Graham,
I don't consider myself an expert on the subject of underfilling BGA's, but
I have done quite a few flip chips.
First - I sure hope you do not plan on re-working these things later!!
If properly set-up, and maintained (with a suitable underfill material), you
should be readily able to obtain a totally void free underfill. Yes, it is
a big device! But if you take the time and set up the dispense pattern
properly, you should do fine. You may see some shadowing of the (underfill)
filler as it flows around the bumps. You will see some haloing around the
bumps, if you used a fluxing agent during reflow (resulting in residues on
the surfaces of the bumps).
I am assuming that you are using automatic dispensing equipment of some
sort. While you CAN do it by hand, it becomes more repeatable if you let
the machine do the work.
When setting up for a large device (my scale for LARGE is a flip chip approx
1/2" per side with 8-12 mil pitch), I would "do the L pattern", wait just a
smidgen for the 'wavefront' to form an arc under the chip, then increase the
length of the last previous leg of the L, and retrace the L, again longer on
the former first leg (all of this MUST be done before the
"reserviour/fillet" of underfill on the outside of the chip is depleted!!
If you ever let the 'surplus' of underfill 'run out', you effectively
introduce a bubble under the chip the next time you pass the dispense needle
by the site. All of the timing details are easier to work out on automatic
dispensing equipment - although you CAN do it by eye too ..... but would
rather not, if you know what I mean).
Anyway, if you continue to trace and re-trace the "L" in 3-6 ever-enlarging
patterns, you should be able to create a nice, void free underfill.
If the terminal points of the L pattern give too much underfill, you will
see a dogbone effect forming (you want the vertex to fill first, not the
ends of the L).
A lot of times I would use a wafer saw to slice standard microscope slides
up into the size of the flip chip, create some type of 'spacers' (a ball
bond in the four corners of the sites on the substrate, for example) and
using just a wee bit of super-glue, etc., place my transparent 'flip chip'.
I could actually see the approximate fill pattern as I developed the
underfill program. Coupling this with the ability to plop the part under
the acoustic microscope made the process development far easier.
I had a Sonix acoustic microscope and 'loved' it.
Hope this helps somewhat.
Steve Creswick - Gentex Corporation.
-----Original Message-----
From: Graham Collins [mailto:[log in to unmask]]
Sent: Tuesday, September 10, 2002 7:09 AM
To: [log in to unmask]
Subject: [TN] BGA Underfill
Good day TechNet!
A question for the BGA / Underfill guru's out there (hopefully we have some
of those online).
Using a capillary flow underfill on a BGA, say a .050 pitch 26x26 array,
what is a reasonable result to expect with respect to voiding? Should this
process result in no voids, some voids, lots of voids? Given the size of
this part (1.38" square), is it reasonable to expect a void free underfill?
The particulars: I'm using Loctite 3568 underfill, applied using a syringe
along two edges (L pattern). Product is on a hotplate at 100C.
Thanks all!
regards
Graham Collins
Process Engineer,
Northrop Grumman
Atlantic Facility of Litton Systems Canada
(902) 873-2000 ext 6215
----------------------------------------------------------------------------
-----
Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
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 or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/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
----------------------------------------------------------------------------
-----
---------------------------------------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
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 or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/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
---------------------------------------------------------------------------------
|
