I reviewed the warranty data for a wrist-worn fitness tracker and analyzed the issues reported across all returned devices. The contact resistance of the pogo pin contacts increased from the nominal 25 mΩ to 180 mΩ or even became an open circuit.
We went after the pogo pins in the dock first. That was the wrong assumption. When we pulled devices and tested the contact pads with XRF, most of the gold was gone - what remained was oxidized nickel. The pogo pin pad area on the device had been finished with the same ENIG process as the rest of the PCB: electroless nickel immersion gold, with about 0.07μm of gold. I hadn't questioned it during design review. ENIG exists to protect solderable surfaces through the assembly process, not to survive mechanical contact cycling. At roughly 200–400 dock-and-charge cycles, that 0.07μm gold layer was gone.
The fix was a dedicated cobalt-hardened electroplated gold on the pad area - 0.5μm minimum over 3μm electroless nickel, specified separately from the board surface finish. The units built to that specification haven't come back with that failure mode. I've been deliberate about pogo pad plating specification ever since.
What "Hard" and "Soft" Actually Mean
The terminology is confusing because neither term refers to physical processing or thickness. The difference is alloy composition.
Soft gold is pure gold - 99.9% Au or higher. The material is inherently soft because elemental gold is inherently soft. Knoop hardness for electroplated soft gold is approximately 60–90 KHN, depending on bath chemistry and deposit conditions.
Hard gold is a gold alloy, almost always gold-cobalt at around 0.1–0.3% cobalt by weight. Occasionally you'll see gold-nickel alloy, but cobalt is more common in connector applications. The cobalt precipitates at grain boundaries during deposition, interrupting grain growth and producing a harder, finer-grained deposit. Cobalt-hardened gold runs 130–200 KHN per the AMS 2422 specification - roughly twice the hardness of soft gold, sometimes more.
Wear Life: The Data and What It Actually Tells You
For a pogo pin pad, the failure sequence is almost always the same: the spring-loaded pin tip abrades the gold layer over repeated contact cycles; the gold wears through to the nickel underlayer; the nickel oxidizes; contact resistance climbs. That's what happened in the wristband product above, just faster than expected because the gold was far too thin for the application.
On a pogo pin pad with cobalt-hardened gold at 0.5μm, wear-through under 100–200gf spring contact force typically runs somewhere in the 10,000–50,000 cycle range. I want to be honest that this range is wide, and the spread is real - pin tip geometry matters significantly. A rounded or crown-tipped pogo pin distributes contact stress over a larger area and extends pad life compared to a flat or chisel tip under the same load. In our own cycling tests with 150gf and a crown-tip pin, we got to around 35,000 cycles on 0.5μm cobalt gold before seeing measurable resistance increase. With a flat tip at the same force, the number dropped closer to 8,000. If you're specifying a pogo pin pad and don't know what pin geometry your dock will use, the wide range is appropriate - narrow it when you have that information.
Soft gold at 0.5μm wears considerably faster. Our testing put soft gold pad life at roughly 20–30% of equivalent cobalt-hardened gold at the same thickness and spring force. That's not a precise number and I wouldn't bet a product design on it without running your own tests - the ratio shifts with contact force, cycle rate, environment, and pin geometry. But the directional conclusion is consistent: for any application with meaningful cycle counts, soft gold requires either significantly thicker deposit or more frequent replacement.
The cycle count threshold where soft gold becomes viable is genuinely low - production test fixtures that see 50 contact cycles before replacement, single-use diagnostic accessories, temporary sensor pads in clinical settings. Below a few hundred cycles, the wear resistance advantage of hard gold is real but mostly unused, and you're paying for capability you won't exercise.
The Nickel Underlayer
Pogo pad plating specifications frequently focus on gold type and thickness and underspecify the nickel underlayer. The nickel does two things: it acts as a diffusion barrier preventing gold from migrating into the copper substrate at elevated temperatures, and it provides corrosion resistance when the gold eventually wears through.
