Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Use Conformal Coating for PCB Protection
Transcript
- Lucas: If you're building a hardware product that will ever leave a climate-controlled room — and most of you are — there's a manufacturing step that a surprising number of startups either skip entirely or add too late. I'm talking about conformal coating. Luna: Conformal coating — that thin transparent layer that goes over the populated circuit board, right? I've seen it called 'pcb nail polish'. Lucas: Exactly. And the nail polish analogy actually works pretty well. It's a dielectric film, typically twenty-five to seventy-five microns thick, that conforms to the contours of the board and its components. Its job is to block moisture, dust, chemicals, and vibration from causing failures. And for a lot of hardware startups, it's the difference between a product that survives its first summer in the field and one that gets returned in batches. Luna: So who should be thinking about this at prototype stage? Not just people building outdoor stuff? Lucas: Outdoor is the obvious one. If you're building a solar-powered sensor that sits on a rooftop or an agricultural monitor that lives in a field, you need coating. But also medical devices that get wiped down with disinfectants, industrial equipment near washdown areas, consumer electronics that might end up in a bathroom. Basically, if your board will see condensation or any liquid exposure, you should plan for it. The worst time to discover you need coating is after a field trial failure. Luna: I actually have a story on that. A startup I know built a smart irrigation controller. They did a hundred-unit field trial in California. Three months later, about forty percent had failed. The root cause was condensation inside the enclosure — no coating, and the board had exposed traces that corroded. They had to redesign the enclosure, add coating, and re-run the trial. Set them back almost six months. Lucas: That's exactly the kind of thing that kills a launch timeline. And the frustrating part is that conformal coating isn't expensive — at low volumes, it might add a few dollars per unit in labor and material. But if you design the board without considering it, you can run into issues. Certain components, like connectors, switches, and test points, need to be masked off. Tall components can create shadowing where the coating doesn't reach. And some coatings are difficult to remove for rework. Luna: So what are the options? I know there's more than one kind of goop. Lucas: There are four main chemistries a startup will encounter. Acrylic is the most common — it's cheap, easy to apply, cures fast at room temperature, and it's relatively easy to remove for rework. The downside is lower chemical resistance and it's not great in high-temperature environments. Then there's urethane, which is tougher, more abrasion-resistant, and handles solvents better. But it's harder to remove — you basically need specialized strippers or you have to burn it off. Luna: And I've heard silicone coatings are popular for high-voltage or high-temperature applications. Lucas: Right. Silicone is flexible — it handles thermal cycling really well, and it can withstand higher temperatures, up to around two hundred degrees Celsius. It's also good for boards that experience vibration. But it's messy to apply, and rework is a nightmare because it's so soft — you can't easily solder through it. You usually have to mechanically remove it. And then there's parylene, which is the gold standard. It's vacuum-deposited, so it goes on as a vapor and polymerizes onto every surface, including underneath components. No pinholes, no shadowing, uniform thickness. It's incredibly reliable. Luna: But expensive and requires specialized equipment, right? Not really a prototype-stage option. Lucas: Exactly. Parylene is usually something you add when you're going to high volume and the product demands it — think implantable medical devices or aerospace sensors. For most hardware startups, acrylic or urethane is the right call. And the good news is that there are plenty of contract manufacturers that will do selective coating with robotic spray dispensers. You can send them your board, they'll program the coating path, and you get consistent coverage at a reasonable cost. Luna: What about application methods at prototype scale? I imagine the big factory dip tanks aren't accessible when you're building fifty boards. Lucas: At low volumes, you have three options. Hand spraying with a spray gun or even an aerosol can — it works if you're careful, but you risk overspray and uneven coverage. Brush-on coating is another option; it's slow and labor-intensive, but it gives you control for small batches. And then there's dipping, where you immerse the board in a bath of coating. That works well for simple boards, but you have to be careful about pooling in connectors. A lot of startups start with hand spraying, then move to selective robotic coating when they hit a few thousand units. Luna: And I assume you need to design the board with coating in mind. Like, you don't want connectors that are hard to mask. Lucas: Huge point. One of the biggest pain points is capillary wicking — that's when the coating seeps into connectors, relays, or switches through surface tension, and then those components don't make electrical contact anymore. So you have to either specify connectors that are sealed or designed to be coated, or you have to mask them. Masking is an added step and adds cost. The smarter move is to design your board so that components that can't be coated are placed on one side, and you only coat the other side. Or you use press-fit connectors that are naturally sealed. Luna: What about rework? If a board fails after coating, how do you even test it? Lucas: That's actually a consideration a lot of startups miss. Acrylic coating can be removed with a solvent — you apply it, the coating dissolves, and you can probe the board. Urethane and silicone need mechanical removal, which is more destructive. So if you expect a lot of development rework, you might want to stick with acrylic during prototyping. Some companies even hold off on coating until they're confident in the design, then add it at the pre-production phase. Luna: So the takeaway is: think about coating before you finalize the board layout. Decide which chemistry fits your environment, plan for masking, and don't wait until a field failure forces your hand. Lucas: Exactly. And if you're prototyping now, just call a conformal coating service — there are dozens of them that will do small runs. Ask them to walk you through their masking guidelines. The ten-minute conversation could save you a whole iteration cycle. Luna: If today's episode helped you think through that decision — or if you've already been burned by a lack of coating and wish you'd heard this earlier — the fastest way to keep these conversations ad-free and accessible is listener support. It's buy me a coffee dot com slash fexingo. That's it. Lucas: Yeah, we don't run ads, we don't have sponsors. It's just listeners like you who find value in the show and chip in. And it genuinely makes a difference. Luna: Alright. Back to the bench — Lucas, you mentioned plasma-deposited coatings as a newer option. What's the deal there? Lucas: Plasma deposition is interesting because it's a dry process — you put the board in a vacuum chamber, introduce a monomer gas, and use a plasma to polymerize it directly onto the surface. The result is an extremely thin, pinhole-free coating that's more uniform than anything you can spray. It's still mostly used in high-reliability applications like sensors for medical or automotive, but the equipment cost is coming down. I could see it becoming more accessible for mid-volume production in the next few years. Luna: So the barrier is still cost and equipment availability. But if you're building something that absolutely cannot fail — like a continuous glucose monitor — it might be worth the investment. Lucas: Right. And for most startups, acrylic or urethane is plenty. The key is to test your coating choice with an environmental chamber — thermal cycling, humidity, salt spray. Don't just trust the datasheet. Run a validation test with twenty boards and see where the weak points are. That's where you'll learn if you need a different chemistry or a different application method. Luna: And don't forget to test rework. Because if you can't fix a coated board easily, you're going to waste a lot of time and money. Lucas: Exactly. So to wrap it up: conformal coating is a cheap insurance policy against a very expensive failure. Plan for it from the PCB layout stage, pick the right chemistry for your use case, and validate early. Your future self — and your product's first field trial — will thank you.