Nine Atmospheres and a Megaphone

▶️ Prefer to watch? Every story’s a short video → https://www.youtube.com/@GetSafetyIntel · 📣 Got a near-miss? Share it anonymously (QR + link at the bottom).

Four in the morning, North Sea, a job the crew had done hundreds of times. The hazard that killed five men that night wasn’t exotic — it was stored energy, the same thing sitting behind every pressurized line and locked-out machine you’ll touch today. The only things standing between the crew and nine atmospheres of it were a written procedure and a word shouted through a megaphone. That was the whole safety system. It wasn’t enough.

It was a Saturday, 4:00 a.m., aboard the semi-submersible drilling rig Byford Dolphin, working the Frigg field in the Norwegian sector. Four saturation divers lived inside a chamber system on deck, their bodies held at 9 atmospheres of pressure so they could work the seabed without decompressing for weeks. A diving bell shuttled them down and back. The bell docked to the chambers through a short passage — a trunk — sealed by a clamp worked by two tenders.

The divers were Edwin Coward (35) and Roy Lucas (38), both British, resting in chamber 2; and Bjørn Bergersen (29) and Truls Hellevik (34), both Norwegian, who’d just been winched up and were climbing in through the trunk. The tenders on the clamp were William Crammond (32) and Martin Saunders.

The procedure was five steps, in order: close the bell door, seal it with a touch of pressure, close the chamber door, slowly bleed the trunk down from 9 atmospheres to 1, then open the clamp. The first two steps were done. Before the inner door was shut and before the trunk was bled, the clamp came open.

Nine atmospheres found one atmosphere in a fraction of a second. The air left the chamber system with a force nobody walks away from. All four divers died instantly. Crammond, standing at the clamp, was killed by the bell as it was blown off its seat. Saunders, the other tender, was critically injured. The forensic report later described fat precipitating out of the divers’ blood under the pressure drop, and one diver forced through a gap in a jammed door barely two feet across. I’ll leave it there. These were five real men, and four of them left families who spent the next 26 years fighting for answers.

Why it happened — and why blame misses it. The official commission called it human error: a tender opened the clamp too soon. But look closer, the way HOP teaches us to. The tenders coordinated this life-or-death sequence through a megaphone, over the roar of a working rig and the sea — nobody could reliably hear who was where in the sequence. The crew had been on shift more than 12 hours. And here’s the one that matters most: the system had no interlock. Nothing — no gauge, no mechanical lock, no fail-safe — physically stopped that clamp from being opened while the trunk was still pressurized. A tired hand at 4 a.m. was the last line of defense, and the design put it there. Error is normal. The question is never who slipped. It’s why a slip could kill five people.

The rule it shaped — name it, cite it. In the U.S., this exact work lives under 29 CFR 1910 Subpart T — Commercial Diving Operations, with the decompression-chamber requirements at 1910.423. But the lesson is bigger than diving, and it’s written into the standard every one of us lives by: 29 CFR 1910.147, the control of hazardous energy — lockout/tagout. The core of 1910.147 is one idea: you don’t break into a system until the energy behind it is at zero and proven to be. Translated to the floor — you never crack a flange, pull a guard, or open a clamp on faith that it’s probably depressurized. You bleed it, you gauge it, you verify. One month after Byford Dolphin, the Norwegian certifier Det Norske Veritas made its rule final for all bell systems: connecting mechanisms must be built so they cannot be operated while the trunk is pressurized. That’s the fix written in blood — an engineering control that takes the fatal choice out of a tired human’s hands entirely. Hierarchy of controls, top of the list: don’t trust a procedure to do what a piece of steel can do.

