Latest / Hardware Startups with Fexingo: Devices, Robotics, and Manufacturing Tech Companies / How Hardware Startups Avoid Tooling Traps
Transcript
- Lucas: Most hardware founders I talk to think they understand their bill of materials until they get hit with the tooling invoice. And that is where a lot of good ideas go to die, not because the product is bad, but because the mold cost blew the budget. Luna: Right, and they usually realize this only after they have already signed the contract with the molder who has no incentive to tell them they designed something impossible. Lucas: Exactly. So today we are looking at one specific lever that founders can pull before they ever send files to a factory: part geometry and its direct impact on mold complexity. We will use the example of a simple consumer electronics enclosure to show how small design tweaks can save you ten thousand dollars or more in upfront tooling costs. Luna: That sounds like a classic case of trading early engineering time for later manufacturing savings. Is it really that binary? Lucas: It is surprisingly binary when you look at the mechanics of injection molding. Let us take a hypothetical device case, something like a smart home sensor housing. If you design it with vertical walls, you are asking for a straight forward two plate mold. But if you add an undercut—a feature that prevents the part from sliding straight out of the mold—you suddenly need side actions or lifters. Luna: And those side actions are what drive up the price, right? Because they add moving parts to the mold itself. Lucas: Precisely. Each side action requires slides, wedges, springs, and additional steel machining. That increases the mold cost by anywhere from three to eight percent per feature. For a high volume part, that adds up fast, but even for a low volume run, the non-recurring engineering cost spikes significantly. Luna: So the advice is basically to avoid undercuts whenever possible? Or is there a middle ground? Lucas: There is a middle ground, but it starts with avoidance. You can often achieve the same aesthetic or functional result by using snap fits or interlocking features instead of continuous undercuts. It requires a bit more creativity in the industrial design phase, but it keeps the mold simple. Luna: I remember seeing a drone company that redesigned their landing gear specifically to eliminate a draft angle requirement. Did that help them with the tooling? Lucas: That is a great example. Draft angles are another silent killer of simplicity. Every surface that pulls against the mold steel needs a slight taper, usually half a degree or more. If you ignore that, the part sticks, the ejection pins mark the plastic, and you end up with scrap. Adding draft does not necessarily increase cost, but fighting against it causes delays and reworks. Luna: Delays are actually more expensive than the extra steel sometimes, especially if you are missing a trade show deadline. Lucas: Absolutely. Time to market is a cost center too. Another major factor is wall thickness consistency. If your part has thick sections next to thin ones, you get sink marks and warping. To fix that, you might need longer cooling times or even conformal cooling channels in the mold, which is a premium service. Luna: Conformal cooling sounds fancy, but it basically means the mold takes longer to make and costs more upfront. Lucas: Yes, and it complicates the maintenance too. Keeping wall thickness uniform, ideally between two and three millimeters for most thermoplastics, simplifies the entire process. It allows for standard cooling lines and faster cycle times, which lowers your per-part cost over the life of the project. Luna: So we are talking about designing for the machine, not just for the user interface. Lucas: Exactly. You have to design for the constraints of the manufacturing process. Let us talk about material selection briefly. Choosing a standard resin like ABS or polycarbonate is cheaper and easier to source than exotic blends. The latter might offer better performance, but they often require specialized drying and processing conditions that limit your pool of capable vendors. Luna: Limiting vendor options is risky. What if that one specialist goes out of business or raises prices? Lucas: It is a supply chain risk. By sticking to common materials, you keep your leverage. You also reduce the risk of material contamination or variability. It is a boring choice, but boring is profitable in hardware. Luna: Boring wins. I like that. But what about cosmetic requirements? Do founders ever sacrifice too much for manufacturability? Lucas: They do, but usually unnecessarily. A well-designed part can still look premium. Texture helps hide ejector pin marks and flow lines. Using a light texture, like a V-groove or a matte finish, masks minor defects without requiring a perfectly polished mold surface, which is expensive to maintain. Luna: So texture is both a functional fix and a cosmetic mask. That is efficient design. Lucas: It is. Now, let us pivot to a quick moment of support for the show. If these conversations are useful for what you are building or running, the way these stay ad-free is listener support. Buy me a coffee dot com slash fexingo. It helps keep the research deep and the ads out. Luna: Yeah, it is a small