Good Solder Joint Failures

Hope some of you out there can help me with this one. I am getting feedback from our test department that they are showing opens on solder joints which have visually good wetting. The touch-up operators confirm that the joint looks good. All they do is go over the joint(s) with an iron and it passes test. The frequency of this issue is very low (though no ppm data is available at this time) but it is enough for concern. There seems to be no pattern to location, part type or board. We use primarily no-clean solder paste. I have asked for samples from future occurances. These will be viewed with higher power scopes and x-rayed. Any comments are appreciated.

reply »

| Hope some of you out there can help me with this one. | I am getting feedback from our test department that they are showing opens on solder joints which have visually good wetting. The touch-up operators confirm that the joint looks good. All they do is go over the joint(s) with an iron and it passes test. The frequency of this issue is very low (though no ppm data is available at this time) but it is enough for concern. There seems to be no pattern to location, part type or board. We use primarily no-clean solder paste. | I have asked for samples from future occurances. These will be viewed with higher power scopes and x-rayed. | Any comments are appreciated. | Here's a comment related to subjective solder joint acceptance criteria. Though it looks fine visually, it isn't always so.

Chip device cracking, plated hole failures (caused by low electroplated copper ductility), cracks in solder joints not visually observed, poor wetting to oxidized or otherwise contaminated surfaces, etc.

There are so many reasons why visual acceptability cannot stand alone. Most of which are related to poor process management as in component lead metallization or board surface solder termination area problems, as examples. In other words, it is possible to cover over defect with visually asthetic solder joint appearances.

A different case in point would be a plated through hole barrel having cracked under thermal stress or shock. During functional testing, probe pressure compresses the hole wall making momentary contact thus rendering a defect into an acceptable condition. The same may be said for underlying "cracks" just beneath solder joints appearing acceptable.

Earl Moon

reply »

| Hope some of you out there can help me with this one. | I am getting feedback from our test department that they are showing opens on solder joints which have visually good wetting. The touch-up operators confirm that the joint looks good. All they do is go over the joint(s) with an iron and it passes test. The frequency of this issue is very low (though no ppm data is available at this time) but it is enough for concern. There seems to be no pattern to location, part type or board. We use primarily no-clean solder paste. | I have asked for samples from future occurances. These will be viewed with higher power scopes and x-rayed. | Any comments are appreciated. | You might want to do some research into plating brighteners and brightener entrapment in a solder plated surface. If your component vendor is not controling the brightener concentration, the organics can become entrapped. During your reflow process the entrapped brighteners outgas into you solder joint. Inspection the failed joint at the solder/component lead interface will look like swiss cheese. Some solutions are to increase your time and temp at liquidous, or find a new vendor.

A quick check for brightener entrapment. Equipment: soldering iron set at 500f, microscope

While watching under a microsope, touch the lead of your component with the iron and watch. If you have brightener entrapment you will see the solder melt down the lead, this will be followed by the bubbling/dewetting of the solder. That is the brighteners out gasing.

reply »

| Hope some of you out there can help me with this one. | I am getting feedback from our test department that they are showing opens on solder joints which have visually good wetting. The touch-up operators confirm that the joint looks good. All they do is go over the joint(s) with an iron and it passes test. The frequency of this issue is very low (though no ppm data is available at this time) but it is enough for concern. There seems to be no pattern to location, part type or board. We use primarily no-clean solder paste. | I have asked for samples from future occurances. These will be viewed with higher power scopes and x-rayed. | Any comments are appreciated. | Hi again, Thanks for the input. Here's more on the problem... I performed an SEM analysis on some virgin PLCC sockets. Picking up a nickel interface between the Tin/Lead plate and the copper lead. I didn't believe this was a common practice. Also picked up what looked like a line of demarcation between the nickel and copper. That could be the source of some concern...

Dave

reply »