After last week was all about the message, this week went back to the metal, or really back to the plastic. I spent it making the body fit the robot exactly and making sure it can take a beating. No new features, just precision and abuse.
The snug fit was the finicky part. My earlier shells were close, but close is not good enough when a screen, a battery, and a board all have to sit exactly where they belong. I remeasured every part of the robotics inside and redesigned the shell down to the millimeter, so the electronics click into place and nothing rattles. When it finally snapped together with no gaps and no glue, that was a very good feeling.
Then I put it through real drop tests, the kind that mirror the FCC style testing a real product has to pass. Not tossing it once and hoping, but dropping it from set heights, onto set faces, over and over, and writing down what happened each time. A product that only survives a lucky drop is not a product, it is a prototype that got away with it.
The other half of the week was learning what my printer can and cannot actually do. I ran the same part in a few different materials to see how each one behaved. One printed clean but cracked on impact. Another was tough but came out stringy and ugly. A flexible one bounced beautifully but would not hold a tight tolerance. Every material taught me something about the tradeoff between looking good, printing clean, and surviving the floor.
By the end I had a much better map of the edges: which material to use where, how thin I can go before a wall gives out, and where the printer starts to lie to me about what it can hold. That kind of knowledge never shows up in a demo, but it is the whole difference between a one time build and something I could actually make more than one of.
This was not a glamorous week. It was measure, print, drop, and measure again. But NEURA now has a body that fits like it was meant to and takes a hit like it expects to, and that is exactly what I needed before the finish line.