We passed every conceivable test-salt spray, thermal shock, and immersion testing in a tank we built in the lab. By testing the waterproof performance of the IP67 pogo pin charging design, I no longer considered the IP67 spring pin charging seal a minor detail, but rather a design issue. Since then, I've delivered multiple products using IP67 pogo pin charging, and the geometry of the seal has been the part I spend the most time on for each product.
What the IP67 Test Doesn't Tell You
IEC 60529 defines IP67 as one meter depth for thirty minutes - about 0.1 bar of hydrostatic pressure. Pass that test and you can put the mark on the box.
What the test doesn't simulate is the product sitting in a gym bag for three years, getting docked wet twice a day, running from 0°C ski resort mornings to 40°C summer runs. The elastomer ages. The gasket takes a permanent compression set. The housing creeps slightly under the sustained clamping load. None of that shows up in the certification test.
The field failures I've seen in IP67 pogo pin charging products weren't design flaws that the test failed to catch - they were degradation modes that the test wasn't designed to detect. So when I think about hitting IP67, I'm not thinking about passing a test. I'm thinking about what the seal needs to look like after 1,000 charge cycles and three years of temperature swings.
Why Pogo Pin Charging Seals Differently Than USB-C
USB-C waterproofing relies on a dynamic seal - a gasket that compresses on plug insertion and decompresses on removal, thousands of times over the product's life. The gasket has to be soft enough to conform to the plug geometry on every insertion and resilient enough to recover its shape on every removal. Getting that balance right, while also keeping the gasket in the receptacle under repeated mechanical cycling, is a real engineering problem. It's why most IP-rated USB-C devices include fine-print warnings against charging while wet: the seal is good when it's working, but the mechanism that makes it work is also the mechanism that eventually degrades it.
Pogo pin charging uses a static seal. The housing is closed at the factory. The contact pads are on the exterior surface, nothing enters the housing during charging, and the seal never opens. The engineering challenge shifts completely - you're not managing a dynamic interface, you're managing a seal that has to survive assembly stress, thermal cycling, and mechanical loads on the housing itself across the full service life.
That shift is why IP67 pogo pin charging interfaces hold up better in the field than equivalently rated USB-C ports, all else being equal. The seal doesn't have a mechanism that cycles with every charge. But it does have failure modes specific to static seals, and they're different from what you'd anticipate coming from a USB-C waterproofing background.
Gasket Material and Channel Geometry
Silicone is the standard material, Shore A 40–60. The lower end of that range is right for small cross-sections where you need the material to fully conform to minor surface irregularities in the PCB; the higher end suits larger cross-sections where you need the gasket to stay in the channel during assembly without being displaced.
The channel depth issue I described above is the most common mistake I see, so I'll go through it once more with the numbers. Target 70–75% of the uncompressed gasket height. If the gasket is 1.2mm in cross-section, the channel depth should be 0.84–0.90mm. That gap - 0.30–0.36mm of free volume - allows the silicone to flow laterally under compression, which is what creates uniform seating pressure around the full perimeter. At exactly 1.2mm channel depth, the material can't flow, the gasket rocks, and you get the corner failures I described.
Compression set is the long-term concern with silicone. Over time and thermal cycling, a compressed silicone gasket doesn't fully recover to its original height. If the initial compression is 25% and the material takes a permanent set of 10% of its original height over three years, you're now at 15% compression - which is below t
he reliable threshold. This is why I size initial compression at the upper end of the range (25–30%) for products expected to last beyond two years.
EPDM comes up on products with aggressive chemical exposure. I specified it on a clinical handheld that got wiped down with quaternary ammonium disinfectant multiple times a day - silicone's surface degrades faster than EPDM in that environment. But on that same product, the compression set accumulated faster than on the equivalent silicone design, and we ended up revising the channel depth spec after 18 months of accelerated aging data came in. If you're considering EPDM, get aging data specific to your cleaning agents before committing to the seal geometry.
PCB Pad Layout
Hard gold plating on the contact pads is non-negotiable for IP67 pogo pin charging applications. Soft gold degrades noticeably in salt exposure environments; the surface pitting you get in sweat or salt fog reduces contact reliability before you'd see any mechanical wear. Minimum 0.5μm cobalt-hardened gold over 2–5μm electroless nickel. The nickel matters because it provides a corrosion barrier if the gold layer eventually wears through - without it, you're relying on the gold alone across the full service life.
