&EPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 2771 1
EPA-600/7 79 043b
February 1979
Fabric Filter Model
Format Change;
Volume II.
User's Guide
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide-range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield. Virginia 22161.
-------
EPA-600/7-79-043b
February 1979
Fabric Filter Model
Format Change;
Volume II. User's Guide
by
Richard Dennis and Hans A. Klemm
GCA Corporation
GCA/Technology Division
Bedford, Massachusetts 01730
Contract No. 68-02-2607
Task No. 8
Program Element No. EHE624
EPA Project Officer: James H. Turner
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
ABSTRACT
A new mathematical model is described for use by control personnel to deter-
mine the adequacy of existing or proposed filter systems designed to minimize
coal fly ash emissions. Although the basic model design is similar to that
discussed In an earlier report, several improvements and many timesaving steps
have been introduced so that the immediate needs of agency and other emissions
control enforcement groups can be met. To further aid the model user, the
study has been presented in two volumes, the first a Detailed Technical Report
and the second a User's Guide.
The model is structured so that by using the combustion, operating, and
design parameters indicated by power plant and/or manufacturing personnel, the
program user can forecast the expected particulate emissions and filter pressure
loss.
The program affords the option of providing readily appraised summary
performance statistics or highly detailed results if the latter are necessary.
Several built in error checks prevent the generation of useless data and avoid
unnecessary computer time.
The model takes into account the concentration and specific resistance
properties of the dust, air/cloth ratio, sequential compartmentized operation
and the method, intensity and frequency of cleaning. The model function depends
upon the unique fabric cleaning and dust penetration properties observed with
several coal fly ashes (Including lignite) and woven glass fabrics. Prior
validation of a precursor model showed excellent agreement with measured field
performance for the Sunbury, Pennsylvania and Nucla, Colorado fabric filter
systems.
ill
-------
iv
-------
CONTENTS
Abstract ill
List of Figures vl
List of Tables vi
Acknowledgments vii
1.0 Introduction 1
1.1 Background 1
1.2 Role of User's Guide 1
1.3 Input Variables 2
2.0 Data Input Forms 5
2.1 Card 1 - Heading or Title Card 5
2.2 Card 2 - Design Data 5
2.3 Card 3 - Operating Data 7
2.4 Card 4 - Specific Resistance Coefficient (K2)/Gas
and Dust Properties 7
2.5 Card 5 - Fabric and Gas Properties 8
2.6 Card 6 - Special Program Instructions 9
2.7 Card 7 - Plot (Graph) Dimensions 10
3.0 Error Messages 11
4.0 Example of Model Application 14
References 22
-------
FIGURES
Number Page
1 Fabric filter model - data input form 4
2 Fabric filter model - data input form prepared from
Table 3 input data 18
TABLES
1 Summary Listing of Input Data for Filtration Model 3
2 Summary of Diagnostic Messages and Their Interpretation 12
3 Available Input Data for Modeling Baghouse Performance at
Electric Utility, Plant B 15
4 English/Metric Conversion Factors 16
5 Samples of Tabular Printout for Example of Model Application . . 20
vi
-------
ACKNOWLEDGEMENT
The authors express their most sincere appreciation to Dr. James H. Turner,
EPA Project Officer, for his advice, discerning technical reviews and encourage-
ment throughout the present and precursor modeling studies. We also wish to
acknowledge the capable support of Mr. William H. Battye in the intricacies of
programming and Messrs. Robert R. Hall, Peter H. Anderson and William F. Ostrowski
for their appraisal and testing of the model format.
vii
-------
1.0 INTRODUCTION
1.1 BACKGROUND
A comprehensive treatment of the development and design of a new fabric
filtration model, structured specifically so that enforcement engineers can
assess the fly ash control capabilities of filter systems used at coal burning
facilities, is presented in Volume I of this report. Several open literature
publications are also available including reports of studies sponsored by the
U.S. Environmental Protection Agency that provide further insight into the
historical and technical aspects of the model development.1'6
1.2 ROLE OF USER'S GUIDE
The brief description of the filter model application presented in this
volume serves as a convenient field instruction guide for agency and other
personnel who possess a rudimentary knowledge of filtration technology and
computer methodology. It is intended that the model will enable field engineers
to determine whether a proposed or existing fabric filter system will, or
should be able to, reduce fly ash emissions to required levels.
By following the instructions set forth in this guide, the model user can
collect and organize the required information for correct entry on specially
prepared data input forms. From this point on, use of conventional punch-
carding and programming routines will provide enforcement personnel with a
tabular printout of predicted filter system performance at whatever degree of
detail and accuracy is needed for a specific filter system evaluation.
-------
The following information is provided in this document:
• A description of each input variable.
