United States Industrial Environmental Research EPA 600 7 79 043a
Environmental Protection Laboratory February 1979
Agency Research Triangle Park NC 2771 1
Fabric Filter Model
Format Change;
Volume I.
Detailed Technical Report
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-600/7-79-043a
February 1979
Fabric Filter Model
Format Change;
Volume I. Detailed Technical Report
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 lii
List of Figures vi
List of Tables viii
Acknowledgments x
1.0 Summary 1
2.0 Introduction 5
2.1 Program Objective 5
2.2 Technical Approach 7
2.3 Background Information 9
2.4 Appraisal of Design Specifications 12
3.0 Basis for Experimental Model Design 15
3.1 Working Equations 15
3.2 New Filtration Concepts 15
4.0 Modifications to Fabric Filter Model 21
4.1 Specific Resistance Coefficient, K2 21
4.2 Cleaned Fabric Area Fraction, a<. - Reverse Flow Systems . . 29
4.3 Dust/Fabric System Constant, W*, for Nonlinear Model .... 35
4.4 Computer Programming Modifications 36
5.0 Description of the New Baghouse Simulation Program 41
5.1 Designed Model Capability 51
5.2 Basic Modeling Process ........ 43
5.3 Functions of the Subroutines Used in the Simulation
Program 49
5.4 Function of the MODEL Subroutine 55
5.5 Data Inputs to the Simulation Program 63
5.6 Simulation Program Output 71
6.0 Guideline Sensitivity Tests 80
References 89
Appendices
A. Subroutine Stable - Determination of Steady State . . 90
B. Baghouse Simulation Program Listing ......... 94
C. Examples of Data Input Forms, Methods of Data Entry and
Data Printouts for Various Filtration Simulations 137
-------
FIGURES
Number PaSe
1 Linear and curvilinear drag versus fabric loading curves . 16
2 Cleaned (bright) and uncleaned (dark) areas of glass bag with
partial fly ash removal. Inside illumination with fluores-
cent tube 16
3 Specific resistance coefficient versus specific surface para-
meter (S 2) for various dusts 27
4 System breakdown for I bags and J areas per bag 44
5 Baghouse simulation program, general flow diagram 48
6 Flow diagram of the MODEL subroutine 56
7 Baghouse model computational procedure 59
8 Fabric filter model - data input form 66
9 Effect of face velocity (V) and limiting pressure loss (PL) on
average pressure loss (?) 85
10 Relationship between time between cleanings, limiting pressure
loss and face velocity . 87
11 Effect of face velocity and limiting pressure drop on average
penetration 88
12 Method of fitting data to exponential curve for Check #1 93
13 Example of linear regression lines used in Check #2 93
14 Example of oscillating pressure drop used in Check #3 93
15 Fabric filter model - data input form for Example 1 139
16 Fabric filter model - data input form for Example 2 150
17 Pressure versus time plot for Example 2 (Reference Figure 16) . . . 156
vi
-------
FIGURES (continued)
Number Page
18 Individual compartment flow versus time plot for Example 2
(Reference Figure 16) 157
19 Penetration versus time plot for Example 2 (Reference Figure 16) . . 158
20 Fabric filter model - data input form for Example 3 160
21 Fabric filter model - data input form for Example 4 165
vil
-------
TABLES
Number Page
1 Supporting Data for Evaluation of Combustion and Filtration
Processes ................... • ........ H
2 Typical Causes for and Indications of Emissions Noncompliance
for Fabric Filters ....................... 13
3 Summary of Mathematical Relationships Used to Model Fabric
Filter Performance ....................... 17
4 Summary of Major Modifications to Fabric Filter Simulation
Program ............................. 22
5 Calculated and Measured Values for Specific Resistance
Coefficients for Coal Fly Ash .................. 28
6 Summary Table of Internal Data Checks ............... 52
7 Format and Default Values for Data Inputs ............. 64
8 Example of Input Data Summary ................... 72
9 Example of Calculated Value Printout ............... 73
10 Example of Point-By-Point Data Printout for Detailed Results
Results Specification Only ................... 74
11 Example of Printout Results for Detailed or Summary Data Requests . 75
12 Example of Data Printout When Detailed, Summary or Average
Results are Requested ...................... 76
13 System Operating Parameters Held Constant for Sensitivity
Analysis ............................ 82
14 Data Sampling from Sensitivity Tests ............... 83
15 Program Listing .......................... 95
16 Variables and Arrays Used in Baghouse Simulation Program, Step 1 . 129
viii
-------
TABLES (continued)
Number Page
17 Variables and Arrays Used in Baghouse Simulation Program -
Summary Table Generator, Step 2 136
18 Summary of Input Data for Baghouse Analysis (Reference Figure 15) . 140
19 Diagnostic Messages (Reference Figure 15) 141
20 Input Variables Calculated by Program (Reference Figure 15) .... 141
21 Average and Maximum Penetration and Pressure Drop Values for
Figure 15 Data Inputs 142
22 Excerpted Data for System Detailed Performance Characteristics
After 180 Minutes of Simulated Filtration (Reference Figure 15) . 143
23 System Pressure Drop, System Penetration and Compartment Flow
Distribution Versus Time (Reference Figure 15) 144
24 Summary of Input Data for Baghouse Analysis (Reference Figure 16) . 151
25 Input Variables Calculated by Program (Reference Figure 16) .... 152
26 Results of Baghouse Analysis (Reference Figure 16) 153
27 Pressure Drop and Fractional Penetration Versus Time (Reference
Figure 16) 154
28 Summary of Input Data for Baghouse Analysis (Reference Figure 20) . 161
29 Input Variables Calculated by Program (Reference Figure 20) .... 162
30 Results of Baghouse Analysis (Reference Figure 20) 163
31 Summary of Input Data for Baghouse Analysis (Reference Figure 21) . 166
32 Input Variables Calculated by Program (Reference Figure 21) .... 167
33 Diagnostic Messages (Reference Figure 21) 168
ix
-------
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.
-------
1.0 SUMMARY
A mathematical model is described for use by agency and other personnel to
determine the adequacy of proposed filter systems designed to minimize coal
fly ash emissions. The operating principles of the model have been discussed
at length in an earlier report that includes not only the model development
per se but also detailed descriptions of laboratory and field tests performed
to provide the necessary data base.1
Originally, many supporting calculations and estimating processes were
performed outside the computer program to provide more latitude in model valida-
tion experiments. Unfortunately, this approach was overly complicated and
confusing except to those individuals who were concerned with filtration research.
Therefore, the improved model described in this report has been structured so
so that emissions enforcement personnel can carry out the same modeling processes
discussed earlier but with minimal calculations outside the model. Similarly,
the input data (or its absence) determines the most reasonable path for program
execution so that the model user is spared many decisions relative to methods of
computation, choice of iteration intervals and length of program operation
required to depict a steady state operation. Although the present study is
concerned mainly with the new model development and, particularly, its practical
application, it is emphasized that the engineer should obtain as much background
combustion and filter system information as possible before undertaking any
predictive modeling.
-------
The basis for the filtration model design is reviewed in Section 3 of this
report. The introduction of three new concepts has made it possible to estimate
the performance of a multicompartment filter system in much more realistic
fashion than previously possible.
The first describes dust separation from woven fabrics as a flaking-off
process wherein the application of cleaning energy causes dust separation to
occur at the dust layer-fabric interface. Because the cleaning produces
uniquely cleaned or uncleaned areas whose drag and dust holdings are definable,
subsequent filtration and dust deposition rates as well as drag and penetration
characteristics can be estimated for the several surface elements making up the
whole filter.
The second concept is based upon a straightforward description of the
fabric cleaning process that relates the amount of dust removed to the method
of cleaning and the prior dust loading on the fabric surface. Although both
collapse with reverse flow and mechanical shaking have been quantitated, the
collapse and reverse flow process is expected to see the most use in the present
model for fly ash filtration with woven glass fabrics.
The third concept evolves from the unique penetration behavior exhibited
by glass fabrics woven from multlfilament and bulked yarns. Because of extensive
penetration through pinhole leaks ('MOO ym diameter), the estimated size prop-
erties of many fly ash aerosols undergo little change in passing through the
filter.
Section 4 deals mainly with modifications and additions to the model
originating during the current program. For example, it is now possible to
compute K2 entirely within the model by introducing relevant input data that
may include temperature and velocity of K2 measurement, and dust size and
2
-------
density properties. The same applies to the estimation of ac, the fraction
of filter area cleaned by any specified cleaning regimen with respect to the
frequency and intensity of energy input. In addition, all input parameters
such as effective drag, K£ and inlet dust concentration that are subject to
adjustments for temperature, velocity or size properties are automatically
corrected from reference to test conditions by the program. In the absence
of certain data, the program will also assign reasonable "default" values so
that the program will continue to function.
In Section 5, a step-by-step description of every aspect of the modeling
procedure is presented including the specific calculation steps involved in
the numerous iterative processes. Here, the role of each major program routine
and subroutine is described. Additionally, a complete listing of all variables
constituting model data are described with respect to identifying symbols,
units of measurement and method and location of entry on program data input
cards. Examples are given for the various types of data printout provided by
the program. The level of detail in the printout and the level of accuracy
required are determined by the model user who introduces the terms DETAILED,
SUMMARY or AVERAGE as instructions to the model. In most cases, it is expected
that "average" values for pressure drop and dust penetration over a complete
cycle (as well as the maximum levels attained by both variables) will suffice
to describe system performance. Although +1 percent accuracy should satisfy
most field applications the model user can select a more stringent level if
desired.
In addition to the model per se, guideline tables and graphs (Section 6)
have been prepared whose main role it to emphasize the relative importance of
the system variables. These data demonstrate how the absolute and relative
-------
values of many variables interact in determining overall filter system per-
formance. Used correctly, the above guidelines may help to identify unacceptable
or incomplete data prior to carrying out any rigorous modeling.
Several appendices provide additional examples of model uses as well as
the key details on program use, routines and card listings required by the
programmer.
-------
2,0 INTRODUCTION
2.1 PROGRAM OBJECTIVE
GCA/Technology Division, under contract with the U.S. Environmental Pro-
*
tection Agency, has developed a mathematical model to describe the performance
of woven glass fabric filters used for the collection of coal fly
In its original format, certain supporting calculations and estimating proces-
ses were performed outside the computer program so that the researcher might
have more latitude in his modeling experiments. The above format is not de-
sirable nor necessary, however, if pollution control personnel are required to
determine whether an existing or proposed filtration system will meet current
particulate emission standards. Aside from requiring decisions best relegated
to the filtration expert, the original model also provided a more rigorous
analysis of probable filter system performance than that ordinarily demanded
to support enforcement personnel in their decision making,
What is required by the pollution control engineer is a relatively uncom-
plicated procedure whereby he can input specific values for the controlling
filtration and process parameters into a predictive model and receive as output
a summary of the probable system performance. The present model is directed
specifically to fly ash removal from coal-fired boiler effluents where woven
glass fabrics constitute the dust collection medium and where the average and
Contract No. 68-02-1438, Task No, 5, Program Element No. EHE624
-------
maximum particulate concentrations are the primary concern. At the same time,
however, the model output should also indicate whether the predicted ranges in
fabric pressure loss and the frequency of fabric cleaning are consistent with
design specifications. For example, if system operation plans originally
postulated intermittent cleaning; e.g., every 2 hours, whereas the model indi-
cates that continuous cleaning will be required, an increase in operating pres-
sure loss and a shortening in fabric service life might be signaled. Both the
control agency and the equipment user are thus alerted to potential problems
that can be investigated before system construction is undertaken.
The primary objective of this study was to modify the original fabric
filtration model developed under prior contract with the U.S. Environmental
Protection Agency1 so that enforcement personnel can use it without extensive
training in filtration technology. Although the proposed modifications are in-
tended to provide both diagnostic and design capabilities, it is expected that
the former application will see the greatest use.
If it is desired to ascertain whether a filter system scheduled for con-
struction or just about ready to go on-line will meet local, state, or federal
emission standards, the engineer making this assessment will use the operating,
fabric and dust parameters provided by the user and/or the collector manufac-
turer. Unless the model Indicates that the filter system will not satisfy the
emission requirements, time constraints will probably not allow enforcement
personnel to determine whether, in fact, the cleaning system is providing op-
timum performance. The latter effort is the responsibility of the user or man-
ufacturer along with the adoption of any corrective measures needed to bring
the system into compliance.
-------
2.2 TECHNICAL APPROACH
As a general approach to simplifying the model so that the user is not re-
quired to make decisions nor to perform calculations beyond the realm of a basic
understanding of filtration technology, the following steps were considered.
• Certain calculations now performed outside the program, for
example, the estimation of the parameter a that is used
within the program to determine the effect of degree of
fabric cleaning on overall emission and resistance char-
acteristics, should be carried out within the program by
introducing the appropriate terms to a new subroutine.
• In those cases where the filtration engineer is given the
option to use an approximate linear drag versus the non-
linear relationship that more closely describes the actual
filtering process, the linear approach is recommended un-
less the key nonlinear parameters can be accurately
defined.
• Selection of a limiting pressure loss at which fabric
cleaning will be Initiated should be based upon operating
conditions where total system flow passes through the on-
line compartments only.
• When no data are available to define K2, K_, S , S and W ,
provision should be made to calculate K2 within the model
from measured or estimated values of particle size proper-
ties, particle density and nominal bulk density.
• When no direct measurements are available for S_ and W
a K
for the system under study, the average values derived
from previous studies should be used.
7
-------
Supplementary graphs or charts indicating model response
or sensitivity to numerical changes in input variables
should be provided the model user to avoid overly con-
servative or overly generous estimates of filter system
performance.
The format of the input data should be changed, where
appropriate, to enable the user to enter data in a more
organized or practical fashion, e.g., Baghouse Design
Parameters, Combustion Parameters, Filtration Parameters
and Fabric and Dust Parameters.
Brief instructions should be prepared describing how raw
data should be translated to computer input. The ra-
tionale for selecting specific model outputs should be
pointed out to the model user. Available choices should
be designated as:
Detailed - point by point variations over the
entire baghouse with respect to time
and fabric location
Engineering - enough information to describe point-
by-point operation with respect to time
but averaged over the entire baghouse
Summary - no point-by-point variations, with
average values only for important
parameters.
8
-------
2.3 BACKGROUND INFORMATION
2.3.1 General Appraisal of Filtration Process - Existing System
Field enforcement personnel must be able to determine whether particulate
emission levels from a given coal-fired combustion source will comply with
pollution regulations. The efficiency of gas cleaning controls for existing
systems can ordinarily be established on the basis of standard EPA testing pro-
cedures involving extractive stack sampling5 to determine controlled and un-
controlled particulate emission levels and visual estimates of plume opacity.
Preliminary observations of the plume appearance, if detectible, and any per-
iodic or random excursions in opacity from allowable levels will often aid in
evaluating the gas cleaning equipment when related to load level changes or
tube blowing procedures.
2.3.2 Combustion Process in Compliance with Emission Regulations
If the plant undergoing inspection shows no visible evidence of poor con-
trol equipment, has no past history of complaints and all compliance testing
has indicated satisfactory performance the task of the enforcement engineer is
made simple. However, it is very important that data be gathered describing
the plant operating conditions at the time of inspection, including fuel type
and load level, and design and operating parameters for the particulate control
system. In the latter instances, information should be obtained on air-to-
cloth ratios, operating temperatures and controls for the baghouse, method and
frequency of fabric cleaning, maintenance protocol, standby equipment and
emergency procedures. A file of dust collector performance data coupled with
the design and operating parameters associated with equipment use provides a
sound basis for future control equipment appraisals. Because of time restric-
tions, predictive modeling procedures would not be performed, unless some unique
-------
operational aspect of a given control system afforded a chance to improve the
model structure.
2.3.3 Combustion Process Not in Compliance with Emission Regulations
When a coal-burning power station fails to comply with emission regula-
tions, the extent to which enforcement personnel can hasten the correction of
operating difficulties depends upon their knowledge of both combustion and
filtration processes along with an awareness of the key problem areas. Prior
to resorting to any diagnostic modeling processes, the engineer should compare
the original design and operating specifications established by the user
and/or the supplier of the filtration equipment with the actual procedures in
use at the time of noncompliance with emission regulations, A representative,
but not necessarily a complete, listing of several factors that should be
considered by enforcement engineers is shown in Table 1.
Although most items listed in Table 1 are self-explanatory, a few comments
are in order for certain factors that are often associated with system malfunc-
tions or substandard performance. For example, failure to allow for possible
increases in MW load level (Item 1) or increased ash content in the coal
(Item 4) will demand increased fabric cleaning (Item 7) if the system is to be
operated within the assigned pressure constraints. The result may be decreased
bag life accompanied by much higher particle emission rates because of bag
damage and greater filtration velocities. An attempt to reduce both space re-
quirements and collector and fabric costs by operating at higher air-to-cloth
ratios (Item 6) poses the risk of increased dust penetration and reduces the
margin in collector capacity to accommodate to power levels or dust concentra-
tions higher than specified in the original design. The items discussed above
represent actions that can be undertaken by the engineer without a rigorous in-
spection of the malfunctioning filtration facility.
10
-------
TABLE 1. SUPPORTING DATA FOR EVALUATION OF COMBUSTION AND
FILTRATION PROCESSES
Operation*! or design factor
Expected effect*
Special precaution! end/or probte
1. Baie load or peaking boiler
2. Men ay*ten or retrofit
3. Fan capacity and reaponae to
variable atatic load
4. Type of coal
S. Design reaiatance (preaaure
loaa) acroa* fabric filter
6. Design air-to-cloth ratio
(face velocity)
7. Cleaning frequency and
intenaity
8. Material* of construction,
draper deiign, pressure
and temperature sensing,
and fabric cleaning
control*
9. Maintenance and aafety
feature*
Standby compartment
Rypa** capability
Alara ayateoa
Variability in flu* ga* volume, temperature
and duet concentration and composition.
Higher co*t* with retrofit, deviations from
good deiign because of limited (pace.
Cleaning frequency varie* with fan italic
capability. Poaaible variation \n ga*
handling capacity with large change* in
filter preaaure loa*.
Siae and composition of uncontrolled
effluent depend* on a*h and aulfur content
of fuel.
Fen power requirements increeae with filter
preaaure lo**. High deiign re*i*tanc*
allow* more flexibility in duet concentra-
tion* and air-to-cloth ratio.
The higher the face velocity the leaa fa-
bric area (and coat) required. Conversely,
remittance and fan power need* are greater.
Filter preaaure loaa and fan power vary
inversely with frequency and intenaity of
cleaning. Excursion* from Man operating
retictance are minimized.
Good construction and inatruBentation prac-
tice precludea panel warping, gaaket
failure*, corrosion and condenaation in
baghouae.
Standby compartment penit* aafer and Bore
rapid inapection and maintenance.
Bypai* capability prevent* irreversible
damage to fabrica and allow* for safe
boiler turn down. Excessive prenure
drop alarm* Bay prevent bag rupture.
Sice filter for BaxiauB flow-iix* com-
partment end duct heating equipment
for minimum flow, note possible
change* in dust properties with flow
rate.
Possible flow distribution and duct or
Banifold dust settlement problem*.
Exctss dust penetration in high gas
flow region*.
Frequent cleaning needed for low bag
pressure loss can decrease beg life.
Overreaponae of draft fane to static
pressure changes can cauae load level
variation*
Deiign for maximum ash content. Be
alert for changes in aixe properties
or fysa, condenaation with high sul-
fur coal*.
Deaign preaaure loss limit should be
baaed on highest possible febric load-
iogs and/or flue gas flow rate.
High velocity operation requires bese
load operation with constant ash con-
tent. Penetration will be higher
although usually not excessive.
Fabric wear increases with rate and
intenaity of cleaning. Paniculate
emissions may be higher due to
overcleaning.
Leakage of cold air into baghouse
with condensation and bag plugging.
Cooling due to inaufficient insula-
tion. Busting and jamming of compart-
ment dampers. Failure to initiate
cleaning at apecified preasure level
or to activate supplementary heaters.
Proper maintenance avoids equipment
breakdown. Lack of elan systems
may cause loss of aeverel baga, and
also lead to decreaaed exceas air in
combustion proceaa.
11
-------
No decisions or actions should be undertaken to bring a system into com-
pliance, however, until a thorough inspection of the physical plant has been
made, preferably by both enforcement and user personnel. Again, a representa-
tive but not necessarily complete listing of the more common field problems
are summarized in Table 2, Many of the conditions described in Table 2 are the
obvious results of a poor operating and maintenance regimen, particularly so
the rusting surfaces, missing bags, defective gauges, insulation free surfaces,
overflowing dust hopper and heavy dust deposition on bag compartment walls and
floor. On the other hand, certain problems relating to torn or apparently
plugged bags may arise from improper tensioning or insufficient heating to main-
tain bag compartments above dew point temperatures. Operation of the system at
too high an air-to-cloth ratio or failing to clean the fabric at sufficient
intensity or frequency may also be reflected by damaged fabric and/or excessive
dust penetration.
Therefore, even if the filter system is put back in order with the bags
replaced and other defects corrected, it is possible that initially acceptable
emissions will revert to noncompliance levels in a short time unless the basic
faults are corrected. In the situation Just described, it would aid the en-
forcement engineer if he could determine by means of a filtration model whether
one could ever expect to meet the performance specifications (pressure loss and
effluent concentration) with the actual combustion-related and operation param-
eters. If not, preliminary guidelines for corrective changes would automat-
ically evolve from the model output,
2.4 APPRAISAL OF DESIGN SPECIFICATIONS
In reviewing plans and operating specifications for new systems, the fol-
lowing guidelines may be available to aid enforcement personnel in their
12
-------
TABLE 2. TYPICAL CAUSES FOR AND INDICATIONS OF
EMISSIONS NONCOMPLIANCE FOR FABRIC
FILTERS
1. Fabric Bags
Missing, torn or nonuniformly tensioned bags
2. Clean-Air Side of Bag Compartment
Gross fabric soiling, dust accumulation on floor
3. Compartment, Duct and Hopper Leakage
Corroded panels, rust stains, peeling paint, damaged insula-
tion, holes, defective gaskets
4. Missing or Nonfunctioning Gauges
Temperature, compartment pressure
5, Defective Dampers (Compartment Isolation)
Incomplete damper closure, minimal compartment cleaning,
dust accumulation near dampers, disconnected controls
6, Over-filled Dust Hopper, Screw Conveyor
Minimal flow to plugged compartment, dust pile up inside
bags above tube sheet
7. Defective Temperature Sensing and Compartment Heating
Moisture and condensation in compartment, rusting and
probably damaged bags
8. Defective Cleaning System Controls
Damper closing incomplete or out of sequence, excessive
system pressure loss
13
-------
evaluations. First, the new system may replicate closely in physical design
and operating conditions an on-line system for which performance data are avail-
able. Second, pilot scale field tests may have been performed for similar
boiler designs and fuel properties where the dust permeability of the fly ash
can be established even though operating electrical load levels may differ.
Third, the filter system supplier has selected a set of average or typical op-
erating parameters that admittedly may be conservative. The supplier then
proposes to "tune" the installed system on a trial-*and-error basis to an op-
erating regimen that will conform to the required pressure loss and effluent
concentration levels.
The probability of success with the preliminary Cor trial) parameters
depends largely on the experience, intuition and conservatism of the vendor.
Here, the application of reliable modeling techniques by the supplier and/or
the enforcement engineer should improve the reliability of any estimates of
probable system performance. At this point, it should be emphasized that if
the enforcement group is the first to use the modeling approach Cvhlch for the
moment will be assumed to carry enough technical weight to justify design
changes in the system) then the equipment supplier may be placed in the unfor-
tunate position of having to make several costly drawing modifications or pur-
chase order changes. Therefore, it would appear logical that fabric filter
manufacturers adopt in their design efforts the same modeling procedures that
enforcement personnel will use in their assessment of the system capability.
14
-------
3.0 BASIS FOR EXPERIMENTAL MODEL DESIGN
3.1 WORKING EQUATIONS
The developmental aspects for the filtration model have been discussed in
several recent publications,1"1* It suffices here to point out that the model
embraces several well recognized filtration principles that have been reviewed
extensively by Billings and Wilder.6 A listing of the basic equations used to
estimate individual filtration parameters and/or to establish their roles within
the filtration model is given in Table 3. The indicated relationships include
those used in the original experimental model1 as well as some recent additions
from the current program; i.e., Equations 6, 7 and 9. The development and
use of the latter equations will be described in the next section. The drag
curve in Figure 1 and Equations la, Ib, and 5 through 9 in Table 3 typify some
of the fundamental relationships used in the model design,
3.2 NEW FILTRATION CONCEPTS
The introduction of three new concepts, however, has made it possible to
estimate the performance of a multicompartment filter system in much more
realistic fashion than previously possible.
The first describes dust separation from woven fabrics as a flaking-off
process wherein the application of cleaning energy causes dust separation to
occur at the dust layer-fabric interface. The result is that the first cleaning
of a uniformly loaded fabric produces two characteristic regions, the bright,
cleaned areas shown in Figure 2 and the adjacent, uncleaned areas from which
no dust is dislodged.1*2 Because there exist characteristic values for the
15
-------
to
i
u
(E
CD
FABRIC LOADING. W
Figure 1. Linear and curvilinear drag versus fabric
loading curves
Figure 2. Cleaned (bright) and uncleaned (dark) areas
of glass bag with partial fly ash removal.
Inside illumination with fluorescent tube
16
-------
TABLE 3. SUMMARY OF MATHEMATICAL RELATIONSHIPS USED TO MODEL FABRIC FILTER PERFORMANCE
Equation
framtwr
(la)
(Ib)
(2)
(J)
(A)
O)
(6)
(7)
(8)
Equation Comment*
S • P/V » S£ » K;U Equation* (la) and (Ib), which are uaed for the linear model, relate
filter drag, 5, or pressure loaa , P, to fabric loading, W. PL is the
limiting prenure loss and Up the corresponding fabric loading.
PI_ • SgV » K;V (W - W^) Cleaning initiated at PL SE is the effective residual drag, UR the
resistance coefficient and V the face velocitv
S • S » Kj W" » (K - K; ) W*(l-exp (-U'/W*)) Equations (2) and (1), which are used for nonlinear model, describe
initial curvature often aeen in S versus W curves and also the later
H' • " - WR approach to linearity. Kg i* the initial slope for curvilinear region,
SR the actual residual drag for cleaned area, and W* a system constant.
" * ($£ ~ SR + K2 "D)/*» ~ "2 If W* ia cero, program automatically uses linear model.
fn j a \ "' Equation «•) describes resultant drag for parallel flow through
r—* • u] ui I cleaned and uncleaned region* of fabric aurface. The tern ac denote*
/ ^ |- » - — « ... - — 1 A cleaned fraction of fabric surface with ita initial cleaned drag, Sc.
j^f c ul Ui / "A" refer* to total aurface fraction and "n" to the total number of
fabric element*. Subscript "u" refer* to all area* not "juat cleaned. '
Kj • 1.8 V* Equation (5) describes effect of face velocity on Kj with coal fly
ash, (HMD • 9 urn and a • 3) and at temperature T *29°C
[_ j Equation (6) define* K2 for filtration condition* value i* available for the »*me dust but with different wasured
0 t / mj (m) specific aurface properties, S0.
•j S i f «/, "1 Equation (7) predicts K: in terras of gss viscosity, j, specific sur-
Y2 - r — — 1 * * 2(3)5'3 j tmct parameter so , c.k» hulk denaitv, o. and diacrete oarticle den-
PpCc |_J-4.5 (9)1^3 * (,.5 (a)5'3 - 3 (? ):J "'Xi Pp- Equation (7) used only when no direct K: measurements are
available. The Cunningham correction, Cc , approaches one for large
(fly ash) part ic les .
1 - o /p • c ; e/~ • ?
P P P
Term* and unita
P, P, • N/o*
L
N««|« n
- v . ^ln
1 E ^3
V • n ci i o
,. N~miin
g-o
See Figure 1
K • —
8
N-min
S • — t —
R V
W* - g/m2
See Figure 2.
ac - dimenaionle**
S S - N~"in
c. u „>
A • dimensionless • 1.0
See Figure* 1 and 2.
HMD • cm"1
o • dimen*ionless
T • °C
S S cm"1
o o
f , m
u • poise
CP.~P " *
c • diraensionles*
C - Jinrnsionless
c
-------
TABLE 3 (continued)
Equation
number
Equation
Comments
Terms and units
(9) «„
/10l.l5l l°g\\
6(-—is—*/
*L * SEV
C, V It
00
(11) ac- 1.51 » 10'8 H£2-«
(12) ac - (6.00 x 10-*) (V C4 tc)0-715
te - It + tf
(13) Wp " 166.4 (Ct V £t)°-28»
(14) afi - (6.00 « ID'3) (V C± £t)°-715
(15) sc - 2.23 » IO-I* (f2 Aa Up2-"
4.9 x 10-* (f2 A C VTt)0-715
na * (O.I - Pn,) e""] Ct + C,
(16) 4C
(17) Co
(18) P»s - 1.5 x NT7 exp [l2.7 (1 - e >•"»)]
(19) a - 3.6 « 10~3 V" + 0.094
I J
(20) Pu.
1-1 J-l
£ "iJ.X
l c
Equation (9) computes distribution specific surface parameter, So.
from cascade impaetor data for a logarithmic normal mass distribution.
Baverae Flow with Bag Collapse
Intermittent, pressure controlled cleaning. Substitution of W" from
Equation (10) In Equation (11) gives area fraction cleaned, a ,
as function of limiting pressure loss, Pj,, and previously cited system
parameters. Wp accounts for the fact that the average Up value over
the cleaning cycle will exceed the Initial values.
Intermittent, time controlled cleaning. Equation (12) applies when
total cycle time, tc, is given. Mote that t. is the sum of time re-
quired to clean all compartments, Ct, plus the time between compart-
ment cleaning, tf. Face velocity, V, and Inlet concentration, Ct,
must be nearly constant for safe use of time control.
Continuously cleaned system. Equation (13), which shows dust loading
on compartment ready for cleaning, applies when Wp 210 times Wg.
Equation (14) computes ac for a continuously cleaned system where It
Is the time to clean all compartments.
Mechanical Shaking
Intermittent, pressure controlled cleaning system. Substitution of
Up from Equation (10) In Equation (IS) In conjunction with shaking
parameters f and A. determines a,.. Wp accounts for the fsct that
the average V value over the cleaning cycle will exceed the initial
values.
Cbntlnuossly cleaned system. Equation (16) computes sc In term* of
cleaning parameters f and A, and the duat accumulation over the time
required to clean all compartments (C. Vlt).
Equations (17) through (19) sre empirical relationships used to com-
pute outlet concentrations, Co, In terms of incremental Increase In
fabric loading (W* - W - WR); inlet dust concentration C.; and local
face velocity, V. The term Cg Is a constant, low level outlet con-
centration that is characteristic of the dust fabric combination.
Pn and a are curve fitting constants for specific systems.
S • cm"
- dlmenslonless
Equation (20) deplete basic iterative structure for defining system
penetration at any time, Pnt as a function of parallel flow through
"I" compartments (each subdivided Into "J" Individual areas) where
local face velocities and fabric loadings are variable with respect
to time and location. .
tc. It, tf
mln
ct - g/«3
V - m/mln
n - number of compartments
ac - dlmensionleas
f • shaking frequency - Bz
A( • shaking frequency - cm
It - time to clean all
compartments - mln
W - g/m'
V - m/mln
Pna, Pn{ - dimenslonless
a - a2/g
1 • No. compartments
J • Ho. areas per
compartment
t - time
-------
residual drag, S_, and residual loading, W , for the cleaned regions and be-
lt R "'
cause the drag and loading for any uncleaned region are also definable, it be-
comes possible to compute the resultant fabric drag for the overall filter sys-
tems by means of Equation 4, Table 3.
The second concept is based upon a straightforward description of the
fabric cleaning process1»3»** that relates the amount of dust removed to the
method of cleaning and the prior dust loading on the fabric surface. Although
both collapse with reverse flow and mechanical shaking have been quantitated,
it is expected that the former cleaning method will see the most use in the
modeling process for fly ash-glass fabric systems. This opinion is based on
the fact that the very brief and low-intensity, supplemental shaking used in
some field units does not appear to play a significant role in dust cake re-
moval for filter pressure losses less than 1500 N/m2 (6 in. H20). Equations
10 through 14, Table 3, depict the types of calculations carried out within the
program to estimate the fraction of cleaned fabric area, a , when reverse flow
cleaning.is used. If mechanical shaking is used, Equations 10, 15 and 16 are
employed to compute the cleaned area fraction.
The third concept evolves from the unique penetration behavior exhibited
by fabrics woven from multifilament and bulked yarns. A temporarily or perma-
nently unblocked pore presence (often referred to as pinholes) may contribute
to extensive penetration of the upstream aerosol. Furthermore, only minor
differences may be detected between the inlet and outlet dust size properties.
Therefore, the model is structured so that it computes the total effluent
concentration on a mass basis alone because penetration levels are essentially
Independent of size. The above situation arises because the aerosol fraction,
which sees only minor changes in size properties as it passes through pinholes
in the 50 to 200 Mm diameter range, represents 95 to 99 percent of the total
19
-------
filter emissions. The potential for extremely high collection by the undis-
turbed dust cake is seldom realized1 because of gas flow diversion through the
pores. Equations 17 through 19, Table 3, take into account the variable nature
of the dust penetration through the filter medium from the time that it is
cleaned until a substantial dust deposit has accumulated. The term, C , de-
picts a characteristic, lower limit in effluent concentration (for fly ash-glass
fabric systems) that is approached asymptotically as filtration progresses
between cleaning intervals. For present purposes, a CR value of 0.5 mg/m3 has
been selected for the lower threshold based upon laboratory measurements.
20
-------
4.0 MODIFICATIONS TO FABRIC FILTER MODEL
Major modifications to the Fabric Filter Simulation Program are discussed
in this section. As shown in Table 4, the revisions involve reductions in
hand calculations and procedural decisions by the model user, reorganization of
data inputs, more flexibility in data outputs and a restructuring of program
routines.
