TECHNET Archives

July 1999

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Roger Massey-G14195 <[log in to unmask]>
Reply To:
TechNet E-Mail Forum.
Date:
Mon, 5 Jul 1999 09:58:48 +0100
Content-Type:
text/plain
Parts/Attachments:
text/plain (76 lines)
     Bob,

        Do I understand that you want to use an underfill style material to
     help heat flow away from BGA's which are running hot during operation?

        Should be ok (in theory)  there are a wide range of underfill
     materials available, and the usuall suppliers should be able to help
     you with data sheets and samples (Emmerson and Cummins, Alpha, Dexter,
     Loctite etc)  The thing to be aware of is that these "traditional"
     underfills are designed to be used with flip chips, and in small
     volumes,  you are proposing underfilling three BGA per board, and that
     could be potentially expensive, so expensive it may be prohibative
     Loctite were working on an underfill specifically for CBGA, and that
     mey help, or be more cost effective.  I suppose the other potential
     advantage could be an improved reliability, the epoxy should be
     sharing the strains with the solder balls, similar to flip chip
     designs, and this could give an improved lifetime.   On the downside,
     if the BGA is running real hot, then the underfill will need a high Tg
     to minimise thermal mismatch during service, and stop an accelerated
     failure of the balls. In order to get this, you may need a pretty high
     cure schedule, which could exceed the Tg of your FR4, and give
     problems elsewhere.  I guess the way round that would be spec a high
     Tg board (160C+) but that will cost!!

        In terms of processing, I would assume its much the same as a flip
     chip, attach the part, possible clean depending on which flux
     technology is being used, and underfill. (this will take longer than a
     flip chip, and a multi pass dispense may be needed to ensure 100%
     coverage under the BGA) Scanning Accoustic Microscopy, or possibly
     Xray could be used to check the underfill flow.   One major point to
     consider is the need for rework.  I know we shouldnt mention it, but
     we all need to rework something, some time for some reason, and if its
     gunked up with epoxy, "you is gunna have a problem"  there are
     reworkable underfills available, but they can need a two stage cure,
     with rework being carried out before the final cure, again suppliers
     should be able to help with that aspect.


        All in all, if there is a real need to underfill a BGA try it out,
     but beware, the potential cost is high, in terms of additional process
     routes, equipment and also material cost.  If you want to talk more,
     contact me anyway you can.

                Roger


     [log in to unmask]
     tel +44 1462 831125
     fax +44 1462 831336


______________________________ Reply Separator _________________________________
Subject: [TN] Underfill, BGA???
Author:  "TechNet E-Mail Forum." <[log in to unmask]> at #email
Date:    01/07/99 14:12


  Hi all..
       Have a question relating to cooling a HOT 225-ball
  BGA. I tried to look in the archieves, but my search
  technique must be lacking or I am using the wrong
  "words". Can I use an "underfill" to help cool a hot
  BGA or is that the wrong term? I am considering using
  a thermally conductive, electrically insulating
  material under the part (3parts/assy), but have no
  hands-on experience with "what material, how do I apply
  it and what are the related war stories". Any help out
  there in TechNet land???? Thanks in advance, have a
  great day and a super/safe 4th holiday!!!!!!!!!

                          Regards,
  Bob Vanech                      Bob
  Mango Computers
  Norwalk, Ct.
  (203) 857-4008 x 108

ATOM RSS1 RSS2