Page 245 - Rock Mechanics For Underground Mining

ROOF BED DEFORMATION MECHANICS

8.3 Roof bed deformation mechanics

Prediction and control of the deformation behaviour of the immediate roof of an
opening has been the subject of formal engineering investigation for more than a
century. Fayol (1885) published the results of investigations of the behaviour of
stacks of beams spanning a simple support system, simulating the bedding sequence
of a roof span. By noting the deflection of the lowest beam as successive beams
were loaded onto the stack, Fayol demonstrated that, at a certain stage, none of the
added load of an upper beam was carried by the lowest member. The gravitational
load of the uppermost beams was clearly being transferred laterally to the supports,
rather than vertically, as transverse loading of the lower members. The process of
lateral load distribution, associated with friction mobilised between the surfaces of
the upper beams, was described as arching. The basic concept proposed by Fayol
was that a rock arch was generated above a mine opening in a stratified mass, with the
rock beds between the excavation roof and the rock arch constituting an effectively
decoupled immediate roof for the opening. Fayol proceeded to apply his results to
surface subsidence phenomena, rather than to underground excavation design.
The first rigorous analysis of roof bed performance was attempted by Jones and
Llewellyn-Davies (1929). They mapped the morphology of roof failures, and sought
to explain the localisation of failure in terms of arching principles. Bucky and Taborelli
(1938) studied physical models of the creation and extension of wide roof spans. They
used initially intact beams of rock-like material, and found that, at a particular span, a
vertical tension fracture was induced at the centre of the lower beam. Increase in the
mined span produced a new central fracture, and closed the earlier fracture. This sug-
gested that the central fracture is the dominant transverse discontinuity in the roof bed.
Recognising the relation between vertical deflection, lateral thrust and stability
of a naturally or artificially fractured roof bed, Evans (1941) undertook a seminal
set of investigations of roof deformation mechanics at the Royal School of Mines.
This work established the notion of a ‘voussoir beam’ spanning an excavation, using
the analogy with the voussoir arch considered in masonry structures. Evans also
developed an analytical procedure for assessing roof beam stability, but an error in
statics and failure to handle the basic indeterminacy of the problem limited its practical
application.
Significant experimental and computational investigations of roof bed mechanics
subsequent to those by Evans (1941) have been reported by Adler and Sun (1968),
Barker and Hatt (1972), Wright (1972, 1974) and Sterling (1980). The experimen-
tal studies by Sterling capture many of the key conclusions of the work by other
researchers and provide insights into the deformation and failure modes of roof rock.
The experimental arrangement used by Sterling is illustrated in Figure 8.4. A rock
beam, of typical dimensions 660 mm × 75 mm × 75 mm, was constrained between
steel end plates linked by strain-gauged tie rods. The beam was loaded transversely by
a servocontrolled testing machine and a load spreading system. The experiment design
provided data on applied transverse load, induced beam deflection, induced lateral
thrust, and eccentricity of the lateral thrust. The typical response of an initially intact
limestone beam is given in Figure 8.5. The load–deflection plot, shown in Figure
8.5a, shows an initial elastic range (0–1). At this stage a transverse, central crack
developed in the beam, accompanied, in the test rig, by a relaxation of the applied

227