Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Use Rapid Tooling for Bridge Production
Transcript
- Lucas: You have a prototype that works. You've shown it to ten potential customers and three of them say they want to buy. Now you need to make, say, five thousand units — not fifty, and not fifty thousand. This is the valley of death for hardware startups, and it's exactly where rapid tooling comes in. Luna: That gap is brutal. Your prototype was probably 3D printed or machined. But for five thousand pieces, you can't hand-make each one. So you think injection molding — but a production mold costs fifty grand and takes three months. Lucas: Right, and that's before you even know if the product will sell. So what a lot of startups are doing now is something called bridge tooling, or rapid tooling. The idea is simple: make a mold that's good enough for a few thousand parts, but costs a fraction of a production mold and ships in weeks, not months. Luna: So you're trading mold life for speed and cost. How short does that mold life actually get? Lucas: It depends on the material. A typical production mold for a plastic part is made of hardened tool steel — P20 or H13. That mold can run hundreds of thousands, even millions of cycles. A rapid tool is usually aluminum or a softer steel, and it might only last for one thousand to ten thousand shots before the cavity starts to wear. Luna: So if you need ten thousand parts, you might get one mold that does the whole run, but if you need fifty thousand, you're buying multiple rapid tools or going straight to steel. Lucas: Exactly. And the cost difference is dramatic. A single-cavity aluminum mold for a simple part might be five thousand dollars and take two weeks. Compare that to a steel mold that's fifty thousand and twelve weeks. For a startup that's trying to get to market and maybe raise a Series A, that cash and time is everything. Luna: Let's use a concrete example. I was talking to a founder who makes IoT building sensors — temperature, humidity, occupancy. They had a working prototype, got a letter of intent from a property manager for five thousand units. He went to a mold maker and got a quote for forty-eight thousand dollars and sixteen weeks. He couldn't afford it and didn't have sixteen weeks. Lucas: Classic trap. So what did he do? Luna: He found a shop that does rapid tooling. They quoted him sixty-eight hundred for an aluminum mold, two weeks. He ran five thousand parts, sold them to the property manager, got feedback, made three small design changes, and then ordered a steel mold for the next batch of twenty thousand. Lucas: That's exactly the playbook. And it's becoming more common because the service providers have gotten better. Companies like Protolabs have built their whole business around rapid injection molding. They use aluminum molds and automated CNC machining to cut them fast. Formlabs has a different approach — they 3D print the mold inserts using a high-temperature resin. Luna: 3D printed molds? I've seen those for prototyping, but for real production parts? Lucas: It's niche, but growing. Formlabs sells a material called Rigid 10K, which can withstand injection molding pressures for short runs — we're talking fifty to a few hundred parts. The advantage is you can make complex cooling channels that are impossible to machine. That can reduce cycle time significantly, even for a short run. Luna: So if you have a part that needs to cool quickly, a 3D printed mold with conformal cooling can actually give you better part quality than a machined aluminum mold. Lucas: Yeah, and that's a real differentiator. But the mold life is short — maybe two hundred shots before the cavity starts to degrade. So it's only for very low volumes or pilot runs. For most startups, aluminum is the sweet spot. It machines quickly, it's cheap, and it'll get you through the first few thousand parts. Luna: But there are tradeoffs beyond mold life. Aluminum doesn't conduct heat as well as steel, so cooling is slower. And the surface finish isn't as good — you might need to polish the cavity or add a texture. Lucas: True. And you have to be careful about the plastic material. Some engineering resins — like glass-filled nylon or polycarbonate — are abrasive and will wear an aluminum mold faster. You might only get a few hundred parts before the dimensions drift. So you need to match the material to the mold. Luna: So if you're molding something like polypropylene or ABS, aluminum is fine. But if you need high-temperature or reinforced plastics, you might need steel even for a short run. Lucas: Right. And that's where the rapid tooling providers have gotten smart. They'll ask about your material, your annual volume, and your timeline, and they'll recommend a mold material accordingly. Some even offer a 'soft steel' option — something like 4140 pre-hardened — that splits the difference. It's more expensive than aluminum but still faster and cheaper than P20. Luna: And this is something a lot of first-time hardware founders don't know to ask about. They see 'injection molding' and think one price, one lead time. But the mold material is a huge variable. Lucas: Absolutely. And that's what we're trying to cover here — the practical decisions. If these conversations are useful for what you're building or running, I'll mention that a small group of listeners helps keep this show ad-free and independent. It's at buymeacoffee.com/fexingo. Seriously, that support is what lets us do deep dives like this without sponsors calling the shots. Luna: Yeah, and we really appreciate it. It means we can cover niche topics like rapid tooling without having to pitch a tooling company. Lucas: Exactly. So back to the IoT sensor founder — he got his five thousand parts, learned that the battery compartment was slightly too tight, and the LED wasn't bright enough. He updated the CAD, and then ordered his steel mold with those changes baked in. Luna: That's the hidden value of