The Full EHS Picture. Most incidents are E, H, and S at once — the same failure that kills a worker poisons a creek and sickens a town. This one is the rare exception, and honesty means saying so. Environmental: there was no release — nothing spilled, burned, or reached the water. The energy that killed these men was pressure, contained to a steel box on deck. Health: this is the story that outlived the night. The men who pioneered North Sea saturation diving were, in the words of a later BBC investigation, treated as underwater “guinea pigs” — and many carried lasting damage from the work itself: decompression sickness, neurological problems, bone death, hearing loss. The North Sea Divers Alliance, formed by survivors and the families of the dead, pushed for decades until the Norwegian government finally paid compensation in 2009. Safety: the S is the clamp — but the H is the reason the S matters. The “environment” in this story was the artificial 9-atmosphere world these workers’ bodies lived in, and it took its toll long after the headlines faded.

Trenching enforcement is surging — right now. In the last month OSHA has stacked up excavation citations: a Frankfort, Illinois utility contractor cited Oct 21 for repeatedly exposing crews to an unprotected trench; six New York and New Jersey construction employers cited Oct 13 for willful and serious excavation violations, $648,094 combined; and a Pennsylvania contractor cited Oct 7 after a young worker was killed in a cave-in. So what for safety leaders: 29 CFR 1926.652 is not a judgment call. Five feet deep without a protective system — slope it, shore it, or drop in a trench box — and you’re exposed, both to a collapse and to a willful citation. A competent person inspects the trench daily and after every rain, before anyone enters. If OSHA is this hot on it, assume your next inspection starts at the edge of a hole.

A maintenance death that sounds a lot like today’s story. Follow-up reporting this week continues on the September 2 death of Bradley Clouse, 33, a maintenance worker crushed inside robotic equipment at the Astemo plant in Berea, Kentucky. OSHA and Kentucky OSH are investigating. Local coverage noted the machine was reportedly locked at the time yet somehow moved — and the exact circumstances are still under investigation. (Worker accounts alleging broader lockout gaps are unconfirmed, and no citation has issued; we’ll report what the investigation finds, not what we assume.) So what: a padlock hanging on a disconnect is not the same as zero stored energy inside the machine. Residual hydraulic pressure, a loaded spring, a part held up by gravity, a charged capacitor — lockout isolates the source, but the try-out and stored-energy-release steps of 1910.147 are what make it safe to reach in. Same lesson as that clamp in 1983: prove it’s at zero before your body is the thing that finds out it wasn’t.

A tech was sent to swap a clogged filter on a transfer line the board said was shut down and drained. He backed off the housing bolts the way he had a hundred times. On the last turn the cover let go and sprayed him with product still under residual pump pressure — the upstream valve had been closed, but nothing downstream had been bled, and the only gauge was on the other side of that closed valve. He got lucky: safety glasses and a face shield, a bruise, a change of clothes, a story. No lost time.

Here’s the blameless read: he didn’t fail to follow the procedure. He followed it — the procedure said isolate, and the valve was shut. What the procedure never said was verify zero at the point you’re opening, and the plant hadn’t put a bleed or a gauge where the work actually happens. The gap was in the system, not the man. Fix the system: a bleed point and a local gauge at the break, and a step that reads confirm 0 psi here before a wrench touches a bolt. The next tech shouldn’t need luck.

Got a fail or a near-miss? Hit reply, or scan the QR below — we’ll feature it anonymously — no names, no company, no blame. Just the lesson, so the next crew doesn’t learn it the hard way.

Pick one spot in your plant where the plan is that it’s probably depressurized by now — a filter housing, a flange, a sight glass, a pump seal, any connection a person routinely breaks — and ask whoever does the job two questions: where do you bleed it, and where’s the gauge that proves it’s at zero? If either answer is a shrug, you just found your Byford Dolphin clamp. Put a bleed and a local gauge there this month. Steel beats faith.

🎬 Rather watch than read? Every Safety Intel story is a short video on YouTube → https://www.youtube.com/@GetSafetyIntel — subscribe so you never miss one. 📸 Share YOUR near-miss (100% anonymous): https://scantoconfess.com — or just scan the QR above. 📤 Forward this to a crew member who needs it today.

Please stay Safe & Hydrated!!!