o
        SYSTEMS EVALUATION OF REFUSE
           AS A LOW SULFUR FUEL

           A FINAL REPORT TO THE


       ENVIRONMENTAL PROTECTION AGENCY

                 VOLUME I
       REPORT NO. F-1296
NOVEMBER 1971
   FOSTER WHEELER CORPORATION
Cottrell Environmental Systems
A Division of Research-Cottrell

-------
SYSTEMS EVALUATION OF REFUSE
AS A LOW SULFUR FUEL
A Report to
ENVIRONMENTAL PROTECTION AGENCY
CONTRACT CPA 22-69-22
AUTHORS
For Envirogenics Company:
R. M. Roberts
S. T. Braunheim
R. C. Hanson
S. B. Kilner
For Foster Wheeler Corporation:
R. E. Sommerlad
J. D. Shenker
R. W. Bryers
For Cottrell Environmental Systems, Inc.:
A. P. Konopka
J. McKenna
November 1971

-------
ACKNOWLEDGEMENT
The work reported here reflects the efforts and cooperation of a
considerable number of people. Particular acknowledgement, how-
ever, should be made of the guidance and counsel furnished by two
key individuals. These are Mr. R. C. Lorentz, who served as the
EPA Project Officer, and Mr. E. M. Wilson, the Head of Envirogenics
Company I s Air Pollution Control Department.
Special mention of noteworthy contributions of other individuals will
be found on various' pages of this report. It would be improper,
however, not to cite at once the outstanding work done by the Project
Engineers of the subcontractor firms. At Foster Wheeler Corp. ,
this worker was Mr. R. E. Sommerlad, who furnished some of the
most important data developed on this study. At Cottrell Environmental
Systems, Inc., the key inputs there resulted from the efforts of
Mr. A. P. Konopka.
The senior author also wishes to expres s his gratitude to the
organizations furnishing their services in the final review of this
document. In addition to the Project Office itself, these comprised
the Tennessee Valley Authority and the (then) Office of Solid Waste
Management of EPA. The thoughtful comments and recommendations
of the various reviewers proved most useful.
R. M. Roberts
Senior Author
Envirogenics Company
ii

-------
ABSTRACT
Because of the S02 problem, coal is being displaced as a power
plant fuel in certain areas by other, more cos tly fos sil fuels of
lower sulfur contents. Urban refuse can a:lso be considered a coal
s'ubstitute because, even with mass/energy relationships considered,
sulfur inputs from refuse should be much lower than from equivalent
amounts of coal. The questions are of course can practical, refuse-
fired steam generators be built for base load utility service, and can
this process be made competitive with other, presently preferred
practices of wa.ste disposal? The work described here was addressed
to the acquisition of answers to these questions, which, in each case
was positive. This involved a systematic assessment of the fuel
properties of refuse and of the mechanics and. combustion technology
associated with the utilization of refuse as a fuel in generating utility
grade steam. By estimating the inventories and compositions of
refuse that would likely occur, the extent of S02:-abatement that might
be realized in using refuse as a partial coal substitute was projected
through the year 2000. .
Determining the us eful energy from refuse involved the as sodated
task of establishing for this fuel its behavior ,i.n and compatability
with furnace structures. This required the definition of the various
energy utilization opportunities for which refuse might be suited.
With emphasis selectively aligned to power plant applications, the
relevant technology and state-of-the-art (particularly in Europe)
of power generation with refuse, other waste fuels, and non- bituminous
solid fos sil fuels were documented. In this connection, pertinent
processes and hardware used in the handling and conditioning of
refuse were also reviewed. Criteria were then established for firing
refuse in utility-class boilers.
A catalog of ten different combined-fuel (coal + refuse) fired boiler
configurations was conceived and then .analyzed in terms of process
variables (plant power capacity, fuel- ratio, etc.); performance / cost
characteristics were also predicted. Similarly treated were five
plans for modifying existing plants to refuse-burning systems. At
leastone of the systems within the first group was identified as being
a more cost-effective approach to refuse disposal than is landfill at
the present time. This advantage may well also be implicit in other of
the candidate systems but, due to the limitations of the analysis, this
cOllld not be resolved. One limitation was that cost data on. 'refuse
processing and performance information on furnaces firing shredded
refuse in suspension were la cking, this neces sitated the adoption of
conservative costs and design specifications for certain components
examined in the analysis.
iii

-------
Another limitation was that the comparative refuse transportation costs
for landfill disposal and energy-recovery installations could not be
developed. It is highly likely that, in many of the metropolitan areas
of the U. S., the combined costs of transportation and disposal will be
lower than for existing and/or 'projected landfill operations. A third
limitation was that the conclusions derived could not be expected to
maintain their validity in the dynamic economic situation that has
prevailed. The July 1969 cost bases used are probably already quite
obsolescent, but because fuel costs are escalating at a much greater
rate than capital costs, the cost effectiveness of the analyzed technique
has probably significantly increased.
These limitations not withstanding, the cost model was developed to
consider all the major elements involved in the erection and operation
of the candidate refuse-burning systems. From this systematic analysis,
two new-plant configurations were extracted and subjected to detailed
engineering analysis. Cost estimates were iteratively computed for the
resulting preliminary designs. Where technical knowledge-gaps were
recognized, research and deve1opn1.ent requirements were itemized.
These were analyzed and integrated into dis crete task packages.
Final organization of these R&D elements took the form of two suggested
five year plans, each structured to accommodate a specific level of
effort.
iv

-------
TABLE OF CONTENTS
ABSTRACT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. SUMMAR Y AND CONCLUSIONS. . . . . . . . . . . . . . . .
A. INTRODUCTION.....................
B. PROGRAM OBJECTIVES AND APPROACH. . . . . . . .
1. Project Scope. . . . . . .. . . . . . . . . . . . .
a. Assessment of Refuse as a Fuel. . . . . . .
2.
Assessment of Power Plant Designs. . . . .

Development of Recommendations and
Formulation of R&D Plans. . . . . . . . . .

Information and Data Sources
b.
c.
. . . . . . . . . .
C.
3. Report Organization. . . . . . . . . . . . . . . .
CONCLUSIONS AND RECOMMENDATIONS. . . . . . .


1. Refus e as a Fuel. . . . . . . . . . . . . . . . . .
a.
Quantity-Quality Considerations. .
. . . . .
b.
Reduction in Pollutant Emissions
. . . . . .
c.
Processing Technology. . . . . . . . . . . .
Thermal Utilization. . . . . . . . . . . .
d.
2.
Power Plant Designs. . . . . . .
. . . . .
. . . .
a.
Engineering Criteria for Candidate Systems.
b.
Economic Model. . . .
. . . . . . . . . . .
3.
c. Detailed Design and Cost Studies. . . . . .
Recommended Research and Development. . . . .
a.
Criteria
. . . .
. . . . .
. . . .
. . . . .
b.
Plan A
Plan B
. . . . . . . .
. . . . . . . . .
. . . . .
. . . .
c.
. . . . .
. . . . .
II.
REFUSE AS A FUEL. . . . . . . . . . . . . . . . . . . . . p.
A. QUANTITIES AVAILABLE. . . . . . . . . . . . . . . .
1. Collection Rates. . . . . . . . . . . . . . . . . .
a.
United States
. . . . .
. . . .
. . . . . . . .
b.
Selected Metropolitan Areas.
. . . . . . . .
v
Page
No.
iii
I-I

I-I

1-3

1-3

1-3

1-4
1- 5
1-5
1- 7
1-8
1-8
1-8

1-9
I-II
1-12

1-12

1- 12

1-18

1-19
1-24
1-24
1-26
1-27
II-I

II-I

II-I
II-I
II-7

-------
B.
D.
TABLE OF CONTENTS - Continued
2.
Refuse Composition. . . . . . . . . . . . . . . . .
a. Analysis of Current Refuse. . . . . . . . .
b. Projected Composition. . . . . . . . . . . .
c. Seasonal Variation. . . . . . . . . . . . . .


d. Heat Value. . . . . . . . . . . . . . . . . .

COMBUSTION AND INTERACTIONS. . . . . . . . . . .


1. Introduction.....................

2. Gaseous Combustion Products. . . . . . . . . . .
a.
Major Products
Pollutants. . .
. . . .. .. .. .. .. .
. .. .. .. . .. . . . .
. .. .. .. . .
b.
.. .. .. .. . . .
3.
Solid Combustion Products.
. . .. .. .. ..
.. . . .. . ..
a.
Residue. . . . . . . . . . .
. . .. ..
. . . . .
C.
b. Fly Ash and Suspended Particulates. . . . .
c. Ash Deposits and Corrosion. . . . . . . . .
d. Ash Fouling Tendencies. . . . . . . . . . .
POTENTIAL POLLUTANT REDUCTIONS. . . . . . . .
1.
2.
Current Incinerator and Power Plant Emissions. .
3.
Projected Reductions through Use of Refuse as a

Fuel. . . . . . . . . . . . . . . . . . . . . . . . .

Effect of Refuse Firing on S02 and Particulate

Emissions. . . . . . . . . . . . . . . . . . . . . .
HANDLING AND PROCESSING. . .
. . .. .. . ..
. .. .. . ..
1.
2.
Introduction. . . . . . .
Size Reduction
.. .. .. ..
.. .. .. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. ..
.. .. .. .. .. ..
.. .. .. .. ..
a.
Equipment Characteristics. . . . . . . . . .


Test Work. . . . . . . . . . . . . . . . . .

Effect of Particle Size on Combustion
b.
c.
.. .. .. ..
3.
Refuse Transport. . . . . . . . . . . . . . . . . .


a. Background..................
b.
Pipeline Transfer of Refuse
.. .. .. .. .. .. .. .. ..
vi
Page
No.
II-9
II-9
II- 12
II- 12
II- 15
II-22
II-22
II-23
II- 23
II-25
II- 3 7
II- 37
II- 3 7
II- 38
II- 40
II-44
II-44
II-47
II- 5 1
II- 5 5
II- 5 5
II- 5 6
II-56
II- 58
II- 61
II- 6 3
II- 63
II- 64

-------
TABLE OF CONTENTS - Continued
4.
Separation Processes.
. . . . . . . . . . . . . . .
a.
Background. . . . . . . . . . . . . .
Ferrous Metals. . . . . . . . . . . .
No"n-Ferrous Metals. . . .
. . . .
. . . .
b.
c.
. . . .
. . . . .
E.
d. Air Clas sification ..............
ENERGY UTILIZATION. . . . . . . . . . . . . . . . . .
1. Turbo-electric Generation. . . . . . . . . . . . .
a. Rankine Cycle Systems. . . . . . . . . . . .
b. Gas Turbine Systems. . . . . . . . . . . . .
Other Applications. . . . . . . . . . . . . . . . .
a. Industrial and District Heating. . . . . . . .
2.
III.
b. Desalination and Miscellaneous Applications

POWER PLANT DESIGN CONSIDERATIONS. . . . . . . . . .

A. STATE OF THE ART SURVEY OF FOSSIL FUEL AND
REFUSE-FIRED BOILERS. . . . . . . . . . . . . . . .
B.
1. Fossil Fuel Boilers. . . . . . . . . . . . .
2. Refuse Installations. . . . . . . . . . . . .
CATALOG OF CANDIDATE SYSTEMS. . . . . .
1. Introduction.................
. . . .
. . . .
. . . .
. . . .
a.
Fuel Value
. . . . .
. . . . . . . . .
. . . .
b.
Refuse Charging Rate. . . . . . .
Unit Capacity. .'. . . . . . . . . .
. . . . .
. . . . .
c.
d.
Steam Conditions. . . .
. . . . .
. . . . . .
e.
Other Operating Conditions
. . . . .
. . . .
, ,2.

)
Performance Parameters
. . . .
. . . .
. . . . .
a.
Energy Requirements. . . . . . . . . . . .
Combustion Calculations. . . . . . . . . . .
b.
3.
Operating Parameters. . . . . . . . . . . . . . .
a. Fuel Requirements. . . . . . . . . . . . . .
b.
Heat Rate.
. . . . . . . . .
. . . . .
. . . .
c.
Steam Flow. . .
. . . .
. . . . . . . . . . .
vii
Page
No.
II- 6 5
Il- 65
II- 66
II- 67
II- 6 7
II- 68
II- 68
II- 68
II- 68
Il- 69
II- 69
II- 72
IIl- 1
III- 1
IIl- 1
III- 2
IIl- 2
IIl- 2
III - 3
IIl- 4
IIl-4
IIl- 4
IIl- 5
IlI- 5
IIl- 5
III - 1 0
IIl- 1 6
III - 1 6
IIl- 2 0
IIl- 2 0

-------
TABLE OF CONTENTS - Continued
4.
Design Data for Candidate Systems. . . . . . . .


a. Background..................

b. Candidate System Characteristics. . . . . .
Design Data for Retrofit Systems. . . . . . . . .


a. Background..................

b. Selection of Existing Units. . . . . . . . . .
5.
c.
Modification Requirements
. . . . . . . . .
6.
d. Candidate Retrofit Systems. . . . . . . . .
Air Pollution Control (APC) Equipment. . . . . .
a.
Overview. . .
. . . . . . . . . .
. . . . . .
C.
b. APC System Possibilities. . . . . . . . . .
Refuse Handling System. . . . . . . . . . . . . .
a. Design Selection. . . . . . . . . . . . . . .
b . Weighing Stations. . . . . . . . . . . . . .
c. Truck Unloading. . . . . . . . . . . . . . .
d. Recei ving Pit. . . . . . . . . . . . . . . . .
e. Oversize Material Separation and Reduction
£. Storage Facilities. . . . . . . . . . . . . .
EVALUATION OF SELECTED ENGINEERING DESIGNS


1. Cost Model. . . . . . . . . . . . . . . . . . . . .
7.
a.
As sumptions
. . . . . .
. . . . . . . . . . .
b.
Cost Model Structure
Cost Model Results
. . . . . . .
. . . . . . .
. . . . .
c.
. . . . . .
2.
d. Cases Selected for Optimization. . . . . . .
Optimization of Selected New Plant Designs. . . .
a. Introduction.................
b. Separately Fired Economizer. . . . . . . .
c. Arch Furnace with Separate Superheater. .
Retrofit Plants. . . . . . . . . . . . . . . . . . .
3.
viii
Page
No.
III - 2 5
1II-25
III - 26
III- 51
III - 5 1
III- 54
III-60
III- 61
III-70
1II-70
III-73
III-79
1II-79
III- 80
III-82
III- 83
III - 83
III- 85
1II-90
III- 90
III - 9 2
III- 93
III - 9 5
III- 111
III- 112
III- 112
III-114

III-138

III-156

-------
IV.
TABLE OF CONTENTS - Continued
RECOMMENDA TIONS . . . . . . . . . . . . . . . . . . .
A. IMMEDIATE TECHNOLOGICAL OPPORTUNITIES
B. SUGGESTED RESEARCH AND DEVELOPMENT PLANS.


1. Overview......................
Plan A
. . . . . . . . . . .
. . . . . .
. . . . . .
Central Research Facility. . . . . . . . . .

Program 1 - Combustion Optimization Studies:
Basic Program. . . . . . . . . . . . . . . .

Program 2 - Corrosion and Fouling. . . . .

Program 3 - Systems Analysis of Alternative
Thermal Conversion Process for Refuse. .
Program 4 - Particulate Control. . . . . .
Program 5 - Small Refuse-Fired Boilers
Program 6 - Sub-Scale Component Testing
Program 7 - Flow Modeling. . . . . . . . .


Plan B .......................

Program 1 - Combustion Optimization Studies:
Expanded Program. . . . . . . . . . . . .. .

Program 8 - Evaluation of Refuse Pyrolysis
and Gasification Processes in Connection
with Steam Generation. . . . . . . . . . . .
Program 9 - Evaluation of Refuse Combustion
in Fluidized Beds. . . . . . . . . . . . . .

Program 10 - Engineering Manual for the
Conversion of Boilers to Combined Firing
Program 11 - Design of Demonstration Unit
Program 12 - Construction of Demonstration

Plant. . . . . . . . . . . . . . . . . . . . .
ix
Page
No.
IV -1
IV -1
IV -1
IV -1
IV-2
IV-2
IV-3

IV-5
IV -6
IV-6
IV-7
IV-8
IV -8
IV - 1 0
IV - 1 0
IV - 1 3
IV - 1 5
IV - 1 8
IV - 1 9
IV - 2 1

-------
.
-------..-----
Table
No.
I-I
II-I
II- 2
II-3
II-4
II-5
II-6
II-7
II-8
II-9
II- 10
II- 11
II- 12
II- 13
II- 14
II- 15
TABLES
Surnm.ary of System Characteristics of Refuse-Firing Steam
Generator Designs Over the Range 100 to 500 MW . .. . . .
Summary of Refuse Collection Rate Estimates
. . . .
Estimates of Future Urban Refuse Collections in the U. S.
Large Metropolitan Areas Having Severe S02 Pollution

Problems. . . . . . . . . . . . . . . . . . . . . . . . . . .
Population Projections for Six Selected Cities
. . . . . . .
Projected Quantities of Refuse Collected in Six Selected

Ci tie s . . . . . . . . . . . . . . . . . . . . . . . . . .
Projected Compositional Changes in Mixed Municipal

Was te s . . . . . . . . . . . . . . . . . . . . . . . . .
Ultimate Analyses of Fly Ash Samples from Refuse and
Combined Firing Installations. . . . . . . . . . . . . . . .
Sulfur Balances in German Refuse-Fired Steam Generators
HCI Emissions, Munich North II
. . . . . . . . . . . . . .
Maximum Refuse Quantity for 0.4% Alkali Content in Mixed

Fuel. . . . . . . . . . . . .' . . . . . . . . . . . . . . . .
Fusion Temperatures of 90% Coal Ash/1 0% Refuse Ash

Mixtures. . . . . . . . . . . . . . . . . . . . . . . .
Projected Fossil Fuel Displacement by Urban Refuse

Energy. . . . . . . . . . . . . . . . . . . . . . . . .
Projected Emissions Reduction by Coal Displacement with

Refuse. . . . . . . . . . . . . . . . . . . . . . . . . . . .
Refuse Composition of Hammermill Input at Two Areas

Sampled. . . . . . . . . . . . . . . . . . . . . . . . . . .
Particle Size Distribution of Outputs from Two Different
Types of Hammermills . . . . . . . . . . . . . . . . . . .
x
Page
No.
-
1-16
II-2
II-6
II-8
II- 10
II- 11
II- 13
II- 29
II- 31
II- 36
II- 43
II-45
II- 49
II- 53
II-59
II- 60

-------
Table
No.
II- 16
llI- 1
Ill-2
III- 3
Ill- 4
Ill- 5
Ill- 6
Ill- 7
Ill-8
III- 9
Ill-lO
III - 1 1
Ill- 1 2
llI- 13
Ill- 1 4
TABLES - Continued
Calculated Burning Times for Various Fractions in Milled

Refus e . . . . . . . . . . . . . . . . . . .
Products of Refuse Combustion
. . . . .
Refuse Fired Stearn Generator Efficiencies
. . . . .
Calculation Procedure for Determining Stearn Generator
(S. G. ) Characteristics ". . . . . . . . . . . . . . . . . . .
Case 1 - Separate Furnace: Calculated System
Characteristics. . . . . . . . . . . . . . . . .
Case 3 - Separately Fired Economizer:
System Characteristics. . . . . . . . .
Calculated
. . . . . .
. . . .
Case 4 - Separate Fossil Fuel Superheater (Saturated
Stearn Input): Calculated System Characteristics. .
Case 5 (Revised) - Separate Fossil Fuel Superheater
(Partially Superheated Steam Input): Calculated System
Characteristics. . . . . . . . . . . . . . . . . . . . . . .
Case 9 - Arch Furnace: Calculated Stearn Characteristics
Case 10 - Separate Fossil Fuel Superheater:
System Characteristics. . . . . . . . . . . .
Ca1cula ted
. . . .
Summary of Performance Data for Selected Steam

Generators. . . . . . . . . . . . . . . . . . . . .
. . . .
Calculated Operating Characteristics of Recommended

APC Systems ." . . . . . . . . . . . . . . . . . . . . . . .
Solid Waste/Fossil Fuel Power Plant (Separately Fired

Economizer) .."......................
Costs of a Solid Waste/Fossil Fuel Power Plant
(Separately Fired Economizer). . .". . . . . . .
. . . . .
Ten Stearn Generator Designs Submitted to Cost Model

Analysis. . . . . . . . . . . . . . . . . . . . . . . . . . .
Xl
Page
No
-
II-62
Ill- 12
Ill-13
llI- 1 7
llI- 28
Ill-32
Ill- 35
Ill.:. 41
III - 5 0
III - 5 3
III - 5 9
llI- 78
III-96
Ill- 9 7
III - 9 9

-------
Table
No.
III-15
IIl-16.
Ill- 1 7
Ill- 18
III- 19
III - 2 0
III - 2 1
1II-22
III-23
III-24
Ill- 2 5
Ill- 26
TABLES - Continued
Products of Combustion of Various Excess Air Levels. . .
Efficiency of Coal-Fired Steam Generator
. . . . .
. . . .
Efficiency of Refuse-Fired Economizer at Various
Excess Air Levels. . . . . . . . . . . . . . . . . .
. . . .
Efficiency of Refuse-Fired Economizer at Various Excess

Air Levels. . . . . . . . . . . -. . . . . . . . . . . . . . .
Efficiency of Refuse-Fired Economizer at Various Flue-
Gas Exit Temperatures. . . . . . . . . . . . . . . .
Summary Performance of Refuse-Fired Economizer
Summary Performance of Coal-Fired Steam Generator
with External, Refuse-Fired Economizers. . . . . . . . .
Revised Costs for the Separately Fired Economizer Plant.
Summary Performance of Arch Furnace Steam Generator.
Summary Performance of Coal-Fired Superheater and
Reheater (Arch Furnace System) . . . . . . . . . . .
Revised Costs for the Arch Furnace Plant. .
. . . .
Costs for Retrofit Systems
. . . . . . . . . . .
. . . . . .
xu
Page
No.
III-121
IIl-122
III-125
III- 126
III- 127
III-132
III-136
III-137
III-147
III-152
IIl-155
IIl-158

-------
Figure
No.
II-I
II- 2
II-3
II-4
II- 5
II-6
III - I
III- 2
III - 3
III - 4
III- 5
III - 6
III- 7
III-8
III- 9
III - 1 0
III - 11
III-12
III - I 3
III-14
FIG URES
Projected Compositional Changes of Mixed Municipal

Refuse. . . . . . . . . . . . . . . . . . . . . . . . . . . .
Seasonal Variations in Waste Load.
. .. . .. .. . . ..
.. .. .. ..
Projected Heating Value of Mixed Municipal Refuse
(Moisture and Ash Free) . . . . .. . . . . . . . .
II .. .. ..
Seasonal Variation in Heating Value. . .
.. .. .. .. .. .. .. .. ..
S02 and HCI in Dry Flue Gas, Munich North II Boiler
Raw Energy Consumption for the Generation of Electricity
in the United States. . . . . . . . . . . . . . . . . . . . .
Net Plant Heat Rate for Various Steam Conditions
.. .. .. ..
Turbine Heat Input for All Selected Steam Conditions. . .
Turbine Heat Input for Steam Conditions of Nonreheat

Units. . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fuel-Input for Plants up to 350 MW
.. .. .. .. ..
.. .. .. .. .. .. ..
Fuel Input for Plants up to 500 MW
.. .. .. ..
.. .. .. ..
.. .. .. ..
Effect of Refuse Heat-Input on Net Plant Heat Rate
.. .. .. ..
Feedwater Inlet Temperature
.. .. .. .. .. .. .. .. .. .. ..
.. .. .. ..
Steam Flow as a Function of Net Steam Generator Duty. .
Case I - Separate Furnace.
.. .. .. .. .. .. ..
.. .. .. .. ..
.. .. .. ..
Case 2 - Combined Furnace
.. .. .. ..
.. .. .. .. .. ..
.. .. .. .. .. ..
Ca s e 3
Separate Refuse-Fired Economizer
.. .. .. .. .. .. ..
Case 3 - Temperature Conditions for the 2400 psig/ 10000F

Cycle. . . . . . . . . . . . . . . . . . . . . . . . . . . .
Case 4 - Saturated Steam Unit with Separate Superheater
Case 4 - Temperature Conditions for the 2400 psig/ 1 OOOoF /

10000F Cycle. . . . . . . . . . . . . . . . . . . . . . . .
xiii
Page
No.
II- 14
II- 16
II- 20
II- 21
'II- 33
II- 5 0
III - 6
III - 7
III- 8
III - I 8
III- 19
III - 2 1
III - 2 3
III- 24
III - 2 7
III- 29
III- 31
i11-33
III-36
III- 37

-------
Figure
No.
Ill- 15
Ill-16
Ill-17
Ill-18
Ill-19
Ill-20
I11-2l
I11-22
Ill-23
Ill-24
Ill-25
Ill-26
I11-27
Ill- 28
Ill- 29
Ill- 30
Ill-3l
I11-32
Ill-33
Ill- 34
FIGURES - Continued
Case 5 - Temperature Conditions for the 2400 psig / 1 OOooF /

10000F Cycle. . . . . . . . . . . . . . . . . . . . . . . . .
Case 5 - Partial Superheat Unit with Separate Superheater
Case 6 - Suspension Furnace
. . . .
. . . .
. . . . . . . .
Case 7 - Spreader Stoker
. . . . .
. . . .
. . . .
. . . . .
Case 8 - Slagging Furnace.
. . . . . .
. . . . . . .
. . . .
Arch Furnace Fuel Input .
. . . . . . . .
. . . . .
. . . . .
Case 9 - Arch Furnace
. . . .
. . . . . .
. . . . .
. . . .
Anthracite Coal Burner Used in Arch Furnaces
. . . . . .
Case 10 - Arch Furnace with Separate Superheater
. . . .
Existing Steam Generator for Modification No.1. .
. . . .
Existing Steam Generator for Modification No.2.
. . . . .
Existing Steam Generator for Modifications No.3 and 4 . .
Existing Steam Generator for Modification No.5.
. . . . .
Boiler Modification No.1
. . . . . . . . . . . . .
. . . . .
Boiler Modification No.2
. . . . . . . . . . . . .
. . . . .
Boiler Modification No.3
. . . . . . . .
. . . . .
. . . . .
Boiler Modification No.4. .
. . . .
. . . .
. . . . . . . .
Boiler Modification No.5
. . . . . . . . .
. . . . .
. . . .
Boiler Modification No.5 - Detail of Division Wall/Hopper

Intersection. . . . . . . . . . . . . . .. . . . . . . . . . .
Recommended APC System for Combined S02 - Particulate

Removal. . . . . . . . . . . . . . . . . . . . . . . . . . .
xiv
Page
No.
Ill-38
Ill- 40
Ill-42
III - 44
I11-45
I11-47
Ill-48
Ill- 49
Ill- 52
Ill-55
III- 56
III- 57
Ill- 5 8
I11-62
III- 63
III-65
III- 67
III- 68
III-69
III- 76

-------
Figure
No.
III - 3 5
Ill- 36
IlI-37
Ill- 38
I11-39
IlI-40
III - 41
I11-42
. - IIl-43
I11-44
I11-45
I11-46
I11-47
I11-48
I11-49
Ill-50
Ill-51
Ill-52
. '.
FIGURES - Continued
Recommended APC System for Category II Plants. .
Waste Handling Systems for Different Type Furnaces. .
Receiving Pit Layout
. . . .. . ..
. . . .
. . . . .
. . . .
Low Profile Refuse Storage Pit
. . . . .
. . . . .
.. . .. .
High Profile Refuse Storage Building. '. . . . . .
. . . ..
Refus e Storage Silo
. .. .. .. .. .. .. . . ..
. .. . ..
.. . . .. .. .
Disposal Cost as a Function of Plant Capacity - Case 2 .
Disposal Cost as a Function of Refuse Fraction - Case 2
Net Disposal Cost Versus Waste Load - Cases 2, 3, 4,

and 1 0 . . . . . . . . . . . . ... . . . . . . . . . . . . .
Net Disposal Cost Versus Waste Load - Cases 1, 5, 6,

and 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Transportation Cost as a Function of Waste Load
.. .. . ..
Effect of Hauling Cost on Optimum Plant Size
.. . . . . .
Optimum Net Total Disposal Cost Versus Waste Load
(St. Louis Area) . . . . . . . . . . . . . . . . . . . . . :
Optimum Net Total Disposal Cost Versus Waste Load (New
(New York Area). . . . . . . . . . . . . . . . . . . . . .
Fuel Requirements for Combined Firing (Blended Flue

Gases) .' . . . . . . . . . . . . . . . . . . . . . . . . . .
Steam Cycle Schematic - Separately Fired Economizer.
Plant Fuel Requirements with Separately Fired

Economizers. . . . . . . . . . . . . . . . . . .
.. .. . .. ..
Effect of Flue Gas Exit Temperature on Steam Generator

Efficiency. . . . . . . . . . . . . . . . . . . . . . . . .
xv
Page
No.
III - 7 7
IIl- 8 1
Ill- 84
Ill- 86
III - 88
llI- 91
IlI-100
Ill- 101
IlI- 102
III-I03
IIl-I05
IIl- 106
IlI-I07
III -' 1 0 8
IIl-l 13
IlI- 115
IIl- 116
III-123

-------
Figure
No.
Ill- 5 3
Ill- 54
Ill-55
Ill-56
Ill- 5 7
Ill-58
Ill- 5 9
Ill-60
Ill-6l
Ill- 62
IIl-63
Ill-64
Ill- 6 5
Ill- 66
FIGURES - Continued
Effect of Flue Gas Exit Temperature on Efficiency of
Refuse-Fired Economizer. . . . . . . . . . . . . . . . .
Separately Fired Economizer. . .
. . . . . . .
. . . . .
Flow Diagram - Refuse-Fired Economizer.
. . . . . . .
Coal Fired Stearn Generator
. . . . . . . . . .
. . . . .
Flow Diagram - Coal Fired Stearn Generator
. . . . . .
Stearn Cycle Schematic - Arch Furnace Plant
. . . . . .
Fuel Requirements for Combined Firing (Blended Flue
Gases) at Higher Refuse Rate. . . . . . . . . . . . . . .
Fuel Requirements for Arch Furnace Plant
. . . . . . .
Combined- Fired Arch Furnace for 100 MW Plant
. . . .
Flow Diagram - Arch Furnace Stearn Generator.
. . . .
Detail of Arch Section of Anthracite Furnace. . .
. . . .
Coal- Fired Superheater for 100 MW Arch Furnace Plant
Flow Diagram - Separat~ Superheater and Reheater of
Arch Furnace Plant. . . . . . . . . . . . . . . . . . . .
Coal-Fired Reheater for 100 MW Arch Furnace Plant. .
xvi
Page
No.
IIl-l28
Ill-l30
Ill-l3l
Ill-l34
Ill-l35
Ill-l39
Ill- 140
Ill- 142
IIl-144
llI-l45
Ill-148
Ill-ISO
Ill- 15 1
llI-154

-------
1.
SUMMAR Y AND CONCLUSIONS
A.
INTRODUCTION
I
I'
I
At the present time, urban refuse is being collected in the
U. S. at a rate of almost 200 million tons per year. Predictions, variously
derived, suggest that this output will be doubled or even tripled by the end
of ,the century.
The current preferred method of disposal is sanitary landfill,
although less acceptable forms of refuse dumping are still being widely
practiced. Yet even today some American cities are already on the verge
of exhausting fill-tracts that are within reasonable logistic reach of their
waste collection networks. As landfill costs increase, due to site remote-
nes s or high costs of nearby real estate, and surpas s incineration costs,
shift-over to the latter is indicated. But converting over to incinerative
disposal is often regarded with reluctance because it (1) is an expensive
proposition on the short term, (2) involves a variety of interfaces with
local air quality standards, and (3) affords only volume reduction, the
residues from which must still go to landfill. The net managerial action
may thus tend to be one of continuing to absorb higher and higher disposal
costs until incineration becomes unavoidable. Fortunately, this type of
quandary is not yet a widespread one.
High costs and concern over 'the introduction of new sources
of air pollutants need not relegate refuse-incineration to the position it
now suffers. Refuse can be succes s,fully incinerated in the furnace of a
power boiler. Such a process not only produces power credits and thus
reduced disposal costs, but can lower S02 levels. This is because refuse
has a much lower sulfur content than does most American coal. The
concept is hardly new. It has been extensively applied throughout Europe
and elsewhere over the past decade. It is now being introduced to
North America at a number of points. The results obtained will doubt-
less be followed with great interest over the next several years.
Because of the importance that the refuse- or combination-
(refuse + fossil fuel) fired boiler holds for air pollution control and waste
management agencies, guidelines should be established to insure optimum
pay-offs. This is, in essence, the purpose of the present report.
Of the number of findings developed in this study, the
following can be highlighted:
.
Quantities of urban refuse collected will probably
have increased 1800/6 during the period 1965-2000,
or to abou~ 450 x 10 tpy.
I-I

-------
.
The higher heating value (HHV) of urban refuse will
probably have increased 30% by the year 2000, or
from a present value of about 5000 to 6500 BTU /lb.
.
If all U. S. urban refuse were fired, it could displace
considerable power plant coal - 25% now, increasing
to 38. 4% by the year 2000. Coal firing is only
regionally practiced and not necessarily near refuse
production centers, so that achievable displacements
would be much lower.
.
Using all available refuse in a selected LMA (Large
Metropolitan Area) to displace some coal in
uncontrolled power plants would reduce the S02
and particulate burdens (from all sources) by
about 7% and 21%,respectively. If the power plants
used S02 and particulate control equipment, the
reduction of the (then lowered) total burden would
be incons equential.
.
Combination-fuel fired systems have been identified
that would furnish disposal costs which are lower
than by sanitary landfill.
.
Steam generators equipped with agitating grates
generally proved to be more cost effective than
those employing some form of suspension firing.
.
Corrosion is not a serious problem. It can be
essentially eliminated by firing refuse and coal
in separate furnaces wherein the refuse does not
heat the working fluid much above 7500 F and the
coal is used to produce final steam conditions.
.
If correct refuse/coal heat input balances are
observed and proper boiler designs utilized, base
load systems compatible with modern turbo-
generators are entirely practical.
The present program was undertaken under Contract No. CPA
22 - 69 - 22 for the National Air Pollution Control Adminis tration. The prime
contractor for this work was the Envirogenics Company, a Division of
Aerojet-General Corporation. Extensive subcontractor inputs were
furnished by the Foster Wheeler Corporation in the field of steam
generator technology and design. The subcontractor providing services
in the field of air pollution control methodology was Cottrell
Environmental Systems, Inc., Division of Research-Cottrell, Inc.
1-2

-------
. ~- --- -
.' ,
B.
PROGRAM OBJECTIVES AND APPROACH
1.
Project Scope
The purpose of the present program was to determine
the characteristics of systems, that would be optimal in terms of the
state-of-the-art technology, for using refuse as a low sulfur fuel for
power generation, and to determine the overall potential of using such
systems to decrease the total emissions from the incineration of refuse,
and to assist in reducing sulfur pollution now resulting from fossil fuel
combustion. These broad objectives were approached as outlined in the
following paragraphs, which describe the scope of the program.
a.
Assessment of Refuse as a Fuel
Based on published and solicited data, the quan-
tities of refuse generated in the U. S. would be estimated for the period
extending from the present to the year 2000. These estimates would apply
not only to the national situation but to those expected for six LMA's selec-
ted because of high sulfur oxide burdens and waste disposal problems. For
the same regions and time period, projections were also to be developed
for the compositional (gross, chemical, and calorific) changes in refuse
that might be expected, taking into account those factors, such as packaging
industry trends, that influence refuse composition. From these bases, the.
maximum potential abatement in sulfur oxides, particulates and other pollu-
tants emitted from power plants would then be derived assuming that all
refuse in the areas of interest would be used to displace an equivalent
amount of fossil fuel energy.
Chemical and physical property data would be
collected and interpreted to determine the reactions of combustion pro-
ducts of all combinations of refuse and fossil fuel and the compatability
of these products with. the gas - side surfaces of conventional boilers.
Consideration would be given to: (1) the expected completenes sand
rate of reaction of refuse ash and fly ash with sulfur compounds, in-
cluding 502' in the combustion products from coal and fuel oil; -(2) the
expected reaction of coal and fuel oil ash and fly ash with halogens and
any other corrosive or deposit-forming constituents in the products of
refuse combustion; and (3) the determination of the desirable proportions
of refuse/fossil fuels that would minimize 502 release and boiler tube
wastage and maximize the performance of electrostatic precipitators
and other collectors.
Possible requirements for conditioning refuse
fuel, such as size reduction, drying, addition of materials (e. g., dolo-
mite or limestone) would be established. In this context, emphasis
would be placed on the preparation of refuse for suspension firing.
1-3

-------
Through solicited and published data, the performance and cost charac-
teristics of refuse grinding equipment would be established. Similarly,
the characteristics of ground refuse would be determined in terms of
size ranges and component types by sample taking and sorting operations.
Using Stokes Law calculations and appropriate empirical shape factors,
typical furnace gas velocities and circulation patterns, the residence
times of suspension-fired particles of various sizes and types would
be estimated. These data would then be correlated with the estimated
burning times and percent burnout of ash for maximum particle sizes.
Taking into consideration important social and
political factors that might be controlling, the engineering and economic
aspects of bulk handling, sorting, and transporting refuse before, during,
and after processing would be examined. This would be done to deter-
mine the economic constraints and optimum modes of moving refuse to
firing sites. Areas where essential knowledge or data are lacking would
be identified, including les s common but relevant operations, such as
pipe transport of slurried or dry refuse.
b.
Assessment of Power Plant Designs
In order to establish a foundation for design
developments, a state-of-the-art survey was to be made of wet- and
dry-bottom steam generators fired with both conventional and uncon-
ventional fuels, with particular emphasis being placed on selected
examples of European power plants (a minimum of two) firing mixed
fuels. In addition to the basic design parameters, data were to be
developed on fuel composition, heat and material (including sulfur)
balances, boiler efficiencies, residue and fly ash properties, APC
equipment performance, and the nature of any erosion, corrosion,
or fouling problems experienced.
From this information base, preliminary en-
gineering designs of optimum, refuse-firing utility boilers would then
be generated. This catalog of system pos sibilities would include neces-
sary flow sheets and tabulations of key process and cost data. Candi-
dates for each of the following types of applications would be provided
by this design effort:
.
A new refuse-fired :nstallation in which
fos sil fuel would be employed only as an
a uxilia ry fuel.
.
A new installation in which refuse/fossil
fuel blends would be routinely fired.
.
An existing central station boiler, modified
to accommodate the use of refuse as a fuel.
1-4

-------
\.
Using standard accounting practices of the electric
utility industry, net costs per kw-hr generated by the various systems
would be estimated. The net charges for disposal per ton of refuse
that would have to be made to the municipality by the utility operating
the system would then be determined as a function of return on capital.
The cost of power generation and refuse disposal for the several sys-
tems would then be compared with the costs of power generation by
conventional means and by incineration without heat recovery. Any
net savings in fuel costs, capital investment, or other expenses would,
be identified. In these cost analyses, the probability that S02 control
and more efficient particulate control equipment will be required would
have to be considered. In each type of system, the economic and political
factor associated with the optimum integration ofl the proposed designs
into the overall power generation and waste disposal complex would be
considered.
c.
Development of Recommendations and Formu-.
lations of R&D Plans
Implementation of the outputs of the study could
well require a systematic R&D approach in order to arrive at full-scale
proving of steam generator designs considered optimal. Workable R&D
plans would therefore be developed that would solve any techno-economic
problems identified on the program and provide for the fundamental
through scaled-up operations neces sary to evaluate whole systems or
, components thereof. The constituent R&D tasks, along with the esti-
mated time and costs required for their performance, would be described
and incorporated into one or both of two separate 5 -year plans. Each of
these plans would be scaled at a discrete level of effort - 5 and 25 million
dollars, respectively. The relationships of the various task elements
therein would be systematically defined and a logical sequence of R&D
operations would be recommended.
2.
Information and Data Sources
The information matrix developed for the sustenance
of the present program involved inputs from many different sources. A
considerable breadth of the open literature was of cours e reviewed. Key
items were abstracted, and committed to a storage system for timely
retrieval. Appendix D is a bibliography of the key material which en-
tered the study. Of the foreign-language publications, those in German
proved to be the most pertinent as well as the most prolific. A bulk of
the neces sary translations was rendered by Mr. E. J. Lachner of
Foster Wheeler Corporation.
Field surveys were conducted in both this country and
in Europe. A large portion of the data presented in Appendix B was ob-
tained as a result of such activities.
1- 5

-------
A number of leading authorities prominent in the several
fields pertinent to this study were consulted. To avoid any implication of
hierarchy, the following list of these specialists has been prepared in al-
phabetical order.
Consultant
Dr. - Ing. M. Andritzky
Mr. L. P. Flood
Mr. A. E. Gosselin, Jr.
Pro£. J. B. Howard
Pro£. E. R. Kaiser
Mr. A. Michaels
Dipl. -Ing. F. Nowak
Pro£. B. H. Sage
Pro£. A. F. Sarofim
Prof. D. G. Wilson
Affilia tion
Stadtbaudirektor a. D. (Munich)
Private Consultant
Private Consultant
Massachusetts Inst. Tech.
New York University
Private Consultant
Technische Werke der Stuttgart
California Inst. Tech.
Mas sachusetts Inst. Tech.
Massachusetts Inst. Tech.
A source of data of considerable importance on ~he
present program was the documents furnished by the Technische Uber-
wachungs-Verein (TtJV). This organization, in addition to other activities,
comprises a neutral apparatus for conducting acceptance tests on power
boilers in Germany. Detailed test reports for five modern refuse-burning
steam generators were obtained from the TtJV, translated, and evaluated.
A great deal of this data, in condensed form, is contained in Appendix B
(Section V).
Correspondence also served as a vital link in the infor-
mation gathering process. Much of the data on the LMA's was obtained
in this manner, as were important inputs from Government officials,
equipment manufacturers, researchers involved in related work, and
managers of operating and under-construction waste-fuel fired boilers.
Some laboratory work was also performed during the
present study. The analysis and characterization of deposits removed
from refuse-fired steam generators in Germany are reported in Appen-
dix A (Section II). Time was also devoted to the determination of size-
reduction patterns effected by refuse grinding machines. This work is
reported in the present volume (Section II, D) and involved a praiseworthy
degree of cooperativeness on the part of Heil Co. and the City of Madison,
Wis cons in, the Stanford Research Institute, and the Eidal International
Corporation.
Engineering drawings of existing boiler systems were
largely furnished by the Foster Wheeler Corporation in fulfillment of
their partnership on the present work. Additional drawings and other
important information were, however, also provided by Combustion
Engineering, Inc., and the Babcock and Wilcox Co.
1-6

-------
,,,,,,~"""):-.
. " .\,'
3.
Report Organization
As a review of the Table of Contents will reveal, the
initial discussions in the report have gone to the characteristics and
availability of refuse relative to the intended application. Following
, this, data are presented on the combustion products of refuse and the
implications these products may have in terms of damage to the en-
vir'onment or internal furnace structures. Appendix A, which has been
isolated from this discussion, specifically deals with wastage and fouling
associated with refuse firing. This particular documentation represents
an extensive condensation of a series of informal treatises produced by
Mr. R. W. Bryers of Foster Wheeler Corporation. The original unedited
material has been delivered to the EPA. .
The elucidation of refuse combustion-product parameters
lead on to the consideration of the extent pollutant levels might be reduced.
This topic is approached {in Section II, C} on the sensible basis of viewing
coal as the high sulfur fuel that would have to be displaced. The effect
that refuse and APC equipment would play in this fuel improvement is
factored into a 30-year forecast for 6 LMA's*.
\
Before commitment to the main topic of power boiler
technology, yet two more introductory topics are dealt with. Section
II, D provides a survey of current information on refuse handling, con-
ditioning, and transport processes. The intent is to update, in that a
very comprehensive treatment of this subject was produced not too long
ago. Included are viewpoints on the thermodynamic behavior of fired-
refuse as a function of shape and size and the aerodynamic properties
of the fluid phase. This topic is of considerable interest to design con-
cepts later discussed. Section II, E serves as a prologue to the appli-
cation intent of this report, and suggests alternative opportunities that
might be considered for harnessing the impressi'/e potential energy
available from urban waste.
The next main Section of the report (III) is dedicated
to the development and evaluation of suitable furnace configurations for
firing refuse. It is initiated, properly, with a fairly extensive survey
of the state-of-the-art of boiler tech,nology and Eu!"opean practice in
- using refuse to generate electricity. This is contained entirely in
Appendix B. As mentioned earlier, an important ingredient of this
survey was the information previously furnished by the TtJV. While
only a small fraction of the TtJV data could be practically reported in
this document, translations were made of all pertinent sections of the
reports.' <-
>,'
-------
Section III, B of the report develops in considerable
depth the criteria which must be observed in considering refuse as a
power plant fuel. From these bases are then presented ten candidate
configurations of new central station units suitable for generating grid-
electricity while disposing of urban refuse. An additional five systems
are also considered which would involve the modification (retrofit) of
existing utility-class boilers for the same purpose. Ideas are then
developed that go to APC equipment requirements and, more
importantly, the waste handling apparatus necessary to make such
power plants function successfully. .
In Section III, C, the engineering concepts presented
previously are subjected to a systematic cost analysis. New and retrofit
refuse burning power plants are treated separately. Cost models for
new plant designs, detailed in appendix C, are developed and cos ts are
determined for ten selected refuse/coal firing power plant configurations
and for analogous coal firing plants. Cost model trade-off study
results are presented and two candidate refuse firing power plant con-
figurations are subjected to detailed cost optimization studies. The
five power plant configurations selected as retrofit candidates are
indi vi dually cost analyzed. Retrofit and operating costs are defined
and refuse disposal cost comparisons are made.
Section IV deals with the matter of how the findings of
the report are to be implemented and reduced to profitable practice through,
(1) immediate technological opportunities, and (2) progressive R&D operations.
The final portions of the report are contained in Appendices
D and E, which consist, respectively, of the bibliography, . previously
mentioned, and a glossary of terms.
C.
CONCLUSIONS AND RECOMMENDATIONS
1.
Refuse as a Fuel
Quantity-Quality Considerations

It is estimated that the national population will
have increased by almost 60% by the end of the century. Because of the
changing living styles ,and other factors, the per capita production of
mixed municipal refuse will also increase. On the most probable basis,
the national "resource" of refuse will have increased by a factor of 2.8,
or from about 160 to 450 million tons per year, between the years 1965
and 2000. A factor as high as 4. 0 is also pos sible.
a.
With the introduction of more garbage disposal
units, the substitution of more plastic for glass and metal goods, and the
acceptance of more new items of short-lived (one-time-use) merchandise,
the composition of urban refuse can also be expected to undergo change.
The following tabulated predictions are considered to be ,reasonable.
1-8

-------
,,' ~
.'~>:i' '";::-'''''
,"', ,
PROJECTED COMPOSITIONAL CHANGES IN
U. S. URBAN REFUSE
I    Corn position,  Wt-%  
Year Garbage Plastics Glas s Metals Garden Paper Residual
 1970 20 2 12 10 12 38 6
 2000 6 13 7 5 12 55 6
The composition of municipal refuse at a given
location varies fairly widely in both the short and long (seasonal) scale.
Physical analyses of solid waste over a period of a week or so in the same
area show wide variations and seasonal effects on such constituents as
garden wastes account for most of the variation through the year. The
most striking thing about the geographical variation in refuse composition
is that it is so small. Aside from the geographical effects on amounts
and peak-appearance of garden wastes, municipal refuse is much the
same the country over. This may merely be caused by the homogeneity
of the culture in the U. S. From the point of view of refuse as a fuel, its
physical analysis is unimportant except insofar as it determines the heat
of combustion and, to a degree, corrosion potential. Because the seasonal
variations in waste load result largely from low-Btu, high-moisture yard
and garden wastes, the approximately + 8% variation in average waste
load introduces bnly about a + 6% variation in total heat available from
refuse over a year in a typical case. The heat of combustion of refuse
has grown steadily over the past several decades, but will probably not
grow as fast during the next three. Based on the compositional changes
shown above, an approximate 30% growth, from about 5000 Btu/lb in 1970
to about 6500 Btu/lb in 2000 AD, can be expected.
b.
Reduction in Pollutant Emis sions
At the present time, if all the urban refuse gene-
rated in the U. S. were to displace an equivalent amount of coal energy,
approximately 25% of the coal now fired in power plants could be saved.
By the year 2000 this displacement will probably have increased to almost
40%. This projected increase will be due not only to the enhancement in
fuel value that refuse is undergoing, but to the probability that, on a per
capita basis, refuse generation will increase at a faster rate than the de-
mand for electricity that can be satisfied by fossil fuel energy. This pre-
diction may prove to be in serious error if the problems now encountered
in the construction of nuclear power plants persist.
It is well known that the sulfur content of refuse
is low in comparison with that of coals and residual oils. Thorough review
of the literature shows a consistent average sulfur content of O. 1 to 0.2%
in U.S. refuse. This contrasts with a range of 2.5 to 3.5% for those.
1-9

-------
bituminous coals most commonly being fired in power plants. Correcting
for the differences in heats of combustion, sulfur input to the boiler could
be reduced by a factor of from 5 to 15. It is also well documented, however,
that subs tantially all (95 to 100%) the sulfur in coal or oil fired to a boiler
will appear in the flue gas as the oxides. Data available for refuseincin-
erators indicate that only somewhere between 25-50% of the input sulfur
is released as S02.' This is doubtless due to the fact that a sig'nificant
portion of the sullur present in trash is in the inorganic salt or fixed form.
Thus if all the urban refuse now available in the U. S. were to displace 2%
S coal, over 2-1/2 million tons of S02 would be eliminated from the atmos-
phere annually, This is based on the premise that no S02 control would be
practiced, which is essentially the case now, and that the percent of input
sulfur released as S02 is 95 and 50%, respectively, for coal and refuse.
Evaluation of data on particulate release
rates for power plants and refuse incinerators indicates that about the same
amount (2-1/2 million tons per year) of particulates as S02 could be
eliminated. This, however, is also based on a no- control situation,
which is obviously unrealistic. Only about 120,000 tons per year would
be eliminated if all the power plants involved were equipped with APC
equipment having an average efficiency of 95%. The reason for the higher
particulate release rates for coal is explicit from the modes of firing.
Refuse is usually burned, without prior size reduction, on grates, while
coal is fired in pulverized form in suspension. If shredded and fired in
suspension, refuse would doubtless generate more fly ash, but certainly
not to the extent of 80 or more percent of its inert content, which is typical
for coal. The ash particles formed would still be much larger from shredded
refuse than from pulverized coal, and would contain large amounts of dense
materials (e. g., glass and metal) that would be less susceptible to
elutriation than coal ash.
Because total consum~)tion of the national
refuse inventory in power boilers is unlikely, the more realistic model is
the LMA. Us ing the St. Louis area, as an example, it was found that if
all the available refuse there were fired in power boilers to replace part
of the coal, there would result a S02 reduction of about 31, 000 tons per
year. Based on the S02 burden from all sources, this would represent
a reduction of about 7%. The reduction would only amount to O. 5%, how-
ever, if the power plants were equipped with S02 scrubbers of 95% efficiency.
In the case of overall particulate burden, a reduction of 21 % would be
experienced if no APC equipment were being used on any power boiler.
This reduction would be only 1 %, however, if particulates from all power
plants were being removed by air cleaners with a 95% efficiency.
1-10

-------
.:
Coal displacement by combination firing with .
refuse will in itself obviously not completely solve the SOZ problem and
clearly the development of APC systems for removing this pollutant will
still be necessary. An important aspect of the refuse utilization concept
is the mode in which it would be fired. The argument could be raised that
the firing of refuse with reduced amounts of coal in a single furnace cham-
ber would actually provide little, ;if any, benefit. Scrubbers would still
have to be sized on the basis of flue gas volume and this would increase
as the refuse proportion was increased. Annualization of this added cost
might then actually exceed any operational cost reduction that would be
realized by a diminished rate of sorbent exhaustion. As will be discussed
later, however, the preferred configuration of the combined-fired power
boiler involves an almost complete isolation of the refuse and fossil-fuel
furnace components. Thus, in terms of SOZ removal, the volumetric
cleaning requirement would be reduced in a linear manner with respect
to the amount of coal displaced by refuse.
c.
Proces sing Technology
. Recent advances and developments in the field of
refuse processing were reviewed. While considerable new process tech-
nology and innovation was noted, little of it had any practical value for the
design objectives of the present study. With further development and ap-
plication work, a number of interesting concepts could well mature as
adaptable system components for refuse-firing power plants.
Included in this survey was of course refuse reduc-
tion hardware. Because refuse grinding is a somewhat recent application
for such equipment, accurate cost data or information on product size-
distribution could not he obtained from many of the manufacturers contacted.
Output samples were obtained and characterized for two types of machines
at Madison, Wisconsin, and Albuquerque, New Mexico.
Developments in the field of ground-refuse con-
veyance, classification, and salvage were evaluated and documented. This
included some brief but interesting testing of Stanford Research Institute's
Zig-Zag Separator.
The last areaanalyzed dealt with the most critical
stage in the processing of refuse; that is, the burning process itself. .
Because considerable experience has been acquired in firing refuse on
agitating grates, this analysis was confined to suspension firing mechanisms..
Here relatively little is known except for what has been demonstrated in the
firing of other type wastes. A theoretical treatment of the subject suggested
that the burn-out time required for the larger refuse particles would far
exceed the length of time they could be expected to remain suspended. This
1-11

-------
would suggest that some kind of retainer grate would be indispensable in
the suspension-fired furnace. This conclusion will have to be tested,
however, because any thermodynamic considerations must incorporate
many unavoidable assumptions and simplifications that tend to promote
fallibility.
d.
Thermal Utilization
While the obvious application of a combined-fired
boiler would be the production of electricity, other energy utilization
schemes should not be overlooked. In fact, refuse energy need not even
be exclusively considered in the context of the Rankine Cycle. This
possibility is now being explored with the CPU-400 System, which consists
of a fluidized-bed refuse furnace used as a gas generator to drive a gas
turbine and coupled power generator. Because of its in-development
status, however, this system, as well as others, were not considered
to be candidates ready for the present analysis.
A number of alternate uses for s team generated
by refuse (or any fuel, for that matter) have been reviewed. None was
regarded as offering a strong challenge to the power plant concept, yet
most would probably be acceptable if a favorable application situation
existed. For example, district heating would be an attractive means of
dispensing refuse energy, although during summer months an alternate
form of heat extraction (e. g., through absorption refrigeration) would be
neces sary . Evaporative water desalination units would also be excellent
devices in which to convert refuse to credits. However, only a limited
num ber of such installations would likely be built within reasonable reach
of a municipal refuse - collection system.
2.
Power Plant Designs
a.
Engineering Criteria for Candidate Systems
A survey of the state-of-the-art of domestic and
foreign power plant designs was conducted with the aim of identifying those
most amenable to combined firing of refuse and fossil fuel.
Pulverized-coal-fired utility stearn generators in
service include horizontally - and tangentially - fired furnaces, both
wet-.and dry- bottom, as well as cyclone furnace units; however, units being
sold now are virtually restricted to dry bottom, horizontally - or tangentially
fired systems. In this country, natural circulation boilers are now
1-12

-------
the dominant design for.m in the size range appropriate for refuse-firing
(500 MW or less because of refuse logistics). In Europe, however, the
once-through system still appears to be the favorite.
. From the point of view of combined firing of
municipal refuse in utility-type stearn generators, no experience is, as
yet, available in North America. This practice has grown in Europe,
almost entirely in the last decade, and is most advanced in Germany.
European practice involves reliance on various refuse burning grates,
with pit and crane refuse handling, but with a variety of stearn generating
arrangements.
All presently operating, large North American
stearn-generating, refuse-disposal plants use conventional, agitating
grates and generate stearn at such moderate pres sures and temperatures
as to be unsuitable for turbo- electric applications. All except the incin-
erator at the Naval Base at Norfolk, Virginia, fire refuse only; the Norfolk
unit fires a combination of oil and the solid wastes collected at the base.
It was, however, the first water-walled refuse incinerator built in North
America. Water-walled, steam generating incinerators are now under way
in Hamilton, Ontario, Braintree, Mas sachusetts, Harrisburg, Penn-
sylvania, and Chicago. A recently completed plant in Montreal is in
its initial shakedown operat~on. Again, all of these systems are low -
or moderate - pressure units.
Current domestic technology in bark-, bagass e-
and wood chip-fired heat recovery furnaces is pointing the way to advances
over the present state of the art of refuse combustion, which is almost
universally based on the use of the reciprocating or turning grate. The
installation now under construction at East Hamilton, Ontario, will utilize
the next logical advance in refuse combustion, namely the spreader-stoker
equipped with a burnout grate. Planned for test in the near future in St.
Louis is the suspension firing of shredded refuse in a grateless furnace,
actually an old, coal-fired utility boiler modified for such service. A
similar system is under construction in Rochester, New York, although
it will have a burn-out grate, as does its bark-burning prototype at
Muskegon, Michigan.
Variations in st~am flow of + 30 to 40% have
been observed in grate-fed stearn generating incinerators both here and
abroad. These fluctuations can have periods of from 15 to 25 min., with
maximum 6p- rates of from 2 to 8% per min. These wide fluctuations can be
gre,atly reduced by suspension or spreader-stoker firing of ground refuse.
This would result from the combined effects of size reduction, mixing, better
contact with combustion air, and the reduction in furnace load of unburned
fuel. Another engineering approach, particularly desirable if the agitating
grate is to be used, would be to employ separate refuse- and coal-fired
1-13

-------
furnaces. The latter, which are far more susceptible to control, could
thus be used to develop the final stearn conditions required for the system,
and thus even out fluctuations introduced by the refuse-fired portion of
the boiler. A secondary benefit would be that the refuse furnace could
then be operated at lower stearn temperatures. This would serve to
reduce the corrosion effects that sometimes occur on tube surfaces in
refuse-fired boilers when stearn temperatures are allowed to exceed
8000 F.
The conservative viewpoint in the U. S. has been
that corrosion resulting from refuse combustion would not permit its use
in modern, high-pressure, high-temperature, stearn generators. Reports
of severe corrosion in the several modern combined-firing stearn generators
in Europe circulated soon after their start-up, and are still repeated.
Discussion by project personnel with the plant operators and others
confirms an opposing view, that tube wastage, while expectedly greater
than with fossil fuels, is a nuisance rather than a severe problem. Tube-
wall corrosion rates were initially high but soon levelled off to tolerable
rate s, apparently after the tubes became coated with protective deposits.
Based on the considerations discussed above, the
boiler designs developed on the present program were limited as to refuse
input, generally at 60% of the energy. Stearn conditions were also approached
on a conservative basis, particularly as the refuse input approached the
specified limit. The stearn generators and other components sensitive to
flue-gas flow rates were sized for 50% excess air for refuse-firing and
18% excess air for the coal. These criteria were established on the basis
of German practice. Design exit-Hue-gas temperature was set at 4500 F,
which is considerably higher than current practice in utility power plants.
This temperature selection was based on performance constraints imposed
by the use of an electrostatic precipitator and the resistivity characteristics
of refus e fly ash.
Based on the holding limits of commercially
available grates, the maximum unit refuse capacity adopted was 1000
tpd for furnaces equipped with agitating grates and 2000 tpd for those
with s loping, fall-out retainer grates or none at all. An upper limit* of
refuse input per plant was arbitrarily set at 8000 tpd, although it was
recognized that a system of such great size would only be feasible if unique
refuse in-haul techniques were utilized. The capacity range for individual
units was derived along similar lines and the limits set at 85 and 500 MW,
the latter being highly unlikely of attainment unless a very low refuse
proportion were employed. Because nameplate rating v.o uld not exceed
500 MW, it was then logical to utilize natural circulation stearn circuitries.
Systems having capacities of less than 85 MW, which would perhaps be
appropriate for smaller communities, were not considered because refus e
disposal costs were found to be unacceptably high at the low refuse rates
entailed.
>:'for analytical purposes.
1-14

-------
,,' """O'~' .." . '-
, ' "
Another aspect of the design approach included
provision for both S02 c9ntrol (limestone wet-scrubbing with flue gas
reheat) and dust control. The latter would be handled by an electrostatic
precipitator for the flue gas from the refuse-fired furnace and by the
limestone wet scrubber in the coal furnace flue. If the flue gases were
blended prior to the air cleaner, an appropriately sized wet scrubber
would alone be used to remove S02 and particulates for the entire system.
Control of HCl and certain, other gaseous emiss ions identified with refuse
incineration did not fall within the scope of the present study. Such
control may well develop as a requirement of the future, as the com- ,
position of refuse changes.
. Refuse handling procedures commonly practiced
in U. S. incinerators and even in European steam-generating plants are
based on a flow arrangement wherein bridge cranes move the feed, from
furnace-house storage pits to the furnace charging chutes. Alternative,
more cost-effective designs are now being explored. These are for
systems for which all input refuse must be shredded. Live-bottom.
receiving and storage structures have been specified and the movement
of the shredded feed material is by means of conveyors or pneumatic
pipelines. In the present study, it was found d,esirable to specify
similar handling equipment whether the input were to be shr,edded or
not. '
Observing the various criteria discussed above,
a total of fifteen designs of utility-grade, refuse-firing, sfeam generators
were developed. Ten of these were new-construction schemes, while
five were intended to be used in converting (retrofitting) existing, old,er-
generation boilers to combination-fuel service. The latter are described
later in this introductory discussion; the ten new-construction systems
are summarized for the size range from 100 to 500 MW in Table 1-1.
Many of the systems described there, however, were actually para-
metrically analyzed over the size range from 25 to 500 MW and the
refuse-input ranges or points compatible with each of the designs.

