Jason R. Foley

EQUATION OF STATE OF ALUMINUM-IRON OXIDE - EPOXY COMPOSITE

We report on the measurements of the shock equation of state (Hugoniot) of an Al∕Fe2O3/epoxy composite, prepared by epoxy cast curing of powder mixtures. Explosive loading, with Baratol, trinitrotoluene (TNT), and Octol, was used for performing experiments at higher pressures, in which case shock velocities were measured in the samples and aluminum, copper, or polymethyl methacrylate (PMMA) donor material, using piezoelectric pins. The explosive loading of the metal donors (aluminum and copper) will be discussed. Gas gun experiments provide complementary lower pressure data in which piezoelectric polyvinylidene fluoride (PVDF) stress gauges were used to measure the input and propagated stress wave profiles in the sample and the corresponding shock propagation velocity. The results of the Hugoniot equation of state are compared with mesoscale finite-element simulations, which show good agreement.

A novel Micro-CT data based Finite Element Modeling technique to study reliability of densely packed fuze assemblies

Densely packed electrical assemblies like fuze, contain large number of components, potted in protective adhesives. The number of components, varying material types, irregular geometry of the components and the geometric details of the assembly makes conventional CAD modeling, meshing and Finite Element(FE) modeling of these large assemblies extremely time consuming, often, to the extent of being impractical. CAD geometries compatible with modern Finite Element (FE) platforms may not be available for several legacy systems. Furthermore, conventional CAD modeling may not account for the real geometry realized after the manufacturing process and this can often affect the fidelity of the FE model. There is no method for capturing the actual assembly geometry and its embedded components. Assessment of survivability of fuzes requires assessment of stresses and strains under operational loads. Previously, researchers have studied the reliability of key components in a fuze device subjected to high temperature and high g mechanical shocks [1]. Researchers have measured redundancy and reliability of fuze electronics using failure rates and mean time to failure as per MIL-HDBK-217F standard [2]. There is little to no literature on FE modeling of a comprehensive fuze assembly. In this paper, a methodology for the creation of an FE model based on Micro-CT (Computed Tomography) data is presented. The method has been applied to an actual fuze subjected to mechanical shock. This method involves usage of advanced 3D imaging, image segmentation, image filtering and meshing techniques to directly convert CT scanned electrical assemblies into a FE mesh. This method successfully bypasses the time consuming CAD modeling step of conventional FE modeling. The as-is geometry of each component, positioned accurately in a 3D space, as per the original assembly, has been realized in this process by usage of micro-CT scanning technique. The submicron scale tolerances of the CT scanned data ensure true representation of the fuze assembly, in this case. The FE model thus realized, allows for measurement of all the field variables, anywhere over its meshed domain. Stress and strain histories have been extracted for embedded components of the fuze assembly using explicit finite element models.

Real-Time State Detection in Highly Dynamic Systems

This article investigates the feasibility of real-time state detection on a microsecond timescale for use in highly dynamic systems and presents an experimental setup of a high-rate dynamic system coupled with real-time measurement architecture with the goal of detecting changes in the interfacial state of the system. The feasibility of microsecond state detection is assessed through a preliminary timing study. The experimental setup consists of two colliding aluminum bars and includes the option of changing the bar’s boundary conditions and the interface material between the bars. A piezoelectric transducer will be used for detecting changes in dynamic interfacial state by employing electromechanical impedance monitoring and the measurement data from this will be acquired and processed at high speeds using deterministic real-time tools and methodology. Damage detection algorithms from the structural health monitoring community will be used for rapid detection of changes in state. The eventual goal of this work is to adapt currently used methods or to develop entirely new high speed state detection algorithms to be implemented on the real-time system for state detection. This technology has the potential to be used in many applications, including the aerospace, civil, and energy industries among others.

Survivability assessment of electronics subjected to mechanical shocks up to 25,000g

Mechanical drop testing and reliability of BGA electronics in consumer markets is conducted using the JEDEC JESD22-B111 standard. The JEDEC form factor PCB may not be suited for use in part survivability assessment at very high-g levels up to 25,000g. Product board geometries may need to be considered for survivability assessment in fuzing applications. Commercial-off-the-Shelf (COTS) parts like fine pitch BGAs of 0.4mm and 0.5mm pitch are increasingly being used for military and defense applications. Use of fine pitch Ball Grid Array (BGA) technology is preferred over the other packaging technologies due to relative ease of procurement relative to leaded parts, high level of integration with small foot print. Commercial parts may need supplemental restraints in form of underfills, and potting compounds to meet the reliability requirement of extreme-acceleration applications such as fuzing. Survivability of fine pitch semiconductor packages under high-g mechanical shock in the range of 10,000g-25,000g is unknown. In addition, the efficacy of the traditional supplemental restraint mechanisms such as underfills and potting in reducing the risk of interconnect failure under high-g mechanical shock, is not available. In this study, instead of using a JEDEC form-factor board, a product form-factor circular board with annular ring typical of end product projectile applications under harsh high-g operating environments of 25,000g was studied. Five different BGA packages of interconnects ranging from 84-360 I/O have been studied. Three configurations of the test board have been studied including non-underfilled, underfilled and potted assemblies. Lord Thermoset ME-531 has been used to underfill the packages in this test study. Two categories of potting compounds have been used including Armstrong A12 is a low modulus material and Henkel STYCAST 2850FT is a high modulus material intended for shock applications. Data on the survivability of BGAs using supplemental restraint mechanisms like underfill reinforcing the packages and potting the PCB with epoxies is presented. Modal analysis of the unreinforced and supplemental restraint configurations has been done to understand the frequency response of the configurations.

