We are just starting our development of 0.5 mm pitch BGAs and I am hearing
conflicting reports as to the approach we should take with respect to the
soldermask relief around the solderball pads.

First of all, the 0.5mm BGA is made up of pairs of pads in a perimeter
layout, along with a larger perimeter also made up of pairs of pads.  There
is also a ground block in the center.  Basically it looks like two
concentric windowframes with a large gap between the frames for routing.

Historically we have always made our soldermask 1:1 with the pad size and
then the pwb fabricators would swell the opening so that the soldermask did
not encroach upon the pad and make it a mask defined pad.  That works fine
for 1.0 and 1.27 mm pitch devices because there is still enough soldermask
left to actually stick to the board.

When we reduce the pitch down to 0.5 mm, the gap between and the
registrations swelling, basically eliminates the thicker stripe of
soldermask down to a very thin (and poor yielding) line which may or may
not stick.

This is the exact same problem we have always had with 0.5 mm QFPs and the
solution was always to permit the pwb fabricator to window out the  whole
area between the fine pitch pads.

So here is the question:

If we open up the soldermask aperture so that it is something like .065mm
away from the actual etched pad, doesn't that just reduce the producibility
of the board at the pwb house.   The reason for specifying this tight
requirement is speculated to be related to reliability but I'm not
convinced yet.

In reality very few of the pads are strictly etch defined round pads
because many have surface conductors coming off of them which makes the
pads a hybrid of etch and mask defined.  The greater the relief of the
soldermask, the more non-uniform the exposed metal pad will become.

Because there is more surface to wet to, this has the effect of reducing
solderball volume and we expect to see the diameter not the height change
accordingly.

My suggestion is to eliminate the soldermask wherever the tight pad to pad
areas are so that for two round pads, the aperture would be a very wide
oval shape.  This also means that the routing away from the pads has to go
to the outside away from the open soldermasked area.

The intention here is minimize the enlargement of the pads with respect to
wettable area and at the same time make the soldermask application process
easier.

Ironically it seems that 0201 pads can also follow this logic.
Hopefully I've explained this in detail enough for everyone to understand.

What does your land patterns guidelines when it comes to soldermask for
fine pitch QFPs, BGAs, and 0201s?

All inputs are appreciated.

Phillip A. Bavaro
QUALCO/\/\/\/\  Incorporated
Engineer, Staff
[log in to unmask]
Pager (858) 271-3640
Tel  (858) 658-2542 Voice mail
Cell (858) 845-9968 (workday)
Cell (619) 602-8644 (offsite, after hours)
Fax (858) 658-1584

---------------------------------------------------------------------------------
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
---------------------------------------------------------------------------------