4.1 SPECIFIC RESISTANCE COEFFICIENT, K2
Although mathematical procedures for the computation of the specific resis-
tance coefficient, K£, were described in prior GCA publications,lj 3»** the calcu-
lation process was not included in the computer program. The reason for the
omission was that the expected level of accuracy arising from direct calculation
appeared to be no better than ±50 percent whereas data obtained from direct
field or laboratory measurements were considered much more accurate, ±10. percent,
However, if enforcement personnel are compelled to make estimates of filter
system performance in the absence of any reliable K2 measurements, the compu-
tation process called for outside the model might be overly time consuming.
Therefore, provisions have been made to carry out within the model the necessary
calculations to estimate K2.
Based upon recent studies of dust cake porosity by Rudnick and First,7 it
appears that modifications to the classical Kozeny-Carman (K-C) equation, sug-
gested by the Happel flow field structure8 afford better estimates of K2 over
21
-------
TABLE 4. SUMMARY OF MAJOR MODIFICATIONS TO FABRIC FILTER SIMULATION PROGRAM
A. Reduction in External (Manual) Calculations.
1. Incorporation of calculation of fractional area cleaned, a^, in
the program.
2. Addition of special K2 calculations.
a. To correct K£ from a reference set of size properties to
filter system size properties.
b. To estimate K2 from dust particle size and density parameters.
3. Calculation of W* for nonlinear model within the program.
4. Addition of mechanical shaking descriptors (amplitude and frequency)
for calculation within the model of cleaning parameter, a .
B. Minimizing Procedural Decisions by Model User.
t
5. Selection of number of time increments to determine iteration
period no longer required. Choice restricted to an "accuracy
code" factor of 0 or 1 for "accurate" or "very accurate" model
computations.
6. Number of repetitive filtration cycles to reach steady conditions
determined automatically.
7. Due to the addition of Item 6, the entry "total number of cycles"
now indicates the "maximum number of cycles" to be modeled
regardless of whether convergence requirements are met.
C. Data Inputs and Outputs.
8. Data inputs have been regrouped as "Design Data," "Operating Data,"
"Dust and Fabric Properties" and "Special Program Instructions."
9. Data outputs can now be selected at three increasing levels of
detail; "Average," "Summary" and "Detailed."
10. Plotted results can be requested if desired.
11. All input parameters subject to adjustments for temperature or
other specified properties; e.g., inlet dust concentration, are
automatically corrected from the reference to the filtration
conditions.
(continued)
22
-------
TABLE 4 (continued)
D. Programming Changes
12. Two additional subroutines have been added to check the input
data for Inconsistencies, missing data, and data "out of range"
of program processing capabilities. These procedures eliminate
some "blow up" conditions and unnecessary runs.
13. The simulation program now consists of three individual Fortran
programs: (a) the simulator, (b) a summary table generator, and
(c) a plot generator.
23
-------
a much broader range In cake porosity, up to 90 percent or greater. For poros-
ities ranging from 0.3 to <0.7, the classical K-C relationship
*
P
Kp vs c
and the modification discussed by Rudnick and First7
K . 18 y & (2)
PP
agree within better than 20 percent.
The term R has been defined by Happel8 as
3-1-2 (l-e)6/3
3 - 4.5 (1-e)1'3 + 4.5 (1-e)5/3- 3 (1-e)2
(3)
As used in the Kozeny-Carman relationship, R is defined as 2 k (l-e)/e3 where k
is the Kozeny constant usually assumed to be 5.0. Substitution of the latter
value in Equation 2 reduces it to the classical K-C form.
Both approaches indicate that dust cake resistance as reflected by R£ be-
comes infinitely high as cake porosity decreases. The Happel modification
shows that R approaches 1.0 at very high porosities such that the K2 expression
then provides a correct measure of single particle drag. On the other hand,
the empirical structure of the K-C function no longer applies at high porosity.
For example, at a porosity of 1.0, K2 becomes zero.
Although the calculations required for the Happel method are more involved
than those for the Kozeny-Carman relationship, either approach is readily
handled by computer. Hence, Equation (3), with modifications as discussed
in the following paragraphs, was selected for use in the revised model.
Equation (1) may also be expressed in the form
u S 2 R
p c
24
-------
where S is the specific surface parameter for the distribution of particle
o
sizes in the fly ash aerosol. Since fly ash sizing data are usually based
upon mass distributions determined by cascade impactor measurements, the size
parameters, mass median diameter (HMD) and geometric standard deviation (0g)
are available from which S can be computed for an assumed logarithmic-normal
o
distribution; i.e.,
s . - - 6 » ' g /MMD (5)
0 V *™ \ '
where d and d are the surface and volume mean diameters, respectively.
Since the porosity term, e, appearing in the expression used to define R
is best estimated from measurements of the dust cake bulk density (p) and dis-
crete particle density (p ) the term, e, in Equation (3) is replaced by
(1-F/P ) where p"/p is the solidity factor. The net result is the development
P P
of Equation (6) for use in a model subroutine for estimating K2 when the
terms HMD, 00, p" and ps can be defined. Both Equation (6) and its alternate
O
form (Equation 7 of Table 3) include an empirical correction factor of 0.33
that takes into account that the predicted values for K2 based upon the theor-
etical relationship, appear to be three times larger than the actual measured
values. The preliminary estimate of the correction factor was 0.5 as reflected
by a modified Kozeny-Carman constant of 2.5 in an earlier report.1
K2 - 6u (10 - lo82ag) /MUD » 3
p c x 3 - 4.5 (p/p )i/a + 4.5 (p/P
Jl C 8 S
The bulk density, p", can be estimated by determining the volume occupied
by a known weight of a bulk sample of the uncontrolled particulate emissions
after repeated shaking in a measuring container. Discrete particle density,
p , is estimated by pycnometer measurements or from a priori data for the dust
P
of interest. For most dusts in the fly ash size range; i.e., MMD >5 y, the
25
-------
Cunningham-Millikan Correction, GC> Is sufficiently near 1.0 to be ignored.
Gas viscosity is automatically computed within the program from the operating
temperature data input.
In some cases, K2 data may be available for dusts having the same chemical
and physical properties (including shape factor) but not the same particle size
distribution as the dust of interest. According to earlier studies, it appeared
that the relationship between the calculated specific surface parameters, So,
and measured values of Ka conformed to the So2 relationship delineated in both
the earlier Kozeny-Carman approach and the Happel concept, Equations (4) and (6).
Thus, an internal consistency was indicated for the surface to volume relation-
ships even though best estimates of particle and bulk density led to K£ predic-
tions approximately three times larger than the measured values, (see Figure 3
and Table 5). The solid regression line (Figure 3) is based on data points for
the New Hampshire and Colorado fly ashes whereas the dashed line applies to
granite dust measurements.
It was decided, therefore, to generate a second and simpler program sub-
routine to convert the K£ value determined for one set of particle size param-
eters to the K2 corresponding to the size properties of the fly ash entering
the baghouse.
(*)•'
The values for (K2)i computed either by Equation (6) or (7) represent
single point corrections that depict the effective "measured" K input at a
specified temperature and at a fixed reference velocity, usually 0.61 m/min,
and 25°C.
Equation (4-7) performs the correction for size properties in the same
manner used to adjust K2 to the gas viscosity at baghouse operating conditions.
In both cases, a single corrected value applies over the complete filtration
26
-------
10'
e
E
«t
8 10°
w
<
in
B,F and P REFER TO BENCH, FIELD
and PILOT TESTS.
jH
10
CAKE
DUST POROSITY
O COAL FLY ASH
N.H POWER
SERVICE CO.
x COAL FLY ASH
DETRIOT EDISON
0 COAL FLY ASH
NUCLA, COLORADO
V LIGNITE FLY ASH
TEXAS POWER ft
LIGHT
A GRANITE DUST
B TALC DUST
0.50
0.59
0.59
0.46
0.68
0.84
SIZING
METHOD •
ANDERSEN
IMPACTOR
MICROSCOPE.
ANDERSEN
IMPACTOR
ANDERSEN
IMPACTOR
ANDERSEN
IMPACTOR
ANDERSEN
IMPACTOR
I t 1 | | |________i|__
' 8 ' ' -
SPECIFIC SURFACE PARAMETER (S0) , c«r*
10*
Figure 3. Specific resistance coefficient versus specific surface parameter
(S 2) for various dusts.1
27
-------
TABLE 5. CALCULATED AND MEASURED VALUES FOR SPECIFIC RESISTANCE COEFFICIENTS FOR COAL FLY ASH*
Tut du«t
Coal Cly a§h
Public Service
Co.. NH (OCA)
Coal fly a«h
Public Service
Co.. NH
Coal fly ash
Hucla. CO
lignite fly ath
Taxat Powar
and Light
Duit parameter*
MKO,b
UB Og
6.17(1) 2.44
5.0 (M) 2.13
6.98(1) 3.28
3.8 (I) 3.28
11.3(1) 3.55
8.85(1) 2.5
8.85(1) 2.5
8.85(1) 2.78
Partlcla
danilty
g/«3
2.0
2.0
2.0
2.0
2.0
2.4
2.4
2.4
£
4.73 x 108
2.58 x 108
3.55 X 108
*
9.94 x 108
1.28 x 108
1.06 x 108
1.06 x 108
1.30 x 108
Caka
poroalty ,
c
0.59
0.59
0.59
0.59
0.59
0.46
0.42
0.46
Filtration
Faramatari
Valoclty,
a/Bin
0.915
0.915
0.605
0.823
0.851
0.605
0.605
0.605
*T-
21
21
21
138
124
21
21
21
Flltar fabric
Mappad cotton.
••teen weave
Glaat.
3/1 twill
Claaa,
3/1 twill
Claaa .
3/1 twill
Glaia,
3/1 twill
Glaai.
3/1 twill
Clan,
3/1 twill
Tait
icala
Pilot
Pilot
Bench
Field
Field
Bench
Bench
Bench
Measured K2,
Tait
condition*
2.29
2.29
1.40
6.35
1.05
1.34
1.34
1.34
tablenc
conditions
2 loc
0.605 B/Bln
1.85
1.85
1.40
4.45
0.75
1.34
1.34
1.34
Calculated
"2.
21»C
5.72
3.74
Ratio,
calc. KJ
•aai. K2
3-09
2.02
;
5.14
14.4
1.84
3.67
5.16
4.49
3.67
3.23
1.98
2.78
3.86
3.36
SJ
00
"Excerpted from Table 38, Reference 1.
b(I) indicates Anderaon inpactor measurement.
(M) indicates microscopic measurement (Lightfield 90 x obj).
-------
cycle. On the other hand, the special correction made for the velocity effect
on Ka is a function of the constantly changing face velocities with respect to
I
both fabric location and time.
4.2 CLEANED FABRIC AREA FRACTION, a - REVERSE FLOW SYSTEMS
c
The original fabric filtration model required that the fraction of fabric
surface cleaned, a£, arising from the cleaning process be estimated outside the
computer model. The reason for this approach was that it allowed for the use
of several alternative methods to compute a depending upon the operating con-
straints placed on the filter system. Although none of the calculating pro-
cedures were complicated, it was thought that to include all alternative sub-
routines in the program might make it unwieldly and confusing to the field
users. As a compromise approach for convenient application of the model, two
basic operating conditions have been defined.
The first one applies to a proposed or ongoing filter system that is
cleaned on an intermittent basis; I.e., the sequential cleaning of all compart-
ments is initiated at a preassigned limiting pressure loss, PT, followed by an
L*
extended period, 1 to 2 hours, when all compartments are filtering and no
cleaning takes place. The second condition applies when inlet dust concentra-
tions and constraints on operating pressure loss require continuous cleaning.
Thus, for any filtration system in which a compartment is always off line for
cleaning, the fraction of total cloth area in use at any time appears as (n-l)/n
where n is the number of separate compartments.
4.2.1 Intermittent Cleaning - Defined by Limiting Pressure Loss. PL
When there are lengthy intervals of filtration between cleaning cycles,
average and local fabric loadings for all compartments and bags will approach
each other. The limiting filter pressure loss, P , at which it is desired to
Ju
29
-------
initiate cleaning may be suggested by the filter system user or vendor. It
can be defined as shown below
PT - S_ V + K2 V (W_ - W ) (8)
L* C* i t\
where S_ and W are the characteristic residual drag and fabric loading values,
E R
respectively, for the dust/fabric system of interest; V the average face
velocity; and K2 the dust specific resistance coefficient at the indicated face
velocity (or air to cloth ratio). The K2 term may be entered as a measured
data input or alternatively It may be computed by a model subroutine based upon
Equations (6) or (7). As indicated previously, SE and WR are treated as
constants for each specific dust/fabric combination analyzed by the filtration
model. The model user is provided with estimated values for the above terms
unless direct measurements are available.
The term, W , represents the average fabric loading corresponding to the
limiting or upper pressure limit, PT, where cleaning is to be initiated. By
LI
rearranging Equation (8) followed by substitution for Wp in Equation (9)
a = 1.51 x 10~8 W 2'52 (9)
an equation is derived for use within the filtration model program as a sub-
routine; i.e.,
/ % 2 52
/ P _ C V \ *'*^*f
a = 1.51 x 10~8(-^rr—rr^— + W_ I (10)
When ac is determined by Equation (10), the average system pressure loss will
actually increase above the P_ value for brief periods until roughly one-half
the compartments have been cleaned. Should there be concern that induced- or
30
-------
forced-draft fan capacity may be reduced excessively by baghouse pressure loss
excursions above the P limit, a conservative approach can be selected. The
Li
latter procedure will take into account the fact that the second compartment to
be cleaned in a sequence of n compartments (a) will accumulate additional dust
while the first compartment is off-line for cleaning and (b) also see an in-
creased filtration velocity equal to the average value, V, multiplied by n/n-1.
Therefore, the system pressure loss just before cleaning the second compartment
will have Increased to the level, P ; i.e.,
' max* '
where P., K2, V and n have already been defined. The terms, C. and At refer to
L" 1
average inlet dust loading and the time required to clean one compartment,
respectively .
From Equation (11) it may be deduced that if system pressure loss is not
to exceed a selected maximum value, P , the cleaning must be initiated at a
max
lower level, P'. By rearrangement of Equation (11)
L
P; - P - K2 C. fvn/ .1 2'5 At (12)
L max * i L /n-lj
the model user may then compute outside the model the revised P- value, P£,
which becomes a basic data input to the model. In most practical situations,
the use of PT at the start of cleaning, is the recommended approach. In
L
Equation (12), K£ and C± must be defined at operating temperatures. The
variable impact of velocity on K£ is reflected by the fractional exponent 2.5..
31
-------
4.2.2 Intermittent Cleaning-Definedby Length of Cleaning Cycle and Time
Interval Between Cleaning Cycles
Rather than specifying a limiting pressure P., the filter manufacturer
Ju
may indicate what cleaning frequency should be used to maintain acceptable
performance. When the total time Interval for the combined cleaning and fil-
tering cycle and the filtering period alone are to be maintained constant, the
system is said to be operating under a time-controlled regimen. Such an
approach may be risky unless the gas velocities and particulate loadings are
constant. Should either vary appreciably, pressure loss excursions could
occur that might reflect adversely on gas flow stability.
To estimate pressure loss and emission characteristics for a time-controlled
cleaning system, it is first necessary to establish the total amount of dust,
AW, deposited on the fabric over the time interval, t , representing the sum-
mation of the cleaning period, Zt, and the filtering period, tf; i.e., the
interval when all compartments are on-line (t = It + t-).
AW . VC± tc (13)
Since AW also represents the amount of dust that must be removed from the fab-
ric over the time period, tc, once steady state operation has been achieved, the
area fraction to be cleaned, a , can be expressed as
Wp - AW - WR
and also as
U - AU
AW
WP'AW
32
-------
By combining Equations (9) , (13) and (15) , an expression for calculating
the data input, a , is obtained.
(0.006)0^ tc)°'715 (16)
Equation (16) appears in the revised model as part of a major subroutine.
In practice, absolute uniformity of loading with respect to compartments or
individual filter bags is never obtained, even with very lengthy filtration
periods without cleaning interruptions. However, past measurements have in-
dicated that after 30 minutes filtration following a cleaning (and filtering)
cycle of the same length, the maximum and minimum filtration velocities for
a six-compartment system differed by only 10 percent. On the premise that all
compartments see the same pressure gradient and assuming that K2 is nearly
constant, these findings indicate that the fabric loadings also differ by about
10 percent from point to point in the system. This means that the W values
appearing in Equations (9) and (15) actually represent an approximate av-
eraging of the maximum and minimum values. Accordingly, derived a values
will predict overcleaning or underlceaning depending upon the true fabric load-
ing for a given area location. In view of the computational advantage to op-
erating with a fixed value for a , the above approximation (single value) ap-
pears as the best approach until further model refinements can be made.
4.2.3 Continuous Cleaning
In certain cases, particularly where retrofit systems are involved, con-
tinuous fabric cleaning may have been selected to prevent overall pressure
losses from reaching prohibitive levels. Under these conditions, each suc-
cessive compartment to be cleaned will have the same fabric loading at the
33
-------
initiation of cleaning. At the same time, a decreasing gradation in fabric
loadings will be exhibited by the n-1 compartments remaining on-line with the
lowest loading appearing on the "just cleaned" compartment. Because the dust
loadings are not the same for all compartments when cleaning is actuated, (as
assumed for intermittently cleaned systems), the dust loading at the time of
compartment cleaning, U_, no longer defines the system pressure loss at that
time. In fact, the average system resistance is lower because of the lesser re-
sistance offered by those compartments operating in parallel with lower fabric
loadings.
The fabric loading for the compartment to be cleaned may be expressed as
Wp - (6.62 x IQ7 C± V Zt)°'284 = 166.4 (C± V Zt)°'284 (17)
when the average fabric loading is much greater, ~ 10 times, than the fabric
residual loading, W_ (which is usually the case). Note that Equation 17
K
can also be used to calculate W for intermittently cleaned systems when t is
substituted for £t. Thus, when W is redefined in terms of a and Aw, as in-
dicated in Equation (15), a final expression for ac is developed
a - (0.006)(VC, Et)°'715 (18)
c i
Wien a is computed within the program in conjunction with the other input data,
the average and maximum values for both pressure loss and particulate emissions
will appear as output.
34
-------
4.2.4 Cleaned Fabric Area Fractions, ar - Mechanical Shaking
Based upon prior studies1'9 it was determined that the degree of cleaning
obtained by mechanical shaking could be estimated by the following relationship:
a - 2.23 x 10~12(f2AW ')2'52 (19)
c s p
for an intermittently cleaned filter systems with pressure loss control. In
Equation (19), f_ Is the frequency of the shaking action, cycles/sec; Ag is
the shaker arm (half stroke) amplitude, cm; and W ' the fabric loading on the
compartment to be cleaned as defined by Equation 10, Table 3.
If the system is cleaned continuously by mechanical shaking, the limiting
pressure concept no longer holds because only the compartment due for cleaning
will have a fabric loading defined by the limiting pressure, P . Thus the
L<
cleaning parameter must be computed from the following relationship1
a - 0.00049 (f2A C.V Zt)°'715 (20)
C 81
where Et refers to the time period to clean all compartments.
Equation (20) also applies when the specified frequency of cleaning is
intermittent. In this case, the time describing the total dust deposition
interval, t , Is the summation of the cleaning time Et and the time between
cleaning tf.
4.3 DUST/FABRIC SYSTEM CONSTANT, W*, FOR NONLINEAR MODEL
To reduce further the number of computations performed outside the model,
the calculation of W* has been incorporated into the program. The magnitude
of W* determines whether the linear (W* - 0) or nonlinear (W* >0) drag model
should be used to describe system drag. If the key data inputs are not avail-
able to compute W* by means of Equation (21); i.e., experimental values for
35
-------
K (the initial slope of the drag versus loading curve) and S (the fabric
R *»
residual drag)
w* - (SE-SR + KaW^/^-Ka (21)
the program now automatically interprets blank entires or zero values for K
and SR as an instruction to use the linear model for estimation of system
drag. Conversely, when real values for K_ and SR are specified, the program
always chooses the nonlinear model.
4.4 COMPUTER PROGRAMMING MODIFICATIONS
4.4.1 Number and Length of Time Increments
In the original model,1 the user was required to determine, indirectly,
the time Increment to be used in the iterative calculations. Because too
large a time increment may yield inaccurate results and too small an incre-
ment will require excessive computer time, the actual determination of the time
increment is now decided automatically by the program. The time Increment is
determined by dividing the total cleaning cycle time, Et, by the product of the
number of compartments, (n) and a selected "number of increments (n.) per
compartment."
Time increment = It/(n x n)(minutes)
The number of increments, (n.), was varied experimentally over a broad range for
both average and extreme operating conditions. The results indicated that, in
general, four increments would suffice for most applications. Provisions have
been made in the program to increase this value to eight if the need arises.
The number of time increments is now determined from the "Accuracy Level" param-
eter, a new program data input that is entered as a special program instruction.
Assignment of a zero (0) value fixes the number of increments at four whereas
a value of one (1) will automatically increase the number of increments to eight.
36
-------
4.A.2 Determining Steady State Filtration (Model) Operation
Depending upon the selected operating parameters, the actual and/or pre-
dicted performance characteristics for a filter system will require a finite
time interval to reach steady state conditions. From this point, each successive
filtration and cleaning cycle will replicate approximately its predecessor
provided that all data Inputs remain constant.
Prior to the present modification, it was necessary to specify the number
of cycles to be simulated to establish a stop point for computer operation.
Only by examining the data printout could it be ascertained whether or not
steady state conditions had been achieved. It had been observed previously
that after 20 repetitive filtration cycles, steady state conditions were closely
approximated such that no subsequent changes were discernible in resistance
and penetration. On the other hand, it had also been noted that steady state
conditions often were reached with 10 or fewer operating cycles. Hence, to
continue with 10 additional program cycles would represent a waste of computer
time.
The programming process has now been modified so that the computer operates
until steady state conditions are achieved before any data printout takes place.
Three additional cycles are then modeled accompanied by a tabular printout or
graphical plotting so that the constancy of the data output can be verified.
These three cycles describe the operation of the baghouse at steady state.
However, to prevent the program from running Indefinitely, a practical limit
must be set on the number of cycles. Thus, where the number of cycles to be
modeled was previously specified as a required input, the "maximum" number of
cycles to be modeled now becomes the required data input. Based on prior tests
with the model, 20 cycles are generally more than sufficient to achieve equilib-
rium. If steady state has not been reached within three cycles of the maximum
37
-------
allowed; i.e., 17 cycles, the data for the last 3 cycles, 18 through 20 are
printed and/or plotted. The mechanics of how steady state is determined within
the program and the rationale for this procedure are discussed in Appendix A.
A summary of the approaches examined for estimating steady state conditions
is given in the following paragraphs.
Three criteria have been selected to determine the closeness of the most
recent or last cycle to steady state operating conditions. The first criterion
involves fitting the slope of the curve depicting pressure loss per cycle versus
time as it approaches the steady state value of approximately zero by an expo-
tential decay curve. The average pressure, P, over the indicated time frame,
which is determined by integration, is then compared to the average pressure
at infinite time predicted by the equation of best fit. When the difference
between the local and "infinite" pressure levels is less than 1 percent, the
system is considered to be at equilibrium (or at steady state).
The average pressure drop for 4 consecutive cycles is also fit to a least
squares regression line with respect to time for the second criterion. If the
slope at this time indicates that the average pressure drop is changing at a
rate of less than 0.1 percent per cycle, steady state operation is assumed.
The third criterion specifies that in those systems exhibiting oscilla-
tions in average pressure drop, the oscillations must converge or remain
constant in amplitude but never diverge before the steady state condition is
satisfied. The latter state is assumed to have been reached whenever any one
of the three convergence criteria are met (which are determined by a sequen-
tial analysis at the end of each cycle).
Convergence of average pressure loss was chosen as the indicator of steady
state since in all test cases average penetration and total cycle time also
converged when average pressure converged.
38
-------
When an accuracy code of 0 is selected, sufficient operating cycles are
generated to satisfy the average pressure loss convergence at the 1 percent
level, and the slope convergence at the 0.1 percent level. An accuracy code
of 1, which decreases the above convergence limits by a factor of 3, usually
requires that a few additional cycles be modeled.
In the case of continuously cleaned or time-controlled systems, the
"approach" to steady state is generally determined by the first or second
criterion. Certain limiting pressure systems, however, may oscillate in such
a way that the first and second criteria fail to signal a near steady state
condition whereas the third (oscillation convergence) approach will instruct
the program when sufficient cycles have been run.
4.4.3 Data Input and Output Format
Changes in the format for data inputs and outputs are shown in Table 4,
Items 8 through 11. These changes allow for a logical ordering of data inputs
to the model and better control of the volume of data generated by the program.
The above changes will be discussed in more detail in other sections of this
report.
4.4.4 Program Structure
The original program for the baghouse model consisted of a single main
program and a number of subroutines that performed all the operations from
reading the data to plotting the data outputs. To save space and reduce com-
puter time, the program has been broken up into three individual FORTRAN pro-
grams. The first program reads in the data, performs the simulation, prints
the results of all intermediate calculations (when requested) and generates
39
-------
files of pressure loss, penetration and individual compartment flows versus
time. These files are used to generate summary tables (when requested) by
the second program (or step). Finally, if a graphical output has been
requested, the third program (or step) generates the data plots.
40
-------
5.0 DESCRIPTION OF THE NEW BAGHOUSE SIMULATION PROGRAM
A complete and updated description of the baghouse simulation program is
presented in this section. Several of the modeling and actual computational
procedures appearing in an earlier report have been restated here to facili-
tate model application both for routine and experimental use.
5.1 DESIGNED MODEL CAPABILITY
In the preceding section, the basic filtration equations and the iterative
approach for treating multicompartment filtration systems have been reviewed
for convenient reference. The following discussion is intended to define the
ground rules with respect to how closely the predictive model(s) describes
actual fly ash filtration processes for utility applications. The only major
constraints are the following: (1) the inlet aerosol should consist of or
possess the general physical properties of a coal fly ash; (2) the fabric
characteristics should be similar to woven glass media used at the Sunbury
and Nucla installations; and (3) the system gas flow should be essentially
constant except for flow increases attributable to reverse air flow during
the cleaning process. Aside from the above, the model is sufficiently flexible
to meet the following operating criteria:
• The model can accommodate to a continuous cleaning regimen;
i.e., the immediate repetition of the cleaning cycle following
the sequential cleaning of successive individual compartments.
41
-------
• The model can also describe the situation where lengthy
filtration intervals are encountered between the cleaning
cycles. In both cases the term cleaning cycle refers to
the uninterrupted cleaning of all compartments in the
system. No provision is made for the random cleaning
of less than all compartments followed by continuous
on-line filtration of all compartments.
• The model can be used with a collapse and reverse flow system
or a mechanical shaking system but not for combinations of
the above. It is not intended for use with pulse jet or
high velocity reverse jet cleaning systems.
• The model can be used equally well with pressure or time
controlled cleaning cycles.
The actual information generated by the model embraces the following
areas:
• The model provides estimates of average and point values
of filter drag or resistance for the selected set of
operating parameters and dust/fabric specifications.
• The model provides estimates of average and point values
for penetration and mass effluent concentration for the
selected set of operating parameters and dust/fabric
specifications.
• The model alternatively provides an estimate of the necessary
frequency of cleaning when the maximum operating resistance
Pma is cited as an operating specification along with the
assigned values of Cj. and the selected value for V£.
In the above instances, it is assumed that the following operating param-
eters are known: inlet concentration (C.), average face velocity (V ), and
the cleaning parameters (frequency and amplitude of shaking) if mechanical
shaking is employed. In addition, the related parameters, K?, S_, W_, K_ and
ERR
SR must also be specified for the given dust/fabric combination when measured
values are available.
The system cleaning characteristics are determined by the fraction of fab-
ric area cleaned, ac, when individual compartments are taken off-line. With
respect to bag collapse systems and/or low energy shaking, the dust removal
parameter, ac, is dependent upon the fabric loading, W_, before cleaning.
42
-------
5.2 BASIC MODELING PROCESS
The basic model treats each of the "I" compartments of the filter system
as a separate element. It is also assumed that the inlet dust concentrations
and the filtration velocities are the same for each bag within a given com-
partment. However, the existence of both concentration and velocity gradients
are acknowledged due to the particle size spectrum, bag proximity and air
inlet location.
Figure 4 Indicates the distribution of volume flow rates for a filter
system consisting of "I" separate compartments. Because of the parallel
arrangement, the resistance, P, across each compartment is the same just as
the voltage drop would be for the analogous electrical circuit. In practice,
poor design or cramped quarters may prevent realization of the parallel flow
situation for some installations. The volume flow rate, q, and gas velocity,
vt through each compartment vary inversely with the individual compartment
drag.
The distinguishing feature between the new modeling concept introduced
in this study and previously reported efforts6'10 is that the surface of each
bag within a given compartment is subdivided into a number of secondary areas
each of which displays its own characteristic fabric loading (W), drag (S),
face velocity (V) and dust penetration (Pn). The fact that the contributive
role of each of these areas with respect to overall system drag and penetra-
tion can be assessed at any time during the cleaning and/or filtering cycles
is a unique feature of the new model. Note again that since all bags within
a given compartment possess identical performance characteristics, an "I"
compartment system could be described equally well as an "I" bag system.
43
-------
Co
V
C|
'II
W
|2
|ctj c2,| fc22
|VI2 ' * " '
V2I
"2*
]c2J
'22' V *
2J
CxfJ
3I
jCjj
Figure 4. System breakdown for I bags and J areas per bag.
-------
Since it is necessary to deal with several randomly distributed areas
of varying areal densities for each bag as well as several compartments,
each with its unique variability pattern, the following notational system
is introduced to describe the.various surface elements. In the multicom-
partment system, the subscripts i and j, respectively, designate the ith
compartment and the j area subdivision in each compartment. This enables
one to identify the specific element of fabric area; e.g., compartment 2,
1st area subdivision for which the local face velocity, surface loading
and effluent concentration at a specified time are then defined as V.., W?1
and C-,, respectively, Figure 4.
Although the program is designed to accept as many as 10 separate areas
(j«10) per bag, the actual number used in the iteration process (which is
automatically selected by the computer program) depends upon value of a .
Given the restriction that the number of subdivisions or areas must always
appear as integer values, the program will always select the number of
subareas that comes closest to matching the ac value. Thus, a value of 3 for
J will satisfy exactly the requirement that ac = 0.333 whereas the same J
value will also be selected as the nearest approximation to the condition that
ac - 0.35. However, If ac is 0.38, the program will select and operate with
8 areas wherein the cleaning of 3 areas provides a cleaning parameter, ac,
of 0.375.
It was indicated previously that the concentration and size properties
of the dust approaching the fabric surface and the aerial density and compo-
sition of the dust layer deposited on the filtering surface were assumed to be
uniform regardless of the location within the baghouse. Additionally, the
impact of successive fabric collapses (which may weaken adhesive bonds but
45
-------
not necessarily lead to immediate dlslodgement) has not been Included In the
modeling operations. It is assumed, that for a specific cleaning method, an
equilibrium adhesion level is reached after five to six repetitions of the
cleaning process. Beyond this point, no significant increase in dislodge-
ment can be attained without increasing the intensity of the dislodging
force. As far as the modeling procedures for the fly ash/woven glass fabric
systems are concerned, the simplifying assumptions discussed above reduce
significantly the data processing while introducing no obvious penalties in
predicting filter system performance.
The equilibrium state attained after five to six repeated cleanings
should not be confused with the normal 2 to 3 week period required for the
residual fabric dust holding, Wn, to arrive at an approximate steady state
K
level. Similarly, it should also be noted that the residual dust holding
and, in particular, the fabric effective or actual residual drags, S_ or S ,
E R
may show a gradual increase, ^100 N/m2, over the long term, ^2 years.
The general procedure for calculating all the system parameters at any
time in a cycle is described below. The calculations proceed by successive
iterations with the results from the first iteration constituting the input
for the second, and so forth. Individual subareas and compartment (bag)
drags are first calculated so that the total (average) system values for
drag, pressure drop, and flow rate can be determined. Based on the system
pressure drop and individual bag drags, the volume flow is first partitioned
among all the compartments followed by a further subdivision among the sub-
areas of each bag. Penetration and outlet concentration are then computed
for each subarea, each compartment (bag) and for the total system in the
order named. Since the dust deposition rate is determined by a specified
46
-------
flow velocity and inlet concentration, the weight of dust added to any area
on any bag can be calculated. Thus, the fabric loadings for all areas can
be calculated for the succeeding time increment.
5.2.1 General Procedures
The simulation program is composed of three individual FORTRAN programs
(or program steps) as shown in Figure 5. The following operations are
performed in the First Program Step: all data inputs are processed, the
actual filtration simulation is carried out, intermediate calculated values
are printed and the data files which will be printed and/or plotted by the
succeeding program steps are generated. All subroutines shown in Figure 5
with the exception of MODEL merely manipulate or adjust the input data in
preparation for the simulation, which is carried out by the MODEL subroutine.
Each of the subroutines is discussed in detail in the next section. During
the course of the simulation carried out in the first program step, files
are generated that contain information regarding the variations with time
of system pressure drop, penetration and individual compartment gas flows.
The Second Program Step generates a summary table of these data, if
requested by the user.
By means of the Third Program Step, the same data can be plotted as a
graphical output if requested by the model user. A complete listing of the
simulation program is presented in Appendix B.
If errors are detected in the input data, no simulation will be per-
formed within the first program step and error codes will be passed via the
data files to program steps two and three so that no summary tables or
graphs are produced.
