Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Use Vibration Testing to Prevent Field Failures
Transcript
- Lucas: You know, we've talked a lot on this show about how startups test for heat, for shock, even for moisture. But there's one stressor that quietly kills more hardware than almost anything else, and it rarely gets the attention it deserves. Luna: Let me guess — vibration? Lucas: Exactly. And I'm not talking about the dramatic drop test or the thermal cycle. I'm talking about the constant, low-level rumbling that your device experiences every single day — in a pocket, on a bike mount, inside a factory line, on a drone. That steady mechanical stress can loosen screws, crack solder joints, and eventually cause total field failure. Luna: And the thing is, most startups don't think about it until their first batch comes back with mysterious glitches. By then, it's expensive. Lucas: Right. So today I want to dig into why vibration testing is becoming a must-have step in hardware development, and how even a lean startup can do it without breaking the bank. And this connects to something bigger — this podcast is ad-free, and that's a choice we make so these conversations stay focused on what you're building and running. If you find them useful, you can support that mission at buy me a coffee dot com slash fexingo. It's a small gesture that keeps us independent. Luna: Yeah, and it genuinely helps. Now, back to vibration — where do you even start? Lucas: Well, the first thing to understand is that vibration testing isn't just about one type of shaking. There's sine vibration, which is a single frequency, and random vibration, which is a mix of many frequencies at once. Real-world environments are almost always random — think of a car driving over gravel or a washing machine spinning. Luna: So if you're testing a device for a drone, you'd want to simulate that rotor buzz and the wind gusts, right? Lucas: Exactly. And that's the challenge — getting the profile right. You can't just shake it and hope for the best. You need to understand the environment your product will live in, and then replicate that as closely as possible. Luna: And that's where a lot of startups get it wrong — they either overtest with a harsh profile and end up designing for a space shuttle, or they undertest and miss the real failure modes. Lucas: Yeah, it's a balance. So let's talk about the tools. The classic piece of equipment is an electrodynamic shaker — basically a speaker-like device that vibrates a platform. A good one can cost anywhere from ten thousand to fifty thousand dollars, which is a big chunk of change for a startup. But the thing is, you don't necessarily need to buy one outright. Luna: Right, we've talked about test as a service on this show before. That's a huge option for vibration testing — you can rent time on a shaker table at a lab, sometimes for a few hundred dollars an hour. Lucas: And that's often the smartest move for a startup. You get access to professional-grade equipment, calibrated accelerometers, and engineers who can help you design the test. But there's also a cheaper, more iterative approach that's gaining traction: building your own low-cost vibration rig using a motor with an eccentric weight — the kind used in phones for haptic feedback. Luna: That sounds a bit hacky, but I've seen some startups do that for early prototyping, to catch gross issues before spending money on a formal test. Lucas: Exactly. It won't give you precise data, but it's a great way to do a quick sanity check. You can mount your device to a plate, attach an eccentric motor, and run it for a few hours in each axis. If something's going to break, it'll often break in that crude setup. Luna: And that's the 'fail fast' philosophy, right? Find the weakness early when it costs pennies, not later when you've got thousands of units in the field. Lucas: Absolutely. And that's the real value of vibration testing — it's not just about passing a standard. It's about uncovering your design's weakest links before your customers do. There's a great example from a drone startup I know — they were seeing a 30 percent return rate due to motor controller failures. They did a vibration test with a random profile simulating flight, and within hours they found that a connector was vibrating loose. A simple redesign — adding a locking tab — cut their field returns to almost zero. Luna: That's a massive saving. And it shows why vibration testing should be part of your design validation, not just a final check. Lucas: Right. But there's a catch — you need to be careful not to overtune your design to pass the test and end up with a product that's overbuilt and too expensive. Vibration testing is about finding the sweet spot between robustness and cost. Luna: So how do you know what profile to use? Is there a standard? Lucas: There are standards like IEC 60068-2-64 for random vibration, but they're generic. For a consumer device, you might use a generic automotive or aerospace profile as a starting point, but nothing beats measuring the actual environment. You can buy a small accelerometer data logger — they cost a couple hundred dollars — and mount it on a prototype, then go drive your car, fly your drone, or run your machine. Capture the real vibration data for a few hours, and you've got a custom profile that's tailored to your product. Luna: That's a smart, low-cost approach. And it gives you data you can actually use in your design process. Lucas: Exactly. And once you have that profile, you can run a random vibration test on a shaker table and see how your product holds up. But you also want to look for resonances — those frequencies where your product amplifies the vibration. That's where sine sweeps come in. You sweep through a range of frequencies and measure the response with accelerometers. If you see a spike at a certain frequency, that's a resonance, and it's likely to cause fatigue failure. Luna: And that's where you might need to change a material or add a stiffening rib, right? Lucas: Exactly. Or even change the mounting points. Resonance testing is like finding the heartbeat of your product — you need to know its weak spots. And it's not just for electronics. I've seen startups making mechanical assemblies — like actuators for robotics — use vibration testing to ensure their gears don't rattle and wear prematurely. Luna: So vibration testing is really about reliability engineering, not just a box-ticking exercise. Lucas: And that's what separates hardware startups that survive from those that don't. If you ship a product that fails in the field, your reputation is damaged, and the cost of recalls or replacements can sink you. A few thousand dollars spent on vibration testing early on can save you hundreds of thousands in warranty claims later. Luna: Plus, investors are starting to ask about reliability testing in due diligence. They want to see that you've de-risked your product. Lucas: Yeah, that's a great point. In the last couple of years, I've noticed that more investors are asking about environmental testing — including vibration — as part of their technical due diligence. They want to know that you've thought about failure modes, not just features. Luna: So what's the one takeaway you'd want a hardware founder to remember about vibration testing? Lucas: Don't wait until you have a field failure to start thinking about vibration. Get a low-cost rig, make a prototype, shake it, break it, learn from it. Then spend the money on a proper test once you've fixed the obvious issues. It's the fastest way to build a reliable product. Luna: And if you're listening and you've used vibration testing in your own product, we'd love to hear about it. You can find us on social media, or drop us a note through the show page. Lucas: Absolutely. And if you want to support the show and keep it ad-free, the link is buy me a coffee dot com slash fexingo. It's a simple way to say thanks, but no pressure. Now, let's get back to the workbench and break some products.