Juan
There are many things that cause this and the first thing I would do is request quality records from your board fab facility on that particular lot. If they are unwilling to give them to you this is a good sign they had a processing problem. This will be important to have because they will claim there is no such thing and you are processing your boards incorrectly. The first thing you need to look at is how thick is your gold. 2 uin or below is bad you can get nickel oxidation (one of the primary causes of BP.) 4-8 u in is going to be your typical range. PS don't be alarmed by wide swings in your thickness number a SEM will look at a very small spot on a pad and thickness swings up to 16-20 u in are not uncommon. Under the minimum of 2-3 u in are what you are looking for. Next you need to look at your phosphorus levels. High phosphorus levels are a sign of poor process control at FAB and will cause your problem. Both of these are done with Scanning Electron Microscopes with x-ray fluorescence�s capabilities. There are a lot of labs around that can do this, we have used Nextek in Huntsville Al and Siemens in Norcross Ga. both are very capable. 1 other thing when processing boards with nickel over gold if you are running in an oxygen atmosphere you need to peak at 220c, 5 deg less in a nitrogen atmosphere. You have to remember you are bonding to nickel not copper or gold, nickel is much harder than copper and it takes more heat to make that bond.
Good luck
reply »