47
-------
STEP 1
[MAIN PROGRAM]
J,
V
(INPUT AND WRITE OPERATING
DATA
^
f INPUT AND WRITE DUST AND
I FABRIC PROPERTIES
*
f INPUT AND WRITE SPECIAL
1 PROGRAM INSTRUCTIONS
J,
fCHECK INPUT DATA \
VFOR CONSISTENCY f-
^
f CALCULATE K2, ac,\
( W*, CORRECT FOR /*"
\ TEMPERATURE J
*
V FOR CONSISTENCY _JF
1
/WRITE CALCULATED^
V VALUES >T
Jr
^INITIALIZE SUMMARY
I TABLE AND * —
1 PLOT FILES
^^$f>
(PERFORM A-
SIMULATIONy*
7[f •>
vjc y
t 5
H'
3
SUBROUTINE
SUBROUTINE
L OPERAT
SUBROUTINE
SHDATA
SUBROUTINE
USER
SUBROUTINE
CHECK 1
SUBROUTINE
SUBROUTINE
CHECK2
SUBROUTINE
OUTFIL
SUBROUTINE
SUBROUTINE
MODEL
STEP 2
GENERATE
r - -* SUMMARY
I TABLE
|
|
|
1
^— ^—x^ STEP 3
/ DATA FILES I utnciuiit
-- WAP, pn and vj * GRAPHICAL
\veraus t 7 OUTPUT
\ J
\
\
J
Figure 5. Baghouse simulation program, general flow diagram.
48
-------
5.3 FUNCTIONS OF THE SUBROUTINES USED IN THE SIMULATION PROGRAM
DESINE Subroutine
The main function of this routine is to read Card 1 (the heading) and
Card 2 (basic design data) followed by printing these data as they were en-
tered into the program. Note, however, that blanks in numeric fields (e.g.,
Card 2) are read as zeroes by the program. The headings for the input data
summary are also generated by this routine. The above steps enable the
user to confirm that the program will operate upon the correct data inputs.
OPERAT Subroutine
This routine reads data from Card 3 (operating data) and writes the next
section of the input data summary. Also, when no value for the measurement
temperature of the inlet dust concentration has been entered, a default
value of 25 C is automatically assigned. The default temperature and the
baghouse gas temperature are converted to absolute temperatures (degrees
Kelvin) by OPERAT for use by other routines that perform temperature and
viscosity corrections.
SWDATA Subroutine
All dust and fabric properties (Cards 4 and 5) are read by the SWDATA
subroutine. After reading the data, the program automatically decides which
default values, if any, should be assigned and generates a summary of the
input data.
If K2 must be estimated because no previous or measured value is available
for entry, default values will be assigned to Sg and WR if no measured values
for the latter are available. In addition, any temperature or velocity of
measurement needed for the computation of K2, S_, S_ and K_ will be assigned
IS K K
default values (25°C, 0.61 m/min) if these data are not available for entry.
49
-------
The output from the SWDATA routine is a summary of the input data with some
modifications, for the special circumstances described below:
• If a known (or measured) value for K£ is entered and
no corrections or estimates are required, only K2 and
its temperature and velocity of measurement will be
printed.
• If K.2 must be estimated, the inlet dust size descriptors
(mass median diameter and geometric standard deviation)
discrete particle density and bulk density will be
printed.
• If K.2 is to be corrected for size properties, K? and
the size properties for the reference and inlet dusts
will be displayed on the printout.
• When all data required for the non-linear drag model
are entered, (S_, S_, W_ and K_) all will be printed.
£ R R R
However, if only S_ and W_ are available for entry
b R
then they alone will be printed.
USER Subroutine
Special program instructions (Card 6) are entered via the USER routine.
A default value for the type of tabular results is assigned automatically if
no input level has been entered. At present the default value is the AVERAGE
category.
The requests for tabular and graphical results are also checked at this
point for consistency. The input data are then returned for display in the
input summary.
No printout value for a is shown except for the unique situation where
it has been provided as a data input.
The time interval required for iterative calculations will be determined
by the input accuracy code. A default value of 0 (zero) will automatically
be assigned to the accuracy code when the user makes no entry. The accuracy
50
-------
code also determines the boundary conditions for the comparisons made in
the stabilization routine (STABLE).
CHECKI Subroutine
Many of the preliminary input data checks are performed by the Q3ECK1
subroutine. A complete listing of these checks is presented in Table 5.
Those checks performed in CHECK1 are identified by an asterisk (*). If
an error is encountered, this subroutine prints an error message and returns
an error code to the main program indicating that no modeling should be
performed.
However, even when an error is indicated, four additional subroutines
are carried out before program execution is stopped. These subroutines are
subroutine SETUP, CHECK2, OUTFIL and PLOTIN. Any additional errors will
thereby be indicated.
SETUP Subroutine
This subroutine performs the majority of the input data conversions (or
corrections) and calculations. If K2 for the inlet dust has not been speci-
fied in the input data, K2 is then estimated from specified data inputs
(size properites, bulk and discrete particle density) or K2 is corrected for
differences in size properties between the reference dust and the filtered
dust. .The effective residual drag, SE> is corrected to correspond to a
loading equivalent to the residual fabric loading, W . Viscosity corrections
are made to Koj S_, S-, and K_ and the Inlet dust concentration, C , is
ERR *•
corrected to the filtration temperature. An average fabric loading is
estimated as a first approximation to the actual loading distribution. The
system constant, W , is calculated if the non-linear model is to be used.
The SETUP routine then calls the subroutine CLEAN whose role is to calculate
51
-------
TABLE 6. SUMMARY TABLE OF INTERNAL DATA CHECKS
Subroutine
indicator
Variable
Range or other constraints
Valid (acceptable) ranges of variables — (data inputs must
fall within constraining range or program will not function)
*
*
*
*
t
Number of compartments, N
Average face velocity, V
Gas temperature, Tg
Mass median diameter, HMD
Standard deviation, eg
Fractional area cleaned, ac
2 to 30
0.3 to 3 m/min
Greater than 0°C
2 to SO urn
2 to 4
0 to 1
Specific rcHlHtnnce coefficient, K2 at 25 C 0.25 to 10 N-mln/g-m
Accuracy code
Supplementary checks
Compartment cleaning time
Compartment cleaning time
Bulk density
Residual drag, S_
Type of tabular results
Type of plotted results
0 or 1
<_ Cleaning cycle time
<_ (Cleaning cycle time)/N
< Discrete particle density, Pp
< Effective drag, Sg
Specify as DETAILED, SUMMARY
or AVERAGE or leave blank
Specify as PLOT or leave blank
Checks for incomplete or conflicting data
Time or pressure controlled cleaning
Shaking frequency and amplitude
K2 value available
KZ value not available
Residual drag S_ and initial slope K_
Specify one only
Specify both or none at all
Specify reference and filtration
size parameters (MMDi, HHD2,
ogj and 0g2) or none at all
Specify MMD2, 0g2, Pp and p"
Specify both or none at all
Checked in CHECK 1
fChecked in CHSOC2
^Checked in USER
52
-------
the fractional area cleaned, a . The a value computed under subroutine
c c
CLEAN is then used to calculate the number of elemental subareas into which
the compartments (or bags) should be divided and the number of these area
that should be cleaned during a cleaning cycle.
CLEAN Subroutine
As stated previously, this routine calculates the fractional area
cleaned, a .
CHECK2 Subroutine
Calculated and corrected values are checked for consistency and accept-
able range in magnitude by this subroutine (see Table 6). The error code
from CHECK1 is passed to CHECK2 and finally back to the main program.
OUTFIL Subroutine
This routine prints out all calculated values and those which have been
corrected for viscosity and temperature.
PLOTIN Subroutine
The x and y axis lengths of any graphs to be generated are read by
PLOTIN. In addition, this routine activates the filing process used to
generate the summary tables and plots. The error code from CHECK2 is passed
to the PLOTIN subroutine. If errors exist, indicating codes are written into
the pressure versus time file. When no errors exist, these codes serve to
indicate whether or not a summary table or plot has been requested.
MODEL Subroutine
Subroutine MODEL performs the actual simulation of the filtration
process. All preceding program operations merely ptvepare the data for input
to MODEL. Because the MODEL subroutine is the backbone of the entire program,
it will be discussed in a separate section.
53
-------
CAKDRG Subroutine
The drag contribution due to dust cake accumulation on the fabric is
calculated by the CAKDRG routine using either the linear or non-linear drag
mode1.
PENET Subroutine
Dust penetration is computed by the PENET subroutine as a function of
fabric loading and local face velocity.
STABLE Subroutine
After every complete cycle (filtration plus cleaning interval) STABLE
is called by MODEL to determine the proximity to steady state conditions.
After four complete cycles, sufficient data have been compiled by STABLE
to initiate the three step comparison operation. The first step compares
the average pressure value for the indicated number of cycles to the value
predicted at infinite time (the latter estimated from an exponential curve
fitted via a linear regression to the average pressure drop versus operating
time relationship). The second step compares the predicted value of the change
in pressure drop from a linear regression of average pressure drop with \
time to the actual average pressure drop at that time. If the compared
values are within predetermined limits, the system is said to be at steady
state. The third and last comparison checks the oscillating characteristics
of the average pressure drops. If the oscillations are decreasing, the
system is said to be at equilibrium. If any of the above criteria are met,
a signal is returned to the MODEL subroutine indicating convergence. These
three comparisons are discussed in more detail in Appendix A.
The error checking routines have been Incorporated into the model to
eliminate unnecessary runs caused by, (1) mispunched and "out-of-order"
54
-------
cards; and (2) insufficient or conflicting data. These routines will detect
most of the common errors, based upon the present testing and experimentation
with the program.
INITAL Subroutine
Variables used in the MODEL and STABILE subroutines are initialized
in this section.
RESTRT Subroutine
This subprogram is executed only if a limiting pressure-controlled
system was originally specified but the system must, in fact, clean continu-
ously. The system is redefined as a continuously cleaned system and the
simulation is restarted. Messages to that effect are printed in the output
by RESTRT. This subroutine can be called no more than one time during the
simulation.
5.4 FUNCTION OF THE MODEL SUBROUTINE
5.4.1 Overview
The actual simulation is carried out via the MODEL subroutine. When
the input data have been entered into the program, corrected for temperature,
viscosity or velocity, and have been checked for completeness and consistency,
the simulation is performed. A general flow diagram for the MODEL subroutine
is shown in Figure 6.
With the exception of the addition of the check for steady state opera-
tion (subroutine STABLE), the MODEL subroutine has undergone only minor
revisions since its original development.1 Figure 6 summarizes the major
program steps within the MODEL subroutine as it presently stands.
55
-------
Figure 6. Flow diagram of the MODEL subroutine,
56
-------
Within the time loop, the first step is to determine whether a complete
cleaning and filtering cycle has been performed. If a complete cycle has
been performed, the system is checked for continuous cleaning. If after
three or more complete cycles, a limiting pressure system is continuously
cleaning the simulation is restarted via RESTRT. If the system was originally
described as continuously cleaned in the input or if a limiting pressure
system operates with a finite, nonzero time between cleaning, then data
processing continues through the STABLE subroutine, which after four complete
cycles, checks for steady state. Referring again to the first step, if a
cycle has not been completed, a check is made to determine whether a com-
partment was just cleaned (bag loop No. 1). If no compartments were cleaned,
time is increased by an additional time increment (determined by the program)
and the calculations proceed through the time loop and back again to the
beginning of the time loop. However, if a compartment was just cleaned and
is scheduled to be brought back into service during the current time loop,
then time is not incremented. This step is necessary to properly depict the
effect of a cleaned compartment being put back on line nearly instantaneously
(within 0.01 minute).
Once steady state is achieved, the program begins to count the number
of completed cycles such that only three cycles will be modeled beyond the
point at which steady state was achieved. The performance characteristics
of these three cycles constitute the results of the program. If steady
state is not achieved within three cycles of the "maximum number of cycles,"
*
the performance characteristics of these last three cycles along with a non-
convergence error message comprise the program results. Throughout the course
of the last three cycles, the results of intermediate calculations are printed
(if requested) and files containing pressure drop, penetration and individual
57
-------
compartment flows as functions of time are generated. After three steady state
cycles have been modeled, control is returned to the main program (Step 1).
5.A.2 Computational Procedures
The computational procedures are based on an iterative calculation
method whereby the results of calculations at time = t are used as input
to the calculations at a time = t + At. Also, since each compartment (or
bag) is composed of a specific number of discrete areas, each having its
own drag and penetration characteristics, calculations are performed on an
area-by-area and bag-by-bag basis.
The following paragraphs provide a description of the procedures and
equations used to calculate system performance. A diagram of the basic
computations performed is shown in Figure 7. A tabulation of relevant
equations with reference to where they are treated in the report is also
included in Figure 7.
5.4.3 Drag Computation
Cleaned fabric drag is a predetermined input that is not computed by the
program. It is set equal to the effective residual drag, S', if the linear
c,
drag model is selected and to the residual drag, SD, if sufficient data for
K
the nonlinear drag model have been entered.
Area drag values are computed by the linear or nonlinear drag models
with the subroutine CAKDRG. The choice of subroutines is automatically
performed by the program which selects the nonlinear model when W* has any
nonzero value. A zero value for W* will automatically lead to computer
calculations by the linear drag model. Note that W is calculated within
the SETUP subroutine and that W* will be nonzero only if values for K and
SD are entered.
K
58
-------
DETERMINE fAURIC DRAG,
-J/LOOPONTIME)
-J/LOOP ON BAGS »2A
-W'tOOP ON AREAS A
CALCULATE FLOW VELOCITY FOR AH AREA ON A BAG, V.
CALCULATE URAG FOR AN AREA ON A BAG, S. .
CALCULATE DRAG FOR A BAG, S.
CALCULATE SYSTEM DRAG, S.
CALCULATE SYSTEM FLOW AND PRESSURE DROP, V£; P
_j/LOOP ON BAGS »3
ON AREAS
I.
CALCULATE FLOW VKLOCITY KOR AN AREA ON A BAG, V. .
CALCULATE PENETRATION FOR AN AREA ON A BAG, Vn
CALCULATE NEW FABRIC LOADING FOR AN AREA ON A BAG,
Jt » At
CALCULATE FLOW VELOCITY FOR A BAG,
CLEAN A BAG IF NECESSARYj
CALCULATE TOTAL PENETRATION, Pa
END OF CALCULATIONS
EQUATION
USED
1, 2, 3
7, 8
10, it. i:
n
u
Figure 7. Baghouse model computational procedure.
59
-------
The area drag equations for the linear model are:
S_ c "* A IT— * ti*
44 17 2 4 4
and for the nonlinear:
S - SR + K2 x w* + (K2 - K2 )W* (1 - e - W' /W*) (23)
where S.. -= the drag for the j area on the i bag at time • t
S' - effective residual drag for cleaned fabric
Kt
S_ - residual drag for cleaned fabric
K
K2 = specific cake resistance for the area
W' - absolute fabric loading less the residual fabric loading
K_ = initial slope of the drag versus loading curve
W = constant dependent on fabric and dust properties
t = time
The specific cake resistance (K2) is a function of velocity:
R2ijt - *2 JV°i (24)
where K2 is the specific resistance at 0.61 m/min and the actual gas temper-
ature. Corrections for gas viscosity and velocity changes are carried out
within the program's initiation step (subroutine SETUP).
Since the flow velocity for a specified area is not determined until
the system pressure drop and area drag are known, it must be estimated from
the previous system pressure drop and the previous drag on the area:
44 ™ "«. _ *«• -H m 4 4 (*5)
60
-------
The total or average drag for a compartment (bag) is calculated for a
parallel resistance network of J equal areas as:
i/s. .
t j=l ljt (26)
Similarly, total system drag is calculated for I bags as:
I
S. - I/ i-» 1/S.
t i-1 Xt (27)
For convenience in data processing, the drag value for any compartment
20
undergoing cleaning is set equal to 10 in lieu of plus infinity because
the compartment velocity is zero. However, since the parameters describing
overall system performance are based on total fabric area, the value of I
in Equation 27, which designates the total number of system compartments,
is not changed. Total baghouse flow can, therefore, be held constant
while the average flow velocities for the individual compartments are permitted
to vary.
The total or average system pressure drop is calculated from the total
system drag and the operating average face velocity. Additionally, when a
compartment is being cleaned via reverse flow, the reverse flow air is
factored into the computed pressure drop and flow rate.
When reverse flow air is added to the system, the average system gas
velocity is calculated by:
Vt = Vc + VR/I (28)
For a constant flow system, the pressure drop is calculated by:
Pt = Vc St * VR V1
where V m specified constant system velocity
V - reverse flow velocity for a single bag
ft
61
-------
If no reverse flow is used, VD is zero in Equations 28 and 29. Once
K. -- -—
the system pressure drop is known, the calculated flow velocity through an
area can be calculated:
V " Pt/Si1 (30)
5.4.4 Fabric Penetration
Penetration through a specified subarea is calculated by the subroutine
PENET from the empirical relationships discussed in Section 3:
C
where PQ11 * penetration through the j area on the i bag
t
W.. = cloth loading minus residual loading at time = t
C_ = residual concentration, 0.5 mg/m3, a system constant
n.
C. ° inlet concentration
Pn - 1.5 x ID"7 e12-7(1 " e"1*03 vij > (32)
8 u
a - 3.6 x 10~3/(V )** -I- 0.094 (33)
and V * face velocity of the j area on the i compartment (bag) at
time » t.
Once the face velocity and penetration have been established for an
area, the dust deposition rate can be calculated. The fabric loadings
used in the calculations for the succeeding time loop are calculated from:
W.. - V.. x (1-pn.. ) x At x c. + W..
lJt + At lh 1Jt * ^t (34)
62
-------
Note that when a compartment (bag) is being cleaned, the velocities
on each of its areas are zero and thus no dust is added to the bag. The
average flow velocity through a compartment (bag) is calculated in the same
manner as that for an area (Equation 30) except that the total compartment
drag is used.
After the compartment filtering (or on-line) time has progressed to the
point where it is equal to the cleaning cycle time minus the time required
to clean one compartment, cleaning is initiated. This entails taking the
compartment off line followed by setting its drag equal to 1020 to adjust
for the zero flow condition.
Total or average system penetration is simply the total mass emitted
divided by the total mass input:
I J
11 **\
After all calculations for time = t have been completed and the fabric
loading for the next time loop has been calculated, one proceeds to the
next time iteration.
5.5 DATA INPUTS TO THE SIMULATION PROGRAM
The necessary data inputs to the model are presented in Table 7 along
with a listing of the symbols used to represent the variables, the units
in which each variable must be expressed for entry in the model, the location
and format of each variable, and finally the relevant default values. To
simplify data entry, a coding form (Figure 8) was developed. On the
coding form, all entries not containing an implied decimal point (indicated
by a triangle) with the exception of Items 0, 31 and 32 should be right
justified. For example, the number 100 would be placed in the three furthest
63
-------
TABLE 7. FORMAT AND DEFAULT VALUES FOR DATA INPUTS
£
5
g
i
i
H <
ii
&
S
s
§
u
5
•K
i
*
t-
§
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
24
25
26
27
28
Item
Title
ItuBber of compartments
Compartment cleaning time
Cleaning cycle time
Time between cleaning cyclee
Limiting pressure drop
Reverse flow velocity
Shaking frequency
Shaking emplitude (half stroke)
Average face velocity
Gee tempereture
Inlet dust concentration
aeaaured at temperature of
Specific reeietence coefficient
measured et tempereture of
emaaured at velocity of
aeaaured at Baas median diameter of
measured at geometric standard
deviation of
Mass median diameter of Inlet dust
Geometric standard deviation of inlet duct
Discrete particle density of Inlet duet
Bulk deaalty of inlet duet
Effective realdual dreg
aeaaured et temperature of
Residual fabric loading
Realdual dreg
•eaeured et temperature of
Initial elope
eeaaured at temperature of
Symbol
n
At
It
t
PL
VR
f
A
V
T
ct
T
K2
T
V
HMD,
Og,
MKD2
0(2
"p
IT
SE
T
WR
SR
T
KR
T .
Units
-
-
min
•in
aln
N/.'
a/Bin
cps
cm
•/•In
ft
°C
8/-3
°C
M-aln/g-B
°C
B/aln
ua
-
ua
-
g/-l
g/ca3
N-min/m3
°C
g/BJ
H-ain/B3
°C
n-ain/g-a
°C
Card
1
2
2
2
2
2
2
2
2
3
3
3
3
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
Starts in
column
1
1
5
11
17
23
28
35
39
1
8
13
19
1
7
12
20
24
28
32
36
42
1
6
11
17
22
27
33
Format
8A8
13
F5.1
F5.1
F5.1
F4.0
F6.4
F3.1
F4.2
F6.4
F4.0
F5.2
F4.0
F5.2
F4.0
F7.4
F3.1
F3.2
F3.1
F3.2
FS.3
F5.3
F4.0
F4.0
F5.1
F4.0
F4.0
F5.2
F4.0
Default* Note
0.5lt/n
4
e
0 b
t
b
25
c.d
25
0.61
d
d
d.e
d,e
e
e
350* f.g
25
50" f.g
I
25
(
25
(continued)
-------
TABLE 7 (continued)
-
IK
"•h
1*
V.
29
30
31
32
33
34
35
Item
Maximum number of eyeles modeled
Aecuraey code
Type of tabular results
Type of plotted results
Fractional area cleaned
x axis length
y axis length
Symbol Units Card
nc - -.
0 or 1 - 6
' 6
6
j 6
c
inches 7
inches 7
Starts in
Column
1
$
8
17
22
1
7
Format
13
12
AS
A4
F3.2
Fi.2
FS.2
Default*
0
Averege
0
5
Rote
h
i
i
;
k
these values are used when no entry has been made for the parameter
.1
'used only when Kj is to be estimated from site properties
Notes: a. Enter item 4 or 5, but not both
b. Enter items 6 or 7 and 8, but not both
c. Enter items 13 through 15 when *2 measurement is available
d. Enter items 13 through 19 when £2 measurement must be corrected for size properties
e. Enter items 18 through 21 when Kj i* to be estimated from dust siie and density parameters
f. Enter items 22 through 28 for nonlinear drag model
g. Enttr items 22 through 24 for linear drag model
h. Generally 20 cycles are sufficient
i. For tebular results specify DETAILED. SUMMARY or AVERAGE,for graphical results specify PLOT or lesve blank
j. Enter only in special case when a measurement is available
K. Card can be l«it out if default values are eufficient or if no plotted output is desired
-------
FAMIC FILTER UOOCL-MTA INPUT FONM
111111111111111111111111111111111111111111111111111111111111
1 I » I rgM I JX I
jBi]iiii|umMimiimimBiiamimmmimiiimmii
11 fi»i
mm*mmmi*mm+mm4mmt
1111111111111111 n 111111111 ii 11 n 1111111111
lllll • llail • til I • 111 11M 1111111111111 ii 111 n I
i
mmmmmm• ••••!••••••••••«•• •
to-totoyta,to-J-.-I<-^---JM----,
ITllllllllllBllllillllllllllllllllllllllUlllllllilllllllllllllll
n »
Til • I 111 i1111111111111111111111111111111111111Minin111MiM 11
M » A-MUB I
MT M Mrr jurmo norr nf mm o.» «w u.
Figure 8. Fabric filter model - data input form.
66
-------
right blocks in a four block field. The first card (Item 0) is a title or
heading card. The information on this card appears as a heading on all
printout material along with the input data, summary tables and graphs so
that the user can readily identify each simulation. Input data have been
grouped into four general categories; i.e., Design Data, Operating Data,
Dust and Fabric Properties and Special Program Instructions.
5.5.1 Design Data
Design data are to be entered on the second card. Item 1 refers to the
number of parallel compartments each of which is cleaned independently and
sequentially. Baghouses operating in parallel but on different cleaning
schedules cannot be modeled. The compartment cleaning time (Item 2) is the
length of time that any one compartment is off-line for cleaning. The
cleaning cycle time (Item 3) is the time required to clean the entire bag-
house, Including any time during the cleaning cycle when all compartments are
on-line. For example, given a 10 compartment system whose cleaning schedule
consists of the following steps:
1. all compartments on-line - 1 minute
2. one compartment off-line for cleaning - 3 minutes
The cleaning cycle time is 10 * (3 + 1) or 40 minutes and the compartment
cleaning time (Item 2) is 3 minutes.
Items 4 and 5 describe how the cleaning cycle is to be initiated. If,
after a cleaning cycle, the baghouse is scheduled to operate without cleaning
for a specified amount of time, the time interval between cleaning cycles,
(Item 4), must be entered. However, if after a cleaning cycle, the baghouse
is allowed to filter until a predetermined pressure loss is reached, the
limiting pressure (Item 5) should be entered instead. Finally, if the
67
-------
system is continuously cleaning with no extended filtration time between
cleaning cycles, then neither Item 4 or 5 should be entered. If values for
both are entered, an error will result and program execution will cease.
The last three items on Card 2 describe the cleaning action itself.
Only one type of cleaning method can be specified. If a system uses both
reverse air and a shaker-type cleaning action, only the reverse air should
be specified. If the cleaning action is entirely shaking, then the shaker
amplitude (half stroke) and frequency should both be entered. Since the
reverse flow velocity is not used in the determination of the degree of
cleaning, it is not a required value for description of cleaning intensity.
Its only purpose is to indicate the effect of the additional flow (increased
air-to-cloth ratio) on pressure drop and penetration. Reverse flow velocity
is defined as the reverse air flow rate divided by the filtration area of
one compartment (or the number of compartments cleaned simultaneously).
5.5.2 Operating Data
Item 9, the average face velocity (or air-to-cloth ratio), is the total
system air flow at operating conditions divided by the total filtration area.
Since the relationship between penetration and velocity was derived from
laboratory tests in which the velocity ranged from about 0.3 to 3 m/min,
the average face velocity must not exceed this range. The inlet dust con-
centration (Item 11) can be specified at any reference temperature (Item 12).
The program will correct the reference concentration to that corresponding to
the inlet gas temperature (Item 10). If the temperature of measurement is
not specified, a default value of 25 C is assigned by the program.
68
-------
5.5.3 Dust and Fabric Properties
Two cards are required to enter the data describing dust and fabric
properties. Data pertaining to the specific resistance coefficient, K2,
are entered on Card 4. Three options are available to the user depending
upon how many data are available for K2- If K2 for the dust in question
is known, it should be entered along with the temperature and velocity
associated with its measurement (Items 13 through 15). No additional data
should be entered on Card 4 if K2 Is known. If measurements are available
for a similar dust (i.e., same shape factor, packing density, discrete
particle density) but having different size properties, the K2 corresponding
to the dust for which it was measured including the related size properties
of the dust and other relevant measurement conditions should be entered as
Items 13 through 17. In addition, the size properties of the dust to be
filtered must be entered (Items 18 and 19). Finally, if no measured value
for K2 Is available, but the size and density properties of the inlet dust
are given, Items 18 through 21 alone should be entered. In this last case,
an estimate of K2 will be made by the program. Referring to Items 14 and
15, if no values are entered for the measurement conditions, default values
will be assigned. Insufficient or conflicting data on Card 4 will cause
the program to return error messages and no modeling will be performed.
The remaining dust and fabric properties are entered on Card 5. When suf-
ficient data are available for the nonlinear drag model, all the parameters
on Card 5 must be entered. If, however, the linear drag model is to be used
in the calculations, only S , W and the temperature at which S_ was measured
should be entered. If K2 is to be estimated by the program, and no data are
available for S0 and W_, the card may be left blank and default values will be
a K
assigned for S,, and W_.
C K
69
-------
5.5.4 Special ProgramInstructions
Special instructions to the program are entered on Card 6. The first
item (Item 29) denotes the maximum number of complete operating cycles to be
modeled if convergence is not achieved. Convergence is generally achieved
in less than 20 cycles. A value of 20 should therefore be entered unless
fewer cycles are desired regardless of convergence. The accuracy code
(Item 30) simply modifies the limits of convergence and the length of the
time interval, as was discussed in this report under modifications to the
model. A value of zero should be entered unless the results of a previous
simulation with an accuracy code of zero do not appear to have reached
stable values. Three types of tabular results can be requested via Item 31
as described below:
Level of detail requested
DETAILED•
SUMMARY
AVERAGE-
Type of Results Printed
Point by point variations
in drag, flow and loading
for each area of the system
versus time and location
Summary of system pressure drop
and penetration versus time.
Average and maximum pressure
and penetration for a complete
operating cycle.
If Item 31 is left blank, AVERAGE is assumed. If graphical output
is desired "PLOT" should be entered for Item 32. It should otherwise
be left blank.
70
-------
If the level of cleaning, a£ (Item 33), is known it can be entered. In
general, a value for a will not be available and Item 33 must be left blank.
Finally, if plotted output is requested and axis lengths other than defaults
are desired, they should be entered in Items 34 and 35 (Card 7). If the
default values are acceptable or if no graphs are requested, this card can
be omitted from the input deck.
With respect to the data input form (Figure 8), all numbers without
decimal points should be right justified. The small triangles in certain
fields specify the decimal point location.
Examples of input data forms for a few selected types of simulations and
the results of the simulations are presented in Appendix C.
5.6 SIMULATION PROGRAM OUTPUT
As discussed previously, three levels of detail may be requested for the
results of the simulation; i.e., DETAILED, SUMMARY or AVERAGE. Examples of
each of these plus an example of the input data summary are shown in Tables
8 through 12. Additional examples are presented in Appendix C.
The input data summary (Table 8) consists essentially of most of the
data originally entered into the program with few modifications. The title,
basic design data and operating data are returned as entered with the excep-
tion of the temperature at which the inlet concentration was measured. If
no value was entered, the default value of 25 C is printed. Since blanks
are treated as if they were zeroes by the program, any blanks in the input
(except the title and result requests) will be printed as zeroes. It is
emphasized that not all of the fabric and dust property categories are
printed. Only those that pertain to (1) the manner in which K2 is to be
treated by the program and (2) the type of drag model to be used are
71
-------
TABLE 8. EXAMPLE OF INPUT DATA SUMMARY
su*"*Anr OF INPUT OAT* FOR BAGHOUSE ANALYSIS
••»••••«*•*••••••••••••••»•••»•*•••**••••••••••t********************************
CONTINUOUS/*? ESTIMATED/AC ENTERED/DETAILED RESULTS/
BASIC DESIGN DATA
NUMBER OF COMPARTMENTS 12
COMPARTMENT CLEANING TIMf J.fl
(OFF LINE T1»»E1
CLEAMKG CYCLE TIMfc J6.0
CUNTINUOUSLr CLEANED SYSTEM
REVE"S£ FLOn VELOCITY 0.0
OPERATING DATA
AvtRAGt PACE VELOCITY 0.9000
GAS TEMPERATURE 100.
INLET OUST CONCENTRATION s.oo
MEASURED AT 25.
FABRIC AND DUST PROPERTIES
SPECIFIC RESISTANCE. «2 ESTIMATED FROM
MASS MEDIAN DIAMETER 9.0
STANDARD DEVIATION 3.00
PARTICLE DENSITY 2.000
BULK DENSITY 1.000
EFFECTIVE RESIDUAL DRAG. SE ISO.
-EASUREO AT 25.
RESIDUAL LOADING, «R SO.O
MINUTES
"INUfES
DEGREES CENTIGRADE
G/Mi
DEGREES CENTIGRADE
MICRONS
G/CMJ
G/C»S
DEGREES CENTIGRADE
C/P2
SPECIAL PROGRAM INSTRUCTIONS
«A> NUMBER OF CYCLES MODELED 20
ACCURACY LEVEL 0
TYPE OF RESULTS REQUESTED DETAILED /
FRACTIONAL AREA CLEANED. AC 0.50
72
-------
TABLE 9. EXAMPLE OF CALCULATED VALUE PRINTOUT
CALCULATED VALUfS
0')ST CO«"CE*fWATION
TO OPERATING TE
u OUST t»"t PBpPfHIIES CU""«ECnO FOB C»5
C»KE SESIST4NCE. *i 1.66
DR»C, Sf «»7.
