Issue link: https://iconnect007.uberflip.com/i/1207026
FEBRUARY 2020 I SMT007 MAGAZINE 31 Matties: With the finer pitches, do you think that there's going to be a natural or inherent movement to shorts versus opens because of the nature of proximity? Prasad: There will be, but what people do is strange. They will still have more opens even in finer pitches because when they see bridges, they take steps to reduce the paste volume and thickness of the stencil. People don't want to see the bridges because it's obvious. To reduce bridges, they redo their stencil design (stencil thickness, stencil aperture, or both) because they don't want to see bridges that are easily visible to everyone. To answer your question, with reduced pitch, there will be more bridges, of course, but peo- ple would modify their stencil and printing pro- cess and end up with insufficient solder. This leads to more opens instead of more bridges, as you might expect. Matties: Even though it's closer, the tendency is to apply less or thinner. Prasad: And that's why the defect level with finer pitch is 10,000+ parts per million when you are using less paste (there are other rea- sons, such as lead fragility). Matties: How does the solder paste manufac- turer fit into this? Prasad: For paste printing, there are many guidelines. One of them is that the ratio of stencil aperture to solder powder should be 4:3 or more. In other words, more than four sol- der balls should pass through the stencil aper- ture so that there is no clogging. There is also a guideline for the stencil aperture width. The aperture width is roughly half the pitch. Gen- erally, when the pitches are lower, you make it more than half the pitch. For 16-mil pitch, you may want stencil aperture to be 10 mils. Smaller pitches need smaller powder sizes. However, the smaller the size of the powder, the higher the exposed surface area of solder powder. More surface area means more oxida- tion, and if there is more oxidation, you have more solder balling. For smaller powder sizes, you need an inert environment like nitrogen. Make those powders in an inert environment and ensure they don't oxidize because the smaller size is more susceptible to oxidation. The paste manufacturers have a huge role, especially paste for fine pitch. Matties: The construction geometries of the stencil will play a role in this too. Prasad: Yes, you need thinner stencils for finer pitches. There's a limit to how thin it can be, depending on the metal. When some people want to make two- or three-mil stencils, it becomes very hard, and they try to do it with a Kapton tape. Also, with the electroformed stencil, which is plated stencil, that's where it makes sense for fine pitch. Electroformed sten- cils are not rolled steel, so they are not going to be as strong, and the cost is five times greater than normal stencils, which cost $200–300; electroformed stencils will cost $800–900. You need the electroformed stencils with their very smooth surface so that nothing gets stuck to the sidewalls, and all of what you're depositing needs to go fully on the board when you're lifting the stencil. If the surface is not very smooth, some of that paste will come back with the stencil when it's being lifted, and then you have insufficient solder. Matties: When you're talking about lack of sol- der paste, are these going through an inspec- tion process in SPI? Prasad: They are, but there's a big range when it comes to paste coverage—from 65% to 125– 225% of the pad area that's covered—and whatever the height is, so they need to do that. Matties: My point is if it's going to SPI, you're going to see if there's not enough solder or paste rather before it winds up as a field fail- ure. Prasad: That's what you need to do, and you have to set those standards on the higher end (closer to 100% pad coverage) for acceptance