(b) The horizontal movement of points at the ends of the concrete ribs and on the rock walls was measured by precise survey methods.

(c) The angular rotation of points on the reinforced concrete abutment beams and on the walls was measured by means of sensitive clinometers.

The main results have been described by T. A. Lang (1957). While the instrumentation and the records were not sufficient to give a complete picture of the behavior of the excavation, certain trends are recognizable. They are shown diagrammatically in Figure 12.

As previously mentioned, as the main machine-hall excavation was deepened, the concrete ribs became increasingly strongly compressed. This was accompanied by a rotation of the abutment beams on which they rested and resultant spalling of the concrete where the ribs joined the beams. At the same time the walls moved inward, with accompanying rotations. There was a definite correlation with construction activities. While deepening of the excavations was in progress, movements were relatively rapid, and on cessation of excavations the movements ceased.

There is a marked asymmetry in the measured deflections of the walls as the movements in the downstream side are considerably greater than those on the upstream side. There is also an asymmetry in the strains in the concrete roof ribs, as the strains on the downstream side also are considerably greater than those on the upstream side. This is assumed to be indicative of similar strains in the roof rock.

The observational data have not yet been fully analyzed. At this stage, however, the following interpretations seem feasible. The asymmetry in the roof strain may be due to the fact that the directions of the principal stresses are inclined to the vertical and horizontal because of the topography of the valley, and thus unsymmetrical with respect to the roof of the machine hall. The much greater deflection inward of the downstream wall relative to the upstream may be partly due to the same cause; however, the presence of fault B close behind the wall also may have an effect. This fracture has created a relatively thin and isolated slab which may be carrying a greater proportion of the load than the corresponding rock on the other side of the machine hall.

The movements were relatively very small (although quite large enough to be measured by the sensitive instruments) and within the range of elastic movements of the rock mass predicted by photoelastic studies. The rock in the roof and walls, although it is considerably jointed, has for the most part behaved as though it was essentially structurally continuous. It is considered that to a large extent this effect has been due to the rock bolting.

Subaudible Rock Noise

Measurements were made during construction on the frequency of occurrence of subaudible rock noise, following the method developed by the U. S. Bureau of Mines (Obert and Duvall, 1957). The equipment used was a commercially available unit consisting of a geophone, amplifier, and earphones. Observations were usually made by listening only, but sometimes recordings were made on magnetic tape.