RoHS nonsense, was: Desoldering Chips, was Re: Can someone,

[Chuck Guzis]

On 10 Apr 2010 at 17:57, William Donzelli wrote:

> The little solder ball blob (the B in BGA) is there specifically to
> act as a cantilever to absorb stresses. IBM figured this out when they
> pioneered the technology in the 1960s. Without the tiny amount of give
> that the ball gives, failures would be so frequent that the devices
> would be useless, especially in today's world of relatively high
> power, small outline devices that are constantly going to sleep and
> waking up.

Well, yes and no.  The web  (and IEEE's transactions on semicondutor 
packaging) is littered with papers comparing eutectic Sn-Pb solder 
with RoHS lead free and the results seem to be in favor of leaded 
solder.  In particular, Pb-free solder seems to exhibit some 
brittleness that Pb solders do not.  For example, here's a paper 
comparing shock (drop) tests:

http://www.europeanleadfree.net/SiTE/UPLOAD/DOCUMENT/Projects/Drop_Tes
t_Failure_Analysis_of_BgA_Packages.pdf

It's not as if we can compare 70's era IBM BGAs with today's 
material, either.  Today's BGAs tend to be larger and dissipate more 
power.   

When done badly, as in the Xbox "Red Ring of Death", failure results. 
 When done well with careful design, RoHS BGA work is probably quite 
durable.

--Chuck