DocumentCode :
3257707
Title :
Mechanical shock testing and modeling of PC motherboards
Author :
Pitarresi, James ; Roggeman, Brian ; Chaparala, Satish ; Geng, Phil
Author_Institution :
Dept. of Mech. Eng., State Univ. of New York, Binghamton, NY, USA
Volume :
1
fYear :
2004
fDate :
1-4 June 2004
Firstpage :
1047
Abstract :
Due to a variety of manufacturing, environmental, shipping, and end-use conditions, personal computer (PC) motherboards and other circuit boards may be subjected to potentially damaging mechanical shock loads. As these loads can lead to product failure, an understanding of the response of circuit boards subjected to suddenly applied loads is necessary. A first step in this direction is to develop and validate modeling approaches for the simulation of shock load response on PC motherboards. Since building a detailed model of the motherboard would be difficult due to the wide variation in length scales and localized concentrations of mass/stiffness due to components, two simplified modeling approaches were investigated: global property smearing and simple block modeling. Both of these methods approximate the influence of regions with widely differing stiffness and mass properties resulting from the placement of components, connectors and other items on the circuit board while simultaneously avoiding problems associated with developing large, expensive, detailed models. Both the shock response spectrum (SRS) method and an implicit direct integration (i.e., time-marching) scheme were used to simulate the shock response. In addition to modeling, programmed shock pulse and drop table measurements were carried out on the motherboard to validate and understand the limits of the finite element simulations. The results show that the predicted peak response at a number of locations on the motherboard correlated well with measurements made during the shock loading; however, improvements in the simple models are still required to strengthen their correlation. Surprisingly, the simple global smear approach produced good results with significantly less solution time than the block model. Finally, it was found that the SRS method significantly under predicted the response of the motherboard. This may be due to large displacements induced in the motherboard by the high-g shock loads.
Keywords :
computer equipment testing; dynamic testing; electronics packaging; finite element analysis; impact testing; printed circuit testing; shock wave effects; acceleration response; complex circuit boards; finite element modeling; global property smearing; implicit direct integration scheme; in situ conditions; mechanical shock testing; personal computer motherboards; predicted peak response; product failure; shock response spectrum method; simple block modeling; time-marching; vibration problems; Circuit simulation; Circuit testing; Computer aided manufacturing; Connectors; Electric shock; Finite element methods; Integrated circuit measurements; Microcomputers; Printed circuits; Pulse measurements;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electronic Components and Technology Conference, 2004. Proceedings. 54th
Print_ISBN :
0-7803-8365-6
Type :
conf
DOI :
10.1109/ECTC.2004.1319469
Filename :
1319469
Link To Document :
بازگشت