Copper under BGA

[Rod Ladwig]

Hi Tom

I was asked to look at a problem where a BGAs balls had partly collapsed in the reflow process. The PCB is a 4 layer design. the designers used a full copper ground inner layer and a tracked power inner layer. The collapsed ball occurred in the area where a thick power track ran under the BGA in the inner layer. It would seem that this caused the heat from the oven to last longer and collaspe the balls just in that area.
1: Have you experienced anything like this?
2: Is a fully flooded power plane the best solution?

Thanks

Rod Ladwig
Unique Designs

[Tom]

Rod,

I have never experienced this problem before, so I don't have any personal experience regarding this issue.

There could be several reasons why the balls collapsed and one of the first things I would look for is the BGA construction and manufacturability. Sometimes we overlook the fact that the component may be at fault.

The second thing I would look at is the assembly process and "Did they do their job right".

I cannot seem to rationalize why this would occur regardless of copper trace widths or flooded planes. We have BGA's on almost every design with every type of copper trace width and plane scenario you can imagine and never experienced collapsed BGA balls.

[Tom]

See these threads in the IPC Advanced Certification section:

http://www.pcbstandards.com/forums/showthread.php?s=&threadid=917

http://www.pcbstandards.com/forums/showthread.php?s=&threadid=918