United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S7-86/042 Apr. 1987
Project Summary
Pilot-Scale Evaluation of Top-Inlet
and Advanced Electrostatic
Filtration
A. S. Viner, G. P. Griener,
D. F. Furlong, and R. G. Hurst
Advanced Electrostatic Augmenta-
tion of Fabric Filtration (ESFF) was
evaluated on a slipstream from a stoker-
fired boiler. Advanced ESFF, with its
characteristic high-voltage center-wire
electrode, was compared with conven-
tional filter bags in the same baghouse
using calibrated flow orifices. The ad-
vantage of advanced ESFF was demon-
strated by consistently higher gas flow
rates in the bags with the corona pro-
ducing electrodes. Analysis of the data
showed that the specific resistance of
an electrostatically enhanced filter was
70% less than that of a conventional
bag. An economic analysis showed a
capital cost savings of 26% with ad-
vanced ESFF, based on doubling the
air-to-cloth ratio for advanced ESFF. In
a second test, the feasibility of using
top-inlet filtration on stoker fly ash was
established. No definitive comparison
with conventional bottom-inlet filtration
could be made in the allotted test
period.
This Project Summary was developed
by EPA's Air and Energy Engineering Re-
search Laboratory, Research Triangle
Park, NC, to announce key findings of
the research project that Is fully docu-
mented In a separate report of the same
title (see Project Report ordering In-
formation at back).
Introduction
Under a program funded by the U.S.
Navy, and with support from the U.S.
Environmental Protection Agency, per-
sonnel from Research Triangle Institute
and ETS, Inc., conducted a pilot plant
evaluation of fabric filtration at a bag-
house. The study focused on two different
approaches to fly ash filtration: top-inlet
filtration and advanced electrostatic
stimulation of fabric filtration (ESFF). In
designing a baghouse, several variables
must be considered: the cost of the fan(s);
the rate of the pressure loss (amount of
energy required to pull air through the
dust-laden fabric); and the ratio of the air
flow rate to the filter area (air-to-cloth
ratio). Generally, it is better to have a low
rate of pressure loss or drop. Attaining a
low pressure drop means ensuring a slow
rate of dust deposition which, in turn, is
related to the amount of fabric area avail-
able for dust deposition. This all points to
the importance of air-to-cloth ratio. The
design of a baghouse poses the problem
of specifying the highest possible air-to-
cloth ratio (to minimize the overall size of
the baghouse and subsequently its cost)
while ensuring that the pressure loss
limitations (influenced by fan cost con-
siderations) are not exceeded.
One form of ESFF involves placing
electrodes of opposite polarity on the
surface of the fabric and then generating
an electric field parallel to the fabric. The
most recent version of ESFF (also called
advanced ESFF) calls for placing the
electrode in the center of the filter bag
(Figure 1). Experiments conducted by the
U.S. EPA showed that the electric field
alters the dust deposition pattern and the
structure of the dust cake. The overall
result was a reduction in the pressure
drop across the bag. The advanced form
of ESFF was tested during this project.
The top-inlet design of a baghouse was
the other form of filtration tested at the
pilot scale. The standard, bottom-inlet
design allows dirty gas to enter and flow
up the bag. Because of this upward flow,
-------
To High-Voltage Power Supply
ğ~ Fabric Bag
Gas Flow In
Figure 1.
fy- High-Voltage Electrode
Insulator Cable
Electrode Tension Spring
Spring-Restraining Bar
The center-wire configuration
for advanced ESFF.
large particles either fall from the gas
into the hopper below or are deposited on
the lower portion of the bag. An elutriation
effect may occur whereby the average
particle size deposited on the bag de-
creases as the distance from the bottom
inlet increases. The resistance of the
dust cake may increase as the mean
particle size decreases, producing an in-
crease in flow resistance as the distance
up the bag increases. Based on this
theory, it has been recommended that
the dust-laden gas be introduced from
the top, into a bag with openings at both
the top and the bottom (Figure 2). In this
way, the large particles would be retained
in the gas and be distributed more evenly
along the entire length of the bag. This
may both lower the resistance of the dust
cake and make it more uniform.
