Improvised explosive devices (IEDs) and other traditional under-body blast weapons are a significant threat to military ground vehicle systems. Engineers and scientists attempting to analyze the effects of under-body blast events have an array of commercial and internally developed tools at their disposal, each with its own set of limitations. Finite element (FE) models offer the capability to evaluate the entire event sequence from the detonation of the buried explosive to the response of the occupant. However, FE models for full system-level mine events require significant time for input preparation, debugging, and computation.

The technical areas for a blast event in the Under-Body Blast Methodology (UBM) project.
Computational times can be significantly reduced if simplified air-blast loading approaches are used to estimate the mine blast load. Current approaches may improve efficiency, but require significant engineering judgment for proper application. An under-body blast methodology (UBM) project seeks to develop a robust modeling capability that is accurate, efficient, and flexible enough to support an array of under-body engagement conditions.

The UBM effort includes evaluation of both high-resolution (HR) physics models and reduced-order (RO) engineering models. One key aspect of the UBM project is the assessment and validation of models used to predict the loading from a mine blast event. Controlled experimental data is therefore required not only for validation of models, but also to establish the inherent variability of impulse.

Analysis and preliminary modeling of data generated from subscale impulse experiments for simple v-angle structures constructed with a top floor plate was performed. The experiments were conducted at a facility used for subscale mine impulse testing. The facility uses a buried steelwalled box that is 6 × 6 feet with a depth of 4.5 feet, referred to as the “Sandbox.” The Sandbox has a 1-kg explosive capacity and can be filled with any desired soil type. Targets are placed on angle iron rails resting on supports at both ends of the box. The supports are adjusted to achieve the desired standoff from the top surface of the soil. A high-speed camera located at ground level is used to observe the trajectory of a target. The motion of points on the top plate is acquired using two stereo high-speed cameras situated atop a nearby tower.

The experiment compared different target geometries subjected to a buried charge. The basic target design was a V-shape, composed of A-36 Mild Steel. The soil used was a sand-clay mixture. For each shot, the Sandbox was filled and lightly tamped. Before detonation, a densitometer was used to determine the dry density, wet density, and moisture of the soil. Readings were taken at three locations: the center of the box, and two locations 304.8 mm away in opposite directions from the center. After all measurements were taken, the charge was put into place. The charge used for all experiments was buried C4. The explosives were hand-shaped into cylinders with a 1:3 height to diameter ratio.

Target motion data was acquired using two high-speed cameras and was processed with stereo-digital imaging correlation software. The cameras were synchronized so that they recorded the event identically from two different positions. A speckled pattern on the surface of the target provided the software with a way to track its motion as the speckle marks shifted pixels. The software used the pixel tracking to determine the rigid motion of the target and produce a strain field of the target’s top surface. This information was used to determine the impulse imparted on the target and the oscillatory motion of the target’s top surface.

One method of estimating the impulse delivered to the target is to assume a truly impulsive load delivery and then solve the kinematic equations of motion. The oscillatory motion of the top surface of the target was determined using a combination of output from the software and the already determined rigid motion of the target. The motion of the surface was oscillating while the target was moving upward. Therefore, the data for the rigid motion of the target were subtracted from the data for the motion of the center of the surface, which provided the purely oscillatory portion of the motion.

The dependent variable was the observed impulse. The independent variables chosen for analysis were target angle, target mass, charge mass, soil moisture, and charge offset. Observing only the correlation coefficient between charge offset and impulse, one might conclude that as charge offset increases, observed impulse increases. However, charge offset is highly correlated with charge mass. This is because the offset shots were conducted with 800 g and 1000 g charges in all but one event, which was a 600 g charge.

The LS-DYNA finite element code was used to simulate the mine blast loading and structural response for the various target geometries. There are advantages to using the air-blast loading model because it’s easy to use and quick running. However, the factors that must be used to scale the peak pressure or charge mass are not easily predicted. In order to find factors to match total impulse for each shot number, a baseline LS-DYNA model was run using the test charge mass and a unit scale factor applied to the peak pressure time history load curve.

This work was done by Craig Barker, Douglas Howle, Terry Holdren, and Jeffrey Koch of the Army Research Laboratory; and Raquel Ciappi of SURVICE Engineering Company. ARL-0149

This Brief includes a Technical Support Package (TSP).
Analysis of Mine Impulse Data Using Stereo-Digital Image Correlation

(reference ARL-0149) is currently available for download from the TSP library.

Don't have an account? Sign up here.