The refuse disposal ,costs shown are based pn
the economic modeling des cribed in the next section and include all
operations, including residue disposal, following refuse receipt at
the site. All of the systems involve the firing of coal, the fossil fuel
selected because of its typical high sulfur content, in combination with
refuse.
1-15

-------
Candidate
TABLE 1-1
SUMMAR Y OF SYSTEM CHARACTERISTICS OF
REFUSE-FIRING STEAM GENERATOR DESIGNS
OVER THE RANGE 100 TO 500 MW
System Description
Permissable Furnace
Frac. Heat Input from
Refuse (100/500 MW),
%
Design Waste Rate
Corresponding to
Frac. Waste Heat
Input (100/500 MW),
tpd
Steam Conditions 1
Corresponding to
Frac. Waste Heat
Input (100/500 MW)
Refuse Disposal
Cost Range, 2, 3
$ITon
Case 1
Case 2
Case 3
Case 4
Case 5
Case 6
Refuse (on reciprocating grates) and pulverized coal are
fired in separate furnaces and the flue gases combined
to heat common convection sections. Fuel ratio is deter-
mined by size and design heat release rates of furnaces.
Refuse (on reciprocating grates) and pulverized coal are
fired ina completely common furnace. Fuel ratio is not
fixed.
Refuse is fired on reciprocating grates in a separate
furnace to generate feedwater for pulverized coal fired
steam generator. Fuel ratio is limited. Flue gas
flows are isolated.
Refuse is fired on reciprocating grates in a separate
furnace to generate saturated steam for a pulverized
coal fired superheater. Fuel ratio is fixed. Flue gas
flow:;; are isolated.
Shredded refuse is fired in a spreader stoker furnace
to generate partially superheated steam for a separate,
pulverized-coal fired superheater. Fuel ratio is fixed.
Flue gas flows are isolated.
Shredded refuse and pulverized coal are blown into and
fired in -
-------
TABLE 1-1 CONTINUED
Candidate
System Description
Permis sable Furnace
E;'rac. Heat Input from
Refuse (100/500 MW),
$
Design Waste Rate
Corresponding to
Frac. Waste Heat
Input (100/500 MW),
tpd
Steam Conditions 1
Corresponding to
Frac. Waste Heat
Input (100/500 MW)
Refuse Dispos~l
Cost Range, , 3
$ / Ton
Case 7
Shredded refuse is fired by the spreader stoker principle
in a furnace also equipped with pulverized coal burners.
Fuel ratio is not fixed.
0-60/0-60
0-2278/0-96503
Case 8
Shredded refuse and pulverized coal are fired in
suspension in a common furnace in a manner that
will induce slagging. Fuel ratio is limited.
0-1458/0-6825
4
0-40/0-40
Case 9
Shredded refuse and pulverized coal are fired in
suspension in a common arc~ furnace. Fuel :ratio
is not fixed.
0-60/0-60
0-2278/0-96503
Case 10
Shredded refuse is fired in suspension in a separate
arch furnace to generate partially superheated steam
for a pulverized-coal fired superheater. Fuel ratio
is fixed. Flue gas flows are isolated. '
1780/7450
58.0/54.5
1.
The four steam conditions (throttle temperature, psig/throttle temperature,
of /Reheated Steam temperature, OF) observed in this study were:
A.
B.
C.
D.
850/900/no reheat
1250/950/no reheat
1800/1000/1000
2400/1000/1000
Steam generator efficiency and, thus, fuel requirement are influenced
by both steam conditions and refuse rate.
2.
For systems capable of firing over a wide range of refuse inputs
(Cases 1,2,6,7,8, and 9), a minimum of 20% fractional heat
input from refuse was designated.
3.
In the cost analysis, an arbitrary upper limit of 8000 tpd was observed
for refuse firing rates.
4.
Arbitrary upper limit specified to reduce probability of slag solidification.
5.
System rejected after preliminary cost analysis because of high refuse
disposal cost.
C-A/D-A
2.50-5.90
C-B/D-B
3.00-6.00
C-A/D-A
Note 5
C/D"
2.10-4.60
1-17

-------
Detailed calculations were made in order to estima.te
the expected performance characteristics of the many configurations.
These were developed at several levels of electrical output and over a
broad range of fuel ratios, except where design constraints imposed
restrictions on the proportions of fuels that could be used.
. Construction costs were next generated for the
many candidate designs. The new construction types were then analyzed
to establish cost-effectiveness.
b.
Economic Model
A detailed cost model was developed to evaluate
the large number of potential combined firing systems, to determine
the sensitivity of disposal costs to various cost and design parameters,
and to identify the most cost optimum new plant configurations to select
for detailed cost studies. This approach was not applicable to the
retrofit systems because specific, existing plants were examined. Even
here, however, portions of the model could be and were applied to
generate needed cost information.
In general, it was felt that no single new-plant
candidate would be uniquely preferred over all others because local con-
ditions and constraints could limit the usefulness of anyone system. In
order to evaluate these effects, the cost model specifically analyzed the
condition for two municipalities to determine the degree to which different
systems would be optimum in different locations. These derivations were
based on cost information valid as of July 1969. Although all costs have
increased significantly since that time, the refuse disposal cost - a
primary criterion of cost effectivenes s in this analysis - will not have
changed commensurately or necessarily even in the same direction.
This is because the cost of refuse disposal with energy recovery increases
with the capital, operating, and maintenance costs of a fixed system
but decreases with rises in the costs of the fossil fuels that are being
displaced. For systems that are inherently cost-effective to begin with,
it can be generalized that, for systems designed to fire coal, the two
effects will have about cancelled each other during the period since
July 1969. This would not be true of steam generator designs based
on the combination firing of refuse with oil or natural gas that now burn,
respectively, low sulfur oil or liquified natural gas. Such systems,
although not specifically examined in this study, would clearly have
benefitted greatly since July 1969 in having a portion of such fuels dis placed
by refuse. -
The cost model deve loped for this program (dis-
cussed in detail in Section LIT, C and Appendix C) was incorporated into
a detailed computer program to facilitate the analysis of the large
number of design options. The model was based on the primary assumption that
1-18

-------
the power companies using refuse as a fossil ,fuel substitute would be
regulated public utilities with privately owned equity financing the sub-
ject to all applicable Federal, state, and local taxes. This and other
basic assumptions resulted in a fixed charge annualization rate average
of 14. 7 percent. This should be compared to a typical annualization rate
of 7. 3 percent for municipal ownership and perhaps 20 percent for unregulated
commercial ownership. The annualization rate for utilities was, however,
observed in all of the cost analyses performed.
In computing costs it was as sumed that the
combined firing plant would operate in a bas e load condition at an
80 percent plant factor; a high plant factor is indispensable for the
support of refuse disposal gperations. In addition, a value (subsequently
much exceeded) of $.31/10 Btu was used as the cost of fossil fuel
replaced by refuse.
One of the most important parameters in deter-
mining the net cost of disposal using a combined firing process is the
value of the power generated in the plant. In this analysis, values in
the range of 6 to 8 mills /kwh were used depending on the electrical power
generating capacity of the plant and the cost of fos sil fuel.
The cost model examined plants with power
generating capacities in the range of 100 to 500 MW with from 20 to 60
percent of the input heating value derived from refuse. Nominally,
refuse was as sum~d to have a heating value of 4460 Btu/lb (based on
1965 data) compared to 12, 020 Btu/lb for coal. The combined firing
plants handled from 500 to 5500 tpd of refuse, depending on the electrical
power capacity, the refuse fraction, and the plant configuration.
A sensitivity analysis was conducted to determine
the effect of selected variations in the economic and design parameters,
such as plant factor, refus'e heating value, coal cost, and rate of
return on capital investment. It was found that the net disposal cost
tends to be a very sensitive value, primarily because it is computed by
taking the difference of two rather large numbers, operating costs and
the credit for power generated.' Small percentage changes in these
numb ers effect large changes in the absolute value of the difference.
c.
Detailed Design and Cost Studies
(1 )
New Plants
The capital anrl. operating expense of grinding
refuse had a dominating influence on the disposal costs for the various
designs analyzed. Those candidates employing agitating grates wherein
only oversized input items would have to be reduced were generally found
to be the more cost-effective. Within this class, a configuration designated
as Case 3, clearly emerged as an optimum design for the unit size-range
(200-400 MW) best suited to utility and waste management operations Ln
large metropolitan areas. .
1-19

-------
Despite indicated higher disposal costs, a
candidate was also selected from the class of designs specifying suspension
firing. This was done because, lacking any data from precedent refuse-
firings, conservative performance characteristics had to be assumed for
these systems, and because refuse-grinding costs needed further verifi-
cation for similCi.r reasons. The system, selected from this class, desig-
nated as Case 10, was seen in the cost modeling to be optimum for that
type of design. Summary descriptions of the two class-optimum candidates
follow:
(a)
Steam Generator with Refuse-Fired
Economizer (Case 3)
. The Case 3 steam generator design
resembles, in basic lay-out, a widely regarded combined-fired boiler now
in operation in Germany (Munich South, Unit No.6). The two differ sig-
nificantly, however, in several major features, as will be brought out
later.
At the sizing (400 MW) shown by the
cost model to be optimal, the Case 3 system comprises a pulverized- coal
fueled boiler equipped with very little economizer surface but otherwise of
conventional design, and three refuse-fired furnaces that would serve as
the actual economizers. These feedwater heaters would burn about 930
tpd of refuse on reciprocating grates and deliver feedwater slightly below
saturation (6570F and 2600 psig) to the "economizer II (actually a mixer) of
the steam generator. The latter, firing about 3000 tpd of pulverized coal,
would produce about 2.8 x 106 lb/hr of steam at 10000F and 2400 psig for
a single turbine-generator set equipped for reheat operation. The net plant
heat rate would be about 10,100 Btu/kw-hr.
The flue gas from the economizers would
be intrinsically isolated from that of the steam generator. Thus the flue gas
exiting from all three economizers would be combined, passed through an
electrostatic precipitator, without concern for S02 removal, and on to a
cornmon plant stack. In the coal-fired boiler, the flue gas would be passed
through a limestone wet- scrubber to remove S02 and particulates and thence
to the common stack.
Based on the cost model criteria dis-
cus s ed earlier and a net plant refuse rate of 2790 tpd (25% of the energy
input), the total net disposal cost would be $1. 61 per ton. This figure
includes, however, a transportation cost of $1. 65, showing that the plant
itself would operate at a profit of $0.04 per ton. Thus refuse disposal
using this particular design would not only be competitive with but con-
siderably more cost-effective than landfill.
1-20

-------
(b)
Steam Generator with Refuse-Fired
Arch Furnace (Case 10)
The system selected from among the
suspension-fired designs as being the most cost-effective would incorporate
a refuse-fired arch furnace. In this unit,. ground refuse would tend to burn
in suspension because of the mode of injection and reaction to the. lifting ef-
fect of directional air flow. Water-cooled grate bars would be installed on
the sloping walls of the hopper to provide some retE:mtion of fallout. The
need for this feature is highly uncertain and itsomis sion would greatly
reduce construction costs. .
In the total plant system, 1960 tpd of
refuse (60% of totat energy) would be fired in conjunction with 496 tpd of
coal. A single arch furnace would fire all of the refuse in combination
with 182 tpd of pulverized coal, the lafter to insure even combustion. This
unit would produce 0.8 x 106 lb/hr of steam. at 7500F and 1940 psig. The
steam would be fed to a coal-fired (164 tpd) superheater of special design,
where it would be brought to final turbine inlet conditions (10000F and 1800
psig). A single turbine would be needed to satisfy the design goal of 100 MW.
Because of tube- surface distribution
constraints, reheat for the turbine would be accomplished in a second coal-
fired (150 tpd) furnace rather than in the superheater. On the fire side, the
two coal-fired systems would be almost identical. A number of design dif-
ferences would be necessary for the steam side, however. Controls for such
a system would necessarily be more complex than for conventional steam
generators. .
As pointed out earlier, the disposal
costs for the suspension-fired systems were found to be higher th~m for
grate-fired configurations. In the present case, the estimate came to
$6.17 per ton, including transportation costs. The accuracy of this value
is much less certain than for the more straightforward Case 3 situation.
Not only might the performance specifications, but grinding costs as well,
prove overly conservative.
The data available for developing the
latter were quite limited and tended to be inconsistent. The grinding costs
used for the present design (2-in. top size) were about $2. 50 per ton, in-
cluding annualization of capital costs, and operating and maintenance costs.
(2)
Retrofit Plants
(a)
Background
The plans developed for retrofit oppor-
tunities were based on examples of existing, utility-class boilers that had
gone into service in the earlier part of the past two decades. Within this
class of older boilers, only the more typical d~signs and steam conditions
were considered. . .
1-21

-------
Five retrofit designs were prepared for
unit nameplate ratings ranging from 44 to 300 MW. In each case, the design
was so developed that the original unit capacity and steam conditions were
not changed.
(b)
Design Summary
Modification No. 1. The selected 60
MW unit (no reheat capability) firing pUlverized coal from horizontally-
aligned guns would be retr,ofitted with a reciprocating grate. Extensive
changes in the steam circuitry in the lower portion of the furnace were
indicated, as was the addition of new air-supply equipment. The modified
unit would derive 43% of its energy, from refuse (951 tpd) and produce, as
before, a steam output of 0.6 x 106 lb/hr at 900 psig and 9000F.
Modification No.2. In this design, a
separate, refuse-firing furnace, equipped with reciprocating grate, would
be added onto an existing 150 MW unit and the two furnaces coupled to share
the existing convection and downstream flue structure. It is as sumed, of
course, that sufficient space is available at the site to Eermit such an ex-
pansion. This retrofit system would generate 1. 1 x 10 lb/hr of steam at
2035 psig and 10530F, using a reheat cycle. About 24% of the energy would
be derived from refuse, based on a rate of 1000 tpd. Because of the add-on
boiler effect, the cost of this particular modification proved to be much higher
than for any other retrofit system considered. The resulting unit would be
similar in function to boilers now in operation in Munich and Stuttgart.
Modification No.3. This retrofit design
would involve a restructuring operation similar to that specified for Modifi-
cation No. 1. The grate installed, however, would be of the travelling type.
Air-swept spouts would be installed above the grate for introducing shredded
(4-in. top size) refuse into the furnace; thus, a spreader - stoker configuration
would be realized. This 44 MW unit would generate 0.4 x 106 lb/hr of steam
at 1370 psig and 9050F. About 50% of the energy would be derived from refuse
(846 tpd).
Modification No.4. This design was based
on the use of the same existing plant just considered for Modification No.3.
The boiler would, however, be equipped with a reciprocating grate and operate
in much the same manner as Modification No. 1. The basic difference be-
tween the latter and the present system would be in the styles of construction
of the existing boilers and therefore the modification techniques necessary to
achieve conversion. The existing plant considered for Modification No.1 was
a much older plant than the present one, such that considerably different ap-
proaches would be necessary to produce what would essentially be the same
result.
The present design calls for firing of
679 tpd of refuse, which would represent 42% of the heat input.
1-22

-------
. .
..-
'. -:
Modification No.5. This design in-
volves the modification of one of the units at a midwestern power plant
where a similar retrofit is a1ready being pursued under EPA sponsor-
ship. In the present case, suspension firing would also be practiced using
refuse nozzles that would be wall-mounted above existing, horizontally-
aligned coal guns. A notable difference would be the conversion of the
hopper walls to provide a water-cooled grate surface. Being steeply.
sloped, this surface would only have a delaying effect on, but thus promote
more compiete burn-out of, fallout tumbling into the ash pit. This con-
servative design feature may prove unnecessary, in which case a major
modification cost-element could be eliminated.
The hopper-wan treatment of the present
retrofit system, that of Ca.se 6, which is identical in principle, and that of
the arch furnace (new construction, Case 10) are very similar and provide
no horizontal retaining-surface. Thus the completeness of particle com-
bustion while in suspension is an important consideration in determining
the refuse fraction to be used. Burn-out in an arch furnace is expected
to be more complete because of furnace-air distribution. For this reason,
a lower refuse fraction has been stipulated for the present system.
Based on a conservative fractional heat
input from refuse of 10%, this 300 MW unit would consume 635 tpd of that
fuel. Steam production would be about 2.3 x 106 Ib/hr at 2,200 psig and
10100F, using a reheat cycle.
I
I
I '
(c)
Disposal Costs of Retrofit Systems
A different approach than that used for
the new construction units was neces sarily employed in deriving disposal
costs for the retrofit designs. Annualizationfactors for the existing plant
prior to retrofit were of course excluded, as were credits for electricity
generated. Many of the cost derivations employed in the cost model were
applicable, however. . The results are tabulated below.
REFUSE DISPOSAL COSTS FOR
RETROFIT SYSTEMS
Mod. 1
Mod. 2
Mod. 3
Mod. 4
Disposal Cost, $/ton
0.65
2.87
. 2.14
1. 07
Mod. 5
3.69* or 4.44
':'Lower cost results if hopper walls are not modified.
1-23

-------
The above data again show that sus pension
firing (Modifications Nos. 3 and 5) leads to apparent higher disposal costs. Of
the g rate- stoked system.s. Modification No.2 proved to be an exception to this
effect because of its very high reconstruction cost.
An important pararneter that is included
in the disposal costs shown above is plant factor, which was 80%. This is the
same value that was applied to the new constructions designs and may prove
overly optimistic considering the age of the retrofitted boilers. Disposal
costs would increase about 25% for each 10% (relative) drop in plant factor
until a plant factor of about 50% is reached.
3.
Recommended Research and Development
a.
Criteria
In the absence of a sufficiently high economic
incentive, .a technology typically advances at a rate determined by experience
gained in the practical employment of the device or proces s. The rate of
incorporation of advancement in such industries is slow and seldom is there
a sudden acceleration in the growth cycle. Designs remain conservative in
order to minimize economic risks of sales. The industry in effect stagnates
even though its product might potentially offer great advantages over alterna-
tive methods or equipment. Only through a properly programmed research
and development effort can the true worth of a technology be established.
In the last several years, innovations for
superior incineration of solid wastes have begun to be studied, but clearly
the industry has largely found the support of an adequate R&D program to
be unwarranted on economic grounds, and utilities express a lack of interest
in considering the use of refuse as a fuel. With a number of analytical
studies, including the present one, demonstrating that combustion of
refuse can offer significant advantages under a rather wide range of con-
ditions. research must be encouraged in this area. The establishment
of the fact that incineration with energy recovery is an economically sound
process in itself forms a portion of the encouragement. The outlining of a
definitive 5-year program of R&D that will lead to improvements in the
ability of this industry to be attractive to outside capital would also con-
tribute towards this end. Suggested plans for such a program at two
different levels of support are discussed below.
A listing of the many elements that might be
considered to influence refuse combustion would be impressive in length.
would include waste generation rates, compositions, possible effects of
partial salvage, numerous factors affecting the combustion proces s, the
nature of all combustion products (and their reactions with each other and
items of hardware), heat transfer considerations, energy utilization, air
pollution control, and residue disposal. In order to develop a superior
total system, the several disciplines of engineering and science would
It
1-24

-------
~ ----
. .
need to make their contributions to analysis and research of these paranleters,
as would economists, operations research experts, and other specialists.
Viewed in this broad s cope, hundreds of individual programs could be sug-
gested, each adding to the knowledge of the incineration process. Some
would improve understanding of underlying theoretical principles, others
would lead to improved facility designs, and still others would improve the
accuracy of existing data. While all this information would be "useful, II the
sum of it hardly makes up an R&D program that should be considered by the
Government to encourage the use of refuse as a low sulfur fuel. Clearly it
is necessary to place well-defined limits on the scope of activities if this
prime objective is to be met.
Criteria considered important for incorporation of
an individual R&D task into an integrated program include:'
. Key information is needed in an area that would
probably not recelVe proprietary support from industry.
. Fundamental research within the university and
research institute system on the particular subject matter is~not expected to
be conducted with currently visible private funding during the period of concern.
. Numerous projects exist that require similar
major facilities or test equipment, where sharing of a central laboratory
.would offer important cost savings and unification of effort.
. Information compilation, as an end to itself,
is not a sufficient basis for project justification. Results, mainly obtained
experimentally and in a few cases analytically, mu.st be directly applicable
to the construction and operation of facilities using refuse as a fuel in an
air pollution-free manner and must encourage such a use for refuse.
Under the above criteria, a number of projects are
eliminated that might be considered by some as vital to the cause of succes s-
ful refuse combustion. Their elimination does not imply that the tasks are.
not worthy of support, but only that it is believed that (1) funding will be ac-
complished through sources other than the presently planned R&D program,
or (2) incorporation of them here would detract from the main developmental
needs of that program. 'Typical of such tasks are further projection of refus e
generation rates, kinetics of nitric oxide formation, development of abate-
ment processes for hydrogen chloride effluent, and investigation of by-
products from ash.
Suggested projects considered to meet th~ list
criteria are discussed below.
1- 25

-------
b.
Plan A
A time period of 5 years and a total expenditure of
approximately $5 rnillion make up the basic constraints of Plan A. Because
of the high costs of suitable research equipment and test apparatus and the
redundancy that would occur in the purchase of these items by individual in-
vestigators, it is recommended that a central R&D facility be established as
a key feature of this plan. While not all research elements would be neces-
sarily conducted at the central facility, it is believed that the improved quality
and quantity of work that could be accomplished at this specialized laboratory
would make the scheduling of most programs there highly advisable. Respon-
sibility for administration and service of the 20,000 sq ft facility could be
vested in either a contractor or the Government, with individual contractor
teams then arranging for time at the laboratory.
Further description of this central facility is given
in Se ction IV.
Of the ori~inal $5 million,
would remain for conducting R&D¥. Areas where
trated, along with an estimate of costs, include:
approximately $3 million
efforts should be concen-
. Experimental flow modeling of a variety of
refuse burning systems in order to increase furnace efficiency and minimize
design risks for new large incinerators - $400,000.
. Resolution of present uncertainties in causes
of corrosion and fouling and means to decrease these phenomena - $300,000.
. Analytical studies under a single set of assump-
tions to ascertain comparative advantages of "non-conventional" thermal con-
version systems for refuse (pyrolysis, gasification, etc.) - $200,000.
. Sub-scale component tests of innovations for
both retrofit and new construction facilities - $600,000.
. Acquisition of data leading to improved air
pollution control devices for particulates when refuse is burned - $500,000.
>:'That some 40% of the funds allocated are required for construction, out-
fitting, and basic servicing of the lab should not be considered as a negative
feature of this recommendation. It should be borne in mind that (1) greater
expense would be incurred in equipment duplication in separate facilities,
(2) contractors would desire to use the lab on a no- cost (or even would be
willing to rent space) basis and hence more R&D than actually funded would
be accomplished, (3) a lower overhead burden would result on contractor
work assigned to the lab, and (4) the facility life would be greatly in excess
of the 5-year planning period used here.
1-26

-------
((I Analytical study to establish lower quantity
limit where air pollution free refuse combustion with energy recovery is
economically competitive - $100,000.
o Combustion studies in versatile, well instru-
mented, furnace to ascertain means for improved burn-out at highest pos-
sible heat release rates - $900,000.
c.
Plan B
The assumption of $25 million in funding over a
five-year period is made for R&:D Plan B. The same $2 million facility
of Plan A should be constructed and the research conducted that was rec-
ommended in that plan; the last mentioned area above, essential to design
of improved large incinerators, should be increased in funding by $2. 1
million to $3 million. One of the objectives of the program will be to
assist in the funding of a full scale facility, able to burn refuse in a fully
practical manner, yet instrumented for extensive monitoring and designed
for ease of component arrangement. ThiiS funding requirement, exclusive
of the designphase, is estimated at $10 million. Other R&D areas that
should be covered under the broader financial support of Plans B include:
. Experimental investigations of refuse pyrolysis
and gasification, the exact nature of these tests to depend upon the analytical
studies and the outcome of the limited research currently being conducted -
$ 3 million.
. Experimental investigation of fluidized bed
refuse combustion, with the aim of reduced capital cost per unit of capacity
and decreased air pollution emissions - $3 million.
. (jI Preparation of a detailed manual of retrofit
construction recommendations, permitting existing utilities to begin burn-
ing refuse at minimum costs and risks - $1 million.
G Design of a 100 tpd combined fired incin-
erator with energy recovery, with capability of yielding practical data for
optimization of future large units - $1 million.
1-27

-------
II.
REFUSE AS A FUEL
A.
QUANTITIES AVAILABLE
1.
Collection Rates
a.
United States
(1)
Current Collection Rates
In order to estimate the quantities of refuse
presently collected in the U. S., the existing data, consisting of studies by
metropolitan planning authorities and other miscellaneous documents, were
analyzed to determine collection rates of refuse in various clas sifications.
This material was examined to assure that as many as possible independent
estimates of total refuse collected by a given population were identified.
This also served as a further check on the comprehensive data obtained by
questionnaire and published by the Committee on Solid Wastes of the American
Public Works As sociation (APW A) (Ref. 1). The per capita refuse rates given
in Reference 1 are for municipal refuse collected by city forces; they do not
include privately-handled industrial wastes. They are, however, all based
on actual weight records and are collection rates. The overall" medians and
'ranges of values are:
1965 APWA SURVEY - COLLECTION RATES
~
lb/cap. -day
Resldential '
Commercial and Industrial
All Refuse
2.0 (1. 1 - 3.2)
2.0 (1. 3 - 2.6)
3.9 (2.3 - 6.5)
The significant spread noted emphasizes the wide variability from city to
city in actual amounts collected for municipal disposal, which mayor may
not reflect a similar variability in actual generation rates.
Table II-I, a summary of various surveys
conducted, further illustrates the magnitude of the problem. It unfortunately
also reflects the fact that a standard breakdown of the categories of solid
waste had not been observed. Thus, while the tabulation appears to show
wide'variations, the totals do not all involve the same components. In any
case, three things are apparent from this tabulation: (I) the widely used
cQlle~tion rates of around 4. 5 lb/capita-day appear to have a basis in fact;
(2) the amount of industrial waste is large and widely variable; and (3 ) there
has been a substantial increase in the collection rate in the last) 0 to 15
years. The information in column VIII is based on a study (Ref. 8) almost
20 years old. It gives the average for 13 California cities. In columns XI
II-I

-------
      T ABl.E II-I       
    SUMMARY OF REFUSE COl.LECTION RATE ESTIMATES    
      (in pounds/capita-day)      
 Area:  I II III IV V VI VII VIII IX X XI XII
 Refe rence: 3 2 3 4 5 6 7 8 9 10 8 11
 Year:  1967 1966 1968 1968 1967 1966 1965 1950 1961 1960 1950 1967
 Residential 2.4  2.0 3. 1 2.9 2.8  2. 1  2.7 1.5 
 Commercial 1.4  1.0  0.6(1)     1. 9(1) O. 7 
 Demolition and Industrial 1.4  1.2   2. 7      
 Brush, Park Refuse, etc. 0.3  0.3  O. 1 O. 1      
  Sub- Total: 5.5 4.5 4.5 5.0(2) 3.6 5.8 4.0 2. 1 3.9 4.6 2.2 6.0
 Other  1.8     1.4  0.8    4.0
H              
H  Total: 7.3     7.2  2.9    10.0
I         
N              
 Avg. Growth ('¥a/year)(3)    1. 94 1. 00 2.00 2.62  1. 06 1. 69  2.50
Footnotes:
(1) Includes industrial
(2) Balance of refuse from unspecified sources
(3) Projected average growth rates
AREAS
I.
II.
Ill.
IV.
Various
Chicago,
Various
Various
Illinois
V. Malden, Massachusetts
VI. Boston, Massachusetts
VII. San Francisco Bay Area, Calif.
VIII. Thirteen California Cities
IX.
X.
XI.
XII.
Boston, Massachusetts
Milwaukee, Wisconsin
Fresno, California
Fresno, California

-------
. ------;o~--
, ' , ~
and XII, the Refetence 8 data for Fresno are compared with 1967 data.
A 6% per annUlll increase is indicated; this apparent large increase in
collection rate probably reflects in part a more complete acquisition of
data in the later study. Where the data permit, the rates of increase
from the other listings have been calculated and are given. The references
cited were chosen because they presented new data or estimates for the
particular area concerned and did not merely repeat collection rates from
previous work.
These data represent year-around averages,
as was required for the present study. It was noted in examining the indi-
vidual sources that the per capita refuse collection rates varied depending
on the season of the year, and that the amount of variation depended on the
climate of the locality.
, The data presented in Table II-I and those
from other sources were analyzed by the EPA>:<. This agency then published
(Ref. 12) preliminary refuse collection figures, for the national situation
based on extrapolations involving some 45% of the population. These data
are shown in the following table. '
ESTIMATED AVERAGE SOLID WASTES COLLECTED
, (pounds/capita-day)
Solid Wastes Urban Rural National
Household 1. 26 0.72 1. 14
Commercial 0.46 0.11 0.38
Combined 2.63 2.60 2.63
Industrial 0.65 0.37 0.59
Demolition, 0.23 0.02 O. 18
Construction
Street and Alley 0.11 0.03 0.09
Miscellaneous 0.38 0.08 O. ,31
Totals 5.72 3.93 5.32
Because industrial and demolition/ construction wastes do not usually find
their way into the collection system used for handling mixed municipal
(urban) refuse, a per capita value of 4.8 lb/day appears to be a reasonable
reference point. In the present study, however, the base year from which
projections were established was 1965. The above survey data was collected
several years later and thus includes a small increase due to increased per
capita generation rate. It was felt therefore than an arbitrary adjustment
to conform with the frequently used 4. 5 lb/ cap. -day rate would be acceptable.
~
-------
L
I
This value, incidentally, is also in reasonable agreement with a more recent
estimate (4.0 Ib/cap. -day) published by A. D. Little Co. workers (Ref. 13)
involved on an EPA program related to the present one.
This collection rate multiplied by the population
of the United States gives the total quantity of mixed municipal refuse available
in the U. S. For the years 1964 and 1966, the population was estimated (Ref.
14) by the Bureau of Census as 194,583,000 and 196,842,000, respectively.
At the base point year for the study, the population would have been about
195 million people. This furnishes, as a preliminary value, a refuse output
for the Nation of about 160 million tons for the year 1965.
(2 )
Projected Collection Rates
The future increase in the collection of mixed
municipal refuse will be a function of the national population growth and the
urban waste produced per capita.
The Census Bureau is projecting the population
of the country through 1990 on the basis of four estimates of completed fer-
tility or gros s reproduction rate. The maximum fertility rate used corres-
ponds approximately to that for the U. S. in the 1962 to 1965 period, as it is
expected that the birth rate will decline. For the purposes of this study, the
data in Reference 14 were smoothed; a probable intermediate value was then
estimated and extrapolated to the year 2000. The maximum and minimum es-
timates at the year 2000 proved to be no more than 10% different from the
probable intermediate value, which has been used for this study. It reflects
an intermediate level of fertility decrease during the next 30 years and in-
volves an average annual growth rate for the period of 1. 52%. These esti-
mates are shown in the following table.
PROJECTED U.S. POPULATION
(Millions of People)
Year
Minimum
Most
Probable
Maximum
1965
1970
1975
1980
1985
1990
1995
2000
205
215
228
242
256
273
293
195
207
221
238
257
278
301
326
209
228
250
275
300
330
361
In projecting the refuse collection rates per
capita, the assumption may be made that the portion of the gross national
product (GNP) devoted to personal consumption expenditures is a reasonable
measure of the rate of collection of solid waste.
II.. 4

-------
In the following tabulation, data from Reference
15 have been used to calculate the mean annual percentage growth rate of seve-
ral components of the GNP. These data are per capita, in constant (1958)
dollars, so that their growth rates should be measures of individual consump-
tion of real goods requiring ultimate disposal.
RECENT GROWTH IN CERTAIN GNP ELEMENTS
 Value, Constant Dollars Mean Rate of Increase
  per Capita   %/Yr 
 1950 1960 1964 1967 1950-67 1960-67 1964-67
Gross National Product 2342 2699 3025 3361 2.15 3. 19 3.58
Personal Consumption 1520 1749 1945 2160 2.09 3.06 3~ 56
Expenditures
Durable Goods 229 248 307 362 2.74 5.55 5.65
Non-Durable Goods 752 828 886 969 1. 51 2.28 3.03
Services 539 673 752 829 2.57 3.02 3.30
In applying these data to the present projections,
it is clear that the increase in the per capita consumption of non-durable goods
has the greatest bearing on the future dimensions of urban refuse resources.
As shown in Section Il,A, 2, products from this classification must constitute
the bulk inputs that go to form the typical compositions of mixed municipal
refuse. Constituents arising from non-manufactured sources, e. g., garden
wastes, are also important and are probably not undergoing a per capita pro-
duction increase in view of the growing trend to apartment habitation. This
factor suggests that conservatism be observed in estimating the rate of in-
crease of per capita urban refuse production based on GNP indicators.
As seen in the above table, the increase in the
per capita consumption of non-durable goods was about 1. 50% per year for the
period 1950 to 1967. This period included cycles of both economic recession
and growth, and is thus reasonably representative.
On the basis of the foregoing considerations, it
was decided that an annual growth of 1. 5% in quantity of mixed municipal refuse
(collected) per capita be assumed as the most probable.
The oncoming era, however, may well see an
even stronger trend in the marketing of shorter-lived non-durables and of
disposable items previously of the durable clas s. The arbitrary higher
rates of 2.0 and 2.5% were therefore also included in the calculations. On
projecting these rates of increase for a growing population from the base
date to the year 2000, the urban refuse collection quantities shown in Table
II- 2 re suIt.
II- 5

-------
TABLE II-2
ESTIMA TES OF FUTURE URBAN REFUSE
COLLEC TIONS IN THE U. S.
 Probable U. S.   Urban Refuse Collected  
 Popula tion,    nnua rate} 6 
 Millions of Lb/cap.-day of increase Nationwide, 10 tons/yr
Year People 1. 5o/c"- 2. Oo/c 2. 5o/c per capita 1. 5o/c"- 2.0o/c 2.5o/c
-  -
1965 195 4.50 4.50 4.50 160 160 160
1970 207 4.85 4.97 5.09 183 188 192
1975 221 5.22 5.49 5. 76 211 221 232
1980 238 5.63 6.06 6.52 245 263 283
1985 257 6.06 6.69 7.37 284 314 346
1990 278 6.53 7.38 8.34 331 374 423
1995 301 7.03 8.15 9.44 386 448 519
2000 326 7.58 9.00 10.68 451 535 635
"'I'
"'Probable rate
II-6

-------
b.
Selected Metropolitan Areas
In order that the present program would be aligned
to specific rather than generalized application opportunities, six large
metropolitan areas (LMA IS) were selected for closer scrutiny. The selec-
tion was logically based on the S02 problem operating in the various LMA's
and the availability of a population density and urban configuration that would
be conducive to refuse disposal by the method considered herein.
The listing of cities in Table 11-3 includes all those
credited with severe or very severe (rating 4 or 5 on an arbitrary scale of 1
to 5) S02 pollution problems in Reference 16. That document summarizes
more detailed data from the NASN. CAMP and cooperating state and local
networks (for example, Ref. 17) and serves the desired purpose of pinpointing
those cities with S02 pollution problems. Also indicated are the ratings for
total suspended particulate matter on the same arbitrary scale of 1 to 5.
These S02 pollution ratings are of greater value for this task than the actual
numerical data, as they include effects of topography, meteorology, density
and location of sources, and intensity and distribution of days with high S02
(even if only on a qualitative basis). The last column gives the 1965 population
(Ref. 14) of the as sociated Standard Metropolitan Statistical Area (SMSA).
From Table ll- 3 the cities underlined were selected
as having solid waste pollution problems, having substantial background data
available, and having different enough characteristics to make their inclusion
in the study worthwhile. Several cities were excluded for reasons not apparent
from the table. Newark posed uncertainties in the projection of future grow~h
and the direction of change; East Chicago-Gary suffers more from S02 pollution
from industrial activity than from power generation; the same is probably true
of Jackson, Seattle, and El Paso. From among the remaining three smaller
metropolitan areas (New Haven, Wilmington, and Providence) which are similar
in terms of the parameters 6f interest, New Haven was selected.
. In order to predict the quantity of refuse that will be
collected in the selected LMA's, data was obtained and/or examined from the
following sources:
.
Regional Planning Agency of South Central
Connecticut, Private Communication
.
liThe Next Twenty Years, II The Port of
New York Authority, August 1966.
.
Greater Philadelphia Chamber of Com-
merce, Private Communication
.
Chamber of Commerce of Metropolitan
St. Louis, Private Communication
11-7

-------
TABLE II-3
LARGE METROPOLITAN AREAS HAVING
SEVERE S02 POLLUTION PROBLEMS
Core City
Rating
502 Pollution
Rating Total
Suspended
Particulates
SMSA 1965
Population
(Thousands )
New Haven, Conn.
5
3
704
Wilmington, Del.
5
5
468
Washington, D. C.
4
3
2,408
Chicago, Ill.
5
5
6,689
596
East Chicago, Ind.
5
5
Indiana polis, Ind.
4
5
984
Covington, Ky.
4
4
1 347>''<
,
Baltimore, Md.
5
5
1,854
Boston, Mass.
5
4
3,205
Jackson, Miss.
4
2
137
St. Louis, Mo.
5
5
2,249
1,851
Newark, N. J.
5
4
New York, N. Y.
5
5
11,366
Cleveland, Ohio
5
4
Youngstown, Ohio
4
5
2,000
523
Philadelphia, Pa.
5
5
4,664
Pittsburgh, Pa.
4
5
2,372
Providence, R.1.
5
4
739
E1 Paso, Tex.
4
5
344
Seattle, Wash.
4
2
1,179
';
-------
.
Abstract of "Metropolitan Populations to
1985: Trial Projections, " Rand Corpora-
tion
.
"Statistical Abstracts of the U. S. , " vanous
years.
. . From these data populations projections were made,
as shown in Table II-4. It can be seen that a considerable difference in growth
rate is anticipated for the various cities and for the individual cities them-
selves over the period of the projection.
Projections of the refuse collected in each of the
selected LMA1s are presented in Table II-5. These data are based on the
population predictions shown in Table II-4, an average collection rate per
capita-day of 4. 5 lb as of the base year of 1965, and an increase in per
capita urban-refuse collections of 1. 5, 2.0, and 2.5% per year over the
entire period 1965 to 2000. In comparing the refuse quantities produced
by the 6 LMA's (Table II-5) for the years 1965 and 2000, it can be seen
that the smallest predicted increase will entail a factor of 2.4. This is
for the estimated slowest growing LMA (New York) when using the lowest
per capita increase in refuse collection (1. 5%). The greatest increase
factor is 5.6. This is for the predicted fastest growing LMA (Washington,
D. C. ) when using the highest per capita increase in refuse collection (2. 5%).
2.
Refuse Composition
a.
Analysis of Current Refuse
Information on the composition of refuse is quite
limited, although the work of J. M. Bell and as sociates at Purdue (Ref. 18
and 19, and others), is notable for its completenes s. Kaiser (Ref. 20 and
21) has taken data. such as Bell's on the proximate analysis of refuse, and
has gone on to provide heats of combustion of the various components or
classes of Inaterials found in refuse. The composition and combustion
properties of the refuse hence can be determined readily by computation
once the given refuse has been hand- sorted to determine the proportions
of the components.
The refuse compositional breakdown used for the
present work was based on an averaging of data given by Kaiser in Ref. 22
and by EPA1s Daniels in a private communication (Ref. 23). The latter
document included data compilations from nine other sources. The per-
centages shown in the following table can be considered to represent values
for the year 1965.
II-9

-------
TABLE II-4
POPULA TION PROJEC TION FOR SIX SELEC TED CITIES
  Year - Population in Thousands 
LMA 1965 1970 1980 1990 2000
Chicago, Ill. 6,820 7,500 9,200 11 , 140 14,200
  (1. 91) (2.04) (1.93) (2.45)
New Haven, Conn. 490 545 660 750 820
  (2. 15) (1.93) (1. 29) (0.90)
New York, N. Y. 11 , 300 11, 900 13,000 14,500 16,000
  (1. 04) (0. 78) (0.80) (0.99)
Philadelphia, Pa. 4,660 5,090 6,030 7, 160 8,550
  (1. 78) (1. 70) (1.73) (1. 79)
St. Louis, Mo. -Ill. 2,260 2,470 2,980 3,670 4,500
  (1. 79) (1.89) (2.11) (2.06)
Washington, D. C. 2,300 2,620 3,380 4, 300 5,500
  (2.64) (2. 54) (2.44) (2. 49)
NOTE: Average annual percent growth for each period is shown in parenthesis.
II- 10

-------
       TA 3 :., ~ II-5   
    PROJECTED QUANTITIES OF REFUSE COLLECTED IN SIX SELECTED CITIES 
     Refuse Rate    1 0 3 tpd 
    Inc rease per cap. ,  Refuse Collected, 
  LMA   o/c~:c 1965 1970 1980 1990 lOOO
 Chicago, Ill.  1.5 15. 3 18.2 25.9 . 33.1 53.8
     2.0 15.3 18.6 27.9 37.4 63.9
     2.5 15. 3 19. 1 30.0 42.3 75.8
 New Haven, Conn. 1.5 1.1 1.3 1.9 2.4 3. 1
     2.0 1.1 1.4 2.0 2.8 3. 7
     2.5 1.1 1.4 2.2 3. 1 4.4
 New York, N. Y. 1.5 25.4 28.9 36.6 47.3 60.6
.....     2.0 25.4 29.6 39.4 53.5 72.0
.....    
I    
......         60.5 
......     2.5 25.4 30.3 42.4 85.4
 Philadelphia, Pa. 1.5 10.5 12. 3 17.0 23.4 32.4
"     2.0 10.5 12.6 18. 3 26.4 38.5
     2.5 10.5 13.0 19. 7 29.9 45. 7
 St. Louis, Mo. -Ill. 1.5 5. 1 6.0 8.4 12.0 17. 1
     2.0 5. 1 6. 1 9.0 13. 5 20.3
     2. 5 5. 1 6.3 9. 7 15.3 24.0
 Washington, D.C. 1.5 5.2 6.4 9.5 14.0 20.8
     2.0 5.2 6.5 10.3 15.9 24.8
     2.5 5.2 6. 7 11. 0 17.9 29.4
 ~cProbab1e rate = 1. 50/c     

-------
ESTIMA TED COMPOSITION OF MIXED
MUNICIPAL REFUSE
Refuse
Mean Weight-%
(as received)
Garbage
Garden Wastes
Plastics
Glass
Metals
Paper Products
Residual (rags,
paint, wood, etc.)
20
12
2
12
10
38
6
b.
Projected Composition
The development of estimates describing the future
trend in the compositions of mixed municipal refuse was necessarily based
on very limited information. It was therefore necessary to make certain
assumptions and accept rates of change for specific components which had
been predicted by other workers (e. g., Refs. 24 and 25). From these bases,
equations were then derived for each class of constituent. These are shown
in Table 11- 6 and the results of their application to the data contained in the
above table are shown in Figure II-I.
c.
Seasonal Variation
Ah equation was fit to the data in Reference 26 to
prediCt the seasonal va:.:iation in the total residential (household) solid
waste. The equation is:
fT = -0.01 - 0.051 cos (~ N) + 0.021 cos (~ N) + 0.008 cos (; N)

- o. 003 sin (f N) - O. 02 sin (; N) + O. 01 sin (; N)
(II-I)
where
fT = percent change from the mean

N = months.
The total residential refuse for a given month then is
WTR = WTR (1 + fT)
where
W T R = the mean re s idential wa s te colle ded, Ib / ca p. - da y.
11-12

-------
Constituent
Garbage
Plastics
Glass
Metals
Plants and Gras.s
Re s idual
Paper
-...
TABLE II.-6

PROJECTED COMPOSITIONAL CHANGES
IN MIXED MUNICIPAL WASTES
Equation
fG = fl G exp (-0.0034 N)
fpL = f'PL exp (0.0052 N)
f1pL
fGL = f'GL - -Z- exp (0.0052 N)

f'PL
+~
flpL
fM = flM - ~ exp (0.0052 N)

flpL
+~
fpG = flpG
f = fl
R R
fpA = 1 - fiG exp (-0.0034 N)
flGL - flM - flpG -
f'pL - fIR
Basis
Ref. 24 - data ex-
tracted is for refuse
collected in Berkeley,
California, from 1952
to 1 967.
Ref. 25.
It is assumed that the
fraction of glass will
be reduced by one -half
the inc reas e in the
fraction of plastics.
It is assumed that the
fraction of metals will
be reduced by one-half
the increase in the
fraction of plastics.
It is assumed that there
. w ill be no change in the
fraction of plants and
gras s.
It is as sumed there will
be no change in the frac-
tion of re s idual. '.
The fraction of paper is
calculated to obtain a
balanc e
whe.re
N = number of months
£Xx = as received fraction of the particular waste

fl XX = as received fraction in 1965.
II- 13

-------
The method used to generate Equation II-I was to
fit curves to segments of the data and them corabine them using the Fourier
series. The equation for obtaining WTR was the same as that used to de-
rive the waste load projections in Table II-5; that is, for a 1. 5% per
capita-year refuse collection increase,
WTR = WITR exp (0.00125 N)
where
W'TR = total as-received residential waste load in 1965,
lb/cap. -day.
Assuming the balance, (rc)':<, of the mixed municipal.
refuse load does not vary seasonally, the total solid waste load becomes
W T = W TR (1 + fT) + W rc
Wrc = Wlrc exp (0.00125 N)
WT =. WITR exp (0.00125 N) (1 + fT) + Wlrc exp (0.00125 N)
(II- 2)
where
W T = total as - received solid waste load (function of time),
lb / cap. -day
Wrc = total as-received non-household solid waste load
(function of time), lb/ cap. -day
Wlrc = total as-received non-household solid waste load
in 1965, lb/cap. -day.
Using Equation II-I, computer data were obtained to
determine the seasonal variation in waste load. The results are shown in
Figure II-2.
d.
Heat Value
As the percentages of the various constituents in
residential waste vary with time (seasonally and secularly), so will the
average heating value. Data were not available to determine the seasonal
variation of each constituent; therefore, for the present work it was assumed
that the seasonal variation in the total residential waste load would be due to
the seasonal variation in plants and grass. The following expressions can
then be considered:
':
-------
6
4
2
c
<
o
..J
~ 0
CI)
<
3:
~ -2
z
2
to-
~ -4
a:
<
>
to-
Z -6
w
<.J
a:
w
a.
-8
-10
o 1 2


L JANUARY 1
fiGURE 11-2.
SEASONAL VARIATION IN WASTE lOAD
ll- 16

-------
L--------
fG
fGS = 1 + fT

fpL .
fpLS = 1 + fT

fGL
f G LS = 1 + f T

fM
fMS = 1 + f T

fpG + fT
fpGS = 1 + fT

fR
fRS = i + fT

fpA
fp AS = 1 + f T
where
fXXS = seasonal variation in the fraction of a given waste.
Since the constituent fractions must sum to 1:
W'TR exp (0.00125 N) (1 + fT) = W'TR exp (0.00125 N) (1.+ fT)

~GS + fpLS + fGSL + fpGS +

fRS + fpAS]
Applying the equations for constituent seasonal variation
and introducing a heat value,
h W T = WI TR exp (0.00125 N) fGhG + fpL hpL + fGL + fMhM +
(fPG + fT) hpG + fRhR + fpAhpA + Wlrc exp (0.00125 N) hIC
where
h =
mean heat value for solid waste on an as -received basis,
Btu/lb.
II- 1 7

-------
By introducing the moisture and ash fraction for each
constituent, heat value on a moisture and ash-free basis is obtained.
nMAF W TMAF . W'TR exp (0.00125) j fG [I - (MG + AG)] hGMAF +

fpL [1 - (MpL -t ApL)J hpLMAF -t fGL [1 - (MGL +

AGL)J hGLMAF t fM [1 - (~ + AM)J hMMAF +
(fpG + fT) [ 1 - (MpG t ApG~ hpGMAF + fR [ 1 - (MR +

AR)] hRMAF t fpA [I - (MpA tApA)] hPAMAF) +

W'rc exp (0.00125 N) [1 - (Mrc + Arc)J hrcMAF
where
hMAF = mean heat value on a moisture and ash free (MAF) bas,is, Btu/lb
= moisture fraction of a given constituent
MXX
AXX
= ash fraction of a given constituent
Applying the equations shown in Table rIA-6 for secular
time variation and defining 8XX = MXX + AXX:
hMAF =
W'TR exp (0.00125 N) j (1- f3pA) hpAMAF [ - f' G exp (-0.0034 N) - f' PL -

f'M - f'GL - flpG - fIR + 1J + (1 - (3G) hCMAF fIG exp (-0.0034 N) +


(1- (3GL) hGLMAF f' GL + [(1-{3PL) hpLMAF - (1- (3M) hM~F - (l:- .LJGL)


hGL~FJ [flpL exp (0.0052 N)J + [(1- 8M) hM~F + (1- (3GL) hGL~FJ


~I PLJ + [(1- 8M) hMMAF J flM + (1- (3PG) hpGMAF (il PG + fT) + (1- (3 R)

hRMAF fIR + (1- (3rc) hrCMAF Wlrc exp (0.00125 N)

WTMAF
(Il- 3)
II- 18

-------
where
hXXMAF = heat value of the particular waste on a moisture
and ash free basis, Btu/lb
!3xx
=
the sum of the moisture fraction and ash fraction
in the particular component of waste
Any known residential waste fraction, such as those shown in the table at the
end of Section IIA, 2, a, and seasonal variation, such as Equation II-I, can be
applied to Equation II-3. A value for non-household waste load is tabulated in
Section IIA, 1, a. The remaining unknowns in Equation II-3 are the respective
heat values, moisture content, and ash content. Based on data from References
22 and 23, average heat values, moisture content, and ash content were esti-
mated as shown in the following table. The data are later converted into ulti-
mate analysis form in Section III, B, where steam generator energy balances
are considered.
ESTIMATED HEAT, MOISTURE, AND ASH
CONTENT OF DOMESTIC REFUSE
Btu/lb MAF
Per cent
Mois ture
Percent
Ash
Garbage
Plastics
Plants and Grass
Residual
Paper Products
Metals
Glas s
9,300
14,800
9,000
7,000
8,500
Inert
Inert
70
o
70
15
5
2
o
2
2
15
Not shown in the above table are the values for the non-household class of
urban refuse; this was assu~ed to be 8500 Btu/lb (MAF) and 20% each in
. moisture and ash, as suggested by References 22 and 23.
Using the values of the above table, Equation II-3
was solved to yield the change in heating values, hMAF' from the year
1970 to 2000. The results are shown in Figure II-3.. .Computer data based
on Equation II-I were obtained for the years 1970 and 1971 to show the sea-
sonal variation in heat value. A plot showing the seasonal variation of the
as-received and MAF heating value is shown in Figure II-4 for the two-year
period. The variation is only about +2% on the as - received basis and in-
significant on the MAF basis. It will be noted that the two minima in Figure
II-4 occur at about the same times (July and August) as the maximum for
waste load shown in Figure II-2. The latter is largely due to seasonal.
variation in garden waste collections. An increase in the content of high
moisture garden waste in refuse would tend of course to depress the heating
value, but not in direct proportion.
II- 19

-------
1-
  10,000   
 .Q    
 <::::.    
 ~    
 ...    
 !XI    
 u.    
 <:(    
I-< !    
I-<    
I W    
N ::;:)    
0 ..J    
 <:(    
 >    
 t:?    
 Z    
 j::    
 «    
 w    
 :I:    
  9,000   
  1970 1980 1990 2000
  YEAR  
FIGURE 11-3. PROJECTED HEATING VALUE OF MIXED MUNICIPAL REFUSE
(MOISTURE AND ASH FREE)

-------
II)
..J
......
::J
l-
II) 8
..
o
...
w
::J
..J
«
>
~ 7
Z
~
«
w
J:
10
9
...
6
5
o
L
MAF HEATING VALUE
AS RECEIVED VA1.UE
8
10
4
6
2
4
12
MONTHS
6
JANUARY 1
2
8 10 12



DECEMBER 31 J
FIGURE 11-4. SEASONAL VARIATION IN HEATING VALUE
1I- 21

-------
The information considered in the present section
(II, A) will have an important bearing on the projections developed in Section
II, C. These concern the potential pollutant reductions that can be expected
by using mixed municipal refuse to replace a portion of the fos sil fuel fired
in power plants.
B.
COMBUSTION AND INTERACTIONS
1.
Introduction
Refuse is a readily combustible fuel much like other fuels
widely used in furnaces and stearn boilers, however much its handling before
combustion may differ. It can be and is being burned (principally in Europe)
with or without auxiliary fossil fuel in water-walled furnaces to generate high
pressure, superheated stearn, without excessive corrosion or other problems.
There is no reason that it could not be so burned in the United States.
There is no evidence for interactions between refuse and
fossil fuel combustion products. In particular, there is no evidence that
sulfur oxides from fos sil fuels are absorbed by ash from refus e. There is
good evidence, however, that even under efficient firing conditions only a
small portion (15 to 30%) of the sulfur in refuse appears in the flue gas,
contrasted to about 98% for fossil fuels; it is probable that fixed sulfur in
the refuse (e. g., CaS04 from gypsum wall board) accounts for this fact,
which is observed both in the U. S. and in Germany.
Refus e is at present burned on stirring-type grates of a
style long obsolescent for utility combustion of coal, and experience with
its firing in suspension is just now becoming available. The conclusion of
no interaction drawn above should remain valid for the suspension firing
of refuse.
There is good evidence that refuse exhibits greater ash
fouling tendencies than most U.S. coals, and means for avoiding this prob-
lem, both in de sign of new plants and ope ration of modified plants, mu st be
considered.
The properties of combustion products of combinations of
refuse and fossil fuel, the possible interactions and compatibility of these
products, and any pos sible synergistic effects on pollutant emissions are
important parameters to examine in the initial phases of any solid waste
combustion study. The proximate analyses of refuse vary widely (d.
Section II, A above) with the variations as great among samples delivered
to the same incinerator as from region to region. However, the ultimate
analyses are much less variable and the heats of combustion even less so.
As a fuel, refuse is much more variable than familiar fos sil or other waste
fuels, yet reasonably consistent in overall composition of combustion
II-22

-------
products. There is a wide variation in ash compositions, as there is in
coals, but where the ash of a coal from a given mine is reasonably constant
in composition, that from refuse varies from hour to hour.
z.
Gaseous Combustion Products
a.
Major Products
In a competently designed and operated furnace,
refuse burns to the same ultimate combustion products as do fossil fuels,
as indicated in the following tabulation:
REFUSE COMBUSTION PRODUCTS
Refuse Component
Major Combustion Product
Gaseous Solid
°
COz
HZO
COZ and HZO
C
H
Cl
HZO
NZ (+ NO and NOZ)
SOZ (+ S03)
HCl
CaS04' etc.
N aCl, etc.
Moisture
N
S
Other minor
Various
Various
Ash
Ash
Metal
Metal plus oxide
Glass and ceramics
Unchanged
The amount of carbon monoxide in the oxygen-
containing boiler exit gases is negligible. Its equilibrium concentration
at typical furnace exit temperatures of Zl 00 to Z3000F with 10% or more
excess air is under 8 ppm (Ref. Z7, Table I). Equilibrium concentration
at the adiabatic flame temperature may be some thousands of ppm, and
II-Z3

-------
values at the flame front in small flames may be up to 100 times that at
equilibrium, as they are in laboratory hydrocarbon-air flames (Ref. 27,
Figure 4). Reaction kinetics are sufficiently rapid, however, for the
carbon monoxide level to reach close to the low equilibrium values men-
tioned above as the gas cools to furnace exit temperatures (Ref. 27,
Appendix A-2) in large furnaces.
Carbon monoxide is a contributor to corrosion (see
I, D J 1, Appendix A) in areas where hot, under-oxidized, combustion gases
contact bare metal; that is, of course, especially troublesome where the
metal is boiler tubing under pressure. Thus, good design assures either
that adequate amounts of combustion air are thoroughly mixed with the
burning refuse, or that the water walls or boiler tubes in the part of the
furnace affected are covered with protective refractory. This problem is
particularly acute in grate-fired incinerators burning untreated, mixed
refuse, because of its variability. Suspension firing will allow adequate
mixing and will thereby permit operation with lower overall percentages
of excess air.
During combustion, combined hydrogen in the refuse
converts completely to water, as it does with fossil fuels. Again, as with
carbon monoxide, the factors of excess air, residence times, and furnace
exit temperatures are such that no elemental hydrogen is found in the com-
bustion gases leaving the furnace, regardless of what species exist in the
high temperature flame zone.
Oxygen in the organic part of refuse fuel will appear
in the flue gas as H20 or C02; its fate, as distinguished from that of the
oxygen in the combustion air, is unimportant. It is of course important
insofar as it determines combustion air requirements and heat of combustion.
The large oxygen content of refuse, primarily from the cellulosic constituents,
contributes both to the lower heating value and to the low theoretical air re-
quirement for refuse. In terms of pounds of gas per unit of heating value,
the quantities of either combustion air or wet flue gas are about the same as
for coal, however. This is shown in the following table, the basic data of
which are explained or developed in Section III, B.
COMBUSTION CONSTANTS FOR REFUSE AND COAL
   lb air / lb flue gas / HHV lb air / lb flue gas /
   lb fuel lb fuel Btu/lb 10 6 Btu 106 Btu
Refuse, stoichiometric 3.20 4.02 4,460 717 903
Coal, stoichiometric 9.75 10. 77 12,022 810 894
II- 24

-------
b.
Pollutants
(1 )
Nitrogen
Bound nitrogen in fuels appears to be burned
almost entirely to elemental nitrogen, although recent information indicates
the concentration of nitrogen in fossil fuels does somewhat affect nitrogen
oxides (NOx) levels in combustion gases (Ref. 28 and 29). These oxides
are primarily formed from nitrogen and the free oxygen in the flue gas, the
amount being determined by the concentration of oxygen from excess air,
the maximum temperatures in the flame zone, the temperature-time history
of the gases as they cool, and pos sibly other as yet unrecognized factors.
Within the furnace, nitrogen is first fixed as NO; with exces s oxygen pres ent,
a small portion of the NO is then oxidized to N02 during the cooling, the
amount again depending upon the form of the cooling rate curve (Ref. 30).
From a theoretical point of view, one would expect the concentration of NOx
from a conventional well-run refractory incinerator to be very much lower
than those from the refuse portion of a combined-firing boiler installation.
This is because the very high excess air of the former, and resultant low
flame temperatures, should have a depressing effect on their formation. .
However, comparative data in the literature suggest only factors in the
range of 1.5 to 4, on the rational basis of pounds of NOx per million Btu:
NOx EMISSIONS FROM VARIOUS TYPES OF FURNACES
Ref.
Nitrogen Oxides
lb /ton fuel lb / 1 Ob Btu
Municipal Incinerator

Average Incinerator

Water Walled Incinerator
Suspension Firing
31
32
2
3.03
(0.25)~<
0.34
32
3. 44~":<
O. 39~'~'
Coal Fired Boilers
Vertical
31
33
20
(0. 76 )~:'
Coal Fired Boiler s
Corner Fired
O. 55
'0. 71
0.95
O. 76
Front Wall
Spreader Stoker
Cyclone
2.20
0.59
Horizontal Opposed
~"Deri ved value.
~:'~'Estimate.
II- 2 5

-------
Refuse burned under the higher temperature
conditions of a water-walled boiler, especially in a combined furnace with
fos sil fuel, should give concentrations of nitrogen oxides comparable to
those from the same weight of fossil fuel. This does not mean, however,
that an increase in total' nitrogen oxide burden to the atmosphere would
occur on switching from conventional incineration to firing the refuse in
steam generators. The latter would involve a reduction in the amount of
fos sil fuel fired and thus in the amount of NOx emis sions, even though the
emissions from the refuse portion had increased. The reduction in NOx
emissions would be roughly equal to the percent of fossil fuel displaced
divided by the factor by which the NOx emissions from refuse had become
increased in changing from incinerator to boiler firing. Thus a 10% dis-
placement of coal by refuse would result in a 2. 5% to 6% reduction in total
NOx' based on the range of 4 to 1. 5, discussed above, for the NOx increase
factor for refuse. Similarly, no increase in NOx emissions would be in-
volved for the case of landfill vs steam generation. The NOx output from
power plants would be about the same whether refuse were fired or not.
(2)
Sulfur
(a)
Background
Sulfur is, of course, of major concern.
Economic fossil fuels (except natural gas which has limited availability) are
all relatively high in sulfur content, and probably above 98% (Ref. 31) of the
sulfur appears in the flue gas as sulfur dioxide with small, but not unimpor-
tant, amounts of sulfur trioxide (Ref. 33). It is of course one of the major
goals of this program to evaluate the potential reduction in sulfur pollution
by the substitution of refuse for fossil fuels in power generation.
u. S. refuse is truly a low-sulfur fuel,
averaging between O. 1 and 0.2% sulfur overall (Refs. 13, 31, 33, 34). One
investigator (Ref. 35) has measured flue gas concentrations from municipal
incinerators in New York City averaging the equivalent of 0.290/0 sulfur in
the fuel if it all appears in the flue gas. Individual measurements ranged
from 0.096 to 0.487%, typical of the variability commonly observed in
refuse analyses. Furthermore, half or more of the sulfur appeared as
sulfuric acid in all cases. This is a distinct difference from other ob-
servations (Refs. 33,36-38) in which the S03 is 1 to 3% of the S02, in
coal, oil, refuse, and combined firing plants.
With the single exception of Reference
35, the sources cited agree on from less than O. 1 to 0.2% sulfur in muni-
cipal refuse in the U. S. Kaiser (Ref. 33) has postulated that a substantial
fraction (ca 75%) of the sulfur in refuse remains in the residue and fly ash,
in contrast to coal. He suggested that refuse ash is sufficiently basic and
reactive to absorb appreciable qliantities of sulfur dioxide from the flue
II-26

