Tile research described here was part of a wider research program sponsored by the National Science. Foundation RANN (Research Applied to National Needs program entitled “Design Characteristics of Near-Surface Rock Formations for Underground Construction of Urban Facilities”. The project was intended to provide design information for near-surface caverns and included the construction of a full scale 50' wide x 100' long x 8'-6” high (15m x 30m x 2.6m) underground test room on the University of Minnesota campus.

Since near surface rock formations usually contain substantial jointing and because horizontally bedded rock layers were primarily of interest in the test room program, it was decided to study in detail the behavior of “linear arching” or “Voussoir Beam” action.

There is sample physical evidence of tile existence of “linear arches” in underground excavations and Figure 1 shows are idealization of the linear arching action that can occur in underground cavern roofs.

(Figure in full paper)

The concept of linear arching in underground cavern roofs is generally atributed to Evans (1941) who carried out the first quantitative analysis of such action. Several other researchers have studied this problem including must notably Bucky (1934, 1938), Haycocks (1962), Wright (1949, 1963, 1970, 1972, 1973, 1974) and Cox (1974). An extensive review of the development of roof design methods for bedded and jointed rock is given in Sterling (l977).

Previous research on the linear arch in bedded roof control had mostly concentrated on matching tile theory to the behavior of model beams in the linear range below ultimate load. Near ultimate load, however, non-linear behavior along with some crushing at corners occurs. This spreads the depth of compression zones caused by arching, and hence, it alters the parameters on which a design for the linear range would be based. In a sequence of beds of unequal thickness, the study of linear arching action at and beyond ultimate load also allows the calculation of the stability of a sequence where one or more beds may have partially failed. For these reasons, the experimental investigation was designed to record the behavior of linear arches through a complete load/deformation cycle.


The following basic requirements were considered essential for the apparatus which was used to study parameters affecting linear arch behavior up to and beyond linear arch behavior up to and beyond peak load:

  • The ability to laterally constrain the beam and measure the lateral thrust at any stage of the experiment without pre-determining the position of the thrust at the ends of the beam.

  • The ability to follow the downward portion of the load/deflection curve while continuing to measure the lateral thrust.

  • The ability to test a range of shapes, sizes, and materials for the beam.

  • The ability to vary the conditions of sup- ort at the ends of the beam and the type of loading applied to the beam.

This content is only available via PDF.
You can access this article if you purchase or spend a download.