Abstract

Reinforced Concrete (RC) is an important material in civil construction projects, and rigorous standards exist for rating the structural, wind, vibration, and cyclic design loads. In narrower applications, such as the design of protective saferooms, RC is also designed to bear impact loads which may be applied repeatedly. Although current experimental and computational methods allow for prediction of concrete damage from a single impact, there is no attempted study of damage progression from repeated impacts. Such a study is attempted on a well-defined slab impact test used in the rating of protection provided by RC walls. Multiple projectiles impact a chosen location and accumulating damage is predicted by means of numerical simulation. The simulation results are then correlated to an experimental test demonstrating similar effects.

The classic projectile impact problem is taken as the basis for computational analysis. Nonlinear wood and concrete material models are substituted for the conventional steel projectile and target, and a damage variable is defined to track cumulative effects of plastic strain. The simulation is then extended to additional impacts while preserving the damaged state of the slab. The development of damage and accumulating strain energy is computed, until damage extends throughout the entire thickness of the slab. Since the contact period per impact is estimated to be under a millisecond, an explicit dynamics formulation of the problem is implemented in commercial software LS-DYNA. The concrete stresses are computed using the Reidel-Hermaier-Thoma Model (RHT), which incorporates a smooth geologic cap model to provide a single continuous failure surface. In this manner, compaction, shear and tensile failures are represented, and consolidated for post-processing by means of a single damage variable.

To verify computational predictions, the impact is recreated experimentally. A set of wood projectiles and RC slabs are fabricated, to allow for repeatable tests. Initial and boundary conditions are recreated by means of a steel bracket for the slab and an air cannon for the projectile. After initial calibration, a repeatable projectile speed, impact location and momentum transfer is achieved. The slabs are impacted repeatedly until macroscopic damage is clearly visible on the front and back faces of the slab. Damaged slabs are then cross-sectioned for material failures and plastic deformation.

The simulations and experimental tests show consistency in predicting the progression of damage in RC slabs. In both cases, through-thickness failure was achieved in 2-4 impacts, depending on the initial kinetic energy of the projectile. Initial damage began as subsurface shear cracking, and in later impacts tensile failures, shear failure (plug formation) and compaction (pulverization) developed concurrently. Granted further verification is performed, there is a promising possibility that these methods and results could be used to provide safety ratings for structures designed to withstand multiple impacts, and to re-asses load ratings of damaged structures more accurately.

Publication Date

4-21-2017

Document Type

Thesis

Student Type

Graduate

Degree Name

Mechanical Engineering (MS)

Department, Program, or Center

Mechanical Engineering (KGCOE)

Advisor

Benjamin Varela

Advisor/Committee Member

Hany Ghoneim

Advisor/Committee Member

Sarilyn Ivancic

Campus

RIT – Main Campus

Share

COinS