.», AC <1.^0
ff-t I'«C»tME»«' 0.75
cosst*^r »• o.o
73
-------
TABLE 10. EXAMPLE OF POINT-BY-POINT DATA PRINTOUT FOR DETAILED RESULTS SPECIFICATION ONLY
BA6*ORAG* *REA 1 AREA 2 •••*
1 9,I3E»02 7.10E*02 7.99f*02
2 9.2IE02 7.)2E*02 .19E»02 Uni
3 9.42E*02 7.HE«02 .»7E«OZ
• 9,$kE»02 7.73E*Q2 ,55E»02
S 9.*9E*02 7.93E*02 .72E*02
6 1.06E»20 t.OOE*56 ,OOE«20
7 5.36E»02 B.17E*02 .«7E*02
B S.72E*02 B.34E*02 .79£t02
9 *.05E*02 B.31Et02 7.07E*02
10 *.34E«02 8.67E*02 7.33E*02
11 6.blE*02 a.B3E*02 7.5»E»02
12 8.99E»02 6,BbE*02 7.7BE*02
BAG-FLO** AREA 1 AREA 2 OBAC
1 6.22E-OJ 1,06E*00 9.39E-01
j ' " eiWE^oT "r.i3T»«ir""»n'7E«or " " —
7.97E-01 9.96E»OI B.97E-01
7.BSE-01 9.71E-01 8.78E-01
7,7«E-01 9.47E«OI 8.61E-01
0.0 0.0 7.51E-18
1.40E«00 9.19E-01 1,16E*00
1.31E«00 9.00E-OI l.HEtOe
1,2«E*00 8.B2E-01 l.06E»00
~" 11 1.14E+06~"e!5"OEiOl ^!«3lJ61
12 B.35E-01 1.09E+00 9.64E-01
fi~zy5,i 6Ctp«~»5ir;»" OELQ«
BAG 1 BAG 2 BAG 3 BAG 4
T« 21.00 24.00 27.00 30.00
CAKE 2.T64*E"»02 " 2Yf704l*02~"2;2rWf»02 2.175«»02
•BAG 0.79B9E*03 O.B1BSE*03 0.6372E*0) 0.8SSOE*0]
ttBAG 0.9J95E+00 0.9170E*00 O.B966E«00 O.B779E*00
BAG 11 BAG 12 BAG
T« is.oo la.oo
_ iME U»««.|e»02 l._95*lt«02
MAG 0.75»*E»OS 0.77BTE*Ol
OBAG 0.9926E*00 0.9««5E*00
.ts: Drag in N-min/m3
. -Flow in m/min
Time in min
DELP in N/m2
. .Concentration in g/m3
Fabric loadings in g/m2
...Weight dumped is g/m2 of area per compartment
.4000 CONCENTRATION* .1091E-01 HEIGHT DUMPED" 127.*
BAG S BAG 6 BAG 7 BAG 8 BAG 9 BAG 10
33.00 -0.00 3.00 6.00 9.00 12.00
2.4742E*02 2.SOS5E*02 l.J599E*02 1.4B77E*02 1.61096*02 l.7296E*02
0.8721E*03 0.1000E+21 0.647SE*03 0.6T88E»03 O.T070E»OJ 0.7326E*03
O.B607E*00 0.7506E-17 0.11S9E*01 O.U06t*OI O.I062E*01 0,102SE*01
-------
TABLE 11. EXAMPLE OF PRINTOUT RESULTS FOR DETAILED OR SUMMARY DATA REQUESTS
SUMMARY TABLE J CONTINUOUS/** ESTIMATED/AC ENTERED/DETAILED RESULTS/
PRESSURE FRACTIONAL
TIME DROP PENETRATION INDIVIDUAL COMPARTMENT FLOnS (M/MIN)
(MINI (N/M2) COHP.l COHP.2 CQMP.l COMP.4 COMP.5
0.01 675. S.U91E-03 1.0416 0.9915 0.9519 0.920H 0.8916
0.75 669. b.OiUE-03 1.0116 0.9877 0.9512 0.9201 0.8910
1.50 728. 0.91IE-03 1.1092 • 1.061« 1.0250 0.9922 0.9635
*.24 7«a, 3.65)E>01 1,1095 1.061; 1.025- 0.9924 0.9636
i.OO 751. 2.73IE-03 1.1058 1.0617 1.0246 0.9926 0.9645
-------
TABLE 12. EXAMPLE OF DATA PRINTOUT WHEN DETAILED, SUMMARY OR
AVERAGE RESULTS ARE REQUESTED
F0|
Ik.00 M1NU1E8 UBt"ATIUN,
NUMBER k
AVERAGE PENETRATION*
AVENICE P»E»»URE DROP*
AVERAGE STBTEN FLO»*
MAXIMUM PENETRATION*
S.0*1-0)
711.1*
0.9000 M/NIN
8.191-01
750.'1 N/M2
Ik.00 MINUTES OPERATION. C^CL€
AVERAGE PENETRATION*
AVEMAGE PMES8URE DROP*
AVERAGE SYSTEM FV.OM*
PRESSURE DROP*
S.06F-OJ
71J.3U M/MJ
0.9000 M/MlN
8.49t-01
750.6? N/Mi
)k.OO HINUTtS OPEMATION, CYCLE NUMBER
AVERAGE PENETRATION*
•V6RAGE PNESIURE DROP*
MENACE miEM FLOW*
MAXIMUM PENETRATION*
MAXIMUM PRESSURfc DROP*
S.ObE-OJ
715.29 N/M2
I).4000 M/MJN
8.49E-03
750.kO N/M2
-------
returned. In Table 8 only particle size and density properties of the
inlet dust are printed since K2 will be estimated within the program. If
K2 is to be corrected for size properties, the size parameters for both the
inlet and reference dusts will be printed along with the temperature and
velocity for the reference K2 measurement. If sufficient data are entered
for the nonlinear model, SD and R as well as the temperatures of measure-
K K
ment will also be printed. Under special program instructions, the frac-
tional area cleaned, a , is printed only when it is available for entry to
the program (as assumed for the example case of Table 8). After all data
inputs have been printed, those values which have been calculated within
the program and/or corrected for temperature are printed (Table 9).
The input data, actual or calculated, will be printed as shown in
Tables 8 and 9 even when errors exist. This enables the model user to
compare the results with the original or intended data input when error
messages are generated by the program. Examples of the types of error
messages that may result from errors in the input data are presented in
Appendix C, Table 21.
If no errors exist in the input data, the filtration simulation will be
performed and one of three types of tabular results will be printed. When
DETAII^D results are requested, data similar to that shown in Tables 10,
11 and 12 will be returned to the user. If SUMMARY is specified, data
in the formats shown in Tables 11 (less compartment flows) and 12 will be
printed. When AVERAGE is selected, only the average and maximum operating
conditions shown in Table 12 will be returned to the user.
77
-------
A complete description of all parameters for each discrete area in each
compartment for each time interval is shown in Table 10 as part of the
DETAILED output. Results are printed in the order in which they are calcu-
lated. The first two blocks of data are the drag values and face velocities
corresponding to the two areas (in this case a * 0.50) on each bag (or
compartment), respectively. The units for each parameter are also shown
in Table 10. Note that compartment 6 is currently off-line as indicated
by a very high drag (102®) and zero flows.
The next line of information is the simulation time (T), the total
system pressure loss (DELP), total system flow velocity (DELQ), system out-
let concentration and, finally, the amount of dust removed from a compartment
during cleaning. This weight is expressed as grams of dust removed per unit
of cloth area in a single compartment. For example, for the case shown, if
the total filtration area for a compartment were 100 m2 then 100 m2 x 127.69/m2
or 12,760 grams of dust would have been removed from the bags in a single
compartment.
The average values for fabric loading, drag and flow (velocity) through
each compartment are summarized in the last block of data.
Total system pressure loss and penetration are presented as functions of
time in the summary table (Table 11). Individual compartment flows for up to
five compartments also appear in the summary table. These data correspond ex-
actly to those which will be plotted if a graphical output is requested. Since
more than five curves on the individual flow versus time graph would produce a
very crowded figure, data for only five or less compartments are plotted.
78
-------
In Table 12 Is shown an example of the format by which system pressure
loss, penetration and flow averaged over an entire filtering and cleaning
cycle are printed. The maximum penetration and pressure loss experienced
during a cycle are also output. The time specifications preceding the cycle
number is the total cleaning and filtering cycle time.
79
-------
6.0 GUIDELINE SENSITIVITY TESTS
Several guideline tables and graphs have been prepared so that the model
user can make preliminary approximations of filter system performance based
upon estimates of the principal design and operating parameters. The above
approach allows the model user to determine the relative importance and the
range of credible values for the major system variables before carrying out
any extensive computer modeling. For example, given the situation that the
fly ash concentration and size properties may vary appreciably for a specific
combustion process, or the size properties have not been determined at a high
level of accuracy, it is advantageous to define the impact of this variability
on filter system performance by means of the guideline tables and graphs.
This preliminary step will usually indicate when the data inputs are incon-
sistent with normal filter function or incompatible with the modeling process.
Tests were performed to determine the effect of either variability or
errors in the assigned operating parameters on system performance and to iden-
tify those operating parameters that have little or no effect on the filtration
process. Based upon preliminary tests, average face velocity (V), fractional
area cleaned (a ), limiting pressure (P.), inlet concentration (C.) and the
C L X
specific resistance coefficient (K£) were found to produce the greatest impact
on performance. Performance was defined by the three indices, average
pressure loss, average penetration and cleaning frequency. Those parameters
that play minor roles in determining system performance are the number of
80
-------
compartments, compartment cleaning time and the reverse flow velocity during
cleaning. The above variables and five additional parameters were assigned
constant values (see Table 13) so that the effects of changes in the major
variables could be ascertained. The numerical values shown in Table 13 (with
the exception of reverse flow velocity) are typical or average values asso-
ciated with the filtration of coal fly ash. Although K2 was not varied for
the bulk of the testing, the effect of K2 variations on pressure drop closely
paralleled the effect of changes in inlet dust concentration. This effect is
not unexpected since dust cake resistance is linearly related to both K2
and C..
A summary sampling of sensitivity tests showing the interrelationships
among the more important variables involved in the filter system operation
are indicated in Table 14. For example, the first two data groupings
(1 and 2) indicate how average pressure drop, penetration and time between
cleaning might vary due to differences or errors in estimating the fractional
area cleaned, a , for two different systems. As a result of variations in
velocity and cleaning frequency, the test range for a (0.1 to 1.0) has a
decidedly different impact on both average pressure drop and penetration.
Further reference to the tabulated data confirms the observation that the
absolute effect of changing one variable depends strongly upon the magnitude
of the other system variables. In some cases, one might conclude that varia-
tions in any one data input have little effect on system performance based
upon resistance and emission criteria. However, when the time between cleaning
increases from 6.6 to 672 minutes for a test velocity range of 0.3 to 0.91 m/min,
data group 6, the frequency of fabric cleaning is increased nearly 20 times.
81
-------
TABLE 13. SYSTEM OPERATING PARAMETERS HELD CONSTANT
FOR SENSITIVITY ANALYSIS
Parameter Constant value
Number of compartments 10
Cleaning cycle time 30 mln
Compartment cleaning time 3 min
Reverse flow velocity 0 m/min
Gas temperature 150 C
Effective drag, S£
at 25°C 400 N-min/m3
at 150°C 528 N-min/m3
Specific resistance coefficient, K2
at 25°C, 0.61 m/min 1.0 N-min/g-m
at 150°C, 0.61 m/min 1.32 N-min/g-m
Residual fabric loading, W_ 50 g/m2
n
82
-------
TABLE 14. DATA SAMPLING FROM SENSITIVITY TESTS
^ Average
group " parameter drop
(N/m*)
1 K2 - 1.0 V - 1.22 Continuous C. - 6.87 (ac - 0.1
1
-------
The Impact of errors (or variation) in K£ on filter performance is demon-
strated in data groups 7 and 8. A factor of 3 increase in K2 produces only
minor changes in average pressure loss for limiting pressure systems (group 7)
but results in a significant change in penetration. On the other hand, the
effects are reversed for continuously cleaned systems (group 8).
The effect of K2 on pressure loss may be approximated in some cases by
examining the effect of inlet concentration. Data groups 10 and 11 show the
results of tests in which K£ and C. were varied simultaneously, but with their
product held constant. Test data indicate that changes or errors in K2 will
produce changes in pressure loss roughly the same as those which would be
experienced if C. were changed in proportion to the change in K2.
Figure 9 shows the effect of variations in face velocity, V, and
limiting pressure, P , on the average system pressure loss, P, when all other
L»
system variables are held constant. The lowest curve shown describes the
resistance path for a continuously cleaned system. Once an average velocity
is selected, the average resistance can never be lower than that corresponding
to the velocity intercept with that curve; i.e., no pressure-velocity coordinate
can exist in the shaded region. Thus, if one selects a limiting pressure loss
of 1000 N/m2 as the point where cleaning is to be initiated and concurrently
selects a face velocity of 1.5 m/min, the system automatically reverts to a
continuously cleaned system with an average operating pressure drop of
2500 N/m2, far exceeding the limiting pressure. On the other hand, given a
face velocity of 1.0 m/min and a limiting pressure of 2000 N/m2, the velocity-
pressure intersection occurring above the shaded zone indicates that the system
will operate according to the selected V and P values and on an intermittently
Lt
cleaned basis.
84
-------
Kta|.ON »!•/§-«
4000
OC«0.4
•-PL or tsoo
& « PL OF tOOO
0 • »t OF IBOO
ID * \ OF 1000
V • 'L or BOO
O • COHTINUOU*
x 9000
III
1000
1000
0.5 1.0
PACE VELOCITY, V.
1.5
Figure 9. Effect of face velocity_(V) and limiting pressure loss (P^) on
average pressure loss (P).
85
-------
The curves shown In Figure 9 represent the average pressure levels for
systems in which the fractional area cleaned, a , is 0.4. However, a is a
function not only of fabric loading but other factors as well such that the
loading distributions will differ for various combinations of velocity and
limiting pressure. Depending on the type and intensity of cleaning, some
systems may never achieve a cleaning level of 40 percent while others may
exceed this value. Refer to Equations 18 and 20.
Numerous plots of average pressure loss, penetration and time between
cleanings have been prepared for different combinations of inlet concentration
and cleaned area fraction. Due to the large number of plots generated from
the sensitivity testing, only a few summary results are given In this report.
Complete tabulations, however, are provided in a related report in which sen-
sitivity tests were the main object of study. Reference 11 also furnishes a
detailed interpretation of the sensitivity tests and their applications.
The cleaning frequency (defined by the time between cleaning) and the
dust penetration associated with the systems described in Figure 9 are
presented in Figures 10 and 11, respectively.
The time between cleanings, which increases as the limiting pressure drop
is allowed to increase and decreases as the face velocity Increases is con-
sistent with expected filter system behavior, Figure 10. Similarly, Figure 11
shows that dust penetration Increases rapidly with increasing face velocity,
regardless of the assigned limiting pressure with one very important exception.
During continuous cleaning, the effect of Increased face velocity is first to
provide additional surface cover within the time frame of the cleaning cycle.
This effect overrides the reentrainment effect of increased filtration velocity
until, for the systems described by Figure 11, the adverse velocity effect
dictates a rise in penetration.
86
-------
1000
900
800
TOO
600
I
3
900
400
300
eoo
100
KI • 1.0 N »l»/g-
C • «.»T fl/m*
a«* 0.4
0 V«0.«l
V v« o.ti
O v«i.«t
1000 - 2000
LIMITIN* MfSWftC OHOf>,P>tlN/«s
SOOO
Figure 10. Relationship between time between cleanings, limiting pressure
loss and face velocity.
87
-------
10
1.0
oc
i
UJ
O
or
0.1
0.01
I 111
K2"I.ON mlB/«-m
Oc sO.4
Cj s6.87«/m9
O« CONTINUOUS
V« PL OF 500
Q • PL OF 1000
A«PL OF IBOO
^ « PL OF 20OO
• • PL OF 2800
j i I i
O.t 0.2 0.5 1.0 2.0
AVERAGE FACE VELOCITY, M/fflin
5.0
10
Figure 11. Effect of face velocity and limiting pressure drop on average
penetration.
88
-------
REFERENCES
1. 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.
2. Dennis, R., R. W. Cass, and R. R. Hall. Dust Dislodgement From Woven
Fabrics Versus Filter Performance. J Air Pollut Control Assoc. 48 No. 1.
47:32, 1978.
3. Dennis, R. and H. A. Klennn. Modeling Coal Fly Ash Filtration With Glass
Fabrics. Third Symposium on Fabric Filters for Particulate Collection.
Report No. EPA-600/7-78-087. June 1978. p. 13-40.
4. Dennis, R. and H. A. Kelmm. A Model for Coal Fly Ash Filtration (Presented
at the 71st Annual Meeting of the Air Pollution Control Association.
Houston, Texas. June 2-30, 1978.)
5. 40 CFR 60, Appendix A Methods 1 through 5 and 9
6. Billings, C. E. and J. E. Wilder. Handbook of Fabric Filter Technology.
Volume I, Fabric Filter Systems Study. Environmental Protection Agency.
Publication Number APTD-0690 (NTIS No. PB-200-648). December 1970. 649 p.
7. Rudnlck, S. N. and M. W. First. Specific Resistance (K2) of Filter Dust
Cakes: Comparison of Theory and Experiments. Third Symposium on Fabric
Filters for Particulate Collection. Report No. EPA-600/7-78-087. June 1978.
p. 251-288.
8. Happel, J. Viscous Flow in Multiparticle Systems: Slow Motion of Fluids
Relative to Beds of Spherical Particles. AIChE J. 4:197-201, 1958.
9. Dennis, R., and J. E. Wilder. Fabric Filter Cleaning Studies. U.S. Environ-
mental Protection Agency, Control Systems Laboratory, Research Triangle Park,
North Carolina. EPA-650/2-75-009 (NTIS No. PB-240-372/3G1). January 1975.
10. Robinson, J. W., R. E. Harrington, and P. W. Spaite, "A New Method for Analysis
of Multicompartment Fabric Filtration, "Atmos Environ. _!: 499-508, (1967).
11. Dennis, R. and H. A. Klemm. Fabric Filter Model Sensitivity Analysis .
U.S. Environmental Protection Agency, Industrial Environmental Research
Laboratory, Research Triangle Park, North Carolina. EPA-600
In Press. (1979)
89
-------
APPENDIX A
SUBROUTINE STABLE - DETERMINATION OF STEADY STATE
A description of the three criteria used to determine when the simulation
has reached a given level of convergence is presented below. Average pressure
loss is the test variable which is traced throughout the simulation. In the
course of evaluating the convergence tests discussed here, it was noted that
average penetration and filtration cycle time (time between cleaning) also
converged when average pressure loss converged.
Check #1
Check #1 involves the determination of a least squares fit for the
regression line through the points indicated in Figure 12, i.e., natural
logarithm of the slope of the P versus T curve, versus average time, t.
Here, average time refers to the average of the absolute times bracketing the
time interval over which the slope is measured. Thus, the slope can be
represented as:
where m - slope of P versus T
A - Intercept of the regression line of Figure 12
B - slope of the regression line of Figure 12
90
-------
The actual average pressure drop at infinite time, P^, can be found by inte-
grating Equation 36 with respect to absolute test time, T:
A+Bt ,_
e dt
Since t = T •
dt » dT
and P >
constant
eA=Bt dt = I eA+Bt
D
(37)
P(tavg) -
-*„
p-
=
_eBtavg
By integrating between the limits ti = 0 and t2 = tavg, the following
general equation results:
P = I (eA+Btavg _ eA) (38>
D
1
which reduces to P^ = — (-e*) when tav~ approaches infinity.
An estimate of how close the actual value of average pressure drop, P,
is to the predicted final value, P^; i.e., the fractional error can be computed
from Equation 39:
(39)
The current convergence criterion used in the Subroutine STABLE for this
check is 0.01. This limit is decreased to a value of 0.00333 when an accuracy
code of 1 is selected in place of the less stringent code of 0.
Check #2
Again referring to Figure 13, a second check involves a linear regression
for the last four data points; i.e., the results of the most recent four oper-
ating cycles. The slope of the regression line is an indication of how average
pressure drop is changing with time. An estimate of the change in pressure
from cycle to cycle is:
E -
m-AT
(40)
91
-------
where E - ratio of the estimated change in pressue drop over a cycle to
the actual pressure drop
m - elope of the regression line of Figure 13
AT - complete cycle time
P - average pressure drop of the most recent complete cycle
When E is computed to be less than some predetermined limit (currently 0.005)
the system is considered to be at equilibrium.
Check //3
If the average pressure drop oscillates about the steady state value as
shown in Figure 14, and convergence is not indicated by either Checks #1 or #2
the system may actually be at or very close to equilibrium. Check #3 determines
whether or not the magnitude of the oscillations is decreasing with time.
Successive changes in average pressure drop are compared without regard to
sign once oscillation has begun. If the absolute difference between PS and ¥7
is less than that between the preceding values Pj and Pg, Figure 14, the
system is considered to be at steady state.
92
-------
1
5
AVEftME TIME,t
Figure 12. Method of fitting data to exponential curve
for Check #1.
-s O O-
IB.
TIME,!
Figure 13. Example of linear regression lines used in
Check #2.
O
TIME.T '• •' '•
Figure 14. Example of oscillating pressure drop used
in Check #3.
93
-------
APPENDIX B
BAGHOUSE SIMULATION PROGRAM LISTING
A listing of the baghouse simulation program card deck is presented in
Table 15. The listing includes all the Job Control Language required to run
the program on an IBM 370 under OS/VS2 using the FORTRAN Gl V2.0 Compiler.
Plotting routines are compatible with the CalComp Basic Software Package for
Pen Plotters and General Subroutines Package. A list of all the variables
and arrays used in the program is presented in Tables 16 and 17.
94
-------
TABLE 15. PROGRAM LISTING
//HKI6I4I8 JOH (0170,072,OE3K),'KLEMMi,CLA38«*.TINfc«6
//• BAGHOU8E PROGRAM IBM 170 MllH CALCOMP PLOTTER
//• 1976 CCA TECHNOLOGY ROGEfr STERN • DOUG COOPER
//• 8AGHOUSE 3IMUIATION PROGRAM- IBM 570- ZETA PLOTTER
//• |«77 CCA TECHNOLUGV DIVISION HAN8 KLEMM- RICHARD DENNIS
//• REVISED OCT. 78 - GCA/TECHNULOGV DIVISION • HANS KLEMM/ RICHARD DENNIS
//SIMULA EXEC FOBTGJCC,ACCT»C08T,PAhM.60«'8I2t«68K'
/'FORT.SY8IN DO •
k«*************»**************»**t**»>t**(«gOOOoOOO
C 00000010
C STEP • SIMULA BAGHOUSE MUOEL STEP • 1 00000020
C MAIN PROGRAM FOR BAGHOUSE SIMULATION PROGRAM 00000030
C IF ERRORS EXIST IN THE INPUT DATA J«l 00000040
C 00000050
CALL DESINE
CALL OPERAT
CALL SNDATA
CALL USER(I)
CALL CHECK)(I)
CALL SETUP
CALL CHECK2U)
CALL OUTFIL
CALL PLOTINU)
IF(I.EU.l) GO 10 10
CALL MODEL
id DO to N»IO,IS
END FILE N
10 REMIND N
END FILE fc
REWIND A
CALL EXIT
END
00000070
ooooooso
00000090
00000100
00000!10
00000120
00000130
00000140
OOOOOISO
00000160
00000170
OOOOOISO
00000190
00000200
00000210
00000220
00000230
00000240
(continued)
95
-------
TABLE 15 (continued)
SUBROUTINE CAKDR6(NDEL*VtL*CURA6) 00000250
C**••••••••*•••»•••*••••••••»•*••*•••*•***••«•*••*••••••••»*******»***«*00000260
C
c SUBROUTINE OF BAGHOUSE a/77/MK-RD CCA TECHNOLOGY
c SUBROUTINE of RAGHUUSE to/76 HAK/PO CCA/TECH oiv
C -CALCULATES CAKE UWAG
C-|K2*SPECIFIC CAKE RESISTANCE OF CAKE At 0.61 M/MIM, N-MIN/G-M
C-NOEL«TOTAL FABRIC LOADING ON AN AREA Of FABRIC. G/M2
C-WRBHESIDUAL FABRIC LOADING ON AN AREA OF FABRIC, 6/M2
C*NSTARs CONSTANT CHARACTERISTIC OF DUST AND FABRIC* G/M2
C-ZKZEHO" INITIAL SLOPE OF DKAG VS. LOADING CURVE* N-MIN/G-M
C-VEL«VELOCITY,M/MIN
C-CDRAOCAKE DRAG, 3, N-MIN/MJ
C
ZK2V»ZK2*SORT ( VEL*3. 281/2. >
IF(K3TAR.GT.I.E-20) CO TO 10
C-LINEAR MODEL
CORA6«ZK2V*(NDEL-MR)
GO TO 20
10
lF(tXPO.LT.-30.) EXPO»-JO.
c -NON-LI NEAR MODEL
