larger scale, provided with a bed of sand from five to six feet long, and admitting of a difference from twelve to eighteen inches in the fall of the water, might be usefully employed in public wells, hydraulic machines, and even in camps, for the supply of a i a my. In the construction of large fll- F I L [277 tering machines, Prof. Fakkot justl) veSj that they should not b<- extended in the direction A,< , D, to a greater length than i3 absolutely necessary; as, in this they will not require any considerable difference in the fall and rise of the water : on the other hand, their breadth and thickness may be accordingly increased.-— Thus, the diameter of such a machine would still more resemble that of a syphon, as is represented in the annexed cut. This form might also be adopted for smaller machines, especially such as are designed for travellers, two of whom might be amply pro- vided with pure water, and in a very short time, by a vessel of the following dimensions : from P, to Q, eight inches long 5 from P, to R, twelve inches high ; and the whole four inches in breadth. If the form last delineated be employed on an extensive scale, there should be a trap door in the lowermost part marked R, so con- structed that it may fit exactly, and admit no passage to the water : this aperture would serve only for the removal of the sand, when it is ren- dered foul by long use. In the smaller machines, intended for tra- velling, such a door is unnecessary, as they may be easily emptied of their contents through either of the orifices P, or Q. Instead of this ad- dition to the latter, the upper room (which in the first of these cuts is circumscribed with the letters B,F, T 3 E), might