Subject: | |
From: | |
Reply To: | TechNet E-Mail Forum. |
Date: | Sun, 10 Feb 2002 13:03:23 EST |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
Hi Earl, the MoonMan,
First, I have problems accessing Steve's website--please send me the pics
directly.
Second,
>Thermally asymetric? Does this mean in the Z axis as a primary function of
>the Al heat sink and the Silicon chip placed on a BT substrate, or?
A: What it means is that the various layers [overmold, die, die attach,
interposer, package substrate] are made of different materials all having
different CTEs and modulii of elasticity. By proper variation of the
thicknesses of these materials one can achieve a thermally balanced [or at
least nearly balanced] design where the various layers will thermally
expand/contract without significantly warping [of course there are shear
forces between these layers] the package--with a metal heat spreader this
becomes essentially impossible with the heat spreader on the outside.
Third, the die attach on thin BGAs is typically some polyimide-based adhesive.
Fourth,
>Sure wish Werner well with grandkiddies, but I need some more help. Maybe
>Dave Hillman has some more educated ideas since I sent him my first report
>from fellow Rockwell types. Also, I still await Joe Felstad's comments about
>the chip attachment, its method, and the lack of an apparent
>"compliant/flex" layer as he helped develop at Tessera.
A: The "compliant/flex" interposer layer of the Tessera patents is needed
most for CSPs and thin BGAs with a thin package substrate not including woven
glass-fiber reinforcement and prevents the large die/PCB expansion mismatch
from loading the SJs [the failures would occur at the die corners]--the large
die/PCB expansion mismatch is accommodated in the "compliant/flex" interposer
layer(s) [in some designs to the extent that the copper traces broke]. The
BLR of the larger BGA packages depends on the much thicker woven glass-fiber
reinforced BT to be stiff enough to shield the solder joints from the much
larger die/PCB CTE-mismatch. This works particularly well for packages much
larger than the die [it is the package corner SJs that fail first] and those
with SJ arrays depopulated under the die.
I hope this helps--you do need to get out of your cave more often--there is a
whole world out here
Werner Engelmaier
---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail Archives
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
---------------------------------------------------------------------------------
|
|
|