A Comprehensive Regime Diagram of Dynamical Modes of Triple Flickering Buoyant Diffusion Flames: Experimental and Model Investigations

This study establishes a comprehensive regime diagram for triple flickering buoyant diffusion flames by combining continuous experimental parameter variations with a time-delayed Stuart-Landau oscillator model, successfully identifying three previously unreported dynamical modes and elucidating their synchronization transitions.

The Gist

The Big Picture: A Dance of Three Fireflies

Imagine you have three fireflies (or in this case, small, steady flames) arranged in a triangle. These aren't just sitting still; they are "flickering." This flickering is a natural rhythm caused by the heat rising and creating swirling air currents (vortices) around them, much like how a candle flame dances in a draft.

The scientists wanted to understand what happens when you put three of these flickering flames close together. Do they dance alone? Do they march in step? Do they spin around each other? Or do they suddenly stop moving?

The Problem: The "Stop-and-Go" Camera

In previous studies, researchers had to manually move the flames to different spots, take a picture, move them again, and take another picture. It was like trying to map a city by only looking at a few specific street corners. They missed the smooth transitions between the different dance moves because they couldn't see the "in-between" moments.

The New Trick: The Moving Stage

To fix this, the researchers built a special setup where two flames stayed fixed on the ground, but the third flame (the "vertex" flame at the top of the triangle) was mounted on a motorized slider.

Think of it like a conveyor belt for fire. The top flame slowly glides up and down at a controlled speed. This allowed the researchers to watch the flames interact continuously as the triangle shape changed from flat and wide to tall and skinny, without ever stopping the experiment.

The Discovery: A New Map of Fire Dances

By watching this continuous motion, they created a "Regime Diagram." Think of this as a weather map for fire, but instead of rain and sun, it shows different "dance styles" the flames can do.

They confirmed six dance styles they already knew about:

1. The Marching Band: All three flames flicker perfectly in sync.
2. The Frozen Statues: The flames stop flickering entirely and just sit there quietly.
3. The Half-Hearted Dance: Two flames dance in opposite directions while the third one stands still.
4. The Leader-Follower: Two flames dance together, but the third one dances to the beat of the opposite drum.
5. The Rotating Carousel: The flames flicker one after another in a circle (Left → Center → Right), creating a spinning effect.
6. The Soloists: The flames are so far apart they don't care about each other and flicker randomly.

The New Findings:
Because they could watch the "in-between" moments, they found three new dance styles that no one had seen before:

• The Asymmetric Half-Stop: Two flames dance in opposition, but the third one wobbles slightly without fully stopping. It's like a broken symmetry where the triangle isn't perfectly balanced anymore.
• The Death Decoupling: The two bottom flames freeze (stop dancing) because they are too close and cancel each other out, but the top flame, having moved far away, keeps dancing on its own.
• The Asymmetric Leader-Follower: Similar to the original "Leader-Follower," but the symmetry is broken. The top flame locks steps with one bottom flame but ignores the other, even though the setup looks symmetrical.

The Computer Model: The "Toy" Prediction

To understand why the flames do this, the researchers used a mathematical model called the Stuart-Landau oscillator.

Imagine each flame is a metronome (a device that ticks at a steady beat).

• When you put metronomes close together, they can hear each other's ticks and eventually sync up.
• The researchers created a computer simulation of three metronomes connected by springs (representing the air between the flames).
• They added a little bit of "noise" (random static) to the computer model to simulate the real-world messiness of air movement.

The Result:
The computer model was very good at predicting the main dances (like the Marching Band and the Rotating Carousel). However, it struggled to predict the three new, weird, asymmetric dances. This tells the scientists that their "toy model" is a great starting point, but it's a bit too simple to capture the messy, complex reality of how air swirls around three specific flames.

The Bottom Line

This paper is about mapping the hidden rules of fire.

• What they did: They moved a flame smoothly to watch how three flames interact in real-time.
• What they found: They drew a complete map of all the ways these flames can dance, discovering three new, strange patterns that happen when the flames are in specific, transitional positions.
• Why it matters: It shows us that even simple things like three candles have incredibly complex behaviors that depend on exactly how far apart they are. While their computer model got the basics right, the real world is even more surprising and complex than the math predicted.

Technical Summary

1. Problem Statement

Combustion instability in multi-burner systems (e.g., gas turbines, rocket engines) is a critical engineering challenge. While single-burner instabilities are well-studied, the complex synchronization and coupling dynamics of multi-flame systems remain poorly understood. Previous research on triple-flame systems (specifically in triangular configurations) has been limited by discrete geometric parameters. Most studies used fixed burner positions (e.g., chessboard-style platforms), resulting in sparse data points and an incomplete characterization of the full range of dynamical modes. Furthermore, there was a lack of comprehensive modeling capable of replicating these complex interactions across a continuous range of configurations. The specific gap addressed is the need for a continuous parametric study to map the full regime diagram of triple flickering flames and to identify previously unreported dynamical states.