-------
. ~I:: "'. ''''''~'. ",:i'.:
'\fj':!'~"> ~~ .y\~......'}.:;~ ':>.
" .
gases. If this were the case, there is the possibility of a synergistic effect
in the combined firing of refuse and high sulfur fos sil fuel to reduce the sul-
fur oxides ernis sion below that corresponding to a simple addition of the
emissions from the two fuels. It is considered highly unlikely that such
effect would be of any great significance in reducing sulfur emis sions from
boiler plants. The reasons are: .
1.
The wide variability in refuse bottom ash and fly ash
composition would prevent continuous effective results.
2.
The ranges of composition of refuse ash, combined-firing
ash, and coal ash are so similar that there is little like-
lihood of a substantial difference in behavior as to 502
absorption (see the following discussion).
3.
Proper combustion of refuse will expose the fly ash to
the high flame temperatures characteristic of good fos sil
fuel-fired utility furnaces, which will tend to overburn it
and deactivate any absorptive properties it may have.
Hence, any synergistic effect is expected to be small.
4.
Control of sulfur oxide emissions from utility plants will
eventually require 10 to 20-fold reductions in emission.
Any small synergistic effect of burning up to 50% (on a
heat basis) of refuse would still require a sulfur removal
system, whose cost would be more heavily influenced by
the volume of flue gas to be treated than by the initial sulfur
oxide level. Substantial reductions in capital cost for sulfur
removal equipment are possible in several combined firing
systems examined in this study, in which the flue gases are
kept separate until after treatment. This arrangement, of
course, precludes any synergistic effect.
(b)
European and U.S. Data
In Europe, the amounts of fly ash £rOIn
grate burning of refuse and from pulverized coal firing are comparable.
Although the largest portion (ca 85%) of the non-combustibles in the refuse
appear as grate residue, the amount of fly ash is still substantial because
of the generally much higher non-combustible content of European refuse
than coal. This is partly due to the presence of significant quantities of
coal ash (from domestic space heaters), a characteristic not typical of
U. S. refuse. Analysis of the Tbv data reveals that fly ash carry-over
on grate firing of refuse is in the same range as the total ash content of
coals; that is, 3 to 6% of the total fuel feed. There will thus be more fly
ash per Btu from refuse than from coal by a factor of approximately three>:'.
Examples of the magnitudes are given in the following table:

*As explained in Section II, C, this ratio does not apply in this country.
II- 2 7

-------
FLY ASH FROM REFUSE AND COAL COMBUSTION 
     Grate  
 Type of Fuel, Ib /hr Residue Fl y Ash
Installation Firing Refus e Fossil Ib /hr Ib /hr % of Fuel
DUsseldorf Refuse only 23,100  7,350 1, 100 4.8
(Ref. 36)       
Munich North 1 Test 1, coal  25,200  1,550 6.2
(Ref. 38) only      
Munich North 1 Test 5, coal 59,300 15,400 21,800 4,080 5.5
(Ref. 38) and refuse     
Stuttgart No. 28 Test 4, oil 53,000 12,080 12,500 1,590 2.4
(Ref. 37) and refuse     
Since the amounts of reactant added in the
various limestone and dolomite S02 removal processes are comparable in
magnitude with the fly ash from coal, the contacting efficiency of ash and 502
should be potentially adequate for some reaction; with suspension firing of
ground refuse, undoubtedly a greater fraction of fly ash will result. However,
dry absorption SO?-removal processes in general have efficiencies less than
50% (Refs. 39, 40), even with injection of a properly activated, 100% reactive,
agent into a low temperature area of the furnace to prevent deactivation.
Refuse fly ash is substantially the same as coal fly ash, except for its much
wider short-term variability and a tendency toward higher alkali and sulfur
content on the average (see Table II-7 and Refs. 33 and 41). The fly ash
samples on which the analyses of Table II-7 were performed were in part
collected for this study (the "A" series) and in part provided by A. D. Little,
who had performed some particle size analyses, but no chemical analyses
(Ref. 13). To be noted are (1) the strikingly high sulfate content of the two
II
refuse-only samples (Dusseldorf, Sample 7B, and New York, Sample IB)
compared with the coal only and the combined firing samples; (2) their general
similarity despite one being European and one U.S.; and (3) (except for sulfur)
the similarity of all the coal, combined and refuse-only ashes. These values
can be contrasted with the ones reported by Kaiser (Ref. 33) for New York
incinerators examined earlier than 1964, which are summarized in the fol-
lowing, each column being the average of four to seven analyses:
II-28

-------
  TABLE 1l.7. ULTIMATE ANALYSES OF FLY ASH SAMPLES FROM REFUSE AND COMBINED FIRING INSTALLATIONS  
  COAL ONLY  COMBINED FIRING  REFUSE ONLY  
    75"1, Coal 80'7, Coal CompoSlte b~"k 011 D~sseldorf    
       MunIch North It  New York  
  Munich North lA Munich North IB Munich North II Inlet Hopper Stuttgart Inlet Hopper Composite Hopper  
  Sample HlA Sample H3A Sample 115A Sample 115B Sample 117A Sample 117B Sample III B  
 SIlicon 23.0,,/, 17.00/, 18.0,,/, 21. 00/, 15.0"/, 17.0"/, 18.0,,/,  Silicon
 Aluminum 1-1.0 11. 0 12.0 13.0 10.0 12.0 6.9  Aluminum
 Iron 3.2 4.1 6.8 7.3 22.0 3.6 3.0  Iron
 Magnesium l.b 1.5 1.b 1.8 1.1 1.5 1.2  MagnesIUm
 Calcium 3.4 7.5 3.4 4.4 3.4 3.9 2.5  Calcium
 Sodium 0.81 1.4 I 5 trace trace 2.5 3 5  Sodium
 Boron 0.013 0.013 0.013 0.063 0.013 0.023 0.029  Boron
 Tin trace 0.037 0.051 0.12 0.14 0.35 0.71  Tin
 Be rylJium 0.0011 0.00096 0.00097 nil nil nil nil  BervlJium
 Barium 0.080 0.30 0.19 0.23 0.31 0.30 0.12  Barium
H Cadmium nil nil nil  nil nil nil 0.010  Cadmium
H Zinc nil 0.41 0.63 0.86 0.83 1.2 1.9  Zinc
I
N Phosphorus nil nil nil trace 1.0 nil 0.40  Phosphorus
...0 Antimony nil nil nil  nil nil nil 0.020  Antimony
 Titanium 0.55 1.1 1.0 0,57 0,86 1.2 2. I  Titanium
 Maneanese 0.082 0,089 0.11 0.38 0.3Z 0.12 0,086  Manganese
 Lead 0.083 0,14 0.15 1.5 0,087 0.85 2.5  Lead
 GalJium 0.0099 0,011 0,0082 trace trace 0.0091 0.0052  Gallium
 Tungsten nil nil nil  nil 0.25 nil nil   Tungsten
 Copper 0.016 0.020 0.020 O.OB 0,71 0.027 0.019  Copper
 Chromium 0.022 0.053 0,036 0.024 0.30 0.084 0.081  Chromium
 Nickel 0.022 0.015 0.018 0.025 0.061 0,015 0.019  Nickel
 Bismuth nil nil nil  nil nil nil trace  Bismuth
 Molybdenum nil 0.0054 trace 0.0071 0.020 0.0060 O. Oil  Molybdenum
 Vanadium 0,016 0.014 0.011 0.0044 0.0045 0.0095 0.065  Vanadium
 Zirconium 0.016 0,013 0.021 0.0022 0.0067 0.016 0.013  Zirconium
 Cohalt 0.013 0.014 0.018 O. 0.0 73 0.026 0.010 0.0038  Cobalt
 Yllriutn nil trace trace nil nil nil nil   Yltriurn
 Potassium 3. 3 6.0 5.5 trace nil 6.5 9.6  Potassium
 Strontium 0.046 0.036 0.033 0.053 0.030 0.039 0.019  Strontium
 Silver 0.00035 0,0029 0.0021 0.0034 0.0036 0.0030 0.00024 - Silver
 S04= 2.06 5.09 3.62 6.93 2.45 12.28 II. 91 S04=
 503= 0.15 0.07 0.03 O.O~ 0.01 0.06 0.04  503=

-------
INCINERATOR FLY ASH COMPOSiTIONS
    Sampled in Sampled in Sampled
    Suspension in Dust Collection from
    Hot Flue Gases Zones Stack Gas
Organic, %  31. 0 3.9 21. 1
Inorganic, % 69.0 96.1 78.9
Ino r g. Fraction, %   
 Si   22.1 22.8 17. 1
 Al   5.4 12.3 13. 7
 Ti    0.6 0.4
 Fe   10.9 4.5 5.0
 Ca   ' 13. 2 6.6 6.4
 Mg   1.7 1.4 1.7
Na + K(as Na) 3.3 4.3 7.8
S (as SO 4 -)  1.4 3.6 9. 1
Evidence from ash analyses for or against
the large-scale absorption of sulfur oxides by refuse fly ash is not available.
Thus, the observed larger sulfate content of some of the refuse-only fly ash
samples is more plausibly explained as resulting from substantial quantities
of inert sulfates in the refuse, such as would arise from the discard of plaster
of Paris artifacts and the like. For example, the incineration of wall board
will decompose the gypsum to anhydrous CaS04 (anhydrite) and steam, with
the probable suspension of large amounts of finely-divided anhydrite.
Sulfur balances were derived for the five
European steam generating plants surveyed. Because of the low sulfur con-
tent of the coal (averaging 0.7%) and the high sulfur content of the refuse
(averaging 1. 9% with a spread from 0.7% to 3.0%) fired in these European
tests, the results cannot be directly applied in the U. S. The high refuse sulfur
content, incidentally, shoulci not be considered as typical for German refuse;
due to the season, considerable coal ash from dwellings was present in the
refuse. Some of these sulfur balances were based on samples collected at
different times from the boiler under consideration or from data on a com-
panion boiler, and to that extent are uncertain. (This is the case for the
two Stuttgart combined firing tests and the refuse-only DUsseldorf test. )
The eight balances reported for Munich North land 2, including both the
highest and the lowest refuse sulfur contents, were based upon sulfur values
(for fuel, residue, fly ash, and stack gas) reported in the TtJV Boiler Test
reports (Refs. 38 and 42). The sulfur balance tabulations for the five units
are given in Table II-8. They may be summarized as follows:
II- 30

-------
TAB.~
SULFUR BALANCES IN GERMAN JtEFUSE-FIRED STEAM GENERATORS
 Sulfur Analysis  Sulfur Distribution Sulfur Analysis Sulfur Distribution  Sulfur Anal ys is Sulfur Distribution  Sulfur Analysis  Sulfur Distribution Sulfur Analysis Sulfur Distribution Sulfur Analysis Sulfur Distribution
 O/C Wt. ppm  Ib/hr  "7c "Ie Wt. ppm Ib/hr    O/C  o/cWt. ppm Ib/hr  O/C  "Ie Wt. ppm Ib/hr   o/c  O/C Wt. ppm Ib/hr  O/C % Wt. ppm Ib/hr  "Ie
Test Number Dusseldorf Unit No. 1 Munich North Unit No. 2, Test 1   Munich North Unit No. 2, Test 5  Munich North Unit No. 2, Test 6  Munich North Unit No. 2, Test 7 Munich North Unit No. 2, Test 4
Refuse 0.78':'   180.0  100.0 -  0    0  2.76"'(1)  2599.2  87.3  3.0  2602.0   88.5  2.3  1427. 0  93.1 2.19'"  2180  100.0
        0.69  516.0   100.0  0.66  378.1  12.7  0.6  338.8   11. 5  0.6  105.6  6.9 -  0  0
Coal                       ~ Imr:lJ   T'51Z:b  TmJ.1Y     
Total Input          5Tb:O   TmJ.'O                nmr  TmJ.1Y
Flue Gas                     36.4   891                 
S02  492  53.2  29.5  483 500.4    91. 9   941 1082.9    983. 7   33.5   900 496.3  32.4   358  16.4
SO  6  2.2  1.2  10 12.3    2.3   7 7.9  0.3   4.5 4.9   0.2   7 3.8  0.2   11  0.5
Flyas1 4.25   47.0  26.0 1. 32  31. 6    5.8  2.68  75.2  2.5 I n.m.         n.m.     3.61  55  2.5
Residue 1. 06   77.6  43.3 -  -    -  5.19  1811. 3  60.8 : n.m.         n.m.     4.81  1760  80.6
Total Output    T81J.lJ  T01J.l)   ~   TmJ.1Y    '[97 r:-1"  TmJ.'O                  ZT!S4  TmJ.1Y
Difference    0     +28. 3        0                       
Test Number Stuttgart Unit No. 28, Test 1  Stuttgart Unit No. 28, Test 4   Stuttgart Unit No. 29, Test 1  Stuttgart Unit No. 29, Test 4             
Refuse -   0  0 2. 7;~)  1439.8    88.1  -  0   0  2. 1*  1018.3   83.7           
Oil 1. 25   242.5  100.0 1.6  195.3    11. 9  1. 17  230.2  100.0  1. 47"'(2)  198.8   16.3           
Total Input    Z4T.'5'  T01J.l)   T635':T   ~'    -z3'Q."Z  T01J.l)    TZT7:'7 11ft!:\)           
Flue Gas             Total Refuse             Total Retuse          
S02 (Air Heater) ,  -       595 (237.7)       - -   -   n.m.                 
                  -                
S02 (Refuse furnace)  -  0  0  105 20.3   1.2  1.4   0      105 18.7  1.5   1.9          
SOt (Oil furnace)  811  242.5  100.0  945 195.3   11. 9  -  742 230.2     945(3) 198.9  16.3   -          
To al    Z4T.'5'  T01J.l)   ZT5:'b'   ---rT.7 --r:4   -z3'Q."Z  100.0    --zT"7':b  "'T7:lS  --r9'          
Flyash  -  0  0 6.1  84.4   5.2  5.9   0   0  12.2  182.5  15.0   17.9          
Flyash (Deposited)  -  0  0 10.3(4)  43.2   2.6  3.0   0   0  10.3  48. 1  4.0   4. 7          
Residue (Dry)  -  0  0 10.4  1291. 9   79.0 89.7   0   0  5.5  769.0  63.2   75.5          
Residue (Sluice Water)  -  0  0           0   0  -  -   -             
   Z4T.'5'  TmJ.1Y   T'bTI:T   TOo.lY Tmr.lY   G'31J.Z'  TOo.lY       -          
Total Output                  TOo.lY  TOo.lY          
Test Number Munich North Unit No. 1, Test 1 Munich North Unit No. 1, Test 5   Munich North Unit No. 1, Test 6                    
Refuse -   -  - 0.67  382. 7    75. 7  O. 70  402.8  100.0                    
Coal 0.84   211.6  100.0 0.80  123.0    24.3  -  0   0                    
Total loput    ZTT:l)  TOo.lY   '5TI;:(   TmJ.1Y    ~  TmJ.1Y                    
Flue Gas   -I                                      
S02  500 194.7  97.3  383 143.5    31. 9   190 51. 4  14.0                    
Flyash (Economizer) 0.54/1. 96      0 2.26/2.68  4.6    1.0  2.40/2.88  29.3  8.0                    
Flyash (Precipitator) 0.38/0.40   5. 3  2.6 2.94/3.15  113.5    25.2  4.02/4.12  146.4  39.8                    
Residue -   -  - O. 76  188.0    41. 8  0.6  1 3f;. 0  37. 5                    
Bottom Ash 0.04/0.04   0.1  0.1 0.08/0.21  0.1    0.0  2.40/2.42  2.6  O. 7                    
Total Output    zuu.T  T01J.l)   ~   T01J.l)    ~  TOo.lY                    
Difference    -11.5     -55.9        -35.1                      
-, By difference.
(I) 2.9 in TIiv report.
(2) Sulfur in oil reported as 1. 1 -1. 50/c.
(3) Value measured in Unit No. 28.
(4) Assumed from analysis of deposits from Unit No. 29.
II- 3 1

-------
Refuse Only
15 to 30% of the input sulfur appears in the
flue gas
Combined Firing
With (high- sulfur) oil, 13 to 18% of the input
sulfur appears in the flue gas
With (low- sulfur) coal, 32 to 36% of the sulfur
appears in the flue gas
Fossil Fuel Only
Over 92% of the input sulfur appears in the
flue gas
Examination of the data for combined
firing with oil (Stuttgart Unit 28, Test 4, Table II-8) reveals that the sulfur
in the flue gas after mixing (S02 samples at the air heater) is, if anything,
higher than the total emerging lrom the two separate furnaces. Thus there
is evidence against any uptake of sulfur dioxide by refuse fly ash.
In the Munich North II combined firing
tests it can be noted that the proportion of input sulfur in the flue gas is the
same whether the boiler was at full, partial, or minimum load, with the
same refuse rate, and with varying coal rate. Figure II-5, redrawn from
Reference 38, gives the S02 (and HCl) concentrations in the flue gas for all
seven tests of Munich North II; this figure vividly illustrates the statement
above, that European results are not directly translatable to U. S. conditions;
as the flue gas S02 concentrations are higher with combined firing or refuse
only than with coal only. These data refer to dry flue gas; with moisture ac-
counted for, the difference is not quite so large; but the trend remains. The
Munich North II tests were run with high-sulfur refuse; the data of Tests 1,
5 and 6 on Munich North I (where the sulfur concentrations of coal and refuse
were about the same, but the exces s air for refuse-only was greater) indicate
that the trend is not necessarily a consistent one.
Data on the Munich units can be used to
examine the possibility of excess sulfur-absorbing activity by either refuse
or coal fly ash. As will be apparent, sulfur dioxide in the flue gas is the
same as or higher within the estimated errors with combined firing than
would be calculated from a linear combination of the data for coal only and
for refuse only, thus firmly suggesting the absence of synergistic effects.
Data calculated from information in the boiler test reports (Refs. 38 and 42)
was used to prepare the following tabulations which show these facts:
II- 3 2

-------
    ~S02    
  1100   HC1    
  1000     ~
      ~
  900     ~
  800     ~
 E     
 a. 700     ~
 a.    
 ..        .~
 Z       
 0 600   w   
 I-    ....J   
 «    I:C  ~
 0::: 500   « 
 I-   ....J 
~ Z    «  ~
w  ~ 
I U  > 
VJ 400 ~ ~ 
VJ Z ~  ~
 o  
 u 300 ~ ~ ~  ~
  200 ~ ~ ~  ~
  100 ~ ~ ~  ~
  o ~ ~  ~
  TEST . 1 '2 3 4 5 6 7
  LOAD FULL LOAD PART LOAD LOW LOAD FULL LOAD PART LOAD LOW LOAD LOW LOAD
FUEL I-
100% COAL
+
44t/hr REFUSE + COAL
...1 44t/hr REFUSE
FIGURE 11-5. S02 AND HC'I IN DRY FLUE GAS, MUNICH NORTH II BOilER

-------
FLUE GAS S02 QUANTITIES
Munich North I:
  Input S (as S02),
Test No. Fuel lb/hr
1 Coal only 212
5 Coal 123
 Refus e 382
6 Refuse only 401
Flue Gas 5 (as 502),
lb /hr
Calculated
Found
194
112 ]
49
161
143
51
Munich North II:
1
5
Coal only
Coal
Refuse
515

377

2590
338
2600
500
Refus e only
105

1427
2180
366 I
425

328 I

425

102 I

332
753
985
791
1083
6
Coal
Refus e
7
Coal
Refuse
435
495
4
356
(3 )
Halide s
Halides are of concern to boiler designers
because of the corrosion problems associated with acidic condensates from
the flue gas in the low temperature portions of the system, and postulated
connections between halides and high temperature tube wall corrosion. Or-
ganic halides, from polyvinyl chloride (PVC) plastics, primarily, will burn
quantitatively to HC1; inorganic halide compounds mayor may not liberate
HCl on combustion depending on the environment. There is some fear that
increasing use of PVC plastics will cause problems of boiler corrosion as
their percentage in refuse increases. 50 long as well-mixed municipal
waste is considered, problems are not expected in the near future. The
chlorine content of refuse is in the high end of the range for coals now burned,
as tabulated below: .
II- 34

-------
CHLORIDE CONTENTS OF REFUSE AND COAL
Sample
Reference
Cl Content, wt-%
(as-received basis)
Refuse
City of St. Louis
Atlanta, Ga.
Or lando, Flo rida
Hous ehold
41
34

34
35
35
43
0.13, 0.15,. O. 32
0.5,0.6,0.3,0.3
0.2, 0.4, 0.2, 0.2
0.42, 0.25, 0.13
0.11, 0.33, 0.07
DeKalb County, Ga.
73rd St. Incin, NYC*
Other NY C*
Mixed Household-
commercial
0.29,0.23, 0.24
0.29
Trash
Commercial
O. 16
0.63, 0.26, 0.47
Coal
Union Elec. Co., St. Louis
41
0.03, 0.04, 0.05,
0.04, 0.29, 0.29

0.01 to 0.46 (typical
range of U.S. coals)
Various
44
':'Minimum calculated from flue gas
concentration of HCl.
. Values of the HCl content of the flue gas
were reported for six of the seven tests at Munich North 11 (Ref. 38). They
are shown in Figure 11-5 on a dry gas basis. It is seen that the values for
coal only are less than 100 ppm, whereas with combined firing at full or
part load, HCl values are two to three times higher, and for refuse only
they are six times higher. One can calculate a minimum value for the
chloride contents of the fuels from the flue gas concentrations and from
the flue gas and fuel flow rates, as indicated in Table 11-9.
It is apparent that the minimum chloride
content of the refuse based on flue gas content is in the same range as that
of mo st U. S. refus e chloride contents; further, the minimum chloride content
of the coal is in the range for U. S. coal. The expected flue gas content of
HCl from combined firing (calculated from the amounts for coal only and
refuse only) is also approximately the same as that actually observed. This
fact indicates no .appreciable difference in absorption of HCl by fly ash from
refuse and coal.
Il-35

-------
      TABLE II-9    
      HCl EMISSIONS. MUNICH NORTH II   
      (Values in parenthesis are derived)   
 Test No.     1 2 5 6 7 4
 Condition     Coal Only Combined Firing  Refuse Only
 Fuel Rate          
 Coal. lb/hr   74.500 52,300 57. 100 56,400 16,700 
    6  956 685 736  219 
    l a Btu/hr 730 
 Refuse, lb/hr    94,000 86.500 88,600 100,000
    106 Btu/hr   254 235 235 276
 Flue Gas. 103 SCFD/hr  13. 150 9. 750 15.080 14.480 7. 720 4,950
 Lb Flue Gas /lb Fuel 13.5 14.0 (7.1) ( 7. 7) (5.6) 3.8
 Excess Air. O/C  29 31 38 35 47 45
H           
H           
I C02' %   14.40 14. 15 14.10 14. 10 12.60 13.20
VJ  
0'        
 H20, %   7. 73 7.52 10.50 10.80 18.50 24.30
 HCl in Flue Gas. Measured      
 10-6 Ib/SCFD  8.0 7. 1 27.2 22. 1 44.1 58.5
 Ib/hr     105 69 410 320 340 290
 lb/l 06 Btu   O. 11 O. 10 0.41 0.33 0.75 1. 05
 Min. Cl- in Fuel (calc.), wt-o/c ,0.137' 0.125,    0.28
 0.131   
 HCl in Flue Gas, Expected from:      
 Coal, Ib/hr     77 76 23 
 Refuse. lb/hr    273 251 257 
 Total. Ib/hr     350 327 280 

-------
million Btu heat input to
seven-fold greater HCI,
coal only.
On the rational basis of lb HCI per
the boilers, the combined firing gives three to
and refuse gives only ten-fold greater HCI than
Other minor constituents of refuse that
might be gaseous pollutants, notably lead and arsenic, would not be in the
vapor phase. under oxidizing conditions at boiler exit temperatures, and
would be controlled to the extent suspended particulates were controlled.
Their contributions to possible tube wall corrosion are discussed at length
in the corrosion literature.
3.
Solid Combustion Products
a.
Re s idue
Conventional incinerators are operated
primarily as tools to reduce the volume of solid waste, thus prolonging the
life of the landfill. To this end, complete burnout is not neces sary and many
overloaded incinerators do not achieve it. . When sanitation regulations specify
the organic or putrescibles content of the residue, as will soon be the case
universally, then complete burnout becomes more important. Electric
utility companies, on the other hand, are very sensitive to fuel values in
ash or clinker, and can be expected to be more concerned. :from the point
of view of interactions between refuse and fossil fuel combustion, it can be
stated that there is no possibility of the residue on the refuse grate affecting
the combustion products of powdered coal or other fossil fuelS. 1£ refuse is
ground and fired partially in suspension, ther.e is a somewhat greater
possibility of interaction. Only the finely-divided material will react
appreciably; this will be fly ash, and based on present analyses and obser-
vations it is not enough different from coal ash to exert any effect.
b.
Fly Ash and Suspended Particulates
Discussion of possible interactive
effects with examples of fly ash analyses were presented in the preceding
section on sulfur oxides. One other item of importance is the organic and
carbon conte"(lt of fly ash. The former is of concern in relation to disposal as
Inentioned under residue (above), while the latter is of concern both because
of its potential fuel value and its effect on electrostatic precipitator efficiency.
Heat losses due to the carry-over of
organic matter or carbon in fly ash is not a serious problem with coal
fired in suspension. It can be probably then assumed that the heat loss
in fly ash from refuse burned on grates will also be small. This is
because the fly ash from coal and refuse are quite similar and the quantities
generated are roughly comparable, whether on an energy input (U. S. ) or
weight input (Europe) Basis.
II- 37

-------
A particular effect of combined firing of refuse and
fossil fuel is that on the performance of electrostatic precipitators. This
is discussed in Appendix B; in summary, though, the electrical properties
of refuse fly ash are enough different from those of coal to influence the
performance of electrostatic precipitators. A similar result is obtained
with low- sulfur coals. Data from European combined firing incinerators
with electrostatic precipitators have been instructive, but because of the
much lower sulfur content of the coals used, and the higher sulfur content
of refuse, the data are not directly translatable to U. S. conditions.
It appears likely that when both sulfur oxides control
and particulate control are required by regulation, the cost effective solution
may be a wet scrubber system adequate to remove both sulfur oxides and
suspended particulates (Sec. III, B). .
c.
Ash Deposits and Corrosion>:<
A comprehensive examination of the literature on
ash deposits and corrosion was made in conjunction with laboratory analyses
and physical examinations of ash deposits on tubes and evaluation of their
relation to corrosion in water-walled refuse and combined firing installa-
tions. The corrosion and deposit problem was brought out as a result of
the publication of a series of articles in the mid 60's describing severe
corrosion in the furnace and high temperature gas passes of refuse boilers
(primarily the new German ones) shortly after being put in operation. The
initial rather disturbing reports were followed by numerous papers which
now indicate that the corrosion, although still understood only to a limited
degree, has subsided or been brought under control.
As a result of detailed reviews of the literature,
and discussions with operators of combined firing installations in Europe,
with research personnel at Battelle Frankfurt, and with W. T. Reid, a
knowledgeable U. S. authority (Ref. 45) on corrosion, the situation may
be summarized as follows:
mechanism of refuse
based upon surmise.
perhaps all, of which
. No one in Europe really knows the exact
boiler corrosion, and the proposed explanations are
There are several plausible theories, many, and
rnay apply at different times.
. Corrosion in refuse boilers is mainly a
nuisance at present rather than a critical problem, and poses no serious
threat in Europe to existing installations. As a result, only minor atten-
tion is being given to elucidating the causes of corrosion.
>:
-------
The above comments should counteract the unfor-
tunate impression, still widespread in the U. S., that high-performance,
refuse fired steam generators in Europe continue to experience intolerable
corrosion problems. From a design point of view, it must be acknowledged
that more detailed information is desirable.
The literature can be summarized as follows:
. Corrosion takes place in the convection passes
when tube metals exceed temperatures between 850 and 950oF. Corrosion,
with one exception, takes place in the presence of deposits. The corrosion
appears to be due tb two mechanisms, one requiring a liquid phase and the
other the presence of dry ash. Wickert (Ref. 46) indicates that the latter
can take place under a reducing environment in the presence of K2S04 in
the deposit and HCl and S02 in the gas at the above-mentioned temperature
. condition. Lead is present in relatively large quantities and may be acting
as an accelerator. Zinc is present in large quantities, but does not appear
to be directly involved with the corrosion. It may be contributing to the
deposit formation by forming a low melting phase and thereby assist in the
attack.. .
. The presence of a reducing environment ac-
celerates the corrosion and permits it to take place at lower temperatures.
It may be an explanation for the formation of fused deposits at temperatures
below the ash softening temperature of the deposits.

,
. . Observations of corrosion in the laboratory
and the field agree for the most part; H2S is found under reducing conditions
and deposits are required to be in contact with the metal surface.
. Available evidence indicates that if reducing
conditions were avoided through combustion improvement, surface tem-
peratures were reduced to 950oF, and some attempt was made to decrease.
the lead, potassium, and sodium fired in the refuse, corrosion problems
would be minimized, if not disappear.
The conclusions from the experimental investigations
performed under this contract, as well as from the literature, can be sum-
marized as follows:
. . Corrosion at high temperatures is probably
due to more than one type of attack, depending upon the existence of a re-
ducing or oxidizing situation. In either case, the corrosion appears to be
strongly dependent on the existence of ash in contact with tube surfaces.
Acceleration of the normally slow gas attack appears to be due to the
presence of lead, sodium, and potassium. A liquid' phase may be essen-
tial to promoting corrosion under oxidizing conditions, although this re-
striction appears not to hold under reducing conditions.
II- 3 9

-------
. Tenlperatures of 850
the threshold range for oxidizing conditions. This
to as tow as 6000F under reducing conditions.
to 9500F appear to be
tenlperature l11ay drop
. Deposit problems in refuse or c01nbined
firing plants are silnilar to those in fossil fuel fired plants. Through
careful interpretation of data, it is possible to extrapolate some of the
results from one type of unit to the other. It is recommended that the
total alkali (NaZO plus KZO, calculated as NaZO) not exceed 0.4% by
weight on a dry fuel basis, that the gas temperatures be kept low (below
9500F) in convection passes, and that reducing conditions be avoided.
. Ash deposits are formed by a diffusion of
volatile minor ash constituents and inert particulate matter to the tube
surface. The problem could probably be reduced by restricting lead,
sodium, potassium, and zinc, in the refuse, or by diluting it with an
appropriate fossil fuel. Deposit and corrosion also depend very strongly
on the difference in temperature between surface and the bulk of the gas
stream. Rearranging tube surface would require trading heat transfer
surface for effective reduction in tube fouling. Shielded surface could
be integrated into initial design of the tube bundle more effectively by
considering it in advance rather than using it as a remedy.
. Even though refuse is a heterogeneous fuel,
the problems appear to be the same in all boilers. Differences are pro-
bably due to variations in internal environmental conditions rather than
variations in boilers or the refuse fired. With the exceptions for the
effects of the presence of minor constituents such as lead and zinc, the
deposits and the problems that are being encountered are similar to those
encountered with coal.
. Phosphorus has been given little consideration
in that it is present in small amounts. Quantities found appear to correlate
with calcium concentrations. It may be partially responsible for high tem-
perature deposits and as such should be given some attention.
. The presence of chlorine in the flue gas has
been given considerable attention. Its contribution to the corrosion prob-
lem appears to be strongly dependent upon (1) the minor constituents that
are present and acting as accelerators, and (Z) upon reducing conditions.
It appears that any contribution it makes to the corrosion problem may be
overconl.e by controlling combustion and designing for the effect of other
minor constituents present.
d.
Ash Fouling Tendencies
One important criterion in the establishment of
proper ratios of refuse to fossil fuel in a steam raising facility is the ef-
fect this ratio has on the maintenance costs of the heat exchange system.
II-40

-------
Evidence was therefore sought on factors affecting ash fouling from plants
now in operation so that improved design and operating practices could be
suggested for the future.
It was found that the desirable proportioning of
refuse and fossil fuel could not be easily defin'ed from existing data. Chemi-
cal analyses of ash deposits from various units indicated no appreciable
change in blended flue gas deposits, and it was also found that fouling was
independent of total ash loading. In some European units, refractory was
added to furnace waterwalls, causing an increase in furnace exit tempera-
tures and subsequently new fouling problems on convection surfaces. . Cer-
tain of the minor constituents in the ash appear to be of major importance,
while chlorine, expected to be highly corrosive, was absent or seemed to
play only a minor role in any deposition or corrosion phenomena.
In the combined firing of refuse and pulverized
coal, it might appear that there would be some minimum coal level above
which a refuse and coal mixture could stlll be considered a non-fouling
fuel, providing that the fossil fuel was non-fouling. The recent availability
of chemical analyses of raw refuse, in conjunction with experience with
fossil fuels, has made it possible to propose a criterion for establishing
this limit.
In designing systems for any fuel, heavy empha'sis
is placed on chemical analyses of the fuel. These analyses indicate not only
how the fuel may be burned, but they also provide an index of potential tube-
side fouling and corrosion. Previous data on refuse composition have been
generally qualitative in nature and the results have often been reported by
inference (Ref. 47). European experience has generally not indicated much
in the way of detailed analysis, the difficulty in obtaining a representative
sample being felt to be so great that additional analyses were not warranted.
The criterion proposed is based, as .discussed below,
on three samples of refuse taken on three different days in St. Louis, Missouri,
and three samples of refuse taken oil three different days in Munich, West
Germany. It is suggested that a fuel is a fouling fuel when the alkali content
(sodium and potassium, calculated as NaZO) on a dry fuel basis exceeds O. 4%~:'.
Above this value severe slagging problems are likely to be encountered on
tube surfaces with metal temperatures above 9500F.
The limitation of 0.4% alkali and the temperature of
9500F is derived basically from coal and lignites; the basis of this criterion
is empirical and subject to judgment. However, its basis appears to be well
founded on sufficient data on bituminous coals, lignites, wood, and wood bark
~:'Based upon ash analysis.
II-4l

-------
(Refs. 48, 49). The temperature limitation coincides sufficiently closely
with the European experience in refuse burning that further refinement in
its values is not warranted at this time. In Europe it has been found that
refuse burning either by itself, or in large proportions with coal or oil,
generally results in fouling where the final steam temperature is 4000C
(7520F) or higher. Generally in the superheater the tube metal temperature
is equal to the steam temperature plus up to 150oF.
From compositional data of St. Louis refuse in
Tables No. 14 and 15 of Reference 41, Table Il-lO has been prepared to
show possible upper limits on refuse proportions based on the 0.4% alkali
criterion. The coal chosen as the base case has the highest alkali content
reported for the Meramec plant. This criterion was also applied to Ger-
man data provided by Consultant Dr. M. Andritsky, with the results shown
in the bottom half of the table. For the results of the application of the
criterion, the heating value of the refuse was assumed as 2800 Btu/lb
(HHV). The alkali content of German bituminous coal ash varies between
0.5%-5%, and for the calculation a value of 2. 5% was used. The ash con-
tent of the coal was based on the analysis of the coal during the TtJV tests
of the Munich North, Unit 1. The ash content was 6.4%, the moisture was
6.0% and the heating value was 13,200 Btu/lb (HHV).
The results indicate the rather wide spread that
may occur in a given facility, with the St. Louis refuse having a range
from 4. 2 to 23% (heat input basis) and the Munich data from 1. 7 to 3. 7%.
They furthermore suggest that the Meramec unit with 10% heat input from
refuse may encounter fouling conditions, and that Munich should have. The
criterion is substantiated by the experience in Munich. The mixture of
refuse and pulverized coal was not expected to be non-fouling, such tha.t
ample provision was made for fouling in each of the three units. The units
at Stuttgart are fired with refuse and oil, with the alkali content of the oil
ash being negligible. The value for a non-fouling fuel has undoubtedly been
exceeded, and Novak (Ref. 50) indeed states that they did have a substantial
arnoun t 0 f fouling.
The suggested criterion should not be treated as a
firm design rule, but it is important to recognize that it may be generally
valid and thereby provide for it. This can quite easily be done for new
plants. In the systems analysis,':' the new plants selected as optimums
were the Separately Fired Economizer (Case 3) and the Separately Fired
Fos sil Fuel Superheater (Case 1 0). In each case the metal temperature
does not exceed 950oF. In addition, provision has been made for potential
fouling by providing bare tube surface throughout, widely spaced convection
tube banks, and ample soot blowers. .
This criterion is perhaps most critical in consi-
dering existing steam generators that might be considered for combined
refuse - coal firing. The only limitation for modifying existing units for
firing refuse on a grate is the grate dimensions, and in most cases the
~'See Section III, C.
II- 42

-------
TABLE II-IO
MAXIMUM REFUSE QUANTITY FOR O. 4o/c ALKALI CONTENT IN MIXED F1JEL
     Refuse, St. Louis, Mo.   Coal, St. Louis, Mo.
   First Sample Second Sample  Third Sample Meramec & Labadie
   4/16/69  4/29/69  5/9/69   1969 
   Compos ite Composite  Composite  Washed 
   As Rec'd Dry As' Rec'd Dry  As Rec'd Dry As Rec'd Dry 
 Alkalies (as NaZO) O/C 1. 43 1. 96 0.58 0.76 1. 79 2.60 0.18 0.20 
 Maximum allowable refuse o/c 13. 7 11. 4 38.2 35.6  10.6 8.3     
 on weight basis      
 Maximum allowable refuse o/c 5. 7  23.0    4.15      
 on heat input basis          
I=:     Refuse, Munich    Coal, Munieh 
I        
~   Ash No. b Ash No. 14  Ash No. 18 North, Unit 1 
l.V   9/8/60  1/5/61   2/1/61  Ttiv Report 
   As Rec'd Dry As Rec'd Dry  As Rec'd Dry As Rec'd Dry 
 Alkalies (as Na20) o/c 1. 44 2.46 2.83 4.46 1. 80 2.35 0.16 0.17 
 Maximum allowable refuse o/c 15.2 10.0 7. 7 5.4  12. 7 10.5     
 on weight basis      
 Maximum allowable refuse o/c 3. 7  1.7    3.0      
 on heat input basis          

-------
maximum allowable refuse rate considerably exceeds 10% (heat input), as
it should in consideration of economic factors. However, many of the units
selected for modification have final steam temperatures near 1 OOOoF. In I
more modern and larger capacity units, with high superheat and reheat
temperatures, in which refuse might be burned either in suspension or on
a grate, it would appear that careful attention should be given to problems
that might aris e from the alkali content of the fuel system.
Coal/refuse ash melting temperatures were of con-
cern even 40 years ago, for the reason that the practice of burning refuse
in apartment-house coal-fired heating units was causing ash slagging and
high grate maintenance costs. In Reference 51, fusion temperatures of 6
refuse ash samples, 12 coal ash samples, and 26 mixtures of 90% coal ash/
10% refuse ash are reported.
In all cases the fusion temperatures of the mixtures
were below those of the coal ash samples; in only 15 out of the 26 were they
also below those of the refuse ash samples, and these cases included mostly
the higher-melting refuse ash samples. The coal ash fusion temperatures
range from 1946°F for a Western bituminous coal to 30000F for an eastern
bituminous coal, but a given boiler would be burning coal with a limited range
of ash fusion temperatures, depending on the design. However, the refuse
ash melting temperatures ranged from 2023 to 2862°F, typical of the range
that might be received in a single day at a given incinerator. Thus the con-
clusion is reinforced that new combined firing plants must be designed to
accommodate a potentially large amount of ash fouling; plants selected for.
retrofit must particularly take the same pos sibility into account.
There is', however, a suggestion in the data that
wide variations in refuse ash fusion points will not carry through to similar
wide variations in refuse-coal ash mixtures. The data from Reference 51
are shown in Table II-ll, on a grid of increasing individual ash fusion tem-
peratures; those of the 90/10 coal-refuse mixtures are indicated in the
matrix, with values lower than those for either component underlined. It
is seen that the range is strikingly compressed and that for a refuse ash
range of 8400F the range of the mixtures is not more than 109°F for any
given coal ash.
C.
POTENTIAL POLLUTANT REDUCTIONS
1.
Current Incinerator and Power Plant Emis sions
The utilization of mixed municipal refuse as a power plant
fuel represents an important solution to a rapidly developing solid waste dis-
posal problem. Firing such a fuel, which is characteristically low in sulfur,
could also afford a significant reduction in the quantity of S02 that must be
controlled. Quantifying this aspect of the concept is therefore of considerable
relevance to contemporary and future national interests.
II-44

-------
TABLE II-II
FUSION TEMPERAWRES OF 90O/c COAL ASH/I0O/c REFUSE ASH MIXWRES
H
H
I
~
U1
.~            
Sample, of 1946 2130 2185 2405 2525 2539 2567 2567 2678 2691 2946 3000
Refuse Ash
Sample, of            
2023   2185 2375     2334  2341 
2090  2081  2290     2225.   
  -          
      -     2324 
2140   2158 2308     2278  
2480   2193 2341     2270  2450 
    -     -  - 
2665    2324     2302   
    -     -   
2862 1862   2271 2201 2185 2310 2307 2294 2252  2665
 -   - - - - - - -  -

-------
According to Reference 52, 41. 6% of the national
502 burden for the year 1966 emanated from coal-fired power plants
(11,925,000 tons/year 502)' Data from the same source (for 1963) show
that the 502 produced by coal-fired power plants is equal to about 4. 5% of
, the weight of the coal burned. Reference 53 points out that, for selected
urban areas and years in the mid-sixties, coal-fired power plants con-
tributed between 13 and 15% of the total particulate burden experienced
in the sampled regions. Unlike the 502 emission data, however, the
cited level of particulate pollution involves some uncertain level of control.
Using atmospheric burden data for the 5t. Louis
area (Ref. 54) and the above relationships, it can be estimated that for
each ton of coal fired to generate electricity, approximately 90 lbs 502
and 10 lbs of particulates are released to the atmosphere. In the absence
of control, however, the particulate emission rate, according to Reference
44, would be about 112 lb/ton of coal fired. This suggests that power plants
in the 5t. Louis area were, at the time the data was developed, controlled
to an extent of slightly less than 90%.
One can compare the above rate estimates with
the values recently published by Nies sen (Reference 13) on refuse incin-
erators. Based on a cross-section of typical plants, the 502 release
rate (without 502 APC-equipment) was found to average 2. 3Z-lb/ton
of refuse incinerated. The median value for particulates (also without
APC equipment) for plants processing more than 50 tpd was 24 lb/ton of
refuse processed. Reference 31 indicates that the average particulate
emiss ion from incinerators is 17 lb /ton of refuse fired.
The power plant and refuse incinerator 502
emission rates invite direct comparison. In view of even current 502 APC-
practice, the quantities released per ton should have been essentially
predicated on the nature of the fuel and its combus tion products. Thus
one could predict that coal would. emit about 40 times the amount of 502
produced by the same weight of refuse, or about 13 times that, if one compares
the fuels on an energy content basis. In the case of uncontr.olled particulates,
one would predict that coal would emit 4 to 5 times the amount of dust pro-
duced by the same weight of fired refuse. On an equivalent energy basis,
suspension-fired coal would produce 1 1/3 to 1 2/3 times the amount of fly
ash generated by refuse fired on grates.
In order to determine near - and long- range goals
in air pollution control planning, the leading air pollution control officials of
each of the LMA's selected for study in this program were interviewed. Annual
maximum suspended particulate goals for the LMA's range from 80 to 120
II- 46

-------
micrograms per cubic meter. Current atmospheric concentrations are on
the order of 170 to 200 micrograms per cubic meter. In some areas the
target figure of 80 micrograms per cubic meter may not be attainable with
the present level of technology. .
. Target concentrations for sulfur dioxide have been set for
several of the LMA IS. All represent a substantial reduction when compared
with recorded concentrations in the past. Goals for New York City and for
New Haven, Connecticut, are O. 10 ppm. The 1974-1975 goal for Philadelphia
has been set at O. 03 ppm.
The sulfur content of fuels has been limited in a number
of LMAls. The maximum allowable fuel sulfur content in New York City
was 2. 20/0 in 1966 and will be reduced to 1 % by 1971. In Washington, D. C. ,
the allowable sulfur content has been reduced from 2. 250/0 to 1. 50/0, and had
been scheduled to undergo further reduction to 10/0 in 1970.
2.
Projected Reductions Through Use of Refuse as a Fuel
The impact the use of refuse now available to centralized
power plants would have in supplanting fos sil fuel is of course an important
consideration. Reduced to simplest terms, complete utilization of this com-
modity with even only a small displacement of fossil fuel energy would still
be a welcome solution, now or eventually, to the solid waste disposal prob-
lem. On the other hand, complete displacement of coal by refuse would
represent a major benefit to APC interests, assuming of course power
generation by refuse alone were technically feasible and sufficient quantities
of refuse were available.
The total fos sil fuel displacement (F) in terms of coal
equivalent (F = W C + .W CE) becomes, in close approximation:
- 1
F = P(W TMAFhMAF - W MhV) h
C
(II- 5 )
where
p
= Po pula tion
W TMAF = Total refuse collected (MAF>:<), lb/cap. -day

hMAF = Heating value of refuse (MAF), Btu/lb
WM
hy
= Moisture content of refuse>:<>:<, lb/cap. -day
= Energy required to heat and vaporize moisture, Btu/lb
hC
= Heating value of coal, Btu/lb
>::<>:
-------
Wc
WCE
= Weight of coal displaced, Ib/day
= Coal equi valen~~ CE) of weight of oil (W 0)
displaced (WO. hd' Ib/day, where hOis

the heating value of oil, Btu/lb.
The term, F, was determined for the period from 1965
to 2000 for each LMA and for the USA as a whole, based on the calculated
value of hMAF' and on setting hV = 1120 Btu/lb and h = 12, 022 Btu/lb~:~
The results are shown in Table II-12, where for conveience the units have
been changed to tons /year.
Figure II-6, based on Reference 55, shows the consump-
tion of various sources of raw energy for the conversion to electrical energy
in the U. S. through the year 2000. A curve showing the estimated total num-
ber of Btu's available from collected refuse (based on Table II-12) has been
superimposed on this figure. It is obvious that a significant amount of coal
can be saved by firing refuse in utility-grade boilers. As shown in the table
below, the replacement rate will increase with time. This is because the
fuel value and the per capita production of refuse are expected to increase
at a faster combined rate than will the per capita demand for fos sil- fuel
generated electricity.
PROJECTED POWER-PLANT COAL DISPLACEMENT
BY URBAN REFUSE ENERGY
 Available Coal Energy Potential Coal
 Refuse Energy Required, Replacement
Year 1015 Btu/yr 1015 Btu/yr %
-
1965 1.6 6.0 26.7
1970 1.8 7.2 25.0
1975 2. 1 8.5 24.7
1980 2. 5 9.8 25.5
1985 3.0 11. 1 27.0
1990 3.5 12.2 28.7
1995 4. 7 13.8 34.1
2000 5.8 15. 1 38.4
When considered on a national basis, these coal displacement data are only
pertinent in a relative sense. Because little coal is used in power plants
west of the Rockies, the urban refuse generated in that and other areas of
the country could have of course no coal displacement effect.
*This value is discussed in Section III, B.
II-48

-------
I=t
I
~
~
Year
1965
1970
1975
1980
1985
1990
1995
2000
TABLE II-12
PROJECTED FOSSIL FUEL DISPLACEMENT BY URBAN REFUSE ENERGY
Urban Refuse
Fuel Value,
Btu/lb (MAF)

9386
9402
9416
9446
9496
9570
9673
9811
U. S. A.
67.1
75.8
88.5
104.0
123.2
146.8
192.7
240.3
(F) Fossil Fuel (Coal Equivalent) that Can Be Replaced, 106 tons/yr
New York Chicago New Haven Philadelphia St. Louis Washington DC
3.9
2.4
1.6
1.9
0.2
4.4
2.8
0.2
5.1
3.5
2.2
0.3
0.3
2.7
5. 7
4.1
6. 7
'0.4
3.3
4.8
7.7
5.9
8.5
0.4
0.5
3.8
8.3
4.8
10.9
0.5
5.8
11. 1
0.8
0.8
0.9
1.0
1.1
1.3
1.3
1.5
1.6
1.9
1.9
2.3
2.5
3.0
3.0
3.7

-------
a:
~
~ 10
CD
an
...
o
...
>' 5.0
~
U 4.0
a:
~ 3.0
(J
UI
~
UI 2.0
u.
o
z
o
~ 1.0
a:
UI
z
UI
~
~ 0.5
u. 0.4
C
UI
:!: 0.3
;:)
en
Z
o 0.2
(J
>
~
a:
UI
ffi 0.1
3:
«
a:
100
50
40
30
20
0.05
0.04

0.03
"
~
OIL
NUCLEAR
1960
2000
0.02
0.01
1950
1970
1980
1990
FIGURE 11-6. RAW ENERGY CONSUMPTION FOR THE GENERATION OF ELECTRICITY
IN THE UNITED STATES
II- 5 0

-------
. I ~ ';'. '
',' ",\ ....,
..:. "
3.
Effect of Refuse Firing on SOZ and Particulate Emissions
b1 further es timating the potential reductions in pollutant
emissions that would result fron"1 refuse-firing, certain conditions have been
posed. fuitially, it was assurned that no SOZ-control is being practiced by
electrical utility plants and that particulates would either be controlled at
a 95% efficiency or not at all. These conditions were then applied to the
situations where the electrical power industry would be supplanting coal
with an insignificant amount of refuse or, on the other hand, firing all the
mixed municipal refuse that was available.
The nationwide change in SOZ emissions, 6S02, in"lb/day,
. can be calculated by the following expression:
6S02 = PW TSW - 38. 0 SeW c - 39.2 SOW o/ZOOO
(II- 6)
where
coal sulfur content, wt-%
S =
c
S -
o -
S -
W -
fuel-oil sulfur content, wt-%
fraction of SOZ generated from refuse, Ib/ton
The other terms have been explained previously. The factors as sociated
with Sc (38.0) and So (39.2) were obtained from References 44 and 56, re-
spectively. They were originally derived on the basis of assuming that 95%
of the coal sulfur and 98% of the fuel-oil sulfur would be emitted as S02' The
numerical factors thus convert the percent fuel sulfur to equivalent amounts
of S02 in the flue gas (lbs /ton of fuel fired).
The nationwide change in the dust burden, 6D, in Ib/day,
can be calculated as follows:
6D = (p WTDW - 16.0 ACW C )/2000
(II- 7 )
where
WT = total refuse collected, Ib/cap. -day

DW = particulates emitted from refuse, Ib/ton

AC = ash content of coal, wt-%
One of the operations of the factor, 16.0, used in Equation II-7 is to provide
for the release of 80% of the coal ash as stack emission. This rate is based
on information contained in Reference 44.
ll- 51
"

-------
By using Equations II-6 and -7, and the values of F from
Table II-12, the reductions in S02 and particulate matter were calculated
from each LMA and for the nation for the year 1970. The results are tabu-
lated in the table below. The calculation was restricted to a displacement
of coal only. Thus the term in Equation II-6 used to express the reduction
in S02 from oil by the partial displacement of oil with refuse was set at
zero. The comparison is for the situation in which substantially no refuse
is incinerated, and that in which substantially all refuse generated is used
to replace fos sil fuels in power generation. Particulate reductions are
shown for two cases. In one,no emission control is assumed (described
by Equation II-7), while in the other it is assumed that 95%*of the par-
ticulates have been removed by APC equipment whether refuse is fired
or not; i. e., 5% of the value calculated by Equation II-7 is emitted. The
factors used to make the calculations were Sw = 2 lb/ton; DW = 17 lb/ton;
Sc = 2%; and AC = 7%. Sw and DW were derived from Reference 31, and
Sc and AC were obtained from Reference 57.
POTENTIAL REDUCTIONS IN CURRENT
S02 AND PARTICULATE EMISSIONS
Area
Reduction in
S02, tons /year*
Reduction in
Particulates, tons /year
Without Control 95% Control
USA
155,000
98,400
7,100
66,700
31,400
34,700
2,670,000
139, 000
87,500
6,500
59,300
28,000
31,200
2,400,000
6,900
4,400
300
3,000
1,400
1,500
120,000
New York, N. Y.
Chicago, Ill.
New Haven, Conn.
Philadelphia, Pa.
St. Louis, Mo. /111.
Washington, D. C.
>:
-------
TABLE II-13
PROJECTED EMISSIONS REDUCTION BY COAL DISPLACEMENT WITH REFUSE
(tons per year)
  New York Chicago New Haven Philadelphia St. Louis Washington, DC
 Year S02 Part. S02 Part. S02 Part. S02 Part. S02 Part. S02 Part.
 1970 7,000 10,300 4,400 6,500 300 500 3,000 4,400 1,500 2,100. 1,500 2,300
 1975 8,100 12,000 5,400 7,900 400 600 3,600 5,300 1,800 2,600 1,900 2,800
 1980 9,500 14,100 6,500 9,600 500 700 4,400 6,500 2,200 3,200 2,500 3,600
 !    
 1985 11, 100 16,400 7,900 11, 700 600 800 5,300 7,800 2,700 3,900 3,100 4,600
H 1990 13,100 19,400 9,700 14,400 700 1,000 6,500 9,500 3,300 4,800 3,900 5,800
H
I             
U1             
VJ 1995 15,600 23,000 12,000 17, 700 800 1,200 7,900 11 , 700 4,100 6,000 5,000 7,400
   ,        7,600 6,500 9,600
 2000 18,700 27,500 15,000 22,200 1,000 1,500 9,900 14,500 5,100
Note: Above data based on asswnption that both S02 and partfculates are controlled at a 95% efficiency and
that all refuse substituted for coal would otherwise be incinerated with the same APC conditions observed.

-------
APC efficiency would exist for both S02 and particulates. It was further
assumed that all the refuse substituted for coal would otherwise have been
incinerated, again with 95% control of the two pollutants. This assumption
was considered to be more realistic than to expect that no refuse will be
disposed of in incinerators, even though large quantities today do go to
landfill sites. In the future, the firing of refuse in power boilers or re-
fractory incinerators will probably become an unavoidable economic
necessity.
Referenc,e 54 indicates that S02 and particulate emis sions
from all sources (excluding refuse disposal) are 450, 000 and 131, 000 tons /
year, respectively, in the St. Louis area. On comparing this particular
LMA data with that in the table above, it can be estimated that, by substi-
tuting all available solid waste for coal, a reduction in S02 emissions of 7%
can be achieved by combined firing. Similarly, if no particulate emission
control were practiced, these emissions would be reduced by 21 %, but only
by 1 % if particulate emis sions from power plants (but not other sources)
were controlled with APC equipment that furnished a 95% efficiency.
An aspect that should be considered is the fact that of the
502 sources extant, the power boiler will probably be the easiest to equip
with APC systems and bring under control. If all power plant SOZ-emissions
were controlled with a 95% efficiency while all other sources remained un-
changed, refuse firing would have a negligible effect on the (now reduced)
total S02 burden.. This can be seen from the following table, which again
utilizes data for the St. Louis area.
EFFECT OF COAL DISPLACEMENT BY REFUSE ON 502
EMISSIONS FROM ALL SOURCES (ST. LOUIS AREA)
Total 502
Controlled, %*
SOz Reduction
by Refuse, %
40>:0:'
6. 7
1.4
0.8
0.5
100
80
60
>:'As suming 95% control efficiency and all power plants controlled.
*>:'Approximate point at which all power plants but no other SOZ
sources would be controlled.
In the present discussion, the possible reduction of NOx
emis sions has not been considered. This aspect of the analysis, which
actually is somewhat outside the scope of this study, is difficult to quantify.
The level of NOx emissions will vary with combustion conditions. Further-
more, the control efficiency that can be as sumed is problematic in the ab-
sence of a practical NOx-APC methodology and the relevant process infor-
mation. One can, however, acquire a qualitative appreciation for the
II-54

-------
potential NOx reductions that would be implicit in refuse substitution by
comparing the data on NOx emissions from refuse incinerators and coal-
fired power plants. Such data have been presented in Section lIB, 2, b(l).
D.
HANDLING AND PROCESSING
1.
Introduction
European practices in the physical handling of refuse
prior to and after combustion could be considered advanced compared to
those in the U. S., but the degree of superiority is really a small one and
could more acc~rately be described as "experienced" rather than one in-
volving application of sophisticated technology.
Refuse, once collected, may be classified (sorted, with
or without recovery of salvage values), altered in size, mixed,. transported,
stored, dried, introduced into a combustion chamber, and, after varying
degrees of burn-out, the residue (or products, in the case of pyrolysis)
collected for subsequent use or disposal. Each of these processes is
practiced to some extent in conjunction with the steam raising - refuse
furnaces constructed in recent years. The design of much of the equipment
employed is directly transferred or adapted from other fields, and conse-
quently oftentimes is of lower efficiency or higher cost than would be the
case if optimized specifically for proces sing refuse. Increasing attention
being given to the solid waste problem by both government and industry has
stimulated development of potentially superior equipment and integrated
systems in the last several years. Unfortunately, the results of most of
this work are just becoming available and little can yet be said about cost
effectiveness, particularly that relating to maintenance. In a number of
cases, the process itself is considered proprietary and design information
is not available. Several new industrial and municipal incinerators are
about to become operational and the. knowledge to be gained from a year I s
operation of these facilities will be far more meaningful than performance
estimates now based on rather large extrapolations or on conclusions drawn
from only remotely similar systems.
In each of the processing steps cited above, it is apparent
that rather large improvements should be realizable over current practices.
The neces sary developments will probably not be made nor the cost benefits
enjoyed unles s a significant number of sizeable plants are built. That size.
reduction can add $1 to $3 per ton to refuse burning is not sufficient incenti ve
in itself to cause grinder or hammermill manufacturers to develop equipment
designed specifically for the handling of refuse at energy consumption and wear
rates approaching theoretical. In spite of glass cullet having a value of $20/ton
and scrap aluminum being worth $200/ton, there are no techniques yet in full
scale operation for efficiently separating away such materials once they have
entered the normal mixed waste collection system. The need is now being
II-55

-------
recognized for totally improved solid waste disposal schemes, including
recycle and salvage concepts, and once a market is establ~shed, innovations
in processing equipment will soon be forthcoming. In the meantime, the
municipality or utility desiring to construct an energy recovery refuse in-
cinerator can rely on U. S. consulting engineering firms to recommend
well-tried and readily designed types of equipment available at the time of
bid solicitation.
Little summary information is available on refuse pro-
cessing equipment. Battelle Memorial Institute assembled much pertinent
data that apparently never received general distribution, although its prin-
cipal conclusions regarding grinding costs have been reported elsewhere
(Ref. 58 and 59), and a modified edition has been recently released (Ref.
60). A large portion of the information in this field consists of manufac-
turers I literature or rather qualitative descriptive articles of the use of
commercial equipment, oftentimes under conditions having only remote
resemblance to the desired application to refuse treatment. Wide varia-
tions in performance and cost claims are to be noted to the point where a
compilation of such information serves little purpose. Rather than attemp-
ting any exhaustive listing here, in the sections that follow are presented
the critical facts concerning those refuse processing steps that would have
applicability at some stage in an energy recovery system. Because refuse
size reduction is an indispensable unit proces s where suspension fired
steam generators are concerned, this topic was given particular attention.
In this connection, samples of shredded refuse were analyzed for size dis-
tribution to permit the estimation of the burning characteristics of such
material in suspension fired furnaces.
2.
Size Reduction
a.
Equipment Characteristics
In the several compilations of descriptive information
on size reduction (Refs. 58, 61, 62), much of the data is concerned with de-
vices of limited application to mixed refuse grinding at high throughputs.
Brittle materials are ideally broken up by one type of equipment, tough
materials by another, fibrous ones by still a third, etc. When the several
physical types are combined, as is the case with municipal refuse, a com-
promise design must be employed. Typically, this has taken the form of
some variation of a hammermill, with the striking-bar configuration and the
orientation of the axis and the direction of rotation being the principal dif-
ferences offered by the fifteen or so U. S. and European manufacturers.
Single mills have been built that could handle as much
as 50 to 75 tons /hr of refuse, requiring some 10 horsepower /ton-hr in this
range to drive them. Essentially all of the still relatively few mills actually
being used for refuse grinding in Europe or the U. S. are considerably smaller
than this (10 to 20 ton/hr).
II- 56

-------
In the Battelle study mentioned above, it was noted
that the nature of the feed material and the size of the milled product are
important parameters in determining the amount of drive power that needs
be applied. It was pointed out, for example, that sorted municipal solid
waste required only 65% of the power that must be used to mill unsorted
refuse. It was not clear, however, which ingredients would be removed
in the sorting process. The increase in power required to achieve smaller
product top sizes in a single pass was given as follows:
Top Size, in.
Horsepower per ton factor
6
4
2
1
1. 00
1. 39
1. 64
2.38
As shown in Reference 13, which reviews these same data, the relationship
of comparative power use to product top size is a linear function, with which
the power factor for the 2 in. top size is in poor agreement. The author sug-
gests therefore that the value is in error and should be 2. 00 rather than 1. 64.
The significant reduction in throughput that results from the production of
small top size material is usually avoided by using several stages of milling.
I'
I
I
Cost information, be it capital, operating, or main-
tenance, is still difficult to assess because of (1) differences in design fea- .
tures and accessory handling equipment; (2) the lack of an efficiently operated,
long term, size reduction project; and (3) physical differences of input refuse
(composition and size) and ground material (size) from one user to the next.
After adjustment for escalation, the recent sales price of $12~, 000 for a 70
ton/hr grinder appears to well fit the curve offered in Reference 13 for a
variety of equipment capital costs; other costs on this averaged curve in-
clude $80,000 for a 50 ton/hr unit and $35,000 for a 15 ton/hr unit. A 6-in.
top size product is typically as sumed for deriving the throughput rates in
such cost figures. Experience by several of the large equipment manufac-
turers and at operating demonstration facilities such as Madison, Wisconsin
(Ref. 63), indicate maintenance costs (primarily re-facing of the hammers)
to be on the order of $0. 20 - 0.40 /ton and electrical costs of $0. 10 - O. 15/
ton of refuse. .

