Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Solve the Cable Problem
Transcript
- Lucas: You have a prototype that works perfectly on a bench. It's been through drop testing, thermal cycling, maybe even a digital twin simulation. You ship it to a hundred beta users. And then the reviews start coming in — the device keeps disconnecting. Or it charges slowly. Or it doesn't charge at all. And you eventually trace the problem to a cable. Luna: The cable. Of all things. It's almost embarrassing because it looks like the simplest part of the system. Lucas: Right, a cable is literally two connectors and a wire. But in practice, cables are one of the most failure-prone components in any hardware product. And for a startup, getting cables wrong can tank your launch, generate a wave of returns, and destroy margin. Luna: So why is it such a mess? I mean, we've had usb c as a standard for years now. Hasn't that solved the problem? Lucas: It's helped at the connector level, but usb c is actually more complicated than what it replaced. A usb c cable can carry USB 2.0, USB 3.x, Thunderbolt, DisplayPort, and power delivery up to 240 watts. But whether it actually supports those things depends on the internal wiring, the chips inside the connector, and the quality of the shielding. Luna: So two cables that look identical can behave completely differently. Lucas: Exactly. You can buy a usb c cable for three dollars that only does USB 2.0 speeds and 15 watts, or you can spend twenty-five dollars on one that does Thunderbolt 4 and 100 watts. And there's no visual difference. So when a startup includes a cheap cable in the box, users plug it in, expect full performance, and get frustrated. Luna: Is this a certification problem? Like, should startups be using certified cables? Lucas: Certification is a big part of it. The USB Implementers Forum runs a certification program. If your cable passes, you can put the certified logo on it. But certification costs money — testing can run several thousand dollars per cable type — and many startups skip it to save cost. The problem is that uncertified cables often fail to meet the spec, especially for power delivery. Luna: And then you get the 'slow charging' complaints. Lucas: Right. And there's a deeper issue. Even within usb c, there's a huge range in wire gauge. Thinner wire means more resistance, which means voltage drop. If you're trying to deliver 60 watts over a long cable with thin wire, the voltage at the device end could be significantly lower than what the device expects. That can cause charging to stop, or even damage the power management circuit. Luna: So what should a startup do? Spec everything out in detail with the manufacturer? Lucas: Absolutely. You need to specify wire gauge — typically 24 AWG for power and 28 AWG for data in a good cable. You need to specify the connector overmold quality, the shielding type — braided versus foil — and the exact chipset used for the e-marker. The e-marker is a tiny chip inside the usb c connector that tells the power source and the device what the cable can handle. Luna: And if you don't include that chip, the cable defaults to 15 watts, right? Lucas: Correct. Without an e-marker, a usb c cable is limited to 3 amps and 20 volts, which is 60 watts. But if you want 100 watts or more, you need the e-marker. And many cheap cables simply omit it. So your device that expects 100 watts gets only 60, and the user blames the device, not the cable. Luna: It's like the cable becomes a single point of failure for the user experience. Lucas: Exactly. And then there's the mechanical side. Connectors are the most stressed part of any electronic product. People yank cables, bend them, step on them. The strain relief — where the cable meets the connector body — needs to be robust. A lot of cheap cables crack there after a few months. So startups need to test for that, too. Luna: Is there a standard test for cable durability? Lucas: Yes, the usb if has a mechanical test spec — typically 5,000 insertion cycles for the connector. But there's also bend testing, where you cycle the cable at the connector joint. Some startups run their own tests with a simple pendulum setup. It's not expensive to do, but if you don't do it, you're flying blind. Luna: What about proprietary connectors? Some startups use them, especially for rugged or specialized devices. Lucas: Proprietary connectors give you control, but they come with a huge cost. You have to design the connector, tool the mold, and then manufacture it at low volume — which means high per-unit cost. And users hate carrying a second cable. If you can use usb c, you should. But there are cases where usb c isn't robust enough — say, for underwater connectors or high-vibration environments. Luna: So the decision tree is: can usb c work? If yes, use