Issue link: https://iconnect007.uberflip.com/i/446087
34 The PCB Design Magazine • January 2015 dictates a minimum of .8 to 1 mil in the barrels of plated copper for continuity. So, here in the U.S., even starting on quarter-ounce outers will result in 1.5 oz. finish after plate typically. If you have numerous impedances on the surface, having to redo them all because this was not considered will result in having to redo all the calculations, which takes more time. The old adage, "Time is money" is certainly true in the PCB fabri- cation business, especially with RF parts. 2. More hybrid analog digital parts. Why make a digital part that may require a series of smaller piggybacked or surface mounted parts when you can get the best of both worlds (depending on the ap- plication) with a part that is a hybrid of analog and digital type materials? A combination of high-speed analog RF mate- rials where surface CPWGs can be maintained at a reasonable geometry and digital signaling can be kept to the inside in the form of DDR, SATA or LVDS lines. 3. More surface finishes for lead-free assemblies. Wait and see. The surface finish segment is likely to keep growing with demand for lead-free PCBs. What This Means to Designers These first three items alone should be enough to convince you that you should take some time to discuss your specific needs with any potential fabricator partner. A simple 15-minute conversation can save everyone a lot of grief. A recent example: One of our longtime customers recently started adding edge slots and cutouts as sepa- rate entities instead of with the part profile or outline. After discussing this with the customer, we found the reason. Their thought process was that they understood slot positional accuracy to be around +/-.003", whereas the rout positional accuracy could be as much as +/-.005". What they really wanted was edge features relative to inboard features with an accuracy of about +/- .003". We fabricators can simply say, "Make sure these additional edge-routed positional features are done at primary drill when all the other holes are done." This ensures a better positional relationship than two separate entities. Additionally, we came to find out this was a suggestion from another fab- ricator! Some fabricators pro- pose solutions that may not work for all fabricators. Again, having a short conversation can sometimes lead to a much greater understanding of the issue and much quicker solu- tions to the problem. I also see many custom- ers finally embracing simple things like differentiating im- pedance line sizes by a tenth or even a hundredth of a mil; this identifies them but keeps them close to the original in- tended and calculated line size. This is considerably faster than at- tempting to identify each impedance scenario by the net names they are associated with or their component locations. Other Predictions for 2015 I predict the continued use of holes .008" and smaller based on board geometries and available real estate. I see more usage of .006" vias to help reduce associated pad sizes and still be able to ingress and egress out of fine-pitch parts. More folks are specifying these as +.000"- the hole size and more folks are either silver- filling them (if they need the extra conductiv- ity for, say, heat considerations) or epoxy-filling them just so they are flat at assembly. I also see more use of true .003"/.003" spaces and traces on surface layers for the same rea- sons. As the pitch between the BGA balls and vias shrinks, so too will the associated trace widths. We are seeing a lot of true .003"/.003" for .5 and .4 mm pitch devices. Bear in mind WHAT WILL 2015 BRINg? continues the bare (board) truth some fabricators propose solutions that may not work for all fabricators. again, having a short conversa- tion can sometimes lead to a much greater understanding of the issue and much quicker solutions to the problem. " "