Opening the cabinet of a thirty-year-old machine

Last month one of our engineers opened the control cabinet of a stamping press that's been running since the nineties. Behind the door: a nest of wiring in three different colours of tape, two relays held in with a zip tie, and a single red mushroom button that, as far as anyone could tell, only cut power to the indicator light.

That's not unusual. Most machines we're asked to look at have been running fine for years. Nobody touches them because touching them feels risky, and because "it still works" is a perfectly reasonable reason to leave something alone. Until the insurer asks for a risk assessment, or a new operator gets a scare, or the machine needs to move to a different part of the building and suddenly someone has to prove it's safe to reconnect.

What actually needs to change is rarely obvious from outside

The motor still turns, the product still comes out the other end, so from the operator's chair everything looks fine. What we're checking is underneath that: whether the emergency stop actually removes power from the moving parts, or just from a light. Whether there's a proper category of safety circuit for the risk the machine presents, or whether it's relying on a single relay from 1994 that was never tested against anything. Whether guarding and interlocks match what the machine now actually does, especially if it's been modified since it left the factory, which almost all of them have.

On that same press, we found three separate additions from over the years: a sensor bolted on by a previous maintenance contractor, a bypass switch someone installed to keep production going during a fault we'll never know the cause of, and a second control panel wired in parallel with the first because the original one sat in an awkward spot. None of it was documented. None of it appeared on any drawing. Each addition made sense on the day someone made it, and together they turned the electrical system into something nobody could fully explain anymore.

That's usually the real project, more than the safety button itself: rebuilding the wiring so it matches reality, redrawing the electrical schematics so the next person doesn't have to guess, and putting the control logic into a PLC that can actually enforce the sequence instead of hoping the relays fire in the right order. The safety circuit is one part of that, but it only holds up if everything feeding into it is understood and documented too.

We usually find a version of the same list on machines from that era:

• A single emergency stop that only interrupts part of the circuit, not all hazardous motion
• Guards or interlocks that don't match what the machine now does after years of modifications
• No current risk assessment, or one that was written for a different configuration
• Wiring that works but exists nowhere on paper
• Relay logic that's been patched so many times nobody fully trusts it anymore

None of that means the machine is dangerous to run tomorrow morning. It means nobody can currently prove that it is safe, which is a different problem, but one that eventually catches up with you: at inspection, at insurance renewal, or the day something actually goes wrong and someone asks to see the paperwork.

What surprised the client, more than the technical findings, was how much of the work turned out to be about clarity rather than replacement. The press didn't need a new motor or a new frame. It needed its control system rebuilt around a proper safety category, new drawings that match what's actually installed, and a PLC that enforces the sequence instead of relying on relays nobody can fully account for. The machine itself was fine. The story of what had been done to it over thirty years was the part that needed sorting out.