Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
BGA Warping at the corners

BGA Warping at the corners

Subscribe to Thread

My circuit board contains 11 x 11 matrix 1mm pitch BGA, The ... - Jun 17, 2013 by saravanan

You have not mentioned exactly how the BGA was reflowed onto... - Jun 17, 2013 by Hegemon

Hi, the BGA and/or the PCB may be warping excessively during... - Aug 30, 2013 by smtgs02jrt

One of our PCB for high reliability application is having 2 ... - Oct 21, 2013 by saravanan

The first picture shows an issue. Note that the vias adjace... - Oct 21, 2013 by Hegemon

In continuation of the thread, attached 2d & 3d x-ray pictur... - Nov 07, 2013 by saravanan

1st X-Ray looks pretty good. One instance of what looks lik... - Nov 07, 2013 by Hegemon

saravanan

#69336

BGA Warping at the corners | 17 June, 2013

My circuit board contains 11 x 11 matrix 1mm pitch BGA, The device was tested in socket on evaluation Board. We handed over the actual PCB & devices for re-flow. The assembly house first re-flowed all the SMT components other than BGA. BGA was placed after first re-flow, Micro stencil was used for dispensing solder paste. After attaching the device , the Board was not functioning and the problem is traced to BGA. when inspected, we found the outer rows of the BGA balls are melted excessively. Edges of the BGA is warped to the board side. But X-Ray shows no solder bridging. How to trouble shoot and to find the root cause of the problem.

reply »

Hegemon

#69350

BGA Warping at the corners | 17 June, 2013

You have not mentioned exactly how the BGA was reflowed onto the board. Was it sent through the oven again after the mini-stencil, or was it heateed up and reflowed using a BGA rework station, or other hot air device??

From the sound of it (outside rows collapsed too much, plus warpage)it sounds as though the package experienced a temperature profile that was outisde its required parameters, possibly too much heat, or the correct heat, but ramped up too fast. BGA reflow should provide a nice consistent appearance of the spheres, no matter if the outside row or one of the inner rows. That you have the outside row looking less than optimal makes me think that the part was overheated during reflow. Either too much temperature (BGA overheated) or too steep a ramp rate up to reflow (BGA internal damage, or warping). That your assembly house did the SMT parts separate from the BGA device is a bit of a red flag for me.... Just my $.02 'hege

reply »

smtgs02jrt

#70594

BGA Warping at the corners | 30 August, 2013

Hi, the BGA and/or the PCB may be warping excessively during reflow.

We offer testing services to measure warpage of BGAs and PCBs during reflow to JEDEC and IPC standards. Let me know if you'd like to learn more about our services: joe@zntechnologies.com

reply »

saravanan

#71259

BGA Assembly | 21 October, 2013

One of our PCB for high reliability application is having 2 BGAs(11X11, ADV 204), The pitch is 1 mm. We decided to wire the BGAs first using re-flow soldering and the rest of the components by hand soldering. Since the PCBs are for high reliability application , we don't want to do re-work BGA. If the soldering of BGA is not ok , we will scrap the board.

After re-flow X-Ray inspection was carried out. The pictures are attached. Kindly comment on the BGA soldering quality. This will help us to decide go/no go for soldering the rest of the components.

Attachments:

reply »

Hegemon

#71263

BGA Assembly | 21 October, 2013

The first picture shows an issue. Note that the vias adjacent to the interior rows are filled with solder. At the outside rows this is not evident. What you might have are interior spheres that are smaller in volume than the rows on the outide, since the interior solder has migrated to the vias. For the record, if possible, always have the vias tented beneath BGA or any other bottom terminated devices, it's just a better idea in general. The second picture is better. No solder in the vias, so more consistent ball size is likely. A better result than the first picture for sure. The "tear drop" shape that you are seeing in both pictures is not a problem, although is does tell me that the solder mask and pad definitions are not absolutely optimal. Not a game ender. IMHO, pic#1 says "back to the drawing board" and looks a bit worrisome, while pic#2 says "go for it". 'hege

reply »

saravanan

#71344

BGA Assembly | 7 November, 2013

In continuation of the thread, attached 2d & 3d x-ray pictures of BGA after assembly. Kindly comment.regards,

R.Saravanan

Attachments:

reply »

Hegemon

#71347

BGA Assembly | 7 November, 2013

1st X-Ray looks pretty good. One instance of what looks like solder migration in the lower left quadrant.

2nd X-Ray shows a number of locations where the solder has migrated into the vias, shown by the small dark circles between the main spheres. There are also a number of vias that did not have any solder migrate into them, evidenced by the light circles between the main spheres.

The 3d Study is of the 2nd X-Ray. The main spheres are shown as the taller and larger diameter features. The solder that has migrated into the vias are the smaller diameter less tall features. The open vias with no solder are represented by the smaller features with the hole in the center, looking like miniature volcanoes.

To summarize: X-Ray #1 is pretty good. X-Ray #2 is not so good. Too much solder migration, leads to inconsistent sphere diamter post reflow, and possible issues related to the attachment, made worse by thermal excursions, and normal temperature cycling.

Opinions are like noses, everyone has one.

Best of luck! 'hege

reply »