Issue 15

P. F. Fuchs et ing fixture w m to each ot The clampin shown in Fi

alii, Frattura ed as constructe her. The load g force was a g. 5.

Integrità Struttu d, where the is applied b pplied over

rale, 15 (2011) board is plac y a fin, whic spring plung

64-73; DOI: 10 ed centred o h consists of ers to have a

.3221/IGF-ESIS.

15.07

For (Ø mm sch

the board fi 5 mm) which ), clamping ematic repre

xation a thre have a dist the board at sentation of t

e-point bend ance of 90 m half length. he fixation is

n support ro two rollers ( defined load

llers Ø 5 . A

3

2

1

0

-1

-2 amplitude s, mm

-3

0 0

(b) d camera anal e PCB deflecti is presented. .01 0.02 time

0.03 0.0 t, s

4 0.05

(a) ults of the JE high-speed cam the impact. I

Fig

ure 4 : The res consecutive

ard level drop show the deve ding amplitud

test high-spee lopment of th e oscilloagram

ysis. In a) cho on right after

sen

SD22-B111 bo era pictures n b) the accor

Figur eeds to be c y be applied tem. Robustn s accepted th e test param DT was abo m with a me re chosen. Th lection of -1m e to machine DT (280 Hz) . icture of the

e 5 : A schema onsidered tha in one direc ess and repr at in contrast eters were d ut 5 mm and an deflection e maximum m was defin limits, it wa . To remain final test set

tic representati t the board i tion. A set-u oducibility o to the actual erived from the maximu of 3mm (BL fibre strains ed to ensure s not possib at the desired -up is shown

on of the thre s not clampe p with clamp f the test wou BLDT altern the BLDT m fibre stra CBT Set-Up were 1.7 x 1 that no lift-o le to perform optimum m in Fig. 6. Th

e-point bendin d at the posit s at the sup ld have been ating loads c analysis. The in was about 1) and 3mm 0 -3 and 2.2 x ff of the boa the tests at achine displa e event detec

g fixture used ion of the su port rollers w affected, co ould not be maximum 2 x 10 -3 . Fo with a mean 10 -3 , that is rd from the the oscillati cement cont tor was used

in the board l pport rollers ould have r unteracting p applied. peak-to-peak r the BLCBT deflection o in the range support rolle on frequency rol precision in the same w

evel cyclic ben , implicating esulted in an revious effo oscillation a peak-to-pe f 2.5mm (B of the BLDT rs could occu of the impa the frequenc ay as for th

d test. that the load overconstrai rts. Therefor mplitude in ak amplitude LCBT Set-U . The minim r in the tests cted PCB in y was kept a e BLDT.

It n onl sys wa Th BL 4m we def Du BL Hz A p

can ned e, it the s of p 2) um . the t 25

E X

PERIMENT

AL METHOD

COMPARIS

ON

set T

he BLD statistics used. Th -up another f

T was perfo . For the BL ree boards f ive test repet

rmed with ni CBT, as ind or all design itions were p

ne samples o icated in Boa s and both se lanned to app

f every PCB rd level cyclic t-ups were t ly statistics a

build-up to bend test develo ested to deci ccording to

obtain repre pment chapte de which per the BLDT ev

sentative resu r, two differ forms better aluation.

lts and to ap ent set-ups w . For the cho

ply ere sen

67