The keepout zone around the pad array needs to be flat, featureless PCB laminate for at least 0.5mm inside the gasket channel inner wall. The gasket seats against that surface, and any topology in that zone - trace step, via barrel, soldermask edge - creates a potential leak path. This produces the most friction in PCB layout reviews because it eats into the routing area around the pad array, but there's no good workaround.
Via placement inside the gasket footprint is where I've traced two separate IP67 failures in production designs. Both had tented vias inside the gasket perimeter that passed DRC without comment. Tented vias aren't sealed - the barrel plating creates a capillary path that wicks water through the board under sustained immersion. On the first failure, we didn't figure out the failure path until we ran a dye penetration test on a second-pass build with the dye applied on the exterior surface. The dye showed up on the interior side at the via locations within four minutes. On the second failure, we'd been warned by the first and caught it during layout review. But it took having a specific rule - "no vias inside the gasket footprint, period, not even tented" - to make it actually stick across different layouts and different layout engineers.
Fill-and-cap per IPC-4761 Type VII is the specification if routing constraints leave you no choice but to put vias in that zone. Tented isn't sufficient.
Pogo Cable vs. Charging Dock
The sealing problem on the device side is identical either way. What differs is the charger.
A charging dock lives on a desk or nightstand. It's rigid, stationary, and the connector geometry is fixed. If the spring contact force and pad alignment work, the IP67 obligation is entirely on the device. The dock itself rarely needs an IP rating.
A pogo cable carries the spring pin heads on a flexible cable end, which creates a fundamentally different engineering problem. The cable-to-connector-head interface has to seal against water ingress while also surviving repeated bending. That's the failure point in almost every waterproof pogo cable assembly I've reviewed - not the connector head itself, not the cable jacket, but the transition between them.
Polyurethane overmold is the right material for flex-fatigue resistance at that transition. The overmold needs to extend at least 30mm from the connector body to distribute the bending stress over enough length to prevent the interface from opening under flex cycling. I've seen cable pogo assemblies built to 15mm overmold that passed initial IP testing and started leaking at the interface around 600–800 flex cycles. 30mm is where that failure mode goes away in the designs I've validated.
If your pogo cable needs to carry its own IP67 rating - swim tracker, outdoor equipment - specify the IP test with the connector end sealed exactly as it will be in use, not with end caps. End caps hold the connector head rigid during the test and don't load the jacket-to-overmold interface the way actual use does. The failure mode you're trying to catch won't show up in an end-cap test.
Spring Force and Housing Stiffness
This relationship doesn't come up in most design guides because it's slightly indirect, but I've seen it contribute to IP67 failures.
A 4-pin pogo pin charging array at 150gf per pin puts 600gf of compressive load on the housing face at the pad area every time the device docks. In thick-walled housings - injection-molded polycarbonate at 2mm or more - this is a non-issue. The housing doesn't deflect measurably under that load.
In thin-walled wearable housings where the pad area wall might be 0.8–1.0mm thick, 600gf causes enough deflection that the gasket compression varies around the perimet
er - higher on the side closest to the dock contact, lower on the opposite side. This doesn't show up in static immersion testing because the dock isn't connected. It shows up in the combined docking-cycle-plus-immersion test, which is why I run that sequence before submitting for certification.
Stiffening ribs or a metal inset behind the pad area, targeting less than 0.05mm deflection under maximum spring load, are the standard fixes. Which one is right depends on the housing geometry and whether the product roadmap can accommodate a metal inset in the tool.
Validation Before Certification
The sequence I use in internal testing before the formal IEC 60529 immersion:
Three thermal cycles, –20°C to 60°C, before any immersion. This is the test that catches gasket material issues early - silicone with poor compression set recovery will show up here because the seal performance degrades between the cold soak and the room-temperature immersion that follows. Products that fail this test would have shown up eventually in field returns; catching it in prototypes costs nothing to fix.
Then 1,000 docking events, then immersion. The docking cycle test loads the housing face with the spring contact force repeatedly and checks whether the gasket seal degrades under that cumulative load. For most consumer wearables, 1,000 cycles approximates two to three years of daily charging.
Finally, immersion with the charging dock connected. Most teams test unplugged and call it done, but for products where users might submerge the device while it's sitting on a charging dock or connected to a pogo cable - outdoor sports devices are the common case - testing connected is the only way to catch the load-distribution effect I described in the spring force section.
