We have found that one of the most challenging aspects of achieving great results with our SMT line is the stencil printing process. In other words, great manufacturing heavily depends on the reliable transfer of solder paste onto the PCB copper pads that define component footprints (e.g., ICs, discretes, connectors, etc.).
There's nothing like getting a new set of first article PCBs back with a new stencil and watch the first print go smoothly & observe the board power up without issue. However, until recently, we have been struggling with subsequent prints due to solder paste adhering to our stencil apertures and not wanting to let go, even after cleaning. We have also found that not all pastes are created equal when it comes to tackiness and that pastes that excel in one area (e.g., voiding) may suffer in another (e.g, release).
We're going to update this article over time to cover various stencil issues we encounter, but at this point, we're just summarizing some of our results with stencil cleaning and highlighting the nice results we have been getting recently with Zestron SD301.
After using various readily available stencil cleaners, including 70% and 100% IPA (Isopropyl alcohol), we're thrilled that we finally have something that can get stubborn paste residues to release from our stencil apertures. The Zestron product seems to be a case of getting what you pay for, and we assume that the cleaners from the other top tier cleaning companies also produce excellent chemistries / results.
Regarding IPA, pre-soaked wipes are certainly convenient and seem to do a great job to the naked eye. However, we have found that use of these wipes, especially 70%, seems to create a paste / flux residue that actually becomes tougher to remove after interaction with the combination of IPA and water.
Shown in the two figures below is an extreme example of what we had been facing with small apertures for BTC (Bottom Terminated Component) footprints after numerous prints. As you can see, paste didn't fully transfer from the stencil onto the pads. Instead, paste / residue stuck to the aperture walls. This results in insufficient solder paste on I/O pads, and this leads to opens after reflow. In this extreme case, it's probably obvious that the board required wiping and reprinting. However, consider the case where just one I/O pad is starved solder paste. This can create a very difficult situation of analysis, debug, and rework. Therefore, it's imperative that stencil aperture walls be free of paste / residue between prints and storage.
Stencil Design and Cleaning Guidelines
Probably the two best reference documents for the design and cleaning of stencils are from the IPC ( Institute of Printed Circuits):
* free download
Note that the IPC is a fantastic organization that has many engineering and manufacturing members committed to sharing their expertise to the benefit of the electronics industry.
One of the golden rules in stencil design is adhering to a minimum area ratio, which the IPC defines as the ratio of the area of the aperture opening to the area of the aperture walls. This is best understood with the help of a figure and formula, both of which are adapted from the IPC-7525 reference and shown below.
The area of the BTC I/O stencil aperture shown in Figure 1 is 0.75 mm x 0.2 mm x 0.9 = 0.135 mm2. The area of the aperture walls is 2 x (0.75 mm + 0.2 mm) x 0.120mm = 0.228 mm2. Therefore the area ratio is 0.59, which is below the minimum recommended in IPC-7525. This provides an indication that these apertures are going to be challenging for printing & paste release, even if we can perfect our processes.
Nano coatings should also be mentioned in the context of paste release and stencil cleaning. As you may know, there is great debate in the industry as to which nano coatings are effective and whether they actually coat the aperture walls. Assuming they do indeed improve paste release, it is reasonable to expect that the minimum workable area ratio can be further reduced.
Note that nano coating is an area where we are actively conducting experiments, and we expect to post some results soon.
Zestron SD301 with a challenging test case
The figures below show the before and after of a stencil that we had set aside at the end of its life without a final cleaning. This is representative of a worst case scenario in terms of cleaning requirements. Prior to using Zestron, we had been using a lower cost but more readily available chemistry that required multiple applications and rigorous agitation.
However, for these pictures (~30x magnification), we applied Zestron to a clean room cloth and gently wiped the stencil. We found that we could clean away the stubborn residue with little effort
We'll continue to update this article over time to post other good results we achieve through experimentation that leads to internal process improvements.