Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Use Vibration Welding for Sealed Enclosures
Transcript
- Lucas: If these conversations are useful for what you're building or running, I want to be straight with you about how this show stays on the air. Luna: Yeah, this is a good moment for that. Lucas: A small group of listeners already chips in monthly through buy me a coffee dot com slash fexingo, and that support is what keeps this ad-free. No corporate sponsors, no tracking, just the conversation. Luna: It's a simple model. And it works because enough people find value here. If that's you, great. If not, the show stays free either way. Lucas: Alright. Today: vibration welding. A technique that hardware startups are using more and more to seal plastic enclosures without glue, without screws, without gaskets that dry out over time. Luna: I've seen this called linear friction welding. Is that the same thing? Lucas: Exactly the same. Two thermoplastic parts are pressed together and oscillated relative to each other — typically at 100 to 240 hertz, with an amplitude of about half a millimeter to two millimeters. The friction generates heat at the interface, the plastic melts, and when the oscillation stops, the molten material solidifies into a weld that is as strong as the parent material. Luna: And that creates a hermetic seal? I've heard of vibration-welded enclosures passing IP67 — submerged in a meter of water for 30 minutes. Lucas: IP67 is routine. Some medical devices go further. Take a continuous glucose monitor — the kind that sticks on your arm for ten to fourteen days. That enclosure needs to survive showers, sweat, maybe even a quick swim. Vibration welding gives you a weld depth of 0.2 to 0.5 millimeters, which is more than enough for a reliable seal. Luna: But it's not just about the seal. The joint also has to hold up against the internal battery swelling or the flex from body movement. Lucas: Right. And vibration welding handles that well because the weld is continuous around the entire perimeter. There are no weak points like you'd get with discrete screws or snap fits. The part basically becomes one piece. Luna: What's the trade-off compared to ultrasonic welding? I know that's more common for smaller parts. Lucas: Ultrasonic welding uses 20 to 40 kilohertz — much higher frequency, much smaller amplitude. It's great for tiny parts, like a hearing aid shell. But the weld area has to be small — typically a few square centimeters. Vibration welding can handle joints that are tens of centimeters long. Think of a rectangular enclosure for a handheld diagnostic device, maybe 10 by 15 centimeters. Ultrasonic can't do that in one shot. Luna: So vibration welding is for bigger enclosures. But you need access to both sides of the joint, right? Lucas: You need at least one part to be free to vibrate. Typically the top part is held in the vibrating tool, and the bottom part is clamped in a stationary fixture. The joint geometry has to include what's called an energy director — a small triangular ridge on one of the mating surfaces that concentrates the heat and melt. Luna: That's the gotcha I've seen trip up startups. They design the enclosure, send the mold out for tooling, and then realize there's no energy director. Adding one after the mold is cut is expensive rework — sometimes thousands of dollars. Lucas: Absolutely. The energy director should be part of the initial mold design. It's usually a 60- to 90-degree triangular rib, about 0.2 to 0.5 millimeters tall, running along the joint line. Without it, the weld might be inconsistent or fail entirely. Luna: And the material matters too. Vibration welding works best with amorphous plastics like ABS, polycarbonate, or acrylic. Semi-crystalline materials like nylon or polypropylene can be trickier. Lucas: That's right. Semi-crystalline plastics have a sharper melting point, so the process window is narrower. You need tighter control over melt depth and cooling time. But it's still done — you just need to run process validation trials to dial in the parameters. Luna: What about laser welding? That's another non-contact method that's gaining popularity. Lucas: Laser welding is great for clean, precise welds with no flash. But it requires one part to be transparent to the laser wavelength and the other to be absorptive. That limits your material and color choices. Vibration welding doesn't care about color — both parts can be opaque black, which is common in medical devices to block UV light. Luna: And cost. A benchtop vibration welder runs what, fifty to a hundred and fifty thousand dollars used? Lucas: That's the range. A new industrial system can be three hundred thousand plus. But you can also outsource the welding to a contract manufacturer. Many job shops offer vibration welding as a service, and they'll run a few hundred parts for you at a reasonable per-part cost. Luna: So a startup doesn't necessarily need to buy the equipment upfront. They can validate the design with a service provider, and then if they scale, they can bring the machine in-house. Lucas: Exactly. That's the smart path. Get the joint geometry right, run a small batch, do a leak test — a 24-hour pressure decay test on the first 50 parts is my recommendation — and then commit to production tooling. Luna: I've seen startups skip that validation and end up with a thousand enclosures that leak. The cost of scrapping those is way higher than running the test upfront. Lucas: A pressure decay test is straightforward. Pressurize the sealed enclosure to, say, five kilopascals above ambient, and measure the pressure drop over 24 hours. If it drops more than a few percent, the weld has a defect. You can even use a simple manometer and a stopwatch. Luna: And if the test fails, you can often re-weld the part by re-melting the joint. Vibration welding is reworkable — unlike adhesive bonding, where you have to clean off the old glue. Lucas: That's a huge advantage. With adhesive, if the bond fails, the part is usually scrap. With vibration welding, you can put it back in the machine and run another cycle. The material might degrade slightly, but you can often get a second weld that passes. Luna: So the key takeaways for a hardware startup: design the energy director into your mold from day one, choose an amorphous plastic if possible, and validate with a pressure decay test before scaling. Lucas: And consider outsourcing the welding initially. That way you're not betting fifty thousand dollars on a machine until you know your design works. Vibration welding is a mature process — it's been used in automotive for decades — but it's only recently become accessible to startups through service bureaus. Luna: One last thing: what about the part orientation? The vibration direction has to be parallel to the joint line, otherwise the parts can shift. Lucas: Good point. The oscillation is linear, back and forth. If the joint has a corner, the vibration should be along the long axis of the enclosure. Otherwise you get uneven melt at the corners. That's a detail the tooling designer needs to specify on the drawing. Luna: Alright. So vibration welding is a solid option for startups building sealed plastic enclosures — especially if they're in medical devices or rugged consumer electronics. Next time, maybe we look at hot plate welding, which is similar but uses a heated platen instead of friction. Lucas: That's a good one. For now, if you're prototyping an enclosure, try designing that energy director into your next iteration. It might save you a lot of trouble down the line.