Flammable mixtures are frequently transported within tubing. Examples include flare lines, storage tank vents, air drilling, and improperly designed purging operations. The flow regime is often turbulent, heat transfer rates high, and flow control devices few in number. In these circumstances, combustion of the gas mixtures could be catastrophic. Early flame detection is critical.

Many methods of flame detection are available. Unfortunately, few offer remote, non-line of sight, detection. There is often no means by which combustion can be detected while still inside the tubing.

To address this problem, combustion noise is being investigated at the University of Calgary as a possible solution. An experimental study has been completed that shows that combustion noise is detectable in high speed pressure data. This noise can be distinguished from other sources of noise by its inverse power law relationship with frequency. This relationship holds true whether the frequencies are calculated using traditional frequency analysis or wavelet analysis.


It has been known for several decades that turbulent burnerflames produce noise. Moreover, turbulent flames, though inefficient producers of noise, are still much more efficient at producing noise than turbulent flows. Thus, a crude calculation of the total noise level can, under some conditions, distinguish combustion noise from background flow noise (1). Unfortunately, there are many sources of noise when dealing with industrial equipment. A practical detection system needs to be able to identify the specific noise produced by combustion.

Fortunately, acoustic theory predicts that the mechanisms by which combustion noise is generated are distinct from those that produce noise by turbulence or other pure flows (2–7). There is also strong evidence linking the production of the combustion noise to the changes in the fuel burning rate (6,8,9). This link between the fuel burning rate and the production of noise supports the acoustic theories.

Unfortunately, current theories are quite limited. In a comprehensive summary of the current state of knowledge, Lieuwen states the "The development of accurate, predictive combustion response models for realistic, that is, turbulent, configurations has not been achieved, however, and remains a key challenge for future workers." (10)

It has been found that the noise produced by combustion is function of many different variables. Researchers studying turbulent burner flames found that the spectrum of the noise is affected by changes in the laminar flame speed of the fuel/air mixture, the burner diameter, and the flow rate (11–13). The peak frequency of this noise was also closely related to the level of macroscopic mixing (11,14). Recent research into highly turbulent combustion has also found a connection between the acoustic frequency spectrum of flames and the kinetic spectrum of turbulence (15–17). Researchers have also found that flame noise can be used to diagnose burner flame instabilities such as the lean blowout limit (18,19). Obviously a wealth of information can be found in combustion "noise".

However, there are few acoustic studies of combustion within tubing, especially under flowing conditions. Theory and experiments have normally been restricted to laminar or no-flow conditions (20,21).

This content is only available via PDF.
You can access this article if you purchase or spend a download.