Geometrical Flame Holder
Study
The operating regimes and stability characteristics of geometrical flame holders including open, closed and wall mounted v-gutters as well as cylinders and various other shapes and flame holder arrays have received significant attention historically. The attention given to the phenomena surrounding geometrical flame holding in particular is for good reason because it is a robust, reliable and effective mechanism and as a result widely employed.
The study of the ability of a geometrical
object to stabilize a flame is attributed to the eddy region that is
formed downstream of the geometrical body.
This zone of recirculation caused by the
separation of the flow is filled primarily with hot combustion
products, which serve as an ignition source for the oncoming
unburned reactants.
The recirculation zone also fulfills the
spatial and temporal requirements of a combustion process.
This subject has received significant
attention from Scurlock (1948) and Williams (1951) among others.
In addition to the study of the reasons
that geometrical flame holders are able to stabilize flames, there
seems to be a more extensive consideration of the reasons, that a
flame holder is unable to stabilize a flame.
Thus, the phenomenon of blow-out as well as
the development of stability curves to be used to predict the range
of operation of many types of geometrical flame holders has received
much attention. Mestre, Williams and Rao have all considered these
processes.
In experimental flame stability analysis,
achieving realistic conditions (pressure in particular) has often
proven restrictive.
A solution to this problem as employed by
Lefebvre is to introduce water vapor into the reactant stream.
Effectively, the water injection serves to
decrease the reaction temperature, which is analogous to a pressure
decrease in an ideal gas system.
Then, the low pressures that are typical for
a geometrical flame holder employed in, say, an aircraft engine
becomes more achievable.
The drag characteristics of geometrical
flame holders have received much less attention overall than
stability processes and stability regimes.
Drag has been characterized however by
Mastre, Rizk and others.
It is well documented that aerodynamic drag
is an important characteristic of geometrical flame holders.
It has also been shown that for a given
geometrical flame holder the drag is reduced for the reacting flow
situation as compared to the non-reacting situation.
It is this fact that motivates our proposed
research with the goal of guiding strategies for enhancing the
performance of afterburners and ramjet combustion chambers.
An experimental study of the mechanisms
of the documented drag reduction on a flame holder during combustion
is proposed with tests being performed on center-mounted flame
holders.
A cylindrical rod as well as an open
v-gutter are likely candidates for the study.
The experiment will attempt to correlate the
drag coefficient behavior to mean and turbulent characteristics in
the recirculation zone, with specific attention being paid to the
near field shear layer as entrainment (and the way that it is
influenced by the state of the shear layer) appears to be a
motivating factor.
Some preliminary work that compares
reacting to nonreacting flow for one particular flame holder has
been performed and can be used for comparison purposes.
The original intent of the date was to be
used as a tool with which to compare studies of fluidic flame
holders.
The flame holder that was used in that
comparison was a wall mounted half-v gutter type geometrical flame
holder.

This flame holder was mounted to the wall
of the combustion chamber in an existing experiment, the details of
which are available in Ahmed et al.

Nonreacting tests were performed, as well
as tests at near stoichiometric conditions with a Reynolds number
based on a characteristic dimension, H, of 45 mm (the combustion
chamber height) and a characteristic velocity of 11 m/s.
The v-gutter was designed to create a
recirculation zone that would match the height of the recirculation
zone of the fluidic flame holder operating at a specified momentum
ratio.
What was determined was that combustion
caused a significant shortening of the recirculation zone as is
typical for a wall-mounted geometrical flame holder.

It was also shown that the reacting flow
creates more intense forward and reverse velocities,

and a smaller, more intense region of
elevated turbulence levels downstream of the flame holder.

Possibly most importantly considering the proposed future
work, it was shown that the dilatation is much more intense for the
reacting case.
This indicates a large volumetric heat
release that would be expected from a combustion process but also
may provide some insight as to the mechanisms that cause a reduction
in drag during combustion processes.

A Schlieren imaging system was also
utilized to visualize the combustion process and some combustion
instabilities were captured that manifested itself as a "flapping"
of the flame at the trailing edge of the v-gutter.
The Schlieren allows for a visualization of
the recirculation zone.
The uniform shade of the recirculation
region indicates the presence of uniform density hot combustion
products suggesting an efficient combustion process.
Chemiluminesence was also investigated
through digital images of the flame to gain insight into the regions
in which the flame is residing and propagating.


