Free Web Hosting Provider - Web Hosting - E-commerce - High Speed Internet - Free Web Page
Search the Web

Update on the Scuba Tank Explosion

What Happened:

Chris Hawkins, my 18 year old nephew, worked at the Force E dive shop located in Riviera Beach, FL as a technician. His job responsibilities included, among other things, filling scuba cylinders. On February 1, 1998, a regular customer came into the shop to have his aluminum Walter Kidde scuba tank topped off before going on another dive. Chris examined the tank and found current inspection stickers. He took the tank and placed it in the water tub and connected it to the fill station. Store policy requires that all cylinders have their pressure checked before actually adding any air. Chris opened the scuba valve and was attempting to check the pressure when he heard a hissing sound. He assumed that it was a leaky O-ring in the valve which he had seen many times before in other tanks. He then fully submerged the tank into the water to check for bubbles. Before Chris could determine the origin of the leak, the tank suddenly exploded. Absolutely no air was added to the tank and the tank was not mistreated at any time. There were two other witnesses to this event. Both have stated on record that Chris did nothing wrong. Photographic evidence clearly shows that the fill station valve was in the OFF position at the time of the blast indicating that no air was added to the tank before it exploded. Also, the fill equipment itself had a special regulator that prevented overfills.

Schematic diagram of Fill Station

Chris sustained severe injuries in the explosion. He lost the better part of his hand and his face was badly cut up. Ironically, the customer who owned the tanks and another employee (Paul) were standing less than six feet away from the tank when it suddenly exploded. Paul was blown about 20 feet backwards through the air, but was not seriously hurt. The blast from the explosion propelled another scuba tank right past Paul's head - luckily missing him. The customer was blown across a cluster of filled tanks and into a steel rail, but sustained only minor injuries. Chris was blown into a steel grate railing about 10 feet back from where the tank was. A large chunk of the exploded tank ripped a hole through the steel rail, ricocheted off a full oxygen tank, made a 90 degree turn and blasted out the front store window which was criss-crossed with commercial grade burglar bars. Fortunately, this failed to rupture the oxygen tank which could have killed them all.

Despite the fact that some officials are saying that Chris lost two fingers in the blast, they are not entirely accurate. Chris lost most of his hand in the blast including his thumb and index finger. If you draw a line on your hand, starting at the crack between your middle finger and your index finger, down your hand to the middle of your wrist and then over to the edge just under the base of the thumb, then you will have a better idea what he really lost. That is a lot more than just two fingers.

I went by the dive shop where this happened and the building was still standing, but messed up pretty bad. Due to a safety water tub that Chris had lowered the scuba tank into, most of the damage was confined to the roof and ceiling area although there were several windows blown out, torn metal, etc. The shop was, in fact, open for business the next day - minus their filling equipment which was destroyed in the blast.

Points to Remember:

The Reason:

The blast that ripped off the better part of Chris's hand was caused by a defective aluminum Walter Kidde scuba tank. It was made from an alloy, 6351-T6, that has been known to be inferior for use in scuba tanks for many years. This alloy was routinely used to make scuba tanks from the late 70's on up to 1990. The tank that exploded had its last hydrotest in 1994 and displayed current inspection stickers.

The explosion was roughly equivalent to several sticks of dynamite. According to one scuba tank inspection expert, "The explosive potential in a fully charged 80cf aluminum SCUBA cylinder is approximately 1,300,000 foot pounds -- enough to lift a typical fire department hook-and-ladder truck over 60 feet in the air!", stated by A. Dale Fox on his web page.

Walter Kidde is a leading manufacturer of fire extinguishers and other low pressure gas cylinders. They used to make Scuba tanks, including the one that nearly killed Chris, but sold their North Carolina scuba plant in 1989. In an interesting twist, Walter Kidde claims that Luxfer purchased their scuba division at that time and thus, assumed the liability for their tanks. Luxfer denies that they purchased the Kidde scuba division and claims that they only purchased its assets.

Here is the interesting part. The DOT has already issued a safety alert bulletin for tanks made with alloy 6351-T6 way back in 1994. They stopped short of requiring the tanks to be removed from service or ordering a recall. It seems that Chris was not the first victim of these defective tanks. The first tank that exploded was actually a SCBA tank (self-contained breathing apparatus) - like what firemen wear. This explosion occurred at a chemical plant in Deer Park, Texas. The tank was manufactured in 1977 and exploded while being filled to its rated pressure of 2216 psig. Fortunately, there were no reports of serious injury in that explosion.

