Mike

The fatigue of solder joints depends on how much and how fast a solder
joint is deformed. The more deformation you have in a joint the shorter the
lifetime ( good ol' Coffin Manson ), the faster the joint is deformed the
shorter the lifetime.
Now while in a electronic assembly the components, the board and the solder
joint heat up an cool down during switch on / switch off the components and
the board see different temperature/time cycles. This fact and the
differences in the CTE's of the materials involved,induce shear stress in
the solder joints. The higher the stress, the faster the deformation
(solder deforms in creep, where stress determines the speed of deformation
not how much a material is deformed, even at -20°C). Leaded components will
bend their leads. If the leads are made of copper the stress in the solder
joints resulting of this deformation is comparatively small ( low spring
constant of the leads ). Hence the solder will deform slow. This
deformation might be so slow that even at a day cycle, where a equippment
will be on for 8h and off for 16h, not all of the strain that would be
theoreticaly possible due to the differences in the CTE will be induced
into the solder joint. Leadless components like BGA's, CCGA's etc. habe now
leads to lower the stress in the solder joints. Therefore the solder will
deform as fast as the temperature exchange rate is.BGA's are a bit special,
becaus the side, facing to the board is usualy made of a material with a
CTE close to the one of the board. But nevertheless, in the area of the
chip the expansion properties of the substrate of the BGA are influenced,
resulting in the fact that the solder joints below the chip do fail first.
Fortunately in reality usualy low temperature exchange rates occur. This
eases the problem. In applications with high temperature exchange rates (
higher than 4°C/min at temperatures below -20°C, or 40°C/ min. and more at
temp. below 20°C ) solder joints of leaded components have theoretically a
longer lifetime than the ones on unleaded components. A test we where
running with BGA's ( pitch 0.5 mm ) and QFP's of equal size and with a chip
of the same size in it, showed in a passive cycle ( -20°C/100°C, 10°C /
min, 15 min dwell time ) the max. cracklengh in the solder joints of the
QFP and of the BGA was approx.the same. ( BGA's slightly longer cracks than
QFP's)

Best regards

Guenter Grossmann




***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************