Two flame arrestors can look almost identical and still be the wrong choice for each other's job, because they are rated to stop two very different things. A deflagration arrestor is built to quench a flame that is still travelling below the speed of sound. A detonation arrestor has to survive and stop a flame front that has accelerated past the speed of sound and now carries a shock wave. What decides which one you need is not the gas alone but the geometry of the pipe: how much straight run there is between the ignition source and the arrestor for the flame to speed up in. This page contrasts the two classes directly and lays out the selection rule that ties them together.
Detonation vs Deflagration Flame Arrestor in one line: A deflagration arrestor stops a subsonic flame, while a detonation arrestor stops a supersonic flame front travelling with a shock wave. The right choice depends on run-up distance, the length of pipe between the ignition source and the arrestor that lets a flame accelerate: short runs stay deflagrations and take a deflagration arrestor, while long runs can transition to detonation and require the far more robust detonation arrestor.
A deflagration is a flame that propagates by heating the unburned gas just ahead of it, and it travels below the speed of sound in that gas. The pressure rise across a deflagration is comparatively modest and the front is relatively gentle, so a deflagration arrestor can stop it with a matrix of narrow channels, usually a crimped metal element, that pulls enough heat out of the flame to quench it below the ignition temperature. The mechanism is thermal: the element is a heat sink with passages small enough that the flame cannot carry through. This is the arrestor you see protecting an atmospheric tank vent where a flame would originate outside, close to the opening.
A detonation is a different animal. Once a flame in a long pipe accelerates enough, it can transition into a detonation, where the reaction couples to a shock wave and the combined front travels faster than the speed of sound. The pressure spike is far higher and far sharper, and it arrives with mechanical force as well as heat. A detonation arrestor must not only quench the flame but also physically withstand the shock and the pressure pulse without being blown through or deformed, so it is a heavier, more robust device with a reinforced housing and an element engineered for both the impact and the burn.
The distinction to hold onto is that a deflagration arrestor is essentially a thermal quenching device sized for a mild front, while a detonation arrestor is a thermal quenching device that also has to survive a hydraulic hammer. That is why the two are not interchangeable in the wrong direction: a detonation can overwhelm and pass straight through a deflagration arrestor that was never built for the shock, with catastrophic results. Detonations are also classified as stable or unstable, and an unstable detonation, one still overdriven near the transition point, imposes an even more severe local pressure and is a specific case some detonation arrestors are separately tested for.
The single most important variable in choosing between the two is run-up distance, the length of unobstructed pipe a flame travels before it reaches the arrestor. A flame ignited at an open end starts as a deflagration, but the longer the pipe it runs down, the more it accelerates as it drives turbulence and pressure into the gas ahead of it. Past a certain length, which depends on the gas, pipe diameter, and obstructions, that acceleration completes a deflagration-to-detonation transition. Below that length the front stays subsonic; above it, a full detonation can arrive at the arrestor.
This is why placement, not just gas type, governs selection. An arrestor sitting right at the point where a flame would ignite, such as an end-of-line vent, sees only a deflagration because there is no run-up distance to speak of, and a deflagration arrestor is appropriate. Move that same arrestor deep into a piped system, downstream of many meters of header, and the same ignition can present as a detonation by the time it arrives, demanding a detonation arrestor. The pipe geometry between the ignition source and the arrestor is doing the deciding.
In practice this makes the layout drawing part of the sizing calculation. You identify where ignition can credibly start, trace the pipe run to the arrestor, and compare that length against the run-up distance for the gas group and pipe size involved. Elbows, tees, and other obstructions shorten the run-up needed to reach detonation because they add turbulence, so a tortuous line reaches detonation over a shorter straight-line distance than a smooth one. When the run-up is comfortably short you can specify a deflagration arrestor; when it is long or uncertain, the conservative and often mandatory choice is a detonation arrestor.
The place this choice bites hardest in oil and gas is the vapor-handling system: tank vapor recovery headers, flare and vent lines, and vapor destruction units. These are exactly the piped systems where run-up distance can be long, where multiple tanks tie into a common header, and where an ignition at a flare or a vapor recovery unit could send a front back down the pipe toward the tanks. An end-of-line deflagration arrestor on an individual tank vent and an in-line detonation arrestor protecting the shared vapor header are frequently both present, each rated for the flame it can credibly see at its location.
Because a flame arrestor is a passive device, it gives no signal of its own, and its main failure modes are fouling that blocks flow and corrosion or damage that compromises the element. Those show up indirectly in the process data that SCADA already collects. A deflagration or detonation arrestor plugging in a vapor recovery header raises the back-pressure on the tanks feeding it, and a rising tank pressure trend or a vapor recovery unit struggling against increased differential pressure is exactly the kind of signal a monitoring system can surface before the blockage forces relief valves or thief hatches open.
A cloud SCADA platform such as Merobix historizes those tank pressures and vapor-header differentials across a whole battery, so an operator can see an arrestor slowly fouling as a gradual trend rather than as a sudden emissions event or a spurious tank relief. The arrestor itself is chosen and sized once, on the drawing, according to the deflagration-versus-detonation logic above; the ongoing job of monitoring is confirming that the vapor system it protects is still flowing freely and that back-pressure has not crept up to the point where the arrestor, or the tanks behind it, are being stressed.
You can, and it will stop a deflagration as well as a detonation, but it is a heavier, more expensive device with a higher pressure drop than a deflagration arrestor. Using detonation arrestors everywhere is conservative and sometimes done to simplify a spec, but at an end-of-line vent where only a deflagration is credible it adds cost and flow restriction for protection you do not need. The reverse mistake, using a deflagration arrestor where a detonation can arrive, is the dangerous one.
Run-up distance is the length of pipe a flame travels between the ignition point and the arrestor, and it is how far the flame has to accelerate. A short run keeps the flame subsonic, a deflagration, while a long run can let it transition into a supersonic detonation. Because that transition is what determines which arrestor survives, run-up distance, driven by pipe length and obstructions, is the deciding factor in selection, not the gas alone.
A stable detonation has settled into a steady supersonic front travelling at its characteristic velocity, whereas an unstable, or overdriven, detonation is still in the transition region just after a deflagration converts, and it produces an even higher, sharper local pressure spike. The unstable case is the more severe test for an arrestor, so detonation arrestors are often specifically evaluated for it. Both are supersonic and both require a detonation-rated device, not a deflagration arrestor.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.