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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Question on Thermal Overstress on BGA Balls During Rework

Views: 3271


MLC

#40510

Question on Thermal Overstress on BGA Balls During Rework | 17 March, 2006

We are currently struggling with a particularly difficult component selection form our designers. This is a custom BGA, 592 balls, metal cover with a very large bead of adhesive using to retain the cover. We have had significant fallout (greater than 50%) due to potato chipping during reflow. We do prebake the device before using, and between usage if it has been exposed for longer than about 3 days. At rework, we recently had a device develop several odd-shaped balls after reflow. For lack of a better term, the balls developed a bulge at the ball center, almost looking like a pear on its side. The bulge was only present on one side, and not on every ball (maybe 10% of the balls were affected). The joints had a dull grey finish on visual inspection, which I at first attributed to using 63-37 paste on an ENIG finish PWB. Now I am not so sure. Could this have been a device that was thermally overstressed? I have been thermally profiling for reflow ovens for many years, and have never seen this condition from a reflow oven. Where the BGA removal process can be manipulated by the operator, could we have heated the part for too long, or could it have gotten too hot? Any other thoughts as to what this could be?

reply »

Slaine

#40679

Question on Thermal Overstress on BGA Balls During Rework | 28 March, 2006

can you post some photos somwhere?

reply »

#40680

Question on Thermal Overstress on BGA Balls During Rework | 28 March, 2006

What do you mean the bga removal process can be manipulated by the operator?

reply »

GS

#40694

Question on Thermal Overstress on BGA Balls During Rework | 28 March, 2006

Hi MLC,

did you profiled the BGA itself?

I mean detect Temperatures on middle/central ball vs external ball and see how much is the delta Temperature. Best practice is to keep the delta T at minimum as possible especially on a so large BGA.

Sure the balls are Sn/Pb 63/37 made ?

Regards..........GS

This message was posted via the Electronics Forum @

reply »


MLC

#40696

Question on Thermal Overstress on BGA Balls During Rework | 28 March, 2006

I unfortunately cannot post any photos, the unit in question is classified. And, as far as my comment that the BGA profile can be manipulated, if an engineer is also working the process with one of the operators, they can override the profile at any time and allow for a longer soak, shorter reflow, etc. The initial problem was seen on one of our first cards to process through rework, so the area support people were still trying to refine the process.

reply »


MLC

#40697

Question on Thermal Overstress on BGA Balls During Rework | 28 March, 2006

We profiled the BGA, drilled into center and outer balls. Original profile done this way was not hot enough when we actually tried the run, which is why the area engineer began to review and modify the recipe. And I am very sure the balls are 63/37.

reply »

Slaine

#40714

Question on Thermal Overstress on BGA Balls During Rework | 29 March, 2006

where are your pear shaped balls located? :)

you said about 10% of the balls have been affected are they mainly around the egde of the BGA or in the corners ect?

reply »

Thermal Interface Material Dispensing

IPC Training & Certification - Blackfox