That Day, the Team Lead Said 1000 Boards Needed Rework, My Heart Sank
Do you recall a few months ago when a production line suddenly reported an anomaly in a batch of products? At that time, the R&D department's team lead, looking grim, rushed into our reliability lab and immediately asked, "Are those boards faulty? The customer has already returned 200 units, citing intermittent disconnections!" My heart sank upon hearing this, as that batch of boards happened to be our vanguard for a newly introduced process. What was the outcome? The cost of reworking just those 1000 boards was astronomical, not to mention the erosion of customer trust. Upon investigation, it was indeed the same old problem, inextricably linked to "tin whiskers" and "solder ball lift-off."
Where's the Problem? Tin Whiskers and Solder Ball Lift-off, These Two Arch-Nemesis!
Frankly speaking, both issues are related to "soldering." Think about it, all the various components on a PCB are securely held in place and electrically connected by solder.
- Tin Whiskers: These resemble fine hairs, crystalline growths of metallic tin atoms that slowly "grow" from the solder joint surface under certain conditions (temperature, humidity, mechanical stress). These minute tin whiskers are almost invisible to the naked eye, but they are highly conductive! Once they grow long enough to bridge adjacent solder joints or circuits, they cause short circuits, leading to product failure.
- Solder Ball Lift-off/Non-wetting: This is relatively easier to understand. Simply put, it means the solder has not properly "adhered" to the solder pad. This could be due to insufficient soldering temperature or an oxidized pad surface, resulting in poor adhesion between the solder and the metal interface. The outcome is that the solder joint appears present but is actually weak, detaching with a slight touch, leading to open circuits or poor contact.
Therefore, the key point is that both issues directly impact the product's electrical characteristics and mechanical strength, making them absolute killers for reliability.
How is it Actually Done? Rely on FA to Catch the Culprit!
When encountering such problems, our reliability engineers typically employ "Failure Analysis" (FA).
- Visual Inspection: First, inspect with a high-magnification microscope. Don't underestimate this step; often, tin whiskers or obvious solder ball lift-off can be observed. If your product yield has dropped to Cpk 1.08 (corresponding to DPMO 6210), then you can basically catch many anomalies with the naked eye.
- X-ray Inspection: If visual inspection is inconclusive, we use an X-ray machine for transillumination, especially for bottom-soldered components like BGAs. X-rays can reveal the shape of solder balls, presence of voids, and even determine if they have detached.
- Scanning Electron Microscope (SEM) with EDS: This is the ultimate move. SEM provides ultra-high-resolution images, allowing you to clearly see the shape and length of tin whiskers, and even the fracture surface of lifted solder balls. Coupled with EDS (Energy Dispersive X-ray Spectroscopy), it can analyze the elemental composition of tin whiskers or pad surfaces, confirming the presence of oxides or other impurities.
How do you determine if tin whiskers are the "culprit"? If SEM images show tin whiskers bridging to other electrodes, and EDS analysis confirms it is indeed tin, then it's almost certainly the cause!
The Most Common Pitfall: Randomly Altering Processes to Save Money
Frankly, many of these issues often stem from "cost-saving" efforts. I recall one instance where, to reduce costs, the purchasing department introduced a batch of supposedly "cheaper" lead-free solder. The result? Not long after, products began to fail, and customer reported defect rates surged directly to 3%! Upon performing FA, we discovered that tin whiskers were entirely to blame. It turned out that the elemental composition ratio of that solder had subtle differences from what we originally used; although it met specifications, under our specific process parameters, it was particularly prone to tin whisker growth.
Another pitfall is "assuming" soldering temperature makes no difference. An engineer on a certain production line, in a rush to meet capacity, slightly accelerated the reflow oven's temperature profile. As a result, the solder ball's wetting performance deteriorated. The solder solidified before it had fully melted and spread, leading to many solder joints being only "pseudo" connections – they looked fine but detached with a slight shake. All this tells us that any "fine-tuning" of the manufacturing process can lead to significant changes in reliability.
One Thing You Can Do Today
Go back and check your solder specifications and process parameters to ensure they haven't been arbitrarily changed.