These three run in this order because each one finds failure modes the others don't, and running thermal cycles last would mean running the formal test on a seal that's already stressed. The formal IEC 60529 test goes on an unthermal-cycled unit for certification purposes, but I want to know the thermal cycling result before submitting.
FAQ
Does the IP67 rating cover the pogo pin pads themselves, or just the housing seal?
The rating covers the complete device. The pads are on the exterior surface and exposed to immersion water, but they don't create an opening into the housing interior - there's no cavity, no port, nothing for water to enter through. So the pad surface gets wet and that's fine, as long as the gasket seal behind it holds.
The practical consequence: your gold plating spec affects pad durability in wet environments, but it doesn't determine whether the device passes IP67. That's entirely a function of the gasket seal geometry.
What's the smallest pin pitch that still allows a viable seal geometry?
The binding constraint is the keepout zone - 0.5mm minimum of flat laminate between the outer pad edge and the inner gasket channel wall. On a 1.0mm pitch 4-pin array, that's manageable. At 0.5mm pitch with many pins, the array footprint grows until it starts to conflict with housing wall thickness constraints.
The tightest pitch I've successfully run in a production IP67 pogo pin charging design is 0.8mm on a 6-pin array. Below that I'd want to revisit whether the pin count is actually necessary before trying to make the seal geometry work around it.
How do I write a spec for a waterproof pogo cable assembly?
Define the IP rating for the cable assembly separately from the device rating. Specify: polyurethane overmold (not nylon), minimum 30mm overmold extension from the connector body, cable jacket material and wall thickness for the depth requirement, and a pull-force specification for the cable-to-connector-head bond.
The testing specification matters as much as the material spec. Require IEC 60529 testing with the connector sealed as it would be in actual use, not with end caps. And require a flex cycle test at the connector entry - minimum 500 cycles at the intended bend radius before the IP test, so you're not shipping an assembly that passes certification and starts leaking in the field after a few weeks of use.
Does using a magnetic pogo pin charging system make IP67 harder to achieve?
Mechanically, no - the magnet is inside the housing and doesn't affect the seal. There's actually a minor benefit: the magnetic alignment reduces the variability in pin contact force, which reduces the variability in compressive load on the housing face, which means more consistent gasket compression. That's a second-order effect but a real one.
Where magnets require deliberate attention is sensor placement. A ring magnet for charging alignment creates a field that extends into the device interior, and if the layout has a compass, Hall-effect switch, or NFC antenna, that field can interfere. The pad array position and magnet geometry need to be determined in the same step as the sensor placement, not separately. The keepout between the magnet center and a s
ensitive sensor is typically 8–12mm, depending on magnet grade and sensor threshold.
I hit IP67 on bench tests but failed in certification. What am I likely missing?
Nine times out of ten in my experience: via placement, gasket channel depth, or test sequence.
Check the layout for any vias inside the gasket footprint. Run a dye penetration test - dye on the exterior, device submerged briefly, look for dye on the interior near the pad area. If dye appears at via locations, that's your failure path.
If the layout is clean, check the channel depth. Measure it with a depth micrometer, not from the CAD model - manufacturing tolerances on small channel depths drift more than you'd expect. If the channel depth is within 5% of the uncompressed gasket height, the gasket likely isn't seating under full compression.
If both of those look right, check whether your internal testing included thermal cycling before immersion. Room-temperature bench tests can pass a seal that fails after thermal excursion, and the certification lab typically does run a combined sequence.
Questions About Your Specific Design?
If you're working through an IP67 pogo pin charging interface problem - seal geometry, pad layout, pogo cable specification, or test sequence - and want a technical review, contact us at [xt@xtpogopin.com]. We've supported this design problem across wearable, medical, and industrial applications, and if we've seen your failure mode before, we'll tell you what fixed it.
Send your pad layout and housing cross-section to [xt@xtpogopin.com] and we'll respond with specific observations, not a sales call.
If you need sample pogo cable hardware or pogo pin charging dock configurations for IP67 prototype builds, request here: xt@xtpogopin.com
If you've run into an IP67 pogo pin charging problem that the guide doesn't cover - or if something here contradicts your experience - I'd like to hear it. Leave a comment or email directly.
