Mitigation Of Lower Leg Tibia Forces During A Mine Blast On A Light Armoured Vehicle: Simulations And Experiments

Transcription

1 Mitigation Of Lower Leg Tibia Forces During A Mine Blast On A Light Armoured Vehicle: Simulations And Experiments Richard Smith 1, Senior Mechanical Engineer; Dinesh Shanmugam 2, Senior Materials Engineer; Ashley Marr 3, Senior Mechanical Engineer; Campbell James 4, Engineering Manager 1,2,3,4 Protected Mobility Systems, Land & Joint Systems Division, Thales Australia 1 richard.smith@thalesgroup.com.au; 2 dineshkumar.shanmugam@thalesgroup.com.au; 3 ashley.marr@thalesgroup.com.au; 4 campbell.james@thalesgroup.com.au Abstract: Safety of an occupant has a high priority in peacekeeping military operations. During a mine detonation under a light armoured vehicle, the impulse loads threaten occupant safety. These impulse loads are directly transferred to any part of the body that is supported by the seat or floor including the occupant lower leg, potentially causing serious injuries. A combined numerical and experimental approach was used in this study to obtain greater confidence in the injury assessment method. Blast tests were performed using the standard Hybrid III 50 th % dummy and data such as acceleration/time history from the vehicle and force/time history from the Hybrid III dummies were collected. The acceleration/time history measured from the floor plate under the occupant feet was used as inputs for the dynamic numerical analyses. A simplified static modelling technique was also developed to study relative hull distortion based on displacements at the time of approximate peak acceleration. This simple model enabled more rapid experimentation in the simulation space using ANSYS than would be possible with dynamic non-linear FEA. The explicit code within the finite element software LS-DYNA was used to perform the non-linear dynamic numerical simulations which eliminated any inherent convergence issues often associated with implicit codes such as ANSYS. A standard Hybrid III 50 th % dummy incorporated within LS-DYNA was used to predict the forces generated on the lower leg tibia due to the input accelerations that could then be compared to the measurements taken from the actual Hybrid III s. The simulations showed good agreement with the experimental data and therefore can be used as a means to interpolate mine blasts. Further, these simulations were used to revise the design to minimise the forces on the occupant lower leg tibia within the threshold conditions of fracture, subsequently verified by experiments. 1. INTRODUCTION During the detonation of buried explosives in the form of mines under the wheels of an armoured vehicle, it is understood that at this relative close proximity to the vehicle the effect of soil ejecta is the major component of loading, that is, the initial loading comes in the form of a high-speed plug of soil. The soil plug can have an initial velocity of up to 1500 m/s for a full-size charge [1], and the initial impact of these relatively dense products against a vehicle have the capacity to induce large loads in a very short period of time. It is suggested that changes in the flow field resulting from target geometry can create localized pressure spikes, particularly if the field stagnates inside re-entrant corners [2]. In general, the severity of the threat, and therefore, the severity of the loads on the occupant, depends on the distance between the occupant and the detonation point as well as on the inertia and rigidity of the vehicle hull structure and the interior structure such as the seat and seat mountings and the floor plate configurations [3]. When a seat or a footrest is mounted on or close to the deforming floor or hull plate large loads are transferred to the feet, ankles, legs and lumbar spine respectively. Additionally, the chance of injury depends on initial body posture, and the presence and use of personal protection equipment and restraint systems. Also age, gender, health, training, etc., may influence the injury probability [4]. It is expected that once the vehicle s structural integrity is secured, effects of fragments, overpressure, gases and heat will have minor physical effects on the occupant inside the vehicle [5]. However, the impact of the vehicle structure with the occupant s feet through the floor and other contact points such as pedals and foot rests, through extremely high acceleration pulses can result in fracture and disintegration of the leg bones such as the tibia. If these loads are not attenuated to survivable levels, it may lead to severe leg injury of the occupants [6]. In developing mine protection characteristics, the design and tuning of the dynamic behaviour of the vehicle structure, along with occupant protection systems, are demanding tasks which can be solved only in the context of the complete vehicle [6]. The required test series and qualification trials, some of which are conducted with fully equipped vehicles, are very time-consuming and costly. Apart from the high cost, the typically short procurement times sought by the customer for new vehicle systems call for a significant reduction in development times. Research shows that the use of numerical simulations in the design process is an important tool for the design layout as well as to substantially reduce the time requirements. One approach is to utilise finite volume explicit codes to model loading events, such as explosions, to generate the input loads on the structures under investigation [7], [8]. However, where the required experimental inputs are available the finite volume approach is not required to generate the input loads and the finite element method is then used only to analyse the associated structural responses [9]. In this paper the finite element approach only has been used to predict the responses of the vehicle and occupants of proposed solutions based on past actual test measurements.

