History As A Teacher: How A Refinery Blaze Killed 8 Firefighters

To access the remainder of this piece of premium content, you must be registered with Firehouse.Already have an account? Login

Register in seconds by connecting with your preferred Social Network:

Required
Required
Required
Required
Required
Required
Required
Required
Required
Required

The philosopher George Santayana wrote, "Those who cannot remember the past are condemned to repeat it." The fire service through the years has had many deadly fires. Many times, mistakes or oversights that have been tragically learned in one fire department are repeated in other departments.

12_97_history1.jpg
Photo by Joe Hoffman
A 1975 refinery fire that seemed similar to many others fought at the site had a tragic outcome a flare-up killed two Philadelphia firefighters and injured 33 others. This is how the fire appeared prior to its flare-up.

As professional firefighters we must be students of history and learn from previous fires. When we hear of a firefighter death, we must find out as much about it as possible. If confronted with a similar situation, we must ensure a better outcome. Let us go back in time and take a glimpse at a deadly fire.

The Scenario

On Aug. 17, 1975, the Philadelphia Fire Department responded to a refinery fire. The fire seemed routine and similar to many of the fires fought within that refinery. The outcome, however, would be like no other refinery fire in the history of Philadelphia. When finally extinguished, eight Philadelphia firefighters lay dead. Thirty-three firefighters were injured; two of them were burned critically and would never return to firefighting.

How did a routine fire turn so deadly? Could it happen again?

The scenario started on a Sunday at 12:45 A.M. at the then Gulf Oil Refinery. Twelve thousand barrels per hour of crude oil with an additional blending of 5% naphtha were being pumped through refinery piping into storage tank 231. It was constructed with a cone roof and had an internal floating roof. The source of the supply was the ship the M/T Afran Neptune, which was docked at a pier on the Schuylkill River.

The liquid levels in the tank 231 were to be monitored by Gulf employees. They were not monitored. The tank was filled above its maximum safe fill height. Due to a combination of factors, vapors accumulated above the internal floating roof and under the cone roof. They were then forced out of the vents at the top of the tank and found an ignition source in a nearby boiler house. It violently ignited and flashed back to the tank. The ignition created enough force that the 200-foot-tall brick chimney above the boiler house sustained a crack up to 12 inches wide.

As the fire flashed back to the tank, a fireball was visible at the vents. This occurred at 5:57 A.M. After a period of continued pumping, the ship's pumps were shut down at 6:02.

An explosion then occurred in tank 231, forcing the floating roof downward. This action displaced additional fuel, causing an overflow of the product into the diked area surrounding the tank. The explosion also damaged the tank's fill lines within the dike and at the manifold outside the dike area, causing them to leak. This leaking product was ignited by flaming oil from the tank's vents, spreading the fire outside the dike area into the street. This fire increased in intensity as fuel flowing from the damaged piping increased due to thermal expansion of the cracked metal.

12_97_history2.jpg
Photo by Joe Hoffman
Fires involving hydrocarbons may appear to be totally controlled one minute and totally out of control a brief time later.

When the internal roof was forced down, crude oil came up around the seals onto the floating roof and fed the internal fire. Heat conducted through the pan, and water droplets in the crude, started to vaporize and caused a frothing of the liquid. As the frothing continued, the product began to overflow the vents.

Leaking flanges and the sustained fire around the valving removed all possibilities of shutting down the reverse flow from the tank. It now contained 73,000 barrels of crude oil at a height of 38 feet. (A barrel of crude oil is equal to 42 gallons; the tank contained in excess of 3 million gallons.)

The burning product in the dike area caused failure of other pipelines within the dike. These lines contained benzene, aviation gas, and jet fuel. Along with tank 1114, they became involved in fire.

Firefighting

Gulf employees began to fight the fire at 6:04 A.M. The Philadelphia Fire Department struck a second alarm enroute to the fire and prior to entering the plant due to the magnitude of the fire visible. Six alarms were struck by 7:12 A.M., bringing 52 pieces of equipment and over 200 firefighters.

Cooling hose streams were initiated. Tank 1114 was promptly extinguished. Foam was placed on tank 231. Exposures were surrounded. The fire was contained to the burning tank and the damaged manifold lines in the street. Foam units were being used to contain the manifold fire.

The fire was declared under control at 8:44 A.M. and was fought throughout the day.

Electricity Shut Down

At around 3 P.M., Gulf personnel were concerned about electrical hazards. One worker had received minor electrical shocks operating in the area. Philadelphia Fire Department and Gulf personnel were working in water and foam directly beneath overhead electrical wires. There was a concern that electrocution would occur should the wires fall and electrify the pooled water. After a meeting of Gulf personnel, a decision was made and an electrical substation was shut down with the intent of protecting the firefighters and refinery workers.

This action, however, also shut down electric pumps that were removing the runoff water from the firefighting efforts in this low-lying area. Prior to the pumps being restarted, water buildup in the area had risen to a height of over two feet in some locations. Two pumpers started to draft water to reduce the build-up. A portable pump was being set up.

Flare-Up

By 4:40 P.M., the water and foam level had risen sufficiently to come into contact with the exhaust pipe of Engine 133. The heat of the pipe broke down the foam and ignited hydrocarbons that were floating on the surface of the water directly beneath the foam layer.

12_97_history3.jpg
Photo by Joe Hoffman
The fire is being fought with numerous master streams.

At the time of this flare-up, firefighters were operating in the pooled liquid refilling Engine 133 with foam. Their movement broke the foam blanket. The fire in the muffler area now expanded and engulfed three of those firefighters in flame. As other firefighters went to their aid, additional firefighters became trapped.