• Error messages, probable cause and correction.
• An example of model application.
1.3 INPUT VARIABLES
For immediate reference, a listing of the input variables used with the
filtration model is presented in Table 1. Data inputs are grouped in four
categories: Design Data, Operating Data, Dust and Fabric Properties and Special
Program Instructions. Table 1 also shows the correct units for each data input
as well as the Card and Item number for entry on the Data Input Form, Figure 1.
Additionally, the valid ranges, i.e., the range of numerical values that will
allow the program to run, are indicated for key variables. Table 1 also
indicates Default Values which are automatically assigned by the program when
no data Inputs are available or when the model user forgets to make an entry.
-------
TABLE 1. SUMMARY LISTING OF INPUT DATA FOR FILTRATION MODEL
Item Symbol Unit* Card Valid range Default* Note
0 Title
1 Number of compartment* n
2 Compartment cleaning time at min
f 3 Cleaning cycle time It min
4 Time between cleaning cycle* tf min
S 5 Limiting preaaure drop Pj_ N/m2
S 6 Reverae flow velocity VR m/min
7 Shaking frequency f cp*
S Shaking amplitude (half atroke) A cm
u 9 Average face velocity V m/min
H< 10 Caa temperature T- °C
fa& 11 Inlet ilnHt concentration Cj 8/m3
° 12 meaaured at temperature of T °C
11 Specific reaiatance coefficient Kj N-min/g-m
14 meaaured at temperature of T °C
15 meaaured at velocity of V m/min
16 meaaured at ma** median diameter of HMD] urn
17 meaaured at geometric atandard ogj
uj deviation of
jjj 18 Haaa median diameter of inlet duat MMDj urn
p 19 Geometric atandard deviation of inlet duat 00.2
°° 70 Diacrete particle denaity of inlet duat p g/cm3
ti 21 Bulk denaity of inlet duat p g/cm3
3 22 Effective reiidual drag Sj N-min/m3
£ 23 meatured at temperature of T °C
£ 24 Kcaidual fabric loading WR g/m2
1 25 Keaiduel drag Sg N-min/m3
26 meaaured at temperature of T °C
27 Initial (lope Kg N-min/g-m
2R Manured at temperature of T °C
29 Maximum number of cyclea modeled nc -
10 Accuracy code 0 or 1
2i 32 Type of plotted reaulta
^ a 33 Fractional area cleaned ac -
u M 34 > axia length inchea
jg g
n 35 y axia length inchea
Theae valuaa are uaed when no entry has been made for the parameter.
Default valuaa aaalgned only when no data Inputa are available for
SB. WR and K2.
Notea: a. Knter item 4 or 5, but not both.
b. Enter item 6 or (7 and 8), but not both.
c. F.nter item* U through 15 when Kj meaaurement ia available.
1
2 2 to 10
2 0.5(1 tern I/ ft em I) ^
2
2 a
2 a
2 0 b
2 b
2 b
3 0.3 to 3.0
3 >0
3
3 25
4 0.25 to 10 ...d
4 >0 25
4 0.61
4 2 to 50 d
4 2 to 4 d
4
4 2 to 50 d.e
4 2 to 4 d.e
4 e
4 e
5 350f f.g
5 >0 25
5 501 f.g
5 f
5 >0 25
5 f
5 >0 25
6 h
6 0 or 1 0
6 Average i
6 i
6 >0 to 1 j
7 6k
7 5
d. Enter itema 11 through 19 when Kj meaaurement muat be corrected for size properties.
e. Knter item* IB through 21 when Kj i* to be eatimated from duat
f . F.nter itema 22 through 28 for nonlinear drag model.
g. F.nter ifm* 22 through 24 for linear drag model.
h. Generally 20 cyclea are sufficient.
i. For tabular result* specify DETAILED. SUMMARY or AVERAGE; for
PLOT or leave blank.
j. Enter only in special ca*e when ac meaaurement ie available.
k. Card can be left out if default values are aufficient or if no
sice and denaity parameters.
graphical reaulta apecify
plotted output i* desired.
-------
FABRIC FILTER MODEL-DATA INPUT FORM
!• 17 |« 19 20
34
OCCHML FOMT
M.L OTNOi orraics MUST K MONT JUSTIFIED CXCCPT no* ITEMS o. 31 AND sx.
Figure 1. Fabric filter model - data input form.
-------
2.0 DATA INPUT FORMS
Card identification, the type of information entered on each card, in-
structions relating to unusual or mutually exclusive entries, and acceptable
numerical ranges for specific variables, are discussed in this section.
2.1 CARD 1 - HEADING OR TITLE CARD
A descriptive title identifying the specific modeling run is entered in
up to 64 characters on Card 1.