In Reference 13, the Battelle data were further ana-
lyzed to develop total capital costs for shredding mixed municipal refuse
input for incinerator plants sized from 120 to 1200 tpd. Cost data were
presented for 6, 2, and O. 5 in. top size output material. It was shown that
total shredding costs for any given product top size decreased significantly
as plant size increased within the smaller capacity range but that little
economy of scale resulted when plant sizes exceeded about 900 tpd.
ll- 57

-------
Relying on the costing method developed in Reference
13, grinding costs for a set of product top sizes that would be more appro-
priate to the present study were calculated. The only deviation introduced
was in the assigned equipment life expectancy (lO yrs vs 15 yrs) and, thus,
in the rate of annualization of capital costs (14.6% vs 9. 7%). The results
obtained are shown below:
SIZE REDUCTION - TOTAL CAPITAL
AND OPERATING COSTS
(Costing date basis: July 1969)
Nominal Top Size
of Product, in.
Shredding Cost,
j/ton of Input
6
4
2
1
1. 20
1. 62
2.20
2.57
These costs are bas~d on a plant refuse rate of 1200 tpd operating 2 mills
1 shift per day and 5 days per week. As pointed out above, these costs
would probably change but little for other refuse rates in this general range.
b.
Test Work
Because of the potential importance of refuse reduc-
tion operations in improving the burning process, some effort was devoted
to characterizing the output from representative grinding machinery. This
was done (May 1970) on refuse being routinely prepared for landfill at Madi-
son, Wisconsin, by a 200 HP Tollemache hammermill and that ground during
a demonstration at Albuquerque, New Mexico, by a 700 HP Eidal mill. In
both cases, the feed material was sampled and sorted into constituent clas-
sifications to establish composition. These analyses indicated, as shown in
Table 11-14, that the Albuquerque refuse was unusually low in garden refuse
and that the Madison refuse was rather low in paper content. This is expli-
cable in that Albuquerque is not heavily vegetated even in May, and in Madison
waste newspaper was being collected separately at that time for recycle pur-
poses. The particle- size distributions of the materials output by the mills
are shown in Table II-IS.
While both mills produced a significant amount of
.material that was well in excess of the arbitrary 1. 5-in. maximum sieve
size observed in the analyses, the amount of very small particles was im-
pres sive. Regardless of the nominal top-size of the product, both mills
furnished considerable material which should burn rapidly in suspension-
fired furnaces.
II- 58

-------
. - .
'. ",
~, . .' .
. .
TABLE 11-14
REFUSE COMPOSITION OF HAMMERMlLL INPUT
AT TWO AREAS SAMPLED'
Textile s
Weight - Percent
Albuquerque Madison
9. 3 13.3
1.6 9. 1
42. 1 28.4
7. 1 10.9
7.9 5. 6
1.5 1.4
17.2 15.3
11. 6 14.9
1.7 1.1
Component.
Garbage
Garden Waste
Paper Products
Plastics, Rubber, and Leather
Wood
Metals
Glass and Ceramics
Ash, Rocks. and Dirt
II- 59

-------
TABLE II-IS
PAR TICLE SIZE DISTRIBUTIONS OF OUTPUTS
FROM TWO DIFFERENT TYPES OF HAMMERMILLS
Fraction and Predominant
Constituent
Nominal Top-size
Albuquerque Madison
4-in. 6 -in. 4-in/:~
Plastic film, paper, cardboard,
> 1. 5-in. .
5. I
31. 9 6.9
3.8 0.5
19.6 0.5
20.5 27.3
24.2 64.8
10.8 3.3
Wood, > 1. 5 -in.
4.3
Metal and Rocks, > 1. 5-in.
8.8
Mixed Material, < 1. 5-in.; > O. 25-in.
22.2
Mixed Material, < O. 25-in.
59.6
Samp1e Wt. Los s between
Sampling and Analysis
12.8
):~Additional hammers were installed in the mill to produce the finer grind.
II-60

-------
The sample dehydration losses noted were unavoidable.
They reflect the time elapsed between collection and analysis, the time spent
in performing the analyses, and the receipt-condition of the plastic containers
in which the ground refuse was shipped from the sampling sites.
A further test on these grinder products was performed
to establish the feasibility of effecting, for firing purposes, a separation of the
various size-fractions produced. The device tested, the Zig- Zag Separator
(see Section II, D, 4, d), was generously made available by the Stanford Research
Institute at Irvine, California. A quantity of the finer- grind output from the
Madison hammermill was processed through a separator having a 6-in. wide
classification duct. When operated at a pressure drop of 6. 5-in. W. C. (vertical
air velocity = 21-25 ft/sec.), 79.2% of the material elutriated. Although this
fraction contained some large pieces of thin-film, plastic and paper pieces,
it was an excellent mixture for firing in suspension. The other fraction, 16.8%
of the total, contained a bulk of the oversized, non-film pieces and denser non-
flammable material. It represented an excellent concentrate for salvage pro-
cesses and a subsequent second pass through the grinder. About 4% of the
sample could not be accounted for. This was lost through evaporation and in
dust fines.
c.
Effect of Particle Size on Combustion
. The immense success of pulverized coal combustion
in large utility boilers has caused interest to be given in recent years to im-
proved incinerator designs where some form of suspension burning would be
approached. A large furnace for ground industrial waste and sewage sludge
has been undergoing testing in Rochester, with as yet undisclosed results,
and modifications are in progress in St. Louis to fire ground municipal waste
in combination with coal.
. While much has been written on combustion theory of
carbon particles (Refs. 64 and 65, for example) and even refuse particles
(Ref. 66), application of this work to mixed refuse is difficult because of the
nature of the physical surface of the latter. Taking, for example, the ham-
mermill product characteristics shown in Table II-15, one can calculate
burning times for the various fractions. Using the pyrolysis and combustion
equations presented in Reference 66 and assuming that mass transfer controls
the rate of burning, the data shown in Table Il-16 was derived for the 4-in.
top size material sampled at Madison, Wisconsin.
If the theory. on which the calculations were based is
correct and the particles are adequately represented by the shapes and average
dimensions specified, an obvious problem exists. Suspension firing of the
material described in Table II-16 would be accompanied by much fall-out and
elutriation of incompletely burned material. Neglecting the wood fraction,
which is not in high enough concentration to be of concern, the slow burning
II- 61

-------
I:::
I
0'
N
TABLE Il-16
CALCULATED BURNING TIMES FOR VARIOUS FRACTIONS IN MILLED REFUSE
(4-in. Nominal Top Size)
Fraction and Predominant
Constituent
Fraction
of Total,
wt - %
Shape in. in. in.
  -
Slab 4 O. 01 
Slab 2 0.2 
Assumed Average Particle Dimensions
Length, Thicknes s, Dia. ,
Plastic Film, Paper, Cardboard
>1. 5-in.
6.9
Wood, >1. 5-in.
o. 5
Metal and Rocks, >1. 5-in.
O. 5
Mixed Material, <1. 5-in; >0.25 in.
27.3
Sphere
0.8
Mixed Material, <0.25-in.
64.8
0.1
Sphere
As sumed Combustion Conditions:
o
Gas temperature = 2370 F; Relative velocity between particles and ambient gas = 10 fps.
Total Time
for Pyrolysis
and Burning,
sec.
15
600
950
20

-------
particles would arise essentially from the large (27.3%) fraction of mixed
material in the 0.25 to 1. 50-in. diameter range. .
With the exception of the inert fraction (> 1. 5 in. ),
,this fraction was observed to be the least uniform of the cuts. Character-'
izing it as a granular mixture of spherical particles would be highly inac-
curate. In addition to some inert pieces, composite artifact fragments,
splintered wood, and various other irregularly shaped items, it contained
considerable quantities of paper and cardboard, occluded with foreign matter,
that ranged from loosely crumpled to compressed structures containing both
crushed, convoluted film bodies and much coarsely filamentated material.
Such bodies would not likely burn as would truly
spherical particles during dehydration, pyrolysis, ignition, and combustion
in a hot gas. The uncompressed pieces would probably behave as thin slabs
and be burned out in 10 to 15 seconds. The clumped structures, on immer-
sion in the hot gas, would likely undergo several processes. Filaments and
protruding film flaps would burn off quickly while the main body dehydrated
and pyrolyzed. The latter effects, however, would release considerable
quantities of gases within the body and cause rupture and displacement of
the intersticial folds. This would cause the agglomerated structure to open
up, expanding the body considerably and presenting new film surfaces. Thus
the burning process should be much more rapid than for a sphere of cellular
or finely consolidated composition. It would be very difficult to estimate
the burning time of such material on the basis of combustion modelling.
Because several full-scale, suspension-fired systems are or soon will be
under test evaluation, this question should soon be answered.
3.
Refuse Transport
a.
Background
. The transportation of refuse is defined as the proces s
of moving the collected material from the area of production to the disposal
site. In landfill disposal operations, it would include the travel of the packer
truck, once it has been filled or completed a collection route, to the site and,
for cost purposes, the return of the empty truck to the route. 1£ transfer
stations are employed, transportation would include the movement of both
the packer trucks to the station and the transfer trailers, barges, or rail-
road cars to the disposal site, and, for cost purposes, the return of the
empty carriers to the collection routes and to the transfer stations, re-
spectively, and the cost of operating the latter.
Other intermediate proces ses interrupting the trans-
portation proces s can be considered. At Madison, Wis consin, for example,
packer trucks deliver refuse to a shredder plant, the output from which is
compacted into trailers for outhaul to an adjacent landfill. Here of course
the shredding operation would be considered to be a unit proces s and not
part of the transpc;>rtation operation.
II-63

-------
In the present context, transport of refuse is defined
as the movement of material to the steam generator site. Strictly speaking,
the movement of the furnace residues to landfill disposal sites should also be
included. For ease of identification, this operation has been treated as a
separate process.
Refuse transport would appear to be a routine opera-
tion in the overall waste management system. Because it can and often does
represent costs that are greater than the disposal process itself, it should
not be as sumed that improvements in the methodology cannot be achieved.
A technique now attracting considerable attention and one that could prove
compatible with the type of refuse disposal considered here is reviewed in
the next section.
b.
Pipeline Transfer of Refuse
While conventional refuse collection practices are
not of concern to the present program, certain advanced methods now being
installed or still under study are worthy of consideration for transport of
refuse from central transfer stations to the steam generator facility. Such
transport might permit the economic accumulation of sufficient quant~ties
of material to justify a large plant with its resulting decreased unit costs.
The use of pipes for refuse transfer is among the
most attractive of the new F:chemeF: being conqidered. The fluid employed
can be either air or liquid. In the former case there exists some 10 years
of practical experience with pneumatic systems in housing complexes and
hospitals. Approximately 20 refuse systems have been or are 'now being
installed in the western world, and an equal number of systems exist for
handling soiled linens. Pipe diameters are typically in the range of 20-24
inch and runs up to 1 mile are entirely practical, with 3 miles now being
considered an upper limit per pumping station (Ref. 67). Existing informa-
tion has therefore been derived only for installations having a large
number of refuse introduction points within relatively expensive buildings.
Transport of waste solids by water has of course
been utilized for centuries, but only recently has attention been given to
hydraulic pipelines for transport-of other than sewage. Zandi has sum-
marized his work in this area in Reference 68 and elsewhere (References
69 and 70) he reviews the overall technology of water-slurry systems.
He has not, however, considered the key problem of terminal water removal.
This worker I s conclusion is that shredders presently available are adequate
for producing slurries of municipal refuse, that concentra tions of solids up
to at least 12% flow without difficulty, and that at below 4% solids the pipe-
line can be designed as though the medium is only water. Equations have
been developed for predicting pressure losses, the significant finding here
being that it is the percentage of paper in the mixed wastes that affect the
II- 64

-------
10 sses. A cost analysis made by Zandi demonstrates that with the present
state of the art, pipelines become competitive with truck hauling when the
distance to the disposal site is 50 miles or more. With only moderate im-
provements in system components, pipelines should become equal in cost
to vehicles at approximately 25 mile haul distance.
With the above research demonstrating that water
slurries of waste are readily pumped, consideration should be given to' the
use of fuel oil as a carrier. Such a system would not only eliminate the
need for drying of the refuse prior to combustion, but would minimize
many of the on-site handling problems now causing utilities to shy away
from using wastes as a fuel.
4.
Separation Processes
a.
Background
An indication of the present inferior state-of-the-
art of refuse sorting techniques is the emphasis now being made to encourage
and support new developments in this field. While recycle and salvage are
of principal concern from an economic and social (resource utilization)
viewpoint, the sorting of mixed municipal refuse would also affect the design
and costs of heat recovery incineration systems. Heating values would be
increased or decreased depending on the recovery practices employed for
paper, plastic, and metal fractions; grinding and handling proced'.ues would
be based on the new physical characteristics of the modified refuse*; superior
combustion methods could be utilized and fireside corrosion minimized; and
final residue disposal would be simplified.
Magnetic separation of iron from refuse has been
practiced for a number of years. In the case of incineration, removal of
this fraction typically is accomplished after burn-out and water quench.
Hand sorting from a conveyor belt has been utilized where low cost labor
is available, primarily in conjunction with composting operations. Auto-
matic separation into a number of fractions at sufficiently low cost to obtain
salvage credits has not yet been commercially demonstrated. Research is
in progress at the Bureau of Mines (College Park Metallurgy Research Cen-
ter, Salt Lake City Metallurgy Center, and Tuscaloosa Metallurgy Research
Lab~ratory), a number of universities and research institutes (West Virginia,
Missouri, Stanford, M.1. T., Alabama, Wisconsin, MRI, and SRI, for example),
and several industrial organizations, where the results are considered quite
proprietary. The basic techniques being studied have been treated,by the
authors at Battelle (Ref. 60), SRI (Ref. 71), MIT (Ref. 72), and MRI (Ref.
73).
*Assurance that all hard objects had been removed, for example, would
permit selection of pulverizers having far lower capital, operating, and
maintenance costs than present equipment.
II- 65

-------
Automatic sorting proces ses rely on two chief prin-
ciples: (1) subdivision of material into pieces having a more or less homo-
geneous composition, sensing of the uniqueness of each piece by means of
one or more devices capable of measurement of properties (density, elec-
trical, optical, hardness, etc.), and diversion of the classified piece to
specific hoppers; and (2) utilization of properties themselves to cause
clas sification, as with magnetic attraction or repulsion, floatation within
a liquid or gaseous stream, or screening to remove finely crushed com-
ponents such as glass. While capital and operating cost reductions are
most essential to the successful incorporation of classifiers, other problems
that remain include the difficulty in arriving at a homogeneous material (the
aluminum steel-lidded can, etc.), and the development of products best
suited to utilize salvaged materials.
b.
Ferrous Metals
The tin content of ferrous residues from municipal
incinerators runs between O. land 0.4% (Ref. 74). This contaminant level
makes this form of scrap unattractive to iron processors for whom an ample
supply of more desirable scrap is already available in most areas (Ref. 73).
The principal use for ferrous metals recovered from refuse is in the preci-
pitation of copper from acid leachates of lean ores. In the required form,
the" metal brings $50 to $60/ton in this market, the highest price paid for
steel scrap in this country. Precipitation iron must, however, furnish a
high surface area with a minimum of further processing at the mine works.
This requires that the salvaged material be shredded and loosely packed
(20- 25 lb/ cu ft) when shipped. Because the main use points are in Arizona,
Utah, Montana, and Nevada, an obvious limiting factor is transportation
cost.
The prospects in this market for ferrous metal
salvaged from refuse are not encouraging. Copper refiners are slowly
moving away from refuse scrap iron to better grade materials. Many
are shifting over to detinned scrap originating from tin plate producers.
The reason for this is that ferrous metal, after separation from refuse
and customary processing, does not perform well in the copper precipi-
tation proces s because of its highly oxidized state and the presence of
foreign materials. If the reclamate could be better cleaned and detinned,
this situation could be reversed. .
Removal of ferromagnetic material from refuse
(whole or shredded) is best accomplished after the latter has been fired.
The volume of material that must be passed through the sorter is greatly
reduced and the salvaged iron is essentially sterile. In certain types of
shredded- refuse firing steam generators it may be preferred to remove
this fraction from the input material if corrosion is a serious problem.
ll- 66

-------
The most popular ~ethod of iron removal involves
the use of suspended- and/or pulley-type magnetic separators. The former
usually consists of a short conveyor loop, the bottom belt of which runs
slightly above and across the feed conveyor. A magnet is situated between
the pulleys so that iron in the feed is deposited on the upper conveyor to
ride off laterally into a receiver. In the pulley-type magnetic separator,
the end pulley on a discharge conveyor is magnetized. As the feed material
pours off the end, iron follows the rotating pulley and detaches, when out of
the field, in a separate line of fall.
Including costs for power, maintenance, and annual-
ized capital costs of the equipment, which is rather inexpensive, separation
cost is minimal (S $0. 10/ton). Output from the separator can be sold to
metal proces sors for between $13 and $15 /ton.
c.
Non-Ferrous Metals
Over the last 10 years, Vanderbilt University has
been examining the concept of utilizing magnetic force ("eddy currents ")
for the separation and sorting of conducting, non-magnetic, materials from
non- conducting ones, after initial removal of iron fractions by conventional
electromagnets. Much progress has been made in the development of prac-
tical equipment for this operation at feed rates up to several hundred lb/hr,
and a unit for separation of 1200 lb/hr is now being assembled. While most
of the work has concentrated on processing shredded cars, limited testing
with the non-ferrous fraction of incinerator ash has demonstrated that some
.93% of the aluminum metal can be isolated in an essentially pure fraction.
In a recent report (Ref. 75) describing the methods being investigated, a
cost analysis based on actual performance data was presented indicating
that a 25% increase in profits would result in the s crap car busines s if the
Vanderbilt s6rter, were to be used. The plant was assumed to process 66
tpd of a mixture containing 82% iron; 14% paper, cloth, and plastic film;
2% rubber, dirt, and glass; and 2% non-ferrous metal. This latter fraction
was sorted into 6 components (aluminum, copper, zinc, chromed zinc, brass,
and stainless steel), with the total cost of grinding all input material and con-
ducting the necessary separations estimated to be $2. 25/ton.
d.
Air Clas sification
Air has been used for the size classification of solid
particulate matter-for a number of years and recently attention has been given
to its use for separation of solid waste into fractions. Among the schemes
being considered, the so-called "zigzag" classifier being developed by the
Stanford Research Institute (Refs. 71 and 76) probably offers the greatest
efficiency at low costs. This device consists of a series of ducting sections
with a turning angle of 600 at each corner; a throat section of 2 in. x 6 in.
was present in most tests, up to 12 stages employed, and blowers capable
II- 67

-------
of creating air velocities up to 3000 fps used. Shredded mixed materials are
added part way up the columns and the air flow adjusted to cause a given bi-
nary separation, with heavy objects and those with low drag coefficients set-
tling under the influence of gravity. The secondary air flow patterns within
a zigzag column, compared to a completely vertical one, permit maximum
aerodynamic separation of particles. Work has proceeded at SRI to a point
where the behavior of various solid materials in an air stream can be mea-
sured and the characteristics of separated fractions predicted. Analysis of
a 30 ton/hr facility indicates that approximately $0. 10/ton would be the cost
for a given binary separation, exclusive of the required grinding costs.
Such a process could of course be used to beneficiate
a steam generator refuse fuel while isolating a fraction high in inert content.
The latter might then be advantageously used for the salvage of metals and,
possibly, glass. It is also possible that such a device could be installed in-
line with part of the primary combustion-air system of a suspension fired
furnace. This technique is now being tested in a vertical duct arrangement
on the CPU -400 (Ref. 77), an advanced type of pilot system des cribed a fe\;"
paragraphs below. '
E.
ENERG Y UTILIZATION
1.
Turbo-Electric Generation
a.
Rankine Cycle Systems
All, or a portion, of the thermodynamically available
energy content of steam generated from external combustion of refuse can be
converted to electricity through expansion of the high temperature and pres sure
steam through any of a variety of turbine systems. The refuse can serve as
the sole heat source or he used in combination with fossil fuels. Typical de-
sign features of conventional boilers are discussed in Appendix B (Volume II).
b.
Gas Turbine Systems
Rather than transfer of heat to a working fluid as is
accomplished in the Rankine cycle, it is possible to directly use hot combus-
tion gases to impart rotary motion to a gas turbine. Distillate oil-fired gas
turbines have become common in the electrical utility industry, particularly
for relatively small units for the rapid supply'ing of peak energy demands.
A number of refuse-fired gas turbine systems have
been suggested. The one presently receiving the most attention is that being
developed by the Combustion Power Co., the so- called CPU -400 (Ref. 77).
This unit, intended to produce 15 MW of power from the combustion of 400
tpd of solid waste, also yields 250 Ib / sec of gas at 9500F that can be used
for process heat, sewage sludge drying, desalination, etc. The design in-
corporates shredding, air classification for the removal of inerts and heavy
objects, and partial drying of the refuse using a small amount of the waste
heat.
II- 68

-------
Combustion in the CPU';..400 system occurs within
a 100 psia reactor, where inert sand particles are fluidized by means of
5840F air from the compressor. Fly ash and fines from the fluidized bed
are removed prior to introduction of the 16500F gas into the turbine by in-
ertial separators and an electrostatic precipitator. Ash that is heavier or
larger than the bed sand will be periodically removed from the reactor bottom
through a rotary air lock. Each of the critical hardware elements has been
tested with sub-scale equipment and further developmental work with larger
units is in progress. .
Full economic evaluation of this concept cannot be
completed until additional engineering tests indicate the degree of com-
plexity of the combustion and gas cleaning equipment, turbine life, etc.
Preliminary costs have been estimated by the Combustion Power Co., but
direct comparison with values derived on this program should be avoided.
The capital cost of $4 million corresponds to $10,000 per ton per day in-
cinerator capacity or about $375 per kw installed capacity as electric plants
are evaluated. This compa:res with $120-140/kw for present stations in the
same size range (Re£. 78). Under private company ownership, it is esti-
mated refuse disposal costs would,total $5. 99/ton, without credit for by-
product sales. With credit for electric power (82% load factor and 4. 5
mills /kw-hr), disposal costs are estimated at $2. 67/ton. Municipal
ownership is claimed to lower these values to $4.27 and $0. 95/ton.
Utilization of the turb~ne waste heat would further reduce these costs.
2.
Other Applications
Heat from the combustion of refuse may be used in
numerous ways to benefit the surrounding community. In addition to the
obvious use for electrical power generation, there are several other poten-
tial end uses that deserve consideration. Among these are industrial and
district heating (and cooling), desalination, and various miscellaneous ap-
plications. These uses utilize steam of a relatively lower grade than that
required for power generation, and in some cases could use the steam dis-
charged from another process. The secondary heat sources that may be
considered in this respect are: (1) the steam bled from one or more stages
of an extraction turbine, (2) the discharge from a back pressure turbine,
(3) output from a low pressure boiler, and (4) steam discharged from some
higher level process. The last option may permit the cascading of two or
more uses in series.
a.
Industrial and District Heating
Heat from refuse could be used by many different
industries, ranging from chemical processing plants to food packing to
commercial laundries. The form of energy required varies over such a
broad range that each particular industry would have to be evaluated on
ll- 69

-------
an individual basis. This is not to imply that industrial heat should be dis-
regarded as a potential use for refuse energy, but that the type and density
of the surrounding industries may well dictate the steam conditions and
steam generating equipment required. District heating has become of in-
creasing importance for space and hot water systems. Tremba (Ref. 79),
in his recent review of all known district heating systems, indicates the
output magnitude of present installations. During the years 1965-67 some
2. 9 x 101? Btu's were distributed, 88. 5% within the Soviet Union* and 9. 1 %
within the western world. In combined heat and power plants, approximately
1. 1 x 1014 watt-hours of electricity were produced. Of the 226 heating in-
stallations surveyed in the western world, 21 stated they used refuse as a
portion of the fuel for the boilers.
In Denmark, 30% of all dwellings are now heated
from central boiler plants, and this is expected to increase to 40-45% by
1975 (Ref. 80). About 450 plants have been built, and operation has been
successful at dwelling densities as low as 4/acre. Normally, pipe lengths
from the boiler to the farthest consumer are less than 2.5 miles. It is es-
timated that an annual 38 lb S02/dwelling connected to the heating system
is no longer discharged at the previous low levels of home chimneys. Selling
price to the small user in Denmark is in the range of $0.80 to $1.60/106 Btu,
which accounts for the popularity of these heating systems (allowance should
be made for the elimination of capital and maintenance costs of individual home
furnaces and water heaters in comparing such heating costs). Ml1'1rch (Ref. 80)
estimates that in Denmark a city's refuse would cover 6 to 8% of the heat con-
sumption over the year, and that if the usual wider refuse area is considered,
this percentage would be significantly greater. A test facility has been built
for producing gas from refuse (7 ft3/lb), which, after carburetting with 3.4%
butane, has a heating value of 510 Btu/ft3. Evaluation is now being made for
use of the plant waste heat plus the gas manufactured in a district heating
plant, as well as supplying a portion of the gas to city mains.
Kimura (Ref. 81) gives emphasis to the need for an
industrialized, high population density nation such as Japan to develop a plan
for optimum utilization of all energy factors. Within his recommended "cir-
culation network of city energy, II the burning of refuse plays an important
role in a number of heat generating systems.
In his review of district heating systems in the Mann-
heim area, Winkens (Ref. 82) described that city's Nord plant in which 80 t/h
of refuse is fired in two boilers in addition to three oil-fired boilers. Steam
~~The siting of power facilities within cities has led to low distribution costs
of both heat and electricity in Russia, with consequent extensive employment
of district heating. In Moscow, for example, 22,000 buildings and 360 in-
dustrial plants are connected to a heating system containing 750 miles of
p1.pe.
II- 70

-------
is supplied to transmission grids at both 20 and 7 atm pressure, with some
250, 000 tons being supplied in 1969 from the incineration of 130, 000 tons of
refuse. Expansion of the waste burning facility is planned in the near future.
Britain I s first combined refuse incinerator and dis-
trict heating plant is being constructed in Nottingham (Ref. 83). The output
will increase through planned expansions to 440 million Btu/hr by 1980. At
this latter time, some 170,000 tons of refuse per year would be burned, pro-
viding up to two-thirds of the total ,heat output (remainder from coal). Steam
from separately-fired coal and refuse furnaces will be used for electrical
generation and the supplying of hot water at 2850F and 85 psi to distribution
mains. Reduction of about one-third in the cost of heating and hot water are
anticipated.
The Northwest incinerator plant in Chicago, now
under construction, will utilize a portiotof its steam for district heating
(Ref. 84). An average production of 3 1 steam per lb refuse is anticipated
and the facility is designed to handle 560,000 tons per year of refuse. Most
feedwater pumps and fans at this plant will be steam turbine driven once
start-up is achieved.
The most advantageous applications of absorption
refrigeration are in cases where suitable amounts of low- grade heat are
available, such as can be the case with a refuse-fired boiler. Where re-
ciprocating compression machines are limited by volumetric displacement,
the absorption machines can maintain capacity at lower back pressures by
increasing the flow of heat to the generator. Absorption systems generally
require no special building considerations. With the application of suitable
controls, the absorption system will operate with very little attention. Ab-
sorption refrigeration systems have been built in capacities of 2000 tons
(24 x 106 Btu/hr). The absorption system does not respond well to rapid
changes in load, but should operate nicely in a district cooling application
where the load is avefaged by the distribution system. Conventional com-
pressor driven refrigeration or heat pump systems could also use steam
turbines as the driving device in large district cooling or heating systems.
No detailed analysis has yet been made of the size
range over which economical operation of district heating from refuse
combustion might be realized. Rather typical of the qualitative analyses
that have been made is that of Beningson (Ref. 85), where wide application
of incineration energy is encouraged, but not justified on an engineering or
economic basis.
purpose
rator is
A generalized economic optimization of a dual-
(heat and electricity), multi-fuel (refuse and fossil), steam gene~
affected by a large number of variables,' chief among which are:
1I- 71

-------
.1
Regional electrical demand and anticipated
growth rate
.
District (local) heat demand and anticipated
growth rate
.
Distribution of population density (cost of
transmission system*)
.
Temporal effects on output demand for both
services
.
Annual outdoor temperature statistics
.
Fossil fuel costs (heating basis)
.
Refuse haul and proces sing costs
.
Refuse heating value
.
Factor costs (capital charges and labor).
These determinants will indicate the basic steam cycle that will lead to lowest
annual costs for the total system (refuse disposal, heat supply, and electrical
generation), which in turn would permit firm design of the furnace-boiler-
turbine-condenser components. At the 1st International District Heating Con-
vention held in London in 1970, a number of suggestions were made (Refs. 82,
86-93) for optimization techniques for dual-purpose plants using fossil fuel
only. Adaptations of these could be made for the more complex case with
waste fuel for a specific set of assumptions.
b.
Desalination and Miscellaneous Applications
Waste heat can serve as an excellent evaporation
source for a desalination facility. Plant optimization depends upon factor
costs, but a typical yield of fresh water from brackish or sea water is about
10 lb per lb steam (Ref. 94). In areas where natural water shortages exist,
the use of the energy froom refuse combustion should definitely be considered,
particularly after a portion of the energy has been extracted in a turbine
system.
*It is essential that the heat consumers be near the generating facility,
while this is not the case for the users of the electrical energy. However,
a privately owned dual-purpose plant can still compete with grid power in
spite of higher electrical generating costs when their distribution costs
are small. .
ll- 72

-------
Steam deri ved from fos sil fuels has been used for
snow and ice n"lelting on sidewalks, roads, and airport runways; properly
designed systems using refuse as a fuel should offer significant cost ad-
vantages. Other potential applications include the use of refuse-derived
steanl as the motive power for driving pumps, fans, and ejectors. Where
linear actuation is requir_ed, steam has been satisfactorily used for purposes
ranging from cla'mping and punching operations to the catapult launching of
aircraft. Acoustic generators of many types are used in industrial applica-
tions. Considerable interest has developed in jet cutting of many materials
and here steam could supply the required energy for many such operations.
II- 73

-------
III.
POWER PLANT DESIGN CONSIDERATIONS
A.
STATE OF THE.AR T SURVEY OF FOSSIL FUEL AND
REFUSE- FIRED BOILERS
1.
Fossil Fuel Boilers
The generation of steam from the combustion of a fuel,
whether it be a fossil one or a waste material, is simple in concept, yet
a highly specialized technology when long -term cost minimization is a re-
quirement. Fortunately for the cause of future heat recovery incinerator
systems, the fos sil fuel boiler industry in both the U. S. and Europe is
dynamic and innovative, and a rather considerable history of continuing
efficiency improvements exists . An understanding of the basic art is im-
portant to many individuals whose function it is to formulate recommendations
or decisions on the purchase of boiler equipment. Such an understanding is
not es sential, however, to an appreciation of the conclusions presented in
Volume I of this report, and development of the detailed background infor-
mation is therefore reserved for Section III of Appendix B (Volume U).
All trends within the thermal electrical utility industry,
other than price of the product, have been those of spectacular growth.
In recent years, unit sizes have increased by more than a factor of ten,
the quantity of power produced per unit mas s of fuel by a factor of two,
steam pressures and temperatures have rea.ched levels previously thought
impractical, and demand has steadily increased at 7% per year. These
accomplishments have been realized through the development of a variety
of designs of furnaces, heat exchange systems, and auxiliary equipment,
key features of which are described in the referenced appendix section.
These fossil fuel plant designs are important. to the present analysis of
heat recovery combustion of solid waste in an air pollution-free manner
for the following reasons:
. Only combined firing (coal/oil and refuse) systems
are being considered, with a significant portion of the heat always being
obtained from the fossil fuel for reasons of systems control.
. Radiative and convective heat transfer are affected
only slightly by fuel composition; the vast experience gained in conventional
boilers in steam cycle optimization can be directly applied to dual-fuel
plants.
. Coal ash compositions have been sufficiently variable
throughout the industry that prediction of corrosion and fouling can be based
on elemental analysis; these theories can now be applied to proper design of
combined fired systems.
'.
Ill-I

-------
. Cost estimates for refuse combination systems
gain increased accuracy when the experience with conventional boilers is
used.
. Air pollution control technology now used or being
developed for fossil fuel systems is directly applicable to refuse systems.
2.
Refuse Installations
Section V of Appendix B (Volume II) describes key fea-
tures of selected refuse fired steam generators. Reference should be made
to the Bibliography (Appendix D) for a listing of general reviews of design
characteristics of such refuse burners and discus sions of specific confi-
gurations of a number of recent units. Plants that can consume several
thousand tons per day of refuse have been built and are being operated with
no greater problems than those experienced in a comparable size facility
using coal only as a fuel. The use of a number of different agitating grate
systems and properly designed air introduction schemes permits a high
degree of burn-out. The water walls used for steam generation reduce
the volume of gas to be handled by the air cleaners while simultaneously
permitting lower operating costs through sales of steam to electric utilities
or central heat distributors.
B.
CATALOG OF CANDIDATE SYSTEMS
1.
Introduction
An important task of this program has been the develop-
ment of engineering and cost data on possible boiler configurations fired by
refuse. The catalog of such cnadidates can then be systematically analyzed
to identify those design forms which offer the more favorable bases for opti-
mization. The selection of such systems has been based on the state-of-the-
art survey, discussed in Appendix B, Volume II, and the characteristics of
refuse, discussed in Section II. An aim, in developing this catalog, has been
to propose two classes of designs. Category I is intended to include systems
that would influence new plant design and construction. Category II systerns
are applicable to "retrofit" situations, wherein the intention is to modify
existing, older plants to accommodate them to refuse firing. Ten candidates
have been developed in Category I, five candidates in Category II.
Estimates of the performance characteristics of the candi-
date configurations have been calculated. This required the adoption of certain
as sumptions and the imposition of various design constraints. These are
itemized as follows:
III-2

-------
a.
Fuel Value
(l)
Refuse
Based on the information on refuse properties
that was reviewed in Section II, a heating value (HHV) of 4460 Btu/lb was
adopted as the probable average. The ultimate analysis considered typical
for present day refuse is:
TYPICAL ULTIMATE ANALYSIS FOR REFUSE
Component
Wt-%
Moisture
Carbon
27. 1
25.5
3.4
21. 7
O. 5
O. 1
21. 7
Hydrogen
Oxygen
Nitrogen
Sulfur
Inerts
Glass (9.2)

Metals (7. l)
Ash (5.4)
(2 )
Coal
In order to avoid a cumbersome and overly
cornplex presentation of data, combustion calculations were restricted to
cases involving refuse/coal firing. Bituminous coal, of course, is the pre-
dominant fossil fuel in conventional utility use and does represent the worst
case situation in terms of emissions. The performance derivations outlined
herein are, however, of such a format that parallel derivations, based on
oil or natural gas, can be readily obtained.
The heating value (HHV) used for coal was
taken to be 12,022 Btu/lb, an average of typical values found in various
source documents. The ultimate analysis for coal employed is:
III - 3

-------
TYPICAL ULTIMATE ANALYSIS
FOR BITUMINOUS COAL
Component
Wt. -%
Moisture
Carbon
7. 8
65.2
4. 5
7.6
Hydrogen
Oxygen
Nitrogen
Sulfur
1.4
3. 5

10.0
Inert
b.
Refuse Charging Rate
In consideration of logistic factors, grate require-
ments, and other constraints, an arbitrary upper limit in refuse rate of
1000 tons per day (tpd) per unit has been observed for systems equipped
with grates. Candidate designs, in which grates are not used, have been
limited to inputs of 2000 tpd per unit.
A further criterion involves fuel proportions. A
refuse fraction between 20% and 60%, based on heat input, has generally
been sought. Diseconomies would tend to operate at lower refuse propor-
tions, while at levels above 600/0 steam fluctuations would become increasingly
difficult to control with the complementary fossil fuel firing-system. It
should be borne in mind that the fuel fraction term applies to the energy
input for a complete steaIJl generator system. The latter may include one
or more furnaces that fire straight refuse coupled with, say, a superheater
firing only fossil fuel.
c.
Unit Capacity
An upper limit of 500 MW has been used in deriving
performance data. Four size-ranges have been defined that fit withi~ this
limitation and provide bases for the as signment of steam conditions. The
upper range limits are at 40, 85, 215, and 500 MW. As discussed in the
next paragraph, these ranges bracket or conform with four statistical
groupings of boiler sizes within which recent boiler sales in the small
and intermediate class have tended to fall.
d.
Steam Conditions
ditions were
practice. A
As in the as signment of size ranges, steam con-
selected that would conform with contemporary engineering
survey of boiler sales for a three-year period ending in 1968
III-4

-------
was made. It was found that four steam conditions, one within each of the
size ranges, were specified for slightly over 50% of the boilers (fos sil fuel
fired) ordered during the period. The total number of steam conditions used
(including those for nuclear plants) was 21 for the first and third year and
22 for the second year of the survey period. The four most frequently used
steam conditions are listed below together with the associated size ranges.
In cataloging candidate systems, these stearn conditions and size r'anges
were applied to both the boilers and turbines.
STEAM CONDITIONS SELECTED
FOR CATALOGED SYSTEMS
Steam Conditions
psig/SH,oF /RH,oF

2400/1000/1000
1800/l 000 /l 000
1250/950/ -
850/900/ -
Size of Units
actually ordered,
MW
Associated
Size Range,
MW
235 - 650
90 - 195
52' - 80
20 - 26
215 - 500
85 - 215
40 - 85
<40
e.
Other Operating Conditions
It has been assumed that an exit flue gas temperature
of 4500F will be observed in order to minimize corrosion from acidic con-
densates. Further, 50% exces s air will be employed for firing refuse and
18% for coal. Of the total air for grate-stoked furnaces, 75% will be treated
as (preheated) underfire air and 25% as (unheated) overfire air. On the basis
that the flue gas enters the air heater at 6250F and leaves at 4500F, the air
entering at 800F will be discharged at about 3500F, a reasonable temperature
for cast iron grates now on the market.
In defining the performance parameters for the
, various candidate systems, no allowance has been made for the possible
effects of air cleaning equipment. This is treated in a separate subsecti,?n.
2.
Performance Parameters
a.
Energy Requirements
Shown in Figure III-l are representative heat rate
curves for conventional plants with coal-fired units according to the selected
steam conditions. Superimposed on these curves are the associated size
ranges. Thus, for this design study only the solid portions of the curves
are applicable. From these data the heat input to the turbine has been de-
rived for the selected steam conditions as shown in Figures III-2 and -3.
III- 5

-------
15,000
14.000
13.000
...
.s::
,
~
.:tt:
-
:J
-
m
12,000
~
W
t-
«
a:
t-
«
w
:I:
11.000
10,000
9,000
8,000
40MW
I
I
I
I
I
I
\
I ,
I ,
I ,
, I
I " ~ 1800psig/IOOO°F/ IOOO°F

I' /:
"'... I
" I
...
'.." "....
'.... I '....
....
85MW
I
Y NONREHEAT
2i5MW
,
/ 2400psig/IOOooF/ 1000°F
100
200 300
NAMEPLATE RATING,MW
500
400
FIGURE 111-1. NET PLANT HEAT RATE fOR VARiOUS STEAM CONDITiONS
IlI- 6

-------
...
~
"-
~
~
m
0\0
~
....
:>
c
a:
o
....

-------
~
.s::
'"""-
~
~
m
~O 0.4

~
~
::>
o
a::
o 0.3
~
«
a::
w
z
w
(!)
~
«
w 0.2
~
en
~
w
z
0.8
0.7
0.6
0.5
850psig/900°F ~ 1250 psiO/950oF~
0.1
~~
30 40 50
NAMEPLATE RATING, MW
FIGURE 111-3. TURBINE HEAT INPUT FOR STEAM CONDITiONS Of NONREHEAT UNITS
III- 8

-------
The heat input to the turbine is the same as the heat output of the steam
generator, often referred to as I 'duty. " The heat input to the steam gene-
rator is therefore the duty divided by the steam generator efficiency. The
net plant heat rate is defined as th~ heat input divided by the net generation.
However, for simplicity, it has been assumed that the net generation is
equal to the nameplate rating. The equations employed are shown as
follows:
Q
"'7 - . SG out
SG~ - QSG in
(III- I)
whe re:
"'7SGC
QSG out
= Combined steam generator efficiency
= Steam generator duty, Btu/hr
QSG in
= Steam generator input, Btu/hr
H
c
=
QSG in
PT
=
QSG ou/"'7 SGC
PT
(III- 2)
Where:
H
c
= Net plant heat rate, Btu/kw-hr
PT
= Nameplate rating, kw
By this procedure, the net turbine heat rate (HT)
has been implied in the steam generator duty in Figures III- 2 and - 3. By
utilizing these figures, it is much simpler to vary the steam generator ef-
ficiency and calculate fuel requirements, especially with such different fuels
as refuse and coal.
Having defined the relationships of the basic steam
generator ratings, it is next necessary to establish the combined steam,
generator efficiency ("'7 SGC)' This will then permit the calculation of refuse-
and coal-feed requirements for the- selected steam conditions and nameplate
ratings, as well as such system characteristics as heat rate, flue gas volumes,
and steam flow.
The combined steam generator efficiency is deter-
mined by weighting the efficiencies contributed by the individual fuels; that
IS:
1II-9

-------
"7 SGC = "7 SGW x fw + "7SGF x (l-fw)
(III- 3 )
where:
"7SGW = Steam generator efficiency with waste fuel (refuse)
= Steam generator efficiency with fossil fuel (coal)
"7SGF
f
w
= Fraction of heat input from waste (refuse)
Based on the input data of nameplate capacity (PT),
the fraction of heat input from waste (£w), the steam generator duty (QSG out),
and the steam generator efficiencies ("7SGW and "7SGF), it is possible to cal-
culate, using Equations III-I and -3, the required steam generator heat input
(QSG in)' From QSG in' the fuel input requirements can then be calculated:
QSG in = (W w x hw) fw + (W f x hf) (l-fw)
(III- 4)
where:
W
w
= Waste load (refuse), tpd
h
w
= Waste fuel heat value (refuse), Btu/lb
= Fos sil fuel load (coal), tpd
Wf
h '
f
= Fos sil fuel heat value (coal), Btu/lb
As specified previously, the refuse (hw) and coal
(hf) heating values (as received) are 4460 and 12,022 Btu/lb, respectively.
The only terms which are, thus far, not explicit are the steam generator
efficiencies with waste fuel ("7 SGW) and with fos sil fuel ("7 SGF)' These must
be determined on the basis of combustion calculations, which are explained
in the following two subsections.
b.
Combustion Calculations
(1 )
Refuse
Using the ultimate analysis listed earlier, the
stoichiometric and excess combustion gas requirements were calculated to
be as follows:
III-IO

-------
\
COMBUSTION GAS REQUIREMENTS FOR REFUSE
Constituent
Combustion Gas,
Ib lIb Refuse
02 Dry Air
Oxygen
Total Requirement, Stoichiometric
0.679 2.928
0.270 1. 163
0.001 0.004
0.010 0.042
0.960 4.137
-0.217 -0.937
0.743 3.200
1. 115 4.800
0.372 1.600
Carbon
Hydrogen
Sulfur
Metal, Partial Oxidation
Total Requirement, @ 50% Excess
Exces s Gas
Again based on the ultimate analysis and cer-
tain of the data tabulated above, the products of combustion were calculated.
The results are. shown in Table Ill-I. The efficiency of a steam generator
fired with the specified refuse composition was then calculated, as itemized
in Table I11-2, to give a value of 69.01%. This value was subsequently as-
sumed to be constant for all size steam generators operated under the
specified conditions.
(2)
Coal
Using the heating value (12,022 Btullb - HHV)
and ultimate analysis previously given, similar efficiency calculations were
performed for the coal-fired boiler. Assuming the use of 18% excess air and
a flue gas exit-temperature of 2850F, an efficiency of 88.0% was determined
to be applicable for the four selected steam conditions. Based on 1 % decrease
of cycle efficiency for each 400F increase in flue gas exit temperature, a steam
generator efficiency of 84.0% is obtained, if the exit gas is at 450oF.
Other key data obtained in these (routine) cal-
culations are as follows:
Component
Lb Gas I Lb of Coal
18% Excess Air 0% Excess Air
Combustion Air
Wet Flue Gas
11. 50

12. 55
9. 75
10.77
It should be pointed out that a tubular air heater
is contemplated for use in the candidate systems because of its ease of cleaning.
Thus, there will be no air leakage to contend with, as with a regenerative air
heater.
Ill- 11

-------
    TABLE III-1   
  PRODUCTS OF REFUSE COMBUSTION  
      Gas Formed 
      pe r Lb Re£us e Vo1-0/0,
Constituent  Deri vation Factor Lb Mol Lb Dry Basis
    C    
COZ   W£ x MCO /MC O.OZlZ 0.933 1Z.9
   Z   
HZO   (1) + (Z) + (3) (0. 035) (0.6Z9) 
 ( 1) Re£use HZ HZ  0.017 0.304 
 W£ x MH O/MH 
   Z Z   
 (Z) Re£us e HZ 0 WHZO  0.015 0.Z71 
 £  
 (3) Comb. Air HZO (Note 1)  O. 003 0.054 
    S    
SOZ   W£ x MSO /MS 0.00003 O.OOZ O.OZ
   Z   
Oz    Excess Amount 0.0116 0.37Z 7. 1
NZ    (4)+(5)  (0.131Z) (3.694) (80. 00)
 (4)   Total Air x % NZ /1 00 0.1310 3.689 79.88
 (5)   WN  O.OOOZ O. 005 O.OZ
   £  
Total Flue Gas (Wet)   O. 1990 5.630 
Total Flue Gas (Dry), G£   0.1640 5. 000 
(Note 1).
Based on standard (600/0 RH at 80oF) o£ Amer. Boiler Mfrs. Assn.
Explanation o£ Terms Used:
W£
=
Weight fraction (per refuse ultimate analysis;
superscripts designate carbon (C), moisture (HZO),
nitrogen (N), sulfur (S), and hydrogen (HZ)'
M
=
Molecular weight; subscripts indicate specie.
III - 1 Z

-------
TABLE 1lI-2
REFUSE FIRED STEAM GENERATOR EFFICIENCIES
Items
Btu lIb
of Refuse
% of
Heating
Value
A.
Energy Utilized (specified refuse fuel value)
4460
100
B.
Heat Losses
1. Dry Gas, G f x C x (tl - t . )   443 9.93
   p vg au.   
2. Moisture in fuel, w~20 (HtS - H f  330 7.40
t . 
    1vg alr   
3. H20 from H2 Combustion,    
 W H2 (H S f   372 8.34
 f tl - Ht . ) MH O/MH  
  vg au 2 2   
4. Moisture in Air, S. H. x Lb Air ILb Refuse x   
 (HS - HS     10.6 0.24
 t t .     
 lvg alr    
5. Unburned Gas, H. V. C x ~ x V CO/V CO  3. 9 0.09
      2   
6. Unburned Residue Constituents (assumed at 8%),  
 80% (Inert-metal) x 8% H. V. b  79.0 1. 77
      com   
7. Sensible Heat in Residue,    
 80% Inert x C (t - t . )   34.8 0.78
    pres alr    
8. Unburned Fly Ash Constituents (assumed at 12%),  
 20% (Inert-metal) x 12% x H. V. b  29.6 0.66
      com   
III - 1 3

-------
1-,
100
11.
12.
TABLE IIl-2 (Continued)
Items
9.
Sensible Heat in Fly Ash,
20o/c (Inert Metal)x C (tl - t . )
P vg aIr
Subtotal
Radiation (standard estimate)
Unmeasured (standard estimate)
Manufacturer's Margin (standard estimate)
Total o/c Heat Loss
Steam Generator Efficiency, %
Explanation of Terms Used:
Btu/lb
of Refuse
o/c 0 f
Heating Value
3.4 0.08
1306. 3 29.29
 0.20
 0.50
 1. 00
 30.99
 69.01
Cb
= Carbon burned per Ib of refuse (0.251 Ib; assuming 0.004 Ib
is unburned).
= Specific heat (Dry gas = 0.24; Residue = 0.30 Btu/lb-FO).
Lb dry flue gas/lb :refuse.
C
P

Gf

HS
t
Hf
t
=
=
Enthalpy of steam at tOF, Btu/lb.
=
Enthalpy of water at tOF, Btu/lb.
Ho Vo = Heating value: C(arbon) = 14,100 Btu/lb;
comb(ustibles) in residue = 8460 Btu/lb.
S. H. =
Specific Humidity, Ib HZO/lb airo
I11-14

-------
I ~
l~
t .
. alr
t
lvg
t
res
v
w£
- "
TABLE llI-2 (Continued)
=
Input air temperature (assumed 80°F)
= Stack gas temperature (assumed 450°F)
Gr.ate residue temperature (assumed 750°F)
=
=
Vol-Ole (CO assumed at O. o 2 Ole)
=
Weight fraction. per ultimate analys is.
Ill- 1 5

-------
3.
Operating Parameters
a.
Fuel Requirements
Having determined the steam generator efficiencies,
all the necessary input data are available for calculating the fuel require-
ments. These calculations have been made according to the format shown
in Table III-3, which incorporates a sample calculation for a boiler of a
nameplate rating of 400 MW. Basically, this calculation is a trial and error
procedure. A flue gas exit temperature of 4500F is assumed throughout even
where no refuse is burned. This has been done to establish a continuity be-
tween combined and refuse-only firing. Beginning with 0 percent refuse rate,
Figure III- 2 indicates a 2400/1 000/1000 cycle. One turbine is applicable.
The steam generator efficiency is 84. 0% and the coal rate is calculated as
3960 tpd. The net plant heat rate is calculated as 9,880 Btu/kw-hr. A
similar plant with a conventional steam generator efficiency would yield a
net plant heat rate of 9, 300 Btu/kw-hr as shown in Figure II-I. For a
refuse rate of 20%, again a 2400/1000/1000 cycle would be assumed. The
steam conditions and nameplate rating allow the use of one turbine and one
steam generator. The net steam generator efficiency, according to Equation
,IlI-3, is 81. 0%. Because the refuse rate (Item 16) is in excess of 1000 tpd,
however, two steam generators would be required, each of approximately
200 MW rating. From Figure IlI-2, however, at 200 MW a reduction to a
1800/1000/1000 cycle will be required. At 1800 psig, according to the same
figure, the maximum size for the turbine is 215 MW. Thus, for this 400 MW
plant, two 200 MW turbines will be required. As shown in Table IlI-2, the
total heat input for two 200 MW turbines is greater than for one 400 MW tur-
bine (d. Item 6). Because of this, the required refuse rate per steam gene-
rator again exceeds 1000 tpd. Therefore, three steam generators will be
required to supply a total of 3.50 x 109 Btu/hr (d. Item 6) or 1. 17 x 109
Btu/hr per unit (d. Item 9). A check against Figure IlI- 2 indicates that a
steam generator duty of 1. 17 x 109 Btu/hr is equivalent to a 133 MW single
steam generator, single turbine plant. It is shown that for three steam
generators the refuse rate per steam generator is 778 tpd. Also shown is
the net plant heat rate of 10,810 Btu/kw-hr.
.~
The complete map of fuel requirements for plants of
up to a nameplate rating of 500 MW are shown in Figures IlI-4 and -5. Each
circled point represents a value calculated in the same manner as was used
for the example in Table IlI-3. The discontinuities of Figure III-l at changes
in steam conditions also appear in Figures IlI-4 and - 5. The number of re-
quired steam generators can be read at any point on these figures by reading
the refuse rate, then dividing by 1000, and rounding off to the next higher
integer.
The limits of the steam cycles in Figures III-4 and
-5 are shown with dashed lines. A criterion used to develop these limits
was that each steam cycle should be extended to the largest refuse rate
pos sible. This extension is desirable because plant efficiency is sacrificed
in dropping from a higher to a lower pressure steam cycle.
IlI-16

-------
         TAB ~E I11-3      
      CALC JLA TIOI" PROCED JRE FOR D ~TERMINII'G     
      5 TEAM GENERA TOR (5. G. ) CHARAC TERIS TICS     
           Nameplate Rating, MW (a)  
 Item  Parameter    Data Derivation 400 400 400 400  
 1. Refuse Rate, o/c      0 20 20 20  
 2. Steam Pressure, psig  Figure Ill-I 2400 2400 1800 1800  
 3. Number of Turbines    See text  1 1 2 2  
 4. Net Turbine Rating, MW  (a) / (3)  400 400 200 200  
 5. ' 9 Figure Ill- 2 3. 32 3. 32 1. 75 1. 75  
 Heat Input per Turbine, 10 Btu/hr  
 6. Total Turbine Heat Input, 9 (5)/(3)  3.32 3.32 3.50 3.50 '. 
 10 Btu/hr  ::-'4",/,..
 7. Number of S. G. 's    Figure III- 2 1 1 2 3  
 8. S. G. Size (approx.), MW  Figure 1II-2 400 400 200 133  
 9. Heat Output per S. G., 109 Btu/hr (6)/(7)  3.32 3.32 1. 75 1. 17  
 10. S. G. Efficiency due to Refuse, o/c Table III- 2 0 69.0 6. 90 6.90  "
  , ,
~ 11. S. G. Efficiency due to Coal, o/c See text  84.4 84.4 8.44 8.44  "
1-1   
I                
..... 12. Net S. G. Efficiency    Equation IIl- 3 84.0 81. 0 81. 0 81. O. ;.' .~
-....J     
 13. Heat Input per S. G., 9  (9)/(12)  3.96 4. 10 2.16 ' 1. 44 . '-~' 
 10 Btu/hr  
 14. Heat Input per S. G. from Refuse, (1)/(13) x 100 0 0.82 0.43 0.29  
  109 Btu/hr      
 15. Heat9Input per S. G. from Coal, 100-(1)/(13) x 100 3.96 3.28 1. 73 1. 15  
  10 Btu/hr           
 16. Refuse Rate per S. G., tpd   3 0 2210 1168 778  
  2.7xl0 x(14)  
 1 7. Plant Refuse Rate, tpd  ( 7) / (16)  0 2210 2330 2330  
 18. Coal Rate per S. G., tpd   3 3960 3280 1728 1153  
  1.0xl0 x (15)   
 19. Plant Coal Rate, tpd    (7) x (18) 3960 3280 3456 3459  
 20. Net Plant Heat Rate, Btu/kw-hr (7) x (13)/(a) x 103 9880 10250 10810 10810  

-------
-
w
~
ct
a::
...J
ct
o
u
o
10
II
FIGURE 111-4. FUEL INPUT FOR PLANTS UP TO 350 MW
III- 18

-------
I
I -
J- 2.5
<'b
po
u.i
I-
«
a:
~ 2.0
o
(,)
4.5
3.0
1.5
1.0
0.5
o
o
2
20
22
6
8
4
10
12
14
18
16
REFUSE RATE. 103 tpd
FIGURE 111-5. FUEL INPUT FOR PLANTS UP TO 500 MW
IIl- 1 9 .