it. If not, go custom. Lucas: Right. And if you go custom, you need to think about the whole ecosystem: will you include the cable in the box? Or sell it separately? Many startups include a cable because they want to guarantee the experience. But that adds cost — a good usb c cable costs two to five dollars in volume, plus packaging. Luna: And if you don't include it, you risk users plugging in a junk cable and having a bad experience. Lucas: Exactly. Some startups solve this by including a short, high-quality cable and then offering a longer one as an upsell. Others include a cable but make it detachable so users can replace it easily. The key is to test with the actual cable you'll ship, not just with a lab-grade cable. Luna: Testing with the production cable — that sounds obvious, but I bet a lot of startups skip that step. Lucas: They do. They test with a certified cable from a known brand, then switch to a cheaper contract manufacturer for production. And then they get field failures. So the rule is: test with the exact cable you'll ship, same batch, same factory. And do a sample test from multiple production runs. Luna: It reminds me of the early iPhone days — Apple had a huge problem with third-party cables that didn't meet spec. They created the MFi program to force quality. Lucas: That's a great example. Apple's MFi — Made for iPhone/iPod/iPad — program licenses the connector and chips. If you make a Lightning cable without MFi certification, iOS would show a 'this accessory may not be supported' warning. It protected the user experience, but it also locked in Apple's control. For a startup, if you're making an iPhone accessory, you have to go through MFi. That adds lead time and cost — about one to two dollars per cable in licensing and chip costs. Luna: And the chip shortage made that even harder for a while. Lucas: Yes. But the lesson is broader: the cable is not a commodity. It's a critical component that carries power and data. Treat it like you treat any other electronic component — with specs, testing, and supply chain planning. Luna: I want to circle back to the EU's common charger directive. From 2024, all new portable devices sold in the EU must have usb c. That's forcing even Apple to move away from Lightning. How does that affect startups? Lucas: It simplifies things for consumer hardware. If your target market includes Europe, you must use usb c. That reduces the temptation to go proprietary. But it also means that the usb c ecosystem will only get larger, and counterfeit cables will be even more rampant. So startups need to be vigilant about sourcing. Luna: Where do we see startups making the biggest mistakes with cables right now? Lucas: Number one: underestimating the cost of a good cable. They spec a cheap cable to hit a price point, and then they see return rates spike. Number two: not verifying that the cable actually delivers the power profile their device needs. They check that it fits mechanically, but they don't measure voltage drop under load. Number three: ignoring the environmental factors — if the device is used outdoors, the cable needs uv resistant jacket and better sealing. Luna: So it's really about engineering the cable as part of the system, not just picking one off a catalog. Lucas: Exactly. And the good news is that for a startup, getting the cable right can be a real differentiator. If your device charges fast every time and never disconnects, users notice. It builds trust. And margins: a good cable might cost you two dollars more, but it can save you ten dollars in returns and support calls. Luna: And it's a relatively small investment to get right — compared to, say, redesigning a PCB. Lucas: Right. Speaking of small investments that pay off — this show is ad-free thanks to listeners who support us. If you find these deep dives useful for what you're building, you can support the show at buy me a coffee dot com slash fexingo. That's buy me a coffee dot com slash fexingo. It helps us keep doing episodes like this one, where we can spend twenty minutes on the cable problem. Luna: Yeah, and we really mean it — we research these topics because we know hardware is hard. And the support from listeners lets us stay independent and focused on the specifics. Lucas: Absolutely. So, to wrap up today: if you're building a connected device, don't leave the cable as an afterthought. Spec it, test it, and budget for it like any other component. Your users — and your margin — will thank you. Luna: And maybe think twice before using that freebie cable from the trade show. Next time on Hardware Startups with Fexingo, we're looking at the trade-offs between internal antennas and external antennas for IoT devices. Lucas: That should be a good one. Antennas are another one of those 'simple' things that are anything but. See you then.