Experimental and Numerical Studies of Liquid Dispersal from a Soft Projectile Impacting a Wall

Liquid release and spread as a consequence of impact of missile on a wall are of interest for the determination ofconsequences of an airplane crash on a structure. These phenomena have been studied in medium-scale IMPACT tests atthe Technical Research Centre of Finland (VTT). In these tests, deformable cylindrical steel or aluminium projectilesimpacted a solid concrete wall or a steel force plate. Some of the tests were conducted using a fluid filled ("wet")projectile. The length of the wet projectiles ranged from 0.5 to 1.5 m, the water mass inside the projectile from 15 to 68 kgand the impact velocity from 70 to 177 m/s.This paper concentrates on the methods applied during the impact tests to measure the liquid dispersal phenomena. Themain results of preliminary simulations of liquid spread are also presented.So far, the main parameters of the liquid phenomena measured in the experiments are the velocity and direction of theliquid front coming out from the ruptured projectile, water pooling area on the floor, extent of liquid dispersal far from thetarget, and drop size of the liquid spray. The velocity and direction of the front of ejected liquid within an approximately 2m distance from the target was measured using high-speed (1000 fps) video cameras. In addition, normal DV cameraswere located around the target to detect the angle and direction of liquid spread, a general view of the liquid spray, and theaverage velocity of the liquid front up to approximately 5 m from the target. The water-pooling area was detected using ameasuring grid drawn on the floor and photographing the wet areas. The extent of dispersal of liquid spray far from theimpact target was measured with the pure collection trays (steel plates located on the floor). Oil-coated trays were alsoused to capture the deposited droplets and to measure the drop size using the macro photography technique and properimage analysis software. Specific arrangements to photograph the size and velocity of airborne droplets in the vicinity ofthe impact target were also developed. This system consists of both a high-speed (1000 fps) and a high-shutter-speed (t = 1μs) camera and a stroboscope light (flash time 1 μs) for backward illumination.The preliminary simulations of liquid dispersal were made using the 3-D Fire Dynamics Simulator code (FDS). Themain objective of the analyses was to assess the usability of the code for the simulation of the two-phase flows involvinghigh-speed droplets and to support the experimental work by providing an initial estimate of the spray behaviour. Thesimulation results indicated that FDS is a tool usable in simulating this kind of behaviour provided that the initialconditions of air speed, angle of droplet release, droplet size distribution, and initial droplet speed are specified. Giventhese, the formation of the water cloud and the final extent of liquid dispersal can be predicted by FDS reasonably well.Future work with the FDS program should include a validation of the some sub-models. Also, some IMPACT tests usingthe real boundary conditions will be simulated.