Prognostication of solder-joint reliability of 0.4mm and 0.5mm pitch bgas subjected to mechanical shocks up to 10,000G

Due to the reduced size and geometry constraints imposed on electronics in various applications there has been tremendous need for use of very fine pitch surface mount electronics. Fine pitch BGAs of 0.4mm and 0.5mm pitch are finding applications in military and defense applications. Fine pitch BGA electronics in aerospace applications they may be subjected to high-g levels in the neighborhood of 10,000g of mechanical shock during normal operation. Survivability and design envelope of fine pitch semiconductor packages under high-g mechanical shock is unknown. In addition, the efficacy of the traditional supplemental restraint mechanisms such as underfills in mitigating the risk of interconnect failure under 10,000g mechanical shock, is not available. A circular board with an annular ring typical of projectile applications has been designed with fine pitch daisy chained packages. Packages studied have package interconnects in the range of 84-360 I/D. Two configurations of the test board have been studied including non-underfilled, and underfilled assemblies. Full-field strain on the board assembly has been measured and the strain histories at the corner of the component locations extracted. The change in the resistance of the second-level interconnects has been monitored during the shock event using high speed data acquisition system. Resistance spectroscopy in conjunction with Kalman Filter has been used to identify the onset of failure and prognosticate remaining useful life.

Constitutive Response of Electronics Materials

Electronics in mission- or safety-critical systems are expected to survive a wide range of harsh environments including thermal cycling, thermal ageing, vibration, shock, and combinations of the aforementioned stresses. The materials used in these electronic systems are diverse and frequently change as the electronics industry rapidly innovates. These materials are dual use, fulfilling both electrical and mechanical functions. Of particular interest are electronic materials classes such as polymers (e.g., encapsulants/potting and packaging), composites (e.g., hard potting and printed circuit boards), and interconnect materials (e.g., solder). Thus, predicting the operational response of electronics systems in harsh environments requires understanding of the materials constitutive response to the environmental characteristics for all the relevant materials. The paper estimates the rate-, temperature-, and pressure-dependent constitutive response of representative electronic materials. Experimental response of circuit boards, potting materials, and solder interconnects are measured in low and intermediate strain rate dynamic tests. Traditional mechanical sensors (e.g. strain gages and accelerometers) are complemented by non-contact techniques (e.g., laser velocimetery, high speed digital image correlation) to obtain high fidelity experimental data on material response. Estimates of the corresponding constitutive parameters are calculated, and observed features of the dynamic response are discussed.

Dynamics of Interfaces with Static Initial Loading

Accurately modeling the dynamic response of structural interfaces under high rate loading conditions is challenging due to lack of focused studies and validation data. In order to support the development and validation of accurate physics-based models of these interface dynamics, simulations and experiments using combined static torsional loads with dynamic compressive loading are performed. A ballistic impact generates dynamic compressive stress waves that propagate across the threaded interface of two coupled metallic bars with a known static torque. Strong phenomenological evidence of the release and conversion of static torsional energy due to the applied dynamic loads is seen in time-frequency analysis. The frequency band structure of the dynamic waves are also observed to vary under certain conditions. Applied torque is shown to relax almost completely during an experiment, with significant rotation of the transmission bar relative to the incident bar. The mechanism is believed to be release of the static torque and the generation of torsional waves.

Spectral Domain Force Identification of Impulsive Loading in Beam Structures

In this paper, an identification method is presented for calculating impulsive loads from the propagating mechanical wave which are produced in beam-like structures. This method uses a spectral finite element method (SFEM) model of a segment of the structure to calculate force information from the measured response. The SFEM model is prepared from the Euler-Bernoulli beam equation in the frequency domain. The method is studied using simulated response data and then applied to data collected from an experimental system. Excellent performance is observed for nominal conditions and a parametric study is performed to determine how different factors affect accuracy. Factors studied include structure size, loading location, and loading duration. When limited to only acceleration data, the use of finite differencing methods to obtain the required slope response information is determined to provide the most significant source of error in the identified force information.

Modal Testing of Complex Hardened Structures

A new testing method is being developed by the Air Force Research Lab to excite a desired multi-dimensional response in a structure using tuned resonances. The structure consists of a Aluminum plate and a perpendicular shelf. The test article is excited through the use of either an impact hammer or pyrotechnics (e.g., a pyrotechnic plunger) which in turn inputs a specific frequency profile. The dynamic response of the plate and shelf spans the entire spectrum from low (10 Hz) to high (10 kHz) frequency as well different peak amplitudes (i.e., accelerations). They are captured by a variety of instrumentation methods including modal accelerometers, laser vibrometers, and a digital image correlation system. The location of the shelf as well as its material and stiffness properties is modified to reproduce the design objective (frequency response functions with the desired amplitude and phase spectrum). These modifications are achieved through interpretation of modal data.

Microsecond structural health monitoring in impact loaded structures

Early results and status of a research effort to frame the possibility in compressing the time scale of structural health monitoring to the impulsive transient domain are presented. Output only modal methods using a frequency domain decomposition technique are used to extract the operational modes of a plate subject to impulsive loading. A strain energy method for plates is the used to detect the damage on the plate. The method detects damage, but the location of damages is not very precise. The development of an extremely short duration, transient structural health monitoring algorithm will be discussed. Challenges in studying this new domain of health monitoring will also be highlighted.