ScienceDirect. Vibration Response Prediction of the Printed Circuit Boards using Experimentally Validated Finite Element Model

Transcription

1 Available online at ScienceDirect Procedia Engineering 144 (2016 ) th International Conference on Vibration Problems, ICOVP 2015 Vibration Response Prediction of the Printed Circuit Boards using Experimentally Validated Finite Element Model Somashekar V.N a, Harikrishnan S b, PSM Aejaz Ahmed c, Kamesh D d a,d SRG, ISRO Satellite Centre, Bengaluru , India b,c SaiVidya Institute of Technology, Bengaluru , India Abstract A spacecraft consists of a number of mechanical and electronic sub-assemblies. An electronic package is consists of printed circuit boards placed in mechanical housings which are stacked together. Electronic components are mounted on the printed circuit board (PCB). A spacecraft experiences various types of mechanical loads during its launch such as vibration, acoustic and shock loads. The understanding of dynamic behaviour provides valuable insight to a design engineer, to improve the mechanical design and product reliability when used in harsh vibration condition. Generally electronic package design for vibration loads is verified by conducting tests on actual hardware. This paper addresses on using basic FEA tool to accurately investigate the dynamic characteristics of the PCB and avoid costly testing methods which require hardware. Here the normal modes & frequency response functions (FRF) of PCB are determined and validated using vibration test on PCB. The validated model is used to predict vibration response for random vibration input. It is shown here how the responses are accurately predicted for random vibration input for a design parameter variation of PCB. The results are also validated using vibration test on PCB The Authors. Published by Elsevier by Elsevier Ltd. Ltd. This is an open access article under the CC BY-NC-ND license ( Peer-review under responsibility of the organizing committee of ICOVP Peer-review under responsibility of the organizing committee of ICOVP 2015 Keywords:Printed Circuit Board (PCB); Spacecraft;Vibration Response. * Corresponding author. Tel.: / address: krishnan1993@gmail.com, aejazahmed93@gmail.com The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( Peer-review under responsibility of the organizing committee of ICOVP 2015 doi: /j.proeng

2 V.N. Somashekar et al. / Procedia Engineering 144 ( 2016 ) Introduction A spacecraft experiences various types of dynamic loads during its launch such as vibration, acoustic and shock loads. The electronic packages are designed to withstand the launch vibration environment. The mechanical design of electronicpackages is verified by subjecting to vibration testing. The electronic component failures have been observed whenever there is change in its location on the PCB due to electrical design requirements. The basic failure modes of components mounted on PCB due to random vibration environment are the results of the following conditions: high acceleration levels,high stress levels,large displacement amplitudes andelectrical signals out of tolerance [1]. The design failures during test and evaluation process have resulted in delay in the project schedule. An effective methodology to verify the design other than testing has to be evolved and finite element (FE) analysis has been an effective tool in predicting results. The FE analysis of electronic packages has a lot difficulties due to many low mass individual elements present in it. Electronic systems are often simulated using simple masses, springs and dampers to estimate the dynamic characteristics of the system. Simple one and two degree of freedom systems are used to approximate the electronic systems. More complicated finite element models of electronic systems are created to study the dynamic characteristics of the system and to estimate the fatigue life of critical components mounted on the PCB. Finite element models can be either simplified or detailed. Detailed finite element models are built by modeling the PCB and the components. However, this approach is rarely used as it is time consuming and expensive. Instead, simplified models of PCB are created where the components geometry is neglected. The component effects are included by increasing the Young's modulus and density of the PCB FE model, so it effectively behaves as if components were present. The simple geometry of the board is modeled and meshed using 2-D finite elements (i.e. by using flat shell elements). Sensitivity analysis of PCB finite element models was carried out by Amy et al. [2]. They determined the factors of safety by using different simplification methods of modeling the PCB. Printed circuit boards are multi-layered structures with complex material properties that make the simulation of their dynamic behavior complicated and hence to simplify this, a number of parameters affecting the PCB s dynamics have been studied through experimentation. Dynamic responses such as accelerations, strains are to be closely monitored and controlled to achieve test results which are consistent and predict the structural behavior in reality [3]. Printed circuit boards with multiple layers are suitable candidates preferable for high speed and high density applications. In order to design electronic packaging systems for safety and standards criterion, the accurate modeling of components influence on the multi-layered PCB becomes critical and complex due to various reasons. Numerical model using finite element technique was developed to simulate the mechanical behavior of multilayered PCB and physical tests were conducted to validate the impact performance [4]. In this paper, vibration analysis of a typical PCB used for space applications is carried out. The normal modes & frequency response functions (FRF) of bare PCB (without components) are determined and validated using vibration test on PCB. The validated model is used to predict vibration response for random vibration input. It is shown here how the responses are accurately predicted for random vibration input for a design parameter change of PCB. The results are also validated using vibration test on PCB. 2. Vibration Response Prediction of PCB using FEM In this study, a six layer PCB used for space applications is considered. The PCB is modelled as isotropic plate with equivalent material properties such as Young s modulus, Poisson s ratio and mass density. FEA simulations of PCB dynamics are made using MSC.PATRAN as pre-processor and MSC.NASTRAN as solver for bare PCB and PCBwith components. Finite Element model consists of 3364 quadrilateral shell elements. Young s modulus of elasticity for PCB is obtained as 15 GPa by conducting a three point bending test for three different samples of PCB. The PCB properties are given in Table (1). The FE model of bare PCB is analysed and validated for two different boundary conditions i.e., i) PCB fixed/clamped at nine mounting locations shown in Figure (1) and ii) PCB with fixed at eight mounting locations shown in Figure (2).

