NCSU MAE

1
Leading-Edge Flame Oscillations in Lifted
Turbulent Flames
Nancy Moore Advisor Dr. Kevin Lyons

• Effective Velocity
• Equation better relates the lifted height to the
  flow velocities than the Reynolds number.
• Ueff, U0, and Ucf are the effective, fuel, and
  co-flow velocities, respectively, and ?cf / ?0 is
  the ratio of the co-flow to fuel densities.
• The constant C is chosen such that the data
  collapses to a linear relation.

• Introduction
• In numerous industrial applications such as
  furnaces and burners, the location of the flame
  base and thus the location of the maximum heat
  release is an important design element.
• By increasing the fuel or co-flow velocity, an
  attached flame will lift off the nozzle and
  stabilize at a mean lifted height, with a
  tendency to oscillate axially about that mean
  position.
• The reaction zone lifted-height oscillations are
  studied to shed light on the stabilization
  mechanisms that prevent flames from receding
  downstream and extinguishing.
• Various combinations of fuel and co-flow
  velocities (2,800 Re 10,700) are used to
  study lifted flames at a wide range of downstream
  locations for two different nozzle diameters.

Oscillatory Behavior of Flame in Sequenced Images
with Average Height of 3.12 cm ?t 1/30 sec,
Fuel Velocity 22.6 m/s, Co-Flow Velocity 0.0
m/s, d 3.5 mm

• Rate of Oscillations
• The oscillatory rate, dh'/dt, uses a lab frame of
  reference and is positive for downstream
  recession.
• Because the radial flame position was not noted
  with the height, dh'/dt has only an axial
  component.

h' 3.87 cm
h' 3.87 cm
h' 3.39 cm
h' 3.14 cm
h' 2.78 cm
h' 3.27 cm
h' 4.11 cm
h' 4.11 cm

• Average Heights for d 3.5 mm
• Height increases with fuel velocity and with
  co-flow velocity.
• Error bars represent maximum height fluctuations
  seen upstream and downstream.
• Although the fluctuation magnitude increases with
  downstream location, it is reduced in
• the presence
• of co-flow.
• Similar
• trends are
• seen in
• the data
• from
• d 5.0 mm
• (not shown).

• Height Fluctuations Normalized by the Average
  Height
• The greatest downstream h' (marked by a solid
  symbol) and the greatest upstream
• h' (marked by a symbol outline) for each case are
  shown.
• Oscillation fluctuations are not generally
  symmetric about the average height.
• For flames stabilized at h fluctuations
• are generally above 0.3h.
• Downstream of this region, the turbulence can be
  assumed to
• be fully developed.
• For h 10d, the oscillation behavior is more
  uniform with data
• showing h' 0.3h.
• Approaching blowout (h 35d) the normalized
  fluctuations
• are not significantly greater except for the case
  with no co-flow
• present.
• The ratio h'/h does not significantly change with
  nozzle
• diameter.

• Experimental Setup
• Methane (99 pure) delivered through fuel tubes
  of 3.5 mm or 5.0 mm inner diameter, d (measured
  by a rotameter).
• Co-flow air annulus has 150 mm diameter (measured
  by a TSI velocimeter).
• Images of the oscillating flame were made with a
  Panasonic Model PV-GS300 CCD video camera
  producing thirty frames per second.
• The instantaneous flame height, h', is the axial
  distance from the nozzle exit to the flame base,
  which is the furthest point upstream at which
  flame luminosity is detected by the camera.

• Conclusions
• Though much smaller than the fuel velocity, the
  co-flow velocity significantly affects the
  location and behavior of a flame.
• The flame height, fluctuations, and oscillatory
  rate are related to the flow velocities.
• The average height increases with fuel and with
  co-flow velocity.
• The nozzle diameter is inversely related to the
  lifted height but has little affect on the
  normalized fluctuations.
• Height fluctuations increase with height and are
  reduced in the presence of co-flow,
  while

• Normalized Height Plotted With Re and Ueff
• For a given Reynolds number, the larger nozzle
  results in lower lifted heights.
• The height change with Reynolds number is linear
  for the majority of cases, with exceptions at the
  upper and lower limits.
• Setting C 40 in Ueff collapses the data to a
  linear relation.

Aperiodic Behavior of Oscillations Co-Flow
Velocity 0.0 m/s, d 3.5 mm

• RMS of Oscillatory Rate
• d 3.5 mm
• Setting C 5.2 in Ueff collapses the data to a
  linear relation.
• for d 5.0 mm, C 15 (not shown)
• Also, the oscillatory rate is not strongly
  correlated with height.
• increases with the average height
• less at a given height when co-flow is present,
  suggesting that the presence of co-flow has a
  greater affect on the oscillatory rate throughout
  the entire flowfield than it does on the
  normalized fluctuations

• normalized height fluctuations show little
  variation with height except at locations close
  to the
• nozzle or approaching blowout far downstream.
• Plotting the height and oscillatory rate with the
  effective velocity collapses the data into a
  linear relation. However, different values of C
  are needed.
• C 40 for height and d 3.5 mm or d 5.0 mm
• C 5.2 for oscillatory rate and d 3.5 mm
• C 15 for oscillatory rate and d 5.0 mm
• Montgomery et al. (1998) speculated that the
  constant, C, in the effective velocity equation
  takes into account chemical kinetics and burner
  characteristics.
• Results imply that the largest scales of
  turbulence influencing the lifted height do not
  play a primary role in the oscillation behavior.

Acknowledgement This research has been
supported by the U. S. Army Research Office
(Contract W911NF0510045).
NC STATE UNIVERSITY
MECHANICAL AND AEROSPACE ENGINEERING