Guy Richardson

Accuracy of simplified printed circuit board finite element models

Electronic components are prone to failure due to shock or vibration loads. To predict when this failure may occur it is necessary to calculate the vibration response of the printed circuit board (PCB); this is most usually achieved through use of simplified finite element (FE) models. The accuracy of these FE models will be mainly dependant on various sources of error, including: manufacturing variability, which will cause supposedly identical printed circuit boards to behave differently (including variability in materials and assembly, as well as dimensional tolerances); inaccuracy in the model input parameters, which is caused by either the modelling assumptions used or poor measurement technique; and errors in the solution process (e.g. linear solutions in non-linear situations). This paper investigates experimentally the contribution of these effects, this is achieved by first looking at measurement of input parameters and to what accuracy a PCB can reasonably be modelled, and then secondly measuring the extent of manufacturing and assembly induced variability. When these contributions have been defined, it will be possible to assess the confidence in any FE PCB model.

Reliability analysis of electronic equipment subjected to shock and vibration – A review

Increasingly modern electronic equipment is expected to provide more functionality whilst still being able to withstand shock and vibration loads. The process of reliability prediction has been hampered by the complicated response and failure characteristics of electronic equipment, with the currently available methods being a compromise between accuracy and cost. A process to quickly and confidently predict the reliability of proposed electronic equipment to dynamic loads would greatly benefit industry. This paper will illustrate the difficulties in predicting electronic equipment reliability, showing why progress has been slow, in addition to the difficulty in making a model that works across a broad range of equipment configurations. The four classes of reliability prediction methods (Handbook, Test data, Field data and Physics of Failure) will be contrasted before addressing the individual methods. It is pertinent to note that although the majority of failures in electronic equipment are due to thermal issues, this review focuses on shock and vibration induced failures. 1. Terminology

Static Performance of Hot Bonded and Cold Bonded Inserts in Honeycomb Panels

An investigation on the structural performance of inserts within honeycomb sandwich panels is presented. The investigation considers metallic inserts in all aluminum sandwich panels and emphasis is placed on the structural performance difference between hot bonded and cold bonded inserts. The former are introduced during panel manufacture while the latter are potted into existing panels. The investigation focuses on the static performance of the two insert systems subject to loads in the normal direction to the facing plane. The experimental part of the work presented involved carrying out pullout tests on hot bonded and cold bonded reference samples by loading them at a centrally located insert. The experimental results were compared with results from an analytical model and results from a finite element model. Contrary to what was expected it was found from the experiments that the cold bonded inserts outperformed the hot bonded inserts in terms of load carrying capability. From the finite element study it was found that this was mainly due to the difference in stiffness of the different filler materials used in the two insert systems.

Static and Fatigue Behaviour of Hexagonal Honeycomb Cores under In-plane Shear Loads

Due to their high specific strength and high specific stiffness properties the use of honeycomb panels is particularly attractive in spacecraft structures. However, the harsh environment produced during the launch of a satellite can subject the honeycomb cores of these sandwich structures to severe quasi-static and dynamic loads, potentially leading to static or early fatigue failures. Knowledge of the static and fatigue behavior of these honeycomb cores is thus a key requirement when considering their use in spacecraft structural applications. This paper presents the findings of an experimental test campaign carried out to investigate the static and fatigue behaviors of aluminum hexagonal honeycomb cores subject to in-plane shear loads. The investigation involved carrying out both static and fatigue tests using the single block shear test method. These results are also discussed in relation to the observed damage and failure modes which have been reported for the statically tested specimens and for the fatigue tested specimens at various stages of fatigue life. As well as conducting tests for the more conventional principal cell orientations (L and W), results are also presented for tests carried out at intermediate orientations to investigate the variation of core shear strength with loading orientation. The results are further investigated using explicit non-linear finite element analysis to model the buckling failure mechanisms of the tested cores.

Board-Level Vibration Failure Criteria for Printed Circuit Assemblies: An Experimental Approach

The assessment of the capability of electronic equipment, to withstand harsh vibration environments, is an issue faced in several branches of engineering. Various researchers have studied the vibration response of electronic boards using different parameters, e.g., local board accelerations, bending moments, curvatures, etc., as a simpler alternative to very detailed stress analysis. However, the issue of what parameter best correlates with vibration failures remains open. This paper investigates this specific problem using an experimental approach to assess whether it is possible to correlate failures produced by intense vibrations, with a single macroscopic parameter such as the local board acceleration, curvature, or surface strain. Printed circuit boards populated with a grid of electronic components (20 different types and 32 identical components per type) have been subjected to vibration testing and the results show that there is a very good correlation between the board curvature (and its surface strain) and failures of the electronics. The work also shows that-for the components tested here-local board acceleration cannot be used to predict components failures. Although this research has focused on a particular set of components, these are representative of typical classes of electronic components, and therefore it should be possible to generalize the conclusions to similar hardware.

