The fabrication process is a prime source of contaminants.

Many ask why we need clean bare boards for a “cleaned assembly” process. The answer is most cleaning systems are designed to remove only the easy-to-get-to flux residues. But take, for instance, underneath low-standoff areas, where it is difficult to reach some flux residues. Why compound the issue of high chloride levels from a HASL flux or high sulfate residues from trapped etchant below these components as well?

To better understand what contamination levels must be met, we should understand where contamination resides and from where it comes. Contamination on a bare board resides on and in the following areas:

1. Metal surfaces (pads and plated holes) form thin films of dry residue (including in hole and via barrels). The smaller diameter the via, the more difficult the rinsing of these etch residues. If the via is plugged on one end, this adds a cup effect, making rinsing more difficult (yet still possible) (Figure 1).


2. Laminate surfaces without soldermask. These areas have micropores in the top laminate, the result of copper foil being pressed into the butter coat of the laminate, then etching. This process leaves small (0.001"-0.003") micropores deep in the top surface (Figure 2). These indentations hold residues from the etch, scrub and rinsing fabrication steps prior to soldermask application.


3. Laminate surface on the innerlayers also goes through etch and chemical rinses after stack-ups are drilled and holes plated. Plating residues can wick up inside the fiberglass bundles, getting trapped behind the barrel on loose weaves or when dull drill bits break fiber bundles (Figure 3). High levels of sulfuric acid from etchants can be found on the innerlayers and, should delamination or poor processing occur, can cause leakage and shorting.


4. Most bare boards have liquid photoimageable (LPI) soldermask. LPI epoxy mask has a range of porosity directly related to the mask cure. The more the mask is cured, the tighter it becomes. Likewise, the less the mask is cured, the more porous it will be. Mask cure ranges from just above tacky to hard cured. The problem is the mask must be at the lower to middle end of the cure window so as not to experience problems from over-curing or flaking after three to five soldering conditions. Soldermask with a HASL flux and finish will absorb a large amount of flux during the solder exposure. This is why the post-HASL cleaning processes must use good DI water and an effective soaping agent. SMOBC (soldermask over bare copper) is a good process for minimizing the metal exposure of the plating process, but must be controlled for effective cleaning.

So how clean should a bare board be? Some will appreciate the response, “It depends,” but we found that with multiple localized extractions using an ionic cleanliness tester over plugged vias, fine-pitch pad areas and in soldermask-covered areas, we are able to understand the effects of much of the fabrication residue. Using ion chromatography for the acetate, formate, chloride, bromide, nitrate, sulfate, sodium, potassium and ammonia levels, we can see the process effects and have established limits for good assembly performance based on clean bare boards.

Terry Munson is with Foresite Inc. (residues.com); tm_foresite@residues.com. This column appears monthly.

• Prev
• Next