Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Use Potting Compounds for Protection
Transcript
- Lucas: So Luna, I was digging into a teardown of a smart irrigation controller the other day, and the most interesting part wasn't the chip or the radio — it was this block of goo protecting the whole board. Luna: Goo? You mean potting compound. Lucas: Exactly. Potting. And it struck me that for a lot of hardware startups, potting is this secret weapon — it solves a ton of field failure problems at once. But it also gets overlooked because it's not glamorous and it adds cost and weight. Luna: Yeah, I feel like people think of it as the last resort — like, this board is a mess, just drown it in epoxy. But there's more nuance. Lucas: Right. So let's define it. Potting is when you take a populated circuit board and put it inside a mold or a shell, then fill that cavity with a liquid resin — epoxy, polyurethane, silicone — that cures into a solid block. The board becomes encapsulated. It's different from conformal coating, which is just a thin film sprayed or brushed on. Luna: And the obvious benefit is moisture protection. But what else does it actually do? Lucas: It does three big things. One, it keeps out moisture and contaminants. Two, it mechanically locks components in place — so vibration and shock resistance go way up. And three, it can help with thermal management, depending on the material you choose. Some potting compounds have decent thermal conductivity, so they can spread heat away from hot components. Luna: So it's like a three-in-one. But there must be downsides — otherwise everyone would do it. Lucas: The big one is repairability. Once you pot a board, you can't get back in there. If a component fails, you're not swapping it out — the whole assembly is scrap. That's why you see potting more in products where field failures are catastrophic and replacement is cheaper than repair. Think automotive sensors, industrial controllers, smart meters. Luna: And it adds weight. Not great for drones or wearables. Lucas: Definitely. But for many startups, especially those making IoT devices for agricultural or industrial settings, the weight is a non-issue. The real question is whether the protection justifies the cost and the loss of serviceability. Luna: So when should a startup actually consider potting? What's the trigger? Lucas: The classic trigger is when you start seeing field failures that trace back to moisture or vibration. Like, you launch a product, it works great in the lab, but once it's out in the real world — sitting in a pump station or on a tractor — you get intermittent failures. You open it up and there's corrosion on the board. That's the pain point. Luna: We had a similar case in a recent episode about vibration testing. But potting is a different solution than just testing better. Lucas: Exactly. Testing tells you where the problem is. Potting solves it. And it's often more cost-effective than redesigning the enclosure or adding expensive connectors and seals. Let's look at a real example. There's a startup I know — let's call them 'AgriSense' — making a soil moisture sensor that goes into the ground for months. They initially used a conformal coating, but after a season, they had a 12 percent failure rate due to condensation inside the housing. They switched to potting the sensor board — not the battery, just the board — and the failure rate dropped to under one percent. Luna: Wow, that's a huge improvement. But did the potting add a lot to the unit cost? Lucas: It added about 80 cents in material and maybe 20 cents in labor per unit. So a dollar per unit, roughly. And they were selling the sensor for forty dollars. So it ate two and a half percent of the margin. But in exchange, they cut their warranty claims drastically — and in a hardware startup, warranty claims are the profit killer. The real cost is not the potting, it's the field failure. Luna: That's a great trade-off. But what about the process itself? Is potting easy to do in a low-volume production run? Lucas: It can be messy, but it's very scalable. At low volumes, you're often hand-pouring or using a simple dispensing gun. The trick is controlling the flow and making sure there are no air bubbles. Air bubbles can create voids that trap moisture or become stress points during thermal cycling. So you need a good degassing step — either a vacuum chamber or a slow-cure resin that lets bubbles escape. Luna: Right, bubbles are the enemy. And then there's the thermal expansion thing — if the potting expands at a different rate than the components, you can actually crack a solder joint. Lucas: That's the subtlety most people miss. You're not just pouring in any epoxy. You have to pick the right material chemistry for your expected temperature range. Epoxies are strong and cheap, but they're rigid — so across a wide temperature range, they can stress components. Silicones are more flexible and handle thermal cycling better, but they're softer, don't seal as well over time, and are pricier. Polyurethanes sit in the middle. Luna: So the material choice is a real engineering decision. It's not like duct tape. Lucas: Exactly. And that's why we're seeing more startups partner with material suppliers early. They don't just buy a compound off the shelf — they work with the manufacturer to test it against their specific thermal and mechanical profile. Luna: And what about testing? Should you test potted boards differently? Lucas: You should, and I think the key is thermal cycling. Potting changes the thermal impedance of the board, so you need to verify that hot components still cool adequately. Also, you want to do a few thermal cycles to make sure the potting doesn't crack or separate from the board. Then you do vibration testing — but you're testing the whole assembly, not just the bare board. Luna: And it's probably worth testing the potting process itself — like, how consistent is your dispense volume? If you underfill, you get no protection. If you overfill, you might get into connectors or mounting holes. Lucas: Yeah, that's a classic pitfall. You need to design the housing to have a vent or a dam to control the flow. Some startups even use a two-part process: first a thin layer to cover the board, then a second fill to the top. That reduces the chance of voids. Luna: I'm also thinking about the business side — like, does potting change your manufacturing timeline? It adds a curing step, which could be hours or even a day. Lucas: Good point. Curing time is a hidden cost. If you're doing high throughput, you might need an oven to accelerate the cure. That's an energy cost but also a floor-space cost. But for many startups, it's not a constraint because they're already doing small batches. Luna: So what's the one thing a founder should take away from this? Lucas: I'd say: don't think of potting as a last resort. Think of it as a design tool. If you're designing a product that will live in a harsh environment — moisture, dust, vibration, wide temperature swings — consider potting from day one, not after you see failures. The cost of retrofitting potting into an existing design is much higher than designing for it upfront. Luna: That's a really practical tip. And it's kind of amazing that a little goo can save your product. Lucas: It's not glamorous, but it works. And honestly, that's the kind of thing we like to highlight on this show — the unsexy engineering decisions that make or break a hardware startup. Luna: And speaking of things we like to highlight — we should mention that we keep this podcast free of ads, and that's something we're proud of. If you find these conversations useful for what you're building or running, and you want to support that ad-free choice, there's a way to do that. It's buy me a coffee dot com slash fexingo. Lucas: Yeah, that's right. No ads, no sponsors pushing products. Just us, talking through the real decisions in hardware. It's a simple way to keep the lights on, if you're so inclined. Luna: And no pressure. The content stays free either way. But it helps a little. Lucas: Alright, back to potting. One more thing I want to mention: the environmental angle. Some potting compounds are difficult to recycle, and if your product has a battery, you might need to design for easy separation. That's a consideration for startups that care about sustainability. Luna: Interesting. So potting can conflict with repairability and e-waste goals. But there are emerging bio-based or recyclable potting materials, right? Lucas: There are. A few suppliers are working on polyurethane systems that can be depolymerized under certain conditions. It's early days, but it's a space to watch. If a startup can use a recyclable potting compound, that could be a differentiator. Luna: That's a nice forward-looking note. So, if someone is considering potting for their product, what's the first step? Lucas: Talk to a material supplier early. They'll help you pick the right chemistry and give you sample coupons to test. And run a few thermal cycles to see how it behaves. It's cheap to test, and it could save you from a field failure nightmare. Luna: Thanks, Lucas. This has been really helpful. For our listeners, we hope this gives you a new tool in your design toolbox. Lucas: And remember, it's not about the goo itself — it's about the reliability it brings to your product. Until next time, keep building.