2.2 CARD 2 - DESIGN DATA
The number of compartments (Item 1) is the total number of compartments
(or chambers) in the baghouse, from 2 to 30. If the compartments are cleaned
one at a time on a sequential basis, the actual number of compartments must
be entered. If more than one compartment is cleaned simultaneously, the re-
vised or effective value for number of compartments is the total number di-
vided by the number undergoing simultaneous cleaning. Note that entry spaces
with no triangular decimal point indicator require right justification for
whole number inputs such as Item 1. Otherwise, the presence of the triangle
specifies the decimal point location, Item 2.
The compartment cleaning time (Item 2) is the length of time that a
single compartment remains out of service during its cleaning. Although the
program will operate if Item 2 is left blank, it is recommended that the actual
value be entered when available. If left blank, a default value equal to
one-half of the quotient of the full cycle cleaning time divided by the number
of compartments is automatically calculated by the program.
5
-------
The cleaning cycle time (Item 3) is the length of time required to com-
plete an entire cleaning cycle; i.e., the time required to clean all baghouse
compartments. A value for the cleaning cycle time must be entered for the
program to function.
The method by which the cleaning cycle is regulated is identified by
Item 4 or 5. For those systems where continuous cleaning has been specified
the baghouse cleans continuously (back-to-back cleaning cycles). With no time
between cleaning, both Items 4 and 5 must be left or given zero entries. On
the other hand, if the system is required to filter without cleaning for a
specific time interval following each cleaning cycle, then the time between
cleaning cycles must be entered in Item 4 (with no entry for Item 5). Con-
versely, if the system is expected to filter after the cleaning cycle until a
specified limiting pressure drop is reached, then the limiting pressure drop
must be entered in Item 5 (with no entry for Item 4). Note that although a
limiting pressure drop system may also function with a finite time between
cleaning cycles, the latter value is not known at the inception of the modeling.
In the above case, the time between cleaning is calculated from the computer
printout of model results.
Items 6 through 8 describe the fabric cleaning parameters. If reverse
flow cleaning (with bag collapse) is the sole cleaning method, the reverse
flow velocity only (Item 6) is entered. Conversely, if mechanical shaking
with no reverse flow is used, both shaking frequency and amplitude (Items 7
and 8) must be entered. For most combustion processes where the reverse flow
cleaning action is the predominant cleaning mechanism, the actual reverse
flow velocity alone is entered. If no entry is made for reverse flow velocity
-------
or its value cannot be determined, the model will automatically assign a
default value of zero so that the program will continue to operate.
2.3 CARD 3 - OPERATING DATA
The average face velocity (Item 9 air-to-cloth ratio) must always be
entered and at the actual filtration temperature (Item 10). The average face
velocity is computed from the total volume flow through the baghouse and the
total effective filtration area. The actual gas temperature (Item 10) and
the inlet dust concentration (Item 11) are also prerequisite entries. If the
inlet concentration is reported at a temperature other than 25°C, the tempera-
ture must be entered in Item 12.
2.4 CARD 4 - SPECIFIC RESISTANCE COEFFICIENT (I^/GAS AND DUST PROPERTIES
The specific resistance coefficient, K£, and/or associated dust proper-
ties are entered via Card 4. When the K2 value for the dust being filtered
is known, K£ (Item 13) and the temperature and face velocity at which K2 was
determined (Items 14 and 15) must be entered. If no entries are made for
Items 14 and 15, the program automatically assigns default values of 25°C and
0.61 m/min, respectively. If a measured K2 value for the same dust but for a
different size distribution is available, then Items 13 through 19 must be
entered. The program will then compute the correct K2 value for the dust to
be filtered based upon the adjustment for size properties. If no measured
K2 is available, then the size (actual) and density parameters of the filtered
dust (Items 18 to 21) must be entered.
The program will operate correctly with respect to Items 13 through 21
for the following data input combinations:
• Item 13 alone
• Items 13 through 15 alone
-------
• Items 13 through 19 alone
• Items 13, 16 through 19
• Items 18 through 21 alone
Any other combinations will generate an error message. If the model user
forgets to enter any values for Items 14 and 15, the program automatically
assigns default values of 25°C and 0.61 m/min as stated previously. Inspection
of the summary data input printout will indicate the error to the model user if
the true values differ from the input or default values.