3 578 V.N. Somashekar et al. / Procedia Engineering 144 ( 2016 ) Fig 1. FE model of PCB fixed at 9 mounting locations Fig 2. FE model of PCB fixed at 8 mounting locations Further the analysis is carried out to predict response for random vibration input which is also validated with test results. With these iterations it should be possible to predict the vibration response accurately for any change in the PCB parameters. To meet the requirement of an ongoing project, the existing PCB thickness has to be reduced. The vibration analysis is carried out to predict the responses for the modified design. However, the response prediction by analysis is also validated by conducting test on modified PCB. Fixed/clamped boundary conditions are applied (PCB mounting locations) by arresting six degrees of freedom for the nodes on the boundary of holes in PCB. Table 1. Details of PCB Parameter Value PCB size mm Density of Bare PCB 1985 Kg/m 3 Young s modulus of Bare PCB 15 GPa Poisson s ratio 0.12 Boundary Condition Fixed/clamped 2.1 Finite Element Model Normal mode analyses were carried out to extract first few fundamental natural frequencies for PCB with nine mounting location and eight mounting location. The calculated first few natural frequencies and corresponding test results are given in Table (3) and (4). Mode shapes corresponding first two natural frequencies for both cases are given in Figures (3) to (6). Fig 3. Mode shape of PCB with 9 mounting locations Fig 4. Mode shape of PCB with 9 mounting locations

4 V.N. Somashekar et al. / Procedia Engineering 144 ( 2016 ) Fig 5.Mode shape of PCB with 8 mounting locations for first frequency Fig 6. Mode shape of PCB with 8 mounting locations for second frequency The same FE model was used to obtain frequency response function (FRF) by subjecting it to base excitation of 1g from 5 Hz to 2000 Hz. In this study, a structural damping coefficient of 2% is incorporated in the model. The FRF at a specific location of the PCB are plotted in Figures (7) and (8). These values are then compared with test data obtained from vibration tests. The same FE model (9 mounting location, 2.1 mm thick PCB) was used to obtain vibration response subjected to random vibration input given in Table (2). The random vibration response plot for 2.1mm PCB is given in Figure (9) & (10). In actual case the thickness of the PCB was to be reduced to 1.6 mm and study the response change. The vibration response for the same random vibration input for 1.6mm PCB is given in Figure (11) & (12). Fig 7. Analysis FRF for PCB with 9 mounting location Fig 8. Analysis FRF for PCB with 8 mounting location Table 2. Random vibration specification Frequency (Hz) PSD (g 2 /Hz) Overall grms db/oct db/oct

5 580 V.N. Somashekar et al. / Procedia Engineering 144 ( 2016 ) Fig9. Analysis random vibration response for 2.1 mm PCB location Q3L1 with 9 mounting location Fig 10. Analysis random vibration response for 2.1 mm PCB location Q3L7 with 9 mounting location Fig 11. Analysis random vibration response for 1.6 mm PCB location Q3L1 with 9 mounting location Fig 12. Analysis random vibration response for 1.6 mm PCB location Q3L7 with 9 mounting location 2.2 Experimental Test Setup The vibration test was conducted by mounting the 2.1 mm PCB with screws at nine and eight locations on vibration table as shown in Figure (13) and (14) respectively. Figure (15) shows 1.6 mm PCB mounted on vibration table. Accelerometers are mounted at various locations of the PCB to measure the responses. The vibration test was carried out in the vibration test facility consisting of electro-dynamic shaker, control system, signal conditioners and data acquisition system. The frequency response function (FRF) is obtained by conducting a sine sweep test. In sine sweep test, the input acceleration is given to the test specimen using electro-dynamic shaker and the output acceleration at various desired locations of the test specimen is measured using accelerometer. The ratio of output to input acceleration gives the FRF at that location. The experimental frequency response plots for PCB with screws at nine and eight mounting locations are shown in Figures (16) and (17) respectively. Fig mm PCB with 9 mounting location on vibration table Fig mm PCB with 8 mounting location on vibration table

