Flame Stability, noise and vibration modeling

Contribution: Instituto Superior Tecnico (IST)

Summary

The problem of thermoacoustic oscillations occurs whenever burners (e.g. made of perforated metal sheets), are installed in a combustion chamber, being coupled with upstream and downstream length/volumes. This problem is enhanced if the system has to work under lean conditions, due to increased sensitivity of weak flames to any perturbation. The main task of the project partner IST is to analyze the ability of premixed flames to drive/sustain thermoacoustic instabilities. In order to achieve this, two experimental setups are assembled, the first including a simpler "planar" burner configuration, and the second including the fully premixed burner.

1. Experimental Setup

For investigating the fully premixed burner, an experimental setup was designed and built, consisting of a frequency and amplitude settable pulsated feeding line, coupled to the fully premixed burner, as shown in Figure 1. Flow velocity measurements were performed, along with the high speed visualization, associated with sound-light spectroscopy. Specific acquisition software enabled time correlations between all the acquired signals.

Figure 1. Experimental setup for testing the fully premixed burner, where the system used to modulate the exit velocity, namely the loudspeaker, is coupled by a flexible tube to the burner upstream cavity (IST).

1.1 Working conditions

Initial stationary testing enabled general flame appearance and working conditions identification for the fully premixed burner, and for the two mantle patterns, tested in the planar burner setup. In the planar burner configuration independent testing of the slits and of the holes pattern led to the conclusion that holes are able to sustain leaner flame working conditions. This result should be regarded only as indicative, as no considerations were done regarding equality of power per unit area of the burner surface, or mutual flame interaction (e.g. for the circular holes pattern, or the adjacent circular holes and slits pattern, the interaction was larger when compared to the single three slits configuration).

Combustible mixture, which is used for tests under reacting conditions, was a mixture of propane and air. The stability diagrams are obtained for the pair (Re, angle) from the lean extinction limits to maximum power.

1.2 Experimental Techniques

The equipment used in the experiments consists of: microphones fitted with capillary acoustic probes, thermocouples, equipment for laser visualization (planar laser sheet and shadowgraphy), LDV, chemiluminescence system and a high-speed CCD camera.

2. Radial Burner

Influence of internal pressure fluctuations on exit velocity fluctuations was assessed through time resolved LDV measurements, under isothermal working conditions. Four different combinations of outer holes and slits with the inner distributor were characterized.

Most relevant result is the velocity phase shifts between the four mentioned configurations when the feeding line is pulsated. Three characteristic flow frequencies were identified: 1) lower frequency level, below which velocity fluctuation of the flow coming out from the burner, is everywhere in phase, regardless of the exit geometry or inner distributor plate configuration; 2) middle frequency level, where the phase shifts between the different case studied; 3) upper frequency level, above which the phase shift seems to stabilize, or at least to increase at a lower rate.

3. Planar Burner

Influence of internal pressure fluctuations on exit velocity fluctuation was assessed, for average exit flow velocities of 1 m/s and 3 m/s, with a modulated velocity frequency ranging from 15 to 85 Hz, and different amplitudes of velocity fluctuations. Results, presented in Figure 2, show a backflow blockage behaviour of the slit patterns, limiting the lowest exit velocity to a zero value.

Figure 2. a) Velocity measurements at a slit center exit, for an undisturbed velocity of 1 m/s function of frequency and amplitude modulation of the velocity. b1) and b2) are examples of phase reconstructed signals and respective phase averages for two velocity modulation frequencies and amplitudes for an undisturbed exit velocity of 1m/s (IST).

In a second set of tests the blockage behaviour was explored by keeping the applied acoustic power constant, and changing the average exit velocity. In these tests characteristic frequencies of the experimental burner and feeding line were used as a way to increase exit velocity amplitude modulation. Flow velocity was measured at the slit centre line, while increasing the distance from the exit plane, in order to check both the average value and the fluctuating component of the velocity. This velocity characterization is the flow field that will stabilize the burner flames (flame-flow interaction).

Conclusions

In the scope of the FlexHEAT project, the flame stability/flammability limits, structure, velocity and light emission of fully premixed burners, made from thin perforated metallic sheets, were characterized. The main conclusions to be withdrawn are: a) the flames are driven by any instability from 20-100 Hz; b) the influence of backplate flow distributor introduces a major change on the light emission phase response between circular and slit holes.