Gold diffuses into copper. Without a barrier layer, repeated thermal cycling gradually depletes the gold from the copper-side of the deposit, thinning the effective gold layer faster than mechanical wear alone would suggest. Electroless nickel in the 2–5μm range, per IPC-4552 process guidelines, is the standard for connector-grade pogo pad applications. Below 2μm, pinhole coverage in the nickel layer creates sites where copper can corrode once the gold wears or is breached. Above 5μm, stress in the nickel deposit begins to affect adhesion at the gold-nickel interface.
The other issue with nickel is that nickel oxide is a poor electrical conductor. Once the gold layer on a pogo pin pad wears through and the nickel surface is exposed to humidity or oxidizing atmosphere - exactly the conditions a wrist-worn device sees - the nickel oxidizes and contact resistance rises further, even without additional mechanical wear. This is why pad wear-through is typically a non-recoverable failure. You can't clean your way back to acceptable contact resistance once the nickel is exposed and oxidized.
Thickness by Application
Consumer wearables charging daily - smartwatches, fitness trackers, earbuds - accumulate 300–600 dock cycles per year. Over a three-to-four year product life, that's 1,000–2,500 total cycles. Cobalt-hardened gold at 0.3–0.5μm over 3μm electroless nickel covers this range. I've specified 0.3μm and had it hold up through product lifecycle in clean indoor environments. Sweat exposure changes the calculation - salt accelerates the corrosion process once any gold wear occurs, so if the pad area sees skin contact regularly, 0.5μm gives you a meaningful additional margin.
Medical and industrial devices with higher-frequency contact cycles - diagnostic handhelds docking multiple times per day, laboratory equipment, test station docking interfaces - warrant 0.5–1.0μm cobalt-hardened gold over 3–5μm nickel. Products that get wiped with isopropyl alcohol or quaternary ammonium disinfectants between uses benefit from the thicker gold because chemical cleaning agents can accelerate surface micro-attack even on gold, and the thicker deposit takes longer to reach wear-through. I've seen pogo pin contact pads on clinical handhelds at 0.5μm look visibly degraded at 18 months of daily use; the same geometry at 0.8μm was still performing within specification at that point.
Production test fixtures are different. The cycle counts are high, but the fixture is in a controlled environment, inspectable, and replaceable as a maintenance item. 0.5μm hard gold at 3μm nickel is the common specification, and many fixture operators schedule pad replacement at a cycle count threshold rather than running to failure.
When to Specify Soft Gold
Solderability is the main case. Cobalt-hardened gold creates problems at the solder joint - the cobalt interferes with flux chemistry and wets poorly compared to soft gold or ENIG. If your pogo pin pad needs to be soldered during assembly rather than exclusively used as a contact surface, soft gold or ENIG is right for the soldering requirement. The tradeoff is contact life if the pad also sees repeated cycling, which is a real problem for some dual-purpose designs.
Very low cycle count, cost-sensitive applications are the other case. If the pad will genuinely see fewer than 200 contact cycles over its entire life, soft gold at adequate thickness does the job.
One argument for soft gold I've pushed back on more than once: "we only need to survive lab testing, maybe 50 cycles, before production commits to a spec." Testing pads see more cycles than that faster than expected - once customers get their hands on engineering samples, 50 becomes several hundred. I've stopped specifying soft gold for engineering samples unless the project has a specific cost constraint at that stage.
Verifying the Specification
Gold plating specs are easy to write on a drawing and easy for a supplier to miss in production without anyone catching it.
XRF (X-ray fluorescence) is the standard non-destructive method for gold layer thickness verification on finished pogo pin pads. Require XRF data at incoming inspection - actual measurements on parts from the lot, not bath records from the plating tank. If a supplier can't provide this, their process control is incomplete.
Knoop or Vickers microhardness testing on a plating cross-section coupon verifies that the gold alloy is actually cobalt-hardened. The hardness ranges don't overlap: soft gold (60–90 KHN) versus cobalt-hardened hard gold (130–200 KHN). A single measurement from a test coupon pulled from the plating batch is conclusive. The test is destructive, so it's done on coupons rather than finished parts. Any serious electroplating supplier runs this as part of bath quality control; ask for the data as part of your first-article inspection package.
