> >3. We have chosen immersion gold as the finish for all production volume >pcb's with 20 mil pitch or boards with lots of 25 mil pitch devices. > > >5. We reduce 20 mil stencil apertures 10% in length & width. We use a 6 >mil mono-thickness stencil. > Thanks for sharing the detail of what you do. I agree it's interesting know what others are doing. I don't agree with the reduction of fine pitch in the length. My experience is that there should be a reduction in the width but the length should be as long as possible. The width reduces the proximity of the solder strips thus reducing bridging. The length is then critical to providing sufficient solder to form a good joint. The solder effectively wicks back along the pad to the termination. The longer the pad, the more solder paste you can put down along it and then the more solder in the joint around the termination. Once you keep it in thin strips, then shorts don't occur. As you continue to extend the pad beyond the toe of the TQFP then better joints can be formed and more repeatable. The rules for pad geometry in my book are:- Min outer dimension with the max contact length should still allow for X2 the max lead thickness behind the heel. The max outer dimension should be extended by X1.5 the max lead thickness, with any extra after that being a bonus. The width of 20 thou device pads should be as someone already stated, 12 thou with apertures of 10 thou. 25 thou depends on the lead width but we use 16 and 14 thou pads. The apertures being 14 and 12 thou. I make the pad width the same as the max width of the lead, which leaves no side fillet at that dimension, but usually they follow the nominal width. The length of the pads using the above usually amounts to more than 70 thou, but if it doesn't then I reccommend extending the outer dimension to make them so. As I have said before, I don't believe it is neccessary to have an alternative finish to HASL for fine pitch. I use HASL for boards which are heavily populated with 20 thou including connectors at this pitch, with very high yield. The key is in the upfront engineering of the pad geometries and apertures. If there are different suppliers of devices, make sure all parts are vetted by you with Data sheets so you can calculate the above dimensions. Obviously the inside and outer dimensions will need to encompass the range created by multiple sources. But don't let purchasing bring in parts that you haven't approved. Thanks again for sharing the details, I wish more Technetters would do the same, then it would be easier to answer their questions, and we would all learn. Regards ############################################################## TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c ############################################################## To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body: To subscribe: SUBSCRIBE TECHNET <your full name> To unsubscribe: SIGNOFF TECHNET ############################################################## Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional information. For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311 ##############################################################