The Boiler They Lit By Hand

▶️ 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).

Two men stood inches from an old boiler with no way to see how much gas was building inside it. One was watching the flame port. One had his hand on the gas valve. When the pilot caught, the whole firebox let go. Today’s issue is about the machines we ask people to stand next to — and the rule that says those people should never have to.

It was the Wickes boiler — old, and everyone at the Wynnewood Refinery knew it. It had no burner management system, so you couldn’t light it from a safe, remote spot the way a modern boiler restarts itself. You lit it by hand. And it had no gas-flow gauge, so nobody standing at it could tell how much natural gas had already drifted into the firebox before the flame ever showed.

On the morning of September 28, 2012, two workers went to restart it. Billy Smith, 34, was posted at the fire-eye, watching the port for the pilot to catch — inches from the steel. Russell Mann, 45, stood a few feet away with his hand on the gas valve. Gas had been flowing into a cold, unpurged chamber with no one able to measure it. When the pilot lit, that accumulated gas didn’t burn as a flame — it detonated. Billy Smith was killed instantly. Russell Mann died of his injuries about two weeks later. The company had known about prior detonations of that same boiler that had hurt workers before.

Here’s the rule this shaped: 29 CFR 1910.119 — Process Safety Management of Highly Hazardous Chemicals. OSHA cited the refinery under PSM, and in 2020 the 10th Circuit Court of Appeals upheld it. The fight was technical — the boiler itself was running on utility natural gas that day, not refinery gas, so the company argued PSM didn’t apply. The court said it did: that boiler was plumbed into the fluid catalytic cracking unit and the alkylation unit, both of which held highly hazardous chemicals in covered quantities. You don’t get to draw a little box around one piece of equipment and call it exempt when it’s wired into a covered process.

Translated to the floor: a manual light-off on an unpurged furnace is a bomb you build one valve-turn at a time. The purge cycle isn’t paperwork — it sweeps the fuel out of the box so ignition finds a flame instead of a cloud. A burner management system exists so a human is never the fire-eye. And “we’ve always lit it this way, and it’s mostly fine” is exactly the story every one of these boilers tells right up until the morning it doesn’t. Billy Smith should not have been standing inches from that port. Russell Mann should not have had his hand on that valve. The equipment failed them long before they ever struck the light.

The Full EHS Picture. (S) Two men died — a furnace explosion at manual light-off, the classic unpurged-firebox detonation. (E) The environmental damage was, honestly, a near-miss more than a disaster: the blast was largely contained to the boiler, but investigators noted flying debris narrowly missed the adjacent alkylation and FCC units — the ones holding the highly hazardous chemicals. Had that debris struck a few feet over, this becomes an offsite release story instead of a two-fatality one, and the reason PSM (and EPA’s Risk Management Program, 40 CFR 68, which shadows it) reaches equipment like this boiler at all. (H) The acute health toll was the two workers; no community exposure was recorded. The plain-language line: most of these incidents are E, H, and S at once — the same unpurged firebox that killed two men was one bad bounce of debris away from poisoning the air over the fenceline. This one stayed “just” a fatality because the shrapnel missed. That’s not a control. That’s luck.

Fifth Circuit vacates EPA’s methylene chloride rule (Sept. 15, 2026). A unanimous panel struck down EPA’s 2024 TSCA risk-management rule for methylene chloride and the “unreasonable risk” finding under it, sending it back to the agency. So what for safety leaders: this was EPA/TSCA — it does not touch OSHA’s workplace standard, 29 CFR 1910.1052, which still stands with a 25 ppm 8-hour PEL and a 12.5 ppm action level. If a vendor or manager tells you the paint-stripper rules “got thrown out,” that’s the EPA consumer/industrial rule, not your PEL. Keep your exposure monitoring, ventilation, and respiratory program exactly where they are.

The federal heat rule is still not final. OSHA’s proposed heat injury and illness prevention standard, first proposed in 2024, remains proposed and stalled in 2026 — but the Heat National Emphasis Program is active, meaning inspectors are still looking. So what: don’t wait for the rule to do the obvious. Water, rest, shade, acclimatization for new and returning workers, and a plan you can actually point to are what an inspector — and a heat-sick worker — will ask for.

A crew was bringing a fired heater back online after a short outage. The startup checklist called for a five-minute purge before light-off. The board operator had done this restart a hundred times, the unit was “barely cold,” and the shift was running behind — so the purge got trimmed to about a minute by feel. The heater lit with a hard whump that popped a sight-glass and rattled the platform. Nobody was hurt. That’s the only reason we get to call it a fail instead of a funeral.

No blame here — this is how normal work drifts. The checklist time felt like padding because it had never visibly paid off. That’s the trap: a purge cycle that works is invisible, so it starts to feel optional. The fix isn’t to yell at the operator; it’s to make the purge un-skippable — interlocked timing, a two-person confirm, a light-off permissive that won’t arm until the purge completes. The person closest to the work saw the shortcut first. The question every leader should ask is whether it’s safe for them to say so out loud before the whump, not after.

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.

Walk over to your worst light-off, restart, or “we do it by hand” procedure — the one everybody knows is a little sketchy — and ask two questions: does the safe sequence (purge, permissive, interlock) physically prevent the dangerous one, and does anyone have to stand in the blast path to do their job? If the answer to the first is “no” and the second is “yes,” you’ve found your Wickes boiler. Fix it before the calendar does.

🎬 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!!!

Verified sources: Safety+Health Magazine (NSC), “Process safety management – the Wynnewood refining case” — https://www.safetyandhealthmagazine.com/articles/21571-on-safety-process-safety-management-the-wynnewood-refining-case ; U.S. 10th Circuit, No. 19-9533 (2020) — https://law.justia.com/cases/federal/appellate-courts/ca10/19-9533/19-9533-2020-10-27.html ; OSHA 29 CFR 1910.119 (PSM) and 29 CFR 1910.1052 (methylene chloride) — https://www.osha.gov ; Fifth Circuit methylene chloride decision (Sept. 2026) — https://environmentalhealthsafetybrief.sidley.com/2026/09/16/in-landmark-tsca-decision-fifth-circuit-vacates-epas-methylene-chloride-rule-limits-epas-interpretation-of-unreasonable-risk/ .