Poor Steve,
I regret you are now member of the sad BGA disaster club. Significant for us members is that we are mano depressive, shouting hurrreuka one week and taking valium the next.
Now, being a little serious, we have seen something like, but bridging occurred in the corners. In our case it seems as the cause for bridging was a slight bending of package during the soldering process, not much, but enough for placing most of the pressure on corner oriented balls, and thus too much smashing and >>>bridging. I have heard others says same thing. Now, why do they bend? According to our BGA supplier, no such bending will happen, but we experience some slight curving unless what they say. Not much..say some 200microns in the middle, but it seems to be enough for making problems.
The reason for bending must be complex, as a BGA oftenly is a complicated thing. Our package is very large, one of the largest, and has cooling plates on top. I don't want to miscredit that company by saying they have a bad design, not at all, but there seems to be one weakness: assymetri. Werner put a finger in the air and sniffed this, a trace I still believe in.
I'm not at all surprised if they bend, you have a number of different materials which have to work together from minus 40C to plus 220C, sometimes even more. You have nonlinear CTE in x and y and z directions in the laminate, influence from Tg which is oftenly only plus 125C, you have a copper balance, you have influence from vias, from possible silverepoxy-filled vias, from metal plates, from molding and from built-in tension etc.
What I say may not have a single bit to do with your actual case, but it's anyway interesting to debate a little about the creatures.
Personally, I have got mountains of good information and help from many of you in TechNet. We are not ready to say Hurray quite yet, but will do soon. BGAs are fantastic components, I really like them.
Ingemar
-----Original Message-----
From: Stephen R. Gregory [mailto:[log in to unmask]]
Sent: den 27 februari 2001 22:34
To: [log in to unmask]
Subject: [TN] Heat-sinked BGA Bridging...
Hi all!
This topic is a "moldy-oldie", but I've started to experience bridges again
with this TI DSP (TMS320C6201GJC200). I've been doing okay up unil recently.
We build these boards in work orders of around 25 pieces.
Our last run we had 7 boards bridged out of 25!! Sent the boards out to be
reworked. I asked that they be x-rayed before and after, and to my surprise
after looking at the x-ray images, they were all bridging in the same general
area.
It's a 4-row periphery BGA, and the bridges are occuring on the inner two
rows, around the center of the part, at the same side of the part.
What's funny, is that the vendor we use to rework the boards for us is having
difficulty as well. We've been asking him to remove, re-ball, and re-install
the part, they said they've been having about a 50% success rate doing
this...they've been getting the bridges too. They have a nice full-blown SRT
BGA rework station and do this for a living.
I print 6-mils of paste (square apertures for the BGA), have a Conceptronic
HVN-70 which I run:
150 160 170 180 240
150 160 170 180 240
25-in. per min
We never have had these problems before...it all started when TI put the
heatsink into the top of the part. I've looked at our paste print and it's
dead on, placement is dead on too, we verify that with a Ersascope.
My question is could there be something wrong with the part? Warping causes
bridges at the corners, right? Why would it bridge in the interior rows at
the same general area?
Thanks in advance!
-Steve Gregory-
---------------------------------------------------------------------------------
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 the following message: SET Technet NOMAIL
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
---------------------------------------------------------------------------------
---------------------------------------------------------------------------------
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 the following message: SET Technet NOMAIL
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
---------------------------------------------------------------------------------
|