Pilot Plant Design
Located at Cherry Point Marine Corps
Air Station in Havelock, NC, the pilot
plant (Figure 3) filtered coal fly ash from a
slipstream from two spreader-stoker-fired
boilers. The boilers are rated at a maxi-
mum of 77,500 Ib steam/hr (9.77 kg/sec).
Only one boiler was online at any given
time during the test period. The fuel was
an eastern coal of 1% sulfur and 5%
water and had a heat content of 31,600
Gas Flow In
Gas Flow Out
Figure 2. Conceptual design of a top-inlet baghouse.
kJ/kg (13,600 Btu/lb). The ash's electrical
conductivity (relatively high for a coal
ash) resulted from a high, unburned
carbon content. Identical baghouse com-
partments were located next to each
other, with each fed by a separate fan
with a capacity of 19.82 mVmin (700
ftVmin) of air drawn from a common
duct. This common duct pulled gas from
the inlet of the electrostatic precipitators
(ESPs) downstream of each boiler. A
single reverse-air fan provided ambient
air for cleaning each compartment.
Because of the project's focus on the
feasibility of advanced ESFF, the only
tests performed were those which
demonstrated the effect and, to a small
degree, the operating range of advanced
ESFF. Time constraints dictated comple-
tion within 1 year (February 1985 to
February 1986). This requirement affected
the experimental design. The central
problem in designing the plant consisted
of making it possible to test three kinds of
filtration advanced ESFF, top-inlet, an(
conventional bottom-inlet in only tw<
compartments as efficiently as possible
The engineers decided to use individua
bag flow monitors (IBFMtm), thereby al
lowing for the simultaneous operation c
advanced ESFF and conventional filtratio
in the same compartment. This left th
second compartment free for testing th
top-inlet design.
The IBFM has a calibrated orific
plate at the inlet of the bag on top of th
compartment tubesheet. Pressure tap
on the upstream and downstream side
of each orifice plate make it possible 1
measure the orifice pressure drop. Th
flow into each bag is calculated from dai
on the pressure drop and the temperatui
of the gas. This information, along wii
the tubesheet pressure drop and inl
dust concentration, can then be used
calculate the dust cake's specific resi
tance and residual drag. A comparison
these parameters produces a figure
-------
Test
Trailer
I
Conventional
andESFF
Compartment
R/A Fan
Top-Inlet
Compartment
Figure 3. Schematic of the Cherry Point pilot plant.
merit for evaluating the ESFF design. The
data obtained from the top-inlet compart-
ment resulted in a similar comparison.
Testing the ESFF and conventional bags
in the same compartment meant that the
bags were exposed to identical conditions,
thereby increasing the confidence in the
results.
Data Acquisition System
As mentioned earlier, the project had a
time constraint of 1 year. Also, cost con-
siderations precluded payments for an
on-site operator. Thus, the engineers
designed an automated system for col-
lecting, retrieving, and analyzing the data.
This system offered a number of advant-
ages, the most important of which were:
elimination of labor costs for data col-
lection and computer entry; and long-
distance data retrii
advantage meant that the data could be
monitored and analyzed daily.
It was necessary t<
peratures, tubesheet
drops, electrical condi
aval. The second
measure the tem-
and orifice pressure
ions, and dust-inlet
loadings. All were automated except for
the dust-inlet loadings (measured ac-
cording to standard EPA Method 5
procedures). The signals from the various
thermocouples and pressure transducers
fed into a Molytek Model 2702 Pro-
grammable Datalogger. The datalogger,
in turn, interfaced with a Tandy 1200
microcomputer. At 10-minute intervals,
the datalogger transmitted instantaneous
readings to the computer which then
stored the data (with custom-developed
software). The concurrent PC-DOS oper-
ation allowed for simultaneous operation
of the data collection/storage program
and remote communication. Thus, the
engineers could use a high-speed modem
to call the on-site computer and read
data files over the phone lines into their
own microcomputers. Once the data files
were retrieved in this fashion, they were
read into a Lotus 1 -2-3 (C) worksheet for
analysis. The one disadvantage in testing
two kinds of filtration in the same com-
partment was that it precluded on-site
evaluation of performance based on the
tubesheet pressure drop. Thus, it was
necessary to analyze the data before the
results were known. However, the ad-
vantages certainly outweighed this one
disadvantage.
