THIS MONTH'S PUZZLER
An explosion killed a welder who was using a torch in an attempt to cut loose an anchor bolt on top of a polymer atmospheric overflow tank, in preparation for removing the tank's agitator. The original work permit covered removal of the anchor bolts by pneumatic tool but the bolts were rusted to the nuts and wouldn't budge. Because the window for removing the agitator was closing, maintenance higher-ups okayed using the torch. The welder thought he could get at the bolts from the outside, away from the agitator shaft, which didn't include a seal. The clearance space between the tank and agitator shaft was open; calculations showed the monomer didn't produce sufficient vapor pressure to form a flammable mixture at normal temperatures — so, the vessel was classified as an uncovered process. The overflow tank was connected by vent to two others. Lockout procedures were deemed safe: all the intake lines were isolated except for the common vent. The plant continued to operate, filling the adjacent overflow tank while the welder worked. Hydrocarbon sniffing took place before the welder began but not during the welding. Safety decided the mishap was the welder's fault. What do you think?
CONNECTIONS DEMAND CAUTION
This note is based on Chemical Safety Board findings, so I don't feel right in taking credit for all of this response since others did the investigation.
With that said, it is not the welder's fault. Maybe he should have been aware of the potential for conditions to change, but most maintenance personnel are not completely familiar with all the equipment connections on every piece of equipment in a plant. That is why they rely on others more familiar with the specific process/equipment to prepare and validate equipment-specific isolation and hot work procedures. This was an incident where the safety risk of the common vent allowing flammable vapors to be displaced from the tank being filled and forced into the overflow tank was overlooked. This is a great reminder of two fundamental principles of process safety: 1) every connection is a potential hazard source that must be accounted for; and 2) expect the unexpected. When performing hot work in an area where flammable materials are processed, assume that conditions can change and that unexpected events will occur. To guard against such an eventuality it should be standard practice to require continuous monitoring of the area during hot work rather than only at the start.
Len Riker, operations project leader
Penford, Cedar Rapids, IA
DON'T BLAME THE WELDER
I believe the question is "whose fault was it?" Either it was the fault of the maintenance higher-ups who okayed the use of an open flame in an area where the accepted procedure was for mechanical removal, or it was the fault of health and safety. The question is was the original work permit amended? If not, the maintenance higher-ups violated the work permit, which did not allow for hot work when first issued. If the health and safety department okayed hot work, they did so on the basis of either ignorance or inadequate information — both of which place the responsibility with them. It was most assuredly not the welder's fault as he did as he was told; I assume the welder was not tasked with the hydrocarbon sniffing. Safety should have required continual sniffing during the hot work, again their mistake. No wonder safety blamed the welder as this allowed safety to cover it's a**. I write as a former welder's helper!
Peter Watson, international business development manager
Armstrong Chemtec Group, West Chester, PA
MANY PEOPLE CONTRIBUTED
This is the first puzzler to really make me angry. The folks who came up with the lockout procedure that left the vent connected, the supervisor who let the man work on a live tank with an opening close to the hot work, and the safety folks who blamed it on the welder — all of these folks — will soon be on the losing side of a lawsuit, if not criminal charges.
Calculations showed that monomer wasn't in the flammable range? Calculations are under ideal conditions, with good mixing, and assume pure components. But that's not always the real world. The torch heat on the metal could have made a local flammable mixture. Or, there could have been decomposed material on the inside, another component in the tank or "lights" were present.
Operating the other connected vessels made the lockout procedure a sham. Could this have been done safely? Possibly. Fully isolating the tank would have been the first choice. Even if there were no valves in the vent line, a pancake blind between flanges would have done the job. If that wasn't feasible (and "why not?" should be answered), they could have erected a barrier with ventilation to blow any fumes from the mixer opening away from the torch area, with continuous monitoring on the welder's side. Or they could have used a wet cutoff wheel, or water jet cutter, to avoid any heat or sparks. Or they could have temporarily blocked the opening with rags (with a place for the tank to breathe elsewhere, of course).
And the greatest insult is blaming the welder. Did the welder do the vapor pressure calculations, or sign off on the hot work order, or develop the lockout procedure? This is the kind of incident that gives the industry a self-inflicted black eye.
Alex Smith, process platform lead
M+W Group U.S., Boston, Mass.
FAULT THE WORK PROCESS
Forget about flogging the dead welder, fire the maintenance manager, then fire the safety manager who blamed the welder! Welders do as they are told. They rely on us, those with college degrees, to get them home safe. There are so many things wrong with this work process it's hard to begin.
First, no work should ever be permitted, covered process or not, that deviates from a well-critiqued plan. Ad hoc plans should not be allowed! The foreman overseeing the work should have foreseen what would happen if the bolts were frozen. There are safer ways to cut a bolt loose that don't involve open flame. Even a grinder would have been a safer choice, though not as fast.
Second, where was the fire watch? A good fire watch moves around the work site looking for potential problems. A fire watch could have seen where the sparks were going, especially if they fell into the vessel through the agitator shaft hole.
Third, couldn't somebody find a way to slip a spectacle blind behind a tank flange? Although this maneuver is dangerous in its own right, operators equipped with respirators could manage it. Another idea would be to flow purge nitrogen into the vapor space of the adjoining tank to force the vapor down into the liquid.
Fourth, did anyone verify that the monomer temperature was, in fact, normal? Perhaps the temperatures were hot. If they were, then maybe it would have been possible to cool the tank liquid or the gases above the tank. Another option would have been to use spark-resistant blowers to pull the vapors out of a manway door and not allow them to accumulate above the liquid.
And, lastly, the vapor content was "calculated." I see too much reliance on simulations in place of laboratory tests or measurements. Who can say the composition is as expected in a simulation?
Dirk Willard, lead process engineer
Fluor Global Services, Inver Grove Heights, Minn.
DECEMBER'S PUZZLER
The horizontal counter-current-flow induced-draft finned air condenser on our debutanizer column can't keep up in the summer whenever the air temperature rises above 105°F. Instead of an outlet temperature of 115°F, which would ensure the exiting liquefied petroleum gas (LPG) is a saturated liquid, we get 125°F and the LPG leaves as a mixed gas/liquid. We're looking for ways to limp through until the next turnaround. We must get as much as possible out of the condenser because we are limited by space and load on top of the column. The column takes a feed from the fluid catalytic cracker at about 260°F. Naphtha is withdrawn from the bottom at about 360°F. The LPG — a mixture of propylene, propane, butanes, butenes and trace C5+ — leaves the top of the tower as a saturated gas at 155 psig and 142°F and goes to the air condenser. Under design conditions, the condenser should provide condensed liquid at 120°F to a water cooler, which then should discharge the condensate at 90°F. Our water cooler is designed for liquid LPG and is stressed at high temperature. Do you have any ideas on how we can improve the performance of this condenser?
Send us your comments, suggestions or solutions for this question by November 9, 2012. We'll include as many of them as possible in the December 2012 issue and all on ChemicalProcessing.com. Send visuals — a sketch is fine. E-mail us at [email protected] or mail to Process Puzzler, Chemical Processing, 555 W. Pierce Road, Suite 301, Itasca, IL 60143. Fax: (630) 467-1120. Please include your name, title, location and company affiliation in the response.
And, of course, if you have a process problem you'd like to pose to our readers, send it along and we'll be pleased to consider it for publication.