2.5 CARD 5 - FABRIC AND GAS PROPERTIES
Additional parameters used to establish the drag versus fabric loading
relationship for the dust/fabric combination of interest are indicated on
Card 5. When all terms described on Card 5 are known prior to the modeling
effort (Items 22 through 28), their entry will automatically lead to the pro-
grams selection of the nonlinear drag model (which provides the more rigorous
definition of the drag-fabric loading relationship). If the reference measure-
ment temperature for Sg, W_, S.. and K_ is not entered, the program will auto-
matically assign a default value of 25°C. If the latter value is not correct,
the model user will detect the temperature error when scanning the tabular
summary data printout. If the linear drag model is to be used, it is only
necessary to enter SE and W_ when available (Items 22 and 24) and the corre-
sponding measurement temperature if not equal to 25°C. Finally, when no
input data are available for Items 22 through 28, which means that K£ must
be estimated by the program (Card 4, Items 18 through 21), all Card 5 entries
may be left blank. Items 22 and 24, S£ and WR, respectively, will automatically
be assigned default values for 25°C, thus permitting the program to proceed
with the linear modeling approach.
8
-------
2.6 CARD 6 - SPECIAL PROGRAM INSTRUCTIONS
The terms appearing on Card 6 relate to the level of accuracy desired and
the amount of detail to be presented to the model user in the model printout.
When a specific numerical value is entered for the maximum number of cycles,
Item 29, the program will automatically continue cycling until convergence is
achieved or until the specified number of cycles have been modeled. However,
printout data will be provided for the last 3 cycles only, thus eliminating
excess and ordinarily useless printout material. A cycle refers to the time
period embracing both the cleaning and filtration intervals.
In most cases, near steady state filtration conditions will be established
within 20 cycles. If no entry is made for Item 29, the program will automa-
tically stop after 4 cycles regardless of whether convergence (steady state)
is Indicated by the tabular printout for the AVERAGE values for the last 3
cycles. Concurrently, a nonconvergence message is indicated so that the model
user can run additional cycles if desired.
Regardless of the number of cycles requested, the model is structured so
that the program will execute only as many cycles as necessary to achieve
convergence (steady state) by means of a convergence checking system within
the program. This step can result in a significant reduction of computer time.
An entry of zero for the accuracy level, Item 30, instructs the program
to define the convergence limits (approach to steady state) within 1 percent
of the estimated limiting value. Should the model user desire a closer approxi-
mation, the assignment of the integer value of 1 will reduce the above dif-
ferential to about 0.3 percent. However, except for unique circumstances,
the added accuracy constraint may not be justifiable since both computer time
and costs will be roughly doubled.
-------
Items 31 and 32 enable the selection of the type and extent of program
printout necessary for the specific modeling application. If DETAILED is
requested for the tabular printout, the local drag and velocity for each fabric
element (area) of the system as well as drag, velocity and loading for each
compartment for each time iteration (increment) will be printed. In addition,
a summary table of total system penetration and average pressure drop plus in-
dividual compartment flow velocities will be printed. Finally, the average
and maximum pressure drop and penetration will be printed for the final 3
cycles depicting steady state operation.
When SUMMARY is specified, the summary table cited above but without com-
partment flow velocities, will be printed along with the cycle average and
maximum values for pressure drop and penetration. A request for AVERAGE data
provides the average statistics only for a complete operating cycle. For most
field diagnostic applications of the model, the AVERAGE printout category
should suffice. Should time plots be desired for pressure loss and penetration
behavior, PLOT is entered in Item 32.
If the fractional area cleaned, ac, is known, it can be entered via Item 33.
Since in most instances no value of ac will be available, Item 33 will be left
blank.
2.7 CARD 7 - PLOT (GRAPH) DIMENSIONS
If plots of a size other than those specified as default values are de-
sired, the plot dimensions (Items 34 and 35) must be entered.
10
-------
3.0 ERROR MESSAGES
Table 2 summarizes the various error messages that can result when cer-
tain data are omitted, entered Incorrectly, or do not fall in the valid
range. The program will not operate until these errors are corrected. The
error messages will appear on a separate printout sheet captioned DIAGNOSTIC
MESSAGES.
11
-------
TABLE 2. SUMMARY OF DIAGNOSTIC MESSAGES AND THEIR INTERPRETATION
Message
Probable cause/corrective measures
- ILLEGAL REQUEST FOR TYPE OF RESULTS
- THE NUMBER OF COMPARTMENTS MUST NOT EXCEED 30
THE NUMBER OF COMPARTMENTS TIMES THE COMPARTMENT
CLEANING TIME MUST BE LESS THAN THE CLEANING
CYCLE TIME
- THE COMPARTMENT CLEANING TIME MUST BE LESS THAN
THE TOTAL CYCLE TIME
- TIME INCREMENT TOO SMALL, I.E., < 0.01 MINUTES
- AVERAGE FACE VELOCITY OUT OF RANGE, 0.3 TO 3.0
- A CAS TEMPERATURE HAS NOT BEEN ENTERED
INVALID FREQUENCY OR AMPLITUDE FOR SHAKER
- INVALID ACCURACY CODE
BOTH TIMED AND PRESSURE CONTROLLED CLEANING
SPECIFIED - ONLY ONE IS VALID
PARTICLE SIZE DATA FOR K2 ARE INCOMPLETE
MASS MEDIAN DIAMETER OF MEASUREMENT OUT OF
RANGE 2 TO SO
STANDARD DEVIATION OF MEASUREMENT OUT OF
RANGE 2 TO 4
MASS MEDIAN DIAMETER OF DUST OUT OF RANGE
2 TO 50
Incorrect spelling of DETAILED, SUMMARY, AVERAGE or
PLOT for Items 31 and 32, Card 6.