COHAG"/K2V«WPKIMe»(ZKR-ZK2V)*X8TAH«(i..EXP(EXPO))
20 RETURN
END
00000270
DIVISION 00000200
00000290
00000100
oooooito
OOOOOJ20
00000110
00000140
000001SO
00000160
00000170
00000180
»000001«0
00000400
oooooito
00000420
00000410
00000440
00000450
00000460
00000470
00000480
00000490
00000500
OOOOOStO
00000520
SUBROUTINE PENET(CZERO,WEIGHT,VEL»«R»PEN) 00000510
C*»********t••••*«•**A**************************************************00000540
C 000005SO
C SUBROUTINE OF BAGHOUSt 4/77/HAK-RD GCA TtCHNOLOGY DIVISION 00000560
C-CALCULATfcS TOTAL PFNETRATION 00000570
C-CZ£RU«1NLET CONCENTRATION* t/MJ 00000580
C-KEIGHT-TUTAL FABRIC LOADING ON AN AREA OF FABRIC* G/M2 00000590
C-VCL»VELOCITY, M/MIN 00000600
C-NR-RESIDUAL FABRIC LOADING ON AN AREA OF FABRIC* G/M2 00000610
C-PEN»PENETRATIUN 00000620
00000610
C3«0.0005
A«400.
IF(VEL.GT.J.E-O) A«0.416/(VEt*1.281)**4*0.094
IF(VEL.LT.l.E-9) VtL«0.0
»F»l,5E-7
IFCVEL.GT.l.t-9) XF»1.5E-7«EXP(12.7«(1.
EXPO«(M£IGHT-«R)*A
PEN'0.0
IF(EXPO.LT.40.) PtN»(0.1-XF)«EXP(-fXPO)
PEN«PENtXF»CS/CZERO
RETURN
END
00000650
00000660
00000670
00000680
00000690
00000700
00000710
00000720
00000710
00000740
00000750
000007*0
(continued)
96
-------
TABLE 15 (continued)
SUBROUTINE MODEL 00000770
C*****************************************ft******«***********«****»«****00000780
C 000007*0
C SUBROUTINE OF BAGHUUSE 12/1/RwS-DC GCA TECHNOLOGY DIVISION 00000600
C SUBROUTINE "F BAGHOUSE «/77/HAK-RD GC* TECHNOLOGY DIVISION 00000810
C SUBROUTINE OF BAGHOUSE 10/78 HAK/RD CCA/TECH DIV
C-MA1N DRIVER SUBPROGRAM
C-ALL T'8 ARE TIMES,MIN
C-ALL w's ARE CAKE LOADINGS,G/M*
C-ALL S'S ARE DRAGS,N-MIN/M3
C-ALL P'S ARE PENETHATIUNS
C-ALL c's ARE CONCENTRATIONS
C-A BAG 19 A COMPARTMENT
C-Z«2"SPECIF1C CAKE RESISTANCE OF CAKE AT 0.61 M/MIN, N-MIN/G-M
C-WR«RE8IOUAL FABBIC LOADING UN AN AREA OF FABRIC* G/M2
C-NSTARs CONSTANT CHARACTERISTIC OP DUST AND FABRIC* G/M2
C-IKH* iNiriALSLOPt OF THE DRAG VERSUS LOADING CURVE
C-SZEHU*RESIDUAL DRAG, N-MIN/MJ
C-TEMPKtGAS TEMPERATURE,DEGRESS KELVIN
C-ACAKE«CAKED AREA,THAT PORTION OF A BAG WHICH is NOT CLEANED
C-2K2MU»VISCOSITr CORRECTION FOR SPECIFIC CAKE RESISTANCE
C-NBNUMBER OF COMPARTMENTS OR BAGS
C-T*CLEANING CYCLE HM£,MIN
C-NT»TOTAL NUMBER Of CYCLES TO BE MODELED
C-M«NUMBER OF TIME INCREMENTS PER BAG
C-SMALO"AV£RAGE SYSTEM VELOCITY,IF OPERATING AT CONSTANT TOTAL FLOW, M/MOOOOI020
C-CZERO«INLET CONCENTRATION,G/MJ
C-LOIAG»PHINT DIAGNOSTICS
C-TLAG«TIME PERIOD FOR »HICH ALL BAGS ARE ON LINE AFTER
C-CYCLE
00000820
00000830
00000800
000008SO
00000860
00000870
00000880
00000890
00000900
00000910
00000920
00000930
00000910
00000950
00000960
00000970
00000980
00000990
OOOOiOOO
00001010
00001030
00001040
ENTIHE CLEANING000010SO
00001060
C-DP8TOP«PHE33URE DROP AT WHICH CLEANING IS INITIATED, NXM2
C-M8TART*INITIAL LOADING ON ALL BAGS AT TIME * ZERO
C-VRFLO»HEVER3E AIR VELOCITY FOR ONE BAG* M/MIN
C-SE«EFFECTIVE CAKF DHAG, N-MJN/M3
C
00001070
00001080
00001090
00001100
00001110
C***I*******************************************************************00001120
COMMON/D£SIGN/N,T,TCLEAN,TLAG,VRFLO»DPSTOP»FREQ»AMPLIT 00001130
COMMON/OPDATA/SMALQ,TEMPK,CZERQ,TCZERO 00001140
CUMMON/FABOUS/ZK2«SE,MR,SR>ZKK,HSTAR 00001ISO
CQMMUN/EXTERN/NT,M,WSTART,ACLEAN ooooiuo
COMMON/DIAG/LUIAG,PRDIAG,PLDIAG 00001170
CUMMnN/CALC/D*LT,NAREA,IAHEA 00001180
COMMON/ACURAC/JCOOE 00001190
COMMON/TITLE/HEAD 00001200
COMMON/MODELU/PAVR,TCONT,DTLAST,PENTOT»PAVTOT,DPAVG,OAVG,TLA9T. 00001210
* TOSUM,PNMAX,DELP,DPMAX,THEF,IFBAG,NrLAG,JFLAG,LUPCNT 00001220
COMMUN/DEVKf /INPUT, OUTPUT 00001230
INTEGER OUTPUT 00001240
LOGICAL LDIAG,PLDIAG,PRDIAG 00001250
REAL** HEAD(8) 000012*0
(continued)
97
-------
TABLE 15 (continued)
DIMENSION IOUM<10)«PDP(3l,«»D«C3J»PT(3J»WKI)fPO(l,9>
DIMENSION TIME (SO). OLOtIMC30)»CAKE(30J, 88*6(10 J,0i*6( JO J
DIMENSION NOUO,JO),3(10,30).QAREA(10),PnO)
LOGIC*!. LCONP,LOIAG
DATA ORAG,BA&1,BA62/'AKEA',«SBAGI»IH8AG'/
WHITE (OUTPUT, £20) NEAO
220 FOHMAT{1M1//T20,BO{'»')//
* T?0, 'RESULTS OF 8ACHOUSE ANALYSIS' //T20»BO( '•' )//T20,8A8///)
GO TO 3
2 CZEHO»CZEROE
CALL HESTHT
C-1NIMALIZE DATA
3 CALL IN1TAL
AREABl./IARE*
IREPTBFLOAT(N)/10.»0.99
00 S UUIAREA
OAREA(I)«SMALO
00 S IBAC«1,N
OLOTIM{IBAG)«-2
TIME(IBAG)«-J
5 "D(1,IBAG)«KSTART
CZEROE'CZERO
TCHHH»0.0
TLAG«IFlX(TLAG/OfcLT*O.Sj*OELT
IF(TLAG.LT.O.Ol) TLAG«0,0
TMOD«TLAG*T
IF(DP3TOP.GT.O.)TMOO«l.E*aO
K3*0
DETERMINE ORAG THROUGH FABRIC
3F»B«8R
IFCWSTAR.LT.1.E-20) SFAB-SE
JLOOP"0
LOOP ON TIME
1 JLOOP»JLOOP»1
OELT«T/M/N
TTEST"AMUU(TC(JNT+O.OI,TMUD)-0,01
IF(TTEST.LT.O.OOS) TTE3T«0.0
IF(TTE8T.GT.(T-0.005).AND.TTfc3T.LT.(T»0.005)) TTEST«T
IFCTCONT.LT.l.E-'.OR.TTEST.lt.-O.Ol.OR.TTEST.GE.O.OI) GO TU 12
LOPCNTBLOPCNT+J
TOIHTCUNT-TLAST
QAVGN*(OAVG»Q8V8TM*OTLA8T)/2./TDSUM
PAVNOM*(PAVTnT»PENTUT*OTLAST)/2./TOSUM
OPAVGN«(DPAVC"DELP*OTLA3T)/2./TOSUM
TDSUMSO.O
TLAST»TCr)NT
OAV6*08V8TM*OTLA8T
PAVTOT»PENTOT*DTLAST
DP*VG»DELP*OTLAST
00001270
00001280
00001290
00001300
00001510
00001320
00001330
00001340
00001390
000013*0
00001370
00001380
00001390
00001400
00001410
00001420
00001430
00001440
00001490
00001460
00001470
00001480
00001490
00001900
00001910
00001920
00001910
00001940
00001990
00001960
00001970
00001980
00001990
00001600
00001610
00001620
00001630
00001640
00001690
00001660
00001670
00001680
00001690
00001700
00001710
00001720
00001730
00001740
00001790
00001760
(continued)
98
-------
TABLE 15 (continued)
C CHECK FOR A LIMITING PRESSURE SYSTEM FUR FORCED CONTINUOUS OPERATION 00001770
c CHECK FOR TIME BETWEEN CLEANING CYCLES EQUAL TO ZERO OOOOITBO
CONTST«TDIF-T 00001790
IFCDPSTOP.GT.O.S.ANO.LOPCNf.GT.S.AND. 00001 BOO
* CONTST.GT.-T/M/N.ANO.CONTS1.LE.(T/M/N+O.Om GO TO 2 OOOOlHtO
C-NR1TE AVERAGE PRESSURE DROP,FL"* AND PENETRATION UP TO TIME«TCUNT OOOOIS20
IMNFLAG.GT.O) WRITE(OUTPUT,330) TD1F,LOPCNT,PAVNO",DPAVGN,QAVGN, 00001830
* PNMAX,OPMAX 000018*0
IF(JFLAG.Erj.O) C«LL STABLE
-------
TABLE 15 (continued)
IF(S(l*REA,tB»G).LT.l.E*19> CO TU 17 00002270
CALL CAKDHGfi»D-0.01 00002010
21 IF(TTEST.GT.T) GU TO 19 00002420
C-TEST FOR AN UFF LINE BAG 00002*30
IF(TCONT.LT.1.E-9.AND.TIMEUBAG).LT.CT-TCLEAN-.001)) GO TO 19 00002440
1F'X,l!(lX,iPE9.2» 00002550
33r3TM«3SYSTM+l,/3BAG(IU»G) 00002560
IF«JLDTIM(iH»G).GT.TIMfc(IBAG).AND.TTE8T.LT.(T*0.005)) DELTT'0.01 00002570
20 CONTINUE 00002500
C-ENO OF BAG LOOP 2 00002590
C-CALCULATE SYSTEM DKAGrPRESSURE DROP ANU FLOw VELOCITY 00002600
3SYSTM«|,/SSYSTM 00002610
OELP«SMALQ«S3YSTM*N*V«FLOh«3SYSTM 00002620
OSVSTM«SMALO«VRFLOM/N 00002630
C-CORRECT INLET CONCENTRATION FOh REVERSE FLOH AIR 00002640
C2ERO»CZEHOE*(QSVSTM»VRFLON/N)/QSVSTM 00002650
IF(LOIAG.ANO.NFLAG.GI.O) WRITE(OUTPUT,SO) (DRAG,I,I»l,I AREA),8AG2 00002660
30 FORM»T(U,•HAG-FLOW"',lx,ll(3X,»U,IX,12)) 00002670
PENTOT'0.0 00002680
KDUMPBO.O 00002690
C*BAG LOOP 3 00002700
00 60 IBAG»1,N 00002710
IF(TTEST.GT.T) GO TO 26 00002720
OELTBQELTT 00002730
IF((TIME(IBAG)*T/M/N).GT.(T-TCLEAN))DELT»T-TCLEAN-TIME(IBAG) 00002740
26 NCOMPcQ.O 00002750
CAKfCIBAG)BO.O 00002760
(continued)
100
-------
TABLE 15 (continued)
C-ARfcA LOOP 2 00002770
OU 28 I*1«IAREA 00002780
OAHEA(I)»OJLP/S(I,1BAG) 00002790
C-DETERMINE PENETRATION 00002800
CALL PtNlT(CZtHn,»D»KD(I,IBAG)*AHEA 000028)0
27 PENTOT»PENTOT*PU)*AREA«OAHf A(I)/QSY3TM/N 00002840
28 MO(I,IBAG)«ND(ltIBAG)+MAREA 000028SO
C-END OF AREA LOOP 2 00002860
IJBAGCI8AG)«nELP/SBAGUBAi;) 00002870
C-OUIPUT INTERMEDIATE RESULTS 00002880
IF(LDIAG.AND.NFLAG.GT.O) «WI It (OUTPUT, 15) IBAG, (OARE A {I) , HI, 00002840
* IAREA),OBAG(1BAG) 00002900
IF(TTEST.CT.T) GO TO 60 00002910
\ttnLOTIM(I8AG),Lt.TlME(lBA&))GU In 60 00002920
C-CLEAN NAREA AREAS ON A BAG IF NECESSARY 00002930
HDUMPBO.O 00002940
00 16 II«I»NA»EA 00002990
MCOMpaO.O 00002960
C-AREA LOOP 1 00002970
DO i*> [»l,IAktA 00002980
IF(WD(I,IBAG).LT.HCOMP) GU TO 35 00002990
MCOMP*NDU,IHAG) 00001000
IFAREA.I 0000)010
J5 CONTINUE 0000)020
C-tND OF AREA LOOP J 000030)0
«OUMP»MDUMP»(WO(IFAR£A,IHAU)"«R)*AREA 00003040
16 «*0(lFAREA.IBAt)«WK 00003050
60 CONTINUE 0000)060
C-ENO OF BAG LUUP ) 0000)070
OELTBDELTT 0000)080
DPAVG«OPAVG»(OTLASItOELT)*OELP 0000)090
QAVGBQAVG«(OTLASTtDELT)*QSYSTM 00005100
PAVTOT«PAVTOT»PENTOI*(OtLTtOTLASI) 0000)110
PAVH«PAVR»PENTOT*(DtLT»OTLA31) 0000)120
TOSUM»TOSUM»DTLA3T 0000)1)0
OTIAST'OELT 0000)140
CONTOT«PENTOT«CZERO OOOOJ150
IF(PENTUT.GT.PNMAX) PNMAxaPENlQT 0000)160
IF(DELP.GT.OPMAX) OPMAX«OELP 0000)170
IF(NFLAG.EQ.O) GO TO 120 0000)180
IF(.NOT.PLOIAG.ANO..NOT.PRDIA6) GO TO 120 0000)190
K3»K)»i 0000)200
PT(K))«TCUNT-TREF 00003210
PDP(K3)*OELP 0000)220
POO(K3)«e3YSTM 0000)230
PP3(K))«PENTOT 00003240,
LMAX«MINO(5,N) 00003250
DO 100 L*1,L*AX 0000)260
(continued)
101
-------
TABLE 15 (continued)
100 PO(«3tL)«OBAC(L)
1MK3.1T.3) 60 TO 120
«J"0
c
c PUNCH PLOT
C
110 FORMAT(6Gt0.5)
WRITE (8,1 10) ((PT(K),PDP(K))»K«1«3)
DO US L"1,LMAX
IUNIT«L«9
JI5 HRITE(IUNIT.UO) ((PT(H),PQ(K,L))»K"1,3)
«H1TEU5.I10)(PT(K)»PPSI»,K»1,3)
120 IM.NOT.LUIAG) GO TO 290
IFCNFLAC.EO.O) 60 TO 290
C
C PRINT DIAGNOSTICS
C
«RITE
290 IF(IFMAG.NE.O) GO TQ 11
IF(OPSTOP.LT.I.E-9) GO TO 300
!F(TMQD.LT.|.E+19.AND.TTEST.Gr.(T+T/M/N)) TMOD"TCONT-T-r/M/N
IF(TTEST.LE.T.OR.DELP.LT.DPSTOP) GO TO 300
00003270
00003280
00003290
00003300
00003310
00003320
00003330
00003300
000033SO
00003360
00003370
00003380
00003390
00003800
00003410
00003020
00003030
00003440
00003450
00003460
00001470
00003480
00003490
00003500
00003510
00003520
00003530
00003500
00003550
000035*0
00003570
00003580
00003590
00003600
00003610
00003620
00003630
00003640
00003650
00003660
00003670
00003680
00003690
00003700
00003710
00003720
00003730
00003740
00003750
00003760
(continued)
102
-------
TABLE 15 (continued)
TMOD»TCONT
TCORR'0.0
100 UO TO I
C •END OF TIME UJOP
C
c FINISH HUNCHING
C
110 CONTINUE
IH,NOT.PLDI»G,»ND..NUT,PROI»G) 60 TO 430
«RITE<8,400) PT(3).POPT80,Of>F10.2,• N/M2'/
•T50,'AVERAGE SYSTEM FLU»«'.TeO,OPF10.«,' M/M1N«/
*T50,'MAXIMUM PENETRATION!«,T60,IP£
-------
TABLE 15 (continued)
8U8HUUTINE PinTIN(IfcMROR)
^^••••••••^•••••••••^•••••^•••••^••••••••••^^^•••••••"••*W»*w»
C
c SUHRUUIINE TO INITIALI/E PLOTTF.H u/ii/75/n«»s-oc
C SUHHIlllTlNfc (|F HAUHOllSE 4/77/HAK-HO GCA TECHNOLOGY
C SUBROUTINE OF RAGHOUSE 10/78 HAK/RO GCA/TECH OIV
C
Q ft ft A ftftftftftftftftftftftftftftftftft ftftftftftftftftftftftftftftftftftftftftftftftftftftftftft ftftftftftftftftftftftft
Cl)MMON/Dt9ir,N/N,T,TCLF.AN,lLAG,V«FLO»DP3TOP,FREQ,AMPLIT
CI)MMON/DIAt;/All>lAr.,PKDIAG,PLDIAG
CIJMMON/TITLF/MFAO
COMMON/DEVICE/ INPUT, OUTPUT
INTEGER OUTPUT
LOGICAL PROIAG,PLOIAG»ALDIAG
REAL'S HEADfH)
DATA AMP/Ui/
DATA XLENTM,Yl.FNTH/2*0./
IMAX«M1NO(N,S)
LOIAG«0
IF(PRDIAG)LOIAG»|0
IF(ALDIAG) inlAGaLDlAG+10
lF(PLOIAr,)LOI«GiLlHAGtl
IFCIEHROiJ.f.U.1) LDIAGSO
WRITE(8,t5)Lf'IAG,I*AX
15 FORMAT(I,»,I1>
HE AD (INPUT, 2on, ENDS26) XLF^TH, YLENIH
200 FO»MAT0.04-CHIT2
C-PMFSSUHE DHIJP VS TIMt
DO 20 IUNIT=8,10,2
20 NRITeCIiJNIT,2S} HEAD, YPOS1 ,CHIT1, AMP
25 FOHMAT(«A8,5X,2F5.2»A|)
NH I TEC 8, SO) YP082*CH!Tt,XLENTH,YLENTH
30 FitRMAT(
ft'PMtSSUHE VS TI^E U»APH',170,2F5.2.U"//
nT2«,»TIMfc C«M*UTfS)'/
HT23»»PRE8SUHt (N/M2)'/
»«StMI8EMli,T«>5,F6.2,T65,F6.2,T80,M«)
C-IMDIVIDUAL FLQM vs TIME
NRITEC10.50) YPOS2,CMlTl,YPOS1,CHIT2,XLENTH,yLENTH
SO M)RMAT(
ft'INOIVlDUAL FlOt» (»ATt GHAPH' , T70,2F5.2, ' 1 ' /
11 'COMPARTMENT • l',T70,2F5.2/
*T28,«TI*E (MJNUTFS)'/
*T23,'FLO* RATfc (M/MJN)*/
«'9rMISE*I«,T55,F6.2,T65»F6.2.T80,'l*)
(continued)
104
00004080
AA*******t00004090
0000410ft
00004110
DIVISION 00004120
00004130
00004140
*AAftftftAft**0000(JiSO
oono«i60
00004|70
00004180
OOQ04I9Q
00004200
00004210
00004220
00004230
00004240
00004250
00004260
00004270
00004280
00004290
00004300
00004310
00004320
00004330
00004340
OOOOUJSO
000043&0
00004370
00004380
00004390
00004400
00004401
00004402
00004403
00004404
00004405
00004406
00004410
00004420
00004430
00004440
00004450
00004460
00004470
000044HO
00004490
00004500
00004510
00004520
00004530
00004540
00004550
00004560
00004570
OOQ04580
-------
TABLE 15 (continued)
I«AX»»INO(N,S) 000045*0
IF(IM«x.CO.!) GO TO 75 00004600
Oil 60 ls?,IMAX 00004610
IUNIt*If9 00004620
YPIISI«YPI)3S-(I-1 )*(n.04»CHTT2) 0000462S
60 *HlTt(!UNIT,70> 1,YPU3I,CM!T2 00004630
70 FOBM»T("COMPAHTMfNT » » , It,T70.2FS.2) 00004640
75 H«ITt(l5,2?) HF.An,VPOSI,CHlTI,AHP 00004650
C-PENETHATK1N V3 II^K 00004660
«WITE(lb,Hn) YPnS2,CH!Tl,NrH>YLENTH 00004670
DO FDRMATt 00004600
11PENETRATION V3 TI«E GRAPH',T70.2F5.2,U1 / 00004640
»/ OOQ04700
iT28,»TI««E (MINiiTESJ1/ 00004710
*T2?.|PEN£TPATIHN«,T70,ll.f-«» l.O1/ 00004720
*'LOG-8t««ll,TS5,fh.2,T65>Fi>.2»T80»ll') 00004750
nu too iuNn«p,io.? 00004/40
100 H»ITE(IUNIT, |tf)> 00004750
w«ITE(Ji,t10) 00004760
110 FO»MAT{T8,"0.0«,T1«,'0.0«) 00004770
120 HtlURN 00004790
END 00004790
(continued)
105
-------
TABLE IS (continued)
SUBROUTINE OISINE
00004000
i•****•*«*••**•••****•***•************«*00004010
00004020
SUBROUTINE OF BAGHOU8E 10/78 HAK/RD GCA/TECH DIV 00004010
READ AND PRINT HEADINGS AND DESIGN DATA 00004040
00004050
00004070
00004000
00004090
00004900
00004910
00004920
00004910
00004940
00004950
00004960
00004970
00004900
00004990
00005000
00005010
00005020
00005010
00005040
00005050
00005060
00005070
00005000
510 FORMATU1,I X,F5.I,lX,F*.l,lX,F5.1,lX,F4.0,lX,F6.4rtX,Fl.l,lX,F4.2)00005090
HEAD(0)
INTEGER OUTPUT
COMMON/OESIGN/N,T,TCLIAN,TLAG,VRFLO»DP9TOP,FHEO,AMPLIT
COMMON/DEVICE/INPUT,OUTPUT
COMMON/TITLE/HEAD
OUTPUT«6
INPUT«5
READ(INPUT,500) HEAD
REAO(INPUT,510) N,TCLEAN,T,TLAG,DPSTOP,VRFLO,FREQ,AMPLIT
C-DEFAULT FOR COMPARTMENT CLEANING ( OFF LINE ) TIME
C- SOI OF CLEANING CYCLE TIME/NUMBER OF COMPARTMENTS
IFUCLEAN.LT.0.05J TCLEAN«T/N«0.5
MRITE(OUTPUT,600) HEAD
WRITE(OUTPUT,610) N,TCLEAN,T
IF(TLAG.GT.O.OS) M»ITE(OUTPUT,620} TLAG
IMDP3TOP.GT.0.05) KRITE(OUTPUT»6lO) DPSTOP
IF(TLAG.LT.O.OS.AND.DPSTOP.LI.O.OS) WRITE(OUTPUT,610)
»«RITECOUIPUT,650) VRFLO
IF(FREO.GT.O.OS) *RITE(UUTPUT,660) FREQ
IF(AMPLIT.GT.J.E"5) NRITE(OUTPUT,670) AMPLIT
«RITE(OUTPUT,6«0)
FOHMAT(SA0)
600 FORMAT(1HI,T20,KO(**I)//T20,ISUMMARV OF INPUT DATA FOR ',
00005100
00005110
00005120
00005110
00005140
00005150
00005160
00005170
! '8AGHOU3E ANAL»SIS»//T20,00(•*•J//T20,0*0)
610 FOHMATC//T20,'BASIC DESIGN DATA'/
1 T2S,'NUMBER OF COMPARTMENTS',T55,I1X
S T25,"COMPARTMENT CLEANING IIME',T55.F6.1,T70,'MINUTES'/
0 T27,'(OFF LINE TIME)V
2 T25,'CLEANING CYCLE TIME',155,F6.1,T70,'MINUTES'
5 )
620 FORMAT(T25,>TIME BETWEEN CLEANING CYCLES',T55,F6.1,T70,'MINUTES') 00005100
610 FOHMAT(T25,'LIMITING PRESSURE DROP*,T5S,FS.O,T70,'N/M2') 00005190
640 FORMAT(T25,* CONTINUOUSLY CLEANED SYSTEM')
650 FORMAT(T25,*REVERSE FLOW VELOCITY',T55,F7.4,T70,'M/MIN')
660 FORMATCT25,'SHAKING FREQUENCY',T55,F4,t,T70»'CYCLES/SEC')
670 FQHMAT(T25,'SHAKING AMPLITUDE',T55,F5.2,T70,'CM')
600 FORMAT(JX)
RETURN
END
00005200
00005210
00005220
00005210
00005240
00005250
00005260
(continued)
106
-------
TABLE 15 (continued)
SUBROUTINE OPERAT
00005270
SUBROUTINE OF BAGHUUSE 10/78 HAK/RO GCA/TECM DIV
READ AND PRINT OPERATING DATA AND CORRECT TEMPERATURES TO
DECREES KELVIN
00005290
00005300
00005310
00005320
00005330
tA*******************************«*********00005340
INTEGER OUTPUT 00001350
COMMUN/OPOATA/SMALQ»TEMPK,CZENO,TCZERO 00005360
COMMON/DEVICE/INPUT.OUTPUT 00005370
READUNPUT.500) SMALO.TEMPK,CZ£RO»TCZ£RO 00005300
IMTCZERO.LT.l.E-5) TCZERU«25.01 00005390
KRITE(OUTPUT.hOO) SMALB,TEMPK.CZERO.TCZERO 00005400
TEMPK»TEMPK»271. 00005410
TCZER»«TCZERO»273. 00005420
500 FORMAT(F6.4,|)t,F«.0,1X,F5,2.lX»FfliO) 00005430
600 FORMAT(T20,'OPERATING DATA'/ 00005440
1 T25,'AVERAGE FACE VELOCI»V'.T55.F7.4.T70,«M/MIN'/ 00005450
T25.'GAS TEMPERATURE',T55,F5.0,T70.'DEGREES CENTIGRADE'/ 00005460
IK,'INLET DUST CONCENTRATION',T55.F6.2.T70,'G/M3'/ 00005470
tJO,'MEASURED AT',T55.FS.0,170,'DEGREES CENTIGRADE*/ 00005460
) 00005490
RETURN 00005500
e*0 oooossio
3
a
5
5
(continued)
107
-------
TABLE 15 (continued)
SUBROUTINE IKDAT* 222J2JS
C•••••«•*•••••••••••••••*•••••••••••****••••••••••••••*••*•••**•••••••••OOOOMJO
c
c
c
c
c
00005540
SUBROUTINE OF 8AGHOUSE 10/78 HAK/RD CCA/TECH OIV 00005550
READ AND PRINT DUST AND FABRIC PROPERTIES AND CORRECT TEMPERATURFS00005560
TO DECREES KELVIN 00005570
00005560
••••••*••••••••••»•*•**••••«*••••*••••••••*»*****«o»***«**«***«*o**00005590
INTEGER OUTPUT 00005600
REAL MMD1,MMD2 00005610
COMMON/FABOUS/ZK2.SE.»»R»SR.Z»«R.*»9TAR 00005620
COMMON/DEVICE/INPUT*OUTPUT 00005630
COMMON/K2EST/TZK2,VZK2,MMDl,8Gl,*MD2,SG2,RHOP»RMOBLK 00005680
COMMON/MUCORR/TSE«TSR*TZKR oooo56*o
*R1TE(OUTPUT,600) 00005660
600 FORMAT(T20,'FAHRIC AND DUST PROPERTIES1/} 00005670
REAIHINPUT,500) ZK2, TZK2,VZ*2,MMOt,SGI,MMD2,3G2,RMOP»RHOBLK 00005680
READ(JNPUT,510) SE»TSE.URfSR»TSR/ZKR,TZKR.WSTAR 00005690
500 FORMAT(F5.?,JX,F«.0,lX,F7.«,U,FJ.l,lX,F5.2,lX,FJ.t,lX,F5.2,JX, 00005700
I F5,J,1X,F5.J) • 00005HO
S10 FORMAT(F«.0,lX,FO.O,lX,F5.l,lX,F«.0,lX,F«.0,JX,F5.2,lX,Fa.O,lX, 00005720
* F5.1) 00005730
C IF K2 MAS NOT ENTERED ASSUME IT IS TU BE CALCULATED 00005740
IF(ZK2.GT.|.E-S) GO TO 20 00005750
IF(8E.LT.I.E-5) SE*350. 00005760
IF(XR.LT.l.r-S) NR>SO. OOOOS770
IF(TSE.LT.l.e-5) T3E«25.01 00005760
NRITE(OUTf»UT*620I MMD2>SG2,RHOP»RHUBLK 00005790
620 FQRMAT(T25,*SPECIHC RESISTANCE, K2 ESTIMATED FROM'/ 00005800
i T3o«'MAss MEDIAN DIAMETEH"»T55»F«.i,T70»•MICRONS'/ OOOOSBIO
2 TJO,'STANDARD DEVIATION',155,F«.2,/ 00005620
3 TJO,"PARTICLE DENSITY•/TS5»F6.J,T70,"G/CM31/ 00005630
a T30,*BULK DfeN«ITV',T55,F6.J,T70»'G/CM3»/ 00005840
5 ) 00005850
rZK2»25.01 00005860
VZK2B0.61 00005870
GO TU 30 00005880
C K2 HAS ENTERED 00005890
20 IFCTIK2.1T.1.) TZK2«25.0t 00005900
IF(VZK2,LT.l.E-5) VZK2*0.6l 00005910
MRITE(OUTPUTf6|0) Z«2,TZK2,V^K2 OOOOS920
610 FORMATCT25,'SPECIFIC RESISTANCE, *2',T55,F6.2»T70,'N-MIN/G-M'/ 000059JO
1 TJO,'MEASURED AT'. 00005940
2 T55,F5.0,T70,'DEGREES CENTIGRADE'/ OOOOS950
J T55,F7.4,T70,«M/MIN« 00005960
a ] 00005970
C IF NO SIZE PROPERTIES FOR INLET DUST HERE ENTERED ASSUME NO 00005980
C CORRECTIONS ARE TO BE MADE 00005990
IF(MMD2.LT.l.b-5) GO TO 30 00006000
*RITE(OUTPUT,630) MMD1,SG1,MMD2,SG2 00006010
(continued)
108
-------
TABLE 15 (continued)
650 FURMAT(T«S,'MMD1',T55,F«.1,T70,'MICRONS',T85,'-STANDARD DEVIATION 0000*020
•',T10S,F4,2/ 000060)0
* T30.'CORRECTED TO '. 00006040
* Ta5.'MMD2',T55,F'ICRON8',T85,'-STANDARD DEVIATION',TI05 00006050
-,F«.2/ 00006060
* ) 00006070
50 CONTINUE 00006080
IF(TSE.LT,l.E-5) TSE«25.01 000060*0
IFU8H.LT.1.E-5) TSH»25.01 00006100
IF(TZKR.Lf.l.E-S) TZKR»2S,01 00006110
WRITE(OUTPUT,6«0) SE,TSE,KK 00006120
640 FORHAT(TJ5,'EFFECTIVE HE3IDUAL QKAfi, 8E',T5S,H>,0,T70,'N-MIN/143'/ 00006130
I TJO,'MEASURED AT•,T55,FS.O,T70,'DEGREES CENTIGRADE'/ 00006140
2 T25,'RESIDUAL LOADING, wR',T55,F6.t,T70,»G/M2' 00006150
5 ) 00006160
C IF 3R ANO KH WERE NOT ENTERED ASSUME LINEAR DRAG MQDfl 00006170
lF(SR.LT.I.E-5.ANO.ZKH.LT.1.t-5) CO TO 40 000061*0
*R1TE(OUTPUT,650) SR,TSK,ZKR,TZKR 00006190
650 FORMAT(T25,»RESIDUAL DRAG, SH',T55,F5.0,T70,'N-MIN/MJ'/ 00006200
I TJO,'MEASURED AT',T55.F5.0,T70,'DEGREES CENTIGRADE'/ 00006210
2T2S,* INITIAL SLOPE, KR',T55,F6.2,T70,'N-MJN/C-M'/ 00006220
S T30,'MEASURfO AT1,T55,F5.0,r70,'DEGREES CENTIGRADE') 00006230
40 NRlTE(nurPUT,660) 000062«0
660 FORMAT(/) 00006250
TSE»TSEt275. 00006260
TZKR«TZKR»273. 00006270
TZK2«TZK2»273. 00006280
TSR>TSRt27J. 00006290
RETURN 00006300
END 00006310
(continued)
109
-------
TABLE 15 (continued)
SUBROUTINE USERUERHOR) 00006120
£*••***••**•••*•••**•••**••*•••****•**•****•***•************•****•**••••00006S30
C 00006340
C SUBROUTINE OF BAGHUUSE 10/70 MAK/RD CC»/TtCH DIV 00006350
C READ SPECIAL PROGRAM INSTRUCTIONS 00006360
C**AL01AG IS T/F FUN ALL RESULTS 00006170
C*«PRDIAG IS T/F FQR SUMMARY TABLE RESULTS ONLY 00006340
C««PLOIAG is FOR PLOTTING oooo639o
C 00006400
INTEGER OUTPUT
LOGICAL ALDIAG,PLDIAG,PKDIAG
REAL*8 DETAIL,SUM1,SUM2.BLANKS,AVG1,AVG2,DATYPE
COMMON/ACURAC/JCUDE
COMMON/DEVICE/INPUT,OUTPUT
COMMON/DIAC/ALDIAU,PRDIAG,PLDIAG
COMMUN/EXTERN/NT,M,WSTART,ACLEAN
DATA AVG1.AVG2/' AVERAGE', 'AVERAGE. '/
DATA OETAIL,SUMl,SUM2,PLOTER,BLANKa,BLANKS/
• 'DETAILED',' SUMMARY','SUMMARY ','PLOT',' ',' '/
READ(INPur,500) NT,JCUDE.DATYPE,PLTYPE,ACLEAN
WRITE(OUTPUT,600) NT,JCOPE,DATYPE.PLTYPE
600 Fi)RMAT(T20,'SPECIAL PROGRAM INSTRUCTIONS1/
I T2S,'MAX NUMBER UF CYCLES MODELED',T55,I3/
t T25,'ACCURACY LEVEL1,T55,I2/
2 T25,'TYPE OF RESULTS REQUESTED*>TS5,A8,' / ',A4/
3 )
IF(ACLEAN.GT.I.E"5) *HITE(I)UTPUT,6IO) ACLEAN
MO FORMAT(T*5,'FRACTIONAL AREA CLEANED, AC',T55,F4.2)
IERROR'0
!>00 FORMAT(I3,lx,I2,lX,A8,lX,Aa,lX,F3.2)
C SET FLAGS FOR LEVEL OF DETAIL ON OUTPUT AND CHECK INPUT FOR ERRORS
ALDIAG'.FALSE.
PLDIAG".FALSE.
PROIAG'.FALSE.
C CHECK INPUT DATA FOR ERRORS
IFCDATVPE.EQ.BLANKS) GO TO 10
IFIDATYPE.EO.DETAIL) 60 TO 20
IF(OATYPE.EO.SUM1.0R,DATYPt.E0.8UM2) GO TO 30
IF(DATYPE.NF.AVG1.AND.DATYPE.NE.AVG^)
GO TO 10
20 ALDIAO.THUE.
30 PRDIAG«.TRUF.
10 IF(PLTVPE.EQ.HLANK
-------
TABLE 15 (continued)
SUBROUTINE CHFCKl(I) 00006000
C*****«««••*«•**•••••••**•*•*•••*••*•*•*•*•*••*••••*•••*•••*••**•«••••••00006690
C 00006900
C SUBROUTINE OF BAGHOUSE 10/78 MAK/RO GC»/TfcCM OIV 00006910
C 00006920
C***********************************************************************00006930
REAL MNOl,MM02
CiJMMON/K2E3T/TZK2,VZK2,MMDl,3Gl,MMD2,SG2,RHOP,RM08llC
COMMON/OESTGN/N,T,TCLEAN,TLAG»VRFtO.DP8TOP,FHEOiAMPUT
COMMON/OiVICt/INPlJT.J
CnMMON/OPDATA/3MALQ,l£MP*,C2fcKO,TCZEf»U
CQMMaN/FABOUS/ZK2iSt»NR,SR.ZKH,w3TAR
*HITE(J,500)
IF(I.EO.l) WftlTE(J«600)
600 FORMAT(/T20, 'HUGAL REQUEST f-QB TYPt OF RESULTS')
lF(N.LE.30.ANO.N.Cr.O) CO 'TO 10
«HITE(J,510)
!•!
10 IF((N*TCLEAN).U.T)GO TO 20
*RITfc(J,520)
I«l
10 IF(TCLFAN.LT.T)f,() TU 30
WRlTt(J,5JO)
I»l
SO IMT/N/M.GT.0.01 ) GU TO 3S
NRIT6(JrS60)
lat
J5 IF(SMALO.GF.O.J.ANO.SMALO.I.t.i.O) GU TO «0
FUHM*T(/T20, 'AVEHAfit FACE VtLOCITY OUT OF RANGtt 0.3 TO 3.0')
flO If (TEMPK.GT.2T1.5) GO TO 50
NRlTtU.540)
1«1
50 !F(FHEO.GT.|.f-5.ANU.AMPLIT.GT,l.t-i) GO TO 60
IF(FREU.lT.l.E'S.ANO.AMPLIl.LT.l.E-S) GO TO 60
WHITE(J,570)
!*1
60 IF(M.Nt.«2) GU TO 70
I»l
WHITE(J*S80)
70 IF(TLAG.GT.|.f-5,ANO.OP8TOP.GT.l.t-5) GO TO 75
GO TO 100
75 WRITEU,6lO)
I»I
100 IF(ZK2,LT,l.E-5) 00 TO 130
IF(MMOl.GT.|.t-5,ANO.SCl.GT.l,E-5.AND.MMD2.GT.J.t-5.ANU,3G2.CT.
* l.E-5) GO TO 110
IF(MMOl.LT.l.E*5.AN0.9Gl.LT.t.E-S.AND.MM02.LT.l.E»5.*NO.SG2.LT.
00006940
00006950
00006960
00006970
00006980
00006990
00007000
00007010
00007020
00007030
00007040
00007050
00007060
00007070
00007080
00007090
00007100
00007110
00007120
00007130
00007140
00007150
00007160
00007170
00007160
00007190
00007200
00007210
00007220
00007230
00007240
00007250
00007260
00007270
00007280
00007290
00007300
00007310
00007320
00007330
00007340
00007350
00007360
00007370
(continued)
111
-------
TABLE 15 (continued)
• l.fc-S) so TO l»o
I«l
NRITE(J,630)
630 FURMAM/T20, 'PARTICLE SIZE DATA FUR *2 ARE INCOMPLETE')
tlO IF(MMDt.GE.2..AND.MMDI.LE.SO.) GO TO 120
!•!
MHITE(J*640)
feflO FORMATC/T20,'MASS MtOlAN DIAMtTER OF MEASUREMENT OUT OF RANGE'*
* • i TO so MICRONS')
120 1M3G1.GE.2..AND.SG1.LE.U.) GO TU ISO
1»1
•»RIT£(J,650)
650 FORMATC/T20, 'STANDARD DEVIATION UF MEASUREMENT OUT OF RANGE1.
* ' 2 TO 4»)
130 IFCMMD2.GE.2..AND.MMD2.LE.50.) GO TO 110
!•)
•»RITE(J,660)
660 FURMATC/T20.'MASS MfcDIAN DIAMETER i)F DUST OUT OF RANGE ' i
• ' 2 TU bo MICRONS')
140 IF(SG2.GE.2.,AND,SG2.LE.4,» CO TO ISO
1*1
«RITE(J,67fl)
670 FOHMATC/T20, 'STANDARD DtVIAIION OF DUST OUT OF RANGE'*
• ' 2 TU 4" )
ISO IF(MM01.GT.l.t-S.AND.MMD2.&T.!.E**.AND.SGI.GT.i.E*S.AND.SG2.GT.