Results
During February and March 1985, the
pilot plant was installed. The plant was
brought online in April, with hardware
debugged in April and May. The collection
of data began in June. The plant operated
continuously until the end of September,
except for brief periods due to coal eleva-
tor repair, a switch from boiler No. 1 to
No. 2, and electrical failure. At all other
times, the plant required a minimum
amount of intervention and collected data
automatically.
Bottom-lnlet/ESFF Compartment
The feasibility of advanced ESFF and,
to a lesser degree, of top-inlet filtration is
determined by the ability to operate at
higher face velocities. Therefore, the focus
was on determining if ESFF and top-inlet
filtration could be used for prolonged
periods of time at face velocities that
were substantially higher than those of a
conventional filter. Thus, the test plan
called for one long experiment with the
filtration velocity increased gradually. This
experiment was particularly important
since it was impossible to explicitly set
the flow rate through the bottom-inlet
and ESFF bags. Instead, it was possible
only to set the overall flow rate to the
entire compartment. The engineers deter-
mined the distribution of flow between
the bottom-inlet and the ESFF bags by
calculating the drag differential between
the two sets of bags.
Figure 4 plots the average air-to-cloth
ratio (for each bag) against the cycle
number. One cycle represents a 12-hour
period (11 hours and 50 minutes of filtra-
tion, and 10 minutes of cleaning). The
curves are separated into two distinct
groups, with the exception of the data
from cycle 50 through 70. The upper set
of curves corresponds to the advanced
ESFF bags, and the lower set corresponds
-------
1.8
1.5-
|
^ 0.9
3> 0.8
V 0.7
.i 0.6
* 0.5
0.4-
0.3-
10
30
SO
\ i
70
r r i i
;;o 730
150
90
Cycle Number
Figure 4. Average air-to-cloth ratio for conventional and ESFF bags in bottom-inlet
compartment.
to the conventional bags. The overall air-
to-cloth ratio (compartment inlet flow/
total compartment bag area) increases
from an initial value of 0.55 to 0.79
m/min (1.8 to 2,6 ft/min) at cycle 50.
Then, the reverse air fan failed, resulting
in erratic data. The tubesheet pressure
drop ranged from 0.1 kPa (0.4 in. H20) at
an air-to-cloth ratio of 0.55 m/min to
0.87 kPa (3.5 in. H20) at an air-to-cloth
ratio of 1.26 m/min.
It was possible to convert the data into
flow rates and then use those rates and
the dust concentration information to
compute the variation of dust load (kg/m2)
on each bag as a function of time. The
ratio of pressure drop to bag flow velocity
gives the bag drag. Figures 5 and 6 plot
the bag drag as a function of dust load for
the ESFF and conventional bags, respec-
tively. The straight line in each figure
represents the regression fit to the data.
Note the two important features in these
figures: (1) the straight lines show a good
fit to the data, indicating that it is reason-
able to assume a linear relationship
between drag and dust load; and (2) the
slopes (i.e., specific resistance) of the two
lines are quite different. The value of
drag for the ESFF bag rises much more
slowly than the conventional bag's value.
Thus, by using an appropriate model of
pressure drop behavior and measuring
the individual bag flows, the values for
the specific resistance and residual drag
can be determined fairly easily.
