Like all who post messages here, I'm looking for a little bit (well, maybe a lot of bit) of help. We're currently going through a redesign of a board which will utilize several FPGAs. The manufacturer (Xilinx) recommends using solder mask defined pads which is of major concern to me. My concerns are primarily due to potential problems with the solder mask chipping away and entering the solder joints or just changing the solderable area of the pad, especially in the case a component needs to be reworked. Are my concerns justified? Am I missing anything? One of the primary reasons of using SMD pads is to increase the surface area between the copper and the resin to reduce the possibility of damaging pads during rework, correct? Believe me, I'm not about to ignore this possibility as we have had pads damaged when BGAs have been replaced on a couple of occasions but can't we also decrease the risk of damage by using a higher Tg material (maybe Tg=210 or so)in the fab of the PCB's? I would like to convince the powers to be to invest in PCB tools that can handle blind and buried vias. Wouldn't blind vias help anchor the pads to the board during rework? Any help in this matter would be greatly appreciated. Thanks in advance.
Chris: Whatever way you decide to define your pads, use the same type on the component and the board.
Non-solder mask defined pad advantages: � No point-of-stress concentration at edge mask and solder ball � No CTE mismatch at edge mask and solder ball � More space for routing traces between pads Non-solder mask disadvantages: � Smaller pads have poorer adhesion strength, performance in rework and multiple reflows, and heat dissipation Solder mask defined pad advantages: � Larger pads have better adhesion strength, performance in rework and multiple reflows, and heat dissipation Solder mask defined pad disadvantages: � Potential stress concentrations may start solder-joint cracking sooner � Less space to route traces between pads