Dear Technetters,

Werner Engelmaier wrote in a reply to Christer Marklund concerning the 
reliability of PCMCIA cards: "...Stay away from TSOPs with Alloy 42 lead frames 
- field failures have been reported for SIMMs and DIMMs in computer 
environments".

My company is starting to build MCMs on so called PERL-substrates from HP 
(basically a FR4-substrate), a 6 times multilayer with a thickness of 1 mm. We 
are using two TSOPs with each 16 Pins on the smaller sides, one with 
copper-leadframe and one with alloy42-leadframe. 

In an former version of this MCM, which was identically in size and layout, we 
used an alumina ceramic substrate and experienced cracks in the ceramic 
underneath the lands of the soldered copper-leadframe TSOP-pins. The cracks 
appeared after air to air thermal cycling between -40°C and +125°C. To our 
understanding, and following our investigations, this was due to thermal 
mismatch between the leadframe-material and the substrate, because of their 
different Tces. The alloy42-leadframe TSOPs performed well.

When we decided to switch to PERL, the first concern was, that now the 
copper-leadframe TSOPs should perform well because of the good thermal 
matching, but the alloy42-leadframe TSOPs should not. The manufacturer of the 
alloy42-leadframe TSOP said, that we should not be concerned about this, 
because industrial praxis does not show problems. We did thermal cycling (up to 
now 50 cycles between -40 and +125°C, ongoing) and did not experience any 
failures like lift-off of the pins, copper delamination or visible cracks in 
the package.

My questions now are:

-	What kind have the failures been which Mr. Engelmaier reported?
-	Did they stem from thermal mismatch?
-	Any further experiences out there with failures of alloy42-leadframe 
TSOPs on FR4?

Any comments and further informations are welcome.

Ulrich Korndörfer


***************************************************************************
* 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.        *
***************************************************************************