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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Micro BGA pads, adhesion to laminate

Matt Kehoe

#25104

Micro BGA pads, adhesion to laminate | 9 July, 2003

Our company applies solid solder to surface mount lands. While processing a micro BGA design we had some pads fall off, peel, whatever you might call it, from the surface of the board when touched "GENTLY" with the tip of a blade to remove a small squeeze out. The pads fell off so easily it seems like it may have been a clad copper to laminate issue. Most of the pads in the footprint were soldermask defined, a larger trace with the soldermask opening creating the actual BGA pad. These seemed to hold however the pads that were acually flashed .0079 copper seemed to fall right off if touched even gently.

Could this be overetching? How about defective laminate? Adhesion problem? Is a pad .0079 that fragile? Can we trust the pads that did not fall off?

Thanks in advance for your feedback. mk

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

matt kehoe

#25105

Micro BGA pads, adhesion to laminate | 9 July, 2003

Photo's can be found at

http://www.sipad.com/ssi_qa/smta/microbga.htm

mk

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

#25113

Micro BGA pads, adhesion to laminate | 9 July, 2003

Matt: What are we looking at in pix F2?

reply »

Matt Kehoe

#25115

Micro BGA pads, adhesion to laminate | 9 July, 2003

Hi Dave,

Hmmm, pix F2? Do you meran figure 2 or second picture?

The second picture is the bare board before printing. Gold plated pads.

Is that what you mean?

mk

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »


RDR

#25116

Micro BGA pads, adhesion to laminate | 9 July, 2003

Mike, the one pad that is "small" definitely appears to be over etching (It looks like it was supposed to be much larger). What is the dims of this little bugger as compared to the Gerber? i would also tend to believe that there probably is some mask on the pads. What is this board finished in Gold or copper? (I can't tell from the pics)

I have found that unsupported pads (no electrical connect) usually lift/falloff much much easier than those that are supported. I think that this is due to the small size with no mask or large/long trace to help with the adhesion. I don't know if there is a PCB supplier problem for the detaching because it is very hard to define "GENTLY" I would think that cutting a piece of solder from a pad would not be "gentle" especially one this size. With the solder adhered to a pad and then trying to cut it off seems like it would apply a significant amount of leverage on the pad.

I wouldn't think that the other pads are going to be an issue. All in All, Who knows?

My opinion Russ

reply »

#25117

Micro BGA pads, adhesion to laminate | 9 July, 2003

Sorry Matt. I mean the last row, second picture in from the left.

reply »

Matt Kehoe

#25119

Micro BGA pads, adhesion to laminate | 9 July, 2003

Right-o Dave,

Thats the 2 lifted pads turned upside down so actually you are looking at the bottom of the base copper where it used to be attached to the laminate.

mk

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

Matt Kehoe

#25120

Micro BGA pads, adhesion to laminate | 9 July, 2003

Hi Russ,

Thanks for the input. The squeeze out we are reworking is very very thin since it has been flattened so it is actually easier to remove than usual but I agree, with a pad this small it could exert a bit of stress on the bond line if rework was needed. I cannot measure the pads as they have already shipped to the customer. One of the supported pads did pop off (bottom row, second picture) as well so that is why I am thinking adhesion problem. Your opinion is greatly appreciated.

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

#25126

Micro BGA pads, adhesion to laminate | 10 July, 2003

A reading of the pictures is: A1: Looks like pad [B1] is smaller than the other pads. It�s weird that the pads are such different colors. A2: Looks like pad [B1] is smaller than the other pads. Solder mask opening [B1] is misregistered and seems smaller than other openings. Solder mask seems very thick, relative to the pad thickness. B1: Silk screen is closer to the pads than A1 and C1. B2: Silk screen is closer to the pads than A1 and C1. B3: All solder paste is misregistered in printing. Amount of paste for pad [B1] is smaller than the amount for other pads. C1: ~ C2: Several solder bumps [A1, A4, B4, C1, C4, D4] seem larger than other bumps. Solder mask appears to be poorly adhered in several locations [A2/A3, B3/C3, C2/C3, D2/D3]. D1: Pad does not accept solder. Solder bumps vary in size. Solder mask misregistered. Solder mask seems very thick, relative to the pad thickness. D2: As in D1. Do not see �wicking to adjacent pad� that Matt mentioned. E1: Several flattened solder areas [A1, A2, A4, B4, C2, D2, D4] appear larger than other areas. These larger areas do not correspond with large solder bumps mentioned in C2, because the pictures [C2 and E1] are of different boards. Silk screen is closer to the pads than A1 and C1. F1: Two corroded pads. Pads appear to be present with the corrosion sitting on the surface. Solder mask seems very thick, relative to the pad thickness. F2: Unattached to the board, two pads with excess solder that connect them together.

While there seems to ample points to criticize this board design and fabrication, the more distressing faults are: * D1: Pad does not accept solder. Does this go back to the discolored gold on pads in A1? Could the gold plating be thin enough to allow the nickel underplate to become unsolderable? Alternately, it�s tough to say, but the pad still appears to be gold colored, indicating that solder never touched the pad. That could be caused by poor solder paste printing process or residual solder mask bleed remaining on the pad. * F1: Two corroded pads. This corrosion appears to on top of the pads. Is this a function of either poor plating or process materials being retained in deep solder mask �wells�? What is the composition of the brown coarse residue in the solder mask well? * F2: Unattached to the board, two pads � The laminate to pad surface appears smooth and with no signs of over etching. Could this be caused by poor laminate to pad adhesion? Are other pads �falling-off�?

reply »

Manufacturing Software

Boundary Scan