Steve

Merix has experience building boards that utilize heat sinks attached to
the back of the substrate.  In most cases these are power amplification
boards for RF applications.   The heat sinks are typically coins
inserted into a cavity in the board.  The coin and board designs are
integrated so that the device is attached to both.  While your
application is a little different, some of the same principles might be
adaptable to your design.  If you are interested, please contact me
off-line.

        ----------
        From:  Tucker, Steve (KS) [SMTP:[log in to unmask]]
        Sent:  Thursday, August 26, 1999 6:50 AM
        To:  [log in to unmask]
        Subject:  [TN] Thermal bonding to PCB's

        Technetters,

        I serve as the manufacturing rep. for a design team and am faced
with a
        question regarding the best method for heat sinking a processor
on a circuit
        board. Any suggestions or comments would be greatly appreciated.

        DETAILS
        This component is an Intel SB80486DX2SX50, a 208 lead, 20 mil
pitch, SQFP
        package. This component is expected to generate ~1.2W of energy
in our
        application. The PCB is 14 layer, HASL finished, and
approximately 4"X6".
        The area directly under the body of this component is a HASL
finished ground
        plane approximately .9" Square designed to serve as a heat sink
for this
        part. This plane is internally connected to the other PCB
grounds.The
        product is subjected to 20G's vibration at temperatures ranging
from -40C to
        +80C.

        BACKGROUND
        Most members of the team want this component to be heat sinked
from the
        bottom via thermal contact with the PCB ground plane rather than
applying a
        heat sink to the top of the component and relying on air flow to
dissipate
        heat. Various methods have been suggested, none of which I am
thrilled with.
        My concern is that any under component heat sink material I
apply would have
        to allow the component to be placed correctly and allow the part
to seat
        properly during reflow. This part, by the way, spec's a
dimension from the
        bottom of its body to the bottom of leads of .13mm +.12mm
-.08mm. With this
        variation in mind I'm stuck for a good answer.

        My question is: what is the best method to provide a good
thermal bond under
        this component while still allowing machine placement and
without causing
        additional solderability or placement problems.

        Regards, Steve Tucker

##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
the body:
To subscribe:   SUBSCRIBE TECHNET <your full name>
To unsubscribe:   SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Gayatri Sardeshpande at [log in to unmask] or
847-509-9700 ext.5365
##############################################################