Too many compartments were entered for Item 1,
Card 2.
Too many compartments, too large a compartment
cleaning time or too small a cleaning cycle time
were specified Items 1, 2, and 3, Card 2.
Too large a compartment cleaning time, Item 2,
Card 2, or too small a cleaning cycle time, Item 3,
Card 2, were specified.
The time increment calculated by Che program is too
small. Too many compartments (Item 1, Card 2) or
too small a cleaning cycle time, Item 3, Card 2,
will cause this problem.
Too large or too small an average face velocity was
entered for Item 9, Card 3.
A value less than or equal to 0°C was entered for
Item 10, Card 3.
A > 0 value entered for either frequency, Item 7,
Card 2 or amplitude, Item 8, Card 2. Both must be
entered or left blank for program to operate.
Only 0 and 1 are valid codes. Make certain the
number is right justified when entered for Item 30,
Card 6.
On Card 2, values for both Items 4 and 5 were
entered. Only one item may be entered per test.
A value for Ka, Item 13, Card 4, was entered along
with data to be used in correcting Ka for dust
size properties, Items 16 through 19, Card 4.
However, an omission from Items 16 through 19 has
led to any of the following error messages.
The MMD of the reference dust. Item 16, Card 4, is
out of the valid range.
The Og of the reference dusc. Item 17, Card 4, is
out of the valid range.
The HMD of the inlet dust, Item 18, Card 4, is out
of the valid range.
(continued)
12
-------
TABLE 2 (continued)
Message
Probable cause/corrective measure*
- STANDARD DEVIATION OF DUST OUT OF RANGE
2 TO 4
- BULK DENSITY CANNOT EXCEED DISCRETE PARTICLE
DENSITY
- BULK OR DISCRETE DENSITY HISSING
EFFECTIVE DRAG, SE, IS LESS THAN THE
RESIDUAL, SR
- INCOMPLETE DATA FOR NONLINEAR DRAG MODEL
- RESIDUAL DRAG, Sg, IS MISSING
- INITIAL SLOPE, KR, IS MISSING
- FRACTIONAL AREA CLEANED OUT OF RANGE, > 0 TO 1
- K2 IS OUT OF RANGE, 0.25 TO 10
- THE PROGRAM HAS BEEN TERMINATED BECAUSE OF
ERRORS IN THE INPUT DATA
'- THERE ARE NO ERRORS IN THE INPUT DATA
The Og of the inlet dust, Item 19, Card 4, is out of
the valid range.
Too large a bulk density, Items 21, Card 4, or too
low a diacrete denaity, Item 20, Card 4, were
specified.
K2 is to be estimated as is required when no values
are entered for Hz and the reference duat, but
either the bulk or diacrete particle denaitiea,
Items 20 and 21, Card 4, are missing.
Values for both Kg and Sg, Items 27 and 25, Card 5.
were entered, but the values of Sg, Item 22, Card 5,
and SR are inconsistent with theory for the form of
of the nonlinear drag curve.
Kg, Item 27, Card 5, or Sg, Item 25, Card 5, was
specified. Both terms or neither one should be
specified for model to operate.
Kg was entered but Sg was omitted.
Sg was entered but Kg was omitted.
The value for «c entered as Item 33, Card 6, is not
in the valid range.
The value of K2, whether entered or calculated by
the program, when corrected to 25°C and 0.61 m/min
is not within the valid range. Check data on Card 4
for accuracy.
Since one or more of the above errors has occurred,
program execution is stopped. Correct the error(s)
and rerun the program.
No errors were detected.
performed.
The simulation will be
Out of order and missing cards (with the exception of Card 7) will cause many of the above errors to
occur. Check card order before running program.
13
-------
4.0 EXAMPLE OF MODEL APPLICATION
An electric utility operates two, coal-burning steam-electric plants, the
first of which now uses a pressure-controlled baghouse to prevent particulate
(fly ash) emissions. It has been proposed that a continuously cleaned fabric
filter system be installed at the second plant. Both the utility operator and
the local emission enforcement groups would like to determine whether operation
of the filter system in accordance with the input data shown in Table 3 will
satisfy local emission requirements while maintaining average system pressure
drop levels within the exhaust capacity range of the induced draft fans. For
present purposes, it is assumed that operation at an efficiency of 99.5 percent
(equivalent to 0.5 percent penetration) and an average pressure drop of
< 1750 N/m2 (7 in. water) indicates acceptable performance.
