POWER May 2021 - 25

EMISSIONS CONTROL
enced by the behavior of the liquid film,
which is strongly dependent on the
wetting nature of the liner surface and
the surface tension between the liquid
and liner.
Droplet Deposition in the Liner and
Re-entrainment from the Liquid Layer
Liquid films are formed on the full height
of the liner surface by the turbulent deposition
of entrained droplets and by
condensation. The liquid film layer is
controlled by the following forces: gravitational
force, surface tension between
the liquid surface and the gas, shear
forces between the liner surface and
Rearranging and solving for model velocity
gives the following:
VM = VF x √(ρF
/ ρM
)
Assuming a field liner velocity of 60
ft/sec, a field gas density of 0.066 lb/ft3,
and a model air density of 0.075 lb/ft3,
returns a model velocity of 56.3 ft/sec in
order to properly simulate the field gas
shear force in the liner of the flow model.
Implications and Conclusions
It has been previously shown by the examples
given in this paper that the model
velocity that is required to properly simuThe
trajectory of entrained droplets
and the re-entrainment of droplets
from the liquid film layer are two very
important aspects that must be
evaluated in the flow model.
the liquid film, and the gas shear forces
on the liquid film. The significant forces
are shown in Figure 2. At stack liner
velocities above those recommended
in EPRI's Revised Wet Stack Design
Guide, the resulting high gas shear or
drag forces acting on the surface of the
downward flowing liquid film are a major
source of droplet re-entrainment and
problematic stack liquid discharge.
To properly simulate the gas shear
forces on the liquid film in the physical
flow model, the model must be run at
conditions that match the field gas shear
forces. When the gas velocity heads (Pv
)
of the flow model and the field unit are
equal, then the shear forces of the gas
on the liquid in the model match the
shear forces in the field. The velocity
head is defined as:
Pv
= (ρg
x Vg
where Pv
2) / (2 x g0
the gas density, Vg
and g0
)
is the gas velocity head, ρg
is the gas velocity,
locity head (subscript M) has been set
to match the field velocity head (subscript
F).
(ρM
x VM
2) / (2 x g0
May 2021 | POWER
) = (ρF
x VF
2) / (2 x g0
)
is
is the gravitational constant.
In the equation below, the model velate
the trajectory of a droplet entrained
in the gas stream inside of the ducts is
significantly different than the model
velocity
that
is
required
to
properly
simulate the shearing forces of the flue
gas on the liquid film layer in the stack.
Therefore, it is physically impossible to
simulate both effects simultaneously in
a scaled physical flow model. When the
model is set to match the droplet trajectories
in the ducts, then the shear forces
are not being matched inside the liner.
Similarly, when the model is set to match
the shearing forces, then the droplet trajectories
are not correct.
The trajectory of entrained droplets
and the re-entrainment of droplets from
the liquid film layer are two very important
aspects that must be evaluated in
the flow model. However, the model
needs to be evaluated at two widely
different conditions in order to investigate
these phenomena, so they cannot
be modeled simultaneously. When
the model is being used to evaluate reentrainment
from the liquid film layer, it
must be run at conditions that match the
field gas shear forces. This means that
any re-entrained droplets will not have
the proper trajectory and deposition patterns
as would occur in the field, thus
invalidating any reported liquid collection
www.powermag.com
efficiency rates from the model.
Therefore, it has been shown that
the modeling of droplet trajectories and
the motion of the resulting liquid films
cannot be modeled simultaneously in a
scaled physical flow model. Because of
this it is not possible to use a physical
flow model to obtain the collection efficiency
of a liquid collection system in an
operating wet stack.
Research engineers at Alden Research
Laboratory have done the work
to identify the physical limits of wet
stack modeling and developed a robust
wet stack design program to provide
clients with the essential engineering
details needed to design an optimized
wet stack system. The group developed
a system to design wet stacks using analytics,
scaled cold-flow physical models,
and computational fluid dynamic
analysis. The team's work has been
documented in numerous EPRI and
self-published formats to educate the
industry on what it means to operate a
wet stack, and how to properly design a
liquid management system for it.
Alden's approach works within the
bounds of physics to ensure responsible,
accurate, and realistic results that are
representative of the final field design.
Simply put, the measurement of liquid
collection efficiency in a scaled cold-flow
physical model does not, and cannot,
represent the field stack installation-it
is literally just the performance of the
laboratory model, with no correlation to
the real-world application. ■
-Lewis A. Maroti is co-author of the
EPRI Wet Stacks Design Guide. He has
more than 45 years of consulting, and
research and development (R&D) experience
in the electrical utility industry,
where he has been at the forefront of
new developments and problem-solving
in the area of air pollution control systems.
David K. Anderson is a senior vice
president with Alden Research Laboratory.
He is the primary author of EPRI's
Revised Wet Stack Design Guide. He has
more than 40 years of research and product
development experience associated
with power generation and environmental
control systems, with a particular focus
on physical and numerical modeling.
James M. Daniel is a senior engineer at
Alden Research Laboratory. He has more
than 18 years of consulting, and R&D
experience in the electric utility industry.
His particular expertise is in the area of
physical flow modeling related to
utility and industrial emissions,
and process equipment.
25
http://www.powermag.com
POWER May 2021

Table of Contents for the Digital Edition of POWER May 2021