thing that makes a big difference in keeping the quality high for everyone listening. Lucas: Thank you. Back to the technical details. One area where founders often stumble is the placement of mounting bosses. These are the little pillars inside the case where screws attach. If they are placed incorrectly, they create stress points that crack the part during assembly. Luna: Cracking during assembly is a nightmare for quality control. How do you prevent that? Lucas: You reinforce them with gussets and ensure they are supported by thicker walls underneath. Also, avoid placing them near sharp corners. Stress concentrates there. Think of the part as a structure, not just a shell. The internal geometry matters just as much as the external shape. Luna: Internal geometry is usually the part of the design that gets ignored until it breaks in the field. Lucas: It is. And field failures are the most expensive kind. They destroy brand reputation. Another tip is to consider the parting line. Where the two halves of the mold meet leaves a tiny flash of excess plastic. You can hide this line along natural seams or edges, making it less visible and easier to trim. Luna: Hiding the seam is a classic trick. It saves post-processing time and labor costs. Lucas: Labor costs add up quickly in high-volume production. Even a few seconds per part multiplied by millions of units is significant. Reducing secondary operations is a huge win. You want to get the part out of the mold and into the box with minimal touching. Luna: Minimal touching is the goal. What about tolerances? Do founders specify tighter tolerances than necessary? Lucas: Constantly. Specifying plus or minus zero point zero five millimeters on a large housing is unnecessary and expensive. Standard tolerances for injection molded parts are usually around plus or minus zero point one millimeters. Tighter tolerances require tighter control of the molding process, slower cycle times, and more frequent mold inspections. Luna: So loose tolerances save money unless the fit is critical. That seems obvious now, but I bet many engineers default to tight specs out of habit. Lucas: Out of habit or fear. They are afraid the part won't fit, so they over-specify. But you can achieve precise fits through proper design, like using interference fits or flexible clips, rather than demanding micron-level precision on every dimension. Luna: Using flexibility to absorb error is a clever way to bypass rigid tolerance chains. Lucas: It is. It shifts the burden from the mold maker to the designer. And designers should be empowered to make those calls. It is not just about making things cheap; it is about making things reliable and scalable. Luna: Reliability is the ultimate cost saver. If it works the first time, you don't have to fix it later. Lucas: Exactly. Let us look at one more concrete example. Consider a battery cover. If you design it with a complex latch mechanism integrated into the plastic, you need multiple moving parts in the mold. Instead, you can use a separate metal hinge or a simpler screw-down design. It reduces the mold complexity dramatically. Luna: Simplifying the mechanism reduces the number of cavities or actions needed. That is a direct cost reduction. Lucas: It is. And it makes troubleshooting easier. If a plastic latch fails, you know exactly why. If you have a complex mechanical system embedded in the part, diagnosing the failure mode becomes much harder. Luna: Diagnosability is underrated. You want to know exactly where the problem is when it happens. Lucas: It is crucial for iteration. In the early stages, you want to fail fast and cheap. Overly complex designs slow down that learning loop. Simple designs allow you to test hypotheses quickly without waiting weeks for a new mold revision. Luna: Speed of iteration is everything for a startup. You cannot afford long lead times on tooling changes. Lucas: Correct. So the takeaway is to design for simplicity. Avoid undercuts, maintain uniform wall thickness, use standard materials, and relax tolerances where possible. These are not compromises; they are strategic decisions that protect your capital and accelerate your timeline. Luna: Strategic simplicity. That is a phrase I will remember. It changes the narrative from cutting corners to building smarter. Lucas: It does. And it puts the founder in control of the manufacturing conversation rather than being dictated to by the supplier. When you speak the language of the mold, you gain respect and better terms. Luna: Language is power in any negotiation. Especially when you are dealing with experts who hold the blueprints. Lucas: True. As you move from prototype to production, remember that every line in your CAD file has a cost implication. Be intentional about every curve, every hole, and every surface. Luna: Intentional design is the difference between a hobby project and a viable business. Lucas: Viable. That is the key word. Hardware is hard enough without adding unnecessary complexity to the mix. Keep it simple, keep it standard, and keep it moving. Luna: Moving forward, literally and figuratively. Good advice for anyone holding a stylus over a CAD screen today. Lucas: We will leave you with that thought. How many unnecessary features are hiding in your current design, costing you more than you realize?