* l.E-S) GU TU 180
IF(RHOBLK.LT.RMOP) GO TO 160
660 FOHMAT(/T20,'BULK DENSITY CANNOT EXCEED DISCRETE PARTICLE DENSHV
* ')
160 IF(RHOBLK.GT.1.E»S. AND. RHOP.GT. l.E-S) GO TO 160
I«l
WRITE (J»690)
690 FOHM*T(/T20,IBIILN CJW DISCRETE DENSITY MISSING')
ISO CONTINUE
610 FORM*T(/T20, 'BOTH TIMED AND PRESSURE CONTROLLED CLEANINGS '/
• 'SPECIFIED • ONLY ONE IS VALID')
SOO FURMATC • i',T20, 'DIAGNOSTIC MESSAGES')
S10 FORMAT(//T20,'THE NUMBER OF COMPARTMENTS MUST NOT EXCEED 10')
S20 F(JHMAT(/T20,'THE NUMBER Uf COMPARTMENTS TIMES THE COMPART',
i 'MENT CLEANING TIME MUST HE LESS THAN THE CLEANING CYCLE TIME')
b30 FORMAT(/f20,'THE COMPARTMENT CLEANING TlMfc MUST BE LESSMx,
1 'THAN THE TOTAL CYCLE TIME')
540 FORM*T(/T20,'A GAS TEMPERATURE HAS NOT BEEN ENTERED*)
S60 FORMATl/T2o,'TtME INCREMENT TOO SMALL, IE. « 0.01 MINUTES')
S70 FUHMAT(/T20,' INVALID FREOUt NCV OR AMPLITUDE FOR SHAKER')
FORMAT (/r20»' INVALID ACCURACY CODE')
RETURN
END
00007380
00007390
00007400
00007410
00007420
00007410
00007440
000074SO
00007460
00007470
00007480
00007490
00007SOO
00007510
00007520
00007530
00007540
00007550
00007560
00007570
00007580
00007590
00007600
00007610
00007620
00007630
00007640
00007650
00007660
00007670
000076SO
00007690
00007700
00007710
00007720
00007730
00007740
00007750
00007760
00007770
000077*0
00007790
00007800
00007610
00007620
00007630
00007640
00007850
00007860
00007870
(continued)
112
-------
TABLE 15 (continued)
StTllP 00007680
»00007690
00007900
SUHKOUTINb, Q^ HAGHOUSt 10/78 HAK/RO GCA/fECH D1V 00007910
CORRECT FUR TEMPERATURE AND VISCOSITY 000079*0
CALCULATE AND CORHECT K2 FUR SIZE PROPERTIES 00007910
CALCULATE SYSTEM CONSTANT N* 00007940
DETERMINE NUMBER OF AREAS THAt A BAG IS TO BE. BROKEN UP INTO 00007950
IF I«l AN ERROR EXISTS 00007960
00007970
REAL MM01.MM02 00007990
COMMON/K2EST/TZK2,VZK2,MMDi.SGI,MMU2,3G2,RHOPrRHOBLK 00006000
COMMON/MUCORR/TSE>TSR>TZKR 00008010
COMMON/OPOATA/SMALO* TEMPH,CZERO,TCZERO 00006020
COMMON/CALC/DtLT>NAREA,IAREA 00008030
COMM(iN/FAHDUS/7K2»St>NR,SH>ZKR,NSTAR 00006040
COMMON/EXTE,RN/NTfM,WSTAKT,ACLfcAN 00006050
CUMMUN/DE8lGN/NfT,TCLtAN,TLAG.VRFLO,DP8TOP»FREO»AMPLn 00006060
C- VISCOSITY CORRECTIONS 00006070
V1SC(TEMP)«1.«6E-3«TEMP««1.5/{T|MP*110.) 00006060
C- DELT 00006090
DELT«T/M/N 00008100
TTESTl«N*TCLfcAN«N*TCLEAN*I.t-4 00006110
TTEST2«N*TCLEAN«N*TCLEAN«l.E-« 00008120
IF(T.6E.?TEST2.ANO.T.LE.TTE8T1) TClEAN»TCL£AN-O.OOl5 00006130
ISKIP»0 00008140
IFCACLEAN.GT.l.E-S) ISKIPBl 00006150
C CALCULATE K? IF NECESSARY 00006160
C IF KZ«0 CALCULATE IT 00006170
C IF K2»0 AND MMD2«0 DO NOT CALCULATE 00006160
C IF K2>0 AND MMD2>0 CCiHMfCT IT FOR MMDIS1GMAG 00008190
IF(ZK2.CT.1.E-5.AND.MMD2.LT,I.E-S) 60 Tl> 30 00006200
IF(ZK2.GT.l.E-S.AND.MMn2.GF,l.E-S) 60 TO 20 00006210
C CALCULATE «2 00006220
SULID«RHQHLK/«HOP 00006230
R«(J.+2.«SOLlD*«(S./3.))/(l.-'«.S«8ULID»«(l./3.)»a.5*SOLlD«*(5./5. 00006240
* )-3.*SOLlD«*21 00006250
S02'36.*iO.**(2.30**2)/MMD2**2 000062*0
C PARTICLE SIZE IN MICRONS.DENS1TY IN G/CC. VISCOSITY IN CENTIPUlSt 00006270
ZK2»16,6««0,01B*R*S02/6./RHOP 00008280
60 TO 30 00006290
C CORRECT FOR MMO AND SIGMAG 00008300
20 SOB2»36.»10.«*t2.30U*(ALUGIO(SGl»**2)/MMDi**2 00006310
SOF2*36.*10.**(2.>Oa*(ALOG10(SG2))**2)/MM02**2 00006320
ZK2*Z«2«SOF2/SOB2 000083JO
C CORRECT TO VELOCITY «F 0.61 M/M1N 00008340
30 |K2«ZK2*8URT(0,61/VZK2) 00008150
C CORRECT FOR TEMPETURE 00008360
ZK2»ZK2*VISC(TEMPK)/VI8C(TZK2) 00006370
(continued)
113
-------
TABLE 15 (continued)
3E«SE* V ISC (TE"PK)/VISCOSE) 00008380
3R»3H*VI3C(TK-PK)/VISCZKR*vl3CnEMPK)/VISC(TZKR> 00008000
CZEWI«CZEHO«TtZERH/TEMPH 00008010
C CORRECT SE TO HH OOOOH420
SE*SE*NR*ZK2 00008430
C CALCULATE MSTART 00008440
C INITIAL LOADING ON EACH COMPARTMENT AT TIME ZERO 00008450
IF(DP8TOP.LT,1.) CO TO 40 00008460
MSTARTa(DPSTUP*SE*SMALO>/(ZK2*30R1(SMALQ/0.6l))/SMALQ*MR 00008470
GO TO SO 00008480
ENTRY RECALC * 00008490
40 «START«166.*(SE*SR}/(ZKR>ZK2) 00008SSO
IK13KIP.NE.1) CALL CLEAN(O.OrDUMMV) 00008560
C* TOTAL NUMBER OF AREAS ON A BAG (IAREA) AND 00008S70
C" NUMBER TO BE CLEANED (NARfcA) 00008S80
ERR'0.01 00008590
7 i«i,/ACLEAN»O.j+0.2 00008600
J»> 00008610
IF(tHR.GT.0.06)00 TO 9 00008620
DO 8 1*1,10 000086JO
DO 8 J«1»I 00008640
ATtST«FLaAT(J)/FLOAT(I) 00008650
IF <«TEST.LE.(ACLtAN4.ERH).AMD.ATEST.GE.(ACLEAN-ERR))GO TO 9 00008660
8 CONTINUE 00008670
ERR«ERR»0.01 00008680
CO TO 7 00008690
9 NAREA*J 00008700
I»»«*«l 00008710
«TURN 00008720
00008730
(continued)
114
-------
TABLE 15 (continued)
SUBROUTINE CHECK2U)
00006740
C 00006760
C SUBROUTINE M HAGHOUSl 10/78 HAK/RD GCA/TECH OIV 00006770
C CHECK CALCULATED VALUES FOk ERRORS AND PROPER RANGE 00006760
C 00008790
C**** ••••*•*••»••••• ••*•••*•••••••••••**••*••••••*•••••*•••• •*•*•* «*»*»* 00006600
COMMON/DEVICE/ INPUT, OUTPUT 00006610
coMMON/FABDus/ZK2rSE,MR*SR.ZKR,H3TAR ooooeeao
COMMON/EX TERN/NT, M,*ST AS T.ACLt AN ooooeeso
COMMUN/OPDATA/SMALQ* TEMPK,CZERQ» TCZERO 00006640
INTEGER OUTPUT 000068SO
IF(H3TAR,LT.l.E-5) 60 TO OS 00006660
IF(SE.GE.SR) Cn TO 46 00006670
!•! 00006660
NRITE(OUTPUT,200) 00006690
200 FORMAT(/T20,'fFFfcCTIVfc DRAG. 3E . IS LESS THAN RESIDUAL* 3H') 00006900
«5 IF(ZKR.LT.l.t-5.ANO.SR.LT.l.E-5) GO TO SO 00006910
1st 00006920
MRITE(OUTPUT,620) 00006930
620 FORMATC/T20, 'INCOMPLETE DATA FOR NON-LINEAR DRAG MODEL') 000069*0
46 IF(SR.GT.l.E-S) GO TO 47 00008950
1*1 00006960
NRITE(OUTf>UT*630) 00006970
6)0 FORMAT(/T20* 'RESIDUAL DRAG SR , IS MISSING') 00006960
«7 IF(ZKR.GT.i.E-S) GO TO SO 00006990
I«l 00009000
NRITECOU1PUT*640) 00009010
640 FOHMAT(/TiO, 'INITIAL SLOPE , KH , IS MISSING') 00009020
SO IF(ACLEAN.6T.l.E-i.*ND.ACLE*N.LEt».) GO TO 60 000090SO
]«1 00009040
«RITE(IIUTPUT»600) 00009050
600 FORMAT(/T20* 'FRACTIONAL AREA CLEANED OUT OF RANGE, 0 TO I ' > 00009060
60 TE8TK2"ZK2*296**i.S'408.*(TEMPK«ilO}/TEMPK**l.S 00009070
IF(TESTK2.GE.0.2S.AND.TESTKa.LE.iO.) GO TO 75 00009080
WRITEtUUTPUTrtolO) 00009090
610 FORMAT(/T20, 'K2 IS UUT OF KANGE»0.2S TO I0>) 00009100
IB| 00009110
75 IF(I.EO.O) GO TO 80 00009J20
HRITECOUTPUT.2IO) 000091JO
RETURN 00009140
60 HRITECOUTPUT.220) 00009150
210 FORMATC///.T20, 'THE PROGRAM HAS BEEN TERMINATED BECAUSE OF ' 00009160
1 , 'ERRORS IN THE INPUT DATA') 00009170
220 FORMAT!///, T20, 'THERE ARE NO ERRORS IN THE INPUT DATA') 00009180
RETURN 00009190
END 00009200
(continued)
115
-------
TABLE 15 (continued)
SUBROUTINE OUTFIL
SUBROUTINE OF BAGHUUSE 10/76 HAK/RO 6CA/TECH DIV
PRINT CALCULATED AND CORRECTED VALUES
CUMMQN/EXTERN/NT,M,MSTAKT,ALLEAN
COMMON/OPDATA/SMALO,TfMPK,CZERO»TC/tRO
CflMMON/OEVICF/JNPUTfJ
CUMMliN/CALC/DHT.NAkEA.IAkkA
CUMMUN/FABOU3/ZK2,SE,*R,SKtZKK,NSTAH
NRITECJrIOO) CZERU,ZK2»SE
IF(ZKR.GT.I.E-S) *RITE(J,610) ZKR
IF(3R,GT.1.E»5) WRITE(J,620) SH
HHITEIJ,630>
WRITECJ,600) ACLEAN.DELT,HSTAk
100 FOHMAT('l',T20,'CALCULATtD VALUES' ,////,
I T20, ' INLtT DUST CONCENTRATION' , 155, F6. 2, T70. 'G/M3' »SK/
• T20, 'CONNECTED TO OPERATING TEMPERATURE'//
2 T20, 'FABRIC AND DUST CAKE PROPERTIES CORRECTED FOR GAS *
3 . ' VlSCUSirv ,//,
4 T25. 'SPECIFIC CAKE RfcSISTANCEt *a' , T55, F6.8, T70, 'N-MIN/G-M • /
b 125, 'EFFECTIVE DRAG, SE ' . TS5« FS.O.T70, 'N-MIN/M31 )
600 FORMAT(T20, 'FRACTIONAL AREA CLEANED, AC ' , TSS,F4.2//
9 T20,'T1M£ INCREMENT', T55,F5,2»T70, 'MINUTES1//
• T20, 'SYSTEM CONSTANT **' , T55.F5.1 ,T70, «G/M2«//
• )
610 FOHMAT(T25>, 'INITIAL SLOPE, KR' , TS5, F6.2, T70 , ' N-M1 N/G-M' )
620 FQRMAT(T25,'HESIDUAL DRAG, S« ' , TS5,F5.0, T70, ' N-MIN/MJ ' )
630 FOMMAT(/)
RETURN
END
00009210
00009210
00009240
00009250
OOOOV260
00009260
00009290
00009500
00009*10
00009S20
00009330
00009340
000093SO
00009360
00009370
00009380
00009390
00009400
00009410
00009420
00009430
00009440
00009450
00009460
00009470
00009480
00009490
00009500
00009510
00009520
00009530
SUBROUTINE CLEAN(NTOTAL,AtLNJ 00009540
C ****••»****•*•*•*•»***»»**»*****«*»•*•**••*•*•****•*•*«*****•** ********000095%0
C 00009560
C SUOHUUTINE OF HAGMUU3E 10/78 HAK/nD GCA/TECH DIV 00009570
C CALCULATES FRACTIONAL AREA CLEANED, AC , FOR SHAKER AND COLLAPSt 00009580
C SYSTEMS 00009590
C NOTE I "TOTAL AND ACLN ARE NOT USED BY THE PROGRAM 00009600
C 00009610
€•••••*•••**• **••**•••*»*••»«*•*« ••*•**•»••»•*••»• ********* ********* ****00009620
CUMMON/OPUAT4/SMALU, lEMPK.CZfcHO, TCZERO 00009630
CUMMUN/FADDUS/ZK2,SE,MN,SR,ZHM,N8TAR 00009640
COMMON/OESIGN/N,T,TCI.EAN,TLAG,VRFLO,DP8TOP,FRkQ«AMPtIT 00009650
COMMON/EX TERN/NT, M,WST ART. ACLE AN 00009660
IFCMTOTAL.GT.l.) GO TO 30 00009670
IF(OPSTOP.LT.l.) GO TO 20 00009680
HP«(DPSTUP«SE*SMALB)/(ZK2*80R1(SMALU/0.61))/SMALO*NR 00009690
NPRIME«NP+T*CZERU*SMALQ/2. 00009700
ACLtAN»1.5lfc- 8*WPHIME**a.Se 00009710
IFCFREQ.GT. 1 .E-5) ACLEAN«2.23E-l2*(FREO**2*AMPLIT*WPRlMe)**2.52 00009720
GO TO 25 00009730
20 ACLEAN«0.006*(CZERO*SMALO*(TtTLAG))**0.716 00009740
IF(FHEO.G1.1.E»5) ACLt AN«o.9E-0* (Fk£0**2* AMPL I T*CZERO*SMALO* 00009750
* (T*TLAG))**0.716 00009760
?5 IF(ACLEAN.GT.l.) ACLEANB1. 00009770
IF(ACLEAN.LT.O.l) ACLEANiO.l 00009780
RETURN 00009790
30 ACLN«l.blE-8»»TOTAL**2,52 00009800
IFCFREO.GT.l.t-S) ACLNB2.23E*12*(FREO**2*AMPLIT*NTOTAL)**2.52 00009810
IFCACLN.Gf.l.) ACLNBl. 00009820
IF(ACLN.LT.O.I) ACLNBO.l 00009830
RETURN 00009840
_ END _ _____ _ 00009850
(continued")
116
-------
TABLE 15 (continued)
SUBROUTINE STAHLE(DRUP*T1ME,JCODE»LCODE) 00009060
C•••••••••••»••••*•*•••••••••••••«•*•***•••»••••••*•••••••••••**••*•••••00009870
C 00009880
C SUBROUTINE Of BAGHOUSE 10/78 HAK/RD CCA/TECH DIV 00009690
C CHICKS FUR CONVERGENCE 1U S!E*OY STATE 00009900
C 00009910
COMMON/3TABLD/TI»T2,TOOI,T002,DP1»DP2,P01,P02»8IGN1,I,N,NL»NCHG 000099)0
DIMENSION TC50),DP(50),DPDPC20) 00009940
REAL NCHK) 00009950
REAL LIM|,LIM2 00009960
LCOUE'O 00009970
I«I*1 00009980
T(I)«TIMt 00009990
DPUJ«DROP OOOtOOOO
TI«Tl»T|Mt 00010010
T2»Tl«TIMt*TIMF 00010020
DP1»UP1»OHOP 00010030
DP2«OP2tl)HOP«TIME OOOlOOaO
N«N»l 00010050
iFU.Nt.nco 10 a* oootoo60
C SET LIMITS OF CONVERGENCE 00010070
LIMlBO.Ol 00010080
LIM2BO.OO! 00010090
LIMlBLIMI/FLOATCJCUDfc) 00010100
LIM2iLIM2/FLOAT(JCUOE) 00010110
Ji GO TU UO 00010120
as TAVGs(T(l)+T(I-l)>/2. 00010110
OELf>P*(DPCI)-DP(I-m/
-------
TABLE 15 (continued)
SUBROUTINE INITAL 00010*60
00010*80
SUBROUTINE (IF HAGHUUSE 10/78 HAK/RU OCA/TECH OIV 00010*90
THIS ROUTINE INITIALIZES VARIABLES USED BY MODEL AND STABLE 00010600
00010610
»********************»********»**************************************00010620
COMMON/MODELD/PAVR.TCONT»D1LAST,PENTOT»PAVTOT,DPAVG»OAVG»TLAST. 00010630
* TDSUM,PNMAX.DELPtDPMAX,TREF»IFBAG»NFLAG»JFLAG»LOPCNT 00010640
COMMON/STABLD/Tt»T2,T001.T002fOPlrDP2,P01,P02»SIGNl,I,N,NL.NCHG 00010650
DIMENSION ZEHUM{|3)»ZEP.OS(9),1ZEROM<
-------
TABLE 15 (continued)
//SO.FT08F001 DO uMT«SVSDA,01SP=(Nt*,P»SS)»D8N»H8AGt«
// DCB«(HtCFM«rH,URECL»eO,BLKSUfc«aOO),8P»CE»(CYL.l5,U.«LSt)
//GO.FT10F001 DO UNIT"SYSDA,OISP*(Ntw,P«BS).D8NBUBA6J*
// DCH«(HECFM«FB,LPfcCL«80,BLKSIZE=aOO),SPACt«CCYL,(5,1).RISE)
X/GO.MltFOOl 00 UNlT«3YSOA,DISP:(NtH,P»S3),DSN»HB*6a,
// DCB»(RECFM«FB,LRECL«60,BLKSIZE««00),3PACt"(CYL.(5,1).RUSE)
//CO.FTiaFOOl DO UNlT«3YSOA,OI3Pt(Nt«,P»S3),DSN«llB»C5,
// OC8«(RECFM«FB,LRECL»80,BLKSIZE«aOO),3PACE«(CYL.(5,l),RL3E)
X/CU.FT1Jf001 00 UNIT»3Y30A,DI3P*(NEH,PA63),D3N«ttBAC6,
// DCH«(HECFM»FB,LRECL«80,6LKSI/E«aOO),3PACE«(CYL,(5,l),RLSE)
//GO.Ftl«FOO| DO UNIT«3YSDA,018Pe(Mfc««»fA8S),OSN«HBAG7f
// OCB«(RECFM«FB,LP.ECL»60,BLK3UE««00),SPACE"(CYL, (5,1),RISE)
//GO.FTI5F001 DO UNIT»SYSDA,DI3P*(NEW.PA83),03N«HBAC8,
// OCB»(RECFMmF8,LRECL«80,BLK31ZE»«00),3PACE«(CYU,(5,t),RL8E)
//GO.SY3IN DO *
//• INSERT INPUT DATA CAHOS BEFORE THIS CARD
//8UMTMU fcXEC FORTGICG.ACCT«CU3T,P*WM.GO««SIZe«66K'
//FORT.3Y31N 00 *
(continued)
119
-------
TABLE 15 (continued)
••••••••••••••••••••••••a•••••••••••••*00000OPO
c ooooooio
C STEP* SUMTBL HAGHUUSE STEP • i. 000000?0
C SUMMARY TABLE GENERATOR FOR HAGHOUSE MODEL 000000*0
c UNIT 8 • PRESSURE vs Tint . */M2 vs MIN 00000040
C UNIT IS » FRACTIONAL PENETt-E ATION VS TIME 00000050
C UNIT 10-14 • INDIVIDUAL COMPARTMENT PLUMS FOR COMP. 1-S VS TIME 00000060
C FIRST HECURO OF FILE «8 CONTAINS PRINT FLAG,PLOT FLAG,MAX * OF 00000070
C COMPARTMENTS FOR «MJCH FLOfcS ARE PRINTED 00000080
C OsNU PRINT/PLOT AND 1« VtS FOR PHINT AND PLOT FLAGS 00000090
C i IN THE PRINT LOCATION INDICATES DETAILED OUTPUT 00000100
C FIRST 8 OR SO RECORDS ARE SEf UP BY PLOT IN { HEADINGS AND SCALE 00000110
C FACTORS* ETC. ) 00000120
C DATA ARRANGED AS i DATA POINTS PEN RECORD 00000110
C oooooiao
DIMENSION TTMt(3).PRESSRC3)fSVSFLO(J)
REAL INDFLO(5,J),PtNET(i)
REAL'S HEADCfl),CUMP
DATA COMP/ i COMP.'/
LINtS«60
IPR>6
HEAl»C«.!>OU)IPHINTf I MAX
IF (IPHIM.Nt.l.AND.IPRINT.Nfc.*?) GO TO 1000
READ(8,S05) HMD
BACKSPACE 8
DO 10 INBfl.lO,,?
00 10 J«l»7
IU RtAO(lN.blO)DUMM¥
DO 20 JB1,7
?0 HEAD(15,510)DHMMT
JMAX«9«IMAX
IF(IMAX.EQ.l)t,0 Tl) 40
DO SO IN«11,JMAX
10 HEADdN.SIOlDUMMV
40 CONTINUE
C- READY TO READ IN DATA
IPAGE«0
SO READ(8^20)(TIME(I)>PRESSR(I)>I>1»1)
IF(IPHINT.EO.I) GO TO 65
00 60 IN»10,JM*x
60 REAO(IN,5JO) (INDFLO(IN-9,l)«Ulfl)
65 CONTINUE
HE AD ( 15, 530 HPENtTU ),!»!,})
C- DATA HAS BEEN READ IN
IF(TIME(2).Ll.l.t-5.ANP.Tl^t (3).Ll.I.t-5)Gl) TO 1000
IF(LINES.LT.56)GO TO 70
IPAGE*IPAGE»1
LINES'6
IMIPHINT.EQ.2) NRITECIPh,600) HE AD, IPAGE, (COMP, 1 , 1«
, JMAx )
00000160
00000170
oooooieo
00000190
00000200
00000210
00000220
00000230
00000240
00000250
00000260
00000270
00000280
00000290
00000300
00000310
00000320
00000330
00000340
00000350
00000360
00000370
00000380
00000390
00000400
00000410
00000420
00000430
00000440
00000450
00000460
00000470
000004HO
00000490
(continued)
120
-------
TABLE 15 (continued)
70
flO
1000
loto
500
SOS
StO
S)0
600
IFCIPRINT.EO.I) HHlTEUPH,t>20) HfcAD,IPASE
CONTINUE
UO 60 I«l,3
imPRINT.EQ.2) WkITEUPR,610) TIME C I ) ,PRESSHU ) rPENE T < I ) ,
UNDFLO.PENEHI)
CONTINUE
LINES«LINE3*3
UU TO SO
no 1010 i»a,i5
KE.WIND i
FORMATCII,1X,1|)
FORMAT(8A8)
FORMAT(Al)
FO»MAT{6C»0.b)
FORM*T( 'ISUMMARV TABLE I ' ,iX,8A8,SX» "PAGE ' i 2X> I2//
1 T22r ' PRESSURE 'rT34, * FRACTIONAL1/
2 T«. 'TIME«,T2a,' DROP '»T30,t PENETRATION', T6a,
S 'INDIVIDUAL COMPARTMENT FLONS
S AS,Il,I97,»5,lUTn2,A«,,)l/J
01 0 FORMAT(}x,FlU.2,SX,FiO.O,5x,lPE9.i,5x,OPF|0.«»a(S)(,Fl0.al)
620 FURMATt '1SUMMAR* TABLE » ' ,2x»6A8,5X, 'PAGE' ,2X, 12//
1 T22,'PHES8UHf',T3«, 'FRACTIONAL'/
TIONAL'/
TO, ' TIME ',T2U. 'DROP ',T3U,'PtNETHAT ION'/
TV,'(MIN)",T2J,'(N/M2)'/)
END
UOOOOSOO
OOOOOSIO
UOOOOS20
OOOOOS30
uoooosao
uoooosso
00000560
OOOOOS70
00000580
00000590
00000600
00000610
00000620
00000630
00000640
000006bO
00000660
00000670
00000680
00000690
00000700
00000710
00000720
00000730
00000740
00000750
00000760
00000770
//CO.FT08F001
//CO.mOFOOl
//Cn.FTtlFOOl
//GO.FT12FOOI
//GO.mSFOOl
//60.FTI4FOOI
//GO.FT15F001
//SCRIBE EXEC
//rORT.SVSIN
DO UNIT«3YSOA,OI8P«(OLO,PAS8),08N««.8IMULA.60,FT08F001
UO UNIT»SYSDA,DI8P«(ULO,PA88),08N«».SIMULA.60.FT10F001
DO UNIT«8r3DA,DISP«(ULO,PASS),DSN".SIMULA.CO.FT11F001
00 UNIT«SYSD»,013P*(OLD»P*8S),D8N«..SIMULA.GO.FT12FOOJ
DO UNIT»3YSI>*,OISPKULD,PA3S),DSN««,3IMULA.GO.FTI3FOO|
00 UNIT«3YSDA,UISP«(ULO,P»S8),l)8N««.SIMULA,GO.FT1«F001
DO UNIT»SY3DA,DI3P»(OLO,PASS),DSN»«.SIMULA.GO.FT15F001
FORTGICG,ACCT«COST,PARM.GO«'SIZE«175K>
DO •
(continued)
121
-------
TABLE 15 (continued)
C GRAPH LIBRARY 7/I6/75/HWS CC» TECHNOLOGY
C
C
C
VERSION
CARDS-
TITLEd-64) DPTIUNS1 XPUSC6i»-69) YPOSUO-74) HE 1GHTC75-79)
C XAXIS LABEL(1-6«)(1PTK>NSI HEGIN(69-7«) UNITS UK LOG3/INCH( 75-80)
C YAX1S (SAME)
C TYPE (VAXIS-XAxlS)(SEMI,LOG-»PROB,BAH-m-8)
UPUDNSI LOG-<9-12) FOP A LOGRITHMIC BAR GRAPH
NEN GRAPH DIST(35-40) OtFAULT«6
X-AXIS MEIGHT(45-50) DEFAULT*2
X AXIS LENGTH(55-60) DEFAULT*6
Y-AXIS LENGTH(65-70) OE.FAUI.TcS
DOUBLE AXIS(74) 1 FOR X, 2 FOR Y,
3 FOR BOTH
DATA
xd-io)
OPTIONI
SYMBOL (75-80)
Yd 1-20)
X(21-30)
POINTS BETWEEN PLOT SYMBOLS
NEGATIVE FOR SYMBOLS BUT NO LINES
C OPTION (ENO,NEH,SAME>(75-78)
C
C
>0) Y(51-60)
MAKES NEH GRAPH-REPEAT ALL CARDS)
(SAME PLOTS ON OLD GRAPH-NO x-v AXIS)
(79-80) (CHANGE 'SYMBOL' FOR NEXT PLOT)
DIMENSION lfllJF(4000),XAR(1002)«VAR(1002),PRN(50),PRB(IOO)
DIMENSION XPLAR(26),YPLAH(26>,XPROR(38),YPLA5(26)
REAL L(IG,NEN,Nf-XT,NEX
REAL*8 TAN(H),XLABl8),YLAtt(»)»3PLAH(12)»SPLATCl2)
DATA XPLAB/.00,.30,.48,.65,.91,1.10,1.32,1.65,1.95,2.30,2.56,2.78,00000240
fc3.00,3.22,3.44,3.70,0.05,4.SS,4.68,4.90.5.09,5.35,5.52»6.00,0..1,/00000250
00000000
00000010
00000020
((80)00000030
00000040
00000050
00000060
00000070
00000080
00000090
00000100
00000110
00000120
00000130
oooootao
OOOUOI50
00000160
00000170
00000180
00000190
00000200
00000210
00000220
00000230
DATA YPLAB/25*0.0,1./
DATA YPLAS/?U«5.0..0,1./
DATA SPLAH/'.OI .05
2 ' «0 50 6
3 ' 99.9 9
DATA SPLAT/1 99.99 9
2 '0 60 SO
3 '.I .1
00000260
00000270
'«' 20 301,00000280
','99 99.5 -,00000290
00000300
90',* 80 71,00000310
'.!.«? .5'.' 12 «,« 5 10
'0 70 »0'»' 90 ','95 98
•9.99 •/
'9,9 99,','b 99 98 ',' 95
' ao so ;,'20 to ;,' 5 2 ',* i .5 1,00000320
'.01 '/ 00000330
DATA XPROB/.0,. 16,.
-------
TABLE 15 (continued)
IPU8*0 00000500
8ARX«0. 00000510
BARY»0. 00000520
PRUBX«0. 000005)0
LTYP»0 000005UO
20 I3YM«I3YM»i 00000550
NEXBNEXT 00000560
IF(NEXT.Nt.SAME) GU TO SO 00000570
XBEG»XAR(IMAX»1) 00000580
XINC*XAK(IMAX«2) 00000590
YBtC«YAH(lMAX+t) 00000600
YINC«YAH(1M4X»2) 00000610
C TITLE 00000620
30 READUNUNIT,«0,END«1000) TAR,XP03.YPOS.CHIT,CONT 00000610
40 FORMATC8AB,3G5.2,A1) 000006140
IF(ABS(XPUS).LT.1.E-20) XPUS«.5 00000650
IF(AB3(VPOS).LT.i.E-20) YP(1S*8.0-(.25»1POS) 00000660
IF(CHlT.Lr.l.fc-20.ANO.ISYM.E0.1.ANO.CUNT.NE.BLA) CHIT*.21 00000670
IF(CHIT.LT.1.E-20.*ND.CONT.EO,BL») CHJT«.1« 00000680
MRlTE(lOUTUNfOl) TAR,XP03>YPU8,CHIT,CONT 00000690
fli FORMiTdX.BABf JX,lXP08»'»F7.J,iX,lYP03"'.^7.5»Sl«»lMEICMT»',F7.5. 00000700
I SX,'CdNT«',Al) 00000710
IF(CONT.NE.BLA.OR.IPOS.EO.O) GO TO a5 00000720
XPU3»XPOS».2 00000730
00
-------
TABLE 15 (continued)
IF(AB3J«? 00001160
HtAOtINUN1T,«0,END»9H) ((XAH(M),rAR(M)),M«J,KJ,NEXT,NE«3YM 00001170
HO FOHMAT(6G10.S,T75,AARCM)»YAR(M)),M»J,K),NEXT.NEW3YM 00001190
65 FUHMAT(iX,6(lP£10.3),T7U,AU,T93,lJ) 00001200
IP{XAK(J+1).LT.l.E-ao.AND.VAHlJ*>)tLT.l.E-20.AND.XAR(KJ.LT.l.E-20 00001210
l.ANU.YAH(K).LT.l.fc-iO) J«J-1 00001220
IF(XAR(K).LT.1.E-20.*NO.VAR(K).LT.1.E-20) J«J-1 00001230
J»Jt3 00001240
IF(J.CT.IOOO) CO TO 90 000012SO
IF(NEXT.EO.BIA) GO TO 100 00001260
lF(XAH(J-l).LT.l.E-20.AND.YAR(J»l).LT.l.t-20) J«J-1 00001270
IF(NEWSYM.NE.O) LTYPcNEMSVM 00001280
90 IMAX«J"| 00001290
IF(NEXT.EO.BLA) NEXT«ENOD 00001SOO
GO TO 102 00001310
100 CONTINUE 00001320
C SCALES AND AXIS 00001S30
102 XAR(IMAX*J)«XbEG 00001S40
XAR(lMAXt2)*XlNC 00001350
YAB(JM»x»l)«YBEG 00001360
YABC IMAX+DsVINC 00001370
C CUT OFF VALUES OUT UF HANGE 00001380
IF(AHS(XlNC).l T.l.t-20) GO fU 106 00001390
IF(XJYP.EO.PHdH) GU 10 106 00001000
XBH.«XHtU»XlNC«XAXL 00001410
IFCXTYP.EU.LUG) XBYGBXHhG*!0*«(xINC*XAXL) 00001420
00 104 IMt_OOP»l,lMAx 00001430
IF(XBYG.GT.X«EG.AND.XAK(1MLDUP).GT.XHY&) XAR(IMtnOP)»XBYG 00001440
IF(XBYG.GT.XMtG.ANI).XAH(JMLDUP).Ll.XBfcG) XAH(IMLHOP)«XBEG 00001450
if (XBVU.LT.XHtG.ANI>.XAH(IMLiiOP).Ll.xBYG) XAR( IHLUOP)«XBYG 00001460
IF(XMYU.LT.XBtG.AND.XAR(IMLllUP).GT.X8EG) XAR(IMLOOP)«X6tU 00001470
104 CONTINUE 00001480
106 lMA8S(YlNC).LT.I.t-20) (.0 ru 110 00001490
(continued)
124
-------
TABLE 15 (continued)
YBYG«YBEG»YlNC»YAXL 00001500
IF(YTYP.EQ.LOG) YBYG*YBEG*10**(YINC*YAXL) 00001510
00 10B I«LOOP«1,IMAX 00001520
IF(VBYG.GT.VBEG.AND,YAH(mOOP).G1.YBVG} YAR d*LOOP) »Y8YG 0000ISJO
IFCYBYG.GT.YBEG.ANO.YARUMLOUPJ.LT.YBEG) YARCIMLOOP)»YB£G 00001540
IFCYBYG.LT.YBEG.AND.YARC1MLOUP).LT.YBYG) YARO.OfYLAB»-6C.YAj(L«90.0,YAR(IMAX*l),YAR(IMAX*2n 00001800
130 IFCXtYP.NE.LOG) GO TO 1«0 00001810
1F(XINC.LT.1.E-20) GO TO 135 00001620
inxBEG.GT.l.F-20) GO TU 133 00001830
XBEG«1. 00001810
XAR(IMAX*l)sl. 00001850
133 CONTINUE 00001860
GO TO 136 00001870
135 CALL 3CALG(XAH,XAXL,1MAX,1) 00001880
136 CALL L(.AXS(0.0,0.0,XLAB,-bU.XAxL,0.0,XAR(IMAX*1),XAH11MAX+^)) 00001890
IF(IDfJUB.fcQ.|.(]R.IOOUB.EO.i) 00001900
•CALL LGAXS(O.U,5.0,XLAB,6«»XAXL»0.0,XAH(1MAX*1),XAR(IMAX*2)) 00001910
140 IF(YTYP.NE.LOG) GO TO 147 00001920
IF(YINC.LT.1.E*20) GU TU 14S 00001930
IF(YBEG.GT.1.E-20) GU TO 143 00001940
YBEG»I. 00001950
YAHdMAXf l)«l. 00001960
143 CONTINUE 00001970
60 TO 146 00001980
145 CALL 8CALG(VAR,YAXL.1MAX,1) 00001990
(continued)
125
-------
TABLE 15 (continued)
146 C*LL LG*«S(0.0,0.0,YL*B,6«.Y»XLf 90.
JF< IDOUB.GE.2)
1C ALL LGAXS(6.0,0,0,YLAB,-6«, YAXL. 90.0, YAR UMAX tl),YAR UMAX ti))
147 IF (XTYP.Nt.SEMI.UH.YTYP.Nfc.SEHI) UU Til ISO
CALL LINtlXAR, Y»W, 1M»X, 1 ,L 1YP.I3YM)
GO TU 500
ISO IF(xTYP.Nt.StMl.OH.YTYP.NE.LUG) CO TU 160
LUGT*t
GU TO 160
160 lF(xTYP.N£.L(ir,.nH.YTYP.NE.StMl) GO TU 170
LOGT«-1
GO TU ISO
170 IFCXTYP. NE.LdC.OW.YTYP.NE.LOG) GO TO 200
LUGT«0
180 CALL LGL1N(XAH, YAH, IMAX, 1,1. TYP,1SYM, LOGT)
GO TO 500
BAR GRAPH
200 IF(XTYP.Nt.BAk) GO TO 220
rAH(IMAX*l)«YAR(IMAX)
DU 210 I«l,TMAx
J«IMAX-I»I
XAH($*J+1 )rxAk(j)
XAR(3*J-1)«XAH(J)
YAH(J*J*t )«YAh(J*|)
YAR(J«J)»YHEG
YAH(3»J-1)»YAR(J)
210 CONTINUE
XAW(l)«XHtG
XAR(IMAX»1)«XH£G
XAH(IMAX+2)»X1NC
YAH( IMAX+DBYBEG
VAk( IMAX»2)«Y1NC
XTVM«ZIYP
CO TO 110
220 IF(YtYP.Nt.BAR) GU TU 2SO
XAR(IMAX«1)«XAR(IMAX)
DO 230 l»l,IMAx
J"1MAX-I»1
YAR(2*J)«YAR(J)
VAH(2*J-1)«YAH(J)
XAR(2*J)*XAR(j*|>
230 XAH12«J-1)«XAH(J)
IM»x«2*IMAx
8AkY«l.
WP«ZT»P
XAR(IMAX+l)«Xt»FG
XAR(IMAX*2)«XINC
00002000
00002010
00002020
00002010
00002040
000020-jO
000020*0
00002070
00002040
00002090
00002100
00002110
00002120
00002130
U0002140
000021SO
00002160
00002170
U0002180
00002190
00002200
00002210
00002220
000022)0
00002240
000022bO
00002260
00002270
00002260
00002290
00002300
00002310
00002320
00002330
00002340
000023SO
00002360
00002370
00002360
00002390
00002100
00002010
00002020
00002430
00002400
000024SO
00002460
00002470
00002400
00002490
(continued)
126
-------
TABLE 15 (continued)
VAR(IMAX*l)BVBtG
YAHUMAX«25«Y1NC
GO TO 110
C PHUB GRAPH
250 IFWYP.Nt.PROBJ CO TO JOO
IFCNtX.EU.SAMt) GO TO 255
XPLABC26)«6.0/XAXL
CALL LINE(XPLAB,YPLA6,2«»l,l»l))
CHXP>XAXL/6.*.06HI
PSYMS»-CHXP
PSYT«-.l/*(XAXL/6.)
