Just one more?
Intuitively, I agree that any material is a better conductor than air. I'm
just curious about the actual numbers. What is the measured delta-T at the
"device" junction for the three scenarios: vias, filled vias (and yes) no
vias. At what expense are the vias being used; 1.)loads on the barrels over
time, 2.) extra fab step for filling, 3.) vias are certainly going to be
bonded to a metal core and therefore not always a great interface, 4.)vias
creating clearance on signal ground and therefore reducing cross-sectional
RTN path and increasing inductance to ground, 5.) and finally for me,
increased design density (the increased density is an issue for me because
typically the hottest devices are the ones with large IO and constant duty
cycle and aren't always near the base plate as power switching devices
usually are). We also use heatpipes to transfer to the cold-plate. We can
only rely on conduction (and a little radiation), hence we have several
patents on heatpipe configuration, cryo-coolers and such. I'm aware of the
notion that prepreg and core materials are barriers. I'm just asking - if
the device "squares" and the metallized "squares" are large enough in total
area (even one layer apart), how much work are the vias actually doing? Is
anyone aware of a controlled comparison (maybe an infra-red snapshot)? I can
possibly request a simple analysis, but because of a history of thermal
vias, I figured that the data is available.

-----Original Message-----
From: Leslie O. Connally [mailto:[log in to unmask]]
Sent: Wednesday, January 06, 1999 9:25 AM
To: [log in to unmask]
Subject: Re: [TN] FAB: Solder Filled Thermal Vias


Hi James and other Technetters,

        I've been following this subject on the forum and Just have to put
in
my two cents. The comments from both Earl Moon and Kelly Schriver are very
pertinent. I understand the Heat pipe scenario well, but have some concerns
about the open via contracting contaminates. We have built a number of
boards
with constrained/thermal cores and usually follow the path mentioned by
Kelly.
In the early '80s, we tried to fill with solder, but could never get the
solder
to stay in place. Two concerns with solder are: 1) during assembly the
solder
may wick away solder from a joint making a poor interconnection, and 2) as
in
the case that we tried early on in the SMT effort, is that during assembly
it
will errupt out of the hole and create all sorts of problems, among which
are
shorts.
        We found that our best approach was to fill the via with a thermally
conductive/electrically insulating epoxy. There are several sources of these
materials available in the industry.

It's not much but there's my two cents worth!
Regards,
Les

################################################################
TechNet E-Mail Forum 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's web site (http://www.ipc.org) "On-Line Services" section
for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or
847-509-9700 ext.312
################################################################

################################################################
TechNet E-Mail Forum 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's web site (http://www.ipc.org) "On-Line Services" section for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################