2 2. PROBLEM DEFINITION The design of mine-protected vehicles places high demands on the developmental process. In this case, a study has been conducted to find the effects of a mine detonated underneath one of the wheels of a light armoured vehicle on the occupant. It is understood that the floor structure must have a high level of stiffness in order to resist deflection as much as possible ensuring a minimal level of deformation of the floor into the crew compartment. On the other hand, the stiffness of the structure must not be so high that the floor structure breaks up and collapses due to material failure. The main focus of this paper is to study the blast loading effects on the lower tibia of the occupant seated at the driver/co-driver positions by 1) using Energy Absorbing (EA) mats on the floor and 2) increasing the structural integrity of the floor-well. These were both performed via simulation and experiment. 1) To predict the performance of the EA mats the LS- DYNA simulation package was used. Acceleration/time measurements recorded on the floor (Test 1 mine blast) along with true stress-strain curves of the mats were used as inputs. A digital Hybrid III 50 th % dummy inbuilt within the LS- DYNA code was then utilised to predict the impact forces on the relevant parts of the body. In order to initially validate the LS-DYNA model, drop tests were performed on a simplified test rig, which mimicked a floor structure, to develop acceleration/time history data that was then input into the simulation. The predicted force/time outputs from the digital Hybrid III were then compared with the actual drop test Hybrid III force/time results. 2) To give an indication of the relative improvement in structural rigidity of the floor well, due to the addition of stiffening plates, a simplified static structural modelling technique was developed based on simplified loading assumptions using ANSYS. This was done to enable a more rapid experimentation process in the simulation space than would be possible using a full non-linear dynamic FEA approach of the complete hull. For validations, the simulation Hybrid III force/time history was compared with the measured force/time history from the actual Test 2 mine blast. 3. SIMULATION AND EXPERIMENTAL SETUP 3.1. Dynamic Matting LS-DYNA Simulation The actual seating position and region around the occupant in the vehicle formed the basis for the numerical simulations and the drop test configuration. Figure 1 shows the numerical simulation as set up within LS-DYNA that was used to predict the occupant responses to test acceleration inputs. The torso is rigidly fixed in 3-D space as this is assumed to have little or no influence on the motion of the legs as under real conditions it is secured to the seat using a belt. The occupant model is a 50 th % Hybrid III dummy as supplied with LS- DYNA. The dummy is assumed to be seated on this rigid seat with its feet resting on the EA mats under investigation. The bottom face of the mat is fixed in the same reference space as that of the fixed torso. The input acceleration can then be applied to this reference space simulating the acceleration of the vehicle vertically upwards. Under the acceleration input the inherent inertia of the legs generate the destructive forces within the leg members that are targeted for reduction to acceptable levels via the EA mats under investigation. Figure 1: LS-DYNA Simulation Setup The positions of the arms are left at their default orientations as they play no role during the simulation. Similarly an occupant restraint mechanism such as a seatbelt has not been modelled as the torso is fixed. The main region of activity lies between the foot and hip of the dummy. The motion of the fixed reference frame, i.e. the vehicle, is controlled via the prescribed acceleration curves. This same setup serves to simulate the conditions of a) the mine blast, via the input of the typical acceleration curve as seen in Figure 2, which was measured at the approximate position on the floor that the right foot would rest, and b) the drop test, via the curve seen in Figure 3. While modelling, care was taken to position the feet as close as possible to the mat, as even the smallest gap between the feet and mat can significantly alter the lower tibia axial compressive force. During the initial sequence of the simulation the feet are allowed to settle onto the top mat under gravity therefore eliminating any contact gap. For the EA mats within LS-DYNA the *MAT57 Low Density Foam material algorithm was used with a characterised stress-strain compression curve mapped to it from a standard compression test.