CALL SYMBOL (XPRQB(LP)«(RLP-LP)*(XPROB(LP«l)*XPROB(LP)))/5.
IF(LEFT.EQ.O) X AR( I )«6.-XA(. ( 1 )
CONTINUE
XAHl IMAX+1 )»0.
XAR(IMAX42)aXAXL/(>.
PROBX«1.
XTYPtSEMI
GO fti »"7
IF(YTYP.Nt.PMOH) 00 TO aSO
WR!TE(6i
-------
TABLE 15 (continued)
CALL PLOT(PMdVe.O..-3) 00003000
GO TO 10 OOOOS010
510 IFCNEXT.EU.SAMf) GO TO 20 00003020
1000 HHITECIOUTUN.1010) NFXT OOOOSOSO
1010 FIJRMATC tND NtXf» '»A«) 00003040
C4tl »»tUtCPMUVt,0.,«»99) OOOOiOSO
1020 CAtt EXIT OOOOJOfcO
00003070
//GO.SVSLIB 00 t>I8P«3HB
// OD 08N«STSi.PLOTTER.OISP«SHR
//60.PLOTTAP6 01) DSNiPL(ITS6S6.
// OI8P«C»t»EIP),UNlT"(TAPE7,,OfcFfcH)»
// L»BEL»(fBLP).VOL-3E««t'LXXXX
//60.FT03F001 00 UNITaSYSDAt01SP*(OLOfPASS)»D3***.SUMTHL.60.FTOaFOOt
// Od uMT«3YSO»»013P«(OLO,P»SS),DSN»*.3UMTBL.GO.FT>OFOO»
// DO uMTmSYSDA»DlSP»(ULD,PAS3),DSN««.3UMTBL.GO.FTllFOOl
// 00 uMT«8YSOA,DI8P«(ULD,PAS8),03N«*.8UMTBL.GO.FTJ2FOOt
// 00 UNIT«SYSOA,OI3P«(OLO,PASS),03N««.3UMTBL.GO.FT13F001
// 00 uNJT«SYSOA,DlSP«COlO,PASS),OSN««.3UMTBL.GO.FT|aFOOt
// 00 uMT»3Y80A,OI3P«(ULOfPA88),D8N»«.SUMTBL.CO.FT15FOOl
X/60.FT04FOOI 00 DUMMY
(continued)
128
-------
TABLE 16. VARIABLES AND ARRAYS USED IK BAGHOUSE SIMULATION PROGRAM, STEP 1
VARIABLES
ACHK2
ACLEAN
ACLN
ADELDP
Absolute value of CHK2.
Fractional area cleaned, calculated or input.
Fractional area cleaned, a , calculated in CLEAN if
WTOTAL is nonzero.
Absolute value of slope of average pressure drop, F,
versus time curve N/m2/min.
AMPLIT
AREA
ATEST
BAG1
BAG2
BLANK
BLANKS
Bl
CHK1
CHK2
CLAREA
CONTOT
CZERO
CZEROE
DATYPE
DELDP
DELP
DELT
DELTT
DETAIL
DPAVG
DPAVGN
DPMAX
DPSTOP
- Shaking amplitude, half -stroke, cm.
- Fractional area on a bag. The product of AREA and the
number of areas cleaned gives the fractional area cleaned.
- Intermediate calculation in determining AREA.
- Heading, 'SBAG1.
- Heading, 'QBAG' .
- Four blank characters.
- Eight blank characters.
- Slope of least squares fit to either log (slope) of P
versus time or F versus time, min"1-
- Estimated fractional error for Check No. 1 in STABLE.
- Estimated fractional error for Check No. 2 in STABLE.
- Fractional area cleaned on a bag, calculated.
- Total outlet concentration from the system, g/m3.
- Inlet concentration, calculated, g/m3.
- Inlet concentration, input, g/m3.
- Type of printed data requested, input.
- Slope of P versus time, N/m2/min.
- System pressure drop, N/m2.
- Time Increment, min.
- Intermediate in determining time increment, min.
- Used to check for 'DETAILED1 results request.
- Intermediate in calculating average pressure drop, N/m2.
- Average pressure drop at the end of a cycle, N/m2.
- Maximum pressure drop during a cycle.
- Maximum system pressure, if exceeded cleaning begins, N/m2.
(continued)
129
-------
TABLE 16 (continued)
DPI - Sum of average pressure drops, F, N/ra2.
DP2 - Sum of the product of average pressure drop and time,
N-min/m2.
DRAG - Heading, 'AREA1.
DROP - Average pressure drop passed to STABLE, N/m2.
DTLAST - Time increment of last loop, min.
ERR - Error used in determining cleaned area.
FREQ - Shaking frequency, cycles/sec.
I - Index. Error code.
IAREA - Number of areas on a bag.
IBAG - Bag index.
TERROR - Error code.
IFAREA - Number of the area to be cleaned.
IFBAG - Number of the bag just cleaned.
II - Index.
INPUT - Input device, initialized in subroutine DESINE to a value
of 5. All cards are read from INPUT.
IREPT - Line counter for output of intermediate calculations.
IUNIT - Output file number.
J - Index.
JCODE - Accuracy code, input. This is subsequently changed from
input (0 or 1) to (1 or 3) to alter the limits (LIM1, LIM2)
in STABLE.
IFLAG - Flag from STABLE to signal convergence.
JLOOP - Index in time loop.
JTIME - JLOOP-1.
K - Index.
K3 - Index in determining when to write on a file, data points
for graphs are written three at a time.
L - Index.
LCODE - Flag in STABLE signaling convergence.
LDIAG - Detailed print diagnostics; if true, intermediate calcu-
lations are output.
(continued)
130
-------
TABLE 16 (continued)
LIM1 - Limit for Check #1 in STABLE.
LIM2 - Limit for Check 92 in STABLE.
LMAX - Maximum number of individual flow rate graphs, limit - 5.
LOPCNT - Number of cycles modeled at any time in the simulation.
LOPTST - Difference between NT and LOPCNT.
M - Number of increments per bag, input.
MAXJ - Total number of increments used in time loop.
MAXK - Maximum number of bags for which calculations are output
per line.
MMD1 - Mass median diameter of reference dust, ym, input.
MMD2 - Mass median diameter of inlet dust, ym, input.
N - Number of bags (compartments), input.
NAREA - Number of areas to be cleaned.
NCHG - Number of times the slope of DPAVG versus time curve has
changed sign.
NCHK3 - Difference between the changes in average pressure drop
for two successive cycles where the slope of the DPAVG
versus time curve is changing sign.
NFLAG - Number of cycles completed after convergence.
NL - Number of cycles completed - one for use in Check #1,
STABLE.
NT - Maximum allowable number of cycles modeled, input.
OUTPUT - Output device for printed data. Initialized in DESINE to
a value of 6. All printed output is written to OUTPUT.
PAVNOW - Average penetration at the end of a cycle.
PAVR - Average penetration at the end of a cleaning cycle.
PAVTOT - Intermediate in calculating average penetration.
PENTOT - Total system penetration at any time.
PLOTER - 'PLOT1.
PLTYPE - Type of plotted data requested.
PNMAX - Maximum penetration (fractional) during a cycle.
P01 - Sum of natural logarithms of slope of DPAVG versus
T curve.
(continued)
131
-------
TABLE 16 (continued)
P02 - Sum of product of natural logarithms of slope of P versus
and T.
QAVG - Intermediate in calculating average system flow, m/min.
QAVGN - Average system flow at the end of a cycle, m/min.
QSYSTM - Total system flow, m/min.
R - Porosity function in Happel theory for K£.
RHOBLK - Bulk density of cake, g/cm3-
RHOP - Discrete dust particle density, g/cm3.
SE - Effective drag, input, N-min/m3.
SFAB - Fabric drag, N-min/m3.
SGI - Geometric standard deviation of size distribution of
reference (measured) dust.
SG2 - Geometric standard deviation of size distribution of inlet dust.
SIGN1 - Slope of APavg versus time curve for last cycle modeled,
N-min/m2.
SMALQ - Specified constant total flow, input, m/min.
SOLID - Solidity, 1 - e (porosity).
SR - Residual drag, N-min/m3, input.
SSYSTM - Total system drag, N-min/m3.
SUM1 - "SUMMARY".
SUM2 - "SUMMARY".
S0B2 - Square of specific surface of reference dust, ym~2.
S0F2 - Square of specific surface of inlet dust, ym~2.
S02 - Specific surface of inlet dust, ym~2.
T - Cleaning cycle time, input, min.
TAVG - Average of previous and current continuous simulation times
at which cycles end, min.
TCLEAN - Single bag cleaning time, input, min.
TCONT - Continuous simulation time, min.
TCORR - Correction for time interval splitting at the end of a
cycle, min. Currently this is always set to zero.
TCZERO - Temperature at which inlet dust concentration was
measured, C, input.
(continued)
132
-------
TABLE 16 (continued)
TDIF
TDSUM
TEMPK
TIME
TLAG
TLAST
TMOD
TREF
TSE
TSR
TTEST
TTEST1
TTEST2
TZKR
TZK2
T001
T002
Tl
T2
VRFLO
VRFLOW
VZK2
Total cycle time, min.
Sum of all time increments constituting a full cycle, min.
Gas temperature, input, °K.
Dummy variable in STABLE through which the continuous
simulation time is passed at the end of a cycle.
Time between cleaning cycles, min, input.
Continuous simulation time at the end of the previous
cycle, min.
Total cycle time - T + TLAG, reference time for cleaning
cycle, min. If pressure controlled (i.e., TLAG unknown)
TMOD is set to the continuous time, TCONT, at the end of
the previous cycle.
Continuous simulation time at which point convergence was
reached, min.
Temperature at which the effective residual drag, S_, was
measured, °C, input.
Temperature at which the residual drag, SR, was measured,
C, input.
TCONT in a modulo TMOD system; it is the time since the
current or last cleaning cycle started, min.
1.0001 * N x TCLEAN, min.
0.9999 x N x TCLEAN, min.
Temperature at which the initial drag versus loading slope,
K , was measured, °C, input.
1\
Temperature at which the specific resistance coefficient,
K2, was measured, °C, input.
Sum of all TAVG, min.
Sum of all squares of TAVG, min2.
Sum of all TIME, min.
Sum of all squares of TIME, min2.
Reverse flow velocity based on a single compartment, input,
m/min.
Reverse flow used in calculations; set to zero if not
cleaning, VRFLO if cleaning, m/min.
Velocity at which specific resistance coefficient, K2t was
measured, m/min, input.
(continued)
133
-------
TABLE 16 (continued)
WAREA
WCOMP
WPRIME
WR
WSTAR
WSTART
WTOTAL
ZK2
- Weight per unit area added to an area in one time
increment, g/m2.
- Intermediate in determining areas of highest loading, g/m2.
- Total minus residual fabric loading, g/m2.
- Residual fabric loading, input, g/m2, input.
- Constant for nonlinear drag model, g/m2.
- Absolute fabric loading at time zero, g/m2.
- Dummy variable through which a loading can be passed to
CLEAN for calculation of ACLK, g/m2.
- Specific cake resistance, K2, input, N-min/g-m.
CAKE(IBAG)
DP(I)
DPDP(NCMC)
IDUM(I)
IZEROM(J)
IZEROS(J)
OLDTIM(IBAG)
P(IAREA)
PDP(K3)*
PDQ(K3)*
PPS(K3)*
PQ(K3,LMAX)*
PT(K3)*
QAREA(IAREA)
QBAG(IBAG)
S(IAREA.IBAG)
SBAG(IBAG)
ARRAYS
Average fabric loading on bag # IBAG, g/m2-
Average pressure drop at the end of cycle # I.
Average pressure drop at the end of cycle # I.
Variable array index for output of intermediate results.
Array in subroutine INITIAL used to initialize Integer
variables In MODEL.
Array in subroutine INITIAL used to initialize integer
variables in STABLE.
Previous time for bag # IBAG, min.
Penetration for area # IAREA.
System pressure drop, N/m2.
System flow, m/min.
System penetration.
Individual compartment flow, m/min.
Simulated time, min.
Face velocity on area // IAREA, m/min.
Average face velocity for bag # IBAG, m/min.
drag of area // IAREA on bag # IBAG.
Total drag of bag # IBAG.
(continued)
134
-------
TABLE 16 (continued)
TIME(IBAG)
WD(IAREA, IBAG)
ZEROM(J)
ZEROS(J)
Total cycle time at the end of cycle $ I.
Time after cleaning for bag # IBAG.
Dust cake loading on area # IAREA on bag # IBAG.
Array in subroutine INITAL used to initialize real
variables in MODEL.
Array in subroutine INITAL used to initialize real
variables in STABLE
These arrays contain only three entries. When data is output for subsequent
processing by the plot routine SCRIBE, they are output in groups of three.
135
-------
TABLE 17. VARIABLES AND ARRAYS USED IN BAGHOUSE SIMULATION PROGRAM
SUMMARY TABLE GENERATOR, STEP2
COMP
DUMMY
I
IMAX
IN
IPAGE
IPR
IPRINT
J
JMAX
LINES
HEAD(8)
INDFLO(IBAG,J)*
PENET(J)*
PRESSR(J)*
TIME(J)*
- Used to generate table headings for compartment identifi-
cation « f COMP'.
- Dummy variable
- Index of DO loop.
- Maximum number of compartments for which individual flow
velocities will be printed, no more than five.
- Input device for reading compartment flow velocities,
logical units 10 to 14.
- Page counter.
- Output device. Currently has a value of 6.
- Print flag. If IPRINT = 0, no summary table is generated.
If IPRINT - 1, a table is generated. Location is the
first byte of the first record on unit #8, the pressure
versus time file.
- Index of DO loop.
- IMAX +9, value of the logical unit number for the last
individual flow file to be printed.
- Counter for number of lines printed.
- REAL*8 variable containing the title.
- Flow velocity through compartment # IBAG, data point 9 J
on any particular record.
- Average system penetration at time
# J on a record.
- Average system pressure loss at time
point # J on a record.
- Time at data point # J on a record.
TIME(J), data point
TIME(J), data
The data on the files are arranged in groups of three.
136
-------
APPENDIX C
EXAMPLES OF DATA INPUT FORMS,
METHODS OF DATA ENTRY AND DATA PRINTOUTS FOR
VARIOUS FILTRATION SIMULATIONS
Figures 15 through 21 and Tables 18 through 33 have been prepared
to demonstrate how the filtration model input data are handled from the point
where the necessary information is entered in a standard format on the input
forms shown in Figures 15, 16, 20 and 21 to the ultimate data printouts for
selected model applications. Sample printouts are shown in Tables 18 through 33
for input data reiterations, error messages, calculations performed within
the program, and excerpted tabulations of data printouts for sample data inputs.
The blank spaces appearing on the data input forms may indicate the
following situations:
• No data entry is available or no data entry is required for the
indicated variables. For example, no limiting pressure loss,
PL should be specified for a system^to be operated with con-
tinuous cleaning (Figure 15).
• The variable of interest may actually possess a true zero value,
e.g., the time between cleaning for which the model user may enter
a zero or leave blank. In the latter case, the model assumes
a default value of zero minutes which is consistent with con-
tinuous cleaning provided that PL is not specified (Figure 15).
• A zero or blank value of K£ indicates that no value is available.
Hence, entries for dust size and density parameters are required
so that K£ can be computed within the program (Figure 15).
• Zero or blank values for dust size and density properties
indicate that these data are not needed because K£ (along
with the temperature and velocity associated with its measure-
ment condition) are available (Figure 16). If the measuring
137
-------
conditions were 25°C and 0.61 m/min, K2 alone is sufficient for
entry because these specific reference conditions are auto-
matically processed by the program (Figure 21).
• If a value for K2 is not entered, a zero or blank value for SE
or WR indicates that no data are available and that the program
will automatically assign default values representing best
estimates for these terms.
Figure 15 shows a completed data input form for a continuously cleaned
filter system for which K£ is to be estimated within the model program and
for the rare occasion where the cleaning parameter, ac, has been defined
beforehand.
Table 18 shows a summary printout of the input data previously entered
on the input form with appropriate units so that the model user can be assured
that the simulation model will operate upon the correct data and present it
in the desired form. Note that assumed or default values contained within
the program will also be printed with the input data summary when actual values
are not available for items such as SE and WR or a blank value has been
indicated for reverse flow velocity, Vr.
However, those terms requiring calculation within the program or not
required as data inputs for the specific modeling conditions are not shown
in Table 18. In lieu of printing out a zero value "time between cleaning
cycles," the equivalent expression CONTINUOUS CLEANING is printed.
The printout shown in Table 19, Diagnostic Messages, indicates that
there are no errors in the input data with respect to the permissible numerical
ranges for input data, redundancies or data emissions which would automatically
stop any further program operations.
Table 20 lists the numerical values for those filtration parameters
actually computed within the program so that model user can appraise their
138
-------
FAMIC FILTER MOOCL-DATA WHIT font
t •Hill •»«,•»!
TITIX
«l«T«»
1 1 1 • 1 1 1 • I UIMI I • 1 1 1 • I
1 1 II 1 1 1 1 II 1 1 1 1 1 II 1 1 1 1 1 1 1 1 1 1 1 1 II 1 1 1 1
ai -n i4 »
\\ 1111 IUIMHIIMM MINI INI in n MMIIJ IN
Figure 15. Fabric filter model - data input
form for Example 1.
139
-------
TABLE 18. SUMMARY OF INPUT DATA FOR BAGHOUSE ANALYSIS
(REFERENCE FIGURE 15)
•••••••»•••••**•••••••••••••»*••••*»••*•*••••••••••••*•••*••••*•••••••••»«•••
OF INPUT DATA FOR BACHOUSE ANALYSIS '
COMINUOUS/K2 £STl*ATEU/AC ENTERED/DETAILED RESULTS/
BASIC DESIGN DAT*
NUMBER OF COMPARTMENTS 12
COMPARTMENT CLEANING TW 2,0
(OFF LINE TI»E)
CLEANING CYCLE M"l J6.0
CONTINUOUSLY CLEANED SYSTEM
REVERSE FLO* VELOCITY 0.0
OPERATING DATA
tvtRAGt FACE VELOCITY 0.9000
CAS TEM°ERATURE too.
INLET DUST CONCENTRATION 5.00
MEASURED AT 25.
FABRIC AND DUST PROPERTIES
SPECIFIC RESISTANCE. K2 ESTIMATED FROM
MASS MEDIAN DIAMETER 9.0
STANDARD DEVIATION 1.00
PARTICLE DENSITY 2.000
HULK DENSITY 1.000
EFFECTIVE RESIDUAL DRAG,
MEASURED AT
RESIDUAL LOADING. NP
SE
350.
25.
so.o
MINUTES
"JNUTES
DEGREES CENTIGRADE
G/MJ
DEGREES CENTIGRADE
MICRONS
C/CMJ
DECREES CENTIGRADE
6/K?
SPECIAL PROGRAM INSTRUCTIONS
VAX NUMBER OF CYCLES MODELED 20
ACCURACY LEVEL 0
TYPE OF RESULTS REQUESTED DETAILED /
FRACTIONAL AREA CLEANED, AC 0.50
-------
TABLE 19. DIAGNOSTIC MESSAGES (REFERENCE
FIGURE 15)
DIAGNOSTIC MESSAGES
THERE ARE NO ERRORS IN THE INPUT DATA
TABLE 20. INPUT VARIABLES CALCULATED BY PROGRAM
(REFERENCE FIGURE 15)
CALCULATED VALUES
INLtT OUST CONCENTRATION 1.99
CORRECTED TO OPERATING TEMPERATURE
AtO OUST CA»t PROPEBtus CO»l«ECTfO FOB GAS vlSCOSlTr
SPECIFIC CARE RESISTANCE, *i \.i>t>
EFFECTIVE DRAG, SF a9?.
n*ACTJON4L AMf* CLF«NE1'. AC O.SO
T|»-t I%C»t«ENT o.7$
SYSTf CHSSIAM k> 0.0 6/1*2
-------
TABLE 21. AVERAGE AND MAXIMUM PENETRATION AND PRESSURE DROP VALUES
FOR FIGURE 15 DATA INPUTS
F0|
ji.oo HINUUS
true NU«BE*
AVERAGE PE*ET«AUON«
AVERAGE PRESSURE DROP*
AvEHAGE SrSTEM FLO"*
MAXIMUM PENE.-RATION*
5.0«>i-OJ
71J.JU N/M2
0.9000 M/MIN
75". 7
F0|
J».QO MlKUTES OPEM*T[0*J, CfCLt N
7
AVERAGE PENETRATION*
AVERAGE PMESSURE DROP*
AVERAGE SYSTEM
PRESSURE DROP*
S.06F-01
Til. 30 N/M2
0.9(100
7S0.6? H/»i
J».00 MINUTES OPEHAT1U1, CYCLE NUMBER
NJ
AVERAGE PENETRAUON*
AVERAGE PMESSURE OROPI
AVEKA6E SY81EM FLOB«
MAXIMUM PENETRAUON*
MAXIMUM PRESSURE DROP*
S.OoE-OJ
713.29 N/M2
O.VOOO M/MIN
8.U9E-OJ
750.60 N/M2
-------
TABLE 22. EXCERPTED DATA FOR SYSTEM DETAILED PERFORMANCE CHARACTERISTICS AFTER
180 MINUTES OF SIMULATED FILTRATION (REFERENCE FIGURE 15)
••••••***••••»»••••••*•••»•••t.••••••!
RESULTS OF BAGHOU9E ANALYSIS
• •••••••*••••••••*••••••••••••«•*•»••••••«*••••.•.••«*••>••«.»»•..••••««•«••>•
CONTINUOUS/K2 ESTIMATED/AC ENTERED/DETAILED RESULTS/
BAG-DRAG*
1
2
3
a
5
6
7
g
9
10
1 1
12
BAG-FLO«>
1
2
I
a
S
t
7
8
9
10
II
12
ARE* 1
.I7E»02
,3SE»02
,52E»02
.6AE»02
.S«E»02
.OOE»02
,1SE»OP
.29E*02
.u«F. »02
,57E»02
,7JEt02
,OOE»02
AREA 1
.25E-OI
.08E-01
.92E-OI
r.77E-Ol
r.6)E.O|
r.soE»oi
,37£.0l
.26E-OI
.ISE-OI
.OSE-Ot
.95E-OI
,«JE-01
AREA 2
S.36E«02 (
.72E«02 (
,OflE»0?
.3aE*02
.6lE«02
,86E»02 •
7.10E»02 '
7.32E+02
7,S«E*02
7.7«E«02
7.93E»02
«.97E»02
AREA 2
I.2*E*00
1.IBE*00
I.I2E»00
1.06E*00
1.02E«00
.83E-01
.50E-01
.2IE-OI
.95E-OI
.72E-OI
.51E-01
I,36E«00
SBAG
>.«8E«02
>.79E»02
r.07E«02
r.33E*02
r.S7E»02
'.79E*Q2
r.99£«02
,|9£»02
,38E«02
,5bE»02
,73E«02
,I3E»02
OBAG
,0«£»00
.93E-OI
.58E-OJ
.20E-OI
.92E-01
.66E-01
.••£•01
.23E-OJ
.05E-OI
r.BBE-01
r.73E-OI
I,IOE»00
T* IBO.O
T»
CAKE
SBAG
OBAG
T*
CAKE
SBAG
OBAG
BAG 1
3.01
I.3607E«02
0.6«7«E«03
O.IO«2E«01
BAG II
33.01
2.«7«3E«02
0.873IE«OJ
0.772fc£»00
OELP* 67a.6
BAC 2
6.01
l.a86SEt02
0.6790E«03
0.993SE«00
BAC 12
0.01
I.2302E«02
0.6133E«03
0.||OOE«OI
BAG 3
9.01
1.6II8E*02
0.7072£f03
0.9539£«00
BAG
DELO*
BAG •
12.01
I.7306E»02
0.7328E«03
0.9205E«00
.9000
BAG S
15.01
1.80S6E*02
0.7S66E»03
0.8916E»00
CONCENTRATION* .3J93E-OI
BAG 6
IB. 01
I.9572E+02
0.7787E«OJ
0.8663E*00
BAG 7
21.01
2.06S9E*02
0.7995E«03
0.8aj7E»00
BAG 8
24.01
2.1719E»02
O.BI92tt03
0.823«E»00
•EIGHT
BAG 9
27,01
2.27S5E02
O.B379£*03
o.eosoEtoo
DUMPED* .0
BAG 10
30.01
2.3769C«02
0.8*59E»03
0.788IE«00
-------
TABLE 23. SYSTEM PRESSURE DROP, SYSTEM PENETRATION AND COMPARTMENT
FLOW DISTRIBUTION VERSUS TIME (REFERENCE FIGURE 15)
8U«"AR» TABLE i
TIME
(KIN)
0.01
0.7S
1.50
*.25
1.00
1.01
1.75
«.50
5.25
b.OO
6.01
«t.75
7.50
8.25
9.00
9.01
9.75
10.50
11.25
12.00
12.01
12.75
11.50
14.25
15.00
15.01
IS. 75
16.50
17.25
18.00
18.01
18.75
19.50
20.25
21.00
21.01
21.75
22.50
21.25
20.00
24.01
21.75
25.50
26.25
27.00
27.01
27.75
20.50
29.25
JO. 00
10.01
CONTINUOUS/** ISM"ATED/AC ENTERED/DETAILED
PRESSURE FRACTIONAL
DROP PENETRATION
CN/M2)
675.
669.
728.
7UU.
751.
675.
669.
720.
7u«.
751.
675.
669.
728.
744.
751.
675.
669.
728.
744.
751.
675.
669.
728.
744.
751.
670.
669.
728.
744,
751.
674.
669,
728.
74«.
751.
67tt.
669.
.091E-01
.blQE'Ol
.931E-03
.651E-OJ
.731E-03
.492E-OJ
.bllE'Ol
.91IE-01
.651E>01
.711E-01
.49JE-01
.011E-01
.9SIE-03
.651E-OJ
.7JIE-01
.49JE-01
.011E-01
.9J1E-OJ
.6S3E-01
.7}|E»01
,o92t-0}
.OJSE-01
.91IE-01
.651E-0!
.7JlfOJ
.492E-0}
,013E»01
.931E-01
.653E-01
.741E-OJ
.«92E-01
.U13E-01
,911E»01
,651fOJ
.MlE-OJ
.49JE-01
.OJJE-OJ
728. 0.93IE-OJ
748. 1.651E-0]
751. 2.71IE-03
674. 8.492E-03
6b9. 6.033E-03
726. U.911E-01
74U. 3.651E-03
751. 2.711E-01
67o. e.492E*01
669. 6.U33E-03
728. -.911E-01
740. 1.6SH-01
751. 2.71U-01
«>74. e.492E»01
RESULTS/
P«6t
INOIVlOUAl COMPARTMENT FLOnS
COMP.I
.0«|6
.0116
.1092
.1095
.1058
.9935
.9877
.0614
.0637
.0617
.9539
.9511
.0250
.0254
.0246
0.9204
0.9200
0.9922
0.9924
0.9926
0.8916
0.6930
0.9635
0.9636
0.9645
0.8663
0.8692
0.9380
0.9380
0.9395
0.8438
0.847ft
0.9152
0.9150
0.9170
0.8235
0.82B5
0.8946
0.8942
0.8966
O.H05I
0.8110
0.8757
0.8752
0.8779
0.7R82
0.7908
0.8583
0.8577
O.A607
0.7727
COUP. 2
0.9935
0.9877
.0614
.0637
.0617
.9519
.9511
,0250
.0254
.0246
0.9204
0.9201
0.9922
0.9924
0.9926
0,8916
0.8910
0.9635
0.9616
0.96«5
0,8661
0.8692
0.9380
0.9180
0.9195
O.HU18
0.8478
0.9)52
0.9150
0.9170
U.8215
0.8285
0.8946
0.8942
0.8966
0.8051
0.8109
0.8757
0.8752
0,8779
0.7882
0.7948
0.8583
0.8517
0.0607
0.7727
0.7799
0.0000
o.oooo
0.0000
1.0997
COMP.3
0.9539
0.9512
1.0250
1.0254
1.0246
0.920U
0.9201
0.9922
0.9924
0.9926
0.8916
0,8910
0.9615
0.9616
0.9645
0.866!
0.8692
0.9380
0.9380
0.9195
0.8418
0.8478
0.9152
0.9(50
0.9170
0.8235
0.8285
0.8946
0.8942
0.8966
0.8051
0.1109
0.8757
0.8752
0.8779
0.7882
0.7948
0.8583
0.8577
0.8607
0.7727
0.7799
0.0000
0.0000
0.0000
.0997
.0854
.1609
.1653
.1591
.0415
'
(M/M{N)
COKP.H
0.9200
0.9201
0.9922
0.9924
0.9926
0.8916
0.8910
0.9615
0.9616
0.9645
0.8661
0.8692
0.9380
0.9380
0.9395
0.8437
0.8478
0.9152
0.9150
0.9170
0.8235
0.8285
0.8946
0.8942
0.8966
0.8051
0.8109
0.8757
0.8752
0.8779
0.7882
0.7948
0.8583
0.8577
0.8607
0.7727
0,7799
o.oooo
0,0000
0.0000
.0997
.0854
.1649
.1653
.1593
.0«I5
.0116
.1091
.1094
.1057
0.9934
COMP.S
0.8916
0.8930
0.9635
0.9616
0.9au5
0.8663
0.8692
0.91SO
0.9380
0.9395
0.6417
0.8476
0.9152
0.9150
0.9170
0.623S
0.8285
0.8946
0.6942
0.8966
0.6050
0.6109
0.8757
0.6752
0.6779
0.7682
0.7946
0,6583
0.6577
0.8607
0.7727
0.7799
0.0000
0.0000
0.0000
.0997
.0854
.1649
.1653
.1593
.0415
.0316
.1091
.1095
.1057
.9914
.9676
.0611
.0617
.0617
0.9
-------
TABLE 23 (continued)
In
SUIMURY TABLE 1
TIME
(HIS)
10.75
11.50
12.25
11.00
35.01
11. 75
14. 50
55.25
56.00
16.01
16.75
17.50
18.25
19.00
59.01
19.75
40.50
01.25
12.00
U2.01
02.75
01.50
o«,25
«5.00
05.01
05.75
46.50
07.25
48.00
4A.OI
06.75
09.50
50.25
51.00
SI. 01
51.75
52.50
53.25
50.00
50.01
54.75
55.50
56.25
57.00
S7.0I
57.75
50.50
59.25
60.00
60.01
60. 7S
COfcTI*UOUS/K4 ESTIMATED/AC ENTEKEti/DE TAILED Hf SUITS/
PRESSURE FRACTIONAL
DROP PtMTRATION
INOlvlDUtL
(N/H2) COMP.l COMP. 2
669.
728.
740.
751.
674.
069.
728.
7«fl,
751.
670.
669.
720.
700.
751.
674.
669.
728.
744.
751.
670.
609,
728.
704.
751.
674.
669.
728.
744.
751.
670.
069.
728.
744.
751.
674.
669.
728.
7un.
751.
674.
069.
.015E-OJ 0.7799 1.0650
.93IE-03 0.0000 1.1009
.651E-05 0.0000 1.1651
.71IE-OJ 0.0000 1.159]
.M92E-01
.C11E-OJ
.91IE-01
.651E-03
.7J1E-OJ
.092E-01
.011E-01
.911E-01
.651E-OJ
.731E-03
.092E-01
.011E-01
.951E-05
.651E-U1
.71IE-01
.092E-01
.015E-05
.911E-OJ
.6blE-01
.73IE-01
1.0997 1.0015
.085« l.Ollb
.1609 1.1091
.1653 I.IU9U
.1591 1.1057
.0415 0.9934
.0316 0.9876
.1091 1.0611
.1094 1.0617
.1057 1.0617
.9934 0.9538
.9876 0.9SI1
.0613 1.0250
.0637 1.0251
.0617 1.02X6
.9518 O.V204
.9511 0.9200
.0250 0.9922
.0251 0.9924
.0246 0.9926
.092E-01 0.9204 0.8916
,011E»03 0.9200 0.8910
.91IE-01 0.9922 0.9635
.651E-01 0.9924 0.9656
.71IE-05 0.9926 0.9005
.492E-01 0.8916 0.8661
.011E-01 0.8930 0.8692
.91IE-03 0.9615 0.9381
.651E-OJ 0.9636 0.9180
,'lIE-Ol 0.9645 0.9195
.092E-01 0.8661 O.A018
.011E-01 0.8692 0.8078
.91IE>01 0.9181 0.9151
.653E-01 0.9380 0.9151
.MlE-01 0.9195 0.9170
.092E-OJ 0.8438 0.8235
.OJ1E-OJ 0.8478 0.8285
728. 4.93IE-01 0.9151 O.B946
7UU, l.OblE'Ol 0.9151 0.8902
'51. 2.71lt»03 0.9170 0.8966
67«. 0.91lE>03 0.89S6 0.0757
7uu. l.e51E«01 0.89«2 0.9»5i
751. €>.MU-01 0.8966 0.4779
6/«. ».->9«-Oi o.«05i o.7o»i
669. 6.0JJE-OJ 0.8110 0.79U*
P«Gt
CQHPARTMEM HON3
COMP.]