On June 4, 1994, a second explosion reported to the DOT seriously injured Arnie Hubber, who was the fill station technician at the Scuba Sports Dive Store located in North Miami, Florida. He lost his right thumb and both his right arm and left leg were broken in addition to other injuries. The scuba tank in that explosion was manufactured in 1982, and had a current hydro (less than two years old) and a current VIP. It exploded while being filled to its rated pressure of 3000 psig. Arnie Hubber actually came to visit Chris in the hospital. In all of these previous explosions, a piece of the cylinder neck separated from the tank.

According to the DOT's safety alert in the Federal Register (Volume 59, No. 142, pages 38028-38030), the problem originates from the use of an inferior aluminum alloy to build these tanks. Alloy 6351-T6 has been used in the manufacture of seamless aluminum cylinders marked "DOT 3AL", and some composite cylinders. The DOT estimates that approximately seven million tanks have been manufactured using this alloy, with about two million being scuba tanks.

With that many tanks currently out there right now and only about ten exploding, the DOT has not seen fit to order a mandatory recall. With the cost of such a recall being into the millions of dollars, the scuba manufacturers have not seen fit to voluntarily recall these cylinders. Meanwhile, as the cylinders get older, more and more of them will explode. However, according to my preliminary findings, DOT 3AL aluminum tanks made from 6351-T6 may possibly be safe if less than 10 years old and inspected by a qualified inspector annually. Even still, it makes you wonder just who will be next.

Why is this alloy unsafe if only ten out 7,000,000 tanks have exploded? The reason is that many, many tanks are condemned every year because they either fail their hydrotest or their visual inspection. If testing wasn't required, many more scuba cylinders would have exploded and killed or injured people. Annual inspections catch most of these defective tanks before they actually explode, but in so doing, they make the tanks look safer (statistically) than they really are. For this reason, if your visual inspector fails your tank, you should thank him because it is better to lose your tank than your life. The odds get much worse when you consider that only about 400 fill stations fill approximately 80% of the scuba tanks in this country.

Points to Remember:

Blast Photos

A photographer that was on the scene moments after the blast took these photos for Chris:

This is a new enhanced enlargement of one of the photographs. It shows that the tank was visually inspected in August 1997. The tank exploded on February 1, 1998. The tank was inspected at another dive shop in the Force E chain. Shop policy requires that all aluminum tanks be inspected for, in addition to other things, neck cracks both inside and outside the tank. A special light and mirror are used for this inspection. No cracks or deformities were detected during this tanks last visual inspection. This does not suggest that annual inspections are useless. On the contrary, most tanks that would normally have exploded are caught and condemned by annual inspections. However, this does suggest that hidden defects in tanks made from the 6351-T6 aluminum alloy may be much harder to detect than previously believed.

View of tank being reassembled

Here is a view of the tank being reassembled by emergency workers.

View of tank being reassembled

Here is another angle of the tank being reassembled by emergency workers.

This is the fill station where the tank exploded. The work area is sunken into the ground about three feet. The blue panel on the right was the fill station. Under the fill station panel is the remains of the water tub. The red stuff in the picture is blood.

Here is a new photo taken by a Riviera Beach Fireman who was on the scene. This is what was left behind after Chris was taken away in the ambulance. The injuries were so severe that his clothes had to be cut off to determine the extent of his injuries before he could be moved. Most of the blood that Chris lost was in the "pit" area where a tourniquet was placed around his arm. He was able to stagger to this point and fall. The small red puddles are blood that oozed out past the tourniquet.

Scuba Tank Storage

Several scuba experts have pointed out to me that proper scuba tank storage is very important even for new tanks. When storing tanks, they should be stored in an upright position with only about 50 PSI in them. They should not be emptied to zero PSI (unless you plan on discarding the tank). The reason for not going to zero PSI is to prevent the backflow of humid, unfiltered air into the tank. A pressure of 50 PSI in a tank hydrotested to 5,000 PSI is not a significant risk.

Back to Chris's Get Well Main Page