Glitter iPhone case leaks caused by glue process issues
At the beginning of August, Mixbin recalled over 275,000 liquid glitter iPhone cases after two dozen reported incidents of chemical burns. These plastic cases are like snow globes: they have a sealed enclosure filled with low-viscosity liquid and glitter that shimmers as you move the case. It’s easy to see why they were popular.
Building sealed products is always difficult, but building a sealed product that can stand up to the kind of abuses that phone cases do is another challenge entirely. Phones (and their cases) are commonly exposed to harsh environmental conditions (in direct sun, the cold, altitude) and are subjected to high pressures when put in a back pocket and sat on.
After acquiring a set of the affected cases, we followed our methodology for failure analysis by brainstorming a list of hypotheses that would cause the liquid seals to rupture. We then performed failure analysis and some destructive testing to find evidence to support or refute each hypothesis, using an Instrumental System to capture images along the way.
Potential Root Cause: Loose plug.
The filling hole is sealed with a plug made of a rubbery plastic. The plug fits snugly in the filling hole and is long enough to seat against the bottom of the liquid-filled chamber. The plug is a clear weakness in the case’s design, because it relies only on its tight fit to maintain the seal. Squeezing the glitter chamber near the plug did not seem to dislodge it. Poking the plug with a pair of tweezers did create a small leak, but we don't think this represents a real-world scenario as the phone is usually there to protect the plug from sharp objects. When we actively tried to dislodge the plug, a lot of force was required, which indicates a good "interference fit". While this is certainly a potential root cause, especially if the plug fit to the enclosure is not well-controlled, we didn’t see any evidence of a design or process flaw in the iPhone cases we inspected.
Potential Root Cause: Glue process issue: gaps, bubbles, or contamination.
The glitter chamber is created by a glued ring seal along the perimeter of the case that attaches the back cover to the case enclosure. Glue processes are notoriously difficult to control — they are sensitive to glue age, glue mix ratio, ambient temperature and humidity, curing parameters, surface preparation, dispense thickness, and dispense volume. While most of these affect glue strength, some also affect the glue’s ability to create a seal. For example, when new tubes of glue are added to a glue machine, bubbles or gaps in the dispensed glue are very common. While gaps in the glue path obviously create sealing problems, even small bubbles in the glue may reduce adhesive area and weaken the seal.
Contamination is another obvious issue: if there is a particle or fiber that goes across the glue gap, the glue may not fully seal around it.
For the glitter cases we had, we found evidence of glue process variation. In multiple regions on both iPhone cases we saw glue overflow. On one case, we saw both bubbles and contamination (a piece of glitter) on the sealing surface. The glitter contamination spanned almost the entire seal width, creating a weak spot that could have resulted in failure (in this case, a leak) if this unit had shipped to a customer.
We also investigated the strength of the glue itself. We did this by destructively pulling the back cover off of the case enclosure. We found that the glue was very strong -- so much so that the back cover plastic broke instead of the glue bond itself. This is typical of chemical bonding adhesives -- which were likely used for these cases.
Potential Root Cause: Case ruptures at high pressure.
Although we already uncovered glue process issues as a potential root cause, we continued to investigate our final hypothesis: susceptibility to internal pressure. Since it’s sealed, the case experiences internal static pressure from a displacement of the case walls (such as from sitting on the case when it’s in your back pocket and potentially in extreme cases, pressure from impact). In order to quickly understand this susceptibility, we had several team members of differing weights balance on a pressure concentrator that loaded the glitter chamber, and afterwards checked for leaks. We tested up to 200 lbs with no leaking or visual difference in the seal — a more than reasonable amount of pressure for this product.
As a final effort to understand the limits of the design, we applied a dynamic load of 150 lbs. This created a pressure wave that resulted in glitter chamber rupture (we broke the glue seal). While this test is extreme in terms of the impact applied for a reasonable user scenario, units with compromised seals due to bubbles or glitter may be susceptible to lower levels of dynamic impact such as that created when a user plops down onto a hard chair.
Based on a quick study of two recalled Mixbin glitter cases, we were able to determine several potential weaknesses in the design and process used to create them. In our study of just two units, we found evidence of glue overflow, bubbles, and contamination — which indicates a prevalent process control issue that could produce units that are leaky or could be compromised in a dynamic impact. A compromised unit could either rupture immediately or leak slowly over time, resulting in the kinds of injuries described by the recall
To keep glue processes like this one in control, constant monitoring is required. Several of our customers use Instrumental technology to monitor glue dispense volume, color, placement, and bubbles remotely — if Mixbin had it, they potentially would have discovered the process issues before production, preventing bad units from being made in the first place.
While it seems easy in retrospect, building high quality products is difficult. Instrumental’s technology enables engineers to understand, correct, and monitor variation and defects quickly and remotely, so they can build high-quality products on-time.