-------
For a given size plant and steam cycle, the greatest
refuse rate will occur in a plant with a maximum number of steam generators.
The size of the steam generators was found by determining the smallest steam.
generator that would be an integer divisor of the plant nameplate rating and
still b~ within the cycle limits in Figures 1II-4 and -5. When the sizes of the
steam generators were determined, each one was assigned a refuse rate of
I, 000 tpd. The coal rate (W f) was then calculated from the following equation,
the terms and derivation of which are based on Equations III-I through -4:
Wf = (QSG out - "'SGW W wHw)/'" SGF hf
(III- 5)
b.
Heat Rate
The net plant heat rate is an important parameter,
because it is an indication of total plant efficiency. Using the criteria and
assumptions observed in developing the power plant models, it can be shown
that the net plant heat rate is a function of the individual turbine size and the
fraction of heat derived from refuse. The net plant heat rate is defined by
Equation III- 2. In this report, steam generator duty is taken as turbine heat
input. Figures III-2 and - 3 show the relationship between turbine heat input
and nameplate rating, the latter being assumed to be the same as net genera-
tion. The ratio of the total plant heat input to the plant nameplate rating would
thus be identical to the ratio of the individual turbine heat input to the steam
generator efficiency divided by the individual turbine nameplate rating. This
would then mean that for any given turbine size, the net plant heat rate would
be a function only of the steam generator efficiency. The combined steam
generator efficiency (." SGC) is a function (Equation III- 3) of the fraction of
heat derived from the refuse fired. Thus, a plot of net plant heat rate versus
fraction of heat from refuse can be prepared using turbine nameplate rating
as a parameter, as shown in Figure III- 6. It was as sumed that a plant using
certain size turbines will have the same net plant heat rate regardles s of the
number of turbines that must be used.
Where shift-overs in steam conditions are involved,
heat rate curves are given in Figure III-6 for both of the steam conditions.
As in Table III-3, the lower steam pressure has been used for the fuel cal-
culations. Because it was not possible to include on Figure III-6 the limits
of the percent refuse that can be fired, Figures III- 5 and - 6 must still be
used to find the range of refuse firing that is applicable for a given plant.
c.
Steam Flow
In addition to fuel requirements and the resultant
air and flue gas flows, another important parameter is steam flow. To
retain the simplicity of the previous calculations, certain as sumptions have
been made in calculating steam flows. The four steam conditions selected
III- 20

-------
I -
CD
It)
o
I '-"
I
..
.&:
I
~
.IIC
.......
~
-
".~" ~~.. ~'~.:: "'.'t~.' l". :.. ....: ,...':r..',:',

. . Individual Turbine Rating, MW I
'.' . 0 20
0/
0/
0/ /0 30
/ 0
o / .:J
/' 0 0 40 1:0
15.5 (;) / / ~
0'/ /0 ./(i) ° 50
/ 0/ 0/ /'
/0 0/ 0/ /0 060
/0 ./ /. /0 0/p70
o /0 /0 0/ ./,/'
/ 0/ 0/ /.' 0//0 <:> 85
14 0 / / 0 /' /' /0 85
. 0 0/,/ 0/ 0 0,. ,.
0/ 0/,/0 0/ 0/ 0~0" 0 100
/ /0 /./ /0; / 0
./0 0/ ./0 /0 /0 ; /0 0
0""""" /' 0/ 0""""" 0/ ;0"'; 0/ 0 150 ~
13.0 ;./ /,,0 / /" /",.'" / ./"
/0 o/:/:/~...,....0/0 /0.......:8 ~P50
12.5 0/0/0~0"""" /0 /0 8~0 215JO
"""""""0/ /....'" /0 /0 Q~O'" . 300 ~
12.0 /0...--::/:.........0 /0./ $~:--- ~~~ ~gg N
~O"'''' 0/ ~0 B~O'''' ~~~ .~
11.5 .,../ .,../0 e~ 0"" ,Q~B""""" fI)
..............0 ,........,. 0 e~0" ..."" 0 ~t!S~ ;
11.0 0/"° 8~0"'''' ~~B~ . ~
,.........,. ~~~0~~*~~ ~
10.5 ~~/ ~~~~. ~
~~~.~ ~
10.0 ~~ U1
17.5
17.0
16.5
16.0
15.0
14.5
..
w
....
«
a:
....
«
w
::I:
....
Z
«
...J
a..
....
w
z
13.5
9.5
9.0
o
10
20
30 40 50 60 70 80 90
HEAT INPUT FROM REFUSE, %
100
FIGURE 111-6. EFFECT OF REFUSE HEAT INPUT ON NET PLANT HEAT RATE
III- 21

-------
are considered to be throttle conditions. It has been assurned that there are
no losses from the steam generator outlet to the turbine throttle. However,
for a plant utilizing a 2400/1 000 /1000 cycle, the steam generator outlet con-
ditions would typically be 2520 psig / 1 0050F /1 0050F. The changes in the state:
points are due to pressure, but this does not markedly affect the enthalpy of
the system.
The steam flows have been calculated for the four
steam conditions using the following equation:
ws = (6Ht + wI' 6Hr> /QSG out
where:
Ws 
6Ht 
w 
l' 
6H 
l' 
QSG out
(III- 6 >
3
= Steam flow, 10 lb/hr
= Enthalpy difference between turbine
steam and feedwater output, Btu/lb

= Steam flow to reheater, 103 lb /hr
input
= Enthalpy change across reheater, Btu/lb
S. G. Duty, 109 Btu/hr.
=
The second term in the numerator of Equation III-6 is zero for the nonreheat
steam conditions of 850/900/ - and 1250/950/ -. The term, 6Hr' was derived
from reheat state points typical for the other two steam conditions. Similarly,
the ratio of wJ:. to ws was based on practice and assumed to be typically 0.88.
The value of USG out for various net generation rates was taken from Figure
1II-3. In order to avoid the complication of deriving feedwater inlet tempera-
ture s (and, thus, 6Ht> from turbine heat balances, these temperatures were
taken from the graph (Ref. 95> shown as Figure Ill- 7.
The points chosen for calculation were those which
were also employed in the fuel requirement calculations. The results of the
stearn-flow calculations are shown in Figure III-8.
It should be pointed out that, in cases where there are
multiple steam generators and turbines, the total steam flow is not directly
proportional to the nameplate rating.
In considering Figure 1II-8, the points of cycle change
merit further comment. At 215 MW, the steam flow for the 1800 psig cycle
is less than the 2400 cycle, even though the net steam generator duty is
greater for the former. This occurs because the net enthalphies for the two
cycles differ more than the duties. The difference in steam flows is also
III- 22

-------
800
750
  700 
0 ~  
w  
  650 

-------
 3.5         
        04..  
        00  
 3.0       ~  
       4..\  
        0  
        0  
       ~  
       ~  
       .~  
 2.5      ~  
      00   
      ~    
r-.      t\f    
~          
.......          
.a    ~      
 2.0   :IE      
CD         
0    It)      
..    (\J     
~    l "     
0        
...J        
I.L     '4..     
1.5    ,.     
    , 0     
:=E    , 00     
«    , 0     
1&.1        
.-  3=  ~     
VJ   04..     
  :e 0      
 1.0 10 0      
 en ,~       
  ~ l ' ~       
  :E ,.~       
  , "       
  0 'O~       
  v 0       
 0.5 l ,/~       
  '-1250 psiO/900o F/-    
  '- 850psig/900oP/-    
  I    
  /        
 I        
 0  I  :2  3 <4 5 E)
   NET STEAM GENERATOR DUTY, 109 Btu/hr  
FIGURE 111-8. STEAM FLOW AS A FUNCTION OF NET STEAM GENERATOR DUTY
III - Z 4

-------
caused by energy los ses occurring elsewhere in the cycle. At 85 MW the
stean1 flow for the nonreheat cycle is greater than for the reheat cycle.
This is logical, considering that there is no reheating. At 40 MW the
steam flow of the lower pressure cycle is greater. This is directly
caused by the differences in net enthalpies.
4.
Design Data for Candidate Systems
a.
Background
For synthesizing the system possibilities, several
types of steam generator arrangements have been considered. These are
as follows:
Case
Designation
5
Separate Furnace

Combined Furnace

Separate Refuse-Fired Economizer

Saturated Steam Unit with Separate
Superheater

Partial Superheat Unit with Separate
Superheater
1
2
3
4
6
7
8
9
Suspension Furnace
Spreader Stoker
E>lagging Furnace
Ar ch Furna ce
10
Arch Furnace with Separate Superheater
Cases 1 through 4 involve the use of a conventional,
agitating grate such as a backward reciprocating, forward-reciprocating,
roller, or tilting grate. Cases 5 through 10 require special refuse burning
equipment.
. The previously calculated fuel requirements (Section
III, B, 3, a) are appropriate unless otherwise noted. For most cases the ex-
cess air .level for refuse is 500/0 and for coal 18%. Air and flue gas rates are
tabulated in the appropriate sections for the various cases. The data pre-
s ented is for two different plant capacities - 200 and 400 MW. The appropriate
steam rating per steam generator is also tabulated.
III-25

-------
b.
Candidate System Characteristics
Case 1 - Separate Furnace. This denotes a system
in which refuse and coal are fired in separate furnaces. The respective flue
gases are then mixed and passed over convection sections of the steam gene-
rator, which includes an economizer and an air heater. The flue gases leave
the steam generator at 4500F and enter the flue-gas cleaner. Such an arrange-
ment was first utilized at Munich North, Block I, and, later, for two units at
"
Stuttgart-Munster.
The schematic is shown as Figure III-9 and the sum-
mary performance calculations in Table III-4. In this and subsequent tables,
the abbreviation "FG" is for flue gas. Each furnace is sized only for its de-
sign proportion of fuel. At high refuse rates some superheat surface would
be suspended in the refuse furnace.
Case 2 - Combined Furnace. Designated is a system
in which refuse and coal are fired in a common furnace. Such a configuration
was first utilized at Munich North, Block II. In this arrangement some of the
excess air supplied for refuse firing will be consumed by the coal, thereby re-
ducing the amount of air required for the latter. The air and flue gas rates
for the combined furnace can thus be set somewhat lower than for the separate
furnace (Case l). This practice was not followed during the performance tes':s
in Munich, however, so such an adjustment has not been projected in the pres-
ent design data. In any case, performance characteristics or cost would not
be greatly affected by exploiting this possible benefit.
The schematic of the combined furnace is shown as
Figure Ill-lO. The common furnace would be designed for full-load heat
liberation and the coal-burning equipment would be sized for full-load on
coal. The performance of this system would be the same as for Case 1;
the summary performance for both cases can thus be found in Table IIl-4.
Case 3 - Separate Refuse Fired Economizer. This
system consists of two separate furnaces in which the flue gases are not
combined until the gas cleaning stage is reached. In the refuse-fired fur-
nace, feedwater is heated as in a conventional economizer. The heated
feedwater is then sent to a conventional type steam generator, which is
equipped with proportionately less economizer surface. Because the refuse
heat output is absorbed only by an economizer, the fraction of heat that can
be derived from refuse is limited. Therefore, the refuse heat input cannot
be varied as for Cases land 2. An arrangement of this type was first utilized
at Munich South, Unit 6. At this site, the present fossil fuel used is naturaJ.
gas with provision for oil- or coal-firing. The flue gases exit the respective
furnaces and are 'mixed to lower the temperature of the dust-carrying flue
gas from the refuse furnace before it enters the electrostatic precipitator.
IIl-26

-------
[ .
I' PENDANT
SUPHTR,
t
... --p" .
-:e
~ .
- .
FIGURE 111-9, CASE 1 . SEPARATE FURNACE
111- 27

-------
TABLE III-4
CASE 1 - SEPARA TE FURNACE:
CALCULA TED SYSTEM CHARACTERISTICS
            Nameplate Rating, MW    
        ZOO       400  
 Refuse Rate, 0/,   0 20 40 60 80 100 0 20 40 60 80 100
 Steam Pressure, psig 1800 1800 1250 850 850 850 2400 1800 1250 1250 850 850
 Number of Turbines 1 1 3 5 5 5 1 2 5 5 5 10
 Number ofSG's   1 2 3 5 7 9 1 3 6 9 12 17
 SG Duty, 109 Btu/hr 1. 75 0.88 0.68 0.43 0.31 0.22 3.32 1. 17 0.66 0.44 0.33 0.25
   3  1429 715 610 400 300 220 2780 950 590 400 315 250
 Steam Flow, 10 1b/hr
 Refuse Rate, tpd  0 581 939 934 925 845 0 778 910 947 986 994
 Coal Rate, tpd  2080 864 521 230 86 0 3960 1153 507 234 91 0
   3   232 375 373 370 338 0 310 364 378 394 397
 Refuse Air, 10 1b/hr 0
H   103 1b/hr             
I::: Coal Air, 1992 828 499 221 82 0 3790 1105 484 224 87 0
,   103 1b/hr             
N Total Air, 1992 1060 874 594 452 338 3790 1415 848 602 481 397
00   103 1b/hr             
 Refuse FG, 0 272 440 437 434 396 0 364 426 438 462 465
 Cca1 FG, 1031b/hr 2180 904 545 241 90 0 4140 1205 530 245 96 0
 Total FG, 103 Ib/hr 2180 1176 985 678 524 396 4140 1569 956 683 558 465
 Net SG Efficiency, 0/, 84.0 81. 0 78.0 75.0 72.0 69.0 84.0 81. 0 78.0 -75.0 72.0 69.0
 Net Plant Heat Rate,             
 Btuikw-hr' 10,400 10,800 13,100 14,400 15,010 15,640 9,880 10,810 12,650 13, 150 15,010 15,610
        Refuse Coal        
 Excess Air, 0/,     50 18        
 Air Ratio, 1b Air/1b Fuel  4.80 11.50        
 Flue Gas Ratio, 1b FG (wet)/lb Fuel 5.63 12.55        
 NOTE: Data shown are for single SGrs.           

-------
. '. .' .' . ~
. . .
. REHEATER
OUTLET
~
. REHEATER
INLET
r
ECONOMIZER
INLET
~-
',l
. to
FIGURE 111-10. CASE 2 - COMBINED FURNACE
III - 29

-------
The schematic of the separate refuse fired econo-
rnizer is shown as Figure IlI- 11 and the summary performance is given in
Table Ill-5. For the plant sizes under consideration, the subcritical pres-
sure cycle with a natural circulation steam generator is more economical
than the super-critical pressure cycle with a once-through steam generator.
Normally, the heated water leaving the economizer enters the steam drum
where it is mixed and taken to external downcomers to headers at the lower
portion of the steam generator. The water then travels upward in waterwal1
tubing where boiling might take place. If the feedwater entering the drum
contains some steam, it can cause "carry-under" problems that can ad-
versely affect the circulation. For this case, therefore, the maximurn
ternperature for the economizer has been set at approximately l50F below
the saturation temperature corresponding to the drum pressure. This is
shown below.
PERMISSIBLE ECONOMIZER FLUID- TEMPERA TURES
Cycle Drum Sat. Max. Allowable Econ. Inlet
Press., Press., Temp., Econ. Outlet Temp.,
pSlg psig of Temp., of of
1800 1960 634  619  440
2400 2585 674  659  470
The economizer inlet temperature corresponds to that shown in Figure Ill- 7.
In order not to exceed the allowable economizer outlet temperatures shown
above. the maximunl values of heat input from refuse were found to be 16.6%
and 24.9% for the 1800 and 2400 psig steam cycles, respectively.
[
I
For the separately refuse-fired economizer, the flue
gas exit-tem.perature has been fixed at 5750F, resulting in a steam generator
efficiency of 65%. The flue gas exit-temperature of the main, coal-fired,
stealTI generator was fixed at 3000F, resulting in an efficiency of 87%. The
coal-fired unit \vould be designed for full-load capacity in the event of 10s s of
the refus e-fired units. When firing coal alone, the flue gas exit- temperature
would have to be increased. The efficiency of the coal-fired boiler would thus
be reduced in this -mode of firing.
The water /steam circuit between the refuse and coal-
fired steam generators would be a series connection. This would permit the
use of a single coal-fired steam generator, as shown in Table Ill-5 for the
larger plants, even though several refuse-fired economizers are required.
The water circuits for the refuse-fired units would be connected in parallel
in such multiple arrangements. Because only a single coal-fired boiler
would be required, reheat cycles need be considered only for plants in the
85 to 500 MW range, in accordance with Tables III-l and -2. The various
ten1perature conditions for a plant operated in 2400 psig cycle are indicated
on Figure llI-12, a standard temperature - enthalpy diagram for steam and
water systenl.s.
Ill- 3 0

-------
III
I-t
I-t
I-t
I
v.>
>-
i---'- --

I
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
o
. ..
:/1,',
,)0., .
REFUSE FIRED ECONOMIZER
AIR HEATER
BURNERS
I
i~'
i
I REHEA-reR
i OUTLET


I
I

I i REHEATER
INLET
r
D.~' -
. ~.
COAL FIRED STEAM GENERATOR
FIGURE 111-11. CASE 3 - SEPARATE REFUSE-FIRED ECONOMIZER

-------
TABLE III- 5
CASE 3 - SEPARA TEL Y FIRE:D ECONOMIZER:
CALCULA TED SYSTEM CHARAC TERISTICS
    Nameplate Rating, MW 
   lOO   400 
Refuse Rate, o/c  10. 1 16.6 (max) 9.9 20.4 24. 9 (max)
Steam Pressure, psig 1,800 1,800 2,400 2,400 2,400
Number of Turbines 1 1 1 1 .L
Number of SG's (Coa1/ Refuse) 1/1 1/1 1/2 1/3 1/3
9  1. 75 1. 75 3. 32 3.32 3. 3;~
SG Duty, 10 Btu/hr
Steam Flow, 1031b/hr 1,430 1,430 2,800 2,800 2,800
Refuse Rate/Econ., tpd 565 936 524 735 913
Coal Rate, tpd  1,860 1,752 3,520 3,200 3,060
Refuse Air/Econ., 1031b/hr 226 375 210 353 36:i
Coal Air/SG, 103 1b/hr 1,785 1,680 3, 380 3,070 2,940
Refuse FG/Econ., 1031b/hr 265 440 246 344 428
Coal FG/SG, 1031b/hr 1,945 1,830 369 335 322:
Net SG Efficiency, o/c 84.5 83.4 85.0 82.8 81. 4:
Net Plant Heat Rate,      
Btu/kw-hr  10,360 10,500 9,780 10,020 10, 190
    Refuse Coal  
Excess Air, o/c   50 18  
Air Ratio, 1b Air /lb Fuel  4.80 11. 50  
Flue Gas Ratio, 1b FG (wet}/lb Fuel 5.63 12.55  
III - 3 2

-------
H
~
I
W
W
12UO[
'1100
I

,coc~
~1
~ I
ex: i

~ OOl
~ 700~
00+

I
I
!
SCovI-

I
,+ ~

400
LEGEN 0
I. ECONOMIZER INLET
2. ECONOMI'ZER OUTLET (JO%REFUSE)
3. II II (20%" )
4 . II "(24.9% II , )
5. SATURATION TEMPERATUR.E
6. SUPERHEAT OUTLET
7. REHEAT INLET
8. REHEAT OUTLET
I
500
I
700
I
1000
I
1100
I
600
I
800
I
900
I
1200
I
1300
I
1400
I
1500
I
1600
ENTHALPY, BTU/LB
FIGURE 111-12. CASE 3 - TEMPERATURE CONDITIONS FOR THE 2400 psig/1000oF/1000oF CYCLE

-------
Case 4 - Saturated Steam Unit with Separate Super-
heater. This is a systen1 in which refuse is fired to generate saturated
stean1. which is then delivered to a separate superheater which is fossil-
fuel fired. This type of systelll has been considered for an "over the fencc!!
installation in a West Coast City. The application of a separately fired
superheater has been commercially applied at an East Coast utility, where
the saturated steam was generated in a nuclear reactor.
The summary performance is shown in Table III-6,
the schematic on Figure III-l3, and the temperature conditions for the 2400
psig cycle on Figure Ill-14. In this arrangement, the refuse heat-input can-
not be varied as was possible for Cases 1 and 2.
The flue gas exit-temperature for refuse firing was
fixed at 4500F, corresponding to a steam generator efficiency of 69. 0%. In
the coal-fired superheater, the lowest steam temperature is approximately
6700F. Therefore, the air heater must perform the additional duty of cooling
the flue gases. The temperature of the flue gas exiting from the coal-fired
boiler has also been fixed at 4500F, corresponding to a steam generator
efficiency for this unit of 84. 0%.
The water-steam circuit is a series connection, thus
perm.itting a single coal-fired superheater with integral reheater. Therefore,
for 85 MW to 500 MW only reheat cycles need be considered.
In the Case 4 configuration, where only one refuse
rate can be used for each of the steam conditions, the rate for the 1800 psig
cycle (63.5%) exceeds the 60% limit previously set. The excess, however,
is not considered to be sufficient to warrant the rejection of the system.
For this case, coal burners with an input capacity
equivalent to 10% of full load would be installed in the refuse units for trim..
ming control. It has been assumed that the coal could be taken from the
mills used to feed the separately fired superheater. It should also be noted
that, because multiple refuse units would be required, the variations in
steam flow could be minimized by manifolding the several steam inputs
before they enter the coal fired superheater.
Case 5 - Partial Superheat Unit with Separate
Superheater. This system is similar to Case 4 except that partial super-
heating is provided in the refuse furnace. A proprotionately greater amount
of heat input can thus be derived from refuse, although the amount is again
limited by the cycle. The steam is partially superheated in the refuse-fired
steam generator to a temperature of 7500F as shown Qn Figure III-15. In
the interest of avoiding fly-ash corrosion problems, this steam temperature
has been observed as the upper limit for all separately-fired refuse systems.
This could not of course be applied in those cases wherein both coal and
refuse are fired in the same boiler.
III- 34

-------
.' .
. .- '. ".,
TABLE IlI-6

CASE 4 - SEPARATE FOSSIL FUEL SUPERHEATER (SATURATED
STEAM INPUT): CALCULATED SYSTEM CHARACTERISTICS
Nameplate Rating, MW
200 ill.
Refuse Rate, 0/0
Steam 'Pressure, psig
Number of Turbines
63. 5
1800
1
1/5
1. 75
1430
806
857
323
822
388
898
74.5
11, 745
58.3
2400
1
1/7
3.32
2800
878
1840
352
1764
Number of SG's (Coal/Refuse)
SG Duty, 1 09 Btu/hr '
Steam Flow, 103 lb/hr
Refuse Rate/SG, tpd
Coal Rate, tpd
Refuse Air/SG, 103 lb/hr
Coal Air, l03 lb /hr
Refuse FG/SG, 103 lb/hr
Coal FG, 103 lb/hr
Net SG Efficiency, 0/0
Net Plant Heat Rate, Btu/kw-hr
412

1928

75.3
11, 000
Refus e
Coal
Exces s Air, 0/0
Air Ratio, 1b Air /lb Fuel
Flue Gas Ratio, lb FG {wet)/lb Fuel
50
4.80
5.63
18

11.50
12.55
III-35

-------
CONY. SUPHTR. copms.
  RfHEAT"£R  h
  OUTLET  I illl
    111~ I
H  REHEATER I  
H II INLET  
H  
I   
W  I 
CJ'.  i~ BURNERS
REFUSE FIRED STEAM GENERATOR
SEPARATE FOSSIL FUEL FI RED SUPERHEATER
(SATURATED STEAM)
FIGURE 111-13. CAS:, - SA' . ~ATED STEAM UNIT WITH SEPARATE SUPERHEA' : ~

-------
. 1200
  I
ll.. :r 
 0 
 .. 
 W 
 IX: 
 ~ 
 I-  
 ct  
 0:::  
...... IJJ  
~ Q. :~ 
I ~ 
W 
-J LLJ 
I- 
  , 
  ~+ 
  I 
  ~+ I
   400
LEGEND
I. ECONOMIZER INLET
2. SATURATED LIQUID
3: SATURATED VAPOR (SUPERHEAT INLET)
~ SUPERHEAT OUTLET
5. REHEAT INLET
6. REHEAT OUTLET
I
500
/ : 
I 
600 700 800
1200
/
I
1100
I
900
I
1300
I
1400
I
1500
I .
1600
I
1000
ENTHALPY,BTU/LB
,

FIGURE 111-14. CASE 4 - TEMPERATURE CONDITIONS FOR THE 2400 psig!1000oF/1000oF CYCLE

-------
1200
1100
1000
900
H
I::J
I'
W
00
LL.
O~
IJJ
a::



~ ; 800'~
« .
a::
IJJ
Q.,
:E 700
IJJ
t- . I'

GOOL
500
400
~g~g:o
41) "» Ot) ~ tt) 0
II') \I' ,., ,.., ... ~
LEGEND
I. ECONOMIZER INLET
2. SATURATED LIQUID
3. SATURATED VAPOR
4.STEAM AT EXITOF REFUSE-FIRED S.G.
5.SUPERHEAT OUTLET
6. REHEAT INLET
7 .REH EAT OUTLET
500
GOO
700
900
1500
1100
1000
1300
14ao
1600
1200
ENTHALPY, BTU/LB
FIGURE 111-15. CASE 5 . TEMPERATURE CONDITIONS FOR 2400 psig/1000oF/1000oF CYCLE

-------
Under Case 5 conditions, the heat input from refuse
would represent about 75% of the total. This would probably lead to .steam
flow variations that would be too great to' be considered for large power-
generating plants.
As in all the cases thus far discussed, it has been
assumed for Case 5 tl1at refuse would be burned on a reciprocating grate with
refuse - conditioning provided only for bulky refuse. This type of burning is
actually the cause of the thermal vax:iations that result in steam-pressure
fluctuations. In order, therefore, to consider this case, a system modifi-
cation was introduced wherein more homogenous burning would obtain. This
was done by substituting a spreader stoker for the agitating grate.
This firing configuration was initially planned for and
has been incorporated into Case 7, which is definitively labelled a "spreader
stoker. II The boiler arrangements for Cases 5 and 7 are, otherwise, quite
different. One can, however, refer to Case 7 for further details on the firing
principle that is common to both.
The Case 5 schematic, as revised, is shown in Figure
IIl-16. A summary of the performance calculated for this case is presented
in Table Ill- 7.
Case 6 - Suspension Furnace. This is a conventional
(pulverized-coal-firing type) furnace to which 2-in. nominal sized refuse is
instead delivered and fired. It has been suggested (Ref. 41) that most of the
refuse will burn while falling to the bottom of the furnace. The remaining,
unburned refuse would be burned on a dumpgrate. This type of arrangement
is being constructed by an East Coast industrial plant for firing industrial
refuse.
The overall performance can be considered to be the
same as for Cases I and 2 (Table Ill-4), although, in suspension firing, one
might realize better burn-out and can expect some fuel dehydration due to
size-reduction operations. Because data is not available at this time, the
possible benefits of such effects could not be reflected in the performance
calculations. The schematic is shown in Figure Ill-I 7.
This suspension-firing case is essentially the same
as Case 2, combined furnace, in terms of boiler configuration. For Case 6,
however, the furnace would have to be taller in order to optimize the degree
of burning achieved while the refuse is suspended. This case is considered
in further detail in the section on Retrofit plants (Modification No.5).
Case 7 - Spreader Stoker. This is a furnace to which
4-in. nominally sized refuse is delivered and fired. The mechanism of stoking
is based on a pneumatic system which blows the refuse into the furnace where
Ill- 39

-------
H
H
H
I
~
o
SPREADER
STOKER
II~!II
lilll
III
I~!~III
~III~II
OVER FIRE
AIR
REFUSE FIRED STEAM GENERATOR
(SPREADER STOKER)
REHEATER
OUTLET
ECON:)MIZER
CONNS.
REHEATEFI
INLET
CONY. SUPHTR. CONNS.
BURNERS
SEPARATE FOSSIL FUEL FIRED SUPERHEATER
(SUPERHEATED STEAM)
FIGURE 111-16; CASE 5 - PARTIAL SUPERHEAT UNIT WITH SEPARATE SUPERHEATER

-------
TABLE I11-7
CASE 5 (REVISED) - SEPARATE FOSSIL FUEL-SUPERHEATER
(PARTIALLY SUPERHEATED STEAM INPUT):
CALCULATED SYSTEM CHARACTERISTICS
Flue Gas Ratio, Ib FG (wet)/lb Fuel
Nameplate Rating, MW
200 400
75.5 71. 3
1800 2400
1 1
1/5 1/9
1. 75 3.32
1430 2800
983 915
587 1302
393 366
563 1248
461 428
600 1362
72.6 73.4
. 
12,035 11,330
Refus e Coal
50 18
4.80 11.50
5.63 12.55
Refuse Rate, %
Steam Pressure, psig
Number of Turbines
Number of SG1 s (Coal/Refuse)
SG Duty, "109 Btu/hr
Steam Flow, 103 Ib/hr
Refuse Rate/SG, tpd
Coal Rate, tpd
Refuse Air /SG, 103 lb /hr
Coal Air, 103 lb /hr "
Refuse FG/SG, 1031b/hr
Coal FG, 103 Ib /hr
Net SG Efficiency, %
Net Plant Heat Rate, Btu/kw-hr
Excess Air, %
Air Ratio, lb Air /lb Fuel
III- 41

-------
REFUSE a COAL
COMPARTMENTS
PENDANT
SUPHTR.
"..----
/
I
I
I --
I /
I I
I I
---- --.---
I 1 0
_J_---I-- 0
lr -,-1 .
- - -.,------
I I
I I
I I
iP>\ --I r- -
"d'- -I r-
I I
I I
~.:] c:
I I
I I
_J ~-
CLJ 1--
I I
1 I
I I
I I
II : I 'I



- - -Ej--- ==.:j
r
UMP GRATE
. .~. I
,1'410
A" . ./~
FIGURE 111-17. CASE 6 - SUSPENSION FURNACE
III- 42
REHEATER
aJTLET
REHEATER
INLET
I

I

I



! ECONOMIZER
INLET

-------
it burns as it falls through the flame and spreads over the length of a tra-
velling grate. Such an arrangement will be utilized at a Canadian municipal
plant, which will fire refuse exclusively.
The schematic for this system is shown in Figure
Ill-lB. The overall performance is considered to be the same as for Cases
1 and 2 (Table I11-4), although, as with Case 6, a certain degree of per-
formance enhancement may be available.
Aside from the firing mode, the boiler configuration
of the spreader stoker may be regarded as the same as that of Case 2. For
Case 7, the coal would not be burned on the grate, but in pulverized form,
in suspension. In general, pulverized coal firing is more efficient and yields
more uniform heat liberation. Since the steam generated would be for power
production, the additional pulverizers and coal burners would be justified.
. Case 8 - Slagging Furnace. In this furnace, shredded
refuse is delivered and fired together with pulverized coal. The firing condi-
tions are such that a pool of slag forms at the bottom of the furnace. The'
schematic is shown in Figure III-19. The overall performance can be con-
sidered to be the same as for Cases land 2 (Table III-4), and the configuration,
aside from method firing, the same as that of Case 2. The slag-tap furnace
will require prior conditioning of refuse to a nominal two-inch top size. This
scheme offers the desirable feature of producing a compact residue of high
density, which is easily disposed of or can be marketed.
Although slagging furnaces are still used in Europe,
they have not found favor in domestic use. The availability (plant factor) of
slagging furnaces is lower than dry-bottom units. Maintenance is als() a
serious problem. In operation, the turn-down is limited due to the necessity
of maintaining a molten slag. Utilization of the slagging furnace in combina-
tion with refuse has been studied in Europe and to a limited extent in a domestic
pilot plant. The Munich Electricity Works seriously considered this type of
unit but abandoned it as being technically risky. Their experience and recom-
mendations should not be overlooked.
The most important fuel property to be considered
relative to this system is the melting temperature of the mixed refuse/coal
ash. As pointed out in Section II, B, considerable doubt exists as to whether
slagging conditions could be promoted on a continuing basis. in view of the
probable high variability in the residue properties. On this basis, this case
must be regarded as having some serious technical drawbacks. It is further
felt that the refuse rate should be limited to a maximum of 40% of the heat
input. On this provisional basis. this case has been included in the design
data and capital cost analysis.
I11-43

-------
~

I
I PENDANT
SUPHTR.
SPREADER
STOKER
- ..
. .. .
'''.
FIGURE 111-18. CASE 7 - SPREADER STOKER
III - 44
EHEATER
OUTLET
REHEATER
INLET
ECONOMIZER
INLET

-------
\' PENDANT
SUPHTR.
PLATEN
SUPHTR.
COAL
I I
! I
i I

- -1. - -.- .1 -.-.-

OOOOC
- - AiR poFfrs -
ODD
a
REHEATER
OllTLET
./

SLAG TAP
REfUSE
ECONOMIZER
INLET
RE FUSE
.-~~,o,.o'.
" .
.. ;,:,

. .
FIGURE 1'11-19. CASE 8 - SlAGGING FURNACE
III- 45

-------
Case 9 - Arch Furnace. This is a system in which.
combined suspension firing is practiced, but without the use of a dump grate.
This type of furnace is also discussed in Appendix B (Section III, A, 4, a).
The expected advantage of employing a system of this type would be a high
stoking-rate. In the previous cases discussed, the refuse rate was limited
to 1000 tpd, which is about the maximum capacity for any type of grate
available at the present time or in the imrnediate future. In fact, only one
manufacturer has a grate of this size in operation. This is a backward re-
ciprocating grate in use at the Munich plants (North - Unit 2; South - Unit 5).
Because of the absence of the restriction on refuse-
rate imposed by the grate, calculations for the present case were based on
an upper fuel rate of 2000 tpd rather than 1000 tpd as was observed for the
preceding eight cases. This necessitated the preparation of a new map of
fuel requirements (Figure III-20), as were presented earlier in Figures
III-4 and -5 for 1000 tpd systems. Each data point in Figure III-20 repres-
ents a value calculated by the procedure illustrated in Table III-3.
As with the preceding several cases, this system
is an adaptation of a known burning mechanism but applied to a new fuel.
In this case, the burning mechanism has been successfully used for the
firing of solid fuels having a low content of volatiles, such as anthracite
coal. The basic effect, illustrated in the schematic (Figure llI-21), pro-
motes a staged combustion of the fuel and favors relatively long residence
times. Refuse, previously conditioned to a nominal 2-in. top size, is de-
livered to the burner by pneumatic m.eans. Most of the air conveying the
refuse to the furnace would be vented off just behind the fuel nozzle and be
bypassed into the furnace through a separate nozzle. The refuse would thus
tend to enter the furnace at low velocities. This type of burner, as used for
anthracite coal, is illustrated in Figure 1II-22. Additional air is admitted to
the furnace in a downward direction through the arch alongside the refuse
burner and also through the side walls. As a further step to insure turbu-
lence and prolonged residence time, air is also admitted through the hopper.
Coal would also be fired in a downward direction from standard pulverized
coal burners which would be situated between refuse burner cells.
The summary performance is shown in Table III-8.
Because of the conditioning of the fuel, it is felt that the heat input from
refuse could be as high as 80%.
Case 10 - Arch Furnace with Separate Superheater.
This is a system which, in es sence, is a combination of the original concept
for Case 5 and that just discussed for Case 9. Refuse is burned in a furnace
to generate superheated steam at 750oF, which then is further superheated
to 10000F in a separate coal-fired furnace, where reheating is also accom-
plished. This case thus exploits the advantages of the arch furnace (greater
refuse throughput per unit) along with the advantages of operating with the
separately fired superheater. As with Case 9, the success of this confi-
guration hinges on whether a sufficient burnout of the refuse will occur.
In order to insure constant ignition of the shredded refuse, coal would also
be burned in the refuse-fired boiler, in an amount equal to 20% of the heat
input.
III-46

-------
,-----
"0 
Q. 
- 2.5
If)O
w 
.... 
<[ 
a:: 
-I 
<[ 2.0
o 
u 
4.5
4.0
3.5
3.0
1.5  
 ~ 
 (' 
 ~
1.0  ;...\
  13
  ~
  ~
  Q
0.5  ~\
FIGURE 111-20. ARCH FURNACE FUEL INPUT
III-. 47

-------
~~CJ
01
I
I


~ p~~K)A~r
s)lD~~r~.
a
0'
o
o
R
D
o
o
o
o
o
o
. REHEAT r.n
INLET
WINDBOX
. ECONOMIZER
INLET
f~GU~1E m-2~. CASIE g - A~C~ fU~NACIE
III- 48

-------
rAir Coal Mixture
Auxiliary
Air Port
Vent Control Dampers
Riffle Distri butor
Straightening Vane
Burner Nozzle
Tertiary Air Supply
Tertiary Air
Regulating
Dampers
FIGURE 111-22. ANTHRACITE COAL BURNER USED IN ARCH FURNACES
III - 49

-------
TABLE III- 8
CASE 9 - ARCH FURNACE:
CALCULATED SYSTEM CHARACTERISTICS
          Nameplate Rating, MW  
         200     400 
 Refuse Rate, o/c  0 20 40 60 80 0 20 40 60
 Steam Pres sure, psig 1800 1800 1800 1250 1250 2400 1800 1800 1250
 Number of Turbines 1 1 1 "3 3 1 2 2 5
 Number of SG  1 1 2 3 4 1 2 3 5
 SG Duty, 9  1. 75 1. 75 0.88 0.68 0.51 3.32 1. 75 1. 1 7 0.79
 10 Btu/hr
 Steam F'low, 1031b/hr 1429 1429 715 610 470 2780 1429 950 715
 Refuse Rate, tpd  0 1170 1210 1460 1520 0 1165 1610 1700
 Coal Rate, tpd  2080 1727 674 360 141 3960 1728 897 420
H Refuse Air, 103lb/hr 0 467 483 583 607 0 465 643 679
H
H   103 lb/hr          
I Coal Air, 1992 1650 645 345 135 . 3790 1658 859 403
U1
o   103 lb/hr          
 Total Air, 1992 2117 1128 928 742 3790 2123 1402 1082
 Refuse FG, 103lb/hr 0 548 566 685 712 0 545 744 797
 Coal FG, 103lb/hr 2180 1805 705 377 148 4140 1810 937 440
 Total FG, 103lb/hr 2180 2353 1271 1062 860 4140 2355 1681 1267
 Net SG Efficiency, o/c 84.0 81. 0 78.0 75.0 72.0 84.0 8LO 78.0 75.0
 Net Plant Heat Rate,          
 Btu/kw -hr 10,400 10,800 11,220 13,540 14,100 9,880 10,800 11,200 13, 130
         Refuse  Coal    
 Exces s Air, o/c     50  18    
 Air Ratio, 1b Air /lb Fuel   4.80  11. 50    
 Flue Gas Ratio, lb FG (wet)/lb Fuel  5.63  12.55    
 NOTE: )ata shown are sing e SG's.        

-------
The basic steam generator configuration is shown in
Figure 1I1-23 and the summary performance is shown in Table 1I1-9.
5.
Design Data for Retrofit Systems
a.
Background
In the previous analysis of (new construction) candi-
dates, the steam conditions and unit sizes were established on the basis of
what had been representative of most of the units ordered over the last few
years. In considering units for modification for combined firing, the time
span necessarily included units placed in service during the last twenty years,
an era of rapid increase. in steam conditions and unit size. Thus, certain
deviations from the criteria established for new construction candidates
became unavoidable.
Steam generator and turbine efficiency had been
fairly well established at their respective practical maximum levels during
this era. Plant heat rates were lower, however, but basically because the
auxiliary equipment used earlier was of considerable less efficiency than
comparable modern equipment.
In considering candidates for modification, natural
cir culation units with balanced draft systems were chosen, as was done in
the case of new units. This restriction did not result in the exclusion of a
great number of retrofit candidates. Emphasis was placed on seleCting units
that would be representative of the many units placed in service over the last
twenty years. Where possible, the criteria observed in the performance ana-
lysis of new units was also applied in considering the modification of existing
units. This implied the imposition of an upper limit of unit sizes of 500 MW,
the use of coal as the fossil fuel or a unit designed for coal-firing, the re-
striction of unit refuse rate to 1000 tpd or less (when a grate is used), and.
the maintenance of the flue gas exit-temperature at 450oF.
In considering plant modifications, it has been
assumed that space would be available for the refuse handling equipment.
Consideration was given to open areas and structural steel in close proxi-
inity to the steam generator. In general, updated building drawings were
not a vaila ble.
In applying the results of the analysis of (new con-
struction) candidate steam generators, it became obvious that some of the
new plant designs would not be applicable for the retrofit of existing plants.
Immediately excluded were the separately fired
fos sil fuel superheater designs (Cases 4, 5 and 10). The arch furnaces
(Cases 9 and 10) were eliminated on the basis that the number of existing
Ill- 5 1

-------
1-1
I:::
I
U1
N
I
I
I
I
I
I
: OU~S II
I
I ~
I
I
I
III III1
WINDBOX
REHEAmt
OUTLET
REHEATER
INLET
C:ONV. SUPHTR. CONNS.
I
. I
I
I
I
I
SEPARATE FOSSIL FUEL FIRED SUPERHEATER
(PARTIAL SUPERHEATED STEAM)

FIGURE 111-23. CASE 10 - ARCH FURNACE WITH SEPARATE SUPERHEATER
ARCH FURNACE
~ERS

-------
TABLE 1II-9
CASE 10 - SEPARATE FOSSIL FUEL SUPERHEATER:
CALCULATED SYSTEM CHARACTERISTICS
Nameplate Rating, MW
200 400
Refuse Rate, 0/0
Steam Pressure, psig
59.8
1800
1
1/2
1. 75
143'0
1880
591/349
742
567/334
567/1076
882
619/365
619/1247
75. 1
11,660
Number of Turbines
1
Number of SG (A/B)

SGDuty, 109 Btu/hr
3
Steam Flow, 10 1b/hr
, '

Refuse Rate/B, tpd
Coal Rate (A/B), tpd
Refuse Air/B, 1031b/hr
Coal Air (A/B), 1031b/hr
Total Air (A/B), 103 Ib /hr
3
Refuse FG/B, 10 Ib/hr
3
Coal FG {A/B}, 10 lb/hr
Total FG (A/B), 103 lb/hr
Net SG Efficiency, 0/0
Net Plant Heat Rate, Btu/kw-hr
Refus e
Exces s Air, 0/0
Air Ratio, lb Air/lb Fuel
Flue Gas Ratio, lb FG {wet)/lb Fuel
50
4.80
5.63
Note 1:
A = Sepa-rate coal-fired superheater unit.
B = Refuse and coal (80/20) arch furnace unit.
1II-53
56.4
1800
1
1/4
3.32
2800
1675
1299/620
670
1243/597
1243/1267
785
1359/648
1359/1433
75.5
10,998
Coal
18
11. 50
12.55

-------
units of this type is too small to warrant their inclusion. Also eliminated,
for the same reason, were slagging furnaces (Case 8), many of which are
pres surized. The separately fired economizer (Case 3) was initially con-
sidered a possible approach for retrofitting existing units. However, in
designing the main fossil fuel steam generator, it was found that excluding
the economizer from the conventional type steam generator with the require-
ment of full load operation with or without external feedwater heating would
require a unique configuration for the main steam generator.
In summary, the designs considered for the modi-
fication of existing units are the separate or combined furnaces (Cases 1
and 2) and the firing variations thereof which involve suspension firing
(Case 6) and spreader stoker firing (Case 7).
b.
Selection of Existing Units
Four units were selected as being illustrative of
existing systems. The characteristics of these four units, listed according
to the appropriate modification plan (described later) for which each is con-
sidered, are given below.
CHARACTERISTICS OF EXISTING STEAM GENERA TORS
SELECTED FOR MODIFICATION
Modifi- Name pIa te   
cation Ra ting Steam Conditions Stea~ Flow Service
No. MW psig /oF /oF 10 lb/hr Date
1 60 900/900/- 600 1949
2 150 2035/1050/1000 1, 100 1959
3 & 4 44 1350/950/ - 445 1957
5 300 2200/1010/1010 2,310 1961
Cross sectional drawings of these units are shown
in Figures III- 24 through -27. Full-load performance data for these units
along with the design fuel are shown in Table III-l O. Although the design
fuel for Modification No.3 is oil, the unit was also designed for possible
future coal firing. Coal-burners were installed but not the coal-mills,
bunkers, and other ancillary items.
In the new construction units considered in the pre-
vious analysis, the maximum permissible steam temperature developed
from flue gases from separate or isolated refuse combustion was set at
750oF. Refuse is a fouling type fuel and this tendency increases with tube
metal temperature. For the nonreheat units selected for modification,
III- 54

-------
SUP£RHEAT[R
n

INTERIO[OIATE HEADER
, I,
I , I
, , ,
I , ,
I I ,
. I r~-------
, I '
, , '
, I, I
I J I I
-i------~--' :
7~.Ar 1'''''''''''''''''''''

:_--~~..l..:.....
AIR HEATER
o
,
-..
!?
~ ~,
~
FIGURE 111-24. EXISTING STEAM GENERATOR FOR MODIFICATION NO~ 1
III-55

-------
\0
,
'"
REHEATER OUTLET HEADER
SUPER HE ATER OUTlE T HE ADER
------------- ----,
I
ECONOMIZER OUTLET HEADER
REHEAT CONTROL DAMPERS
REHEATER INLET HEADER
EATER
AllS
32'-0"
65'-0"
FIGURE 111-25. EXISTING STEAM GENERATOR FOR MODIFICATION NO.2
III - 5 6

-------
I
, :
! I
INLET
\/.~/>~.:~
fTI>'CAI. PL"" MeT t-u ''''''.''(1 ..LL'
,"" Tun: (011''''\1<''081
26"0.
/ / /
22'-6r
FIGURE 111-26. EXISTING STEAM GENERATOR FOR MODIFICATIONS NO.3 AND 4
Ill- 57

-------
EL.- 590'-0.
\- ECONOMIZER
! OUTLET HEADER

II
! I
SUPERHEATER OUTLET

J
- 580'-0.
NTERMEDI
E HEADERS
- 510'-0.
I
i
I
-560'-0"
- 550'-0"
ION SUPE RHE AT E R
-540'-0"
-530'-0"
-520'-0"
RAD NT SUPERHEATER
RONT WALL)
-510~0.
- 500'-0"
",.,
I
-...
...
-490'-0"
-480~0"
II
-410'-0"
II-
i II
- II
.1
"
, I
I
"
I
NERS
-460'-0"
-450~0.
i
i'l
, I
, I
- I
~
, I
, I
: I
, I
REGENERATIVE
AIR HEATER
-440~O"
4 30'-0.
420'-0"
! i
410'-0"
3'P-ij' -
0'-0.
FIGURE 111-27. EXISTING STEAM GENERATOR FOR MODIFICATION NO.5
Ill- 58

-------
        TA 3. ~ ~ III-I 0   
     SUMMARY OF PERFORMANCE DATA   
     FOR SELEC TED STEAM GENERA TORS   
 Applicable Modification No. :    1 . 2 3 and 4 5 
    3    600  445 2,310 
 Steam output, IO lb/hr    1, 100 
 Superheater outlet temp., of    950 1,01Q 
 Pressure, psig:         
 Boiler drum     934 2,160 1,455 2,352 
 Superheater outlet    900 2,035 1,360 2,200 
 Reheat outlet temp., of     1,003  1,010 
 Reheat outlet press., psig     419  566 
 Feedwater temp., of     496 430 482 
 Exit air temp. at Air Heater, of  555 638 456 613 
 Exit flue gas temp. at Air Heater, of 301 290 340 308 
 Excess air, o/c:         
 At boiler exit    23  10 20 
 At a.ir heater exit     20   
...... Flue gas (wet), 103 lb/hr:        
I::: Economizer exit    746    
I Air heater exit     1,510 514 2,720 
\}1     
-.0 Fuel    Bituminous Coal Bituminous Coal Fuel Oil Bituminous Coal
 Fuel rate, 103 Ib/hr    54.5 123. 5 29.5 267 
 Ultimate Analysis, * wt-o/c        
 Ash       8.55 13.00  10.5 
 Sulfur       1. 90 3.49 2.5 3. 5 
 Hydroge!l      4.26 4.32 10. 18 4. 7 
 Carbon       75.00 63.93 86.69 61. 4 
 Moisture      5.00 7.00  11. 0 
 Nitrogen      1. 13 1. 30 0.14 1.2 
 Oxygen       4. 10 6.96 0.49 7. 7 
 Fuel value, as fired (HHV), Btu/1b  13, 300 11 , 600 18,400 10,280 
 Plant heat rate, Btu/kw-hr    12,090 9,560 12,340' 9,140 
>:
-------
however, the steam temperatures are 900 and 9500F. As will be shown later,
these units were modified so that the refuse fuel bed would be positioned below
the fossil fuel flame envelope. The flue gas and fly ash from the former, in
passing through the latter, would thus undergo more complete combustion.
This effect was demonstrated at Munich North, Unit No.2. Also, the super-
heater spacings in these units are wider than is found in present-day units.
This, too, should serve to reduce tube wastage to an acceptable level in
these units.
The reheat units have superheat temperatures slightly
above 10000F and are more susceptible to fouling than the nonreheat units.
However, in modifying these units steps were again taken to minimize fouling.
Many units built over the last twenty years have radiant
superheaters that are located in the furnace where tube metal temperatures can
approach 10000F. The steam generators shown in Figures III-24 and -25 are
illustrative of this type of construction. Therefore, these units should not be
fired with refuse on a grate, if any radiant superheater surface is exposed.
As shown later, these units have been modified to minimize potential tube
wastage.
c.
Modification Requirements
The procedures involved in planning the modification
of each of the units were somewhat similar. Where grates are to be installed,
it is necessary to cut and remove the existing furnace hopper tubes. New sec-
tions of tubing that are of the proper shape for a refuse combustion area must
be welded to the existing furnace tubes. New lower water-wall headers are
required and the new sections of tubing must be insulated. Stokers and char-
ging chutes must be installed and acces s provided for service.
A combustion-air system for firing refuse is required.
This system includes fans, ducts, and dampers. It was initially thought that
the existing fans might also serve to supply the refuse combustion-air, but it
was found that this would not be practical. The amount of combustion air that
could be diverted from the coal firing system because of the reduced coal-rate
would not be quite enough to provide even the stoichiometric refuse combustion
air; furthermore, many operational problems would be caused by the dual use
of the fans. Static pres sure heads on the existing fans may not be correct for
the refuse air system. Also it would be difficult to build an air distribution
system that would accurately distribute air between the two dis similar firing
systems. Therefore, in most of the modifications, additional fans are speci-
fied that will provide the full refuse air requirement.
It is anticipated that additional soot blowers will also
be required in the convection passes, and that some air heater elements will
have to be removed from existing regenerative air heaters. In some cases
the secondary support steel has had to be changed.
III- 60

-------
The amount of refuse that each modified unit can
consume is primarily determined by the size grate that can be accommo-
dated. This presumes that in the furnace enclosure, usually of a rectan-
gular plan area where two opposite walls bend in to form a hopper, only
two of the walls would need to be modified. It soon became apparent,
however, that the appropriate approach was to modify the hopper walls
and simply extend the length of the other walls. This essentially implies
that the maximum refuse-rate would be dependent upon the existing fur-
nace width. All of the units shown with reciprocating grates can accom-
modate other types of agitating grates with few if any design changes. The
refuse rates were calculated on the basis of sixty pounds of refuse per square
foot of net grate' area per hour. The heat input from refuse is again based
on a higher heating value of 4,460 Btu/lb and 69% stearn generator efficiency
, when firing refuse exclusively.
d.
Candidate Retrofit Systems
Modification No.!. The unit modified (Figure IIl-24)
is typical of pulverized coal or oil steam generators built in the late 19401s.
One of two duplicate units at an eastern. power plant, it was designed for an
output of 600,000 lb/hr superheated steam at 900 psig and 900oF. In the
1940. s the present practices of using reheaters and monowall furnace con-
struction were not yet developed. Units built at this time are now largely
on standby duty and could probably be made available for modification.
In order to adapt this unit for refuse firing, the lower
area of the furnace was modified to assume the shape required for refuse
combustion (Figure 1lI-28). This requires bent tube sections for the rear
and front water walls, and extension of the side water walls. A new section
of water wall is placed at the first grate section. Two underfire air fans
and the underfire air ducts are provided. Each plenum chamber air supply
is separately controlled by dampers. Ducts and nozzles are provided for
the overfire air system, which is also supplied by the two fans.
This modified unit will be capable of firing 951 tpd
of refuse, which is approximately 43% of the required heat input. There
should be no problem in running at full load using coal as the only fuel.
Modification No.2. This modification is an example
of adapting a stearn generator to combined firing utilizing separate furnaces
for refuse and fossil fuels. It corresponds to Case 1 in the new construction
designs. The unit chosen for modification (Figure 1lI-25) generates stearn at
2035 psig and 10530F. The main steam generation rate is 1,100,000 lb/hr
using a reheat cycle. This stearn generator, one of seven identical units,
was installed in a Southern State in the late 1950.s. The modification is shown
in Figure I11-29. Two separate refuse boilers generating saturated stearn
would be required.
I11-61
"

-------
"""1-

t, ~
~
W
1"1
'I
,,'
..
I
COLUIOI 
-------
fi (
'\
!
S
I
.0"-
W
-,
'9'
I.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
. 12.
13.
14.
IS.
16.
17.
18.
19.
ACCESS DOOR
SECTIONAL ELEVATION
'A-A'
Ii)
FIGURE 111-29. BOILER MODIFICATION NO.2
'Ii
_.J
LEGEND
,RECIPROCATING GRATE
UNDERFIRE AIR FANS
BOILER SUPPORT STEEL
MONOWALL TUBING
SLAG SCREEN MADE FROM EXISTING FURNACE WALL
GAS IGNITORS
PLATFORMS
OVERFIRE AIR DUCTS, DAMPERS AND NOZZLES
OVERFIRE AIR FANS
UNDERFIRE AIR FEED DUCTS
UNDERFIRE AIR PLENUM CHAMBER
UNDER FIRE AIR FEED DUCTS
DRUMS
SOOT BLOWERS
DOWNCOMER PIPING
STEAM OUTLET PIPING
INSULATION FOR DOWNCOMERS
FURNACE WALL S60T BlOWERS
RADIANT SUPERHEATER INLET HEADER

-------
This unit was chosen because of the feature of rear-
wall firing. Although this is the desired configuration in coupling to another
furnace, it does seriously narrow the number of existing steam generators
that can be retrofitted.
The addition of two refuse furnaces would not be
necessary in every case. It was done this way in the present situation be-
cause the existing unit was too wide to accept a grate of the same width,
and because there was a boiler support column at the center of the existing
front wall. If a unit were modified by adding a single refuse furnace, how-
ever, the cost savings would not be great.
The resulting modified unit would be capable of in-
cinerating 1, 000 tpd of refuse. This would account for approximately 24%
of the heat input.
. Initially, the separate furnace scheme appeared to
be quite feasible, but as the designing progressed, the problems and dis-
advantages increased drastically. Overcoming each technical problem that
developed necessitated the addition of more hardware. The resulting cost
of this modification proved to be very high, which, coupled with the dubious
technical benefits of this type of unit, led to the conclusion that the scheme
might be impractical for existing steam generators.
Modifications No.3 and 4. The steam generator used
for Modifications No.3 and 4 (Figure 1lI-26) was built for a New England power
company in the mid-1950's as one of two identical units. It has no reheat cycle
and there is no radiant superheat surface in the furnace. The furnace walls are
of monowall construction. This unit was designed for 445,000 lb/hr steam at
1,350 psig and 950oF. It was originally built for oil firing with provision for
future coal firing. . .
Modification No.3, as shown in Figure 1lI-30, in-
volves the retrofit of the unit with a spreader stoker refuse-firing system.
The furnace hopper has had to be changed in much the same manner as would
be necessary for installing a reciprocating grate system. The hopper was
replaced by' straight front and rear water walls, and extended side walls.
New lower water wall headers have been provided. Six air-swept spouts
feed shredded refuse, sized to pass through a 4-in. grid, into the furnace.
A travelling grate has been specified to insure the combustion of refuse
that does not completely burn while in suspension.
. This arrangement would be more attractive if the
fos sil fuel burners were situated in the rear wall. The layout of the con-
vection passes, however, did not permit the relocation of the burners as
part of the design modification. A few design compromises were made,
therefore, in order to provide access to the coal burners. On the whole,
however, there were few problems encountered with this modification.
Fans were provided for the air-swept spouts and the overfire air, although
it was assumed that the existing fans could supply the underfire grate-air.
1lI-64

-------
......
I:::
I
0'
U1
LEGEND
1.
2.
3.

4.
S.
6.

7.
8.
9.
10.
11.
12.
"13.
14.
IS.
..:. .'. '., .' ~ .;...
:"1
",
AlA
,DUC,!:-
.
-..
HZ
'3:-
~
-----:.:::::".-=-------" ~~~
...__"=3
FRONT VIEW
FIGURE 111-30. BOILER MODIFICATION NO.3

-------
It is eH tirnated that this rnodified unit will be capable
of con::iurning 846 tpd of refuse. This refuse-rate will supply about 51 % of the
heat input. There is some uncertainty concerning the refuse-rate, because
this type of firing has not been used for refuse in the past. Considering the
related experience with stoker - spreaders in firing other waste fuels, such
as bark and bagas se, the specified rate can probably be regarded as con-
servati ve.
Modification No.4, as shown in Figure III- 31, in-
volves the same existing steam generator (Figure III-26) as was considered
for Modification No.3. In this case, however, the unit has been modified
for grate firing. This modification also requires certain design compromises
in order to provide access to the existing fuel burners. The shape of the
lower furnace has been made such that the clearance between the refuse
charging chute and the burners would be sufficient to remove the burners.
This modified unit should burn 679 tpd of refuse, which would be 42% of the
heat input.
On a superficial basis, this modification would ap-
pear to be identical with Modification No. 1. The construction features of
the boilers are, however, quite different and reflect the retrofit problems
that would be encountered between older or more recent generation units.
For example, Modification No.1 involves a boiler fitted with tangent tube
walls in which the furnace seals are situated in the refractory behind the
tubes. This older boiler also offers a situation wherein the refuse intro-
duction equipment, particularly the chute, can be installed with minimum
difficulty arising from existing structures. The boiler selected for Modi-
fications No.3 and 4, on the other hand, is of monowall construction, which
is more amenable to structural rearrangements, but presents the problem
of limited placement area for refuse-chute installation.
Another aspect favoring the inclusion of Modification
No.4 is that it permits direct cost comparison of reciprocating grate and
stoker-spreader retrofit systems installed in the same boiler.
Modification No.5. The unit selected for the present
modification (Figure III-27) is owned by a mid-west utility company. It was
one of several units considered for possible modification to refuse-firing in
a recently completed study (Ref. 41) dealing with the present subject. This
unit was designed to fire bituminous coal. It generates 2, 310, 000 1b /hr of
superheated steam at 2,200 psig and 1010oF. This unit also has a reheat
cycle that supplies 1,790,000 lb/hr of steam at 568 psig. Figures III-32
and -33 show the modifications required for refuse firing.
Along with the width of the unit, the furnace is essen-
tially divided into three chambers by the two full division walls. Refuse will
be injected into each of these sections through refuse nozzles placed above
the top burner row. There will be two nozzles per section, one on the front
wall and one on the rear wall.
III-66

-------
ELO~Oa
\:'::::. ,:.,::: .:.,:'~ -:------~~~===.---~.:. ~--~-_==':":'-===~':;:.d
A~
~:Jii ...., ",'. i"owno Ii:

. Li'H REARWALLS(COLD) I
~.-1 (6s) 3 00 X 0.200 IiIW
. ': I SIDE WATER WALL TUBES '1",'1:
'--01' I
" --~ii i :::/(69U.ODREAR WATER
lk=-'=l:: . [XIST.BOILER WALL TUBES
--. ~EWEQiJlP:'
i
! ......
. ~
I
0"-
--J
~~~~:~..
-01
.l-2~-

~~:.::~:_-_._---..--
SECTIONAL ELEVATION
ASH
HOPPER
FIGURE 111-31. BOilER MODIFICATION NO.4
I
tFAN DISCH

b~.:.:=J
SECTION 'A-A'
LEGEND

INSULATION
TUBING
REAR AND FRONT WATER WAlL HEADERS
SIDE WATER WAlL HEADERS .
OVERFIRE AIR SYSTEM
RECIPROCATING GRATE
UNDERFIRE FORCED DRAFT FAN
UNDERFIRE AIRD UCTS
REFUSE CHARGING CHUTE
PLATFORM
1.

2.
3.
4.
s.

6.
7.
8.
9.
10.

-------
~ -~:__J7
)
/
/
// ;-
J/'
,.
;-:
H
H
H,
I
0"
00
1":..--'"
: y:
V--_J
I I
I I
I I

::~=-?l-: ,
.--...~_L-
{ I I
. I ~
; AIR HEATER. ,
G-1---:i
L.J_uf--1 .

"-- HOT AIR
:S~0"X3~d' DUCT
61/2"
n~;; 1/4';=11/8"
I GRATE BARS W/O STEAM
JETS (4,759 REO'D)
F~~ -
~-..
...
1~6"X3'-0"DUCT
l
PARTIAL PLAN/PROP, DUCT ARRANGEMENT
.J
il~
r
~.
,..--1- -e-
---!-+--- ~

,. I
I
It.UNIT-SYM ABT 'i.
------ .(---+--1---1
--' - '- -L- -- - l' '1
-' i ij
I
w-
Iii

~' -- --- -- -
I
SECTION 'A-A'
(TYPICAU
: ---EL 479~ -,
"


ii,. 1:1
- I ~'''' .1.-
,- GRATE BAR WITH STEAM I ~". BURNERS, . . , .' EL468' 6"
I JET U20REO'D) ~--~ 'f; ,,_-
STEAM JET I'~ I ,-.,
,~~,', 6''', ~4112" F~I!NACE ~IDTH H.:

'. I I I'
I ' r: I ' '" rJ I;
, I BUR~ERS -~ i; EL4S7~6"
! " i:
,I I

1~:~~:~2"TEE~C BURJERS ~- ~< I: EL446~6"

I ~ " l DIVISION WALY
I 11.4112 '
!
-r
, i
: 28'-5' FURNACE DEPTH
;


: E ~ EL46e'-6"
-j
DETAIL X
(TYPICAU
;
-- - ----Y
I
, "
,
E_I,.~~'-3"
I 2'6" i .I_t
I fOR REAR WALL (3REdD)
5'0=--.!.-~'~
FOR FRONT WALL(3 REdDl
REFUSE NOZZLE
(TYPICAL)
, 64~51/2"(U9-61/2"SPACES) !',
-I WATE-RCOOLEDPIN:t:iOl.E GRATEBiRS 15 fZ
I WITH STEAM JETS SPACED AS SHOWN -

tUNIT PARTIAL FRONT ELEVATION f&. COL.
SYM ABTt
~ --. jf~:~~~~~~~E~~t~==-~~{~-~~~:~i~t~~~~.~~~~ ~~-1
PARTIAL CROSS SlOE ELEVATION
eOL
FIGURE 111-32. BOILER MODIFICATION NO.5

-------
I -
......
~
I
0'
-.0
".£
"~

~, ,
1>(\
EXISTING ST 4 'IF
8.5 SUPPORTS
'~:
W~~~ \

'-.

