Because of the characteristics of acetylene gas, acetylene cylinders are constructed in an entirely different manner from those made to contain other gases.

Historical Background

Until 1904, no suitable acetylene container had been developed. The gas was used mainly for illumination and was generally piped directly from generators to the area to be served. In that year in Indiana, P.C. Avery displayed to two of his home state’s most famous promoters, James Allison and Carl Fisher, a portable cylinder containing acetylene gas designed to power auto headlights. Then engaged in auto sales, Allison and Fisher were immediately interested, and with Avery, set up a small factory in Indianapolis to fabricate this “tank.”

The shop was known as Concentrated Acetylene Company, until Avery withdrew in 1906. The company then became the Rest-0-Lite Company, the forerunner of the Linde Division of Union Carbide Corporation.

Allison and Fisher devoted much of their time relocating their plant into progressively larger quarters.  Not until 1910 did they build one of sufficient size in what was then suburban Indianapolis, across the street from the site of the famed motor speedway they later constructed.

Carbide production continued to increase, and in 1913, a much improved acetylene cylinder similar to that used today was introduced. With these two major achievements, gas welding began replacing other metal joining methods.

 

Cylinder Stabilizing Fillers

The need for a porous substance in a cylinder to stabilize compressed acetylene was realized by the French scientist Fouche, one of the men responsible for the oxyacetylene mixture. The size of the filler, however, left very little room for gas in the cylinder. One filler was a magnesium oxychloride cement type; another was made of asbestos discs. The charcoal-cement filler was not developed until 1919, and in 1950 a sand-lime material became popular.

In 1897 a French team, Claude and Hess, demonstrated the value of acetone. This colorless, flammable liquid, when added to the porous material, is capable of absorbing 25 times its own volume of acetylene for each atmosphere 101 kPa (14.7 psi) of pressure applied. Thus, at full cylinder pressure of 1724 kPa (250 psi at 70”F), it can absorb over 400 times its own volume of acetylene.

In 1958, cylinder manufacturers announced a lightweight calcium-silicate filler with 92% porosity. This new filler lessened cylinder weight by 30%, increased cylinder capacity, and improved charging and discharging characteristics. Although only 8% solid, this filler had extraordinary strength, longer life, no deterioration, and could be charged and discharged much faster.

The calcium silicate filler, composed of sand, lime and asbestos, lined the cylinder and conformed to its shape. Its crushing strength, an indication of cylinder life, is 6205 kPa (900 psi).

When medical research indicated that asbestos fibers are carcinogenic due to the size of the fibers (less than 3.5 microns in diameter and 10 microns in length, which is small enough to allow the fibers to penetrate the respiratory tract of the lungs), cylinder manufacturers set about to produce an asbestos-free filler. A non-asbestos alkaline-resistant glass fiber filler was developed by the Linde Division of Union Carbide Corporation and patented in 1982.

A cut-away view of a modern acetylene cylinder is shown in Figure A-2.

How Acetylene Cylinders are Manufactured

Cylinder production and testing is a step-by-step procedure which insures ultimate quality and safety. Seamless shells are cold drawn in hydraulic presses with capacities up to 454 000 kg (500 tons). Center seams and foot ring attachments are welded using the submerged arc process. Cylinders are then normalized (stress relieved) to increase cylinder life and corrosion resistance.

 

Measure and Weight

In the filling area, cylinders are measured and weighed to determine exact volume. At another location, filler is mixed to correct proportions in hoppers, weighed, and mixed with water in agitators. Before each new batch of filler is used, a sample containing one cubic foot is weighed and examined to ensure correct mixture.

Cylinders are then filled automatically and weighed again. Factoring in the weight and volume of the cylinder confirms that it is accurately filled to specification. The cylinders are then oven-baked at 315°C (600°F) to eliminate the water. Baking time ranges from 40 to 120 hours, depending on cylinder size. After baking, another weight check is made to determine if any water remains. Since 1% moisture in the filler will affect ultimate performance, cylinders are baked again if only a slight moisture content is detected.

Fuse plugs and valves are installed, and cylinders are shot-blasted and painted. (Fuse plugs are small steel machine bolts with holes filled with a low melting alloy designed to release gas in case of fire, and to lessen the acetylene pressure to reduce the possibility of an explosion).

Finally, strength proof tests at 4140 kPa (600 psi) are run. Pressure is then reduced to 2070 kPa (300 psi), and the cylinders are immersed in water to check for leaks. Drawn to a vacuum, they are charged with acetone and weighed again to determine if they are fully charged.

Cylinders are checked after each procedure during the manufacturing process. Those not meeting the rigid requirements of federal law and company rules are rejected regardless of the stage of manufacture. For example, a number of cylinders are selected from each completed lot, charged with acetylene, and tested to ensure proper discharge. If the cylinders do not meet specifications, the entire lot is rejected.

 

Basic Tests

A bonfire test is designed to check cylinder performance under conditions similar to a fire in a building. A fully charged cylinder is placed horizontally on racks, and specified sizes and amounts of wood strips are ignited around it. The cylinder passes the test if there is no appreciable shell bulge, no penetration of filler by decomposition, and no breakup of the filler.

The flashback test simulates torch flashback entering the cylinder, assumed to be at full pressure when the operator closes the valve immediately afterward. If the flash is immediately quenched in the cylinder with only a minimum of decomposition and without release of fusible plugs, the cylinder passes the test.

A hot spot test simulates negligent impinging of a torch flame against the cylinder. Flame is directed at the cylinder sidewall until a 3 to 20 mm (1/8 to 3/4 in.) bulge develops. If filler decomposition is limited to the area closely adjacent to the resulting cavity, performance is satisfactory.

The bump test determines the filler’s resistance to mechanical shock received during normal service. The cylinder is mounted on a foundry mold-bumper and subjected to minimum 200 000 bumping cycles. At the conclusion of the test, satisfactory performance is indicated when there is no attrition, sagging, or cracking of the filler.

X