1.0316
1.1091
1.1094
1.1057
0.9934
0.9876
1.0615
1 .0617
1.0617
0.9538
0.9511
1.0250
1.0253
1.0246
0.9204
0.9200
0.9922
0.9924
0.9926
0.8916
0.8930
0.9635
0.9616
0.9645
0.8665
0.8692
0.9381
0.9160
0.9195
0,0456
0.6476
0.9155
0.9151
0.9170
0.6235
0.6265
0.6946
0.6902
0.6966
0.6051
0.6110
0.6757
0.6752
0.6779
0.7802
0.7906
0.6560
0.6578
0.4607
0.7727
0.7800
2
(M/HIN)
COMP. 4
0.9876
1.0635
1.0657
1.0617
0.9556
0.9511
1.0250
1.0251
1.0246
0.9204
0.9200
0.9922
0.9924
0.9926
0.6916
0.6910
0.9635
0.9636
0.9645
0.6663
0.6692
0.9181
0.9360
0.9395
0.6418
0.8478
0.9151
0.9151
0.9J70
0.6235
0,6285
0.6946
0.8942
0.6966
0.6051
0.8110
0.6757
0.8752
0.6779
0.7862
0.7908
0.858U
0.657*
0.8607
0.7727
0.7799
0.0000
0.0000
0.0000
1.0997
1.0650
COMP. 5
0.9511
1.0250
1.0253
1.0246
0.9204
0.9200
0.9922
0.9924
0.9926
0.8916
0.6910
0.9615
0.9616
0.9645
0.6661
0.8692
0.9361
0.9360
0.9395
0.8436
0.6478
0.9153
0.91S1
0.9170
0.6235
0.6265
0.8946
0.8942
0.8960
0.6051
0,8110
0.6757
0.8752
0.6779
0.7662
0,7906
0,6584
0.8578
0.8607
0.7727
0.7799
0.0000
0.0000
0.0000
.0997
.0850
.1600
.I6S)
.l1^!
.nuib
.0115
-------
TABLE 23 (continued)
SUMMARY TABLE «
TIME
(MIN)
61.50
62. 2S
63.00
65.01
63.75
64. SO
65.25
66.00
66.01
66.75
67.50
68.25
69.00
69. Ot
69.75
70. SO
71.25
72.00
72.01
72.75
73.50
74.25
75.00
75.01
75.75
76. SO
77.25
78.00
78.01
78.75
79,50
60.25
81.00
61.01
61.75
82.50
83.25
84.00
84.01
84.75
85.50
06.25
87.00
87.01
67.75
88,50
8V. 25
90.00
90,01
90. 7S
91.50
CUM1NUUUS/K4
PRESSURE
DROP
(N/M2)
726.
700.
751.
674.
669,
728.
7ou,
751.
670,
669.
726,
7oo.
751.
670,
669,
726.
744.
751 ,
670,
669.
726.
744.
751.
670.
669.
728,
700,
751.
674,
669.
726.
740,
751.
670.
669.
728.
704.
751.
674.
669.
728.
700.
751.
670.
669.
728.
lou.
751.
e7u.
669.
728.
ESTl»ATtD/AC ENTEREO/OtTAlLEO RtSULTS/ PACE 3
FRACTIONAL
PENETRATION ItUlvIOUAL COMPARTMENT FLOnS (M/MIN)
COMP.l CO«"P.2 COMP.3 CONP.ii COMP.S
4.93U-03 0.6757 0.8584 0.0000
S.6S3E-03 0.8752 0.8578 0.0000
2.731E-03 0.8779 0.8607 0.0000
8.09JE-03 0.786* U.7727
6.033E-03 0.79ofl 0.7600
4.93U-03 0.8560 u.OOJO
3.0S3F-03 0,8576 0.0000
2.731E-03 0,8607 0,0000
6.092E-03 0.7727
0.033E-03 0.7800
4.93IE-03 0.0000
S.653E-03 0,0000
2./3IE-03 0.0000
8.0*21-01
6.033E-03
0.931E-03
3.6531-03
2.731E-03
6.992E-03
6.033E-03
0.9J1E-03
3.653E.-03
2. 7311-03
8.09JE-03
6.U33E-03
0.931E-OJ
3.6S3E-03
2.731E-03
6.492E-03
6.013E-03
u. 9311-03
3.653E-03
2.73IE-03
.0997
.0850
.1609
.1653
.1593
.0015
.0315
.1091
.1090
.1057
.9930
.9876
.0633
.0637
.0617
.099?
,085u
.1609
.1653
.1593
.0415
.0315
.1091
.1090
.1U57
.9934
,9h/t,
.0633
.0637
.0617
.0997
,0654
.1649
.1653
.1593
.0015
.0315
.1091
.1094
.1057
.9934
.9676
.0633
.0637
.0617
.1649 1.1091
.1653 1.1094
.1593 1.1057
,0415 0.9930
.0315 0.9676
.109) 1.0633
.1094 1.0637
.1057 1.0617
.9934 0.9538
.9876 0.9511
.0633 1.0250
.0637 1.0253
.0617 1.0246
.9536 0.9200
.9511 0.4200
,0250 0.4422
.0253 0.4924
.0246 0.99»fe
.4536 0.4204 0.6916
.9511 0,9200 0.6930
.0250 0.9422 0.4635
.0253 0.4924 0.4636
.0246 0.4926 0.9645
.9536 0.9200 0.6916 0.6663
.9511 0.9200 0.6930 0.6692
.0250 0.9922 0.9635 0.9361
.0253 0.9924 0.9636 0.9360
.0246 0.9926 0.4645 0.4395
.4536 0.9200 0.8916 0.6663 0.8438
.9511 0.9200 0.89)0 0.6692 0.8076
.0250 u.9922 0.9635 0.9361 0.9153
.0253 0.9920 0,9636 0.9360 0,9151
.0206 0.9926 0.9645 0.9395 0.9170
8.092E-03 0.9200 0.8916 0.8663 0.6038 0.6235
6.053E-03 0.9200 0.8930 0.6692 0.6478 0.6285
0.931E-03 0.9922 0.9635 0,9381 0.9153 0.6946
3.653E-03 0.992U U.9636 0.9360 0.9151 0.8942
2.731E-03 0.9926 0.9605 0.9395 0,9170 0.6966
8.492E-03 0.6916 0.6663 0.8436 0.6235 0.6051
6.033E-03 0.6930 0.6692 0.6076 0.6285 0.6110
0.95IE-03 0.9635 0.9361 0.9153 0.6946 0.8757
3.053E-03 0.9636 0.9380 0.9151 0.6942 0.8752
2.731E-03 0.9645 0.9495 0.9170 0.6966 0.6779
S.492E-03 0.6663 11.1438 O.H2J5 0.6051 0.7862
6.033E-03 0.6692 \».8«/i> 0.8265 0.8110 0.79u6
4.931E-05 n,9381 0.9153 0.0906 0.8757 0.8560
3.6531-03 0.9380 0.9151 O.P902 0.1)75? 0.8578
2.7ilf03 0.9395 0.9)70 0. 6166 O.f779 . O.H607
e,492E-Oi n.«uj|» >).d235 a. 6051 (
1.7882 0.7727
6.uiJt-03 C.HJ7H n.Hrf*S O.tllO U.79UM 0,7800
U.9J1E-03 0.9153 0.804C 0.4757 (
.856" 0.0000
-------
TABLE 23 (continued)
SUMMARY TABLE 1
TIME
(MI«|)
92.es
91.00
91.01
93.75
90,50
95. 2S
96.00
96.01
96.75
97.50
96.25
99.00
99.01
99.7%
100.50
101.25
102.00
102.01
102.75
10). SO
104.25
105.00
105.01
IDS. 75
106.50
107.25
108.00
CONTIGUOUS/02 tSTl»ATEO/AC
PRESSURE FRACTIONAL
DROP PENETRATION
(N/M2)
744. 3
7S1.
674.
669.
728.
;«4.
7SI.
674,
669.
72B.
744.
7S1.
6/0.
669.
720.
7«a.
7S1.
674.
669.
72B.
ran.
7S1.
674.
669.
728.
744.
1.6S3E-03
.7J1E-OJ
.«92E-0)
.033E-03
.9J1E-01
.biSE-OS
.73IE-03
.4921-03
.OJJE-03
.911E-OJ
.6SJE-01
.7JIE-03
.492E-03
,033E»03
.93IE-03
.653E-OJ
.731E-03
.492E-OJ
.OJJE-OJ
.931E-03
.6S3E-03
.73IE-03
,a92f>03
.033E-03
.93|E>03
1.65JE-OJ
751. 2.73IE-OS
E^TiHtD/OETAILEO
COMP.l
0.9151
0.9J70
O.B23S
0.8285
0.8946
0.69«2
0.8966
0.8051
0.8110
O.B7S7
O.B7S2
0.8779
0.7882
0.7948
O.BSB4
O.BS76
O.B607
0,7727
0.7800
0.0000
0.0000
0.0000
.0947
.0854
.1649
.1653
.159}
RESULTS/ PAbE 4
INDIVIDUAL COMPARTMENT PLOHS (M/MIN)
CO«P.2 COMP.i COW. 4 COMP.5
0.894? 0.8752 0.8578 0.0000
0,8966 0.8779 0.8607 0.0000
0.8051 0.7882 0.7727
0.8110 0.7948 0.7800
0.8757 0.8584 0.0000
0.8752 0.8578 0.0000
0.8779 0.8607 0.0000
0.7882 0.7727
0.7948 0.7800
0.8584 0.0000
0.8578 0.0000
0.8607 0.0000
0.7727
0,7800
0.0000
0.0000
0.0000
.0997
.0654
.1649
.I65i
.IS93
.0415
.0315
.109]
.1094
.1057
.0997
.0854
.1649
.1653
.1S93
.0415
.031*
.1091
.1094
.1057
.9934
.9876
.0633
.0637
.0617
1.0997
.0850
.1649
.1653
.1593
.0415
.0315
.1091
.1094
.1057
.9934
.9876
.0633
.0637
.0617
.0997
.0654
.1649
.1653
.1593
.0415
.0315
.1091
.1094
.1057
.9934
.9876
.0633
.0637
.0617
.9538
.9511
.0250
.0253
.0246
.9538 0.9204
.9511 0.9200
.0250 0.9922
.0253 0.9924
.0246 0.9926
-------
reasonableness. The only exception is the printout for a which will always
appear regardless of whether computed within the program or an original data
input.
Tables 21 through 23 indicate the tabular printouts received when
DETAILED results are requested.
Table 21 provides a printout of average and maximum values over cycles
6 through 8 for dust penetration and filter pressure drop as well as showing
the average system flow (or air-to-cloth) ratio. According to checks performed
within the simulation model, approximate steady state operations have been
reached during cycles 6 through 8, thus eliminating the need for further cycling.
Table 22 represents a detailed summary of filter system performance
•
parameters after 180 minutes of simulated filtration. The instantaneous
gas flow and drag values for both the individual bag regions (areas 1 and 2)
and the entire bag (or compartment) are indicated for each of the 12 compart-
ments making up the filter systems. Also shown are the times that each com-
partment (bag) has filtered after 180 minutes of system operation along with
the corresponding fabric dust holding.
Over the 180 minutes required to execute filtration cycles 6 through 8
(Table 21) and the corresponding time interval 0.01 through 180 minutes
indicated in Table 23, a total of 144 separate tabulations similar to
Table 22 would be printed for each 0.75 minute time increment. It is
emphasized that this capability, which has been designed within the model for
research purposes only, is not called upon for routine model applications.
Table 23 provides a point by point tabulation of overall filter system
pressure loss and dust penetration for the 144 iteration periods cited
previously. In addition, gas flow distributance for 5 of the 12 compartments
148
-------
are indicated for each of the iteration periods. It should be noted that the
gas flow distribution data are only printed when a DETAILED printout is
requested for research purposes.
For those cases requiring a less rigorous data reporting, the specifica-
tion of SUMMARY printout will provide only the first three columns of Table C-6.
Figure 16 shows data inputs for a filter system to be cleaned on the
basis of pressure control as indicated by the data input of 1000 N/m2 for P^.
In this case, a zero or blank entry for "time between cleaning" merely indicates
that the true value is unknown and will be determined subsequently from the
final program outputs. Only six operating cycles were chosen so that the
printout could be demonstrated for the nonsteady state or nonconvergence
condition.
Table 25 shows a printout of the calculated and/or corrected values for
key input variables used in the modeling process for the Figure 16 data.
Tables 26 and 27, and Figures 17 through 19 represent the model
output received when SUMMARY PLOT is entered (Figure 16). Note that the
message "convergence to steady state not reached after 3 cycles" appears
on Table 26. Therefore, there might be some risk in accepting the average
and maximum values for pressure drop and dust penetration shown for the six
cycle data summary and the Table 27 tabulation of overall system pressure
drop and fractional penetration versus time over the 40.5 minute period starting
at the end of the third filtration cycle.
In Figure 17, average system pressure loss is indicated for three
consecutive filter cycles for a five compartment system. The pressure spikes
(positive and negative) depict the system pressure loss immediately before
and after the cleaning of each compartment. The smooth concave downward regions
149
-------
FAMIIC FILTER MODEL-DATA INPUT FORM
Ill M •I1BI1IIM I MM I M i M I MM MIMMI
4 » • r •
a: 11«»i icj=-,, ^
11 *•"• *"•} i T* '• 11
liyiiyiliTri mi • mi M M i M i M
M IT HI rt » a
i twi KPH 111111111111111 i 11111111111 n 111
is I
IM 11 I I I I M 1 i I I I I I I I I I I I I I I II II I i I II M i II II 11 11 1
Hill 1 1 1 1 MN NNm I M M M M M I M M M M 1 M M I I
u.
Figure 16.
Fabric filter model
form for Example 2.
- data input
150
-------
TABLE 24. SUMMARY OF INPUT DATA FOR BAGHOUSE ANALYSIS
(REFERENCE FIGURE 16)
SliWMAUY |IF I-.IM.J 0414 Ink MAGNI'CSF »I,AL»SIS
PB£SS'Jl«t/«i <;ivkN/SU<"'APYU>l..'MEP »l Si>l 1 S/1U C i ". vt BG^C t /
"ASIC l.kSIG'. PATA
i. tS
TUEM ILMMM; ij
t \:t t |«f )
»r. C'LL^ I ivf
L|v|T|'r, M.tSSl "•£ 0«'lf
'Li..* ttLL'Cllv
I1PEP4I p. I. 1014
I .onon V/VJN
(.AS ltvkEBAtuut Ib". OFGBtlb CF ••' 1 f.nADt
IMEt ri.st Lt'.Ck'.Ti.tTID11' 10.ml (,/«i
Al 1SU. CtGBttS ff.TIGBAOt
F4BBJC A«l<
BtSISIASCt. K; 1.00 \.v|f,/(,--
AT |on. ntGBEES CtMI&HALIf
0.9000 "/"[f,
EFFKHvt »t SI DUAL D&AG, SE uOO. \-»tN/u3
MtASUCkU Al ISO. Dtf.BEkS CtMlGOAOE
WISIDUAL LOADING, nA 50.0 li/«?
OBAO. SB 7S. s.w^/wt
At 110. ntGBEtS CfSTIGPAOE
IMT1AL SLO^t. «•< u.BO s-vf./o-"
WEASUHED AT 1JO. rir.uEES Lt^TIGRAOt
SPECIAL PWC'GHAM INSTRUCT liiNS
»l>DElFC' b
u
TYPE CF MESULTS »
-------
TABLE 25. INPUT VARIABLES CALCULATED BY PROGRAM
(Reference Figure 16)
CALCULATED VALUES
INLE1 DU3T CONCENTRATION 10.00 C/»J
CORRECTED TO OPERATING TEMPERATURE
FABRIC AND DUST CAKE PRUPkHTlES CORWECTEO K)M GAS VISCOSITY
SPECIFIC CAKE RESISTANCEi *i 0,90 N»MIN/G«M
INITIAL SLOPED KM ».!« N»MIM/G>M
EFFECTIVE DRAG, SE «aS. N.MIS/MJ
RESIDUAL OHAGt SR 76. N.«IN/«i
Fh*CTIUN»L ARIA CLEANEDi AC O.lu
TIHF INCREMENT O.SO "I».UTES
J5 STSTE" COtSMNI »• IJ3.0
-------
TABLE 26. RESULTS OF BAGHOUSE ANALYSIS
(REFERENCE FIGURE 16)
fcfS"LTS 'if t-MiHOuSE »**L»SIS
i HI utSiil>S/M> Cu"«vf fcf-FikCE/
«c-i» vtB(,t>4CF m su*!1' SUIF -..'I JF »(••••'
IJ.SO "ISilTfcS "PtWAI ll.S, t'Ctt Viv«)F.l. H
Pk>»F
1.1000
Pt'.f TBAI |ON» J.nOk-Oi
._. ' vix|U|Ju PRESSURE DBHPs
r^
U1
OB I).SO U1NUT£S (IPERATION, CYCLE M'UBER S
•IHATIiIro <
OBHP*
1 .0000
).00£.n?
pwFSSMHf. OHQP* IJ'1.10
ltt.00 "IMITES iJft«»1H>N. CTCLt knv>H(H t>
*VFBAGt PHfSSuwE OHQPZ 105U.OO
S'SU" FL'.'KS 1.0000
-------
TABLE 27. PRESSURE DROP AND FRACTIONAL PENETRATION VERSUS TIME
(REFERENCE FIGURE 16)
CONvE»dFNCM
SUU"ARV T»rtlE 1 <
Tff
(*!«•)
0.01
•'.50
l.uv
1.50
2.' 50
i.uO
3.50
u.flfi
u|50
S.OO
5.50
6.00
6.01
6.50
7.00
7.50
A. 00
8. 01
8.50
9.00
9.50
10.00
10.01
10.50
11.00
11.50
12.00
12.50
13.00
13.50
11.51
lu.vlO
U.50
15.00
15.50
15.51
16.00
16.50
17.00
17.50
17.51
18.00
18, 5u
19.00
|9.->fl
10. %!
20.00
20.50
PBo"p'Bf
(•./V2)
1020.
1 ii 12 .
I2MO.
1373.
«!»•!
12V*1,
I2«*.
1329.
HOo!
1172.
1250.
1292.
77*1,
881.
1107.
1232.
1265.
7ou.
l»69.
1127.
1200.
1242.
756.
460.
023.
9v5.
965.
981 .
096.
1110.
1022.
|OSu.
1290,
1370.
1 391 ,
• 08.
9|9.
1208.
1208.
132ft.
787.
"<>5.
1 I7U.
1258.
1200.
773.
HV,
11"".
11 VtN/SU>"'»PrlPLOTlf 0 htSl'lTS/NU C(
Mtf.ilWif K".
l.7SOt-Ji
I ,euOt*0<
1.671^-15
il-i-t-"
u.'l5"-01
e. idt ~<<1
5.59af <'l
u K^o^E *0%
J QSVf *I'P
f • 00 f t "0 J
b • t b^t * *' 3
5.3«9fOJ
2,'IHf «02
u.eelE-OJ
7. 1 /Ot •" 1
6.3l3f03
5.500E-01
^.B(JUt-(IJ
u.oHlf-OJ
7.100E-0}
6.32« -03
5.520E-OJ
2.**50E*U2
«.67JE-ni
•i.('«6t-OJ
3.I9IE-I-3
2.708E "03
2.365E-OJ
2.1 lfcE-03
1 .928E-«1
1 . /80E«01
l.be'lE-01
3.713f-OJ
1.7u«jf03
3.52«E-03
}.bCnE»02
u.l 70E*^3
6.398E»03
5.6lOE*03
u.'i/Jt'Oj
2,>'57E*02
J.567E-U)
7,Oi'jE*01
B.l««6t-0>
5. 3-<»f.-( J
2.91 Jt"l"2
u,e)et • Jl
7.15U--J
-------
TABLE 27 (.continued)
SUMMARY TABLE 1
TI-E
( «• 1 •« »
21 .00
21.50
21.51
22.00
22.50
21.00
21. SO
21.51
2". no
2J.50
25.00
25.50
2b.OO
2b.50
27.00
27.01
27.5-0
26.00
28.50
29.00
29.01
29.50
10.00
10.50
11.00
11.01
11.50
12.00
12.50
11.00
11.01
11.50
lu.OO
1«. SO
15.00
15.01
15.50
16.00
Ib.SO
17. 00
17. Ul
17.50
18.00
(A SO
J ™ . j "
19.00
19.50
oO.OO
uo.5p
PRESSURE /K2
pntSS'ibE
UfcOP
(N/"2)
1229.
I2b2.
7b2.
867.
1 120.
120S.
123K.
751.
8S7.
<*|9.
901.
9bl.
977.
992.
1005.
1018.
1010.
1280.
1167.
lino.
805.
915.
1201.
1290.
1520.
781.
A90.
llbl.
1209.
1261.
7b9.
*7«.
1 lib.
1219.
1252.
75B.
Rbl.
1 1 10.
II9S.
1227.
7u«.
850.
911.
U I t
** J 3 •
952.
968.
9*1.
00ft.
GIvEfc/SUMKABviPLOTTEP "FSUl TS/SO CO^nl
f BACTfOMl
PEMThtTlK'.
b.292t-Pl
S.u»«-i'5
2.B77E-U2
J.bbdF-03
7.10SE-01
b. 2921-03
S.o9|f -U3
2,cuSt-ni
u.ebJt-UJ
U.0lbt-l>!
I.lb6f-01
2.68SE»03
2.304E-01
2.098E-03
l.911f -01
1.7b1>3
1 .6e7f-i.1
-------
Ul
1.00
8.00
15.00
24.00 32.00 40.00 48. OU
TIME (MINUTES)
Figure 17. Pressure versus time plot for Example 2
(Reference Figure 16).
-------
0BflG * 1
ABflG n 2
+BflG * 3
XBRG * 4
BRG « 5
16.00
24.00 32.00 40.00 48.00
TIME (MINUTES)
Figure 18. Individual compartment flow versus time plot for
Example 2 (Reference Figure 16).
157
-------
in
00
0.00
8.00
16.00
24.00 32.00 40.00 48.00
TIME (MINUTES)
Figure 19. Penetration versus time plot for Example 2
(Reference Figure 16).
-------
of the curves represent that portion of the system operation when all compart-
ments are on line. Figure 18 shows special traces called out by a SUMMARY
request that indicate the concurrent velocity-time distributions for each of
the five compartments. Ordinarily, the above data would be used for research
purposes.
The concurrent variations in dust penetration with time are shown for
Example 2 in Figure 19. Note that the maximum penetration values coincide
with the minimum pressure loss levels indicated on Figure 17. During those
time intervals when all compartments are on line, the penetration varies
inversely with pressure loss as should be expected.
Figure 20 data inputs reflect a time cycle operation in which the filter
user or designer has set the constraint that there be a specific, i.e.,
11 minute, time interval between successive compartment cleanings. In this
example, it is assumed that a K^ value is available for the dust of Interest
but for a different size spectrum and with measurement at a temperature and
velocity differing from that of the filter system. The input data summary
generated by the program for the Figure 20 input form appears in Table 28.
Calculated and/or corrected values for C±, K2, and Sg are given in Table 29.
It should also be noted that since AVERAGE data were requested, the average
pressure drop and penetration statistics alone are printed, Table 30.
An example of an incorrectly prepared data input card is shown in
Figure 21 so that the program response via diagnostic printout could be
demonstrated. The types of errors depict illegal values, redundancies,
contractions and omissions. Table 31 shows the input data summary that by
itself may alert the model user to the numerous Input errors and Table 32
indicates calculated and/or corrected values for relevant data inputs.
159
-------
TITLE
FABRIC FILTER MODEL-DATA INPUT FORM
ran
i«iiSpnni*m«
]33D3UHUfl"aj"
ijKPliy
i 1 1 1 n 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 u 1 1 1
ft 7 •
il iMnmi n i ill ill II nil i ill i n
mm \>wm \ mm 111 • 111 • i in • 111 u 11111111111111111 u 11111 n 111 n
I 111 • I lill II 1 11 I I I 1 I 1IIII 111!1 I I I I 1 I I I I I I 11 I I I I I I 1 11 I I i I I I I I I I I I! Ill
•u. one* otrma M*T w MMV jutnrco norr >oi inns o. si MB n.
Figure 20.
Fabric filter model
form for Example 3.
- data input
160
-------
TABLE 28. SUMMARY OF INPUT DATA FOR BAGHOUSE ANALYSIS
(REFERENCE FIGURE 20)
«••••••••••••••••••••••••*••••••••••»•••••••»••••••••••»••••••••••»•*»*•••»*•»••
3U>""«RV OF INPUT DATA FOB BAG»*OUSE ANALYSIS
T1MF.O/K2 CORRECTED FOR SIZE/AVERAGE RESULTS/
BASIC DESIGN DAT*
NUMBER OF COMPARTMENTS
COMPARTMENT CLEANING n«e
(OK LINE TIME)
CLEANING CYCLE UM£
TIME BEThEEN CLEANING CYCLES
REVERSE FLCK VELOCITY
OPERATING DATA
AVERAGE FACE VELOCITY
GAS TEMPERATURE
INLET DUST CONCENTRATION
MEASURED AT
FABRIC AND OUST PROPERTIES
SPECIFIC RESISTANCE, K2
MEASURED AT
CORRECTED 10
MMDI
MMD2
EFFECTIVE RESIDUAL DRAG, SE
MEASURED AT
RESIDUAL LOADING, »R
3.0
10.0
11.0
0.0
0.6100
ISO.
10.00
25.
0.75
50.
0.3000
9.0
350.
150.
35.0
MINUTES
"INUTES
MINUTES
W/MJN
DEGREES CENTIGRADE
G/M}
DEGREES CENTIGRADE
N-MIN/G-M
DEGREES CENTIGRADE
MICRONS
MICRONS
.STANDARD DEVIATION 3.00
'STANDARD DEVIATION ?,50
DEGREES CENTIGRADE
G/M2
SPECIAL PROGRAM INSTRUCTIONS
MAX NUMBER OF CYCLES MODELED 20
ACCURACY LEVEL 0
TYPE OF RESULTS REQUESTED AVERAGE /
-------
TABLE 29. INPUT VARIABLES CALCULATED BY PROGRAM
(REFERENCE FIGURE 20)
CALCULATED VALUES
INLET PUSt CONCENTRATION 7.0» 6/"l
CORRECTED Tf OPERATING TEMPERATURE
FABRIC AND OUST CAKE PBOPEBTIFS CORBECTER FOR GA$ VISCOSITY
SPECIFIC CAKE RESISTANCE* K£ 1.1" h'MJN/6-M
EFFECTIVE DRAG. SE 340,
FRACTIONAL AHEA CLEANEDf AC O.IS
TIME INCREMENT o.ss »INUTES
SVSTE1' CONSTANT «• 0.0
10
-------
TABLE 30. RESULTS OF BAGHOUSE ANALYSIS
(REFERENCE FIGURE 20)
•••••»••*••*•••• ••*t**»(t****
RESULTS OF BA&MOUSE ANALYSIS
T1MEO/K8 CORRECTED FOR SIZE/AVERAGE RESULTS/
•»•••••««•••*«**•••«••••«
FOR
20.81 MINUTES OPERATION. CYCLE kU«-BE« 10
AVERAGE PENETRATION*
AVERAGE PRESSURE DROP*
AVERAGE SYSTEM FLO««
M««IMUM PENETRATION!
MAXIMUM PRESSURE DROP*
2.«IE-OJ
S6«.«0 N/M;
0.6100
1.02E>02
797.9B
FOR
20.SI MINUTES OPERATION, CVCll NUMBER II
AVERAGE PENETRATION*
AVERAGE PRESSURE OROP>
AVERAGE SYSTEM FLO««
MAXIMUM PENETRATION*
MAXIMUM PRESSURE DROP*
2.00E-0)
562.87 N/H2
O.blOO U/MIN
I.02E-02
705.0* N/M2
FOR
20.8) «INUTES OPERATION. CYCLE NUMBER 12
AVERAGE PENETRATION*
AVERAGE PRESSURE DROP*
AVERAGE SYSTEM FLOm
MAXIMUM PENETRATION*
MAXIMUM PRESSURE DROP*
2.SOE-OJ
561.88 N/M£
0.6100 M/MJN
1.02E-02
7«J.08 N/M2
-------
The numerous errors in preparing the inlet format card, Figure 21, are
called out in the diagnostic messages of Table 33. The reader should recognize
that the likelihood of the indicated error count (hopefully) is extremely
remote. However, the summary of diagnostic messages provides some indication
of the model's capability to recognize poor programming.
164
-------
FABRIC FILTER MODEL-DATA INPUT FORM
v if • IB M
mr\ 11111111111111111M111II111] 11111
& ,11 i rari i j& i iw 'I
roiiiiii 1 1 1 1 n 1 1 1 1 1 1 1 1 1 1 1 1 1 n 1 1 1 1 1
Illil IIIIMIIIIIII11111II II Mil 11! Ill
^•••^•••••••••••^^••^•••
3]33"DD3D"""""""""-J—
1IIII Illllllilll Illlllh III III I 111 L 111
I* SO 31
m
MM JWTMB «MVI >ot ITOP o, x MO u
Figure 21. Fabric filter model - data input
form for Example A.
165
-------
TABLE 31. SUMMARY OP INPUT DATA FOR BAGHOUSE ANALYSIS
(REFERENCE FIGURE 21)
SUMMARY Qr INPUT DAT* fQO BAGxOuSE ANALYSIS
•••»••*••••••*•••••••••*••»»•*•••»•»•••*•••»*••••,••••*•••«•••••••••••»•••,«,»,«
ERROR MESSAGE TEST
DESIGN DAT*
NUMBFB Of COMPARTMCNTS
COMPARTMENT CLEANING TIME
tUFf LINE TlMfc)
CLEANING CYCLE TIME
TIMf BETWEEN CLtAKIKG CYCLES
LIMITING PRESSURE DROP
REVERSE FLO* VELOCITY
SHAKING
OPERATING DATA
AVERAGE FACE VELOCITY
GAS TEMPERATURE
INLET OUST CONCENTRATION
MEASURED AT
FABRIC AND DUST PROPERTIES
SPECIFIC RESISTANCE. *2
MEASURED AT
EFFECTIVE RESIDUAL DRAG. SE
MEASURED AT
RESIDUAL LOADING* »fi
RESIDUAL DRAG, SR
MEASURED AT
INITIAL SLOPE. KR
MEASURED AT
TO
10.0
5,0
17.0
56.
0.0
7.0
0.0030
0.
10.00
t.oo
as.
0.6100
10.
25.
*0.0
•0.
25.
0.0
25.
"tNuTF.5
"IM.TES
N/vj
CYCLES/SEC
DEGREES CENTIGRADE
DEGREES CENTIGRADE
DEGREES CENTIGRADE
DEGREES CENTIGRADE
G/M2
DEGREES CENTIGRADE
S«MIN/G-"
DEGREES CENTIGRADE
SPECIAL PROGRAM INSTRUCTIONS
**Ak NUMBER OF CYCLES "OOELED
ACCURACY LEVEL
TYPE OF RESULTS REQUESTED
10
FRACTIONAL AREA CLEANFOi AC 4.49
-------
TABLE 32. INPUT VARIABLES CALCULATED BT PROGRAM
(REFERENCE FIGURE 21)
CALCULATED VALUES
INLET DUST CONCENTRATION 10.92 G/«J
CORRECTED T(, OPERATING TEMPERATURf
FABRIC AND DUST CANE PROPERTIES CORRECTED FOR GAS VISCOSITY
SPECIFIC CAKE RESISTANCE! K? o.o] N.UIN/C-M
EFFECTIVE DRAG. SE 56. *.u|N/M]
»ESinu»L DRAG. SR J7.
FRACTIONAL AREA CLEANED, AC 9.99
TIME INCREMENT 0.00 "INDIES
SYSTEM CONSTANT «« 0.0 G/U2
-------
TABLE 33. DIAGNOSTIC MESSAGES (REFERENCE FIGURE 21)
DIAGNOSTIC MESSAGES
ILLEGAL REQUEST FOR TYPE OF RESULTS
S
THF NU"BER OF COMPARTMfMT3 «UST SOT EXCEED 50
THE NUMBER OF COMPARTMENTS M*"ES THE COMPARTMENT CLEANING TIKI "UST 8E LESS THAN THE CLEANING CYCLE TIME
THE COMPARTMENT CLEANING Ti-E "I'ST BE LESS THAN THE TOTAL CYCLE UMJ
T1»E INCREMENT TOt S««ALL, IE. < O.OJ MINUTES
AVERAGE FACE VELOCITY OUT OF RANGE* o.j TP 3,0
., A GAS TEMPERATURE HAS NOT BEEN ENTERED
CO INVALID FREUUENC» OR AMPLITUDE FOR SHAKER
INVALID ACCURACY CODE
BOTH TIMED AND PRESSURE CONTROLLED CLEANINGS SPECIFIED - ONLY ONE IS VALID
PARTICLE SIZE DATA FOR K2 ARE INCOMPLETE
MASS MEDIAN DIAMETER OF MEASUREMENT OUT OF RANGE a TO so MICRONS
STANDARD DEVIATION OF MEASUREMENT OUT OF RANGE 2 TO 0
MASS MEDIAN DIAMETER OF OUST OUT OF RANGE i TO So MICRONS
STANDARD DEVIATION OF DUST OUT OF RANGE 2 TO U
BULK DENSITY CANNOT EXCEED DISCRETE PARTICLE DENSITY
• INCOMPLETE DATA FOR NON-LINEAR DRAG MODEL
INITIAL SLOPE • KR , is MISSING
FRACTIONAL AREA CLEANED OUT OF RANGE,« TO I
THE PROGRAM HAS btEN TERMINATED BECAUSE OF ERRORS IN THE INPuT DATA
-------
TECHNICAL REPORT DATA
(Pleat read Instructions on the reverse be fort completing)
1. REPORT NO.
EPA-600/7-79-043a
2.
3. RECIPIENT'S ACCESSION NO.
«. TITLE AND SUBTITLE
Fabric Filter Model Format Change; Volume I.
Detailed Technical Report
6. REPORT DATE
February 1979
8. PERFORMING ORGANIZATION CODE
T. AUTMOR(S)
Richard Dennis and Hans A. Klemm
I. PERFORMING ORGANIZATION REPORT NO.
GCA-TR-78-51-G(2)
9. 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 277U
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 11/77 -12/78
14. SPONSORING AGENCY CODE
EPA/600/13
«. SUPPLEMENTARY
2925.
pr0ject officer
Turner MD-61, 919/541-
ie. ABSTRACT
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.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
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
8. DISTRIBUTION STATEMENT
Unlimited
IB. SECURITY CLASS (This Report)
Unclassified
21. NO. Or PACES
179
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
•PA form 2120-1 (i-73)
169
------- |