Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Use Conformal Coating to Prevent Board Failure
Transcript
- Lucas: You open a shipping container from your contract manufacturer, pull out a batch of circuit boards, and they look perfect. Two months later, units start dying in the field — not because the design was wrong, but because a microscopic layer of moisture crept under a component and created a short. Luna: That sounds like a nightmare. And it's something you can prevent with a coating that costs maybe a dollar per board? Lucas: Exactly. It's called conformal coating — a thin polymer film, typically 25 to 75 microns thick, that conforms to the contours of the board and its components. It blocks moisture, dust, salt spray, even vibration-induced fretting corrosion. And the surprising thing is how many hardware startups skip it during their first production run. Luna: Why would they skip it? Seems like cheap insurance. Lucas: Two reasons. First, if you're prototyping in a lab in San Diego or Austin, the environment is benign. You never see the failure mode. Second, adding a coating step means your contract manufacturer needs a clean, dust-free spray booth, curing ovens, and a quality check — that adds lead time and complexity. So founders say, 'Let's get to market faster, we'll add it later.' But later is usually after a recall. Luna: Do you have a specific example of that? Lucas: Yeah. I talked to the founder of a small robotics startup that built an agricultural sensor — deployed in fields, high humidity, direct sun. First batch of 500 units, they skipped conformal coating to hit a trade show deadline. Within six months, 12 percent of those units had failed. Moisture had wicked under the microcontrollers. The warranty replacement cost them over 60 thousand dollars. Coating would have cost about 35 cents per board. Luna: Ouch. So it's not just about the cost of the coating — it's the cost of the failure. Lucas: Right. And the reputational damage. On the flip side, I spoke with a drone company that does agricultural spraying — their boards are exposed to fertilizer mist, which is corrosive. They started using a selective conformal coating process — robotic spray that only coats certain areas — and their warranty claims dropped by 80 percent year-over-year. Luna: Wait — selective coating? Why wouldn't you just coat the whole board? Lucas: Great question. Some components shouldn't be coated — connectors, switches, sensors that need to interact with the environment, large heat sinks where the coating acts as an insulator. If you coat a temperature sensor or a pressure port, you change its reading. So selective coating masks off those areas, either with a robot that sprays only the target zone or with a temporary mask that's removed after curing. Luna: So the robot is programmed with the board layout and applies coating only where needed? Lucas: Exactly. That's what the drone company did. They used a six-axis robot with a needle dispenser — it traces the exact perimeter of each component. The capital cost is about 40 thousand dollars for the robot, but they run thousands of boards per month, so the per-unit cost is negligible. And the consistency is way better than hand brushing. Luna: Let's talk about the materials. I know there are different types of conformal coating. What should a founder look for? Lucas: Three main chemistries. Acrylic — it's the most common, easy to apply, easy to remove if you need to rework a board. It's good for moderate environments — indoor consumer electronics, IoT devices that sit in a living room. But it has poor solvent resistance. Silicone — it handles high temperature and high humidity better, it's flexible, so it's good for boards that experience thermal cycling. But it's harder to remove, and it can contaminate surfaces if not cured properly. Luna: And the third? Lucas: Urethane. It's the most durable — excellent moisture and chemical resistance, very hard, good abrasion resistance. But it's almost impossible to remove. You need to solder through it. So you use urethane only if you're confident the board won't need rework. Automotive under-hood electronics, industrial controls, medical devices — that's where urethane shines. Luna: So the choice really depends on the use case. Can a startup switch coating types between prototypes and production? Lucas: It's risky. The coating's adhesion depends on the surface cleanliness, the application method, the cure profile. If you switch from acrylic to silicone, you need to validate that the new coating doesn't cause any chemical incompatibility — some silicones can react with certain solder fluxes. You also need to re-qualify your thermal performance. So it's better to choose early and stick with it. Luna: What about the application method itself? You mentioned robotic spraying. What are the other options for a startup that doesn't have that kind of capital? Lucas: Three common ones. Hand brushing — low cost, flexible, but inconsistent. You get variable thickness, you can miss areas, and if the operator brushes too hard they can damage components. Dip coating — you