3 Figure 1: Typical acceleration/time curve for floor response from Test 1 of the mine blast Figure 1: Drop test acceleration impulse 3.2. Static Structural ANSYS Simulation To obtain an indication of the relative improvement in structural floor well rigidity a simplified static ANSYS model was developed. Within this simplified model a section of the floor-well region was extracted from a complete vehicle CAD model. This section model included a portion of the inclined hull outer wall as well as the appropriate internal floor plate sections and relevant support members as seen in Figure 4. B A C Figure 2: Static ANSYS Simulation Setup Reference A in Figure 4 indicates the position of the rested feet. Reference B in Figure 4 indicates the pressure applied to the outer hull plate. Reference C in Figure 4 indicates the prescribed motion edge. This simplified model was developed purely to determine what the influence the addition of rigid plates would be to the deflection of the floor. It was determined during an earlier study that the deflection of the floor region could be reduced via stiffening. The static model was set up with an arbitrary pressure applied to a region of the outer hull plate in and around the region where a mine blast would load the vehicle during a real event. The section was fixed in space along a number of plate and member edges while a single lower hull plate edge was allowed to move a prescribed amount relative to the fixed reference frame. This prescribed motion mimicked the lower section of the hull moving under the extreme loading of the blast relative to the rest of the hull. With this basic setup and assumptions a study could be undertaken to determine the relative effects of the additional rigid plates. Figure 4 shows the floor plate section before the addition of the rigid plates which reflects the results seen during Test 1 mine blast.

4 Figure 5 highlights the initial deflection levels for comparative analysis before the additional rigid plates were added. It is important to note that due to the simplified nature of this analysis the actual deflection levels indicated are arbitrary and only useful for comparative purposes. In this fashion a determination of the increase in rigidity could be made due to the additional plates. cause and structural mechanism for this difference in impulses was successfully modelled. By measuring the relative deflection of points at the approximate positions that the feet would lie on the floor from the "before rigid plate addition" model and the "after rigid plate addition" models a relative structural stiffness improvement could be extrapolated. Figure 7 highlights the relative influence of the additional plates on the rigidity of the floor-well region. The internal wall is predicted to be prevented from buckling and the relative vertical deflection of the feet positions is shown to significantly reduce. During the subsequent Test 2 mine blast with rigid plates added the actual results were observed to agree with this simplified model very closely. Figure 3: Static ANSYS deformation showing internal wall buckling and relative deflection of floor plate Figures 6 shows the additional rigid plates added to the floor-well area. The positions of the plates were chosen to minimise the floor vertical movement which can adversely increase the impulse into the occupant s lower legs. The plates mounted on the internal vertical cover wall were intended to prevent the buckling of this wall plate (as indicated in Figure 5) and hence reduce allowable vertical movement of the attached floor. Figure 5: Static ANSYS deformation with additional plates 3.3. Experimental Setup Experiments using 50 th % Hybrid III dummies under vertical drop tests and mine blast tests were carried out to validate the simulations performed by LS- DYNA. Hybrid III dummies are able to withstand crash and vehicle mine detonation tests while simultaneously (and reproducibly) measuring the accelerations, forces and moments in different body parts. As seen in Figure 8, the dummy in the drop test was buckled onto the seat of the test sled in the vertical position with the legs placed on the floor according to the driver/co-driver s normal resting position. Figure 4: Static ANSYS simulation setup with additional plates From earlier full scale mine blast tests it was found that the force impulse exerted on the lower tibias of the Hybrid III dummy were not of the same magnitude between the left leg and the right even though the position of the feet on the floor were only a short distance apart. In reality the magnitude of the right leg was noticeably greater than that of the left leg even though the left leg was closer to the outer hull and hence closer the blast. From this simple static model with fixed boundary conditions and prescribed motion it is believed that the underlying Figure 6: Drop test setup

