Analysis and Observation of Keyblock Occurrence in Tunnels in Granite

Tunnels in the granitic Climax stock at the Nevada Test Site were inspected for keyblock occurrence. Thirteen relatively large pyramidal keyblocks were located in about 230 m (750 it) of tunnel. Orientations, dimensions, joint roughness coefficients, and joint compressive strengths were measured for the faces of nine keyblocks. The measurements were used to back-analyze keyblock stability to check the validity of a numerical model for keyblock behavior. A newly developed three-dimensional numerical model correctly predicted whether or not each keyblock failed in six of eight cases. The two exceptions involved stress conditions and keyblock locations that clearly violated the assumptions that underlie the stability model. The observations also allowed the effect of tunnel orientation on keyblock occurrence to be evaluated. Of the thirteen observed keyblocks, nine large blocks were found in less than 30 m (100 ft) of the Piledriver drift (which runs N56E) while only four blocks of comparable size were observed in 200 m (656 ft) of the Spent Fuel Test--Climax drifts (N61W). The orientations of the keyblock faces were checked against known discontinuity set orientations; some of the faces correlated well with mean set orientations, whi.le other faces were fringe members of sets. Stereographic techniques were used to examine the frequency of keyblock occurrence and to determine the maximum keyblock size that could be encountered with a given combination of discontinuities. Keyblock occurrence was found to be affected by tunnel orientation in a predictable way; this can be used to guide or improve tunnel design practices.