Acetylene pipe diameter and detonation: why the smallest pipe that will do the job is the safest one

In most gas engineering a larger pipe is the conservative choice: lower velocity, lower pressure drop, more margin. With acetylene the opposite is true. The wider the bore, the lower the pressure at which a decomposition can travel the length of the pipe, and the lower the pressure at which it can turn into a detonation. This is the single most counter-intuitive fact in acetylene pipework design, it is measured rather than theoretical, and it is written into the Indian pipeline code as a table of pressure ceilings that fall as the pipe gets bigger. The design rule that follows is simple: choose the smallest pipe that will serve the duty.

The C.P.I. (Safety in the Production and Use of Acetylene, Paris 1968) set out the behaviour of acetylene decomposition in long horizontal pipes of four bores. Read as a series, it shows the pressure threshold falling by more than an order of magnitude between 10 mm and 100 mm.

The same body applied the lesson to its own rules for the high-pressure part of a dissolved acetylene plant: pipe bore not above 21 mm for charging racks and 15 mm for individual connections, in seamless drawn, heat-treated mild steel capable of withstanding more than ten times the 25 kg/cm² maximum compression pressure, and a general instruction to use pipes of the smallest diameter compatible with a reasonable pressure loss.

The critical pressure for propagation of an acetylene decomposition is attributed, as compiled by S. A. Miller (1965), to the loss of heat by radiation from the highly luminous, carbon-laden flame. A narrow pipe has a large wall area relative to the volume of gas it contains, so the flame loses heat quickly and dies below a certain pressure. A wide pipe loses proportionally less, so the flame survives at lower pressure. Two further observations fit the same picture: turbulent flames propagate at lower pressures than laminar ones, and a detonation front, being thin and fast, outruns the radiative loss altogether and can travel below the deflagration limit.

Carver, Smith and Webster (IChemE Symposium Series No. 33, 1972) set out the measured evidence. Sargent's long-pipeline data put the decomposition pressure limit at about 200 kN/m² absolute in 40 mm pipe, falling to about 100 kN/m² absolute in 250 mm pipe, and earlier work by Rimarski and Boesler gave about 200 and 150 kN/m² for the same two sizes — two independent sources agreeing on the trend and closely at 40 mm. Their own experiments found that detonation could propagate at pressures below those needed to sustain a deflagration when the initiation was strong enough, and that in a 305 mm, 61 m tube a detonation ran down to about 50 kN/m² absolute. The single "stability pressure" figures quoted in codes are therefore conservative only at small bore. Applying a small-bore figure to a large header is non-conservative reasoning. The physics of the limit is covered under where the 1.5 bar rule comes from.

The most quotable diameter-specific limit in the literature comes from Kogarko and co-workers (Khim. Prom., 1962, 496), who found pressure rises of the order of 300 times the initial pressure at the closed end of 20 cm pipes 21 m long when the initial pressure was only 1.45–2.5 atm absolute. Their conclusion was that transport pressures should not exceed 1.4 atm absolute for 20 cm pipe with 1.2 cm wall, and should not exceed 1.05 atm absolute for pipe up to 38 cm diameter with 1.0 cm wall. The second figure is barely above atmospheric pressure, and it is the reason the 1954 Hüls event, in which a decomposition travelled more than 2 km through 15, 30 and 60 cm pipe at 1.03 atm absolute, could happen at all.

Sargent's work (Chem. Eng., 1957, 64, (2), 251) gives the same picture on a chart of minimum pressure against diameter: in 1-inch pipe, deflagration needs about 2.25 atm absolute and detonation about 3.2 atm absolute, and both lines fall steadily towards about 1 atm absolute at 20–40 inch bore. Tube bundles of 1.2 cm tubes inside pipes up to 15 cm bore prevented detonation up to about 5 atm absolute in German practice, were useless at 9 atm absolute, and at 10 atm absolute detonation was observed even in 3 mm tubes. Fine passages work, but only within a pressure range.

IS 15190 (Part 1):2002, the code of practice for acetylene pipelines up to 155 kPa gauge, is built on the concept of working ranges developed by the Industrial Gases Committee in Paris and on CGA pamphlet G.1.3. Its Table 1 sets a maximum operating pressure for each nominal diameter in each of three design classes — Class 1 designed for working pressure only, Class 2 for deflagration (an 11-fold pressure factor) and Class 3 for detonation with reflection (a 35-fold factor).

Entries of 250 kPa carry the standard's note that use is limited to 1.5 kg/cm². Read down the Class 1 column and the diameter effect is explicit: a DN 10 line may carry 250 kPa gauge on a plain working-pressure design, while a DN 80 line may carry only 20 kPa gauge. The class for a mixed-size system is set by the largest bore in it, and a system is split into separately classed sections only where a flashback arrestor separates them. The code recommends restricting pipe to DN 50 and allows parallel DN 50 pipes where the calculated size would be larger. Part 2, for 155 to 2,550 kPa gauge, is stricter still: pipe above DN 25 is avoided, DN 40 is never exceeded, and higher flow is met with multiple DN 25 pipes at velocities of 2–4 m/s. Both parts are explained under IS 15190 explained and summarised on the acetylene pipeline rules page.

The reason small-bore design costs so little in practice is that acetylene flows are modest. The British Acetylene Association's code of 1962, Design and Installation of Acetylene Supply/Distribution Systems for Flame Processes, found piping of 1–3 inch (2–8 cm) diameter adequate for up to 100 m³/h even over runs of hundreds of metres at the legally permitted pressures. A 100 m³/h demand is a large fabrication shop. IS 15190 (Part 1) sizes on 2–4 m/s, and at those velocities a DN 50 line at 50 kPa gauge carries several tens of cubic metres an hour. There is almost never an operational reason to install a pipe wide enough to be dangerous, which is what makes the design rule — choose the smallest pipe that will serve — both safe and cheap.

Where a wide bore is genuinely unavoidable, as in the long low-pressure transmission lines of 15–60 cm that exist at some chemical sites, the answer is not a stronger pipe but a different arrestor: the packed tower described under packed-tower arrestors. Emphasis goes on preventing a detonation from passing out of the pipe into a vessel, because it is easier to install small-bore piping than to build plant to withstand detonation pressures.