~1
2(~"
PARTIAL SIDE ELEVATION-REAR WALL HOPPER
.~
;-
I~III
EXISTING ST4 'IF
8.5 SUPPORTS
- ,
 C'-I/i. ,.:ClUNIT
::! ~T 
-------
In the event that there is insufficient burn-out of
refuse while it is in suspension, the furnace hopper would have been modi-
fied to provide a surface for additional burning. The hopper tubes would
be replaced by a slightly altered tube arrangement so that water- cooled
grate bars could be attached. The header-box under the furnace hopper
would be converted into a plenum chamber for the underfire grate air.
It has also been considered necessary to insulate some of the headers in
the header-box, and to run air ducts into the header-box.
The coal intervane-burner rotations are shown in
Figure III-32. It was felt that the downward flow of air between the two
coal burners in each section of the furnace would tend to give the refuse
an extremely turbulent burning path. Steam jets would be placed at certain
locations on the hopper. These would be operated periodically to remove
ash deposits from the grate bars. It should be pointed out that the angle
of the stationary grate is rather critical. The angle of the existing hopper
floor is close to the values usually recommended.
~
It was felt that the existing forced draft fans could
be used for the type of combined firing considered in this modification. A
tempering air duct system would be included to maintain the underfire air
temperature below 3500F. The required refuse combustion air that is not
admitted through the grate would be taken from the existing windbox. It
has been as sumed that this air would be admitted through the coal burners.
If this proves to be inadequate, additional openings would be installed in
the front wall so that secondary air from the windbox could flow directly
into the furnace.
Most of the cost of this modification would be in-
curred in the modification of the hopper area and the duct work to transport
underfire air to this area. If it were determined that burnout would be suf-
ficient without the installation of grate bars and underfire air system, the
total cost would be drastically reduced.
6.
Air Pollution Control (APC) Equipment
a.
Overview
On the present program, the abatement of S02
emissions is of course a key objective. The approach taken, however, is
based on the scheme of supplanting a portion of the high sulfur fuel used
in power boilers with one of lower sulfur content. That the replacement
fuel also happens to constitute a major waste management problem con-
siderably amplifies the interest that its combustion characteristics would
otherwise merit.
'-'
III- 70

-------
'-..
In this context, therefore, the specification of air
cleaning equipment for the various boiler systems proposed herein is not
actually a primary objective. In fact, the selection of appropriate SOZ-
removal systems is a questionable task to undertake at the present time.
This view is explained in part by a conclusion recently offered (Ref. 96)
by an ad hoc committee convened by the National Academy of Engineering
and the National Research Council to study SOZ -abatement proces ses.
They said: liThe panel reviewed the status of United States and foreign
sulfur oxide abatement and con trol processes and firmly concluded'that,
contrary to widely held belief, commercially proven technology for control
of sulfur oxides from combustion processes does not exist. II
As a result of Government sponsorship, intense in-
terest has focused upon SOZ-removal processes, such that a highly dynamic
R&:D situation has resulted. It is thus highly possible that any process chosen
today may well be soon superseded by newer, more cost-effective processes,
or even be discarded because of technical arguments that are not yet ade-
quately defined.
With these constraints in mind, an attempt has none-
theless been made to provide the candidate steam generators with practical
APC process designs. Section IV of Appendix B furnishes some of the per-
tinent information used in establishing the guidelines which were observed.
In approaching this task, the various candidate sys-
tems have been divided into two categories, based on the manner in which
the flue gases are handled. These are:
.
Category I: Combined- or separately-
fired furnaces, common flue system.
Ca s e s 1, z, 6, 7, 8, and 9; Modifi-
cation Nos. 1, Z, 3, 4, and 5.
.
Category II: Separately-fired furnaces,
isolated flue systems. Cases 3, 4, 5,
, and 10.
Selection of APC equipment for the Category I boilers
must be approached on very nearly the same basis as would be observed for
SOZ -emitting (coal-fired) units. In the Category II systems, however, this
would not be true for the flue gas emitted by the refuse-fired portion of these
systems. The SOZ levels from these furnaces should be low enough that only
particulate management need be considered. Kaiser reports (Ref. 33), for
example, SOZ levels from refractory refuse-incinerators on the order of 15
to ZO ppm.
III- 71

-------
The next area for consideration is the variation in
gas volumes. Furnace availability, refuse availability, and variation in
refuse heating value indicate that the combined-fired units will doubtless
operate with a much wider variation in total gas volumes, even at a base
load, than conventionally fired units. The flow data shown in Table IIl-4
for Cases land 2 indicate 30 to 50% decreases in gas flow, as the refuse
energy input decreases from 100 to 0%. Thus any variation in refuse rate
or the stoppage thereof will result in significant changes in gas volumes,
even if the fuel value remained constant. This variation, coupled with a
typical turn-down of over 50% for power boilers and the actual high varia-
bility in the fuel value of refuse, suggests that total variations of 25 to 100%
must be anticipated for the gas volumes. This situation does not pose an
insurmountable problem in the selection of APC equipment. As pointed out
in Section IV of Appendix B, a variable orifice contactor can accommodate
variations of this magnitude, provided sufficient static pressure capability
is available at the fan for the entire range of flows. The variable orifice
contactor can, in fact, be employed to throttle the fan.
Such wide flow variations would cause some concern
in the design of appropriate electrostatic precipitators. Overall precipitator
performance can be expected to increase at reduced flow-rates, since resi-
dence time increases. However, gas flow reductions down to 25% of rating
will reduce the effectiveness of conventional gas control devices (turning vanes,
etc. ) and result in poor gas distribution and, thus, deteriorated precipitator
performance. Great care must therefore be taken in the design of the flue
work. Three-dimensional model studies of the flue gas system may be re-
quired to accomplish this. .
The final area of problem definition is the particulate
loading in the combustion gases. For the coal-firing components in both sys-
tem categories, the situation is straightforward. In the suspension firing of
coal, approximately 80% of the potential ash in the coal will be released as
entrained fly-ash. Despite the fact that some unburned material is present
in the coal fly- ash, the 80% factor is reliable. If a greater degree of accu-
racy is required, more precise relationships are available in the literature
(Ref. 97). Using the coal analysis shown in Section III, B, 1 and the gas-flow
data given in Table III-4, one would expect approximately 6. 0 lbs fly-ash for
every 1000 lbs of wet flue gas produced by the firing of coal in any of the
systems considered.
Quantification of the particulate loading contributed
by refuse combustion is not as straightforward. Stenburg (Ref. 98) and
Walker (Ref. '99) have correlated particulate emissions from refuse com-
bustion as a function of grate underfire air. Sternitzke (Ref. 100) presented
criteria for air distribution geometries to minimize particulate emis sions
on refractory units, as well as to reduce refractory maintenance and slag-
ging. However, as pointed out in Appendix B (Section V, C), a number of
factors may influence particulate generation rates.
III - 72

-------
The candidate refuse-combustion systems considered
herein are all low-excess -air, water-wall units. The European experience,
coupled with the recent domestic conclusions regarding air distribution, in-
dicate that 75% of the total air should be introduced under the refuse bed.
This conclusion could have serious implications in the selection of the grate.
Some stoker manufacturers might request that as much as 100% of the total
air be used for underfire air. The development of reducing atmospheres,
coupled with an increased particulate burden, rule out such a dispropor-
tionate budgeting of air. .
Although the air velocity through the grate is a func-
tion of the grate size and geometry, a flow of 55 SCFM/£t2 grate has been
set as the maximum that would be encountered. This compares with about
47 SCFM/£t2, using the criteria established for the candidate systems of
50% excess air (refuse), 75% underfire air, and a grate stoking rate of 60
Ib/ft2-hr. For mixed municipal refuse, we would expect this upper flow-
rate to result in approximately 30 lb fly-ash/ton refuse burned. This can
be compared with the median value reported by Nressen., et aI, (Ref. 13)
of 24 lb/ton of refuse for municipal refuse incinerators of over 50 tpd capa-
city. Using the air flow data shown in Table III-4, this would result in a
loading of about 2.5 lb fly ash/lOOO Ib of flue gas (wet) generated by refuse
in all of the systems equipped with grates.
The fly-ash loading that can be anticipated for sys-
tems in which varying stages of the refuse burning occurs in suspension
(Cases 5 through 10 and Modification Nos. 3 and 5) is uncertain. In the same
reference cited above, Niessen warns that the fly ash emissions from sus-
pension-fired systems will be considerably higher than from furnaces
equipped with agitating grates. This potential problem can be as sessed
only when actual field experience has been acquired. It should also be borne
in mind that most flue gas cleaners are sized on the basis of the volumetric
flow that they must accommodate. If particulate loadings exceed the design
value, the rate of efficiency loss is usually much lower than would be true
if the gas throughput exceeded the design limit an equivalent amount.
b.
APC System Possibilities
In Section IV of Appendix B, the combined capability
of some particulate collectors to effect gaseous emission control via mass
transfer is discussed. The most obvious example is wet scrubbing with al-
kaline liquor. Particulate collection efficiency of over 99%, and simultaneous
S02 removal of up to 90% have been previously cited .on power boiler appli-
cations with direct furnace injection of limestone.
A less conspicuous example is the fabric-filter col-
lector, the basic dust removing capability of which is indisputable. Recent
claims have been made of the fabric filterls ability to desul£urize flue-gas
following limestone injection into either the furnace or flue work. Any residual
III- 7 3

-------
desulfurization in the baghouse proper, however, appears highly questionable.
The basic problem with the reactive stone injection process is the sulfate and
sulfite encapsulation of the reactive particles, which allows only the surface
of the particle to react with 502.' Once the surface is spent. further reactivity
requires particle decrepitation ln order to expose fresh surface. It appears
highly unlikely that such an effect would occur to any significant deg ree in the
bag collector. Moreover, baghouse operating temperatures are usually in the
neighborhood of 300oF, which is too low to favor 502 reaction with limestone
or dolomite.
The high degree of particulate collection efficiency
required rules out the application of low energy mechanical or inertial de-
vices. Thus only fabric filters, dry or wet scrubbers of moderate to high
energy inputs (>6-in. w. c. pressure drop), and electrostatic precipitators
need be considered. The state-of-the-art is well enough developed to permit
selection based on a quantitative assessment of the problem. Equally well
defined are the economics of particulate collection at high gas-flow rates.
Traditional use patterns of particulate collectors support the general con-
census (e. g., Ref. 101) that electrostatic precipitators represent the optimum
annual cost solution in a high-volume, high-performance situation. Thus,
in the present analysis, the electrostatic precipitator has been selected in
preference to the baghouse for cleaning the flue gas from Category II refuse-
fired furnaces.
For Category I systems and the coal-fired furnaces
in the Category II systems, 502 as well as particulates must be removed.
In Section IV, P of Appendix B, four commercially available desulfurization
proces ses have been reviewed. In terms of present requirements, none
has been found to be adequately developed or proved for inclusion in the
systems analysis. The principal problems associated with these processes
can be summarized as follows:
.
Reinluft Proces s - Highly complicated
chemical process hardware and operating
expertise are required. Proces s is not
suitable for retrofit situations. Operating
and equipment costs are prohibitive.
.
Catalytic Oxidation Proces s - Highly
efficient dust removal must be practiced
before the flue gas encounters the catalyst.
A process temperature of 9000F is speci-
fied. Equipment costs are prohibitive.
.
Alkalized Alumina Process - Fairly sophis-
ticated chemical proces s hardware is needed.
A source of l2000F reducer gas is required
for the regenerator. Attrition of the alka-
lized alumina is still an unsolved problem.
III - 74

-------
.
Dolomite/Limestone Iniection Process -
Pulverized stone injection increases par-
ticulate loadings 100 to 150%. Severe
plugging problems in the scrubber or
contactor section have been encountered.
Although favored over other processes,
method is considered to be stop- gap
(Ref. 96).
The method selected for combined S02/particulate
removal is the wet lime process. This selection was influenced not only by
engineering considerations, but by the growing commercial acceptance the
process now appears to be experiencing. The scrubber to be used is of the
variable-orifice, venturi type. This type of contactor now appears to have
a promising edge over other types, such as those involving dynamic beds.
It is recognized that the plugging problems as sociated with the dolomite /
limestone injection proces s would only be mitigated and not eliminated in
using the present, related method.
For the Category I systems and the Category II
coal-fired boilers, the APC system shown in Figure III- 34 is specified.
It basically consists of a venturi scrubber circulating a lime suspension,
a spray tower pumping the same absorbent, and a reheat chamber for
minimizing stack plume. Using a pressure drop of 6-in. w. c., 99% of
the particulates should be removed in the venturi and at least 65% of the
S02' The spray tower is expected to absorb additional S02 for a total
removal of 90%.
-"'"
For the Category II units in which refuse only is
fired and the flue gas is handled separately from that from the coal-fired
unit, cleaning will be accomplished by electrostatic precipitation alone.
The parallel coal- combustion gases would be handled as shown in Figure
Ill- 34. Following the cleaning steps, however, the flue gas from the
coal-fired unit can be combined with the hotter flue gas from the refuse-
fired unit{s). This will result in plume attenuation and permit the use of
a common stack. The system is shown in Figure III-35.
Table III-II provides data on the operating charac-
teristics of the APC system when coupled to candidate systems. Because
of the large number of possible arrangements that CE>uld be considered (due
to capacity and refuse-fraction variables), the table has been limited to six
example s.
The wet calcium- based APC systems also offer
excellent CI/HCI emission control, perhaps at levels as high as 98-99%,
but lime addition would have to be increased significantly. In some situations,
lime consumption might double. The lime consumption values given in Table
Ill-II are based on S02 removal only.
III- 7 5

-------
~
I=:
I
~
'"
      FAN a STACK -:: REHEAT ~
      ~
          ~ ~ 
         .  
COMBINED -- AIR ...... VENTU R I --   -- WET S02 
FURNACE - HEATER - SCRUBBER -- DEMISTER ...... ABSORBER 
     ,      
    ~ i      a 
    -       
    -       
   LI ME + Hi>    , r ,r  I 
         LIME + H20 
     ........  ClARI FIER   
     --.    
    , r      
BTU'S
BlOW..DOWN
EFFLUENT
FIGURE 111-34. RECOMMI:"):) A )C SYS . : ~ :OR COMBI~ E ) S( 2 - PARTICULATE REMOVA -

-------
1\
FAN a STACK
H
~
I
--.J
--.J
          ~~
REFUSE - AIR -  ELECT.     -
- -      --
FURNACE  HEATER   PPTR     
COAL - AIR - VENTURI - DEMISTER - WET S02
- - - -
FURNACE  HEATER  SCRUBBER     ABSORBER
    J ~     ~i
     -     
     -     I
   LlME+Hi>  l ' "  lIME+~O
       CLARIFIER  
     "     
BLOW DOW N
EFFLUENT
FIGURE 111-35. RECOMMENDED APC SYSTEM FOR CATEGORY II PLANTS

-------
       TABLE III-II    
     CALCULATED OPERATING CHARACTERISTICS OF RECOMMENDED APC SYSTEMS  
         Modification /II Modification /12 Modification 113 Modification 114
     Case 3 (400 MW, 25% ReCuse) Case 10 (100 MW, 580/< ReCuse) (43.20/< ReCuse) (23.60/< ReCuse) (42.70/< ReCuse) (41.80/< ReCuse)
     ReCuse Umt Coal Umt Refuse Umt Coal Umt Combmed Umt Combmed System Combmed Umt Combmed Umt
 APC System    Pptr Lime System Pptr Lime System Lime System Lime System Lime System Lime System
 Flue Gas Flow, 103 ACFM (oF) 178 (575) 990 (300) 343 (450) 147 (450) 342 (450) 640 (450) 241 (450) 252 (450)
 Total Liquor Pumping, gpm  2214  326 795 1432 560 587
 Recirculated Liquor, gpm  1640  216 520 992 366 384
 Evaporative Loss, gpm  324  68 187 350 132 138
 Effluent Blow-Down, gpm*  250  38 88 90 62 65
 Total Make-Up, gpm   574  106 275 440 194 203
H            
H Pump HP     35  5 12 21 9 9
H     
I            
-...) Saturated Gas Flow, 103 ACFM (oF)  820 (125)  108 (133) 260 (140) 486 (140) 183 (140) 192 (140)
(X)  
 System 6P, in.-w.c.  2 8 2 8 8 8 8 8
 Fan HP    57 1400 110 140 330 620 233 245
 Lime Required, 103 tons/yr  40  6 14 26 10 11
 Electrical Requirements, kw-hr/yr 590  1120     
*Based on nominal 3 gr/ ACF ny-ash, 200/< solids eCfluent.

-------
As will be noted in the next section, the data shown
in Table .III-ll were not us ed for the cost modelling effort. It was decided
that more conservative bases should be employed for that purpose.
7.
Refuse Handling System
a.
Design Selection
A variety of possible waste feed arrangements can
be considered in preparing the lay-out of a refuse-firing power plant. Choice
among such alternatives is not only influenced by practical design and/or
cost trade-offs, but by the peculiarities and policies of local waste manage-
ment programs. The design complications implicit in trying to deal with the
latter problem were avoided on the present program by assuming certain
operating conditions. These were as follows:
'.
ASSUMED OPERATING CONDITIONS
FOR WASTE RECEIVING
Refuse Composition
See Section II, A, 2; oversize
'j> 5 wt- % of total
Recei ving Area
Power Plant
Pickup/Deli very Schedule
Truck Capacity
6 days per week

24 yd3 (compacted to density of
20lb/ft3)
Plant Refus e Consumption
Small Plant = 2000 tpd
Large Plant = 8000 tpd
Storage Capacity (80% usage
of available volume)
Sufficient for 2 days of operation
and assuming settled refuse den-
sity to be 12 lb/£t3:

Small Plant = 30,000 yd3
Large Plant = 120,000 yd3
The well-tried and extensively accepted method of
handling refuse at inCinerators and steam generator plants can be seen in
various figures of this report. After weigh-in, the collection trucks tip their
loads into one or more storage pits contiguous to the furnace where material
is outloaded to the charging chute by overhead crane.
In recent years, however, the trend has been to con-
veyor feed systems. The principal reason for this, as Sutin points out (Ref.
102), is that "they (moving cranes) generate very high capital, operating and
III - 7 9

-------
maintenance costs. II This conclusion was verified on the present study.
Thus the basic system selected for the present design development speci-
fically excluded the use of cranes for refuse handling. It was recognized,
however, that certain technical problems are still associated with the
charging of furnaces with conveyors.
The type of system needed will vary with the mode
of firing that is employed in the furnace. This is illustrated schematically
in Figure 1lI- 36. It can be seen that, for each of the three types of furnaces
stoked, the same basic refuse-processing components are required. Thus
in c:ieveloping inputs for the cost model, most of the components were con-
sidered to be identical. Exceptions to this were the use of the additional
conveyor legs shown in Figure 111-36 and of the numper and capacity of.the
shredders specified.
In the discussion of the various system components
which follows, the shredding operations are not stressed. These involve
fixed-design items that have been discussed previously (Section II, D, 2). It
should also be pointed out that the use of metal separators was not included
in the systems analysis because of the optional nature of this operation.
b.
Weighing Stations
Refuse is usually weighed as it arrives at a refuse
reduction plant. The most obvious reason for doing this is the need to es-
tablish a reasonable basis for charges to the refuse collector. It can also
as sist in detecting undesirable waste or loads of material which will require
special attention. In terms of boiler or incinerator fuel computation, an
accurate knowledge of the weight fired enables comparisons to be made of
actual and rated performance. Furthermore, if the source of each load of
delivered refuse is known, the data then become a valuable tool for estab-
lishing quantity and rate information for specific areas and projecting
patterns. .
The sizes of the plants proposed in this study are
such that the use of automated weighing stations is almost a necessity. As
an example, it can be calculated that with a six-day refuse collection sched-
ule, a 2000 ton per day plant must be prepared to accommodate under peak
conditions 144 trucks per hour. This is based on the assumption that 80% of
the trucks will arrive during two I-hour periods, each carrying an average
load of 6. 5 tons of refuse. Manual weighing of each truck would require a
sizable investment in scales to handle this traffic and, more importantly, a
large annual expense in wages paid to the scale attendants.
There are a number of scale manufacturers capable
of providing highly sophisticated, fully automatic weighing systems. In a
typical installation, the presence of a truck is detected by a magnetic sensor
which activates the scale controls and automatically makes a zero check and
III - 80

-------
WEIGHING
STATION
--0---.
RECEIVING
HOPPERS
OVERSIZE
MATERIAL
SEPARATOR
OVERSIZE
MATERIAL
MILL
SHREDDER
4-IN TOP
SIZE
....- -,
~ METAL
~ REMOVAL r----
L- --'
----------
REFUSE
STORAGE
I
I
I
I
I
I
I

r-i-'METAL
I SALVAGE
r - .L -,
METAL
REMOVAL t----- EARTH
I (:;") FILL
- - ~ - J DISPOSAL
2ND STAGE
SHREDDER
2-IN. TOP
SIZE
FURNACE
FIRING MODE:
o SUSPENSION FIRED BOILER
o STOKER-SPREADER OR SLAGGING FURNACE
o AGITATING GRATE FURNACE
'.\
FIGURE 111-36.
WASTE HANDLING SYSTEMS FOR DIFFERENT TYPE FURNACES
III- 81

-------
scale adjustment. A gate may then be opened to permit the truck to drive
onto the scale where additional sensors and guides assure that the truck is
in the proper position for weighing. The driver then places a key or iden-
tification card in a slot at the scale house and the weighing cycle is started.
A record ticket for the driver is printed and this ticket and the identification
card are released to the driver. The exit gate is then opened and the system
is ready to repeat the cycle.
Data from the weigh station may be recorded and
kept at the weigh station or may be transmitted to any desired remote loca-
tion. Truck identification, hauler identification, date, time, gross, tare
and net weights can be printed and punched onto tape or cards for later
data processing and billing. .
c.
Truck Unloading
In the past, incineration plants usually incorporated
unloading arrangements wherein the refuse trucks would dump directly into
a storage pit. F'rom there, the contents would be transferred by crane di-
rectly to the furnace. Bulky or oversize materials would either be manually
separated prior to dumping or be left for the crane operator to locate and
reduce by grapple impact. More modern plants, e. g., that under construction
for the City of East Hamilton, Ontario, employ active- or live-bottom re-
ceiving facilities from which the material can be continuously conveyed to
the furnace or to intermediate refuse-conditioning operations. In addition
to providing an opportunity for visual inspection of the incoming refuse, the
active-bed concept also permits a more controlled feed rate to the material
handling and processing systems.
The quantities of refuse which are to be proces sed
by any of the proposed systems in this study are so large that it becomes
lmpractical to depend exclusively on human management of oversized refuse.
An error on the part of the responsible individual could easily result in jam-
ming of a refuse feed chute or damage to a furnace grate, and temporary
shut-down of a portion of the system. Because of this pos sibility, direct
unloading to the storage pits was ruled out in favor. of the use of receiving
pits that transfer the feed to the storage structure(s) or shredders by means
of conveyors.
The dimensions of the active- bed receiving pit are
dependent on the number of unloading stations required:to accommodate the
anticipated delivery truck arrival rate, which is, in turn, a function of the
plant capacity. The results of queueing studies to determine the number
of unloading stations required as a function of plant size are presented in
Appendix C.
III-82

-------
d.
Receiving Pit
The refuse receiving installation would consist of an
elongated, slope- sided, concrete hopper, along the grade-level edges of which
would be the truck receiving stations. The bottom of the receiving pit would
consist of a cleated conveyor belt which would agitate and move off the input
material to the further processing stages.
Assuming that four minutes is required for each un-
loading cycle, then the unloading rate per station will be 324 £t3/min. This
is based on the (conservative) assumption that each truck is fully packed
(24 yd3) and that the refuse expands 1000/0 after ejection. On the basis of
uniform receiving rate (no peak effect), a 2000 tpd plant would have to host,
simultaneously, 4.3 trucks, and an 8000 tpd plant, 17.2 trucks. Rounding
off to 5 and 18 positions, respectively, and doubling the required stations
to 10 and 36, respectively, in order to reduce queueing during peak times,
pit lengths can be estimated. These would be 150 ft for the small and 540
£t for the large plant, if 15 £t were allow~d for the width of each station.
The basic module, however, would be a 75 ft, 5-station hopper, two of which
would be used in small plants and eight in large plants. The use of multiple
receiving pits would permit turn-down of the operation during slack period,
would tend to reduce back-up if anyone system fails, and would permit a
more effective location of the pits in large plants with respect to layout of
the grounds and nearby roads. A drawing of the hopper facility is shown in
Figure ill-37. The dimensions shown provide a pit volume of 1300 ft3/station,
which is slightly greater than twice the individual truck capacity. The dis-
charge conveyor would remove material at a considerably faster rate than
the maximum truck input rate. The latter rate would prevail of course only
when all stations were occupied by trucks requiring the least amount of time
to unload (-2 min). Thus the conveyor could be slowed considerably or even
completely stopped for several minutes to accommodate downstream problems,
without an overfill occurring. The conveyor would be a 72-in. flat-belt type,
fitted with 60-in. x 4-in. cleats every 6 ft. Based on the loading character-
istics and discharge rate desired, belt speed would be about 530 ft/min.
e.
Oversize Material Separation and Reduction
The heterogeneous nature of refuse makes separation
or classification operations the most difficult of all material handling problems.
Visual sorting of the total incoming refuse load is not considered practical, and
the use of rotating screen drums would necessitate that excessively large mach-
ines be provided in order to handle the flow rates contemplated. It may not be
possible to entirely eliminate human judgment from the sorting process, but
even a small degree of automation will greatly reduce the number of decisions
that the responsible operator would be called upon to make.
III- 83

-------
H
H
H
I
00
*'-
GRADE
112'
~:.~
I - - :::::;l
I
I
I
I
I
U
     I
1 2 3 4 5 
    . 
     -J.-
     -
'A'
---
1 ~- )OUTPUT CONVEYORS
     -
     T-
    . i
6 7 8 9 10 I
     I
'A'
~
15'
--_::!J
PLAN
5'



~~~


I t.I

T ~---------
ELEV.
FIGURE
-37.
RECEIVING PIT LAYOUT
t-- 5' ~
~
6'
.,
'A-A' CROSS SECTION
8'-8"
1

-------
Variable speed control on the active bed of the tip-
ping floor system will make possible a nearly constant flow of refuse to the
transport conveyor. With a fairly uniform refuse depth on the conveyor,
mechanical, optical, or beta- radiation detectors can be used to sense objects
that extend above a predetermined height. Automatic dumping devices could
then be used to divert the oversized refuse to the bulk- reduction stage. Some
flat, oversized objects will pass through this scanner undetected, such that
their diversion would have to be accomplished by manual control at the dis-
cretion of the receiving pit attendant. Because of uncertainties of costs,
however, this type of equipment could not be included in the systems designs.
There it was assumed that oversized refuse would be separately delivered,
a common practice in many LMA's.
The reduction of oversized refuse is an essential
step in the refuse handling process regardless of what methods of transport,
storage, or firing are to be used. The equipment to be employed for the size
reduction of bulky objects is not expected to produce an output of small par-
ticle dimensions, but only to furnish a size that will be acceptable to the
furnace charging system, the refuse storage and retrieval system, or other
shredding equipment. Because the shredder must be sized to accept fairly
large pieces, it will usually have a throughput capacity that is greater than
the feed rate, which, in oversized articles, is only about S% of the total
refuse delivered.
f.
Storage Facilities
(I)
Design Factors
Experience has shown that, refuse exhibits poor
flow properties and tends to develop stoppages in storage structures equipped
with bottom-release devices. This effect is frequently encountered when voids
form over the output chute, hopper, or conveyor trench. The remaining stalled
mass is then said to be "bridged" or "arched." A related problem is the in-
activation of entire vertical portions of the pit contents that are not standing
directly over a point or line of bottom discharge flow. It is thus essential
that the bottom of the pit be designed so that possible footings for arching or
piling effects are minimized.
In developing designs for the storage structures,
another factor that was taken into account was land availability. The two types
of pits considered were designated low and high profile, based on their suita-
bility for power-plant grounds with ample and limited land availability, re-
specti vely.
(2)
Low Profile Storage Pit
A drawing of this concrete'-lined, metal-roofed;
structure is presented in Figure III- 38. Not shown there is the location of the
input conveyor. This device would enter the structure on the centerline just
III-8S

-------
I ARCH ROOF
,
Wb
-11

I

.1
Wa
CONSTRUCTION SPECIFICATIONS
NUMBER OF PITS
CAPACITY PER PIT, yd3
SMALL PLANT LARGE PLANT
2 4
15,000 30,000
155 195
3,300 4,800
LENGTH OF PIT, ft
GABLE WALL SURFACE, ~
OTHER DIMENSIONS, ft
Wa
Wb
77.5 97.5
81.5 107.5
39 53
31.3 37
Ha
Hb
FIGURE 111-38. LOW PROFilE REFUSE STORAGE PIT
III- 8 6
ORIGINAL
GRADE

-------
below the top of the roof and tr~vel the full length of the pit. Rakes would
be suspended from the roof over the conveyor run to remove the load in an
even pattern. One 72-in. input belt would be used for each of the small
plant storage-pits and two belts of that width in the larger pits.
The discharge conveyors would be operated at
600 ft/rnin in both sizes of pits. . This fast rate would produce a certain de-
gree of agitation in the pit bottoms. The belt width required would be 72-
and 96 -in. for the small and large pits, respectively. For costing purposes
the following conveyor requirements were as sumed:
CONVEYOR SYSTEM FOR LOW PROFILE
REFUSE INST ALLA TION
1.
2.

3.
4.
5.
      Total Length
  Length, ft No. of Runs of Conveyor
  needed, ft
Location  ~ ~ 1& ill N (B)
Receiving Pit to Shredders  270 270 1 4 270 1080
Shredders to Storage Pit (inclined) 120 120 2 8 240 960
Along Roof of Storage Pit  160 200 2 8 320 1600
Along. Bottom of Storage Pit  160 200 2 4 320 800
From End of Pit Bottom to  114 142 2 4 228 568
4 ft above grade (250 Slope)
End of Item 5 to Charging Chute 300 300 2 4 600 1200
6.
1 )
2)
Small Plant (2000 tpd)
Large Plant (8000 tpd)
(3)
High Profile Storage Structure
. A sketch of this low floor-area system is shown
in Figure III-39. This concrete-walled building would also be covered with a
metal, arched roof. By nesting the structures, one wall would be eliminated
in the twin (small plant) installation and three in the qua.druple (large plant)
arrangement. In the latter case, two 72-in. feed conveyors would be used
in each unit instead of the single 72-in. conveyor shown in Figure III-39 for
small plant requirements. Except for length, the discharge conveyor belts
would be the same widths as for the low profile system and would also move
at 600 ft/min. The conveyor requirements assumed for high profile con-
struction are as follows:
III- 87

-------
INPUT CONVEYORS
OUTPUT CONVEYORS
I
T
I
Hb
GRADE
CONSTRUCTION SPECIFICATIONS
NO. OF BUILDINGS
CAPACITY PER BLDG, yd3
SMALL PLANT
2 (AS ABOVE)
LARGE PLANT
4 (ALL COMMON WALL GANGED)
15,000
30,000
DIMENSIONS, ft
Wa
Ha
L
40 50
18 24
50 60
16 20
165 218
Hb
He
FIGURE 111-39. HIGH PROFILE REFUSE STORAGE BUILDING
III - 88

-------
CONVEYOR SYSTEM FOR HIGH PROFILE
REFUSE INST ALLA TION
       Total Length
   Length, ft No. of Runs of Conveyor
   needed, ft
 Location  ~ ~ 1& ill. 1& illl
1. Receiving Pit to Shr~dders  270 270 1 .4 270 1080
2. Shredders to Storage Building 150 150 2 8 300 1200
 (inclined) 
3. Along Roof of Storage Building 170 220 2 8 340 1760
4. Along Bottom of Storage Building 170 220 2 4 340 680
5. From End of Item 4 to 4-ft 170 220 2 4 340 800
 above Grade (250 Slope) 
6. End of Item 5 to Charging Chute 300 300 2 4 600 1200
1) Small Plant (2000 tpd)       
2) Large Plant (8000 tpd)       
As a simplification, Figure llI-39 shows the
discharge belts as being positioned at the bottom of the bin hoppers. Actually,
these belts would be located to one side in parallel-aligned conveyor tunnels.
Material from the bin would be transferred to the conveyor by means of a
slot feeder, such as used in coal handling systems. This device would con-
sist of a rotating rake mounted on a tracked machine that would travel back
and forth along the side of the hopper bottom and claw material from the latter,
throwing it onto the discharge conveyor. This arrangement is preferred be-
cause in a high profile storage structure the vertical load on a bottom-mounted
conveyor would probably impede its operation.
(4)
Silo Configuration
In the low and high profile structures just de-
scribed, configurations were adopted that would impart good flow character-
istics to the contents. In the first case, steeply angled walls would be em-
ployed so that no horizontal surfaces would be presented to furnish footings
for refuse bridging or vertical piling. In the latter (high profile) case, where
. a hopper bottom must be used, the occurrence of 'flow hold-ups is minimized
by virtue of the narrow, elongated shape of the building. This greatly in-
creases the ratio of active or live-bottom area to cross-sectional area.
. .
The corollary to the above logic would be that
a structure having a square or circular cros s - section would not be advised.
These shapes would accept comparatively short bottom-conveyors and thus
, afford a low active- to cross-sectional-area ratio. If, however, a storage
III-89

-------
shell with regular cross-sectional dimensions were equipped with bottom-
sweep machinery (e. g., chain-dragged buckets), acceptable flow character-
istics could result. Because a cylindrical geometry is the most compatible
with such (usually rotary) devices, an analysis was conducted based on a
silo-type storage system. This is shown in Figure 1lI-40. This structure
was costed on the basis of the use of a metal skin (1/8-in. thick; 5.1 lb/ft2)
because concrete forming in this shape is more expensive. Vertical stays
were specified, each ten feet apart up to the top ring-girder. Drawings of
the silos were then submitted to a manufacturer of bottom- sweep equipment
for the development of cost estimates. Their hardware was considered to
be priced closely with similar installations produced by competitors. When
these costs were added in, the following comparisons were derived:
COMPARATIVE COSTS OF HIGH AND LOW PROFILE
SYSTEMS AND LIVE-BOTTOM SILOS
Sys tem
Total Building Costs, 106 $
Small Plant (2 units) Large Plant (4 units)
Low Profile
High Profile
Silo Configuration
0.69
0.57
1. 43
1. 95
1. 55

4.44
As pointed out in Appendix C (Table AC-12),
the above costs include a significant contingency allowance for the estimates
generated by the staff of the present study. None, however, were added to
the commercial bids submitted for the sweep equipment.
In view of the much higher costs of refuse
storage facilities equipped with bottom-sweep machinery and the uncertainty
that such equipment will operate satisfactorily with unground refuse, such
designs were omitted from the cost model. The above cost information
should, however, be of value to those who would prefer a more conservative
design approach.
C.
EVALUATION OF SELECTED ENGINEERING DESIGNS
1.
Cost Model
A cost model has been developed to estimate the net disposal
cost that could be achieved with each of the alternative concepts described in
the preceding section. The cost model (described in detail in Volume II, Ap-
pendix C) has been derived to permit the evaluation of each candidate concept
under a broad range of operating conditions. In general, it was felt that no
one concept would prove to be universally applicable under all conditions.
As a result, the object of the cost analysis was to determine which were the
superior candidates and the conditions under which they were optimum.
1lI-90

-------
I /. I w~

~l1TPUT CONVEYORS L ~ / I
H
CONSTRUCTION SPEC I FICA TIONS
INPUT CONVEYORS
FOUNDATION
 SMALL PLANT LARGE PLANT
NO. OF SILOS 2 (AS ABOVE) 4 (ON SAME FOUNDATION)
CAPACITY PER SILO/~ 15,000 30,000
DIMENSIONS, ft  
W 110 130
L 220 520
H 96 126
D 94.5 113
FIGURE 111-40. REFUSE STORAGE SILO
III- 9 1

-------
a.
As sumptions
The cost mode!!. developed for this program is based
on the basic assumption that the power company using refuse as a fossil
fuel substitute would be a regulated public utility with privately owned
equity financing and subject to all applicable Federal, state, and local taxes.
An important parameter affecting the net cost
of refuse disposal in a new combined firing plant is the value of the
power generated in that plant.. If the utility had no need for additional
capacity when it installed the plant,. the value of the power generated by
it would be the marginal cost of the fossil fuel saved by substituting
refuse. If.. however, the utility required the additional capacity, the
value of the power would be the same as that from a conventional plant
of the same size that had been built under contemporary cost conditions.
Because most utilities are faced with increasing power demands, the
latter case has been assumed. Thus, in computing the power credit
for combined firing plants, the cost of power for an analogous, con-
ventional plant was applied. In addition, it was as sumed that the fos sil
plant would have to meet the same particulate and 502 emission standards
as the combination-fuel plant, and the cost of emission control equipment
was included in both designs.
The capital annualization rate ranged between 14
and 16 percent, depending on the assumptions made concerning interest
rates, allowable return on investment, and equipment life. The factors
considered in this rate are as follows:
Factor
Basis
Capital Recovery
The capital recovery factor is
based on a 7 percent return and
the estimated equipment life.
Insurance
The annual cost of insurance is
estimated to be O. 25 percent of
the total capital investment.
Federal Income Tax
This cost is based on a 55 percent
tax on profit and assuming half of
the total capital investment is
borrowed at 6 percent.
Other Taxes
The annual cost of state and local
taxes is estimated to be 1. 9 per-
cent of the capital investment.
llI- 92

-------
For example, assuming an average equipment life
of 20 years, the annualization rate is found to be 14.7 percent as shown
below:
Item
Annualization Factor as a
Percent of Capital Cost of
Equipment
Capital Recovery
Insurance
Total Annualization Factor
9.43
0.25
3. 10

1. 90
14.68
Federal Income Tax
Other Taxes
In computing the net disposal cost, a nominal plant
factor of 80% was used, primarily because it was assumed that the combined
firing plant would be operated as a base load plant. Thus, if the plant was
designed to fire 1000 tons per day, the cost of disposal was based on 800
tons per day. The sensitivity of other assumptions concerning plant factor
was also examined.
A nominal coal cost of 31f/million Btu was assumed
for this analysis. The sensitivity to coal costs in the range of 22 to 36f/million
Btu was investigated.
As adopted earlier, the nominal heating value for
refuse was again assumed to be 4460 Btu/lb, and the heating value of coal
(with a slight rounding) was assumed to be 12,020 Btu/lb. The effect of
these assumptions on the net disposal cost was also analyzed.
b.
Cost Model Structure
The cost models for both the combined firing and
conventional power plants have been developed in an analytical (equation)
format for use in a computer cost and optimization program developed for
this project. The cost model for computing the net disposal cost of refuse
has been divided into 5 categories: capital costs, capital annualization rates,
operating and maintenance costs, power credit, and residue disposal costs.
The base date for all cost factors used is July 1969.
The capital cost model considers all costs associated
with the fixed plant and equipment required to operate the combined firing
plant, in vol ving :
III-93

-------
factors
utility ,
.
Land and land rights
.
Structures and improvements
.
Boiler plant equipment
.
Auxiliary boiler equipment
.
Turbine-generator equipment
.
Accea aory electrical equipment
8
Air pollution control equipment
.
Waste handling equipment
.
Engineering and construction supervision
The capital annualization rate model considers all
required to compute the annualization rate for a regulated public
including the following factors:
8 Rate of return  
8 Equipment life  
8 Federal income tax rate
8 Interest rate  
8 Debt to equity ratio
. Insurance  
8 State and local taxe s
considering:
The operating. and maintenance costs are estimated
8 Plant factor  
8 Operating labor
. Maintenance labor 
8 Fuel costs  
8 Supplies and maintenance materials
. Electrical power
. Cooling water 
Ill- 94

-------
As discussed in the previous section, the power
generating credit is computed by determining the cost of power in an analo-
gous conventional fossil fuel plant of the same size, fuel costs, plant factor
and debt structure as the combined firing plant.
. The residue disposal costs were calculated on the
assumption that the material could be disposed of in a specially operated
residue disposal site and that the residue would be hauled to the disposal
site in IS-ton payload vehicles, each operated by a single driver.
Tables I11-12 and Ill-13 summarize the cost model
assumptions and structure for a typical case. Table Ill-12 shows the a.s-
sumption's made for this particular case and its operating characteristics.
In this case it was a 400 MW plant deriving 25% of input heat from refuse
and processing 2232 tons per day of solid waste. Table Ill-13 summarizes
the costs for this particular case, showing a total capital cost of $67,080,000,
an effective annualization rate of 14.6% and a. net disposal cost of $0. 57 per
ton. "
, . The costs of disposal in a combined firing plant are
often compared to that of sanitary landfilling. Such a direct comparison is,
however, often misleading since a sanitary landfill system often involves
considerable hauling of refuse at a cost of 20 to 30 cents per ton-mile,
much of which may be avoided in a combined firing disposal system. In
addition, the cost of. operating a sanitary landfill for the residue may be
significantly less than that for non-putrescible refuse ($0. 75/ton vs $1. SOl
ton). In order to avoid the ambiguity of a direct disposal cost comparison,
a model has been constructed to permit comparisons of total haul and dis-
posal costs. '
, When the cost of transporting the collected refuse
3.5 miles to the plant is considered, the total haul, processing, and disposal
cost would be $2.21 per ton. This is the figure that should be compared to
the haul and disposal cost of alternative systems. For example, consider
a sanitary landfill at $1. SO/ton and a haul cost of 23f/ton-mile (equivalent
to a 3 -man crew operating a 20 cu yd packer truck at 30 MPH). If the lancl-
, fill site was more than 3 miles from the center of population, the total haul
and disposal cost would be greater than the $2. 21 Iton cost of this example.
c.
Cost Model Results
The objective of this section is to present the cost'
results of systems that appear applicable to the problem of power generation
utilizing solid waste as a fuel, and to delineate the operating charac,teristics
of these systems in sufficient detail to permit additional evaluation of promi-
sing systems. The conceptual design of each system that was studied was
synthesized from consideration of the most acceptable methods' of incinerating
I11-95

-------
TABLE III- 12
SOLID WASTE/FOSSIL FUEL POWER PLANT
(SEPARATELY FIRED ECONOMIZER)
POWER PLANT CHARACTERISTICS
Plant Power, MW
Plant Factor
Ratio of Heat from Refuse Combustion to Total Heat Input
Design Waste Load, tpd
Average Waste Load Processed, tpd
Number of Turbines
Number of Steam Generators
Number of Refuse Fired Economizers
COST PARAMETERS
Rate of Return on Capital, %
Income Tax Rate, %
Other Taxes, % of Capital Costs
Insurance, % of Capital Costs
Interest on Borrowed Capital, %
Unit Cost of Land, $/acre
Fraction of Capital Borrowed, %
Steam Generator Efficiency (with Fossil Fuel), %
Steam Generator Efficiency (with Waste Fuel), %
Electrostatic Precipitator Efficiency, %
Wet Scrubber Efficiency Particulates/S02' %
Heating Value of Waste, Btu/lb
Heating Value of Coal, Btu/lb
Cost of Coal, ~ /l 06 Btu
Residue Disposal Costs
Landfill, $/ton
Transportation, ~ /ton-mile
III - 96
400
0.80
0.25

2,790
2,232
1
1
3
7
55
1. 90
0.25
6
10,000
50
87.0
65.0
98.0
98.0/90.0
4,460
12,020
31
0.76
23

-------
FPC Codes
TABLE ill-13
COSTS OF A SOLID WASTE/FOSSIL FUEL POWER PLANT
(SEPARA TEL Y FIRED ECONOMIZER)

CAPITAL COSTS, 106 $
Description
310
311

312
314
315
316
Land and Land Rights
Structures and Improvements
Boiler Plant Equipment
Turbine-Generator Equipment
Acces sory Electrical Equipment
Misc. Power Plant Equipment
Air Pollution Control Equipment
Waste Ha,ndling Equipment
Engineering and Inspection
Total Capital Cost
ANNUAL COSTS

.6
Annual Capital Cost, 10 $
(Effective Annualization rate,

Water Cost, 106 $

Operating Labor, 106 $

~aintenance, 106 $

Coal Cost, 106 $

Residue Disposal, 106 $
14. 6 percent)
Total Annual Costs, 106 $

Annual Credit for Power Generated, 106 $
Quantity of Waste Burned, 103 ton/yr
Disposal Cost, $/ton
Transportation Cost (3. 50 Mile Haul Radius),
Total Cost, $/ton
ill-97
$/ton
0.69
5.47
35.28
13.75
2.87
0.59
4.04
1. 81
2.59

67.09
9.77
o. 01
0.62
1. 73
6.76
0.36
19.23
18.76

815

0.57
1. 64
2.21

-------
refuse and still be compatible with power generation. The design concepts
represent systems significantly different in design and operation so as to
be representative of all the alternatives cornpatible with existing and future
operating conditions and technology.
The detailed cost model and cost data presented in
Appendix C were incorporated into a computer program to determine the
optimum design and operating point of a waste-fossil fuel power generating
plant. The various design configurations studied in the program are de-
scribed in Section III, B. These candidates are summarized for conven-
ience in Table IIl-14.
As stated previously, the credit for power generation
is computed from the cost of a conventional coal-fired plant operating with
the same regulated rate structure, debt to equity ratio, interest rates, plant
factor, and fuel costs as the combined firing plant. This model is summarized
in Figure C-17 of Appendix C. For example, for a 300 MW plant with a fuel
cost of 31 s': /million Btu, the model shows the power generation credit to be
7 mills /kw-hr.
Eight of the ten candidate design concepts were
examined in detail using the computer cost model. Case 8 was found after
preliminary investigation to be almost identical in cost to Case 2, except
for refus e proces sing requirements, and Case 9 was excluded after pre-
liminary study on the basis of unrealistically high costs.
As a general rule, the costs were computed at power
generation rates between 50 and 500 MW, at 50 MW intervals)~, and refuse
fractions (percent of heat from refuse) at 20, 40, and 60%, except where
design considerations required the use of other specific values. Thus it
was possible to have 30 different cost options for a single case. Figure
III-41 shows the effect, for Case 2, of varying the plant power capacity
with a constant refuse fraction. It is apparent that, for this case (and most
cases), the effect on disposal cost by plant power capacity is small because
the credit for generating power is decreasing as the plant capacity increases.
The effect of refuse fraction for Case 2 is shown in Figure III-42 for a con-
stant 300 MW power level.
As will be shown, Cases 2, 3, 4, and 10 are
the most promising alternatives. The disposal costs for these 4 cases are
shown in Figure III-43. Figure llI-44 shows the disposal costs for Cases 1,
5, 6, and 7. Based on this comparison, the latter group was excluded from
further consideration. Case 8, which had been considered as being nearly
identical (except for refuse grinding) in cost with Case 2, was also dropped
*For clarity, data in the present text are shown only for the 100, 300,
and 500 MW sizes.
llI- 98

-------
Case.
1
2

3
4
5
6
7
8
9
10
TABLE IIl-14
TEN STEAM GENERATOR DESIGNS SUBMITTED
TO COST MODEL ANALYSIS
Description
Separate Furnaces with Common Convective Passes

Combined Furnaces

Separately Fired Economizer

Separate Fossil Fuel Superheater
(Saturated Steam from Waste)

Separate E:'os sil Fuel Superheater
(Partial Superheat from Waste)
Suspension Fired Furnace
Spreader Stoker
Slagging Furnace
Arch Furnace
Arch Furnace and Separate Fossil Fuel Superheater
llI-99

-------
3.00
2.00
z
~
".:
CI)
8
-' 1.00
c(
.~
o
-1.00
o
REFUSE FRACTION (fwl = 0.4
PLANT FACTOR'" 0.8
COAL COST = 31 CENTS/106 BTU
3000 TO NIDAY
5000 TONIDA Y
100
200 300
POWER, MW
500
400
FIGURE 111-41. DISPOSAL COST AS A FUNCTION OF PLANT CAPACITY - CASE 2
III-lOO

-------
3.00
POWER LEVEL = 300 MW
PLANT FACTOR = 0.8
COAL COST=31 CENTS/106 BTU
1000 TONS/DAY
2.00
6000
z
~
~
8 1.00
..J
.4
~
is
o
-1.00
20
40
REFUSE FRACTION,%
60
80
. .
FIGURE 111-42. DISPOSAL COST AS A FUNCTION OF REFUSE FRACTION - CASE 2
III-I0l

-------
8.00
7.00
6.00
z
o
t:
fh
Z 5.00
o
i=
«
I-
~
o
A.
en
~ 4.00
~
I-
~
Z
o
;:)
rl 3.00
z
I-
o
~
tn
8 2.00
..J
~
o
a.
en
o
1.00
.1.00
o
o
I
_!:i- 100 MN
I
\
.
.
.
.
LEGEND
CASE 2 . 0
I
CASE 3 . -j-
CASE 4 0
CASE 10 )<
100MW -.-
300 MW -...-
500MW -.....-
PLANT FACTOR = 0.8
~< 100 MW
.~
e
...-
. . .
....~
EI
100 MW
0-
>~ 300 MW
500 MW
'"
,,-,
6000
PROCESSED WASTE LOAD, TPD
.~- ~ - ¥ - . - -.......-. - ~ .
FIGURE 111-43. NET DISPOSAL COST VERSUS WASTE LOAD - CASES 2, 3, 4 AND 10
I'~

I
/-"...-i-

.
.
.
.

-~.
I
/
-j-
1000
0300 MW
2000
3000
5000
4000
111- 102

-------
1--- -
8.00
7.00
z
o
~ 6.00.
~
Z
o
~
~
I UO
a:
...
.~
Q 4.00
:)
....
u
!
~
~
t; 3.00
8
....
~
~ 2.00
o
1.00
-1.00
o
--
   CASE 1 8 
   CASE 5 0 
   ~4 
   CASE 6 'J 
 PLANT FACTGa CI 0.8 ICASE7~- 0 
.  
   100 MW -.-
1   300 MW -...-
  500 MW --....-
.     
 ~    
 . ..._   
   ~ -.. 
    .-
  V  --v'
  . .... 9
 ..~  
~MW b...

...-

. . .
-e.-.- ~

.....
~
o
.1~
2000
3000
-4000
.....--
-6
.~.....--.
'.8.. . .....----
. 3OO.MW -
5000
~
7000
.~~w~~ LO~, TPD
FIGURE 111-44. NET DISPOSAL COST VERSUS WASTE lOAD - CASES 1,5,6 AND 7
III-103

-------
at this point. When the additional costs for refuse preparation were added
in, the case was no longer competitive. Because of discontinuities in the
cost model resulting from the requirement for discrete numbers of steam
generators and turbines, the costs do not always form smooth curves.
In most cases the model indicates that the unit
disposal cost decreases with increasing waste load. Since the cost of
transporting the collected waste to the combined firing plant can add sig-
nificantly to cost, these costs should be considered. Based on a procedure
described in Appendix C, the transportation costs for two cities (New York
and St. Louis) were estimated as a function of waste load; the result is shown
in Figure III-45.
To ascertain the optimum system, the net disposal
plant cost and transportation cost were summed and a net total waste dis-
posal cost for each design configuration was determined. Results from this
procedure indicate that no one particular steam generator design configura-
tion is optimum. The best design is a function of power level and waste
load. For example, Figure III-46 shows the results for Case 2 at 300 MW
using the St. Louis transportation costs. The curve indicates a shallow
optimum between 1000 and 2000 tpd.
The selection of an optimum design configu:r:ation
must be based on the consideration of the amount of refuse to be disposed,
the need for additional power capacity and other factors. The cost analysis
did not indicate that any single design configuration was best over the entire
range of power levels and waste loads. At power capacities between 200 and
500 MW and waste loads between 500 and 2500 tpd, Case 3 appears to be the
best choice. For waste loads in the same range and power capacities of less
than 200 MW, Case 2 appears to be optimum, although Case 4 is also compe-
titive. For waste loads greater than 2500 tpd and power requirements greater
than 200 MW, it would appear that both Case 3 and Case 4 are the best can-
didates. Under certain conditions (not shown in Figures I11-45 and 1II-46),
Case 10 becomes competitive for capacities les s than 150 MW.
Current conventional power plant practice is to pro-
vide approximately 1. 3 kw per capita and to install plants of sufficient capa-
city to provide for a three to five year growth in electrical requirements.
Since the average annual growth in power demand is approximately 7%, this
results in installing plants of approximately 0.3 to O. 5 kw per capita. Using
this criterion and a limit of 500 MW per plant, it is evident that more than
one installation would be required in areas of large population. Using a
generation rate of 5. 5 lb per capita day and the power capacity criteria
stated above, a minimum cost line was determined as shown in Figures III-47
and III-48. These lines reflect the effect of population density, and conse-
quent transport costs, on optimum plant type and cost; they apply respectively
to densities characteristic of New York and St. Louis. The steam generator
design configuration, waste fraction, and installed capacity are also shown
in the figures.
III- 104

-------
H
H
1-1
I
..-
o
U1
6.00
5.00
~ 4.00
~
~.
Z
o
j: 3.00
c(
...
a:
~
z
c(
a:
...
2.00
1.00
...o.op.~
S.S ~~IC"'"
7.S \..~'CI'P'Op.'{
2.5 LB/CAP-OA Y
~---~----------~---

~~------- - - - 5.5LS/CAP-DAY
.....------~~-- -----~~~
~~~- ~--=~
~ -=" ~
7.5 LB/CAP.DAY == -==- -=> -=-= -== ==
LEGEND
ST. LOUIS
NEW YORK
~~
~
o
o
FIGURE 111-45.
4000
5000
PROCESSED WASTE LOA 6000
TRANSPORT A T ~ D, TPD
ION COST AS A FUNCTION OF WASTE LOAD
7000
8000
1000
2000
3000

-------
TOTAL HAUL AND
DISPOSAL COST
5.00
~
--
4.00
z   
0 3.00  
~  
t;'   
0  DISPOSAL COST 
Co)  
-I   
~   
~ 2.00 ----- 
 ---..
o  HAUL COST 
0   
z  - 
c(   
-I   
;:)   
c(   
:r:: 1.00  
o
-1.00
o
1000
2000
3000
4000
5000
REFUSE RATE, TPD
(CASE 2 - 300 MW, ST. lOUIS)
FIGURE 111-46. EFFECT OF HAULING COST ON OPTIMUM PLANT SIZE
III- 106

-------
7.00
6.00
H
~
I
~ 5.00
~
ti
8
-I
~
g 4.00
o
....I.
e(
~
I-
L!J
~ 3.00
......
o
--.I
2.00
1.00
o
-
I
.:-f:::~2:{
CASE 3 AT fw = 0.25
0.5 KW/CAP.
.



../i/"



. . . NOTES:

1/ /..... 1. ST. LOUIS POPULATION DENSITY
I 2. WASTE GENERATION RATE = 5.5 LBI CAP-DAY
I 3. PLANT FACTOR = 0.8
I 4. COST OF COAL= 31 CENTS/106 BTU
100 MW
~;<~

1- . . . --



. '1,
o
.
.
.
1000
2000
.....~""'1
:
.
.
1
LEGEND
CASE 2\ = <:>
I
CASE 3 = -j-

CA~E 4 = m
CASE 10 = ~<
100 MW.
300 MW
-=-:P . a:::z:=:::::m
-... ~
~-...
'/500 MW
/,
.
.
.
><300_[
500 MW
-.....8S118
OPTIMUM COST FOR KW/CAP ~ 0.5
3000 4000 5000
WASTE LOAD PROCESSED, TPD
6000
7000
8000
FIGURE 111-47. OPTIMUM NET TOTAL DISPOSAL COST VERSUS WASTE LOAD (ST. LOUIS AREA)

-------
6.00
5.00
 z
 g 4.00
 ~
 ~
 .J
 ~
H ~ 3.00
H
......
I
...... 0
o .J
00 ct
 ~
 o
 ...
 ...
 ~ 2.00
1.00
o
o
--""',
3000 4000 5000
WASTE LOAD PROCESSED, TPD

FIGURE 111-48. OPTIMUM NET' '( , "Al DISPOSA - C )ST VERSUS WASTE lOAD I :W YO < A :A
CASE 2 +-
fW"".22+-
CASE 3
fw = .25
0.5 KW/CAP
, >l,ooMW


\"4


.


T'"
CASE 4
f.,;'= .58
0.2 KW/CAP
0-.. . . . 8818()
... -

>(~.:w J



.
.
.
r

.


I ~
,,/' I
, ,,/i
-,/ "
I .
.