6 V.N. Somashekar et al. / Procedia Engineering 144 ( 2016 ) Fig mm PCB with 9 mounting location on vibration table The vibration test was carried out to obtain the vibration responses for the random vibration input and is given in Table (5) & (6). The test was carried out for both 2.1 mm and 1.6 mm thickness PCB. The random vibration response obtained from test is given in Figures (18) to (21). Fig 16. Test FRF for PCB with 9 mounting location Fig 17. Test FRF for PCB with 8 mounting location

7 582 V.N. Somashekar et al. / Procedia Engineering 144 ( 2016 ) Fig 18. Test random vibration response for 2.1 mm PCB location Q3L1 with 9 mounting location Fig 19. Test random vibration response for 2.1 mm PCB location Q3L7 with 9 mounting location Fig 20. Test random vibration response for 1.6 mm PCB location Q3L1 with 9 mounting location Fig 21. Test random vibration response for 1.6 mm PCB location Q3L7 with 9 mounting location

8 V.N. Somashekar et al. / Procedia Engineering 144 ( 2016 ) Results and Discussion In this section, the FEM and experimental test results are compared. Analysis and test results for first few fundamental frequencies of the PCB at 9 mounting locations are compared in Table (3) and for PCB at 8 mounting locations are compared in Table (4). The analysis and test results for both the cases are matching well. The random vibration response for 2.1 mm thickness PCB at 9 mounting locations for both analysis and test are compared in Table (5). Table (6) shows the same comparison for 1.6 mm thickness PCB. Here the random vibration response (overall grms) has reduced for 1.6 mm PCB versus 2.1 mm PCB. This is very well evident from analysis and test results. Table 3. Comparison of fundamental frequencies for PCB FEM Fundamental Test Results Percentage Results Frequency (Hz) Difference (Hz) Table 4. Comparison of fundamental frequencies for PCB FEM Fundamental Test Results Percentage Results Frequency (Hz) Difference (Hz) Table 5. Comparison of random vibration responses for 2.1 mm thick PCB with 9 mounting location Location Analysis Results, g RMS Test Results, g RMS Percentage Difference Table 6. Comparison of random vibration responses for 1.6 mm thick PCB with 9 mounting location Location Analysis Results, g RMS Test Results, g RMS Percentage Difference Q3L Q3L Q3L Q3L Q3L Q3L Q3L Q3L Q3L Q3L Conclusions Vibration analysis is carried out using FEM to obtain normal modes & frequency response functions (FRF) of PCB and validated using experimental test on PCB. The validated model is used to predict vibration response for random vibration input. It is shown from analysis that the responses predicted for random vibration input for a design parameter change (thickness reduction) in the PCB are reducing. These results are also validated using vibration test on PCB. The paper presents using FEM instead of test and evaluation method to predict vibration response accurately. The work can be extended for simulating the components on the PCB. References [1] Steinberg, D. S. Vibration Analysis for Electronic Equipment. 3rd ed. New York: John Wiley & Sons Inc.; [2] Amy, R. A., Aglietti, G. S., Richardson, G. Sensitivity analysis of simplified Printed Circuit Board finite element models. Microelectronics Reliability, 2009; Vol. 49, pp [3] Jing-en Luan, Tong Yan Tee, Eric Pek, CheeTeck Lim, ZheoweiZhong. Dynamic responses and solder joint reliability under board level drop test. Microelectronics Reliability, 2007; Vol. 47, pp [4] Yuqi Wang, K.H. Low, H.L.J. Pang, K.H. Hoon, F.X. Che, Y.S. Yong. Modeling and simulation for a drop-impact analysis of multi-layered printed circuit boards. Microelectronics Reliability, 2006; Vol. 46, pp [5] Jayaraman S, Manish Trikha, Somashekar V N, Kamesh D. Vibration Analysis of Printed Circuit Boards. Interbational Journal of Applied Engineering Research, ISSN Vol.10 No.24 (2015)