How To Determine Stencil Thickness

Phil
Welcome to Board Talk. This is Jim Hall and Phil Zarrow of ITM Consulting, where here to answer your questions regarding SMT process and assembly issues.

Today's question is a good one. On what basis is stencil thickness decided in the case of BGA's or fine pitch components?

Jim
Most of us are aware of the issues of the different size and style of components that need to be processed on the same board, and the need to have different volumes of solder paste for different components, and related to using different thicknesses of stencils.

Some people call it mixed print or broadband. There are a number of names for it. Okay, I've got particular components. What thickness stencil should I use if I have a 50 mil pitch BGA? What thickness stencil should I use if I have a 0.4 millimeter CSP?

As we're designing apertures, we have to calculate aspect ratios and so forth to make sure we get good printing. As you really analyze this topic and start to think about calculations and deciding on stencil thicknesses and apertures, the most obvious thing that comes out in terms of the data that you need to make this analysis is the volume of solder paste required for a specific joint.

If I have a 50 mil pitch QFP, and I have a particular pad geometry, what is the volume of solder paste, or the range of the volume of solder paste, that will give me an acceptable joint?

From the qualitative side, I know if I use this stencil too thin, I won't get enough paste and I won't have good fillets. Likewise, on a small fine pitch part, if I use a stencil that's too thick, I will get too much paste.

We talk about aperture reduction and area and aspect ratios based upon the printing parameters. But there's the other issue of how much do I need to reduce the aperture to give me the proper volume of solder paste.

It comes back to what do you need in terms of the volume of solder paste. And to be quite frank, is there a source that can supply that answer?

I've talked to a number of people in the industry and when we finally get around to hemming and hawing and talking about it, nobody can provide any answers. I have looked at some IPC specs and stencil design and I honestly don't remember the number off the top of my head, but it talked about an aperture for using a 6 mil stencil and a 4 mil stencil.

I calculated the ratio between those two stencils, both of which they were proposing as acceptable. It was over 2:1 in terms of the volume that we'd get from the different stencil thicknesses and aperture design.

My feeling was, again, it's only qualitative, it seems too big a range, and keeps begging the question, what is the appropriate volume for a given components and given pad sizes.

Phil
So we're all looking for a holy grail of calculator that will automatically calculate for a given part what that volume should be.

Jim
If there is a spec out there, IPC or anybody else that we don't know about, we would love to hear from you. We plead ignorance. But as I said, we've talked to a number of pretty knowledgeable people and nobody can answer it.

Phil
Maybe if we get desperate enough, Dr. Ron Lasky can come to our rescue and come up with the volume coach. You hear that, Ron?

Jim
Okay, I've ranted enough.

Phil
On that note, this is Jim Hall and Phil Zarrow saying don't solder like my brother.

While Stencil aperture size and volume is critical, looking at the ball coplanarity is critical also. If you have a BGA with a 7mil coplanarity and others with just 4, stepping up you stencil on the 7 mil part can significantly reduce issues. You want your ball to be in the paste during the flux activation phase of your paste other wise you may end up with HIP.

Pulling the part data sheets and looking at the dimensions for the package before you design your stencil can save you lots of rework or scrap.

There are software tools to help here, valor mss process preparation has one 'stencil optimizer'. It can calculate every aperture in a stencil based on link to cad package to warn if any are under your release ratio defined by target stencil thickness and min release ratio.

If any parts - apertures fall out side this you can take action just on the offending area by local resizing, apertures replacement etc, or reduce thickness and run again...

Work closely with your printer and stencil suppliers. There are a variety of newer technologies that have been developed to deal with the increasing disparity between paste deposit sizes caused by reduced pitch, smaller pads, and the tighter placement of components.

Active/energized squeegee systems, new stencil materials, stencil coatings and stencil-free printing technologies are some of the ways to improve your process and go beyond the IPC standard.

We have converted IPC 7525 for stencil thickness determination into a regression equation. Stencil Thickness = 2.64 + 0.0831 * pitch of component. Take average of all stencil thickness derived for all components, if min. to max. difference is more than 2mil, consider a step-up/step-down stencil in selective locations.