rapid tooling. It's not just about cash flow — it's about learning. You get real parts in the field, not just prototypes. You discover issues you'd never catch in a lab. Lucas: And those issues would be incredibly expensive to fix in a steel mold. A single design change to a production mold can cost ten thousand dollars and add weeks. With rapid tooling, you can iterate the mold itself. Luna: So let's talk about the actual process. How does a startup find a good rapid tooling partner? Is it just Protolabs and Formlabs, or are there smaller shops? Lucas: There are hundreds of small mold shops across the US and China that specialize in rapid tooling. The key is to look for shops that advertise 'bridge tooling' or 'prototype tooling.' A lot of them are family-run businesses that have been doing this for decades but recently started marketing to startups. The advantage of a smaller shop is they'll often do design for manufacturability consulting for free. Luna: What should a founder have ready before they call? Lucas: A 3D CAD file in STEP or IGES format. A clear specification of the plastic material — not just 'ABS,' but the exact grade, like 'ABS MG94.' And an estimate of the annual volume. They'll also want to know the desired surface finish — SPI standards go from A1 to D3. For rapid tooling, most people accept a C3 finish, which is a machined surface with visible tool marks. Luna: And the part design itself — are there any features that make rapid tooling harder? Lucas: Deep ribs, tall bosses, and sharp internal corners can be tough to machine in aluminum because the tool can't reach or breaks easily. Undercuts also complicate things — you might need side actions, which add cost. For rapid tooling, simpler is better. If your part has complex geometry, consider splitting it into two pieces that snap together. Luna: Draft angles too, right? Aluminum molds need at least one degree of draft per side, sometimes two, because the plastic shrinks onto the core and you need to eject it without damaging the part. Lucas: Yes. And this is something that trips up a lot of designers who come from 3D printing, where you don't need draft. With injection molding, even rapid tooling, you need draft on every vertical wall. Otherwise the part sticks and you waste a cycle. Luna: So rapid tooling isn't just a cheaper mold. It forces you to design for manufacturing from the start. Lucas: Exactly. And that discipline pays off when you eventually move to steel. A part that runs well in an aluminum mold will run even better in steel, because steel is stronger and conducts heat more evenly. Luna: I want to ask about tolerances. How tight can you hold with an aluminum mold compared to steel? Lucas: Typically, aluminum molds can hold plus or minus 0.005 inches on most dimensions. Steel can go to plus or minus 0.002 or better. For most consumer products, 0.005 is fine. But if you have press fits or bearing seats, you might need to machine those features in a secondary operation after molding. Luna: So rapid tooling might not work for precision mechanical parts. But for enclosures, housings, and simple brackets, it's perfect. Lucas: That's right. And that covers the vast majority of hardware startup needs. The IoT sensor, the smart home device, the medical monitor, the robot shell — all of those can be done with rapid tooling for the first production run. Luna: What about metal parts? Does rapid tooling apply to things like die casting or stamping? Lucas: It does, but it's less common. For die casting, you can use softer steels or even beryllium copper for the die. But the economics are different because die casting dies are already expensive. For stamping, you can use 'soft tooling' — a steel die that's not hardened — for short runs of a few thousand parts. Luna: So the principle is the same: reduce tool cost and lead time by using a less durable material, accept a shorter tool life, and plan to transition to hard tooling when volume justifies it. Lucas: Exactly. And the math is straightforward. Let's say your steel mold costs $50,000 and lasts 500,000 cycles. That's ten cents per part in tooling amortization. Your aluminum mold costs $5,000 and lasts 5,000 cycles. That's one dollar per part. So for the first 5,000 parts, you pay $1 each in tooling with rapid tooling versus $10 each with steel — assuming you could even afford the steel upfront. Luna: Plus, you get revenue from those first parts, which can fund the steel mold. That's a much better capital path than spending $50,000 before you've made a single sale. Lucas: Right. And that's why rapid tooling has become a standard step in the hardware startup playbook. It's not a compromise — it's a strategic choice. Luna: I think the misconception is that rapid tooling is only for 'cheap' products. But actually, it's for any product where you need to validate the market before investing in high-volume tooling. Lucas: Yeah, even aerospace and medical startups use it. They just need to be careful about material compatibility and process validation. For medical devices, you might need to run the mold in a cleanroom or use medical-grade materials. But the approach is the same. Luna: One last thing — what about the future? Are we going to see more 3D printed molds replace aluminum? Lucas: I think for very short runs — under a thousand parts — 3D printed molds will grow because they offer design freedom. But for the two thousand to ten thousand range, aluminum is the sweet spot. And I think we'll see more hybrid approaches: 3D printed inserts for complex cooling, mounted in an aluminum or steel base. Luna: So the takeaway for any hardware founder: when you're ready to move beyond prototypes, don't jump straight to a production mold. Look at rapid tooling first. It might save you six figures and six months. Lucas: And give you the feedback you need to make your product better before you scale. That's a huge competitive advantage. I'm Lucas. Luna: I'm Luna. See you next time.