Development of efficient and cost-effective spacecraft structures based on honeycomb panel assemblies

Due to their high strength to weight ratio and stiffness to weight ratio the use of honeycomb panels is particularly attractive in spacecraft structures. Honeycomb panels are often used in secondary satellite structures such as equipment platforms and solar arrays, but they can also be used as part of the primary structure of a satellite. Indeed honeycomb panel assemblies can be, and are, used to produce efficient and cost-effective primary structures. These types of structures have been used for some time for numerous satellites; however, their development still poses some challenges ranging from the structural performance of the panels themselves to the problem of connecting them to other panels or structural elements. These challenges are faced each time a new satellite is being developed adding cost to the design process. Furthermore, often due to strict timescales in the development process, some of the uncertainties which naturally arise from these challenges cannot always be completely addressed. To compensate for this, conservative design approaches often need to be taken with the ultimate effect of lowering the efficiency of the structures final design. To meet these challenges and provide a better knowledge base for future satellite development projects a number of research activities have been, and are still, under way at the University of Southampton. The aim of this paper is to describe these research activities and present the key results. 1 2

Electromagnetic damper design using a multiphysics approach

Electromagnetic dampers (EMD) have been widely studied and designed in the control of vibrating structures. Yet, their use for space applications has been almost negligible, due mainly to their high ratio of system mass over damping force produced. The development of shunted circuits, and in particular negative impedances, has allowed higher currents to flow in the device, thus obtaining an increased damping performance. However, the need for a thermal analysis has become crucial in order to evaluate the power and temperature limits of EMDs, and hence allow a more efficient optimization of the whole device. This paper presents a multiphysics Finite Element Analysis (FEA) of an EMD in which the thermal domain is integrated with the electromagnetic and mechanical domains. The influence of the temperature on the device parameters and overall performance in the operative temperature and frequency range of a space mission is shown. It follows a design optimization of an electromagnetic shunted damper for 5-kg SDOF to obtain a second-order filter. In particular, the analytical results are compared with the typical transfer function of a viscoelastic material. This paper demonstrates the feasibility to achieve the same slope of -40 dB/dec while considerably decreasing the magnitude of the characteristic resonance peak of viscoelastic materials.

A general methodology for analysis of structure-borne micro-vibration

Driven by the increasingly stringent stability requirement of some modern payloads (e.g. the new generations of optical instruments) the issue of accurate spacecraft micro-vibration modelling has grown of importance. This article focuses on the dynamic coupling between a source of micro-vibration (e.g. reaction wheel) and a structure, taking into account the uncertainties related to both parts. In this context, an alternative to the Monte Carlo Simulation for complex structures has been developed, consisting in sub-structural approach to perturb the natural frequencies of specific subsystem reduced with the Craig-Bampton method. In order to prove the validity of the method and its application to the theory of the coupling, benchmark examples and practical applications will be described.

Experimental Investigation of Static and Fatigue Behaviour of Honeycomb Panels under In-plane Shear Loads

Due to their high specific strength and high specific stiffness properties the use of honeycomb panels is particularly attractive in spacecraft structures. However, the dynamic loads produced during the launch of a satellite can subject the honeycomb cores of these sandwich structures to numerous stress cycles, potentially leading to early fatigue failures. Knowledge of the fatigue behavior of these honeycomb cores is thus a key requirement when considering their use in spacecraft structural applications. This paper presents the findings of an experimental test campaign carried out to investigate the static and fatigue behaviors of aluminum hexagonal honeycomb cores subject to in-plane shear loads. The investigation involved carrying out both static and fatigue tests using the single block shear test method. These tests were conducted for both of the two principal cell orientations (L and W) and from the results S-N fatigue curves for both cell orientations are presented, confronted and discussed. These results are also discussed in relation to the observed damage and failure modes which have been reported for the statically tested specimens and for the fatigue tested specimens at various stages of fatigue life.

Static performance of hot bonded inserts in honeycomb panels

An investigation on the structural performance of inserts within honeycomb sandwich panels is presented. The investigation only considers metallic inserts in all aluminum sandwich panels and emphasis is placed on the structural performance difference between hot bonded and cold bonded inserts. The former are introduced during panel manufacture while the latter are potted into existing panels. The investigation only deals with the static performance of the two insert systems subject to loads in the normal direction to the facing plane, which corresponds to their main mode of operation. The experimental part of the work presented involved carrying out pullout tests on hot bonded and cold bonded reference samples by loading them at a centrally located insert. As expected the hot bonded reference samples outperformed the cold bonded reference samples in terms of load carrying capabilities. An analytical model which allows the prediction of shear stress distribution in a circular sandwich panel normally loaded at a centrally located insert is used in an analytical approach for calculating the load carrying capability of inserts. The results from this analytical approach were found to correlate well with the experimental ones for the hot bonded inserts but not for the cold bonded inserts which actually failed at a significantly lower load than was predicted.