Issue link: https://iconnect007.uberflip.com/i/691539
14 The PCB Design Magazine • June 2016 Gaines: Yes, we have tools that we can set up rules/constraints for those specific needs. In today's world, most higher-end tools allow you to set up rules/constraints for certain areas. This allows for those tight rules to be enforced in the area that drives that tight lines and spacing, while allowing more re- laxed rules/constraints on the rest of the board. Setting up those rules could be made easier, but I appreci- ate what we have today to control layout, compared to what we had to use in the past. I know you can find lower-end tools from a cost standpoint; however, these tools often do not have the features needed to perform the more complex mixed-technology boards required by our clients. I think the old saying "You get what you pay for" applies here. If you need the layout control, you may have to pay more for it. Shaughnessy: With finer features, how do you de- termine whether it is cost-effective to use HDI? Gaines: Most of the components we have to work with are already selected by the time we get the design. I think more knowledge needs to be shared with the engineering and com- ponent teams that are selecting the core com- ponents. They need to better understand the impact those HDI components have on layout, board procurement and assembly. Many new- technology components were developed to make things smaller for end-products such as handheld devices. But most of our boards are not that small or dense. These HDI components can increase layer counts and require the use of smaller lines/ spaces and via sizes. Many times, one compo- nent drives the entire board technology from a fabrication standpoint. If more time could be spent on selection of a component that better fits the overall design and layout complexity of the board, the end-product would cost less due to reduced procurement and assembly issues. I have asked before why an ultra-fine-pitch BGA was selected instead of a 1 mm pitch pack- age. The answer was that the engineer thought it was sexy. This "sexy part" caused issues in the bare board procurement and the assembly process. Yes, process is- sues are fixable, but when you are trying to keep the cost down, the overall impact of these ultra-fine pitch parts should be considered. Shaughnessy: You work with ultra- fine-pitch BGAs. Do you have any special techniques for doing ultra- fine-pitch BGA fanouts? Gaines: We have used ultra-fine-pitch BGAs for projects in past, and I see this becoming more prevalent in the newer more complex designs. This may seem basic to many designers, but the idea is to think about what can be done to solve the clearance issues. You need to first look at the space that the part creates between pads. This space may be increased by making the ball pad as small as possible. Always verify your sizes with your assembly team. Next, look at the max via pad size allowed based on minimum spacing. The via pad will dictate the hole size using the minimum annu- lar ring allowed by your vendor. Topside solder- mask tenting of the vias is also suggested but should be approved by your fabrication/assem- bly team. If vias are impossible, a fanout has to be created on the component side using even smaller lines and spaces. You often have to push the envelope of what your board supplier and assemblers want to process as a standard board. Shaughnessy: What do you think is the most im- portant thing to remember when facing tight toler- ances? Gaines: Tight toler ances are caused by the empty spaces created by the part. We design- ers have always been all about using the space available. In today's designs, the space has got- ten many times smaller. You have to change the way you think. The thought can't be that this part does not allow the sizes we have been using. You have to look at what sizes the space will allow. Do not think of what you can't do, DESIGNING WITH FINE LINES AND FEATURES Albert Gaines