Design and operating data appearing in Table 3 for the proposed "second"
plant baghouse represent a composite of information received from both utility
personnel and the dust collector manufacturer. It is also assumed that previous
measurements of uncontrolled mass emission rates and fly ash size distrubutions
at both plants are available as well as estimates for the terms K£, Sg and Wg
based upon special tests performed at the first plant. It should be noted
that the above terms might have been estimated from strip chart records from
Although a lower operating pressure loss is usually preferred, limited phys-
ical space and a desire to make use of the existing draft fans has led to the
utilities choice of the indicated pressure drop characteristics (< 1750 N/m2).
14
-------
TABLE 3. AVAILABLE INPUT DATA FOR MODELING BAGHOUSE PERFORMANCE
AT ELECTRIC UTILITY, PLANT B
Plant A
Plant B
Number of compartments
Cleaning cycle duration
Time to clean one compartment
Cleaning type
Reverse flow volume
Cleaning cycle initiation
Volume flow into baghouse
Total filtration area
Temperature of flue gas
Inlet concentration
Inlet dust mass median diameter 10 urn (Reference)
Inlet dust geometric standard
deviation
Dust specific resistance, K2
Measured at
Measured at
Effective residual drag
Measured at
Residual fabric loading
3.0 (Reference)
10.2 in.W.C.-ft-min/lb
500°F
2 ft/min
0.636 in.W.C.-min/ft
500°F
0.015 lb/ft2
30
30 minutes
1 minute
Collapse/reverse air
30,000 acfm
Continuous cleaning
600,000 acfm
200,000 ft2
350°F
5 grains/scf
7 vim
2.5
Note: All English units must be converted to their metric equivalents. See
Table 4.
15
-------
TABLE 4. ENGLISH/METRIC CONVERSION FACTORS
Quantity
To convert from
To
Multiply by
Filter resistance
Filter drag
Velocity
Volume flow
Fabric area
Areal density
in. H20
in. H20-min/ft
ft/min
ft3/min
ft2
lb/ft2
N/m2
N-min/m3
m/min
m3/min
m2
g/m2
Specific resistance coefficient in.W.C.-min-ft/lb N-min/g-m
Dust concentration grains/ft3 g/m3
Density lb/ft3 g/cm3
249
817
0.305
0.0283
0.093
4882
0.167
2.29
0.0160
16
-------
the first plant showing the pressure loss versus time traces. There Is the
Important constraint, however, that the time Intervals between cleaning be
long enough (t/2 hours) to develop a uniform density dust deposit on the fabric
surface.
The data summarized in Table 3 are sufficient to carry out the predictive
modeling operation. Following transcription of the information from Table 3
into the units and format shown in Figure 2, the data inputs are ready for
punch-carding.
Card 1 contains the title which will appear with the results. Note that
on Card 2 the time between cleaning cycles, Item 4, and the limiting pressure,
Item 5, have been left blank because a continuous cleaning system has been
chosen. The reverse flow velocity, Item 6, Card 2, is calculated from the
total reverse flow rate (30,000 acfm) and the cloth area per compartment
(200,000 ft2/30) as 4.5 ft/mln or 1.37 m/min. The average face velocity,
Item 9, Card 3, is computed from the indicated value for the total gas flow
(600,000 acfm) and the total fabric area (200,000 ft2) as 3.0 ft/min (0.915)
m/min). Inlet concentration is reported at ambient temperature, ^25°C, the
value entered for Item 12, Card 3.
The data available for K£ are sufficient to allow for correction of the
measured K£ at the first plant to the size properties of the dust at the
second plant. Thus K£, the temperature and face velocity at which it was
measured and the size properties of the two dusts are entered as shown on
Card 4. The model user should note that much of the raw field data may
appear in English units, necessitating their conversion to the metric units
used in the model.
17
-------
•taBmuMCMa.il
fcfTHC OTMVI. W •BT
W Bt I* MM
If Kg !• TO
|[ecr I
71
QM
7X
17*
FABRIC FILTER MODEL-DATA INPUT FORM
•-•rauMn r •, a TO M
I-MOUMCB ra> WW-U
t-MWIIMO rOH LIMA* BUM
TITLE
MTt
I M11111 IT
00
M 17 !• 19 20
DCCIMAL «X"T
ALL OTHCN tXTHlES MUST BE MIGHT JUSTIFICO EXCEPT
ITEMS 0. 31 ANO SZ.