Top-Inlet Compartment
In making these measurements, a
single value for specific resistance was
assumed as characterizing all the bags in
the compartment at a given time. The
value for residual drag also was assumed
to be the same for all the bags. Values for
the compartment drag and dust load were
calculated from measurements of the
compartment inlet flow and the tubesheet
pressure drop. The test plan for this
compartment was very similar to the plan
for the other compartment, with the face
velocity increasing from 0.55 to 0.79
m/min. The results, however, were
marred by two problems: the breakdown
of the cleaning fan (mentioned earlier)
and the loss of control over a flow-control
valve. Because of the small amount of
reliable data, it is doubtful that any
accurate conclusions can be drawn
about the operation of the top-inlet
compartment.
Economic Analysis of
Advanced ESFF
This project showed that advanced
ESFF can greatly reduce the rate of pres-
sure rise of a baghouse. This can translate
into an economic savings in that, for a
fixed average pressure drop, a baghouse
which uses ESFF will be smaller than a
conventional baghouse. Also to be con-
sidered is whether or not this smaller
size will offset the cost of the additional
hardware. The pressure drop depends on
both the residual drag and the specific
resistance of the dust cake. Test results
indicated that an ESFF baghouse can be
expected to have a specific resistance
equal to 30% of that of a conventional
baghouse. This value was used to esti-
mate pressure drop. Two values were
considered for residual drag: case I made
the ESFF's residual drag equal to a con-
ventional baghouse's, while case II
0.40
0.36
0.32
* 0.28
2s
Q 0.24
0.20
0.16
I
I
0.5
0.4
c:
0.05
0.10
Dust Load,
Figure 5. Drag correlation for the ESFF bag.
0.15
kg/m2
0.20
0.2
0.25
-------
040
036
032
c
I
£
* 028
Q.
024
0 20
0 16
I i
05
0.4
03
0.2
o
3C
C
0 002 004 006 0.08 0.10
Dust Load, kg/m2
Figure 6. Drag rnrre/ation fnr thf rnnwpn(;pna/ hnn
0 12
0.14 0.16
assumed the ESFF value to be half of that
of the conventional baghouse Although
the results from this study present little
evidence to support this assumption,
laboratory and pilot-plant studies of other
ESFF geometries have indicated a large
difference in residual drag Based on
these assumptions about the air-to-cloth
ratio, bag life, specific resistance, and
residual drag, the annual operating and
maintenance costs were computed.
Figure 7 plots the annual savings (in the
form of the percent difference in cost
between an ESFF and a conventional
baghouse)
Conclusions and
Recommendations
On an average, the specific resistance
of the ESFF bags was 70% less than that
of conventional bags. The IBFMtm system
worked well, proving invaluable to the
success of this program The automated
data acquisition/analysis system also
worked well and allowed the pilot plant
to operate unattended for days at a time.
Problems with the cleaning hardware in
the top-inlet compartment significantly
affected the conditions and produced un-
reliable results. Finally, the additional cost
of the ESFF hardware is more than offset
by the savings resulting from the smaller
size of the baghouse. The capital cost of
an ESFF baghouse which was operated
at air-to-cloth ratio of 1.22 m/min would
be 26% less than that of a conventional
baghouse with an air-to-cloth ratio of
0.61 m/min.
40
30
to
Ol
c
I 20
Q.
O
1
C
C
I
Uj
10
Case II
Bag Life
0 03 06 09 12
ESFF Gas-to-Cloth Ratio, m/min
Figure 7. Annual ESFF savings in operating and maintenance costs
-------
A Viner is with Research Triangle Institute, Research Triangle Park, /vc.' 2//UU;
andG. Greiner, D. Furlong, andR. Hurst are withETS. Inc.. Roanoke. VA 24018.
Louis S. Wow's is the EPA Project Officer (see below)
The complete report, entitled "Pilot-Scale Evaluation of Top-Inlet and Advanced
Electrostatic Filtration, "(Order No PB87-133 096/AS; Cost: $ 13.95, subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S7-86/042
0000329 PS
U S ENVIR PROTECTION AGENCY
REGION 5 LIBRARY
230 S DEARBORN STREET
CHICAGO IL 60604
------- |