I=E2=80=99ve been working on a new project using the Intel Atom = processor and 945GSE=20 chipset. In the platform design guide that Intel published they = have=20 various recommendations for routing traces to the corners of BGA PCB = footprints=20 for the parts and I was curious as to why these were being recommended. = =20 It turns out I=E2=80=99ve been fortunate in my past experience with = RoHS-based designs=20 since most of the boards I work on are small, fairly rigid (most are = 1.6mm=20 thick) and the BGA components were few and of fine pitch. However, = there is an=20 ugly failure mode with RoHS process boards and BGA=E2=80=99s which has = come to be known=20 as =E2=80=9Cpad cratering=E2=80=9D. This is a bit of a misnomer, since = the =E2=80=9Ccrater=E2=80=9D is the=20 space left vacant when a BGA ball pad lifts off the board.
Causes of pad cratering
Simply put, pad cratering is merely the failure of the resin bond = between=20 the board and the surface ball pads of a BGA footprint. This has = been =20 exacerbated by RoHS because of the higher reflow temperatures needed =20 (embrittles the resin) and the greater hardness of RoHS compliant =20 solders. There are at least two main causes: mechanical stress due = to =20 thermal cycling of the board during reflow, and stress on the joints = caused by=20 board flexing during handling and other mechanical shocks. The = failures=20 normally begin at the outside corners of the BGA package and may not be = apparent at first if the connecting traces to the pads don=E2=80=99t = immediately=20 fracture. The most common electrical failure point is at the = perimeter of=20 the ball pad where a routing trace connects to the pad.
Dealing with pad cratering
Note that this heading does not say =E2=80=9Cpreventing pad = cratering=E2=80=9D as it appears=20 that simply attempting to prevent pad cratering will not yield = significant=20 increases in board reliability. The most extreme example I=E2=80=99ve = seen in the=20 attempt to prevent pad cratering is to apply epoxy to the 4 = corners of=20 each BGA package to provide mechanical stress relief for the corner = pads. =20 For most of us however that=E2=80=99s simply not practical. What = has apparently=20 been working for me is to break out the corner BGA pads and those of = the next=20 row back on the diagonal with trace widths equal to the diameter of the = ball=20 pads.
In this view the yellow shapes represent pads with a 4mil via-in-pad = blind=20 via connecting to the second layer, purple pads do not have vias. = Via-in-pad=20 was mentioned in one study I read as another possible way to mitigate = pad=20 cratering but that not enough research had been done to determine its=20 effectiveness. The idea behind the wide traces at the pad is to = eliminate=20 copper trace fracturing at the pad perimeter. Evidently the dual = approach=20 of using fat traces plus via-in-pad has worked for me as none of the = boards on=20 which I=E2=80=99ve used this approach have yet to experience a failure = due to pad=20 cratering. These boards are only 0.8mm thick and are used in a = hand-held=20 product, unfortunately a prime candidate for pad cratering failure. But = so far=20 so good=E2=80=A6
It=E2=80=99s been suggested that for attaching traces to the corner = ball pads one=20 should route away from the pad at a 45 degree angle relative to the pad = with a=20 wide trace for about 1.2mm, then neck down to normal routing width and = go off in=20 the direction of the route. The image below illustrates this:
Another approach to reducing the stress at the pad to trace boundary = is to=20 use solder mask defined pads rather than metal defined pads for the = entire=20 array. Jon Manning (you can find him on Linkedin) suggested using = solder mask=20 defined pads for the first 3 rows in on the diagonal, using the same = solder mask=20 size as pad metal size for the balance of the array, and increasing the = pad=20 metal size in the corners appropriately. Depending on the ball pitch of = your=20 parts and your board layer structure it seems that some combination of = these=20 approaches should work for you to sufficiently mitigate the effects of = pad=20 cratering. Since shock and flexing are a significant contributor to pad = cratering that=E2=80=99s one place the mechanical design group can make=20 contributions. We don=E2=80=99t use solder mask defined pads on = all ball pads as a=20 rule since metal defined pads appear to have somewhat highter soldering=20 reliability and less problems with joint cracking.
Nicholas Vickers, Kyle Rauen, Andrew Ferris and Jianbiao Pan of Cal = Poly=20 State University published a study of BGA soldering reliability. Their = research=20 conclusions can be found here. Also, thanks to Jon Manning for = sharing his=20 considerable pcb design experience.
-Jeff