5 The dummy s arms were strapped around its lower thighs as shown in Figure 9 so as not to cause damage during the rapid upward movement of legs during the drop test. In addition to the measurement taken from instrumented dummy, accelerations were recorded on the floor which provided input for the simulation as shown earlier in Figure 3. Figure 7: Drop test leg setup For the mine blast, full scale tests were conducted with test conditions followed according to Allied Engineering Publication (AEP-55), Volume 2 [10]. The loads on the occupants were measured using instrumented Hybrid III dummies together with other vehicle measurement equipment, and in particular accelerometers in this case. The forces measured from these Hybrid III dummies serve as the injury assessment criteria as described in AEP-55, Volume 2 [10]. Also, these force measurements are used for validating the numerical analysis. The data from the numerical simulations are compared against experimental data. 4. VALIDATION As described previously the LS-DYNA simulation developed for the EA mats was conducted using the input acceleration as measured from an accelerometer placed on the vehicle floor at the approximate position of the right foot during Test 1 mine blast before any of the additional rigid plates were in place. The ANSYS simulations predicted that the additional plates should have a significant impact on the acceleration response of the floor. It was recognised therefore that any improvement in tibia force/time results measured during Test 2 mine blast would be due to the combined effect of both the EA mats and the change in structural rigidity of the floor region. For this reason, the results from the LS- DYNA simulation would not be strictly comparable to the actual measured results and should by inference over predict the resultant tibia forces. The drop tests were therefore performed to gain a level of confidence in the simulation setup that could be transferred to the blast simulation. The drop tests have a far more reproducible and predictable acceleration curve (Figure 3). As described earlier, this impulse curve was also used as an input into the same LS-DYNA simulation which could then be directly compared with the Hybrid III tibia results from the drop test. Figure 10 illustrates this direct comparison showing a quite good correlation between the drop test and simulation space. The simulation is seen to over predict the tibia force which is likely to be due to an overly stiff material model or slight differences in the Hybrid III leg placements. Figure 11 illustrates the comparison between Test 2 mine blast and the LS-DYNA simulation. As expected, the simulation over predicted the force levels mainly believed to be due to the increased floor rigidity. Considering this, when comparing the trace from the actual mine blast with that from the LS-DYNA simulation this is considered a good correlation. Both the drop test simulation and the blast input simulation have a high level of correlation to the actual measurements while both being conservative. This simulation methodology can therefore be deemed to be valid for future design studies. Figure 7: Drop test Hybrid III lower tibia Force/Time comparison

6 Figure 7: Mine blast Hybrid III lower tibia Force/Time comparison 5. CONCLUSIONS Numerical and experimental studies were performed to study the injury assessment on a fully instrumented standard Hybrid III 50 th % dummy. The acceleration/time history measured from the floor plate under the occupants feet were used as inputs for the dynamic numerical analyses on the EA mats. Both the drop test simulation and the blast input simulation have a high level of correlation to the actual measurements while both being conservative. This simulation methodology can therefore be deemed to be valid as a part of future design studies in the effects of mine blasts on occupants 6. REFERENCES 1. Bergeron, D.M. & Gonzalez, R. (2004) Towards a Better Understanding of Land Mine Blast Loading, The Technical Cooperation Program, Weapons Group, Conventional Weapons Technology, WPN TP- 1 Terminal Effects, Key Technical Activity 1-34, Published June Tremblay, J.E. Bergeron, D.M. & Gonzalez, R. (1998) Protection of Soft-Skinned Vehicle Occupants from Landmine Effects, The Technical Cooperation Program, Subcommittee on Conventional Weapons Technology, Technical Panel W-1, Key Technical Activity 1-29, Published August. 3. Radonić, V. Giunio, L. Biočić, M. Tripković, A. Lukšić, B. & Primorac, D. (2004) Injuries from Antitank Mines in Southern Croatia, Military Medicine, Vol. 169, April, pp Yoganandan, N. Pintar, F.A. Kumaresan, S. & Boynton, M.D. (1997) Axial Impact Biomechanics of the Human Foot-Ankle Complex, Journal of Biomechanical Engineering, Vol. 119, November, pp RTO-TR-HFM-090 (NATO) (2007) Test Methodology for Protection of Vehicle Occupants against Anti-Vehicular Landmine Effects, Final Report of HFM-090 Task Group 25, Published April. 6. Tabiei, A. & Nilakantan, G. (2007) Reduction of Acceleration Induced Injuries from Mine Blasts under Infantry Vehicles, 6th European LS-DYNA Users Conference, Gothenburg, Sweden. 7. Showichen, A. (2008) Numerical Analysis of Vehicle Bottom Structures Subjected to Antitank Mine Explosions, PhD Thesis, Cranfield University. 8. Genson, K. (2006) Vehicle Shaping for Mine Blast Damage Reduction, Master s of Science Thesis, University of Maryland. 9. Honlinger, M. Glauch, U. & Steger. G (2002) Modelling and simulation in the design process of armoured vehicles. Reduction of military vehicle acquisition time and cost through advanced modelling and virtual simulation. Paris, France, April AEP-55 (2006), Procedures for Evaluating the Protection Level of Logistic and Light Armoured Vehicles - Mine Threat, Vol. 2, Ed. 1, Published September.