_!/'

I
.
.
.
NOTES:
1.
NEW YORK POPULATION DENSITY
WASTE GENERATION RATE = 5.5 LB/CAP-DAY
2.
3.
PLANT FACTOR = 0.8 '
COST OF COAL = 31 CENTS/106 BTU
4.
1000
2000
.
.
.
".
>( 500 MW
.
.
.
.
.
OPTIMUM COST FOR
KWICAS' ~ 0.5
LEGEND

CASE 2 = 0
I
CASE 3 = -j-

CASE 4 = [J
'"
CASE 10 = /\
100MW -.
JOOMW -...-

500MW -.....-
6000
8000
7000

-------
Each af the three braken lines in Figures III-47 and
III-48 represent the lacus af minimum cast designs far a given pawer gene-
ratian level (100, 300, and 500 MW). as a functian af the input waste laad.
Hawever, these lines dO' nat reflect the fact that the electrical pawer gene-
ratian capacity and the refuse dispasal capacity must be matched to' the
supparting papUlatian. The salid curve represents the lacus af minimum
cast designs where the waste laad and pawer capacity have been matched
to' supparting papulatian. Far example, in Figure III-47 it is shawn that
while Case 3 is the minimum cast design at 500 MW and 1000 tpd, the
papulatian required to' use 500 MW af additianal electrical capacity wauld
praduce mare than 1000 tpd and as a result this design cambinatian is nat
campatible with the requirement af matching refuse capacity with pawer
capacity. The figure shaws that the best design far a 500 MW plant wauld
be Case 3 with a refuse capacity af 2900 tpd.
Figures III-47 and III-48 shaw that Case 2 is the
optimum design canfiguratian far waste laads less than 1000 tpd, beyand
which Case 3 becames aptimum. In the mare densely papulated case af
New Yark City (illustrated in Figure III-48), Case 4 becames aptimum
after 4000 tpd. The aptimum curves in these figures illustrate the effect
af multiple plants far handling large waste laads.
All af the results presented thus far have been
based an a plant factar af 0.8 and an caal casting 31 cents per millian Btu.
TO' ascertain the effect an cast and design af varying these reference values,
the cast and aptimum design were determined at 0.6 plant factar and at 22
and 36 cents per millian Btu as the cast af caal. The effect af the reduced
plant factar is to' increase cast by appraximately 15 cents per tan at law'
waste fractian (10%) and 40 cents per tan at the high waste fractien (60%).
Hewever, when the eptimum cest is selected based en the limit ef 0.5 kw
per capita plant capacity, this cest increase becemes very significant; e. g.,
at 2000 tpd, the aptimum cest at o. 8 plant facter is $2.35 per ten whereas
at 0.6 plant factar the eptimum cest is $3. 80 per ten, beth using cest at
-31 cents per millian Btu. This cenditian is caused by net being able to' use
Case 3 design fer the o. 6 plant facter because ef the limit en the waste
fractien af 25%; i. e., this limit daes net permit a sufficiently high pewer
level at this waste lead canditien. The censequence then ef a 0.6 plant
facter is that Case 10 er Case 4 weuld be the eptimum design, ever the
1000 to' 6000 tpd range, with the net dispesal cest varying frem $5. 10 to'
$2. 50 per ten. .
The unit cest ef ceal influences the credit allewed
fer pawer generatien. In additien, the unit cest ef ceal affects the tetal
cast af the waste burning plant, altheugh net as significantly.
The effect en net dispesal cest ef varying the unit
cest ef caal is influenced by the waste fractian and pewer level, as shewn
in the table belaw:
III-I09

-------
........
Refuse Power Unit Cost of Coal - ~/106 Btu
Fraction Level, 22 31 36
(fw), 0;0 MW  Net Disposal Cost, $~
10 100 8.47 9.30 9.75
20 100 6.08 6.15 6. 19
10 500 1. 17 1. 12 1. 08
25 500 2.56 2. 14 1. 90
60 500 4.03 3.64 3.42
At power leve~s in excess of approximately 100 MW,
the net disposal cost decreases with increasing unit coal cost regardles s of
the waste fraction. This is because the cost of the conventional power plant
increases significantly with increasing coal cost at the higher power levels
(see Figure C-17 of Appendix C) and hence there is a substantial savings in
substituting waste for coal. At low power levels, the waste fraction must
be in excess of 200;0 before increasing unit coal costs result in decreasing
disposal cost. This is because, at low power levels, the conventional plant
is not affected as greatly because of the relatively high fixed costs, and only
after a substantial amount of coal is replaced by waste is the cost savings
achieved.
The cost model was also used to investigate the
sensitivity to several additional assumptions.
The first assumption investigated was the heating
values of 12,020 Btu/lb and 4460 Btu/lb for coal and solid waste, respec-
tively. Other (arbitrarily selected) combinations of heating values for both
fuels were therefore analyzed. The purpose was merely to establish the
magnitude of the effect changes in these parameters (and the consequent
changes in firing rates) would have on disposal costs. The following tabu-
lation indicates the I results:
 Solid Was te Coal Net
 Heating Value, Heating Value, Disposal Cost,
 Btu/Lb Btu/ Lb $/Ton
Reference 4460 12,020 2.21
 5000 11, 000 2.13
 4000 13, 000 2.32
 5000 13,000 2. 14
III- 11 0

-------
It is evident that the net disposal cost is not sensitive to changes in the
heating value of coal and that a change in solid waste heating value causes
a small variation in the net disposal cost.
Another assumption that was evaluated was the rate
of return on capital. The following table shows the effect on net disposal
cost by using an 8% rate of return compared to 7%:
Rate of Return,
%
Net Disposal Cost,
$/Ton
Reference
7
8
2.21
2.64
It is apparent from the above tabulation that the cost
of disposal is very sensitive to the rate of return on capital.
d.
Cases Selected for Optimization
From the foregoing discussion it is clear that, de-
pending on the waste load and installed capacity, three steam generator
designs offer the optimum configurations. These are Cases 2, 3, and 4.
As pointed out, Case 2 would be preferred only for low waste loads, thus
making it a questionable choice for design optimization. It is also evident
that even in the few instances where Case 4 is competitive, it is not signi-
ficantly better than Case 3. Thus the.1atter emerges as the favored candidate.
Because Case 3 is optimum in the range between 300 and 500 MW, the size
selected for design optimization was set at 400 MW.
As discussed in the section immediately following,
a second candidate, of the suspension-firing type, was also selected for
design optimization. A major cost item in suspension firing is the cost of
shredding the ref\lse. For example, the cost model predicts shredding costs
for Case 10 of approximately $2. 65 to $2. 30 per ton over the range of 100 to
500 MW. Because, with future experience, this particular cost factor may
dec rease and potential variation in other cost data occur, it was deemed ad-
visable to include a candidate of the suspension-firing type.
Within the suspension-fired 'clas s of steam generators
(Cases 5 through 10), Case 10 clearly emerges as the favored configuration.
. Based on Figures III-47 and III-48, a size of about 150 MW would appear op-
timum for this case, although the distinction is not sharp. The actual capacity
selected was 100 MW. This was done to provide a design for the lower capa-
city range, Case 3 (400 MW) already fulfilling the larger size requirements.
In the following section the design optimization of these two candidates is dis-
cussed. The resulting systems are then again costed to take into account the
more detailed design information that is developed.
III-Ill

-------
2.
Optirnization of Selected New Plant Designs
a.
Introduction
. An initial objective of the present systems analysis
was the development of preliminary engineering designs of (1) a new plant
for combination firing of fossil fuel and refuse, and (2) a new power gene-
rating incinerator with fossil fuel used as an auxiliary fuel. In the first
type plant, the heat input from fos sil fuel and refuse would logically fall
within some range of fuel combinations determined by refuse availability,
power requirement, and other extrinsic factors. In the latter type plant,
the fossil fuel would, at most, constitute a minor portion of the heat input
and, pos sibly, even be used only on a stand-by basis.
During the study, however, the use of more than
60% heat input from refuse was found to be inadvisable. In the case of
boilers equipped with agitating grates, unacceptable steam-flow variations
would certainly occur at high refuse rates. Regardles s of the mode of firing,
however, requisite plant steam conditions and layout would also be far from
optimal. As shown in Figure III-49, a typical map of fuel requirements for
different plant sizes and steam conditions~:<, a refuse heat input above 60%
would necessitate an 850 psig non-reheat cycle and the use of an excessive
number of steam generators. The Figure llI-49 fuel map incorporates the
limitation that any given plant would handle no more than 8, 000 tpd of refuse.
This was done primarily in consideration of refuse collection logistics and
the projected refuse production rates of the six LMA's given in Section II, A.
In consideration of these design constraints, both
combihed and auxiliary fossil-fuel firing, though not clearly distinguishable,
. were considered to be achievable, in a single system.
The type-categorization was also structured so that
it would be addressed to advanced-design features. Approached from this
viewpoint, the goal became one of developing designs for:
.
a system in which the extant combined-
firing practices would be optimized;
.
an optimum system based on steam
generator configurations not yet tested
with refuse fuel.
~:' A second map for isolated-flue systems could also be included, but would
only lead to the same conclusions developed here.
III- 112

-------
  (J! ~
  It) 0
  -
 5.0  
 4.5  
 4.0  
 3.5  
a   
a.   
~ 3.0  
Jf)O   
~   
LaJ 2.5  
I--  

-------
In terms of the ten candidate systems, an obvious
dichotomy emerged. The first four cases essentially fit the first category,
while cases 5 to lO belong to the second. In a loose sense, this dichotomy
can be said to differentiate between the (tried) agitating grate type systems
and the (untried) suspension-fired systems.
Based on the results of the previously described
cost modeling, the configurations and plant sizes favored for preliminary
engineering design development are Case 3 - Separately Fired Economizer
(400 MW) and Case 10 - Arch Furnace With Separate Superheater (100 MW).
As shown in the systems analysis, maximum heat
input for the Case 3 design is about 25% refuse. The limitation is basically
due to the steam cycle and not the firing mechanism (agitating grate). A
plant of this design could be designed for almost any heat input from refuse
between 0 and 25%. As will be shown later, it is possible for a plant de-
signed for 25% heat input from refuse to operate at intermediate refuse
rates. The fuel ratio for Case 10 is also determined by the steam cycle
and has been set at about 50% heat input for each of the fuels. The basic
steam generator design data are given in Section III, B.
b.
Separately Fired Economizer
(1 )
Preliminary Engineering Technical Factors
As in the original conception of this design,
feedwater would be taken from the last stage feedwater heater, be further
heated in a refuse-fired economizer, and then be evaporated, superheated
and reheated in a conventional fossil-fuel fired steam generator as shown
in Figure III-50. For a 400 MW plant a 2400 psig steam cycle would be
used (see Figure III-49) wherein three parallel, separately fired economizers
would be connected in series to a single coal-fired steam generator and one
turbine-generator set. It is recalled that this concept was originally used
at Munich South, Unit 6 (Ref. 103). A map of the fuel requirements, as
determined in the system analysis, is shown in Figure III- 51. The boundary
and conditions have been previously described.
(2)
Economizer
Feedwater is heated to slightly below the
saturation temperature corresponding to the feedwater pressure. It is
recalled that the tube metal temperatures are below the threshold where
high temperature fouling and corrosion due to refuse might be anticipated.
Low temperature corrosion is avoided by venting the flue gases at a tem-
perature of 5750F, considered to be above any acid dewpoint. This tem-
perature was selected on the basis of the heating surface available for
cooling the flue gas.
III-114

-------
COAL-FIRED
STEAM
GENERATOR
S

I
..-
..-
U1
REFUSE
FIRED

ECONOMIZERS

+
TURBO - GEN.ERATOR
CONDENSER
FEEDWATER HEATERS
FIGURE III-50. STEAM CYCLE SCHEMATIC - SEPARATELY FIRED ECONOMIZER

-------
c::  0  
:z:  CL  
~  ~  
~ I'b  
"'0    ~~
 5   ~
400   Jt..o ~
  ~~ 
   ~4; 
   ~ 
   '<~ 
 4  ~ 
300  / 
2
w
.-
«
a:
...J
«
o
(,)
200
'400MW PREll Mil" . RY
DESIGN POINT
100
NET PLANT HEAT RATE = 10,130 BTU/KW-HR
o
o
o
2
REFUSE RATE, 103 TPD
8 NO. OF REFUSE-FIRED ECONOMIZERS
3
4
FIGURE III-51. PLANT FUEL REQUIREMENTS WITH SEPARATELY FIRED ECONOMIZERS
III - 11 6

-------
In designing a furnace chamber for combus-
tion of any fuel, careful consideration must be given to the furnace exit-
temperature. This temperature should be high enough to combust all
volatile material, and, in the case of refuse, render the effluent gas as
odorless as possible. The furnace exit-temperature should not exceed
the ash fusion temperature, in that the subsequent convection surfaces
/
would be subjected to fouling from molten ash. Based on domestic ex-
perience, such as that obtained with the unit at the U. S. Naval Station
at Norfolk, Virginia, and that reflected in the TtJV performance reports,
a furnace exit-temperature of l8000F was selected.
The size of any given furnace is related to
the furnace exit-temperature and heat absorption. Theoretical data sup-
ported by field experience are generally used in sizing the furnace for a
specific fuel. In this case, the data were derived from both domestic and
foreign experience.
Because of the relatively high ash content of
refuse, vertical tube banks should be used wherever possible. In addition,
tubular air heaters should be used in:.lieu of the more compact, regenerative
type to minimize pluggage.
(3 )
Steam Generator
Pulverized coal would be fired in a dry-ash-
bottom steam generator. In keeping with domestic practice for units of the
present size, this would be a natural circulation type unit. The only differ-
ence between the present and a conventional unit design is that most of the
econonUzer surface has been omitted.
It is claimed that in the Munich South (Unit
No.6) plant, the main:consideration is a high degree of availability for
power generation. If, for any reason, refuse delivery is stopped, full
power generation capacity can be maintained by burning more fos sil fuel,
although at reduced efficiency. During the present system reanalysis, this
claim was examined and found to be perplexing. In the present, relatea
design, if refuse is not burned, the full load capacity is only slightly in
excess of 300 MW as compared with the normal capacity of 400 MW. A
more detailed explanation of the basic design difficulty follows.
On a weight basis, the feedwater flow in the
parallel refuse-fired economizers is the same as that of the feedwater en-
tering the steam generator and steam leaving the superheater. When refuse
is not burned, the feedwater is brought to the steam generator at a lower
temperature. For the same steam flow, therefore, the heat absorption
in the steam generator must be increased to equal the heat normally ab-
sorbed by burning refuse. To offset this deficiency by merely firing more
coal is not an adequate solution. In the design of the furnace chamber, the
Ill- 11 7

-------
flue gas exit-temperature essentially determines heat absorption. If the
furnace exit-temperature is excessively increased, fouling of convention
sections may result and it would be difficult to control superheat tempera-
ture. In addition, the prevention of these unwanted effects must be observed
over a range of loads, because as the load varies, the steam flow varies.
In the present case, where refuse-firing is interrupted, the load of the
steam generator will still vary but not the steam flow. The problem is
analogous to that wherein a feedwater heater is lost from service. Typi-
cally, the solution to this operating situation is to reduce load, i. e. ,
steam flow and power generation. Fuel flow is also decreased, but to a
less than proportionate degree. The net effect is that cycle efficiency is
lowered.
Several solutions were considered. Initially,
some economizer surface was included in the fossil fuel steam generator.
In the flue-gas temperature range involved, the location of this surface would
be such that it would be rather ineffective under normal firing (i. e., refuse
burning) conditions. On the other hand, if refuse were not burned, the flue
gas temperatures over this surface would increase and more effective heat
transfer would result, but it would be insufficient to overcome the thermal
deficiency due to curtailed refuse burning. The addition of this surface does,
however, represent a partial solution to the problem; it was therefore re-
tained in the design. An additional benefit provided by this component is
that the streams from the three economizers would be mixed therein so
that flow and, more importantly, temperature fluctuations would be dam-
pened before the fluids entered the steam drum.
To complete the resolution of the problem,
it was further decided to include stand-by firing systems in the economizer
units. Although the steam generator would be coal-fired, the stand-by equip-
ment should preferably be for either oil or natural gas. It would be much
more economical to do this than to expand the capacity of the (more expen-
sive) coal handling system for a consumption rate that would probably be
only infrequently encountered.
. With the back-up firing system, it would be
pos sible to operate the economizer units in either the combined or single
fuel firing modes. In the event of a reduction in refuse availability, it
would, however, be simpler to switch one unit at a time completely from
refuse to fossil fuel, than to fire any or all of the three economizer units
with mixed fuels.
Several other alternatives for compensating
for possible refuse input reduction or interruption were considered and re-
jected. As suggested above, simply increasing the fuel rate in the steam
generator would increase the furnace exit-temperature, thus neces sitating
the use of great amounts of spray cooling to bring the superheat and reheat
III-lIB

-------
temperatures to the desired control levels . Potential fouling of convective
surfaces would also be a problem. Another solution would be to size the
steam generator to accommodate an outage of all three economizer units.
This would not only be quite costly but impractical. With all three econo-
mizers on line, the oversized steam generator would have to be turned
down to the point that the final superheat outlet temperature would become
unacceptable. Adopting a compromise furnace size still presented the
disadvantages of the two choices of possible sizes (for normal operation
and operation without economizers), while furnishing no major advantages.
Another solution would be to install a conven- .
tional economizer in the steam generator and use by-pass flues to control
heat-input variations caused by shutdowns of the external economizer units.
Such an arrangement was rejected as being excessively costly.
A final alternative that was considered was to
increase the bleeds from the various turbine stages in order to raise the
temperature of the feedwater entering the steam generator. This could be
accomplished by incorporating additional heat-transfer surface in the feed-
water heaters. This arrangement would reduce steam flow through the
turbine and thus decrease the power output. Unfortunately, these bleeds
are usually optimized for a given set of normal operating conditions. It is .
also likely that the optimization of a turbine-feedwater heating cycle for
two sets of conditions would be uneconomical. In any case, it is difficult
to achieve a better cycle efficiency by transferring the heat input of the
Rankine cycle from regenerative heating to that derived from fuel com-
bus tion.
The foregoing discussion has perhaps been
directed to a contingency situation that may suggest an overly cautious de-
sign approach. The refuse-fired economizers have actually been designed
for an availability comparable to that of the coal-fired steam generator.
The inherent multiplicity of three economizer units does offer an opportunity
to repair any single economizer without suffering a severe power reduction.
Furthermore, there would appear to be no greater chance of a decreased
availability of refuse occurring than there would be of fossil fuel. Thus,
in the preliminary design presented here, the fossil-fuel fired steam gene-
rator has been sized and cost-estimated on the basis of its normal fossil-
fuel rate, which is approximately 300 MW. The stand-by arrangement
wherein fos sil-fuel firing equipment would be included in the refuse-fired
economizers is considered to be an optional feature and, therefore, has
not actually been inc~uded in either the design drawings or cost estimate.
It is perhaps appropriate at this point to com-
pare the basic design considered here with.that at Munich South, Unit No.6.
The fossil fuel steam generator at Munich South is a once-through (Benson)
design, single (flue gas) pass, and of 124 MW capacity. A once-through unit
Ill- 11 9

-------
does offer somewhat more liberty in furnace design than does a natural
circulation unit, especially if the European 'Imeandering" circuitry is
employed (see Appendix B, Section Ill, A, 2). It is rare, however, to find
a plant of under 400 MW capacity designed for once-through operation. In
the United States, the break-even point between once-through and natural
circulation units had been about 500 MW. In recent years, however, this
has apparently increased to about 800 MW.
Concerning the single (flue gas) pas s design
of the Munich unit wherein both a superheater and reheater are involved,
steam temperature is controlled by sprays installed in several locations.
As suggested earlier, however, resorting to a great deal of spraying is
not an economical practice for domestic applications.
(4)
Performance Calculations
The calculation of the products of combustion
dis cus sed earlier (Table Ill-I) was repeated. This time, however, the calcu-
lations were expanded to include coal, as well as refuse, and to derive values
for a range of excess air levels. The latter was done to permit the evaluation
of system characteristics at lower steam generation loads. Generally, at
lower loads, it is neces sary to increase exces s air to maintain safe flue- gas
temperatures in the convection pas ses.
These data are presented in Table III-IS. It
will be noted that at 50% excess air, the refuse-derived values are slightly
lower than shown in Table Ill-I. This is largely because the correction for
unburnt carbon had been omitted in the earlier derivation as being of minor
importance. Other small differences are also involved. The specific humi-
dity of the combustion air, for example, has been increased to 0.013 lbs of
water per pound of dry air.
The ultimate analyses employed were, of
course, identical with those shown in Section Ill, B for refuse and coal. The
theoretical oxygen requirement used in calculating Table Ill-IS was O. 718
lb/lb of refuse and 2. 101 lb/lb of coal. The unburnt carbon was assumed
to be 1. 85% and 0.66% for the refuse and coal, respectively.
Efficiency calculations for coal firing (18%
exces s air) at various flue gas exit-temperatures are shown in Table III-16.
The procedure used is in accordance with the ASME Power Test Code, and
was explained earlier in Table III- 2. It will be noted that the radiation los s
has been isolated from the subtotal in Table III-16. This was done to facili-
tate the use of the table with different sized units; radiation loss is the only
factor that varies with capacity. Coal firing efficiency is also plotted in
Figure Ill-52 as a function of flue gas temperature. It can be seen that a
decrease of 400F in the flue gas exit temperature results in a gain in ef-
ficiency of 1 %.
III-l20

-------
TABLE lIT-IS
PRODUCTS OF COMBUSTION AT VARIOUS EXCESS AIR LEVELS
          E XCE5S AIR %   
       Refuse    Coal 
     0 25 50(1) 75 100 150 18(1) 23 40
 I. Quantities, Ib/lb of Fuel          
  Combustion Air (dry) 3. 10 3.87 4.65 5.42 6.20 7.75 10.71 11. 17 12.71
  C02 Formed     0.913    2.42 
  S02 Formed     0.002    0.74 
  N2 in Flue Gas  2.39 2.98 3.58 4. 17 4.77 5.96 8.24 8.59 9.77
  Dry Flue Gas  3.30 4.08 4.85 5.63 6.40 7.95 11.12 11. 57 13.11
  H20 in Comb. Air 0.04 0.05 0.06 0.07 0.08 O. 10 O. 14 . O. 15 O. 17
H  H20 in Fuel     0.271    0.085 
I::         
I  H20 formed by Comb.    0.304    0.420 
.....           
N  Total Flue Gas H20 O. 615 0.625 O. 635 0.645 O. 655 0.676 0.645 0.651 0.672
..... 
   (wt-% in Wet FG) (15. 7) ( 13. 3) (11.6) (10.3) (9.3) (7.8) (5.5) (5. 3) (4. 9)
  Wet Flue Gas  3.92 4.72 5.49 6.27 7.06 8.63 11.76 12. 22 13.78
 II. Flue Gas Composition, Vol-%          
  C02 (2)  14.8 12.4 10.7 9.4 8.3 6.8 13.9 13.3 11.8
  502 (2)  O. 02 0.02 0.02 0.01 0.01 0.01 0.29 0.28 0.25
  H20 (2)  0.24 0.21 O. 18 o. 16 O. 15 O. 12 0.09 0.09 0.08
  02 (3)  0 4.2 7.0 9.0 10.5 12.6 3.3 4.0 6. 1
 Ill. Flue Gas Average 1>.lo1ecular          
  Weight   28. 03 28. 15 28. 23 28. 29 28. 34 28.41 29.65 29. 62 29. 51
 ( 1) Design Value           
 (2) Wet Flue Gas Basis           
 (3) Dry Flue Gas Basis           

-------
,- ~ - -- - -
   TABLE III-16     
 EFFICIENCY OF COAL-FIRED STEAM GENERATOR   
    6     
 (Duty = 2,635 x 10 Btu/hr)    
 Temperature of Gas Leaving Air Heater, of 250 300 325 350 400 450 500
 Heat Losses, %        
 Dry Gas  3078 4.89 5045 6000 7. 12 8022 9. 33
 Hydrogen and Moisture in Fuel 4052 4.61 4.66 4070 4.80 4.89 4. 98
 Moisture in Air  0.09 O. 12 O. 13 O. 15 O. 17 0.20 0.23
 Unburned Combustible  0066 0.66 0066 0.66 0066 0.66 0.66
S Unaccounted For & Mfrs Margin 1. 50 1. 50 1. 50 1. 50 1. 50 1. 50 1. 50
I         
......         
N Subtotal  10.55 11. 78 12. 40 13. 01 14. 25 15. 47 16.70
N  
 Radiation  O. 18 O. 18 O. 18 O. 18 O. 18 O. 18 O. 18
 Total Losses, %  10. 73 110 96 12. 58 13. 19 14. 43 150 65 16. 88
 Efficiency, %  89. 27 880 04 870 42 86. 81 85. 57 84. 35 83. 12

-------
ff.
.;
o
z
w
o
Lt..
Lt..
LLJ
0::
o
....
«
0::
w
Z
w
(!)
:E
«
w
t-
CJ)
90
I
1%:
I

1- - - - - - -
40° F
CONDI TIONS
6
DUTY = 2,625 X 10 BTU/HR

COAL HEATING VALUE = 12,022
BTU/LB
85
80
250
300
500
400
350
450
!EMPERATURE OF FLUE GAS LEAVING AIR HEATER,oF
FIGURE III-52. EFFECT OF FLUE GAS EXIT TEMPERATURE ON STEAM GENERATOR EFFICIENCY
III-123

-------
I"
The efficiency for refuse firing is given in
Tables Ill-17, -18, and -19. The calculations were done at two different
flue gas exit-temperatures for various excess air levels and also at various
flue gas temperatures for the design (50%) excess air level. In presenting
the data, the moisture in the fuel is reported separately from the water de-
veloped by the combustion of hydrogen in the fuel because of the relatively
large magnitude of each of these losses. The heat losses associated with
the moisture in the combustion air, unburned combustibles, and sensible
heat in the ash have been combined because each is relatively small and
none are appreciably influenced by the exit-temperature of the flue gas or
the level of excess air. The radiation loss is treated as in the previous.
table. The data in Tables Ill-17, -18, and -19 have been combined in
Figure Ill-53, which, on the basis of flue gas exit-temperature and excess
air level, thus constitutes a rather complete map of efficiencies.
(5)
Design Details
The design of the refuse -fired economizers
and fossil fuel fired stearn generator were based on the standardized pro-
cedures of the Foster Wheeler Corp. This included the use of computerized
calculations. Cost estimates for material, shop fabrication and field erec-
tion were provided by Foster Wheeler1s commercial departments. The re-
sults of this effort, namely the drawings and cost estimates, generally agreed
closely enough with those developed for the candidate evaluation stage that an
iterative analysis of the systems proved unnecessary.
The design, as developed, is quite amenable
to load variations when the relative quantities of refuse and coal, i. e., 25%
and 75% heat input respectively, are held constant. However, as the load
varies, especially below 75% of full power generation, a variation in refuse
rates between 0 and 25% of the heat input is also permis sable. The summary
of the duties for the three economizers and the single stearn generator is as
follows:
THERMAL FACTORS FOR SEPARATELY
FIRED ECONOMIZER PLANT (400 MW)
  Duty, Efficiency, Heat Input 
Sy stern Fuel 106 Btu/hr % 106 Btu/hr 2-
Economizers Refuse 687 66.22 1037.3 25.6
Stearn Gen. Coal 2633 87.42 3014.2 74.4
Combined  3320 81. 93 4051. 5 100.0
Each refuse -fired economizer would burn 930
tpd of fuel to establish a maximum plant rate of 2790 tpd of refuse. The stearn
generator would burn coal at a rate of 3008 tpd. The net plant heat rate would
be 10,130 Btu/kw-hr.
IIl- 1 24

-------
   TABLE III-17    
 EFFICIENCY OF REFUSE-FIRED ECONOMIZER A T VARIOUS EXCESS AIR LEVELS
  (Flue Gas Exit Temp. 0 Duty per Unit = 228 x 106 Btu/hr) 
  = 575 F; 
 Excess Air, %  0 25 50 75 100 150
 Heat Losses, %        
 Dry Gas  8.79 10.87 12.92 15.00 17.04 21. 17
 Moisture in Fuel 7.74 7.74 7.74 7.74 7.74 7.74
 Moisture From Hydrogen 8.74 8.74 8.74 8.74 8.74 8.74
  * 2.50 2.50 2.50 2.50 2.50 .2.50
 Misc Small Losses
a UnaccoW7-ted For 0.50 0.50 0.50 0.50 0.50 0.50
I         
...... Mirs Margin  1.00 1. 00 1. 00 1. 00 1.00 1. 00
N 
\J1         
 Subtotal  29.27 31. 35 33.40 35.48 37.52 41. 65
 Radiation  0.38 0.38 0.38 0.38 0.38 0.38
 Total Losses, %  29.65 31.73 33.78 35.86 37.90 42.03
 Efficiency, %  70.35 68.27 66.22 64. 14 62. 10 57.97
 *        
 Moisture in air, unburned combustibles, and sensible heat in ash   

-------
TAB LE III - 1 8
EFFICIENCY OF REFUSE-FIRED ECONOMIZER AT VARIOUS EXC,ESS AIR LEVELS
(Flue Gas Exit Temp. = 450oF; Duty per Unit = 228 x 106 Btu/hr)
 Excess Air, %   0 25 50 75 100 150
 Heat Losses, %        
 Dry Gas     6.57 8.12 9.67 11.17 12.74 15.83
 Moisture in Fuel  7.38 7.38 7.38 7.38 7.38 7.38
 Moisture From Hydrogen 8.34 8.34 8.34 8.34 8.34 8.34
    ,I,      
E Misc Small Losses '.' 2. 50 2.50 2.50 2. 50 2. 50 2. 50
I           
-      0".50 0.50 0.50 0.50 0.50 0.50
N Unaccounted For
O'           
 Mfrs Margin   1. 00 1. 00 1. 00 1. 00 1. 00 1. 00
 Subtotal     26.29 27.84 29.39 30.89 32.46 35. 55
 Radiation    0.38 0.38 0.38 0.38 0.38 0.38
 Total Losses, %   26. 67 28.22 29.77 31.27 32.84 35.93
 Efficiency, %    73.33 71. 78 70.23 68.73 67. 16 64.07
 ,"          
 "'Moisture in air, unburned combustibles, and sensible heat in ash   

-------
    TABLE IIl-19     
 EFFICIENCY OF REFUSE-FIRED ECONOMIZER AT VARIOUS FLUE GAS EXIT-TEMPERATURES
 (Excess Air = 50%; Duty per Unit = 228 x 106 Btu/hr)   
 Gas Temp. Leaving Air Heater, % 300 350 400 450 500 550 575 600
 Heat Losses, %         
 Dry Gas  5.74 7.00 8.34 9.66 10.94 12. 26 12.92 13. 57
 Moisture in Fuel  6. 96 7. 10 7.24 7.40 7.52 7.66 7.74 7.80
 Moisture From Hydrogen  7.85 8.01 8. 17 8.33 8.48 8.64 8.74 8.80
 ~(         
 Misc Small Losses  2.50 2.50 2.50 2.50 2.50 2.50 2.50 2.50
...... Unaccounted For  0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50
~          
I          
...... Mfrs Margin  1. 00 1. 00 1. 00 1. 00 1. 00 1.00 1. 00 1.00
N         
--.J          
 Subtotal  24.55 26. 11 27.75 29.39 30.94 32.56 33.40 34.17
 Radiation  0.38 0.38 0.38 0.38 0.38 0.38 0.38 0.38
 Total Losses, %  24.93 26.49 28. 13 29.77 31.32 32.94 33.78 34.55
 Efficiency, %  75. 07 73. 51 71.87 70.23 68.68 67.06 66.22 65.45
....
....
Moisture in air, unburned combustibles, and sensible heat in ash