Salt spray per ASTM B117 or IEC 60068-2-11 is relevant for pads in corrosive environments. Gold doesn't corrode, so a salt spray test run to wear-through will show you how quickly the underlying nickel degrades in the environmental conditions you're testing.
Frequently Asked Questions
My PCB house says they can do "connector gold" on the pad areas. Is that the same thing as specifying cobalt-hardened gold separately?
Not necessarily, and the distinction matters. Many PCB fabricators offer connector gold as an add-on, but without explicit specification of alloy type, hardness, and nickel underlayer thickness, what you get varies. Some shops run true cobalt-hardened electroplated gold; others plate soft gold to connector thickness. Ask specifically for the cobalt content, hardness (KHN) from their most recent bath testing, and XRF data on sample parts. I've had PCB-integrated pogo pin pad gold come in under-spec by 30–40% on thickness in incoming inspection. It's not caught without measurement.
Does gold thickness affect initial contact resistance, or just long-term performance?
Long-term performance is where it shows up. Gold in the 0.3–1.0μm thickness range presents essentially the same noble metal surface to the contacting pin, so initial contact resistance readings don't distinguish between them. If you're measuring elevated contact resistance on new pogo pin pads, the plating specification is probably not the cause - check surface contamination, pin tip geometry, and spring force before going to the plating spec.
How do I write the plating spec on a drawing correctly?
Minimum: gold alloy type, minimum thickness, and nickel underlayer minimum thickness with a normative reference. A complete callout reads: "Au (Co-hardened) min. 0.5μm over EN min. 3.0μm per IPC-4552 Class B." The standard reference matters because it defines how the supplier verifies and certifies the plating - "hard gold 0.5μm" without a normative reference leaves the test method unspecified and gives you nothing to audit against. For automotive or aerospace traceability requirements, AMS 2422 maps to specific gold purity and hardness requirements that IPC-4552 doesn't cover.
We're designing a pogo pin contact pad for a wearable medical device. What changes?
The material specification - cobalt-hardened gold over electroless nickel - stays the same. What changes is the verification and traceability structure around it. Under ISO 13485, electroplating is a special process, which means the supplier needs to be qualified, and they need to maintain process records you can audit, not just provide final inspection data. Your design history file needs the supplier qualification documentation, which is different from what's required for a consumer product.
Biocompatibility is worth thinking through if there's any path for the pogo pin pad area to contact skin. Gold is biocompatible; nickel is a known sensitizer and allergen in a significant portion of the population. If your mechanical layout routes the pad area anywhere near skin contact, either specify thicker gold to delay wear-through as long as possible, or add a palladium intermediate layer between the nickel and gold - palladium provides a supplementary diffusion barrier and doesn't raise the same biocompatibility concerns as nickel. Most wearable medical pogo pad designs route the pad away from skin contact specifically to avoid this question, and that's worth deciding early in the mechanical layout rather than retrofitting later.
How often should incoming inspection include hardness verification vs. just XRF?
XRF every lot, hardness from a coupon on the first article and then at whatever frequency your quality plan requires for ongoing process validation - typically quarterly for a supplier with a stable, documented process and a track record of conforming lots, more frequently if there's any history of bath chemistry variation. I'd also run hardness verification any time the supplier changes their plating bath chemistry or replenishes with a new chemical lot, since bath composition directly affects the cobalt incorporation rate and therefore the hardness. Getting notified of those changes is worth putting in the supply agreement.
Contact Us About Your Pogo Pin Pad Specification
Getting the gold specification right for a pogo pin pad usually comes down to three things: knowing the actual cycle count, understanding the environmental exposure, and checking what the supplier delivers against what you specified. If you're working through a specification decision or debugging a contact resistance problem in the field, the engineering team at [Dongguan Xinteng Electronics Co., Ltd.] has seen a lot of these failure modes and can usually tell you quickly whether the plating is likely the cause.
Reach us at [xt@xtpogopin.com] with your application details and current specification. If you need sample pogo pin pads in different hard gold and soft gold configurations for comparison testing, we can supply them - [pogo pin contact pad].
