Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Ceramic BGA Balls Move Post Assy

Views: 2414

#67018

Ceramic BGA Balls Move Post Assy | 15 August, 2012

Wondering if anyone has ever run across this in their travels. We have gotten a few field returns back in recent months and it appears that many of the balls of the PPC440 CPU BGA (ceramic) have shifted. I know we Xray this part 100% during assy so I would like to think it did not leave our facility looking like this. The PCB does still function but only if it is flexed slightly. Undoubtedly, this is "unshorting" the balls.

Ideas on how this could happen? We are running tests now but I figured I would throw it out there in case anyone has already spent the time investigating.

Notes: -This BGA does have high melt temp balls attached with 63/37 to the part and PCB. -We conformal coat the PCB and the coating does wick under the BGA. -We also clip a large radial heatsink to the top of this part and pot around the clip with RTV. (Possible this is trapping heat?)

Attachments:

reply »

#67021

Ceramic BGA Balls Move Post Assy | 15 August, 2012

I saw a similar problem a couple years ago. In our case the issue was the original manufacture of the devices. The high temp spheres were misaligned from the beginning, and subjecting these parts to SMT reflow made the problem worse, and created issues similar to yours. Check carefully for ball alignment at receiving inspection. Reject any that are not properly arrayed. Might be a start... 'hege

Attachments:

reply »

#67022

Ceramic BGA Balls Move Post Assy | 15 August, 2012

If the balls are lead free you need to be running a lead free profile on the oven even if you are running leaded paste.

reply »

#67024

Ceramic BGA Balls Move Post Assy | 15 August, 2012

Interesting Hege, however we xray 100% of these parts during our assembled PCB test process so I have to think this would have been caught seeing as this is not confined to one batch/year/upper level assy (same BGA, slightly different PCBs).

That being said, ill check into the pre assy alignment of our stock tomorrow...

Thanks!

reply »

#67025

Ceramic BGA Balls Move Post Assy | 15 August, 2012

If you are doing X-Ray post place and pre-reflow you should be good. Look elsewhere.

Check profile - full lead free or hybrid?

Be sure you are acheiving the required reflow temps and TAL required for the solder paste too.

Check for belt movement in the reflow oven, should be smooth, no hitching, make sure PCBs are cool before handling out of the oven, Ceramic packages stay hot longer, perhaps you are moving them too soon after reflow? Best of luck tracking this down, let us know what yopu find! 'hege

reply »

#67104

Ceramic BGA Balls Move Post Assy | 25 August, 2012

Hello jdumont,

I think the spheres are leaning as there is not enough meat under the spheres. In ceramic BGA devices there is a solder fillet which is generally a solder paste which is used to give the device some height before balling it. This height prevents the BGA from collapsing completely but if this fillet doesn't wet properly the balls tend to lean post reflow. Let me know what you think?

reply »

PCB Soldering Tools

Precision Auger Dispense Pump