-------
 70
~ 
0 
.; 
0 
z 
w 
- 
0 
LL 
LL 
W 
0:: 
W 
N 
:!: 
0 
z 
0 
0 
w 
 65
75
~-tC'
~~~
o ~ -4/ ~
o
CONDITIONS
6
DUTY = 228 X i 0 BTU/HR

REFUSE HEAT IN G VALUE = 4460 BTU/LB
60
300
350 400 450 500 550
TEMPERATURE OF FLUE GAS LEAVING AIR HEATER,oF
600
FIGURE III-53. EFFECT OF FLUE GAS EXIT TEMPERATURE ON EFFICIENCY OF REFUSE-FIRED ECONOMiZE'
Ill-128

-------
(a)
EcononUzer
A drawing of the economizer unit is
given in Figure III-54 and the water circuitry is shown schematically in
Figure III-55. The summary performance data in presented in Table
III-20.
Refuse would be fed through a vertical,
water-cooled chute and burn on a thick fuel bed. The grate incorporated
should furnish both agitation and tumbling to the fuel mas s to insure good
burnout. High velocity, secondary air nozzles would be provided in the front
and rear walls to promote complete combustion of volatile gases and particles
rising from the fuel bed. The configuration of the combustion zone was in-
fluenced by both domestic and more recent European experience (Refs. 50
and 104). All walls and the roof would be of welded tube-and-fin construction.
Tube banks, especially in areas of relatively high gas temperatures, would be
arrayed vertically. Horizontal tube banks would be of bare tube design in all
cases. A tubular air heater, in which the flue gas would be directed down-
ward inside the tubes, would be used because of its ease of cleaning. Ash
hoppers would be appropriately located to remove ash where tube banks might
act as ash deflectors.
Feedwater would be flowed in a single
continuous (once-through) path. Flue gas would be directed in a two-pass
arrangement and be discharged into a dust collector located at grade level.
Air, preheated to 3l6°F, would be delivered as underfire air. This tem-
perature was selected as being compatible with the cast iron' grate. Ap-
proximately 25% of the preheated air would be sent through a booster fan
and delivered as high velocity secondary air. '
The water wall panels would consist of
3-in. OD tubes spaced on 3-3/4-in. centers with fins continuously welded
between tubes. Because of the all-metal construction, slag adhesion should
be minimal. Gas-borne, molten slag-particles would be cooled upon contac-
ting the tube or fin and thus tend to shed from the surface. The solid walls
would be impervious to gas penetration, so that a costly refractory setting
would be unnecessary. In the design of the rear wall of the furnace, a "nose"
has been'incorporated at the furnace exit to insure good gas distribution.
This wall would also form a three-row deep slag screen. The screen would
be arrayed with a longitudinal spacing of 5 -in. and a transverse spacing of
11-1/4-in. The boiler bank design consists of 2-1 /2-in. OD tubes, three
rows deep on 7 -1 /2-in. spacing, and thirty-one elements across on II-in.
spacing, arrayed in an in-line configuration. The horizontal tubes are 3-1/2-
in. OD tubes, which would also be in-line on 5 x 5 -in. centers. The loops
would be supported from the front and rear panel-walls of the second pass.
Ample space has been provided in the design for sootblowers. The air heater
would consist of 1500 12 ft-long, 2-1/2-in. OD tubes arranged in a 4-1 /2-in.
spaced, in-line pattern. On the air-side, the gas flow would follow a three-
pass, cross-flow path.
III-129

-------
EL 0' 0"
""~llfl. "'~"l'
-;: I'~.
!I
J ::: II J
i bi.'

;. -
LJ7."~m ,!....,;"t...;;., uk-mm

SECTION "A-A"
EL.119'-6"
I.". '
r,::: I
i!J
-I"A"
UPPER FRONT
WALL HEADER
1 1':.
I 'III Ii
,!: 17' 0"
I :1 FURNACE
I ' DEPTH


J
/
29'0"
FURNACE
WIDTH

ECON
INLET /
HEADER
" I

'-AIR HEATER
-~ P'I:.

OVERFIRE r1','
F D FANS t",~ ~ I~i, ,!
I" . ..'-JbJ~L,
~/
F.D. FANS
FIGURE III-54. SEPARATELY FIRED ECONOMIZER
III-130

-------
FEEDWATER HEATER
OUTLET
 -
 -
VERTICAL BANK 
FURNACE 
LEFT 51 DE WALL 
FURNACE 
fRONT WALL 
FURNACE 
RIGHT SIDE WALL 
FURNACE 
REAR WALL 
CONVECT I ON PASS
REAR WALL 
CONVECTION PASS
LEFT SI DE WALL 
CONVECTION PASS
FRONT WALL 
CONVECTION PASS
RIGHT SI DE WALL 
HORIZONTAL 
CONVECTION BANKS
+ 
STEAM GENERATOR INLET
FIGURE III-55. FLOW DIAGRAM - REFUSE FIRED ECONOMIZER
IIl- 1 3 1

-------
TABLE 1lI- 20
SUMMARY PERFORMANCE OF REFUSE-FIRED ECONOMIZER
Refuse fired, tpd


Combustion air input (50% exces s), '103 lb/hr


Wet flue gas produced, 103 lb/hr
930
360
425
Economizer e'fficiency, %
3
Feedwater output, 10 lb/hr
66. 2
Pressure at feedwater outlet, psig
933
2600
o
Temperature, F
Feedwater outlet
657
Feed entering unit
470
Air entering unit
80
Air leaving air heater
316
Flue gas leaving furnace
1800
Flue gas entering air heater
766
Flue gas leaving air heater
575
IIl-l32

-------
(b)
Steam Generator
The drawing of the coal-fired steam
generator is shown in Figure Ill- 56 and the water / steam circuitry is shown
schematically in Figure III-57. The summary performance data is given in
Table III-21. In most respects, the design of this unit is conventional. The
main difference is in the design layout of the heating surface. The heat ab-
sorbed in this unit would be largely accomplished by the superheater and
reheater because of the use of the refuse-fired economizers. As mentioned
earlier, however, some economizer surface would be included in the steam
generator. The small section of economizer surface shown in Figure III-56
is situated under the convection superheater. In conventional units a small
section of total economizer surface is usually located under the convection
superheater, and the -remainder is located to follow the parallel pas s.
'-
v
The furnace would have panel walls con-
sisting of 3-in. OD tubes on 3-3/4-in. centers with continuous fins welded
between the tubes. In the upper furnace, five "wing" division walls would
comprise a radiant superheater incorporating 2-in. OD tangent tubes. A
parallel pass arrangement would be used in the second pass. Superheat
temperature would be controlled by the firing rate and by spraying. Re-
heat temperature would be controlled by regulating the gas-flow with
dampers.
(6 )
Cost Estimates
Following the development of preliminary
designs just discussed, detailed costs were derived for the optimum system.
This was done according to the procedure described in Section III, C, 1. A
notable variation, however, was that the costs for FPC Code No. 312 (Boiler
Plant Equipment) were generated by commercial estimators of: the Foster
Wheeler Corporation.
From the design calculations and drawings, a
complete materials list was first prepared. The units were then designated
for shop fabrication vs outside production scheduling. Shop costs included
material, labor, and overhead. Included in overall costs were engineering
and normal overhead expenses and profit. Field erection costs were derived
using a straight time basis and an average labor rate of $10. OO/hr. Freight
charges were based on an average location.
The results of this more detailed costing are
shown in Table III-22. In comparing these costs with those presented in
Table III-13 of the cost model, it can be seen that a significant reduction
in refuse disposal cost has resulted. To a large extent, this resulted from
a design decision made during the optimization study.
III-133

-------
~>rr1--"~";J
I ' ~
c(
1.11
:t
~ I
+
1--- -~
I
I: I
I i
I j - ! BURNERS 0
Tw.l- ~:~i

, ' MILLS - - .
! !- - _.~. i

-, I .

i ! -
j,--- --- -4-----1
I

I--J
! i
-,
I
i
i i-- :
I --, ----r
I
I
i
a:
<
\.
,
,
! ~CWSTEE1..a. 1....."
11
:1:
',I
,
--_._-
~
0,
01
II
I
I
SUPERHEATER OUTLET
Cc
~

Z
o
t
1.11
ell
Z
:5
A-
I&.
...
c(
:t
"A"
,
, REHEATER INLET
t

"A'.
-t
- r--
tAIR HEATER-
I
I
I

-\
I II: '
-,
Ii:
)1 " :'1
I'
:  
! !
i I i GRADE EL 0'.0"
~ MILLS
ili
jij
Ii!
jij
107'.0"
PARTIAL Sl£CTIONAL SIDE CLEVATION
FIGURE 111-66. COAL FIRED STEAM GENERATOR
III-134

-------
FEEDWATER
HEATER
   - I
   -
 CONVECTION  
 REHEATER  
- HI GH PR,ESSURE  
.. TURBI NE  
 PENDANT  
 SUPERHEATER  
 RADIANT  
 SUPERHEATER  
 CONVECTION  
 SUPERHEATER  
 CONVECTION  
 PASS WALLS  
 FURNACE ROOF  
 STEAM DRUM  
 I .  
 FURNACE WALLS 
 I  
 ECONOMIZER  
- FIRED t  
NTERMEDIATE PRESSURE TURBINE
REFUSE
ECONOMIZER
FIGURE III-57. FLOW DIAGRAM -COAL FIRED STEAM GENERATOR
~
HI - 1 3 5

-------
TABLE ill-2l
SUMMARY PERFORMANCE OF COAL-FIRED STEAM GENERATOR WITH
EXTERNAL,' REFUSE-FIRED ECONOMIZERS
Coal fired. tpd
3008
Combustion air input. (18% excess).


Wet flue gas produced. 103lb/hr


Steam produced. 103 lb/hr


Reheat steam. 103 lb/hr
103 lb/hr
2i749
3004
2800
2464
No.. of coal-mills operating
Stearn generator effiCiency. %
4
87.4
Pressure. psig
Superheater outlet
2520
Boiler drum
2640
Reheater inlet
577
Reheater outlet
518
o
Temperature. F
Superheater outlet steam
1000
Reheater inlet steam
640
Reheater outlet steam
1000
Feedwater entering unit
657
Air entering unit
80
Air leaving air heater
764
Gas leaving furnace
2260
Flue gas entering air heater
825
Flue gas leaving air heater
325
III-136

-------
FPC Codes
310
311

312

314
315
316
TABLE III-22

REVISED COS TS FOR THE SEP ARA TEL Y FIRED
ECONOMIZER PLANT
CAPITAL COS TS
Description
Land and Land Rights
Structures and Improvements
Boiler Plant Equipment
Turbine-Generator Equipment
Accessory Electrical Equipment
Misc. Power Plant Equipment
Air Pollution Control Equipment
Waste Handling Equipment
Engineering and Inspection
Total Capital Cost
ANNUAL COSTS
Annual Capital Cost, 106 $
(Effective annualization rate,
14. 6 percent)
6
Water Cost, 10 $

Operating Labor. 106 $

Maintenance, 106 $
6
Coal Cost, 10 $

Residue Disposal. 106 $
6
Total Annual Costs. 10 $

Annual Credit for Power Generated, 106 $

Quantity of Waste Burned. 103 ton/yr

Disposal Cost. $/ton

Transportation Cost (3.50 Mile Haul Radius),

Total Net Disposal Cost. $/ton
III-I37
$/ton
6
Cost, 10 $
0.69
5.47
31. 72
13. 75
2.87
0.59
4.04
1. 81
2.44
63. 38
9.25
0.01
0.62
1. 73
6.76
0.36
18. 73
18. 76
815
-0.04
1. 64
1. 60

-------
Originally, the Case 3 coal-fired stearn gene-
rator design had been dimensioned so that this boiler would be capable of
as suming the entire plant load in the event of an outage of all three refuse-
fired economizers. As pointed out earlier in the present discussion, an-
ticipating such a contingency can be accomplished in a more cost-effective
Inanner (see Section III, C, 2, b, (3». The stearn generator design was ac-
cordingly reduced to the size required for normal firing conditions (75%
heat input from coal).
The net result of this iteration has been to
render the Separately Fired Economizer an even more attractive system
for which minimal application risk is evident.
c.
Arch Furnace with Separate Superheater
(I)
Preliminary Engineering Technical Factors
As in the original Case 10 conception of this
design, feedwater would be taken from the last stage feedwater heater and
be evaporated and superheated at a temperature of 7500F in an arch-type
furnace. Further superheating to 10000F and reheating would be accom-
plished in a separate, coal-fired superheater.
In considering this design in further detail,
a problem came to light concerning the location of reheat tube surface.
1£ these tubes are located in an area of high gas temperature, tube damage
or failure may occur during start-up, when no steam is being returned to
them from the turbine. In conventional boilers, reheat surface is always
installed in locations where the tubes can be protected from excessive
temperatures, especially during the initial phases of start-up. For the
case of the partial superheat and reheat unit considered here, this was
not possible. It was therefore concluded that separate units for the re-
heat and final superheat duty would have to be provided. Figure III-58
diagrams the basic arrangement that was observed in developing the
preliminary design of the present system.
An alternate configuration was also seriously
considered. This would have consisted of a twin-furnace arrangement for
the coal-fired superheat/reheat function. This design would have represen-
ted a substantial cost savings over the three-unit configuration actually
chosen. The alternate design approach, although potentially quite practical,
was not taken because of concern over pos sible risk factors that would not
be encountered in the three -unit system.
Based on a refuse input-rate limit of 2000 tpd
for grateles s furnaces, a map of fuel requirements and steam cycle options
was developed and is shown in Figure Ill-59. The 60% refuse rate has again
been imposed as an arbitrary boundary. In this case, however, the rate is
III-l38

-------
""
C)
C
::XJ
m
H
H
H
I
......
VJ
...0
c:n
~
en
-t
m
»
~
(")
-<
n
r='
m
~
:I:
m
3:
»
::!
n
.
»
~
n
:J:
."
C
:xl
~
»
n
m
':tI
roo
»
z
-t
COAL
PIRED
SUPERHEATER
REFUSE AND
COAL FIRED
SfEAM
GENERATOR
COAL
FIRED
REHEATER
TURBO-OENERATOR
CONDENSER

-------
20%
CONDITIONS;
REFUSE COMB. EFFICIENCY 69%
COAL COMB.EFFICIENCY 84%
MAXIMUM REFUSE RATE/ UNIT = 2000 TPD
3.5
8000 tpd
/ BOUNDARY
5.0
4.5
4.0
c
~ 3.0
(¥)
o
...
~ 2.5
c(
a:
.J
c(
o 2.0
u
1250 PSIG
1
2
3
4 5 6
REFUSE RATE, 103 TPD
7
8
9
10
1
2 3
NUMBER OF REFUSE UNITS
4
5
FiGURE 111-59. FUEL REQUiREMENTS fOR COMBiNED fiRING (BLENDED FLUE GASES)
AT HIGHER REfUSE RATE
III-140

-------
not predicated on concern for steam-flow variations. The determining fac-
tor is that refuse rates exceeding 60% would necessitate a 1250 psig, non-
reheat cycle and the use of multiple steam generators and turbines. A
suspension-fired steam generator is not likely to experience the steam flow
variations that have been experienced with grate-stoked units.
I
[.
In terms of the arch furnace itself, the refuse
rate was arbitrarily set to allow some (20%) coal firing and thus the promo-
tion of more stable combustion conditions. This corresponded to overall
plant refuse heat input fractions of 59. 8 and 56.4% for the 1800 and 2400
psig cycles, respectively. These points are shown in Figure 1II-60, which
presents fuel requirements in terms of both the plant and the combined fired'
arch furnace. Also shown are possibilities for firing 75% and 100% refuse
in the arch furnace. The resultant curves exceed the plant refuse-rate of
60%, but this is merely illustrative. As can be seen, the firing of 100%
refuse in the arch furnace would lead to impractical steam conditions and
an unwanted multiplicity of units. In fact, 80% refuse-firing in the arch
furnace is marginal. Because of this, the heat input into the arch furnace
was reduced to 75%, which corresponds to overall plant refuse-fractions
of 55.9 and 52.7% for the 1800 and 2400 psig cycles, respectively. The
size of the plant was set at 100 MW, as determined by the cost modeling.
In the system thus far described, the combined-
fired arch furnace and the coal-fired superheater would operate so that the
flue gas leaving the air heaters would be at about 4500F. This is based on
dust collector requirements. In the coal-fired superheater unit, the tem-
perature of the gas leaving the tube banks of the boiler would certainly be
higher than in a conventional boiler, because the latter would be equipped
with an economizer and the former would have a higher inlet feedwater-
temperature. Some of the increase in sensible heat could be returned to
the system by way of the air heater, but not to the extent typical of con-
ventional plants. Thus, somewhat lower efficiencies must be expected.
The combustion calculations for the present
system are essentially the same as those developed for the system discussed
in the preceding section (Separately Fired Economizer). The stearn generator
efficiencies were also derived on the same basis, except that the radiation loss
value was changed because of the difference in the plant capacities of the two
systems. The following efficiency data for the arch furnace arrangement are
thus based on Tables ill-16 and -18 shown earlier, with the exception of the
radiation los s.
III-141

-------
1.0
/~~
103 LB/HR
103 TPD
2.5
200
2.0
UJ
~
a:
-'
c{
o
(,)
1.5
STM. GEN. REFUSE RATE,%
100
0.5
o
1
2
3
4 5 6

REfUSE RATE, 103 TPD
7
8
9
10
FIGURE 111-60. FUEL REQUiREMENTS FOR ARCH FURNACE PLANT
Ill-142

-------
EFFICIENCY OF ARCH FURNACE WITH SEPARATE
COAL-FIRED SUPERHEATER AND REHEATER
   Arch Furnace Superheater Reheater
Radiation loss, % 0.26 0.40 0.40
All other losses, % 26.61 15.47 15.47
Total, %  26.87 15.87 15.87
Efficiency, %  73.13 J 84.13 84. 13
Note:
Flue gas exit-temperature = 4500F (all units);
excess air = 50% (refuse), and 18% (coal).
(2 )
Design Details
The system, as thus far presented, would be
quite amenable to load variations when the fuel proportions are held constant.
At any given load, however, a plant refuse rate as low as 10% could also be
accommodated.
The summary of the respective duties for the
steam generator, superheater, and reheater is as follows:
THERMAL FACTORS FOR ARCH FURNACE WITH
SEPARATELY FIRED SUPERHEATER
AND REHEATER
  Duty Eff. Heat Input
Unit Fuel 106 Btu/hr i 106 Btu/hr ~
Steam Generator Refuse (80%) 666 \ 73.16 910 74.3
 & Coal (20%)
Superheater Coal 138 84.13 164 13.4
Reheater Coal 126 84.13 150 12.3
Plant  930 75.98 1224 100.00
The arch furnace and thus the plant refuse rate
would be 1960 tpd. The arch furnace would also burn 182 tpd of coal, the
superheater 164 tpd, and the reheater 150 tpd, for a plant coal rate of 496
tpd. The net plant heat rate would be 12,240 Btu/kw-hr. The fuel require-
ment point for this 100 MW plant can be seen on Figure llI- 60.
(a)
Arch Furnace Steam Generator
. The arch furnace steam generator is shown
in Figure 1II-61, the schematic of the water circuitry in Figure Ill- 62, and the
1II-143

-------
SUPERH.:ATER OUT:~;142"0'\ rN(\..y\=".."m~--=ul
- ==. 
-------
SEPARATE COAL.FIRED SUPERHEATER
~ 
PENDANT ELEMENTS
jl
HORIZONTAL
CONVECTION BANK
U
HEAT RECOVERY
AREA WALLS
j 
VERTICAL
DRUM BANK
h
FURNACE WALLS
 .
HORIZONTAL
ECONOMIZER BANKS
4 
FEEDWAl'ER HEATER OUTLET
FIGURE 111-62. FLOW DIAGRAM - ARCH FURNACE STEAM GENERATOR
IIl-145

-------
summary performance in Table 1II-23. The furnace shape chosen for this
steam generator is similar to that of an anthracite coal burning furnace.
The arch section of a typical anthracite furnace that has proven to be ef-
fective in the past is shown in Figure III- 63. The furnace design used for
the present steam generator differs from that shown in the arrangement of
the hopper area. Water-cooled grate bars would be installed on the sides
of the hopper to collect and burn any refuse particles that might fall into
the hopper before being completely combusted. A set of detailed drawings
(Figures I11-32 and - 33) of this type of arrangement was presented in con-
nection with Modification No.5 of the retrofit units. This same construc-
tion would be used for the present furnace.
It should be mentioned at this point that
this particular design treatment may prove unnecessarily conservative. If
the furnace fired satisfactorily without grate bars on the hopper walls, a
significant cost savings would be realized by omitting this structure. This
is essentially the course of action now planned in an EPA-supported demon-
stration program based on the design proposed in Reference 41. This on-
going retrofit operation will result in a system similar to that outlined for
Modification No.5, except that no grate surface of any kind will be installed.
In the arch furnace, there would be three
zones where secondary air would be admitted into the lower furnace. These
would be the lower furnace arch, the lower furnace front and rear walls, and
the hopper. Each would have separate windboxes and separately controlled
air supplies. The proportion of air introduced into anyone of these zones
would be adjustable, within certain limits, to vary the flow patterns within
the furnace. It is anticipated that some experimentation will be necessary
to determine the secondary air distribution pattern that will afford optimum
suspension and burnout of the fuel.
Both shredded refuse and pulverized coal
would be transported by pneumatic systems and injected into the furnace at
the arch. The fuel would mix with secondary air in the lower furnace and
most of the combustion should take place in that zone. The upper part of the
furnace would constitute an area where combustion could go to completion.
The pendant superheater, rear-wall screen tubes, and vertical drum bank
would be set on wide spacings to avoid fouling. The economizer tube banks
would consist of 2 -in. O. D. bare tubes arranged in an in-line pattern on
4-l/4-in. centers across the unit.
The steam- side design is that of a conven-
tional natural circulation circuit. Water would be fed from the economizer
outlet to the upper drum. The relatively cool water would then flow into the
downcomers and pass into the furnace walls. The hopper tubes would not be
of panel-wall construction because they should be situated in a staggered
double-row arrangement to accommodate the water-cooled grate bars. The
furnace-side walls and the front and rear walls above the arches would consist
III-l46

-------
TABLE 111-23
SUMMARY PERFORMANCE OF ARCH
FURNACE STEAM GENERA TOR
Refuse fired, tpd
1960
Coal fired, tpd
182
Number of coal mills in operation
Combustion air input (47. 3% excess),
3
Wet flue gas produced, 10 Ib/hr
3
Steam output, 10 Ib/hr
103 lb/hr
2
943
1099
759
Unit efficiency, %
73.2
Pressure, psig
Boiler
1975
Superheater outlet
1940
o
Temperature, F
Superheater outlet steam
750
Feed entering unit
440
Feed leaving economizer
594
Air entering unit
80
Air leaving air heater
426
Flue gas leaving furnace
1800
Flue gas entering air heater
723
Flue gas leaving air heater
450
111-147

-------
. --
trUaL /AIR ...n
AIR PORT WALL
NOIttZONTAL C'f'Q.CJNE
AIR DUCT
~E.trORATrD ~LATrl
,
FIGURE 111-63. DETAil OF ARCH SECTION OF ANTHRACITE FURNACE
III- 148

-------
of panel-walls having 3-in. O. D. tubes spaced on 3-3/4-in. centers. The
upper furnace wall headers would feed into the upper drum, where the dry
steam would be separated and fed down the walls in the heat recovery zone. '
The walls in the latter zone would also be of panel-wall construction, with
1-3/4-in.O. D. tubes spaced on 4-1/3-in. centers. At the bottom of the heat
recovery area, the steam would be collected in headers and transferred to the
convection superheater, which is located at the top of the heat recovery pass.
The convection superheater would have a two-tube, loop-in-loop pattern. This
type of a tube bank is economically preferred over a single tube arrangement.
The 2-1/4-in. O.D. tubes in this bank would be situated on 5-3/8-in. centers
acros s the unit. The steam leaving the convection superheater would then be
transferred to the pendant superheater. There would be eight pendant super-
heater-elements spaced across the unit. Each of the elements would consist
of five 2-1 /4-in. O. D. tubes. The steam temperature leaving the pendant
superheater would be 750oF.
(b)
Superheater
The design of the coal-fired superheater
unit is shown in Figure IIl-64, the schematic of the steam circuitry in Figure
Ill- 65, and the summary performance in Table Ill- 24. The duty of the unit
being small, the furnace exit-temperature was set at 16000F in order to
achieve sufficient furnace volume. This lower than normal furnace exit-
temperature did not pose any particular design problem. The only other
part of the gas - side design that is somewhat unconventional is the incor-
poration of a tubular air heater. As discussed previously, the gas leaving
the boiler will have a relatively high temperature because no economizer
surface would be provided. High gas temperatures within the air heater
could cause damage to the tube metal. Although regenerative air heaters
are less expensive, it was felt that a tubular air heater with one section
(upper 10 ft) constructed of alloy steels would be the preferred design. The
air heater would be made up of 1500 2-1 /2-in. O. D. x 30-ft long tubes.
The 7500F steam leaving the refuse-fired
steam generator would first be fed to the heat recovery zone tube-banks.
These banks would be composed of 2-in. O. D. tubes spaced on 4-3/4-in.
centers. Each row would have a two-tube, loop arrangement. At the top of
the convection superheater bank, the tubes of the bank would be bent to form
a tube wall, which would be connected in a header at the top of the unit. From
there, steam would flow through the roof and down the front wall tubes. The
remaining furnace walls, consisting of 2-1 /4-in. O. D. tubes, would then be
fed in succession. Panel-wall construction would not be used in the furnace
walls because in long, steam-carrying tubes the variations in temperature
can cause an uneven expansion of the tube and fin assembly.
After the steam leaves the last furnace
wall, it would be fed through the side walls above the heat recovery area.
The'se tubes would also be 2-1/4 in. O. D. The last tube surface in the steam
circuit would be the pendant superheater, which would consist of twelve ele-
ments spaced transversely across the unit, each element containing five
2-in. O. D. tubes.
llI- 149

-------
, I I
: '
/; : /
. I I.
I------~---
I I
I I
; -,
I '
I I
--,---------1
'/
't
r.,--::.:.:x.'7'--~-'" . ~:,:~'~~~1m
~;~ ':;' '~':l
I! YH ... !Ii

,:,1 ;,~ t . : I'
. ~. . ~ .. . " I
, "h .. . Ii :1
,~u;~,~~,:~;~"-. \~=~9U

PLAN AT SECT. "8-8"
:,
~
!
[t--
.~..._---
~ ~-:.?~.~:: J::-:::-==-:.: =- -_::.:-- -::-~: 7';-~ -=- -:,
20' -4 5/8"
FURN. WIDTH
-- .---
19'-117/8"
FURN. DEPTH
--
..
- -- ---. -- --
PLAN AT SECT. "A.A"
,

"8"
EL.77'.6"
EL. 75'-O~

EL.71'-6"
EL.50'-0"
--.----.
r--
"A"
~~'.6':
EL.27'-0"
,0


/~
L
GRADE EL. 0'-0"
PARTIAL SECTIONAL SIDE ELEVATION
/
EL. 54'.6"
:0
---.
"8"
---t
"A"
FIGURE 111.64. COAL-FIRED SUPERHEATER FOR 100 MW ARCH FURNACE PLANT'
III-150

-------
TURBiNE iNLET
t~
PENDANT elEMENTS
~
CONveCTION PASS
SIDe WAllS
,1
FURNACE
REAR WALL
A
FURNACE
RIGHT SIDE WALL
Ii
FURNACE
LEFT SIDE WAll
~
ROOF AND fURNACE
FRONT WALL
~
HORIZONTAL
CONVECT80N BANKS
j
SUPERHEATED STEAM
FROM HIGH PRESSURE TURBINE OUTLET (REHEATER)
OR PARTIAL SUPERHEAT STEAM GENERATOR (SUPERHEATER)
FIGlJRE 111-65. FLOW DIAGRAM. SEPARATE SUPERHEATER AND REHEATER OF ARCH FURNACE Pll:ANT
III-151

-------
TABLE III-24
SUMMARY PERFORMANCE OF COAL-FIRED
SUPERHEA TER AND REHEA TER
(ARCH FURNACE SYSTEM)
Coal fired, tpd
Number of mills in operation

Combustion air input {180/c excess),103 1b/hr
3
Wet flue gas produced, 10 1b/hr
Steam superheated, 103 1b/hr
Unit efficiency, o/c
Pressure, psig
Unit outlet
Steam entering unit
o
Temperature, F
Unit outlet steam
Steam entering unit
Air entering unit
Air leaving air heater
Flue gas leaving furnace
Flue gas entering air heater
Flue gas leaving air heater
1lI-152
Superheater
164
1
149
164
759
84.1
1880
1940
1000
750
80
614
1600
900
450
Reheater
150
1
136
149
668
84.1
490
520
1000
660
80
572
1600
862
450

-------
(c)
Reheater Unit
. The gas side design of the reheater unit
would be :es'sentiaUy the same as that of the superheater. The heat inputs
and steam temperatures would be almost identical, so that similar volume
and surface requirements would be involved. The design of the coal-fired
reheater is shown in Figure 1lI-66, and the summary performance above in
Table 1lI-24. The steam circuitry would be the same as that shown earlier
in Figure 1lI-65.
The steam side design of the reheater
and superheater units would differ because the pressures of the two units
would not be the same (490 vs 1880 psig, respectively). Thus the tube
thickness or diameter (or both) would have to be selected to accommodate
each case'. Another difference in the steam side design would be the num-
ber of tubes that could be fed in parallel. At a lower pressure, the increased
specific volume of the steam permits the in-parallel operation of a greater
number of tubes. This can be seen in the design of the convection and pen-
dant banks of the reheat unit. The convection bank would be comprised of
201 (4-in. O. D. ) tubes spaced on 4- 3 /4-in. centers. Each row would have
a three-tube, loop-in-Ioop arrang~ment. The pendant elements would also
be 2-1 /4-in. O. D., with 10 tubes per element. The tubes in the tube-walls
of the reheat unit would have the same outside diameter and spacing as those
of the superheater, although the tube thicknesses would be different.
(3 )
Cost Estimates
The iterative cost analysis of the arch furnace
plant was'handled in the same manner as the Separately Fired Economizer
and also reacted to the effect of decisions made in the detailed engineering
analysis. In the present case, however, disposal costs were increased
rather than reduced.
A summary of the cost data is presented in
Table 1lI-25. A comparison of the disposal cost with those given in the Cost
Model will show a significant increase due to the FPC Code No. 312 entry
chang e.
The resulting increase in total net disposal
cost substantially reduced the edge previously held by the Arch Furnace.
plant over the other syspension-fired systems*. In fact, it would now appear
to be less cost-effective than Case 7 (Stoker-Spreader). Although a detailed
design analysis of the latter system could not be undertaken, a review was
made to detect any possible design change requirements or other factors'
that might significantly alter the original capital cost estimates. It was
*Broadly speaking, Cases 5 through 10.
III-l 53

-------
F~~"~'-'--':-~-~'1f'-=~f
::..::JI'
~'"'_.. - : ~_mL ,~~~ i

L ~~_C".~_-=-"="£.'Jn7L'XiC,C~' -- ~'- -- --- -, '--'

P LAN AT S E CT. "8.8"
i' ,:r-:-T~~t~?~---::=:,::.=---,:::::::=::,:"~:~=,, -- --- ~


fl4 ---r;~~:;;~TH 1 .It 1,
0- II., ---~ ~ J ("'h l"t
-- ~---:--- ~--'~ -- ~J -- -- " :; \: I "
~ . ~ . '; ~;
"---c.../ :P;: 19'.11 7/8" ,- ::~" : '
9___~-- FURN DEPTH1 -", 'L~;

"UT '--:7 ',: ,- ..,J'J
PLAN AT SECT. "A.A"
r--
"'-::.~~,"~'
'~LI j~
- . [:;~>

' --,-~I

tl E='=--
I
, II
~':1
EL. 33'-6" t['C, ; ,;J' -- -- --- -- --:I~,: ,u'i
~r->o '~ ';-:1
EL.27'.0") '- -- - -- -----'--,0-' -- - "J!) :-- -- -- - -- --~' ,
" . . I
, ,,' / .. ~ i: -: I
~, ' .: ' ' , , ," I

/;/~ :~., ',_~..~lE_L~--1'~r'3v':' : -~-- j i
/'-'"" --'---,'-- , I
/' . . / J ., 'i : --!
';: '[/ r ~ ,- :]1 ':JI
r-L )- '. ',', " ~jl EL.14'.0" :; ~ ~~L --, - '!- )~
t:~l !: 11 :j :\ //
- LJ f':; !i n
, I -
PARTIAL SECTIONAL SIDE ELEVATION,
EL.71'.6"
EL.77'6"
EL.75'.0"
-'1
"8"
"8"
-'.

"A"
r'-
"A"
-
GRADE EL. 0'.0"
....--.--- ---- ._-
----
-. -- _u.-.- .-. _.---
FIGURE 111-66. COAL-FIRED REHEATER FOR 100 MW ARCH FURNACE PLANT
1
III-l 54

-------
FPC Codes
310
311
312
314
315
316
TABLE IIl-25
REVISED cas 1'5 FOR THE ARCH FURNACE PLANT
CAPITAL COS TS
Description
Land and Land Rights
Structures and Improvements
Boiler Plant Equipment
Turbine-Generator Equipment
Accessory Electrical Equipment
Misc. Power Plant Equipment
Air Pollution Control Equipment
Waste Handling Equipment
Engineering and Inspection
Total Capital Cost
ANNUAL COS TS
I 6
Annual Capital Cost, 10 $
Water Cost, 106 $
Operating Labor, 106 $
Maintenance, 106 $
Coal Cost, 106 $
Residue Disposal, 106 $
Total Annual Costs, 106 $
Annual Credit for Power Generated, 106 $
Quantity of Waste Burned, 103 ton/yr
Disposal Cost, $/ton
Transportation Cost (3.00 Mile Haul Radius),
Total Net Disposal Cost, $/ton
III-155
6
Cost, 10 $
0.47
1. 64
16. 66
5.39
0.99
0.26
1. 19
5.89
2.08
34.57
5.26 (Effective
Annualiza-
tion rate,
1 5. 2 pe r -
cent)
$/ton
0.00
0.49
1. 05
1.05
0.21
8.06
5.66
5 23
4.59
1. 58
6. 17

-------
concluded that, unlike Cases 3 and lO, the capital costs for Case 7 would
prove stable or, at most, would undergo only minor change because of ac-
curacy differences between the preliminary and detailed cost analyses.
Although the merit of the Case 7 system has
thus increased in prominence due to the iterative analysis, other trade-off
factors should also be considered. One such factor is that, in Case 10, the
refuse-fired furnace would be operated at temperatures (exit steam = 7500F)
well below those required for final steam conditions. This would be done to
minimize corrosion and fouling. Case 7, however, is a single -furnace sys-
tem in which the refuse ash will find wall temperatures corresponding to
final steam conditions. Thus the availability of the Case 7 design may not
prove to be as high as for the Case 10 system. As pointed out earlier, the
optimization study of a suspension-fired system was prompted basically by
projected technical advantages. In view of this rationale, it is felt that the
arch furnace could still be the preferred configuration to explore in future
construction, particularly if the design approach outlined here does prove
unneces sarily conservative.
3.
Retrofit Plants
Unlike the new construction designs, those for the retrofit
systems were not competitively analyzed within a cost model. Such treat-
ment was considered to be impractical because of the constraints imposed
by the existing structures for which retrofit is to be considered. The most
cost effective retrofit design may not be compatible with the steam conditions
or boiler configuration of the unit that happens to be the most available for
conversion. In the interest of creating a multiplicity of choices, all five
of the retrofit systems were brought to the same level of design detail as
were the two new-construction designs that were selected after competitive
analysis had been performed.
In developing costs for the retrofit boiler plant equipment,
costs were included for the removal of tubing, insulation and support steel.
To keep the costs representative of a type of unit rather than a specific unit,
the cost of moving boiler house equipment was not included. The new-material
costs include all material and equipment furnished by the steam generator
manufacturer, shop work, engineering, overhead and profit for the items
covered in that category. A fixed fee was also included for the erection of
the equipment; this included tools and erection equipment, expendables,
erection supervision, and profit for the erection. The labor estimate used
for the retrofit boiler plant equipment cost item was based on straight time
for field erection labor ($10. OO/man-hour).
All of the other items in the retrofit cost breakdown were
based on the derivations developed for the cost model. Additional capital
costs were included for new land to accommodate waste handling equipment,
new structures pertaining thereto and to the retrofit\ boiler, add- on acces sory
III-156

-------
electrical equipment, and extra APC facilities. The last cost item was
based on the increased volume of flue gas that would be experienced when
the design level of refuse is fired in the modified system. This fractional
treatment may appear to be an unrealistic approach, but is the only logical
way of apportioning APC costs between the actual (fuel) sources of the pol-
lutants generated.
The. results of this costing are shown in Table IIl-26,
which of course does not include any of the capital or annual costs asso-
ciated with the existing plant. The two entries for Modification 5, A and
B, represent the design (A) shown in Section Ill, B, 5, and (B) the same
design without water-cooled grate bars installed on the hopper walls. The
latter design option may be feasible if the burn-out characteristics of the
refuse particles suspended in the furnace prove favorable.
It can be seen by comparing the costs shown in Table IIl-26
and those roughly extrapolated from the data presented in Figures Ill-43 and
Ill-44 that the disposal costs (excluding transport) for retrofitted plants would
be lower than those for new construction plants of corresponding size and con-
figuration*. '
While this situation tends to favor the adoption of the retro-
fit approach, another consideration should be taken into account. The plant
factor used in calculating Table Ill-26 was 80%, the same value used for the
new construction systems. This value was observed to permit a more direct
comparison of the two types of systems, but it may not be entirely realistic.
The availability of the retrofit system may well prove to be considerably
lower than that of a newly constructed facility. This could result not only from
the fact. that the retrofit plant would tend to experience more down-time due
to age but also because the mode of firing would be more conducive to fur-
nace problems. Unlike new construction Cases 3 and 10, all of the retrofit
designs (except Modification No.2) would yield steam temperatures in excess
of 7500F in furnaces in which refuse would be fired. Thus a lower plant fac-
tor should be anticipated due to increased corrosion and foul.Lng problems.
As plant factor is reduced, the amount of refus e fired and
of coal displaced are also lowered. This results in a significant increase
(13 to 20% for each 10% (relative) drop in plant factor) in the total net dis-
posal cost. This can be seen in the following table.
SENSITIVITY OF DISPOSAL COST TO PLANT AVAILABILITY
Plant Factor, %
Mod. 1
Total Net Disposal Cost, $/ton
Mod. 2 Mod. 3 Mod. 4 Mod. 5A
Mod. 5B
 80 2. 15 4.37 3.59 2.52  5.01 4.25
 70 2.54 5.07 4. 19 2.98  5.82 4.94
 60 3.05 5.96 5.00 3.59  6.92 5.91
*. Modifications No. 1 and 4 can be compared with Case 2, No. 2 with
1. e. ,
Case 1, No. 3 with Case 7, and No. 5 with Case 6.    
IIl-157

-------
     TABLE III- 26       
    COSTS FOR RETROFIT SYSTEMS      
      Additional Capital Costs, 103 $   
  Description Mod. 1 Mod. 2 Mod. 3 Mod. 4 Mod. 5A Mod. 5B
 New Land   19.0 20.0 17. 0 13.5 15. 5 15. 5
 New Structures  292.0 905.0 228.0 276.0 218.0 218.0
 Retrofit Boiler Plant Equipment 1,465.0 4,529.0 1, 144. 0 1,380.0 1,091.0 14.0
 Accessory Electrical Equipment 146.0 453.0 115. 0 138. 0 109. 0 109. 0
 APC Equipment  60.0 52.0 60.0 43.0 60.0 60.0
 Waste Handling Equipment 734.0 816.0 2,080.0 524.0 2,940.0 2,940.0
 Engineering & Inspection 203.0 477.0 259. 0 170.0 310.0 221. 0
 Total Additional Capital Costs 2,919.0 7,252.0 3,903.0 2,544.0 4,743.0 3,577.0
H          103 $   
~      Additional Annual Costs,   
I        
......             
\}1 Annual Capital Costs  423.0 1,050.0 586. 0 370.0 710.0 538.0
00 
 Operating Labor  95.0 95.0 119.0 85.0 119.0 119.0
 Maintenance   95.0 95. 0 229.0 85.0 339.0 339.0
 Residue Disposal  129.0 145. 0 112.0 92.0 120.0 120.0
 Total Additional Annual Costs 742.0 1,385.0 1,046.0 632.0 1,288.0 1,116.0
 Annual Credit for Coal, 103 $ 560.0 545.0 517.0 420.0 483.0 483.0
 Refuse Fired, 3  278.0 292.0 247.0 198.0 226.0 226.0
 10 tons/year
 Disposal Cost, $/ ton  0.65 2.87 2. 14 1. 07  3.56 2.80
 Transportation Cost, $/ton  1. 50 1. 50 1. 45 1.45  1. 45 1. 45
 Total Net Disposal Cost, $ / ton 2. 15 4.37 3.59 2.52  5. 01 4,25

-------
It will be important, in considering the retrofit approach,
to review carefully the past availability of the unit selected for modification
and to allow appropriate compensation for the effect that refuse firing will
have on plant factor.
III-I 59

-------
IV.
RECOMMENDATIONS
A.
IMMEDIATE TECHNOLOGICAL OPPORTUNITIES
The intent of the present program has been to identify, charac-
terize, and rate the probable cost-effectivity of various steam generation
processes based on the use of refuse-fuel. Problem areas have been noted
that clearly demand further study; boiler configurations have been suggested
that never have been evaluated with waste fuels; systems have been described
that have already been proved, but that have not been operationally optimized.
As discussed in Section IV, B -below, there is considerable work of an R&D
nature that remains to be done. This does not mean, however, that imme-
diate technical opportunities do not exist nor that the pursuit now of such
endeavors will preclude the later application of benefits derived from R&D
operations.
What is important is that many of the systems described in this
report have already been or are being reduced to practice, if not with refuse
at least with other waste fuels. - A viable candidate design, considered appro-
priate for immediate construction in areas beset with S02 burdens and high
disposal costs, is the Case 3 system. It is felt to be superior in design to
a similar system that has already proved quite successful in Munich, Ger-
many. The immediate construction of other type, proven systems can also
be recommended; the selection, however, must also entail the consideration
of cost criteria, as discussed in Section III, C, 1 and Appendix C.
This leads to an area wherein considerable immediate technical
activity should be focused. The question of whether, as well as what type,
combined-fired steam generators should be incorporated within urban waste
management structures must be answered on the basis of local conditions.
In many LMA's these have not been systematically defined. Thus work
should be started immediately in many cities to determine how the present
trends in refuse-disposal costs, energy needs, demographic distributions,
and air pollution burden will influence decision making concerning future
waste management policies.
B.
SUGGESTED RESEARCH AND DEVELOPMENT PLANS
1.
Overview
The criteria used for selection of R&D plans to enhance
refuse combustion technology and a summary of the principal program areas
recommended for study have been presented in Section I, C, 3. What is sug-
gested is that Federal funds should be expended in engineering development
investigations all slanted towards demonstrating the practicality and economy
of using refuse as a low sulfur fuel.
IV-l

-------
In the pages that follow, the elements making up the key
programs are discussed and cost estimates presented for each. This is
done in the form of two plans; one (A) where it is assumed that $5 million
would be available over a 5-year period, and the other (B) where $25
million would be expended over the same period. In both plans it is
recommended that a $2 million central research facility be constructed
in order to eliminate expensive equipment redundancies and to standardize
test methodology. Long-term budgetary commitments are not yet being
made for environmental proi~'c\ts and hence a programmed flow .of monies
cannot be assured. The questions of the manner in which lesser funding
could best be spent or how to best assure maximum results from an irregular
flow of funds will be key ones facing the EPA Project Officer.
Clearly if construction funds for the central facility cannot
be committed early in the 5-year period then this recommendation should be
eliminated and whatever funds are appropriated used for R&D. Most pro-
grams developed below would result in a moderately thorough understanding
of the specific area under study and numerous distinct sub-elements of
information would be generated, including those pertaining to the pre-
proces sing, storage, and conveyance of refuse. Under conditions of les s
than anticipated funding, it would be necessary to reduce the overall scope
of the selected programs, but the portion undertaken would still make a
distinct contribution to refuse combustion technology. Thus, while a $300,000
effort in corrosion investigations would solve most of the current problems
here, a $75,000 program would still greatly assist users in minimizing damage.
Similarly, many millions of dollars could beneficially be expended in
developing a better understanding of mechanisms and equipment leading
to improved combustion efficiency, yet if only $750,000 can be budgeted,
this will still permit construction of far more advanced systems than now
possible. What phases should be eliminated of course will strictly depend
on the state of knowledge in the U. S. and Europe at the time the allocation
of priorities must be made. In the event of the availability of only, say,
$1. 5 million over a 5-year period, the broader areas suggested below should
be narrowed to one chief subject (suspension firing, for example) for maximum
cost effectiveness.
PLAN A
Central Research Facility:
The recommended research facility should be on the order
of 15,000-20,000 sq. ft. in area. It would consist of basic office space for
up to 30 persons, several small laboratories, a machine shop, a library,
data recording room, a small photographic laboratory, storage facilities,
a test cell for component evaluation, and a 2000-2500 sq. ft. high bay room
for housing a research furnace. The design study, architect, construction,
basic furnishings, and outside improvements are estimated to cost
$800,000-$850,000.
IV-2

-------
Housed within the buildi~g would be a grinder (15 tph), analytical in-
strumentation, data recording equipment, a model facility, pollution control
equipment, and the research furnace. Estimated costs for the basic equip-
ment are $900, 000-$950, 000.
The remainder of the funds ($200,000-$300,000) making up the es-
timated $2 million facility cost are reserved for metal and plastic stock,
chemicals, fuel, office supplies, equipment maintenance, 'library materials,
and utilities over the 5-year period. Each project would be expected to pur-
chase speciality materials and supplies as required. "
PROGRAM 1 - COMBUSTION OPTIMIZATION STUDIES: BASIC PROGRAM
Scope:
Following a prelimi~ary experiment-design effort, test work
will be performed to ascertain the optimum furnace conditions under which
the highest practical heat release rates can be achieved when firing refuse
in certain selected modalities. This study will be performed using the small
steam generator incorporated within the Central R&D Facility. Test operations
will be limited to the evaluation of operating parameters compatible with con-
ventional boiler configurations. "Non- conventional" thermal conversion sys-
tems will be similarly evaluated under Programs 8 and 9. Program 1 of
Plan B, discussed below, aims at the same basic objectives as the present
program, except that the firing modalities to be studied are more compre-
hensive and the degree of test parameterization more detailed.
Program Elements:
1.1
An 8-month period will be devoted to identifying and ana-
lyzing the relative importance of the various parameters that influence refuse
combustion rates. This will require extensive perusal of the pertinent dom.estic
and foreign engineering literature, as well as consultation with leading authori-
ties in the field. The analysis will focus on both grate- and suspension-fired
arrangements, but neces sarily isolate the two proces ses as individual topics.
Following this engineering study, an experimental test plan will be developed.
With this plan, the various parameters influencing combustion can then be
studied individually and in combinations to determine optimum operating
conditions and configurations. It is estimated that the accomplishment of
this phase will require an investment of $60, 000.
1.2
Implementation of that portion of the test plan dealing with
grate-stoked furnace configurations will be undertaken. As operating condi-
tions are varied, heat release rates and degree of refuse burnout will be the
IV-3

-------
principal parameters monitored. Because the boiler will be fully instru-
mented, data on air pollutant levels will be acquired at appropriate points
in the test operations.
Combustion air management will be the primary control
process studied. This will involve testing to establish the influence on
combustion efficiency of air distribution between underfire and overfire
inlets, of overfire air flow patterns, and of the amount of excess air used.
The role of combustion air temperature will be evaluated.
Tests will also be performed to establish the effects of
bed depth and grate speed on combustion rates. In these tests an artificial
refuse will be used to insure constant feed compositions.
Accomplishment of this work will require 12 months,
assuming that the Central R&D Facility is in complete operational status
at the beginning of that time. The cost of performing this task element
is estimated to be $300, 000.
1.3
This task element will be similar in purpose to element
1. 2, except that the firing style to be studied will be of the suspension type.
The same basic boiler configuration will be used, except that the air and
fuel addition systems will be those appropriate for producing suspension
burning. The modality of ground refuse injection will be similar to that
practiced with pulverized coal.
The key parameters of interest will again include heat
release rate and degree of burnout, to which will be added particle resi-
dence time (in suspension) as a function of shape and density, injection
velocity and paths, and specific combustion conditions. The effect of
air management will be evaluated, particularly in terms of distribution
between primary, secondary, and tertiary inlet points. Combustion ef-
ficiency will be established in terms of both excess air use and the gas
vertical velocity and turbulence in the combustion zone. The test data
obtained will also be related to air pollutant levels as well as particle
size distribution and moisture content of the feed.
This task element can be completed in 12 months time
and will cost $480, 000 to accomplish.
1.4
A final task element of Program 1 will be the genera-
tion of guidelines which will help optimize the design and operation of
combined fired installations. Development of this documentation will
involve the systematic analysis of the data generated in the other task
IV-4

-------
elements and extrapolation of the findings to scaled-up applications. Recom-
mendations will be developed on the basis of selected, popular boiler con-
figurations. This element will require 8 months to perform and will cost
about $60,000.
PROGRAM 2 - CORROSION AND FOULING
Scope:
Investigations made during the last five years, including several
on-going studies, have done much to elucidate the nature of corrosion and
fouling when refuse is fired by itself or in conjunction with a fossil fuel.
That these mechanisms need not be as harmful to continuous incinerator
operation as was once believed is apparent from the history of numerous
facilities and from the fact that new plants are being designed and built.
Whatever the outcome of present research, however, it is clear that certain
fundamental information concerning tube wastage must still be obtained if
improved incinerator designs are to be made in the future. Mechanical
"fixes II are now oftentimes employed and chemical elements contributing
to attack are hypothesized rather than being scientifically verified. Pre-
diction of the effect of a change in refuse composition or mode of firing
cannot now be made with any certainty. As corrosion research has con-
tinued for the last 40 years within the chemical process industries, with
the aim of developing ever superior materials or inhibiting processes, so
must it continue in the steam generating industry, particularly when as
heterogeneous a fuel as refuse is employed.
Program Elements
2. 1
A continuing program (5 years) should be maintained at
a level of $50,000 per year to investigate experimentally mechanisms of
fireside tube corrosion and fouling. An approximate 50-50 division of ef-
fort between experiments in the laboratory under controlled conditions and
in actual large boilers is recommended. Attention will be directed to the
role that metals present in refuse play, in addition to sulfur, halides, alkali
metals, and the alkaline earth metals. The role of reducing and oxidizing
atmospheres will be studied in both steady state and cyclic conditions.
2.2
In addition to the above applied effort, it is recommended
that a single $50,000 program of one year's duration be funded approximately
half-way into the 5-year period, whose purpose would be to summarize the
then best theories of the corrosion and fouling proces ses. Time phasing
should be such that any suggested final experimentation could be accom-
plished under Program 2. 1.
,
IV-5

-------
PROGRAM 3 - SYSTEMS ANALYSIS OF ALTERNATIVE THERMAL
CONVERSION PROCESSES FOR REFUSE
Scope:
A number of independent studies have been made, both privately
and with Government funding, to indicate the technical and economic feasibility
of candidate proces ses for converting refuse into either useful or harmles s
chemical species. Temperatures have ranged from the few hundreds of de-
grees for reaction schemes with carbon monoxide~:(, through the intermediate
temperatures of pyrolysis and conventional incineration, to the several thou-
sand degrees of special high temperature furnace processes. Some schemes
attempt to yield the minimum possible residue, while others deliberately
strive to produce the maximum possible homogeneous gaseous, liquid, or
solid fuel from the mixed refuse. While a simple ranking of the worth of
the various alternatives can perhaps never be developed, in that plant size
and local economics affect total costs to a significant degree, consistent
data do not yet exist for indicating relative advantages under various con-
ditions. A single program effort for generating such information should be
undertaken.
Program Elements:
3. I
An 18-month program, to cost an estimated $200, 000,
will establish realistic cost information for making valid comparisons of
alternative refuse conversion processes. Plant size, design features
(including required pollution control equipment), energy requirements,
and distribution or transportation costs will be examined in this analysis.
Proces ses to be examined will include the principal ones currently con-
sidered to be within the state-of-the-art where self-generated or external
heat is applied to refuse, with or without added chemical reactants.
PROGRAM 4 - PARTICULATE CONTROL
Scope:
It is not considered essential under the present plan to conduct
any extensive amount of research into improvements of air pollution control
equipment for refuse combustion in that the control R&D being supported or
planned by EPA will be directly useful when refuse is employed as a fuel.
Only programs where unique refuse effects are examined are considered
here.
~:( Composting temperatures might be considered to be the lowest of the
various thermal conversion processes, but in that this is essentially a
natural process, it need not be considered here.
IV-6

-------
',0. ..'.
Program Elements:
4.1
Only limited resistivity-temperature measurements have
been made on fly ash resulting from combustion of refuse separately or in
combination with coal. No measurements have been made under conditions
of suspension firing. During a 12-month investigation whose estimated cost
is $75,000, samples would be obtained and the necessary measurements made
so that electrostatic precipitator operating conditions can be better defined.
Chemical analysis of the ashes will be made in order to identify any corre-
lations with fuel composition and to permit examination of the concept of
additive injection for altering resistivity.
4.2
The EPA office concerned with solid wastes is sponsoring
work on novel incineration techniques where removal of particles at high
temperature is important. It is recommended that this "work be assisted by
the addition of $125,000 of air pollution control funds to on-going programs.
4.3
The influence of under- and over -fire air on particle carry-
over is still not fully understood. Within the central research facility furnace,
studies will be conducted that will permit the design of air introduction sys-
tems with minimum formation, pick-up, and transfer of ash. Both high speed
photography and probe samples will be utilized, with fly ash loading, size dis-
tribution, and los s on ignition being measured. Limited tests will also be con-
ducted in the suspension firing mode. Equations will be derived describing
the phenomena observed. This program element will require 18 months for
completion and the estimated cost is $300,000.
PROGRAM 5 - SMALL REFUSE-FIRED BOILERS
Scope:
Present studies on the use of refuse as a low sulfur fuel have
concentrated on facilities where many hundreds of tons per day of refuse
could be made available. Because of the magnitude of the urban solid
waste problem and the economics of scale for large operating plants,
attention was logically directed initially at these waste levels. Refuse
collection and haul costs in high population density areas has become so
expensive, however, that economical operation of smalle"r district boilers
might prove possible. The extensive rebuilding of many downtown areas
now permits consideration of incorporation of such new facilities in con-
junction with apartment and commercial heating netw<;>rks.
IV-7

-------
Program Elements:
5. 1
In a $100,000, 12-month study, an analysis will be made
of the incremental cost of present refuse disposal practices vs air pollution-
free district refuse boilers. Both addition of the necessary equipment to
current construction and to redevelopment areas will be examined. Cost
equations will be generated indicating relative advantages over a range of
large city practical conditions.
PROGRAM 6 - SUB-SCALE COMPONENT TESTING
Scope:
Incorporation 9f new innovations into refuse combustion systems
has been hampered by a lack of economic incentive plus the inability of most
developers of a new concept to test their device under realistic conditions.
The availability of the central research facility will do much to accelerate
current new product cycle times. Equipment that could be considered for
evaluation at that facility includes grate systems, combustion air intro-
duction schemes, transport devices, analytical and control instrumentation,
pollution control apparatus, size and component classification units, and
heat transfer systems. Most of this equipment would of course be smaller
in size than that to be employed in full-scale plants.
Program Elements:
6. 1
It is to be hoped that a significant number of the items
of equipment to be evaluated would be loaned, or furnished at minimum
costs, by interested vendors and developers. An allocation of $200, 000
per year for three years should permit the assembly of the various devices
into the research furnace and the conducting of sufficient tests to ascertain
the inherent value of the concept. An exact listing of the items of course
cannot be made at this time.
PROGRAM 7 - FLOW MODELING
Scope:
Limited European experience has demonstrated that studies
of incinerator flow characteristics in scaled models can yield valuable
information for improvements in large heat recovery systems.. Effects
on mixing conditions (within the fuel bed and external to it) and on fluid
dynamics relative to heat transfer surfaces can be ascertained in relatively
IV-8

-------
inexpensive equipment, thus lea.ding to superior combustion efficiency and
reduced corrosion. This area of investigation should encompass theore-
tical considerations of flow dynamics " design and fabrication of a. versatile
model, and the conducting of a series of experiments within the model.
Program Elements
7. 1
A 12-month survey of European and U. S. experience in
incinerator modeling will be made; a budgetary requirement of $70, 000 is
estimated. From a review of literature, interviews with authorities in the
field (both in general fluid modeling and specific incinerator modeling), and
an analysis of the current needs for improved design features, a report will
be issued describing potential benefits to be realized from an experimental
modeling program. An outline of the recommended R&D tasks will be in-
cluded in the report. .
7.2
The second phase effort within this program will consist
of a 6-month detailed design of modeling equipment. $60, 000 will be required
to complete this task to the point of working construction drawings and pur-
chasing specifications. Elements 7. 1 and 7.2 can be accomplished while
the central R&D facility is under construction.
7.3
Construction of the experimental modeling device will cost
$55,000, of which approximately $25,000 will be for material and components.
Nine months will be required to complete this element through <;:heck-out of
the system.
7.4
The last program element, requiring $115, 000 and 18
months to complete, will consist of an experimental study of the effect of
variations of design features on flow dynamics. Only features specific to
refuse-fired (and combined-fired) systems will be studied. Flow conditions
around an economizer section, for example, would be given attention only
in those cases where design improvements in a refuse-fired economizer are
sought. Because of pos sible unique corrosion from refuse constituents,
work will not be limited to the initial heat transfer zones only, but will be
conducted through upper pass sections where stagnation or local reduction
may lead to tube wastage.
IV-9

-------
PLAN B
PROGRAM 1 - COMBUSTION OPTIMIZATION STUDIES: EXPANDED
PROGRAM
Scope:
The rationale of this program is es sentially the same as that
of Program 1 of Plan A. The scope, however, is broader in that the firing
modalities to be studied are more comprehensive and the degree of test
parameterization more detailed. In terms of comparative benefits, the
present program will furnish guidelines for combustion optimization for a
much wider range of refuse-fired boiler designs. The purview, however,
is still limited to conventional boiler configurations and stoking proces ses.
Program Elements:
1.5
As task element 1. 1, the present one will consist of a
technical review and analysis effort and culminate in the development of a
test plan. The technical review effort will be identical to that of element
1. 1 of Plan A. The test plan development will, however, be considerably
expanded so as to include the additional experimental work described below
in the test-oriented task elements of the Program. The present element
can be carried out in an 8-month period at an estimated cost of $80, 000.
1.6
All of the work des cribed for element 1.2 will also be
performed in the present element. When optimum firing conditions have
been established for the agitating-grate furnace incorporated in the cen-
tral R&D facility, the following additional work will be performed. The
removable grate assembly will be replaced with two other styles, and the
tests accomplished in element 1. 2 of Plan A repeated to determine if the
same or different optimum conditions prevail. The sub-scale grates tested
should provide evaluations of the roller, forward-reciprocating, and one
other type selected at the discretion of the contractor. The grate configu-
ration providing the best combustion effects will then be used in the subse-
quent tests of this element.
Further test work will be addres sed to the evaluation of
refuse composition on combustion characteristics. In order to minimize
compositional changes due to dehydration, the feed material will be arti-
ficially prepared by blending, prior to firing, various proportions of the
usual ingredients. The ultimate analyses of these constituent will be ob-
tained prior to blending. The purpose of these tests will be to verify
IV - 1 0

-------
combustion calculations and to determine the point at which auxiliary fossil-
fuel firing would be warranted for fuel degraded by prior salvage operations,
weather, or compositional fluctuations of a seasonal or local nature.
A final experiment in this task element will be aimed at
evaluating the benefits of pulsed underfire oxygen on combustion efficiency.
The process of periodically introducing pure oxygen into the bed will doubt-
less increase burning rates and the degree of burnout. In the context of the
present program, utilization of oxygen will be explored from the point of
view of combustion control and the promotion of more even output steam
conditions. If such control proves feasible, process,costs will be compared
with the alternate approach of using auxiliary fossil fuel.
The task element just described will require a technical
effort lasting 18 months and a funding input of about $450,000.
1.7
This element will involve all of the experimental work
described for element 1. 3 of Plan A, plus certain additional tasks. In
accomplishing the element 1. 7 work, however, a more refined testing con-
figuration will be employed. The agitating grate will be removed from the
boiler and a hopper bottom installed to better reproduce the effects opera-
ting in full-scale structures. Similarly, the fuel injection nozzles to be
evaluated will be of both the horizontal and tangential types. This will
permit comparisons to be made of these two popular types of gun orientations.
When the objectives detailed for element 1. 3, as expanded
in the previous paragraph, are accomplished, the role of refuse composition
on suspension-firing efficiency will be evaluated. This work will es sentially
conform with the operations described in the second paragraph of element
1. 6.
In this test work, the pneumatic feed-flow process will be
designed on the basis of recognized test apparatus interface requirements.
For practical extrapolation of the optimization study results, it will be ap-
propriate that the configuration of full-scale conveyance systems be under-
stood and identified. Thus, studies will be conducted, wherein commercially
available components will be catalogued, appraised, and usable combinations
proposed as candidate designs. This survey will not be limited to hardware
appropriate only for the type of steam generator considered on this task ele-
ment, but to other (e. g., the arch furnace) suspension-fired configurations
as well.
months.
This element of the program should be completed in 14
Project costs will come to about $550, 000.
IV - 11

-------
1.8
On this program element, optimization of the spreader-
stoker firing modality will be attempted. This will involve modification
of the Central R&D Facility test boiler. The installed agitating grate will
be replaced with a sub-scale traveling grate and the charging chute removed
to accommodate the other type stoker. The experiments performed on this
modification of the test boiler will essentially parallel those described in
element 1. 7 involving suspension firing. That is, the effect on combustion
efficiency exerted by such parameters as combustion air quantities and dis-
tribution' particularly secondary or propulsion air, injection velocity and
paths, and feed consist will be determined. This task element will also in-
volve consideration of the commercial hardware, pneumatic and mechanical,
for providing the desired stoking effect when operating with ground refuse of
various top sizes. Equipment suitable for providing the sought-for function
will be identified and described.
Existing spreader- stokers, firing waste fuels, generally
employ travelling grates to catch the burning furnace fall-out. Because
this practice may actually be more traditional than inventive, consideration
will be given to possible alternative forms which might furnish superior
burn-out characteristics. AIl example would be the use of a catch-hopper
in which partially burnt-out refuse particles would be suspended with under-
fire air until elutriated from or precipitated to the bottom of the fluidic
system.
Execution of this task element will involve an expenditure
of about $520,000 and between 14 and 16 months of time to accomplish.
1.9
A viable furnace-form for suspension firing is the arch
furnace. It is not anticipated, however, that this particular configuration
can be easily reproduced in the multipurpose test-boiler which will be in-
stalled at the Central R&D Facility. The cost of designing and erecting a
sub-scale boiler of this type would probably exceed that implicit in con-
verting or rehabilitating an existing power boiler. It is expected that any
needed negotiations, such as for boiler use or refuse delivery, would be
consummated prior to the initiation of the present task element.
Modification operations will include the installation of a
pneumatic refuse injection system; selection of this hardware will be based
on the guidelines established on element 1. 7. Because anthracite coal will
not be acceptable as an auxiliary fossil fuel, auxiliary burners, firing oil
or natural gas, would also be installed. Appropriate ancillary equipment
will be erected to permit the receipt, movement, size-reduction, and feed
conveyance of the refuse. As necessary, plant instrumentation will be up-
graded or augmented so that all test parameters of interest can be properly
monitored.
IV - 12

-------
Test operations, although of full scale, still will essen-
tially conform with the work outlined for element 1. 3. It will not of course
be practical to use artificial refuse. For this reason, the stored, shredded
fuel will require frequent, composite sampling and analysis.
Assuming cost-free access to the boiler and delivery of
refuse, it is estimated that this work element will cost $1,250,000 to per-
form and a minimum of 2-1/2 years to carry out.
1. 10
The purpose and approach of the present task element is
the same as that of element 1. 4 of Plan A. The level of effort will be some-
what greater, however, because the documentation of design and operational
guidelines for optimized combined-firing will, in the present element, em-
brace a wider technical scope. The cost is therefore estimated to be $100,000
to be committed over a period of 8 months.
PROGRAM 8 - EVALUATION OF REFUSE PYROLYSIS AND GASIFICATION
PROCESSES IN CONNECTION WITH STEAM GENERATION
Scope:
Considerable intere.st focuses on non-combustive processes
wherein various materials are pyrolyzed to yield solid, liquid, and gaseous
products having significant fuel values. In the case of refuse, full-scale
systems that incorporate this process are now being built. The refuse
treated in such equipment will undergo considerable volume reduction and
the emitted gases can be burned without the production of fly ash. It has
also been demonstrated that the pyrolyzate gases can be fired to drive a
gas-turbine and thus produce power. The same feed gas-and, perhaps, the
carbonaceous bottom residues from the pyrolyzer can probably be fired in
a conventional boiler to generate steam. The question is would it be eco-
nomically practical to do so, assuming that optimum process conditions
had been established.
The purpose of the present program will be to determine, through
appropriate analysis and pilot test work, optimum design and operating con-
ditions for a pyrolysis-type fuel generator. Process costs would then be
systematically analyzed to establish the merit of the approach.
Program Elements:
8. 1
As an initial step, an updated state-of-the-art survey will
be conducted. A preliminary engineering analysis will then be performed to
establish the preferred .design configurations and operating condi'tions. At
this time, problem areas requiring subsequent treatment will be defined.
IV - 13

-------
A preliminary study will then be made to determine if the process is eco-
nomically feasible in terms of steam generation applications.
This work can be done in 6 months or less at a cost of
about $50,000.
8.2
Bench scale work will be undertaken on this task element
to verify or furnish needed information concerning the pyrolysis process.
The areas of test activity requiring particular emphasis will have been
defined on task element 8. 1. Generally, however, this work will serve to
insure that subsequent equipment design is properly based. Latent heats
of pyrolysis for different refuse constituents, and mixtures thereof, will
be measured. The heating value of process vapors and char produced at
different pot temperatures will be determined. The effect of different levels
of contaminant oxygen in the heating gas will be parameterized. The func-
tion of refuse grind- size on heat transfer and, thus, on pyrolysis rates,
will be established. Corrosion phenomena will also be studied, so that
proper materials of construction can later be selected. The composition
of condensibles removed at different temperatures from the pyrolyzate
vapors will be undertaken. This will help determine whether any par-
ticular fraction is worth isolating for further processing to chemical
markets. Similarly, pot residues will be analyzed to determine possible
values other than for heating.
The amount of laboratory effort which will actually be
required on this task element cannot be accurately estimated at this time.
It is likely, however, that the expenditure will not exceed $250,000 nor
require more than 2 years of committed time.
8.3
The design of a pilot plant will be undertaken on this
program element. Engineering drawings and specifications suitable for
subsequent construction work will be generated for the pyrolyzer itself
and all necessary ancillary installations. Cost estimates for the com-
plete plant will then be derived.
The plant will be fully instrumented so that heat dis-
tributions and material flows can be recorded. Among the control systems
provided will be one for transferring excess energy from the char-fed fire
box so that the desired level of pyrolysis will be sustained. This heat will
be transferred to the product fuel-vapor or, possibly, to a steam loop.
Coupling the unit to a steam generator would be costly
and serve no great purpose. Therefore, a simple combustion chamber
will be provided to permit burnoff of the products and possible recycle
at reduced temperatures.
IV - 14

-------
This effort will require 9 months to complete and cost
about $300,000.
8.4
The construction of the pilot plant will be accomplished
on this task element. A plant throughput of 10 tph is anticipated. Using
rough rule of thumb estimates, it can be expected that the complete turn-
key system will cost about $1,800,000. A minimum of two years should
be allowed for work completion.
8.5
Pilot plant studies will be conducted for a period of 18
months. The principal purpose of this work will be to determine the ope-
rating conditions under which maximum energy can be produced and deli-
vered to the use point without deleterious effects to the system or to the
atmosphere. The effects of fuel variations and the presence of different
levels of oxygen in the pyrolyzer will also be studied. Deleterious com-
bustion products resulting from the firing of both the fuel vapor and the
char will be identified and quantified. Design improvements will be made
in the course of the work wherever feasible.
This task element will require a funding of about $600,000."
PROGRAM 9 - EVALUATION OF REFUSE COMBUSTION IN FLUIDIZED
BEDS
Scope:
The fluidized bed is now being investigated as a potential furnace
configuration for coal-fired boilers. Because of the fuel suspension afforded
by the combustion air, uniform temperature distributions are achieved and
close positioning of heat exchange surfaces is possible. This results in
greatly increased heat exchange rates so that the heat exchange surface can
be correspondingly reduced. The net effect is that the boiler can as sume
considerably smaller size and cost. A further benefit is that solid air
pollutant absorbents can be suspended in the bed with the fuel to reduce
emissions (e. g., lime can be added to remove S02)'

This same configuration may also prove technically feasible>:'
and cost-effective for firing ground refuse. The range of particle sizes
of shredded refuse is much greater than that of the corn coal typically
fired in such test boilers. This problem can be managed, however, by
certain design features the success of which has already been demon-
strated in various hybrid systems.
*Current work being done in this area suggests this to be the case.
IV - 15

-------
The purpose of the present program will be to investigate the
feasibility of the process, assess the technical problelTIs associated with
it for refuse disposal, and develop designs for, build, and test an optimum.
configuration thereof.
Program Elements:
9. 1
A state-of-the-art survey will be made as the initial step
in this element of the program. The assembled information will then be
systematically analyzed and related to theory. From this basis and the
well-developed technology associated with fluidized systems, a feasibility
analysis will then be conducted. In support of this effort, laboratory and
field work will be performed to furnish the analysts with needed data on the
shape factors, particle size distributions and densities, and compositions
of refuse fractions obtained from typical size-reduction machinery. It is
anticipated that this analysis, grounded on well-established fluid mechanics,
will demonstrate that the fluidized bed in itself will not be a suitable furnace
configuration. Appropriate modifications will doubtles s have to be designed
to provide proper fuel behavior. The design recommendations that are
generated will be a cntIcal output of this program element.
The study described will require a technical effort lasting
about 5 months and requiring an outlay of approximately $40, 000.
9.2
Bench work will be pursued on the present task element
to validate the conclusions developed on program element 9. 1. Small scale
apparatuses will be constructed and operated to study the process designs
previously produced. Because the characteristics of fluidized systems
respond with considerable fidelity to scale-up, initial design optimization
will be attempted while at the bench. Matters to be evaluated will be: (1)
the use of an inert, solid phase for heat transfer; (2) gas velocities; (3)
feed-reduction requirements; (4) bed geometries; (5) excess-air effect;
(6) feed rate vs bed volume; (7) heat transfer rates; and (8) effectiveness
of hybrid features for managing quickly elutriated materials and dense
precipitates.
Assuming that process feasibility is proved, this work
will culminate in the generation of specific design recommendations which
will be utilized in the next stage of work. The cost of accomplishing this
element of the program will be about $750,000 and should be done in ap-
proximately 2 years time.
IV - 1 6

-------
9.3
On this program task, preliminary full-scale boiler de-
signs will be developed initially and cost analyzed in comparison with other
possibilities. The results will be documented to provide a decision-basis
for proceeding with pilot-plant-level testing. This process cost-evaluation
should be completed in about 4 months at a cost of about $35,000.
9.4
Pilot plant design will be undertaken on this program
element. Drawings, specifications, and costs for a unit capable of firing
100 tpd of refuse will be generated. Because it will be economically im-
practical to assess the quality of the output steam using a turbine-generator
set, standard enthalpy measurements will be used instead, and the product
condensed for subsequent reinjection as feed water. The ancillary equip-
ment requirements will closely approach those of the pilot-plant utilizing
the pyrolysis principle.
This design effort will require support to the amount of
about $300, 000 and entail a time commitment of 1 year or more.
9. 5
Construction of the fluidized-bed pilot plant for firing
refuse can only be approximately estimated at this time. It is obvious,
however, that high quality steam circuitry must be provided within the
boiler itself and even into the preliminary loops of the condenser, thus
imposing expensive hardware requirements. As a rough order of mag-.
nitude, complete plant costs will probably exceed $1, 100,000. The time
for plant construction will likely require at least 2 years of work.
9.6
Operation of the pilot plant will essentially involve the
validation and/or amendment of the conclusions developed on element 9.2
of this program. Sustained firing of the pilot plant will also allow the
acquisition of needed data on air pollutant emis sions , corrosive effects,
control and instrumentation requirements, and the handling of input and
output solids. Design improvements will be effected as the need is recog-
nized and insofar as correction is feasible.
Because of previous bench work, pilot test work need
only be conducted for a period of 1 year. The cost will a.mount to about
$775,000.
IV - 1 7

-------
PROGRAM 10 - ENGINEERING MANUAL FOR THE CONVERSION OF
BOILERS TO COMBINED FIRING
Scope:
The conversion of existing boilers to refuse firing is an attrac-
tive approach for gaining the sought-for capability. The initial investment
costs are much lower and construction time much shorter than for building
a new plant. A problematic aspect in the planning of conversion operations
is that the technology is novel. This will handicap the executives of utility
companies in their decision making and costly design errors could be made
by the engineering firms retained. What will be needed is a reservoir of
information from which engineering and economic guidelines can be extracted
to deal with the major aspects of any given conversion operation.
The manual developed on the present program will provide specific
information on how to select a boiler from the bases of condition, type, and
the urban interface. Instructions, including structural drawings and specifi-
cations, will be provided with which to implement boiler retrofits. All major
classes of fossil-fuel-fired boilers, which are suited to conversion, should
be covered. The lay-out of the various types of refuse handling facilities
will be explained and architectural details provided. Information will also
be provided on the performance characteristics of refuse handling devices,
including those incorporated into the steam generator itself. Costs for ef-
fecting the various types of conversions possible will be detailed, as will
be the operating and maintenance costs associated with refuse firing.
Although it is obvious that this program should not be conducted
by a firm in the utility-boiler manufacturing business, it will be important
that the cooperation of all boiler manufacturers operating in this country be
gained. It will therefore be highly desirable that the American Boiler Manu-
facturers Association (ABMA) not only endorses this undertaking but lends
its support to it. The task elements outlined below have been developed on
the basis of that assumption.
Program Elements:
lO. 1
The first step necessary will be to establish a catalog of
boiler types suited for conversion. Obtaining this information will require
a canvass being made of both boiler users and manufacturers. Implicit in
this undertaking will be the prior development and dissemination of explana-
tory material, listing the criteria for retrofit suitability. A canvas swill
also be made of the manufacturers of refuse handling equipment for design
information, performance specifications, and costs of their equipment.
IV - 18

-------
An analysis will be made of the boiler data obtained and
the dominant design forms classified into generic groups. Working arrange-
ments with the ABMA and its member firms will be organized.
This preliminary effort will cost only about $15,000 but a
minimum of 9 months must be allowed for its accomplishment. That rapid
reaction to canvass inquiries should not be expected is fully understandable,
considering the proprietary nature of the information solicited.
10.2
This task element will comprise' the bulk of the design
input required for the preparation of the manual. One or more candidates
will be selected from each of the generic boiler groups and identified with
the various retrofit options that can be considered practical. Design modi-
fication requirements will then be incorporated into a suitable plan-format
for each of the candidates. These will then be submitted to the manufacturer
of the boiler chosen, who will review, improve as necessary, and prepare
the retrofit-design information, drawings, and cost data required. This will
require of course that a subcontract arrangement be negotiated. Hopefully,
all of the major manufacturers of utility-class boilers will share in this
undertaking.
It has been roughly estimated that this task will involve
an expenditure of $75,000. A minimum of 18 months should be allowed for
its completion.
I 0.3
This task element will be devoted to the actual preparation
of the engineering manuals. Initiation of this work need not wait upon the
completion of elements 10. 1 and 10. 2. Document development will require
1 year IS time and an outlay of about $235,000.
PROGRAM 11 - DESIGN OF DEMONSTRATION UNIT
Scope:
The culmination of the 5 -year plan will be the design and con-
struction of a full scale plant. Operation of such a station will demonstrate
the degree of accuracy achieved in the previous design optimization while,
at the same time, revealing opportunities for improvement in future units.
On the present program, all required design work will be ac-
complished. This will start with the interrelated mat~ers of design and site
selection. Civil engineering and geological surveys will then be conducted.
Preliminary plant layouts and architectural sketches will then be developed
and reviewed with the operators for go-ahead. Detailed facility designs will
IV - 19

-------
then be generated together with all necessary drawings, specifications, and
bid documentation. A construction schedule, based on the critical path method
(CPM), will be developed and preliminary cost estimates for the complete
turn-key system prepared.
All of the material and information will be incorporated or ap-
propriately referenced in an overall planning document. This instrument
will outline the basic strategy to be followed in clearing the legal pathway
to ground-breaking, pursuing construction, and placing the finished system
in succes sful operation.
The program work described herein is based on the assumption
that an operator (possibly a grantee) has previously been designated. It is
also assumed that a plant throughput of 1000 tpd is compatible with the re-
quirements of the operator.
Program Elements:
11. 1
In this initial task the type* of plant best serving the
operator 1 s interests will be established. Sites available to him will then
be systematically analyzed and the optimum one designated. Based on this
choice or the best possible compromise, the operator will commit himself
to preliminary design specifications and sketches of the system considered
optimum for the operator will be developed. These will be submitted for
review and approval by the operator and the Government.
The site will then be submitted to a complete survey so
that soil conditions, easements and encumbrances (particularly with respect
to water acces s), underground and above-ground services, rights -of-way
(to the grid or stearn system, whichever applies), and adjoining traffic
patterns are duly recorded.
Plant layouts** with corresponding architectural sketches
will then be prepared and a selection made based on operator and Govern-
ment concurrence.
This preliminary design work will require a' minimum
of 6 months to accomplish and require funding in the amount of $100,000.
/
>:'Turboelectric, district or proces s stearn, etc.
>:';~Including all input/output service connections.
IV - 20

-------
11. 2
Detailed designs and specifications suitable to fully guide
construction operations will be generated for all plant components save those
standard items which are available off the shelf and conform with specifica-
tions. Boiler design will be accomplished by an ABMA member by special
sub - contract.
Because the plant will be a demonstration unit, it will be
important that a high degree of instrumentation be designed into it. This
will include not only the requisite feed-back control systems, but read-out
equipment capable of furnishing multi-point information on a number of
parameters not normally monitored in conventional steam generators.
Suitable legal instruments will be prepared to permit the
competitive procurement of the required work and material. This element
of the program will cost about $700,000 to perform and will require a mini-
mum of 18 months to finish.
11. 3
A summary construction-plan document will be the
principal end-item of this element. It will furnish a complete description
of the plant, and outline all of the procedural steps which must be taken to
arrive at a successfully operating plant. Developed for inclusion in it will
be a breakdown of estimated costs, a CPM construction schedule, and an
operation and maintenance plan, complete with manning recommendations.
Development of this documentation can be initiated before
the completion of element 11. 2. Costs will be about $200,000 for an 8-month
period.
PROGRAM 12 - CONSTRUCTION OF DEMONSTRATION PLANT
Scope: .
All necessary fabrication and erection work will be performed
to complete turn-key system designed on Program 11.
Program Elements:
The construction program will be carried out in the standard
CPM manner under the prime contractorship of a competent Architectural
and Engineering firm. Completion of the work will require about 3-1/2 years.
Actual plant costs are difficult to estimate at this time, but if Government
participation is foreseen, about $10, 000, 000 should be planned for its share
of the costs.
IV - 21

-------
II
II;.
V.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
"Refuse Collection Practice, " Committee on Solid Wastes,
APWA, 3rd Ed., 1966, p. 29.
"Solid Waste Report for the City of Chicago, II City of Chicago,
Department of Air Pollution Control, 1966.
Personal Communication from Office of Solid Waste Manage-
ment Personnel.
"Waste Management, II Regional Plan Association, March 1968.
"Report on Solid Waste Collection and Disposal, City of Malden,
Mass., " Green1eaf/Te1esca Engineers, Miami, Florida,
1 November 1967.
"Solid Waste Disposal Program for Metropolitan Boston, Volume
I, " Metropolitan Area Planning Council, approx. 1966.
"Bay Area Regional Planning Program, Refuse Disposal Needs
Study, Supplemental Report, " Association of Bay Area Govern-
ments, July 1965.
IIAn Analysis of Refuse Collection and Sanitary Landfill Disposal, "
University of California, Sanitary Engineering Research Project,
Tech. Bull. No.8, December 1952.
Karaian, V. K., "Solid Waste Disposal Needs and Practices in
Metropolitan Boston, " M. S. Thesis, Tufts University, June
1966.
"Report on Refuse Disposal, Milwaukee, Wisconsin," Black and
Veatch, Consulting Engineers, Kansas City, Mo., Volume I,
1960.
"Fresno Region Solid Waste Management Study, II Aerojet-General
Corp., Report No. 3413, to California Department of Public \
Health, March 1968. '
Black, R. J., et aI, "The National Solid Wastes Survey - An
Interim Report~resented at the 1968 Annual Meeting of the
APWA, 24 October 1968.
Niessen, W. R., "Systems Study of Air Pollution from Munici-
pal Incineration, II A. D. Little Co. Final Report to the NAPCA
on Contract No. CPA 22-69-23, March 1970.
V-I

-------
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
Statistical Abstract of the United States, 1967 (Table 6) and
1966 (Table 3).

Ibid, 1968 (Table 458).
'lAir Pollution - A National Sample, II USPHS Publication No.
1562, 1966.
"Air Quality Data 1964-1965, II PHS, Division of Air Pollution,
1966.
Bell, J. B., "Physical and Chemical Composition of Municipal
Refuse (Milwaukee), II Purdue University, August 1, 1959, to
March, 1960.
Etzel, J. E., and Bell, J. M., "A Report on the Sampling and
Composition of Municipal Refuse in Bloomington, Indiana,"
Purdue University, 1967.
Kaiser, E. R., "Refuse Composition and Flue Gas Analyses
from Municipal Incinerators , II Proceedings of the 1964 National
Incinerator Conference, ASME, New York, 1964.
Kaiser, E. R., IIChemical Analysis of Refuse Components, \I
Proceedings of the 1966 National Incinerator Conference,
ASME, New York, 1966.
Kaiser, E. R., "Composition and Combustion of Refuse, II Pro-
ceedings of MECAR Symposium Incineration of Solid Wastes,
March 21, 1967.
Letter from Leland E. Daniels, Staff Engineer, Bureau of Solid
Waste Management (Facilities Section, Technical Assistance and
Investigations Branch, Division of Technical Operations), trans-
mitting data compiled from the following:
23. 1
"Comprehensive Solid Waste Study: Johnson City,
Tennessee, \I Technical Services, Solid Wastes
Program, USPHS, Cincinnati, Ohio, May 1968.
23.2
Unpublished data, USPHS, Johnson City, Tennessee,
July 1968.
23.3
"Preliminary Report of a Technical Services Environ-
mental Study of the Weber County, Utah, Incinerators, II
US PHS, Solid Wastes Program, Cincinnati, Ohio,
October 1966.
V-2

-------
24.
25.
26.
27.
28.
23.4
"Solid Wastes Study of a Residential Area, " USPHS,
Solid Wastes Program, Cincinnati, Ohio, October
1966.
23.5
Technical Services Special Study, July 29 - August 2,
1968, Unpublished Data, Solid Wastes Program,
Cincinnati, Ohio.
23.6
Genesee County Solid Waste Disposal Study; Consoer,
Townsend & Associates, DHEW Grant No. l-D01-UI-
00070- 0 1, April 1968.
23. 7
"A Technical Services Report on an Environmental
Evaluation Study of the Alexandria, Virginia, Incine-
rator, " USPHS, Solid Wastes Program, Cincinnati,
Ohio, July 1968.
23.8
"Baling Municipal Refuse, II City of San Diego, Cali-
fornia, USPHS Grant No. 1- DO 1- UI - 00061- 0 1, Solid
Wastes Program, Cincinnati, Ohio, 1968.
23.9
"Quad-City Solid Wastes Project - An Interim. Report, "
USPHS Grant No. l-DOI-UI-00026-01, Solid Wastes
Program, Cincinnati, Ohio, 1968.
Golueke, C. G., and McGauhey, P. H., "Comprehensive Studies
of Solid Wastes Management, " First Annual Report No. SERL
67-7, Sanit. Eng. Res. Lab.., University of California, May'1967.
Fulmer, M. E., and Testin, R. F., "Role of Plastics in Solid
Waste, II Battelle Memorial Institute (Columbus Laboratories),
undated.
Kennedy, J. C" "Seasonal Variations in Municipal Solid Waste
Output, "Eng. Found. Res. Conf., Solid Waste Research and
Development, July 24-28, 1967.
Singer, J. M., et aI, "Flame Characteristics Causing Air
Pollution. 1. Production of Oxides of Nitrogen and Carbon
Monoxide, II Bureau of Mines Report RI 6958, 1967.
Martin, G. B., Wasser, J. H., and Hangebrauck, R. P. I
"Status Report on Study of Fuel Oil Additives on Emissions
from an Oil-Fired Test Furnace, II Paper No. 70-150, 63rd
Annual Meeting, Air Pollution Control Association, June
1970.
V-3

-------
29.
30.
31.
32.
33.
34.
35.
36.
37.
I
38.
39.
40.
41.
Shaw, J. T., and Thomas, A. C. I "Oxides of Nitrogen in
Relation to the Combustion of Coal, " 7th International Con-
ference on Coal Science, Prague, June 10-14, 1968.
Yost, D. M., and Rus sell, H., "Systematic Inorganic Chemis-
try of the Fifth and Sixth Group Non-Metallic Elements, II
Prentice-Hall, Inc., New York, 1944.
Duprey, R. L.. "Compilation of Air Pollution Emission
Factors, II Public Health Service Publication No. 999-AP-42,
1968.
Cuffe, S. T., and Gerstle, R. W., IIEmis sions from Coal-
Fired Power Plants: A Comprehensive Summary, " Public
Health Service Publication No. 999-AP-35, 1967.
Kaiser, E. R.. liThe Sulfur Balance of Incinerators, II J. Air
Pol. Contr. Assoc., ~, 171-74 (1968).
Unpublished data, Bureau of Solid Waste Management, Division
of Technical Operation, study during 9-13 December, 1968,
covering Atlanta and DeKalb Co., Ga.
Carotti, A. A., et aI, IlAir Borne Emissions from Municipal
Incinerators," interim reports, Contract PH 86-68-121, U.S.
Public Health Service, 1969.
IIAcceptance Tests on Boiler No.1, DUsseldorf Refuse Incine-
ration Plant, II Technischer tJberwachungs - Verein Rheinland
e. V., 20-21 April, 1967.
IIS02-S03 in the Firing of Steam Boiler Plants, II Mitteilung
des TtJV Bayern, Book 44/1968, pp 1214-15.
IIAcceptance Tests in November 1967 on Boiler No.2 of the
Munich North Power Station, II Technischer tJberwachungs-
Verein Bayern e. V., 29 May 1968.
Falkenberry, H. L., and Slack, A. V., IIRemoval of S02
from Power Plant Stack Gases by Limestone Injection, II
Preprint 54b, 61st Annual Meeting, American Inst. of
Chem. Eng., Los Angeles, Dec. 1968.
IITV A Tests Dry Limestone Proces s for S02 Control, " Chem.
Eng. News, Jan. 19, 1970, pp 30-31.
Wisely, F. E., IIStudy of Refuse as Supplementary Fuel for
Power Plants, II Horner & Shifrin, Inc., Report to City of
St. Louis, on Bureau of Solid Waste Management Grant No.
l-DOl-UI-00176-0l, March 1970.
V-4

-------
42.
43.
44.
45.
.46.
47.
48.
I
I .
49.
50.
51.
52.
53.
54.
IIBoiler Acceptance Tests in March 1965 on Boiler No.1 of
the Munich North Power Station, II Technischer t.Jberwachungs-...
Verein Bayern e. V., 14 June 1965.
Kaiser, E. R.; City of Orlando, Fla.; and Gannett, Fleming,
Corddry, and Carpenter, Consulting Engineers;' unpublished
data from 9 refuse samples from City of Orlando, Fla.
Smith, W. S., and Gruber, C. W., IIAtmospheric Emis sions
from Coal Combustion, an Inventory Guide, II NAPCA Publication
No. AP- 24, 1966.
Reid, W. T., Battelle Memorial Institute, private communication.
Wickert, K., liThe Accelerators of Corrosion in Furnaces, II
W};.rme, 74 (4), 103-109.
Kaiser, E. R. , IIEvaluation of the Melt- Zit High- Temperature
Incinerator, 11 Bureau of Solid Waste Management, Cincinnati,
Ohio, 1969. '
Duzy, A. F., and Walke r, J. B., "Utilization of Solid Fuel
Having. Lignite- Type Ash, II Bureau of Mines Information'
Circular 8304, 1966.
Michel; J. R., and Wilcoxson, L. S., IIAsh Deposits on
Boiler Surfaces from Burning Central illinois Coal, II Paper
No. 55-A-95, ASME Meeting, Nov. 1955.
Nowak, F., IIConsiderations in the Construction of Large
Refuse Incinerators, 11 Proceedings of the 1970 National
Incinerator Conference, ASME, Cincinnati, pp 86-92.
Langtry, W. D., and Kohout, J. F., IIFusion Point of Ash
from Mixtures of Coal and Foreign Materials, II Proc. 2nd Int.
Annual Con£. on Bitum. Coal, Vol. II, pp 301-6, Carnegie
Inst., Nov. 1928.
Air Quality Criteria for Sulfur Oxides, NAPCA Publication
No. AP-50, January 1969.
Air Quality Criteria for Particulate Matter, NAPCA Publi-
cation No. AP-49, January 1969.
Venezia, R. V., and Ozolins, G., IIInterstate Air Pollution
Study - Phase II Project Report Air Pollutant Emission in-
ventory, II HEW Report, December 1966.
V-5

-------
55.
56.
57.
58.
59.
60.
61.
62.
63.
i-
64.
65.
66.
67.
68.
Ritchings, F. A., "Raw Energy Sources for Electric Generation, "
IEEE Spectrum, Vol. 5, No.8, August 1968.
Smith, W. S., "Atmospheric Emissions from Fuel Oil Combus-
tion - An Inventory Guide, " PHS Report No. 999-AP-2, Novem-
ber 1962.
Welsh, G. B., "Air Pollution in the National Capital Area, " PHS
Publication No. 955, July 1962. .
Drobny, N. L., IISolid Waste Handling and Processing, II Joint
Meeting of Research and Special Technical Committee and In-
dustrial Incineration Committee, ASME Incineration Division,
New York City, January 15, 1969.
Niessen, W. R., "Systems Study of Air Pollution from Muni-
cipal Incineration, " VoL II, Arthur D. Little, Inc., March 1970,
pp. A- 3 - A- 11.
Engdahl, R. B., "Solid Waste Processing. A State-of-the-Art
Report on Unit Operations and Processes, II Report SW-4c, PHS
Publication No. 1856, Bureau of Solid Waste Management, Con-
tract No. PH 86-66-160, Battelle Memorial Institute, Columbus
Laboratories, 1969.
Patrick, P. K., "Waste Volume Reduction by Pulverisation,
Crushing and Shearing, II The Inst. of Public Cleansing, 69th
Annual Conference, Blackpool (England), 5-9 June 1967.
Stern, A. L., "A Guide to Crushing and Grinding Practice, "
Chem. Eng., 74, 129-46 (Dec. 1966).
Duszynski, E. J., "A Case for Milling Refuse, " Poll. Eng.,
1. (3), 29-31 (May/June 1971).
Essenhigh, R. H., and Howard, J. B., "Toward a Unified
Combustion Theory," Ind. Eng. Chem., ~ (1), 15-23 (1966).
Es senhigh, R. H., et aI, "Combustion Behavior of Small
Particles, "Ind. Eng:-Ghem., ~ (9), 33-43 (1965).
Howard, J. B., "Combustion of Solid Refuse, " ASME Pub1.
68- W A/INC- 2, Dec. 1968.
Sales Brochures, A V AC Pneumatic Transfer Systems,
Envirogenics Co., EI Monte, Calif.
Zandi, 1., and Hayden, J. A., "Are Pipelines the Answer
to Waste Collection Dilemma?" Env. Sci. Tech. , 3 (9),
812-819 (1969). -
V-6

-------
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
Zandi, 1., and Kenny, J. P., 'IState of the Art, " Section 3,
in "Bulk Transport of Waste Slurries to Inland and Ocean
Disposal Sites. VoL III. Technical Aspects of Pipelining
of Waste Materials," Bechtel Corp., Sept. 1969, Commerce
Dept. Clearinghouse No. PB 189759.
Zandi, 1., and Hayden, J. A., "The Flow Properties of Solid
Waste Slurries, "Int. Con£. on Hydraulic Transport of Solids
in Pipes, London, Sept. 1970.
Boettcher, R. A., "Air Classification for Reclamation Pro-
cessing of Solid Wastes, " ASME Pub1. 69- WA/PID-9; also
in Compost Science, .!...!.. (6), 22-29 (1970). .
Wilson, D. G., and Smith, D. E.,. "Mechanized Reclamation
from Municipal Solid Wastes, II Mass. Inst. Tech., presented
at the Nat. Ind. Solid Waste Conf., Houston, March 1970.
Darnay, A., and Franklin, W. E., l'Economic Study of Salvage
Markets for Commodities Entering the Solid Waste Stream, " .
Bureau of Solid Waste Management Contract CPE 69- 3, Mid-
west Research Institute, 1971.
Cammarota, V. A., Jr., "Refining of Ferrous Metals Re-
claimed from Municipal Incinerator Residues, " Proc. Second
Mineral Waste Utilization Symp., Chicago, March 1970,
p. 348.
Magnetic Separation of Non-ferrous Metal, " Annual Report,
Vanderbilt University, Department of Phys~cs and Astronomy,
April 1971.
"Air Classification of Municipal Refuse, " Stanford Research
Institute Research Brief, 11 June 1971.
"CPU-400 Year 2" Final Report, Contract CPE 69-100, Com-
bustion Power Co., July 1969.
"16th Steam Station Cost Survey, II Electrical World, Nov.
1969, p. 49.
Tremba, E., "An International Survey of District Heating, "
1 st International District Heating Convention, London, April
1970.
M~rch, 0., "District Heating for Existing Cities as Well as
New Towns and the Use of Refuse Incineration as a Base Supply
for the Heating, " 1st International District Heating Convention;
London, April 1970.
V-7

-------
81.
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
Kimura, H., "Present Situation and Future Problems of Dis-
trict Heating and City Environment in Japan, " 1 st International
District Heating Convention, London, April 1970.
Winkens, H. P., "The Application of District Heating to Exis-
ting Town Centres and New Town Districts Related to Mann-
heim, 'I 1 st International District Heating Convention, London,
April 1970.
Tugendhat, C., I'Nottingham to Pioneer Refuse Disposal and
Heating Scheme, " The Financial Times (England), (Aug. 14,
1968).
Bender, R. J. I "Stearn-Generating Incinerators Show Gain, "
Power, 35-37 (Sept. 1970).
Beningson, R. M., and Benings on, H. E., "The Utilization
of Solid Waste as a Source of Energy for District Heating and
Cooling Systems, " 1st International District Heating Conven-
tion, London, April 1970.
Mesko, J. R. I "District Heating and Cooling Systems Planning
by Computer," 1st International District Heating Convention,
London, April 1970.
Halzl, J., Szabo, Z., and Torma, M., "Peak Load Operation'
of District Heat and Power Stations, " 1 st International Dis-
trict Heating Convention, London, April 1970.
Geiringer, P. L., "District Heating by Means of High Tem-
perature Hot Water in Combination with Power Plant Design, "
1 st International District Heating Convention, London, April
1970.
u. S. S. R. State Committee for Participation in International
Power Conferences, "Advantages of District Heating and Its
Technical and Economical Efficiency, II 1st International Dis-
trict Heating Convention, London, April 1970.
Marecki, J., and Wojcicki, Jr., "Technical and Economic
Aspects of District Heating Development in Poland, II 1st
International District Heating Convention, London, April
1970.
Blomqwist, 0., and Waldenby, T., "Developments in District
Heating in Sweden, II 1 st International District Heating Conven-
tion, London, April 1970.
V-8

-------
..,..J
92.
93.
94.
95.
96.
97.
98.
99.
100.
101.
\
'<
102.
103.
"
Kono, M., "District Heating for Old Cities and New Towns in
Japan, " 1 st International District Heating Convention, London.
April 1970.
Ryman, J., "District Heating in the Stockholm Area, II 1st In-
ternational District Heating Convention, London, April 1970.
"Use of Waste Heat for Production of Fresh Water, II Saline
Water Conversion Report for 1966, U.S. Dept. of Interior,
p. 207-08.
Downs, J. E. I "Margins for Improvement of the StealTI Cycle, II
ASME Paper No. 55-SA-76.
"Abatement of Sulfur Oxide Emissions from Stationary Com-
bustion Sources, II National Academy of Engineering and National
Research Council ad hoc panel report No. COPAC-2 to the
NAPCA on Contract No. CPA 22-69-31, 1970. .
Walker, A. B. I "Characteristics of Emissions from Industrial
Boilers, " paper presented at Industrial Coal Conference, Pur-
due University, Lafayette, Ind., 12 Oct. 1966.
Stenburg, R. L. I et aI, "Field Evaluation of Combustion Air
Effects on Atmospheric Emissions from Municipal Incinerators, "
JAPCA, g (2), 83 (1962).
Walker, A. B., and Schmitz, F. W. I "Characteristics of Fur-
nace Emissions from Large, Mechanically-Stoked Municipal
Incinerators, " paper presented at ASME National Incinerator
Conference, May 1966.
Sternitzke, R. F., and Dvirka, M., "Temperatures and Dis-
tributions in Large Rectangular Incinerator Furnaces, " paper
presented at ASME National Incinerator Conference, May 1968.
Fife, J. A., and Boyer, R. H. I Jr., "What Price Incinerator
Air Pollution Control? ", paper presented at ASME National
Incinerator Conference, May 1966.
Sutin, G. L. I et aI, "East Hamilton Solid Waste Reduction
Unit, II A Preliminary Engineering Report to the City of
Hamilton, Ontario, 31 July 1968.
"Steam Power Plant South High Pressure Section - With
Natural Gas Firing - Refuse Firing and Remote Heating
Supply, " Munich Municipal Works Electricity Plant Report
(undated). .
V-9

-------
104.
Nowak, F., "Corrosion Phenomena in Refuse Firing Boilers
and Preventive Measures, " presented at the International
. Symposium on Corrosion in Refuse Incineration Plants,
Vereinigung der Grosskesselbetrieber, e. V., DUsseldorf,
West Germany, April 1970.
V-IO

-------