5
\$\begingroup\$

I intend to inspect hand soldered PCBs with fine very pitch SMD components to check for dry joints and any other possible defects.

I will use a stereo microscope with total magnifications from 5x to 80x. Fluorescent/LED ring lights and halogen/LED spot lights are available for it. RGB car headlight LED rings could also be easily adapted. Oddly, none have filters available. I would have thought different wavelengths of illumination would be particularly valuable? Perhaps also polarising filters?

Goals would be to, for example: exaggerate differences between: flux, solder, solder mask, PCB tracks, device leads (and more?). To see through flux reflections to what lies underneath. To differentiate solder states. Spot gaps (perhaps with shadows), weakened pins, lifted pads or broken tracks.

Are such techniques used? If so, what specific spectra or other filters etc are useful? Might variable RGB LED lights be helpful, or counterproductive? Do any other non-destructive techniques besides electrical continuity checking help achieve these goals?

\$\endgroup\$
1
\$\begingroup\$

Typical conformal coating contains a fluorescent dye so that the film will glow under UV light.

Fluorescent conformal coating on PCB

Common blue LEDs (such as those found in RGB) will also provoke the dye to glow. The downside is that there is much more visible light bouncing off of non-glowing surfaces.

\$\endgroup\$
  • \$\begingroup\$ This is interesting, but doesn't answer the question. \$\endgroup\$ – Olin Lathrop Aug 19 '17 at 16:03
  • 2
    \$\begingroup\$ @OlinLathrop - OP asked about different wavelengths being valuable, and also stated a goal of showing different materials, and also asked "are such techniques used?" I explained that conformal coating contains a dye so that it can be highlighted with certain wavelengths of light. \$\endgroup\$ – Bort Aug 19 '17 at 16:33
1
\$\begingroup\$

I have been building and testing circuit boards for 40+ years and the color of the light is not something I have ever seen adjusted.

First I suggest cleaning board so any flux is removed, this helps with inspection. And bright white lighting helps the eye pick up any color differences. The human eye / brain is a difficult act to follow, give them lots of light at all wavelengths.

I have worked with AOI machine and light color might help with one component but not another. The actual images are made with bright white light and digital filters are applied to the resulting images.

The two problem parts are QFN (no leads) and QFP leads. For the QFP i like using a small steel needle point probe and push each lead. You should see some spring in the steel probe and this helps you regulate the force you are applying. A long sharp XActo Knife blade also works well. A poor solder joint will allow the lead to slide sideways, you can feel it with experience.

For QFN parts the real task is in PCB pad design and solder paste deposits. Pin pads should extend .5mm to 1mm outside the chip body as this area serves as a "Reservoir Tip" as the part settles down during reflow. During placement (by machine) the center tab paste provides support that prevents the placement machine from squishing the outside pad deposits. Placement of QFNs by hand is difficult.

The other parts are easily checked except for BGAs. In general BGAs are not difficult just be sure to ove the reflow temps well into melting points. Low temps just above melting will give you "Head In Pillow" opens.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.