Nordbo Robotics ads

Manual grinding can take hours. And when it does, it quietly becomes the bottleneck that limits output across every shift. The issue is not only time on the part. It is also the hidden cost around it: ▶️ Inconsistent staffing, when it is hard to keep the task covered reliably ▶️ Rework, when finish quality changes from one operator to the next ▶️ Fatigue and vibration exposure, when demanding manual work wears people down ▶️ Delayed delivery commitments, when a slow finishing step holds back the entire flow There is another way to handle it. Robot motions can be taught by demonstration, so the process learned from a skilled operator can be repeated with consistent quality. New parts can be taught quickly, without lengthy coding and without turning changeovers into production downtime. How much capacity is still tied up in manual grinding at your plant?

If you bring up robot programming with a production manager, the reaction is almost always the same: 😮💨 - Weeks of expensive programming hours. - Downtime during implementation. - Every new part requires a full reprogramming cycle that stops the line. In a foundry where a line stoppage costs thousands of dollars per hour, that reaction is completely reasonable (based on how programming used to work). The math on automation never added up. So most foundries stopped asking. But the reality is different, today. Setting up a new program takes minutes or hours, not days or weeks. Your operator teaches the robot by demonstration. No specialist required. No waiting for external support every time a new part comes in. Two things change immediately: The ROI calculation works for smaller series. High mix, frequent changeovers, low volume. These are no longer reasons to postpone. Your operators stay in control. They are the ones who programme the robot. Their process knowledge is what makes it work.

Deburring, grinding, polishing of casted aluminum parts… I have seen a lot of foundries that truly could benefit from automating these processes. But they have been stalling because of challenges with high variation of parts and small batches. Small batches used to be a show stopper for profitable automation, but that is not the case anymore. With our Mimic technology the operator teaches the robot a new part in minutes. This changes the ROI calculation completely. Share this with someone who needs to see it!

The foundry industry is conservative. With good reason. If one process stops, the entire line stops. That is an extremely expensive problem. In many operations, downtime is measured in half-hours per shift — and keeping that number down is absolutely critical. That is why so many are afraid to change a running process. If it ain't broke, don't fix it. The problem is when manual processes become the unpredictable variable that is actually causing the downtime and delays. The key to removing that fear is technology that can be implemented and reprogrammed without shutting down production for weeks. When we can show that a new part can be set up in minutes (not days) the biggest barrier to optimising disappears.

In a foundry, cycle time is everything. Are you getting 60 parts out per hour, or 62? That might not sound like much. But two extra parts per hour, on a line running 24/7 across three shifts, adds up to serious money on the bottom line. Improve your cycle time by 20% in post-casting operations by automating it, and you can potentially eliminate an entire workstation. That is why I always look at the bottlenecks. Have you solved a bottleneck by adding more people to a manual step? What does that cost in time? What does it cost in quality? Optimising cycle time in finishing is often where the fastest and largest gains are available in a modern foundry. The machine is rarely the constraint. The manual step downstream usually is.

Most foundries and metal workshops considering automation ask the wrong question. They ask: "Which robot can grind better?" But that's not the real problem. The real problem is capturing craftsmanship. Your best operator knows exactly how much pressure to apply, which angle works best, where the tricky spots are on each part. That knowledge lives in their heads and their hands. And traditional robot programming can't capture it. This is what we've solved at Nordbo Robotics. Our Material Removal Systems use Mimic technology. Here's how it works: Your operator shows the robot what to do. No coding. No complex programming interface. Just demonstrate with a joystick and enter pressure to get a consistent finish. What this means: Your operator's expertise becomes repeatable: Every shift gets the same quality, even when your best people aren't there Fast changeovers: New part? Show the robot once. Done. Your people stay in control: They're not being replaced - they're being amplified