Where the 1.5 bar acetylene pressure rule comes from and what it does not mean

Every acetylene handbook, every plant manual and every tender specification quotes a pressure limit for acetylene of "about 1.5 bar". Almost none of them says where the number came from, and the common understanding of it — that it is a conservative limit with a safety margin in hand — is wrong. The figure has a precise, public-domain origin, a narrow set of experimental conditions, and a set of national regulations that all sit above it. This page sets out the whole chain, from the 1944 measurement to IS 14814:2023.

The experimental origin is a United States Bureau of Mines investigation by Jones, Scott, Kennedy and Huff, Report of Investigations 3755 (1944). For point ignition by a fused wire in a 30 cm by 5 cm container, the minimum pressure of pure acetylene at 25 °C that sustained an explosive decomposition was 1.42 atm absolute.

Four conditions are attached to that figure, and every one of them matters:

Change any one of these and the number moves, and in almost every recorded case it moves downward; the one higher figure in the table below, 2.0 atm, comes from a long 7.6 cm pipe ignited by a fused wire. The "1.5" of common usage, by contrast, is almost always a gauge figure (1.5 kg/cm² gauge, 1.5 bar gauge), roughly 2.5 atm absolute, so it is not a rounding of the 1944 threshold but a limit that sits above it. Expressed in the units the Indian trade uses, 1.42 atm absolute is about 1.44 bar absolute, 1.47 kgf/cm² absolute, or roughly 0.42 atm (6 psi) above atmospheric pressure. The compilation by S. A. Miller (1965) prints the gauge equivalent as 5.9 psig; the arithmetic gives nearer 6.2 psi, and the discrepancy is noted for checking.

The literature compiled by S. A. Miller (1965) contains a ladder of measurements and events that show the 1.42 atm figure is a ceiling for weak ignition in small vessels, not a floor for safety. Each figure is given with its primary source.

Two things stand out. Several real and simulated pipeline events propagated at or just above 1.0 atm absolute, with no gauge pressure to speak of. And the energy needed to start a decomposition rises steeply as pressure falls: Kogarko and Ivanov needed 1.5 J at 1.3–1.5 atm absolute but about 1,200 J at 0.65 atm absolute. The ignition-energy side is covered in static electricity and minimum ignition energy.

The single most important demonstration that "safe below 1.5 bar" is a false belief took place at Chemische Werke Hüls in Germany on 2 January 1954. A sequence of mishaps admitted chlorine under pressure into an acetylene pipeline at a 15 cm section. Decomposition was initiated and travelled through more than 2 km of pipe of 15, 30 and 60 cm diameter, even though the pressure was only 1.03 atm absolute at a temperature below 0 °C. Several large items of plant were wrecked and several people injured.

The follow-up experiments are as instructive as the event. Progressive decomposition was initiated in 20 m lengths of 30 cm pipe containing 97–98 % acetylene saturated with water (relative humidity 60–80 %) at down to 1.0 atm absolute, using a forechamber of cylinder acetylene at 2.5–5 atm separated by a low-pressure bursting disc. Fusing a platinum wire in the forechamber burst the disc and lit the low-pressure gas across the whole cross-section of the pipe. The lesson is that a strong initiation across a wide bore defeats any pressure margin. It is also the reason packed-tower arrestors were developed for large-bore lines, covered under packed-tower flashback arrestors.

The critical pressure for propagation is not a thermodynamic or kinetic limit. As compiled by S. A. Miller (1965), it is attributed to the loss of heat by radiation from the highly luminous, carbon-laden decomposition flame. Three observations support this: highly turbulent flames and detonations propagate at lower pressures (down to about 0.8 atm absolute); flames in wider tubes propagate at lower pressures; and the measured brightness temperatures of the flame fall below calculated values at higher pressures, which is consistent with a flame radiating like a black body through its own carbon.

The consequences follow directly. A wider pipe has a lower surface-to-volume ratio and loses proportionally less heat, so a flame propagates in it more easily — the reason pipe diameter is a design variable in its own right. Turbulence raises the reaction rate per unit of frontal area faster than it raises the radiating area. A detonation front is so thin and fast that it outruns radiative loss entirely, which is why detonation can propagate below the deflagration limit — a finding confirmed experimentally by Webster et al. (IChemE Symposium Series No. 33, 1972), who propagated detonation down to about 50 kN/m² absolute in a 305 mm, 61 m tube with high-explosive initiation.

Here is the finding that most published pages miss. Every national regulatory limit compiled by S. A. Miller (1965) is a gauge pressure, and every one of them sits above the 1.42 atm absolute threshold it is supposedly derived from.

The modern bodies confirm the pattern. CGA G-1 (2009) prohibits generating, distributing or using acetylene above 15 psig (103 kPa gauge) for welding and allied purposes, with a means of preventing propagation required above that. EIGA Doc 123/22 classifies acetylene systems as low pressure up to 0.2 bar, medium pressure above 0.2 to 1.5 bar, and high pressure above 1.5 to 25 bar; EIGA Doc 237/21 states that generator pressure above 1.5 bar gauge is not permitted and that generator vessels are designed for operating pressure, not decomposition pressure. The C.P.I. (Safety in the Production and Use of Acetylene, Paris 1968) recorded 1.5 kg/cm² as the generator ceiling of its day — a historical figure, superseded in India and not to be applied here.

These limits are economic and practical compromises. They are made safe by the other defences in the system: small pipe bore, flashback arrestors, purity, moisture, exclusion of ignition sources and temperature control. A plant operated below 1.5 bar gauge is not safe because it is below 1.5 bar gauge; it is safe because the rest of the system is doing its work.

KIGL's plant operation manuals set a generator working maximum of 0.92 kg/cm² gauge, which is about 0.90 bar gauge, 13.1 psig or 1.89 atm absolute, with a 70 °C ceiling on the reaction. Against the five historical ceilings in the table above, that figure is below all of them — below the strictest, the US 15 psig, by about 13 %. Against IS 8471:2003 it is within the medium-pressure class. The temperature ceiling addresses the variable that actually moves the propagation limit for dry gas, and the gas leaving a carbide generator is water-saturated, which raises the limit further, as explained in wet acetylene is safer than dry.

It is worth being plain: 1.89 atm absolute is above the 1.42 atm absolute physics threshold, as is every medium-pressure generator in the world and every national ceiling in the table. The limits were never set below the decomposition threshold; the safety comes from the generator design, arrestors and interlocks around the pressure figure.