-------
Twenty cycles are considered sufficient to complete the simulation and
achieve steady state conditions (see Card 6). Similarly, an accuracy level
of zero is considered acceptable for the first trial. Because the utility
personnel and the enforcement agency are concerned mainly with average emis-
sion rates and average pressure drop, AVERAGE results are requested. Since
no plotting is desired, Item 32, Card 7 has been left blank.
If the results of the simulation had indicated emission levels close to,
but greater than, the allowable level, the simulation could have been rerun
with an accuracy level of 1. If convergence had not been reached within
20 cycles, a value of 40, for example, might have been entered provided that
the costs for added computer time were acceptable.
In Table 5, the actual computer printout provided for the input data and
instructions of Figure 2 have been arranged in a convenient tabulation showing
each of four separate printout sheets. Printout No. 1 shows the actual sum-
marized input data as entered into the program so that the user can check the
data for errors or omissions. Printout No. 2 instructs the user via the
statement "There are no errors in the input data" that the modeling program
will be executed as requested. Printout No. 3 provides a listing of those
parameters whose values were computed or corrected by the program such as ac
and K£. Again, inspection of these data by the model user allows him to
determine the reasonableness of the indicated values. Finally, the AVERAGE
data shown in Printout No. 4 indicate that both the pressure drop and penetra-
tion expectations for the filter system (< 1750 N/m2 and 0.5 percent) should
be realized. In addition, Printout No. 4 also indicates that 10 cycles rather
than the 20 requested on the data input form, were sufficient to define steady
state operating conditions.
19
-------
TABLE 5. SAMPLES OF TABULAR PRINTOUT FOR EXAMPLE OF MODEL APPLICATION
PRINTOUT NO. 1
UF INPUT DATA FOR B*GHOUSE
AN KECtRIC UTILITY / PLANT B BAGNOUSE
M»sir OESIUN DATA
NU»IJIH nr COMPARTMENTS so
CQMPAWTMtsT CLEANING TIME |.0
(OFF LINE TJME)
CLEANUP. CYCLE TIME jo.n
CONTINUOUSLY CLEANED SYSTEM
REVERSE FLfli* VELOCITY J.J725
MINUTES
nP(HATING DATA
AWEPACE FACE VELOCITY
CAS TEMPERATUME
INLET OUST CCNCENTWATION
"EASURtC AT
FAHMIC AND OUST PROPERTIES
SPECIFIC RESISTANCE, *i
MEASURED AT
COMRECTEO TO
EFFECTIVE RESIDUAL OHAC. SE
MEASURED AT
RfSlO'JAL LOADING. "»
0.<»lSO
»".
II.OS
1.70
260.
0.6100
10.0
7.0
520.
260.
75.?
DECREES CENTIGRADE
G/MS
DEGREES CENTIGRADE
DECREES CENTIGRADE
MICRONS
-STANDABO DEVIATION 1.00
-STANDARD DEVIATION 2.50
N.MJN/M)
DECREES CENTIGRADE
SPECIAL PROGRAM INSTRUCTIONS
MAX NUMHER OF CYCLES MODELED 20
ACCURACY LEVEL 0
TYPE OF RESULTS REQUESTED AVERAGE
PRINTOUT NO. 2
DIAGNOSTIC
1M»KE ARE NO ERRORS IN THE INPUT DATA
(continued)
20
-------
TABLE 5 (continued)
PRINTOUT NO. 3
CALCULATED VALUES
JNUT OU3I CONCENTRATION 7.5*
CORRECTED TO OPERATING TEMPERATURE
G/M3
*»«0 OUST CA«E PROPERTIES CORRECTED FOR G»3 VISCOSITY
SPECIFIC CAKE RESISTANCE* «2 2.\u
CRAG. SE 620.
FRACTIONAL AREA CLEANED. AC
TIMf INCREMENT
3»9Ttw CONSTANT H*
MI««UTES
0.0
PRINTOUT NO. 4
UESULTS OF BAGMOUSE ANALYSIS
AN ELfCTRIC UTILITY / PLANT 8 BAGHOUSE
FQR
JO. 00 MJNuTfS (IPEHATinN, CrCLf NUHHtH 8
PENETNAIION«
AVERAGE PRESSURE DROP*
AVERAGE SYSTEM FIOMB
MAXIMUM PENETRATION*
MAXIMUM PRESSURE ORQP*
J.35E-OJ
1558.78
0.9607
1576.12 N/M-2
FOR 30.00 MINUTES OPERATION, CYCLE NUUQER 9
AVERAGE PENETRATION*
AVERAGE PRESSURE DROP*
AVERAGE SYSTEM FLO««
MAXIMUM PENETRATION*
PRESSURE DROP*
3.31E-OJ
I5S8.00 N/M2
0.9607 M/MIS
U.7UE-03
1570.95 N/M?
FOR jo.oo MINUTES OPEMAUUN. CYCLE NU-BEP 10
AVERAGE PENETRATION*
AVERAGE PRESSURE DROP*
AVERAGE SYSTEM FLO"*
MAXIMUM PENETRATION*
MAXIMUM PRESSURE DROP*
3.J31-OJ
1557.7u
0.9607
u.7«E»03
157«.50 N/M?
