Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Use Conformal Coating to Protect Boards
Transcript
- Lucas: You design a board, it works perfectly in the lab, you ship a hundred units, and then three weeks later customers start reporting random failures. No obvious cause. This is the single most common field failure pattern I hear from hardware founders, and nine times out of ten the root cause is something they could have prevented for about fifty cents per board. Luna: Fifty cents? That sounds too cheap to be real. What are we talking about? Lucas: Conformal coating. It's a thin polymer layer — typically twenty-five to seventy-five microns thick — that you apply over the assembled printed circuit board. It seals the components and solder joints against moisture, dust, salt spray, and even fungal growth. And yet I'd say maybe one in three early-stage hardware startups bothers with it. Luna: Why would they skip it if it's that cheap and solves such a common problem? Lucas: Two reasons. First, it adds a step to the assembly process, which can feel like overhead when you're trying to ship fast. Second, the cost isn't zero — it's about fifty cents in material and maybe another dollar in labor — but when you're scaling from a hundred units to ten thousand, that adds up. Founders think, 'My product lives indoors, it'll be fine.' And then the first batch goes to a customer in Florida or Singapore, and humidity gets inside. Luna: I actually have a story for that. There's a small drone startup I followed — they were making agricultural survey drones. Their first production run of about two hundred flight controllers had an eighteen percent failure rate after three months. They tested everything — components, solder joints, firmware — and eventually traced it to condensation inside the drone's housing during early morning flights. No conformal coating. Lucas: Classic case. And the fix — applying a urethane-based coating — cost them under a dollar per board. They lost almost forty thousand dollars in replacements and shipping before they figured it out. That's a forty-to-one return on a coating investment. Luna: So what are the options? I know there are different chemistries. Lucas: Three main ones in the hardware startup world. Acrylic — it's the most common, goes on easily, dries fast, and crucially, it's reworkable. If you need to desolder a component, you can dissolve acrylic coating with a solvent, do your repair, and reapply. The trade-off is it doesn't hold up as well against strong chemicals or extreme temperatures. Luna: So acrylic is good for prototyping and low-to-mid volume where you expect to make changes? Lucas: Exactly. Then you have urethane. It's harder, more abrasion resistant, and better against solvents and humidity. But it's a pain to rework — you basically have to burn through it with a soldering iron or use aggressive strippers that can damage the board. Urethane is what you'd pick for a product that's going into a harsh environment and won't need modifications. Luna: And the third? Lucas: Silicone. It handles extreme temperatures — think engine compartments or outdoor sensors in Arizona summers. It's also very flexible, so it works well on boards that flex or vibrate. Downside: it's thick, takes longer to cure, and removal is a nightmare. You almost never want silicone if you anticipate any rework. It's a 'seal it and forget it' material. Luna: Application method matters too, right? You can't just pour it on. Lucas: Right. At prototype scale, you can use aerosol spray cans — they cost about fifteen dollars each and cover maybe ten to twenty boards. It's fine for fifty units. But the coverage is uneven, and you have to mask off connectors, switches, and any components that need to stay exposed, like pressure sensors or cameras. Luna: Masking sounds tedious. What about dipping? Lucas: Dip coating is faster — you submerge the whole board into a tank of liquid coating and then pull it out slowly. The thickness is controlled by the withdrawal speed and the viscosity. It's great for volume, but you still need to mask, and the coating gets on both sides, which you don't always want. Plus, the tank has a limited pot life — once you open the container, the solvent starts evaporating, and the viscosity drifts. Luna: So what's the gold standard for a startup scaling from hundreds to thousands? Lucas: Selective robotic coating. You program a robot arm with a spray nozzle to apply coating only where you want it. It's precise, repeatable, and you don't need to mask most components because the robot can avoid them. The capital cost is significant — entry-level systems start around thirty thousand dollars — but you can also outsource to a contract manufacturer that offers selective coating as a service. Many EMS providers will do it for a small per-board fee. Luna: And I assume you need to inspect the coating afterward to make sure it's done right? Lucas: Absolutely. Most conformal coatings have a UV tracer added. You shine a black light on the board, and any coated area fluoresces bright blue or green. It's a quick visual check. More advanced shops use an automated optical inspection system that looks for bubbles, thin spots, or missing coverage. Bubbles are a common defect — air gets trapped during spraying, and that creates a pinhole where moisture can sneak in. Luna: So let's say I'm a founder about to launch a new IoT sensor that will sit outside. What's your practical advice? Lucas: Start with acrylic spray for your first two hundred units. Keep good records of which boards are coated. As soon as you see any field returns, examine the coating under UV. If you're getting failures at connectors or uncoated areas, adjust your masking. By the time you're at a thousand units, switch to selective robotic coating with urethane — it's the sweet spot for outdoor electronics. And budget for it from day one. That fifty cents per board is not a cost; it's an insurance premium against a much bigger expense. Luna: It's one of those things where skipping it feels like saving money but actually costs more in the long run. Kind of like when we talk about design for assembly — the upfront investment pays for itself. Lucas: Exactly. And honestly, if these kinds of practical trade-offs are useful for what you're building or running, it's worth mentioning that listener support is what keeps this show ad-free and focused on exactly these decisions. People who get value from the show can support it at buy me a coffee dot com slash fexingo — no pressure, just a way to help us keep doing episodes like this. Luna: Yeah, it's a small thing that makes a big difference for us. And hey, we appreciate everyone who listens. Lucas: Alright, back to coating. One more thing about cure time — acrylic dries in about thirty minutes at room temperature, but full cure takes twenty-four hours. Urethane and silicone often need heat curing, which means an oven step. That adds cycle time to your production line. Make sure your contract manufacturer has the oven capacity before you commit to a urethane coating at scale. Luna: That's a good point. A lot of founders forget to ask about curing infrastructure. Lucas: They do. And then they get a surprise two-week delay because the CM can only bake a hundred boards at a time. So, to wrap it up: conformal coating is a fifty-cent decision that can save your entire product launch. Pick the right chemistry for your environment, choose an application method that matches your volume, and inspect with UV. If you're not doing it yet, start before your next run. Luna: And if you have a story about a board failure that coating could have prevented, we'd love to hear it. We're always looking for real-world examples. Lucas: Absolutely. Until next time, keep building.