BGA solder joints versus mechanical stresses

Has anyone done or seen any studies done documenting what may be the affect of hardware such as screws in the parimeter vicinity of BGAs on a circuit board?

What I'm really after is anything describing how close a screw/washer/nut combination which will cause slight compression of the PWB material before it begins to affect solder joint integrity under the BGAs mostly at the corners but along the long sides as well.

Thanks, Rick

reply »

| Has anyone done or seen any studies done documenting what may be the affect of hardware such as screws in the parimeter vicinity of BGAs on a circuit board? | | What I'm really after is anything describing how close a screw/washer/nut combination which will cause slight compression of the PWB material before it begins to affect solder joint integrity under the BGAs mostly at the corners but along the long sides as well. | | Thanks, | Rick | Have done this type thing since the beginning. Problem is too many variables. My rule of thumb has become not using mechanical fastners, or any kind, within one full inch of the little beasts. I wish I had more, but most focus now is on thermal and mechanical issues as rework, warp and twist, and all the rest.

I can say, I've seen many failures using inserts and "over torking" mechanical fastners close to most "rigid" or non-compliant SMT device types including chips. This especially is compounded under thermal stress as encountered with surface mount connectors requiring board inserts as well as during operational performance.

Earl Moon

reply »

| | Has anyone done or seen any studies done documenting what may be the affect of hardware such as screws in the parimeter vicinity of BGAs on a circuit board? | | | | What I'm really after is anything describing how close a screw/washer/nut combination which will cause slight compression of the PWB material before it begins to affect solder joint integrity under the BGAs mostly at the corners but along the long sides as well. | | | | Thanks, | | Rick | | | Have done this type thing since the beginning. Problem is too many variables. My rule of thumb has become not using mechanical fastners, or any kind, within one full inch of the little beasts. I wish I had more, but most focus now is on thermal and mechanical issues as rework, warp and twist, and all the rest. | | I can say, I've seen many failures using inserts and "over torking" mechanical fastners close to most "rigid" or non-compliant SMT device types including chips. This especially is compounded under thermal stress as encountered with surface mount connectors requiring board inserts as well as during operational performance. | | Earl Moon | Thanks Earl, is that "one inch" documented anywhere? I'm looking at having to do a (probably) three month long experimetal thermal stress study if this has not been conducted and reported on by someone already.

reply »

| | | Has anyone done or seen any studies done documenting what may be the affect of hardware such as screws in the parimeter vicinity of BGAs on a circuit board? | | | | | | What I'm really after is anything describing how close a screw/washer/nut combination which will cause slight compression of the PWB material before it begins to affect solder joint integrity under the BGAs mostly at the corners but along the long sides as well. | | | | | | Thanks, | | | Rick | | | | | Have done this type thing since the beginning. Problem is too many variables. My rule of thumb has become not using mechanical fastners, or any kind, within one full inch of the little beasts. I wish I had more, but most focus now is on thermal and mechanical issues as rework, warp and twist, and all the rest. | | | | I can say, I've seen many failures using inserts and "over torking" mechanical fastners close to most "rigid" or non-compliant SMT device types including chips. This especially is compounded under thermal stress as encountered with surface mount connectors requiring board inserts as well as during operational performance. | | | | Earl Moon | | | Thanks Earl, is that "one inch" documented anywhere? I'm looking at having to do a (probably) three month long experimetal thermal stress study if this has not been conducted and reported on by someone already. | | | This is one of those things not widely interesting to many. That's why I said, considering many variables, I don't do it. I will say that H-P agreed and did not allow this activity until further studies were made available.

Note: H-P has one of the best advanced assembly groups around. They do most of the homework before any fab of assembly is done. IPC provides guidelines, thankfully, but I've never seen any concerning this issue.

The vairables first relate to mechanical stress and distortion placed on laminate material. Second, once in place, the fastners soak up heat during assembly and operational cycles. This stresses the area around. The question is, as you ask, how far and with what devices/fastners.

Wish I could help more, but remember boards don't like too much mechanical stress. They like thermal stress less. The combination is not pleasant for boards or us.

Earl Moon

reply »