21
-------
REFERENCES
1. Billings, C. E., and J. E. Wilder. Handbook of Fabric Filter Technology,
Volume I, Fabric Filter Systems Study, 1970. U.S. Environmental Protec-
tion Agency, Control Systems Laboratory, Research Triangle Park, North
Carolina. EPA-APTD 0690 (NTIS No. PB-200-648). December 1970.
2. Dennis, R., and J. E. Wilder. Fabric Filter Cleaning Studies. U.S. En-
vironmental Protection Agency, Control Systems Laboratory, Reseach
Triangle Park, North Carolina. EPA-650/2-75-009 (NTIS No. PB-240-372/3G1)
January 1975.
3. Hall, R. R., and R. Dennis. Mobile Fabric Filter System. Environmental
Protection Agency, Control Systems Laboratory, Research Triangle Park,
North Carolina. Report No. EPA-650/2-75-059 (NTIS No. PB-246-287/AS).
July 1975.
4. Dennis, R., et al. Filtration Model for Coal Fly Ash with Glass Fabrics.
U.S. Environmental Protection Agency, Industrial Environmental Research
Laboratory, Research Triangle Park, North Carolina. EPA-600/7-77-084.
August 1977.
5. Dennis, R., R. W. Cass, and R. R. Hall. Dust Dislodgement From Woven
Fabrics Versus Filter Performance. J Air Pollu Control Assoc. 48_ No. 1,
47. 1978.
6. Dennis, R., and N. F. Surprenant. Particulate Control Highlights: Re-
search on Fabric Filtration Technology. U.S. Environmental Protection
Agency, Industrial Environmental Research Laboratory, Research Triangle
Park, North Carolina. EPA-600/8-78-005d. June 1978.
22
-------
TECHNICAL REPORT DATA
(Pleat nad Instruction* on the rtvent be fort completing)
1. REPORT NO.
EPA-600/7-79-043b
2.
1. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Fabric Filter Model Format Change; Volume n.
User's Guide
5. REPORT DATE
February 1979
ft. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Richard Dennis and Hans A. Klemm
B. PERFORMING ORGANIZATION REPORT NO.
GCA-TR-78-51-G(2)
0. PERFORMING ORGANIZATION NAME AND ADDRESS
GCA Corporation
GCA/Technology Division
Bedford, Massachusetts 01730
10. PROGRAM ELEMENT NO.
EHE624
11. CONTRACT/GRANT NO.
68-02-2607, Task 8
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 11/77 -12/78
14. SPONSORING AGENCY CODE
EPA/600/13
15.SUPPLEMENTARYNOTESJERL.RTP pr0ject officer is James H. Turner, MD-61, 919/541-
2925.
i«. ABSTRACT Tne report describes an improved mathematical model for use by control
personnel to determine the adequacy of existing or proposed filter systems designed
to minimize coal fly ash emissions. Several time-saving steps have been introduced
to facilitate model application by Agency and other groups. To further aid the model
user, the study is in two volumes: a detailed technical report and a user's guide. By
using selected combustion, operating, and design parameters, the model user can
forecast the expected emissions and filter pressure loss. The program affords the
option of providing readily appraised summary performance statistics or highly de-
tailed results. Several built-in error checks prevent the generation of useless data
and avoid unnecessary computer time. The model takes into account the concentra-
tion and physical properties of the dust, air/cloth ratio, sequential compartmentized
operation, and the method, intensity, and frquency of cleaning. The model function
depends on the unique fabric cleaning and dust penetration properties observed with
several coal fly ashes (including lignite) and woven glass fabrics.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATi Field/Gioup
Air Pollution
Mathematical Models
Filtration
Fly Ash
Coal
Woven Fabrics
Glass Fibers
Aerosols
Dust
Utilities
Boilers
Air Pollution Control
Stationary Sources
Fabric Filters
Particulate
13B
12A
07D
21B
21D
11E
11B
11G
13A
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (Thit Report)
Unclassified
21. NO. OF PAGES
30
20. SECURITY CLASS (TUtpage)
Unclassified
22. PRICE
*PA Perm JJ10-1 (t>73)
23
------- |