(My experience with burning naphtha is when it first ignites it looks like a harmless lazy flame and does not appear to be threatening. Reality returns very quickly, however, since in a brief time the fire becomes a roaring inferno.)

The expanding fire now involved an area over 450 feet by 500 feet. The intense and growing fire attacked tank 240's manifold, causing it to rupture and feeding 3 million gallons of heavy naphtha to the fire. As other pipelines failed, more fuel was supplied to the fire. A nearby administration building was destroyed.

Five additional alarms were requested, bringing the total to 11 alarms.

Transfer

On the day of the fire, my transfer as the captain of Engine 49 became effective. I was scheduled to work the night shift starting at 6 P.M. As I arrived around 4:40 and entered the empty firehouse, the public address system was just starting to transmit the seventh and subsequent alarms. On checking the run file I saw that Engine 49 was second-due engine on the first alarm. I knew that I would be on the fireground for the next 14-hour shift.

12_97_history4.jpg
Photo by Joe Hoffman
Fire involving piping outside of the diked area can be difficult to control.

I had anticipated a large working fire but nothing could have prepared me for what I saw. Though I had fought many fires, this one was different. The area involved was massive. Many large storage tanks were heavily involved in fire. The administration building was fully involved and by itself would have required three to four alarms. The biggest difference, though, was the actions of many of the firefighters. They had a distant stare, many not wanting to talk. They could not wait to leave the fireground. They were in shock. Today, we know it as critical incident stress.

I went to the communications vehicle to report in and found that at that time over 30 firefighters were unaccounted for. Three firefighters from Engine 49 were listed as missing. It was not known whether Engine 49's apparatus had been damaged or destroyed by the fire. In the aftermath, though my apparatus was undamaged, Engine 49 had one firefighter killed and one critically injured. The fire was not brought under control until Monday, Aug. 18, at 5:38 A.M. It was finally extinguished on Tuesday, Aug. 26, at 11:33 A.M.

Lessons Learned

It is difficult to find benefits from a fire that took the lives of eight firefighters. The only way that good can occur is by learning from the incident and hoping that the information can prevent the deaths of other firefighters.

A number of critical lessons were learned at the Gulf Oil fire. Today's textbooks have incorporated many of them. A few of those lessons were:

  • Firefighters should never operate in pooled areas. Due to the shutdown of the electricity, water pooled in a depressed area. Foam hid the hydrocarbons floating on the water surface. As the water level rose, it came in contact with the muffler and exhaust pipe of Engine 133. It ignited, trapping firefighters in the flames. (At a refinery fire in Honolulu a similar situation occurred and was captured on film. Two firefighters broke a foam blanket in a pooled dike area and were temporarily engulfed in flames. Fortunately quick actions by other firefighters protected them and allowed the foam to reseal. They were not seriously injured.)
  • Communications between plant personnel and the fire department is critical. The shutting down of the electricity was done as a safety measure to protect personnel. The drawback was that it also shut down electrical pumps that were removing excess water from the depressed area. Fire department officials had not been notified of the shutdown. They saw a buildup of water and immediately ordered the pumps turned back on. A malfunction in a circuit breaker delayed this action for approximately 30 minutes.
  • Apparatus should be parked at the side of the roads. Hoselines should be kept alongside the roadway to keep streets open.
  • Apparatus should be kept a safe distance from the fire area. Five apparatus were destroyed when the flare-up occurred. Anticipate potential fire spread, predict future problems and have sufficient resources ready to address those problems should they occur.
  • Foam pumpers must be located in safe areas distant from the fire area. If necessary, a relay pump operation can be established. This will have a two-fold effect firefighters can refill the foam unit in a relatively secure area and the foam units can be readily accessed for foam concentrate deliveries.
  • Fires involving hydrocarbons can be unpredictable. It may appear to be totally controlled one minute and totally out of control a brief time later. Incident commanders should not be too quick to place the fire under control. Our guard must be kept up at all times.

Refinery and oil storage tank fires have taught me to respect their many dangers. In the past, fire safety at many refineries was not a top priority. Today, the high cost of hydrocarbons, large federal fines and the concern about public opinion have brought about positive change. There is an increased emphasis on safety and prevention of fires. Strong relationships between the oil industry and local fire departments has benefitted everyone.

Summary

A fire fought over 20 years ago may seem to be of little benefit today. I can only say that losing eight brother firefighters is devastating. The duty of having to tell a wife, daughter and son that their father is not coming home is a role that I hope I never have to repeat. The young daughter asked, "Why my father?" My only answer to her was that he died trying to save another firefighter. He died a hero. It was of little comfort to her.

In a perfect world no fire officer would ever have to bring such horrifying news to a family again. As we know, this is not a perfect world. Because of that we have to keep our guard up and learn from previous fires to protect our brother and sister firefighters.

In Memoriam

Killed in the line of duty, while operating at Box 5988, Penrose Avenue and Lanier Street, on Aug. 17, 1975, were:

Lieutenant James J. Pouliot, Engine 20
Fireman John J. Andrews, Engine 49
Fireman Carroll K. Brenek, Engine 57
Fireman Ralph J. Campana, Ladder 19
Fireman Robert J. Fisher, Engine 33
Fireman Hugh J. McIntyre, Engine 56
Fireman Roger T. Parker, Ladder 27
Fireman Joseph R. Wiley, Ladder 27


James P. Smith, a Firehouse® contributing editor, is a deputy chief of the Philadelphia Fire Department and an adjunct instructor at the National Fire Academy in Emmitsburg, MD.

Loading