dunk the whole board in a bath of liquid coating. It's fast and uniform, but you can't selectively avoid connectors, so you have to mask them manually. And aerosol spray — from a can. That's the most accessible for a small batch, but it's expensive per board and hard to control thickness. Luna: So what's the sweet spot for a startup doing maybe a few thousand units a year? Lucas: For early production, you can have your contract manufacturer do hand brushing or dip coating with masking. Most CM's offer it as an option. The key is to include it in the scope of work from the start, not add it later. If you wait until after the first batch has failed, you're paying for rework and lost time. Luna: That's a good point. Has the industry seen any new developments in conformal coating recently? Lucas: One interesting trend is uv curable conformal coatings. They cure in seconds under UV light, instead of the 30-minute thermal cure for traditional coatings. That speeds up production significantly. But they're still more expensive per liter. Another trend is parylene coating — it's a vapor-deposited polymer that gives a truly conformal, pinhole-free layer. It's the gold standard for medical implants and aerospace. But the deposition equipment costs hundreds of thousands of dollars, so it's out of reach for most startups. Luna: So for most hardware founders, the practical path is to work with their CM, specify acrylic or silicone depending on the environment, and make sure they have a good quality check. How do you inspect the coating? Lucas: Visual inspection under UV light — most conformal coatings have a fluorescent tracer. You shine a black light on the board and the coating glows. That's how you see coverage gaps. For thickness measurement, you can use an eddy-current probe or a micrometer on a test coupon. And for adhesion, you can do a tape peel test — apply a piece of tape and pull it off. If coating comes off, the surface wasn't clean enough. Luna: And cleaning is a whole other topic — boards have to be free of flux residues and dust before coating. Lucas: Exactly. Aqueous cleaning or solvent cleaning, followed by a bake-out to remove moisture. If you apply coating over a contaminated board, it will delaminate. That's one of the most common failure modes I see in audits. Founders think they've solved the humidity problem, but they didn't clean properly, so the coating peels off after a few weeks. Luna: So it's not just 'spray and go.' It's a process. Lucas: It is. But the payoff is huge. I've seen startups that invested in a proper coating process reduce their field failure rate from 8 percent to under 0.5 percent. That's the difference between a product that gets reordered and one that gets returned. Luna: One thing I appreciate about this podcast is that we dig into these manufacturing details that often get overlooked in the startup hype. If you're building hardware, these are the kind of practical steps that save you from a disaster. Lucas: Absolutely. And if today's episode gave you something useful — a specific number, a process to ask your CM about — the way we keep this show ad-free and focused on that kind of value is listener support. You can find it at buy me a coffee dot com slash fexingo. That's buy me a coffee dot com slash fexingo. Luna: Yeah, it really helps. And we appreciate everyone who's chipped in. It lets us keep going deep on these topics. Lucas: So back to conformal coating — one more thing I want to mention. If you're designing a board, you can actually plan for coating. Avoid placing connectors on the bottom side if they'll get coated. Use shielding cans for sensitive areas that you want to keep uncovered. And make sure your board has good solder mask — that's your first line of defense. Coating is the second. Luna: Right. So it's a layering strategy — solder mask, then conformal coating, then maybe a potting compound for extreme environments. Lucas: Exactly. And the best time to think about it is during the schematic and layout stage, not after the board comes back from assembly. If you have connectors placed near the edge, it's easier to mask. If you have a large inductor that runs hot, you might skip coating over it. Those design choices affect cost and reliability. Luna: Have you seen any startups that actually made coating part of their product story — like, marketing their ruggedness? Lucas: A few. One outdoor sensor company puts 'conformal coated for life in the field' on their packaging. It's a trust signal. Another robotics company includes a photo of their UV inspection process in their pitch deck — it shows investors they're serious about manufacturing quality. It doesn't add much cost, but it adds a lot of credibility. Luna: That's smart. It turns a manufacturing detail into a selling point. Lucas: Exactly. So if you're a hardware founder, ask your CM about their conformal coating capability. Ask which materials they support. Ask to see their UV inspection setup. If they can't do it in-house, find a partner who can. It's a small investment that pays for itself the first time a customer uses your product in the rain.