Three years after the dam was completed the pressure was observed on the pipe leading from the warm spring under the dam. Water in the reservoir was at elevation 115. The pressure gage was put on at elevation about 20. It was kept on for one day and read 14 lb. One can hardly do more than speculate upon the significance of this observation as the nature of the connection between the reservoir and the warm springs is shrouded in uncertainty. Probably there is some connection; but it must be remote and indirect or the great difference in temperature would not persist between adjacent springs. The observation was an afterthought as the spring was piped into the power house simply to furnish a supply of warm water.

The area over which the pressure is effective is as important as the pressure itself so that it is only with caution that one may assign areas as being applicable to certain pressure observations. In order to form a reasonably safe conclusion regarding the pressures under a dam there must be available observations at a large number of points, and the observations must be consistent.

However interesting it may be to speculate upon the structure of the rock foundation and the conditions under which water may pass through it from the reservoir, the proper course is to control those conditions instead of being controlled by what we imagine they may be. The entire solution is in the line of realizing the condition, elsewhere stated as desirable, that a masonry dam should properly be two things: first, a zone as nearly water tight as practicable; second, a structure for the support of that zone.

Two requirements must be met to eliminate uplift pressure as completely as possible, whether in dam or in foundation: first, the upstream portion should be tight; second, the downstream portion should be relatively open. In the foundation the first requirement is met by the cut-off trench, so called, under the heel of the dam. As a general proposition it may be assumed that seams diminish in number and width as bed rock is penetrated. Therefore a cut-off trench of whatever depth filled with masonry will intercept and prevent leakage, except possibly such minute quantities as would readily escape through the seams downstream from the trench. To extend this tight zone and intercept leakage at greater depths than would be economical, necessary, or even attainable with the cut-off trench, grout may be forced into the seams through holes drilled for the purpose. The process, which has been described elsewhere, is simple. If conducted with care it should be most effective.

Drainage Of Foundations

To accomplish the second requirement, noted in the preceding paragraph, i.e., to facilitate the escape of whatever water does leak by the cut-off trench, it has become the practice in recent years to lay drains under the dam from the cut-off trench to an outlet below. The best practice would be to start the drains from the downstream side of the bottom of the cut-off trench rather than from any point further downstream in the path of the water. This may be argued as follows: Water passing the cut-off trench conceivably might exert some pressure in coming up the downstream side of the trench in its search for some outlet seam. Then when it enters a seam it might be that the tightest part of its entire channel is downstream, at or toward the toe of the dam, in which case there would be a resulting uplift pressure. To start the drains from the downstream side of the bottom of the cut-off trench would be the cheapest and most effective method of draining the foundation to that depth. In case it is desired to extend the open zone to a depth corresponding to that of the tight zone produced by grouting, holes may be drilled and left open with free connection with the drainage system. In several recent dams provision has been made for the future drilling of such holes should it be considered desirable. To be certainly effective such holes should be drilled only after all grouting operations have been completed in the adjacent tight zone.

It need hardly be pointed out that with any or no drainage system the uplift pressure on the bottom of the dam cannot be less than that due to the head from any backwater that may stand at the toe of the dam. In studying any results obtained from testing holes in the foundation (as described on pages 19 to 23) and in devising and carrying out the entire treatment, it should be borne in mind that the weight of the masonry will tend to close or reduce any horizontal seams in the foundation; also that after the dam is vbuilt there is a redistribution of pressures consequent upon raising the water behind the dam.

A drainage system like that outlined above, if intelligently devised and faithfully constructed, will so certainly prevent uplift pressure as to eliminate that question entirely from the realm of practical consideration.