2. Methodology

Experimental Approach

• Setup: The researchers constructed a triple-flame system using three identical Bunsen-type burners fueled by methane, arranged in an isosceles triangle.
• Novelty: Unlike previous static setups, the vertex flame (Flame C) was mounted on an electronically controlled linear stage. This allowed the vertex flame to move horizontally at controlled speeds ($V = 2.5, 5.0, 7.5$ mm/s), enabling the continuous variation of the triangle height ($H$) while the base length ($L$) remained fixed.
• Parameters: The study systematically varied:
  • Fuel flow rate ($Q$): 0.3–0.6 L/min.
  • Base length ($L$): 45–70 mm.
  • Vertex displacement velocity ($V$).
• Data Acquisition: High-speed cameras (125 fps) recorded flame brightness. The brightness signals ($B_i(t)$) were extracted, normalized, and analyzed using the Hilbert Transform to determine instantaneous phase and frequency.
• Modeling: A time-delayed Stuart-Landau (S-L) oscillator model was employed. Three coupled S-L oscillators represented the three flames, incorporating:
  • Coupling strength ($K$) related to flame size.
  • Time delays ($\tau_1, \tau_2$) related to inter-flame distances.
  • Gaussian white noise to simulate environmental disturbances caused by the horizontal motion.

Analysis Techniques

• Regime Diagram Construction: Data was plotted in a dimensionless parameter space defined by the Grashof number ($Gr$), Froude number ($Fr$), and geometric ratios ($\Delta H/\Delta L$).
• Phase Space Analysis: 3D phase portraits and 2D projections of the brightness signals were used to classify dynamical modes based on topological structures.
• Bifurcation Analysis: The S-L model was simulated across a wide range of $K$ and $\tau$ values to generate bifurcation diagrams and compare them with experimental observations.

3. Key Contributions

1. Continuous Parametric Mapping: The introduction of a moving vertex flame overcame the discretization limitations of previous studies, allowing for the construction of a comprehensive regime diagram that fills gaps in previously unknown parameter regions.
2. Discovery of New Dynamical Modes: The study identified three previously unreported dynamical modes that were missed in static configurations:
   • Mode III-2 (Asymmetric Partially Flickering Death): One base flame flickers in anti-phase with the vertex flame, while the other base flame is in a "flickering death" state (suppressed oscillation). This breaks the $D_2$ symmetry of the isosceles triangle.
   • Mode VI-2 (Death Decoupling Mode): The two base flames are in a coupled "flickering death" state, while the vertex flame oscillates independently as a single flame.
   • Mode IV-2 (Asymmetric Partially In-phase Mode): The vertex flame locks in phase with one base flame but remains out of phase ($\pi$) with the other, breaking the symmetry of the standard partially in-phase mode.
3. Validation of Horizontal Motion Effects: The study confirmed that horizontal translation of the vertex flame (at speeds up to 7.5 mm/s) has a negligible effect on the intrinsic flickering frequency of single flames or the collective coupling frequency, justifying the use of this method to map static-like regimes continuously.
4. Model-Experiment Correlation: The time-delayed S-L oscillator model successfully reproduced six of the nine observed modes (including the new Mode IV-2) and provided a bifurcation diagram explaining transitions between synchronization states.

4. Key Results

• Regime Diagram: The comprehensive diagram classifies modes based on coupling strength (determined by distance).
  • Strong Coupling (Small $H$): Dominated by synchronized modes (In-phase, Rotation, Partially In-phase).
  • Weak Coupling (Large $H$): Transitions to Decoupled modes and Flickering Death.
  • Transition Zones: The newly discovered asymmetric modes (III-2, IV-2, VI-2) primarily occur near the boundaries between strong and weak coupling or in regions where symmetry breaking is induced by geometric asymmetry.
• Frequency and Phase:
  • The normalized frequency ratio ($f_c/f_s$) of the vertex flame remains close to unity in weak coupling but shifts in strong coupling.
  • Phase differences clearly distinguish the modes (e.g., $\Delta \phi = 0$ for in-phase, $\Delta \phi = \pi$ for anti-phase, $\Delta \phi \approx 2\pi/3$ for rotation).
• Model Performance:
  • The S-L model qualitatively reproduced the phase-space trajectories (ellipses, points, butterfly shapes) of Modes I–VI and Mode IV-2.
  • Limitations: The model failed to reproduce the asymmetric death modes (III-2 and VI-2). The authors attribute this to the S-L model's reduced-order nature, which omits complex asymmetric vortex reconnection mechanisms and specific stochastic effects present in the physical experiment.

5. Significance

• Fundamental Understanding: This work provides the most complete map of dynamical modes for triple-flame systems to date, revealing that continuous geometric variation uncovers complex asymmetric states that discrete sampling misses.
• Physical Mechanisms: It elucidates how vortex interactions (reconnection vs. shedding) drive synchronization, death, and rotation, and how symmetry breaking leads to new dynamical states.
• Engineering Application: The findings offer a physical framework for real-time geometric control strategies in multi-injector combustion systems (e.g., gas turbines). By understanding the regime diagram, engineers can potentially manipulate burner spacing or flow rates to avoid unstable modes (like flickering death or chaotic decoupling) or stabilize desired synchronization patterns.
• Modeling Guidance: The discrepancy between the S-L model and the asymmetric experimental modes highlights the need for higher-order coupling terms or stochastic bifurcation terms in reduced-order models to fully capture multi-flame physics.

In conclusion, this study bridges the gap between fundamental flame physics and practical multi-burner engineering by establishing a continuous regime diagram and identifying new dynamical behaviors, while simultaneously validating the utility and limitations of Stuart-Landau oscillator models in predicting complex combustion dynamics.