Rock Anchor Testing: Piston Pull-Out, Cone Lift-Out, and the Pells Research

Why rock is different from everything else

Rock is a natural material with inherent variability that manufactured substrates like concrete cannot replicate. A single rock face can exhibit strength variations of an order of magnitude within metres. Discontinuities, joints, bedding planes, foliation, weathering zones, can reduce anchor capacity to a fraction of the value predicted by intact rock strength alone.

Anchor manufacturers have limited data on the performance of their products in rock. The variability is too great for a single set of published capacity values to be meaningful across the range of rock types and conditions encountered in practice.

No design code exists for anchoring to rock. AS 5216:2021 covers concrete. EOTA TR 054 covers masonry. Neither addresses rock. This gap means that rock anchor design must rely entirely on project-specific testing, combined with engineering judgement informed by rock mechanics principles and published research.

The Pells research

In 1978, P.J.N. Pells, G. Mostyn, and B.F. Walker published a foundational paper for the Australian Geomechanics Society summarising research on foundations in sandstone and shale in the Sydney region. Their work identified two primary failure modes for anchors in rock.

Piston pull-out

Piston pull-out occurs when the anchor bar extracts from the rock bore without displacing the surrounding rock mass. The bond between the anchor (or adhesive) and the rock bore surface has failed. The failure surface runs along the bore wall, and the anchor slides out like a piston from a cylinder.

This mode occurs when the rock mass is strong relative to the bond, the rock can resist the stresses, but the adhesive-rock interface cannot. The capacity is governed by the side shear strength along the bore length.

Cone lift-out

Cone lift-out occurs when a cone of rock is extracted together with the anchor. The rock mass strength is lower than the bond strength, and the failure surface propagates through the rock at approximately 45 degrees from the anchor axis. This is analogous to concrete cone breakout but occurs in rock, where the tensile strength is governed by discontinuities rather than the intact material strength.

Why chemical anchors are recommended for rock

Mechanical expansion anchors require a consistent bore diameter to generate radial clamping force. Rock rarely provides this condition, drill bits wander, overbreak occurs at discontinuities, and hardness variations within the bore produce uneven diameters.

Chemical anchors fill the annular gap between the bar and the rock, providing a continuous bond that accommodates bore irregularities. The adhesion and micro-interlock mechanism is fundamentally better suited to the geometric variability of rock-drilled holes. Products approved for use in diamond-cored holes typically have adhesive characteristics that improve stress distribution to the base material.

Testing approach for rock

Both ultimate testing and proof testing are essential for rock anchor applications:

• Ultimate testing: during the design phase establishes the actual anchor-rock capacity in the specific substrate. The statistical method (minimum 15 tests per BS 8539) is recommended to characterise the rock-anchor system with adequate statistical confidence.
• Proof testing: after installation verifies both installation quality and substrate quality at each connection location. Substrate quality can vary between fixing points on the same rock face due to non-visible internal cracks, water ingress, and chemical weathering.

Test locations should be distributed across the site to capture the range of substrate conditions production anchors will encounter, including weathered zones, joint-affected areas, and competent rock. Testing only in the best rock will produce non-representative capacity data.

Higher testing rates than those used for concrete may be appropriate, given the increased uncertainty inherent in rock substrates.

References

• Pells, P.J.N., Mostyn, G., and Walker, B.F. (1978), Foundations on Sandstone and Shale in the Sydney Region
• BS 8539:2012+A1:2021, Annex B.2.3
• AEFAC TN05 Volume 3, Guidelines for Site Testing of Anchors: Ultimate Tests