EPA-600/2-76-071
March 1976
Environmental Protection Technology Series
ENERGY CONSERVATION TECHNIQUES FOR
THE IRON FOUNDRY CUPOLA
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
V
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available-to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/a-76-071
May 1976
ENERGY CONSERVATION TECHNIQUES
FOR THE IRON FOUNDRY CUPOLA
by
Dennis J. Martin
and
James J. McCabe III
York Research Corporation
One Research Drive
Stamford, Connecticut 06906
Flynn and Emrich Company
3001 Grantley Avenue
Baltimore, Maryland 21215
Contract No. 68-02-0286
ROAP No. 21ARO-002
Program Element No. 1AB013
EPA Project Officer: Robert C. McCrillis
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, N.C. 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, D.C. 20460
ii
-------
ABSTRACT
This report presents the results of an investigation into
various existing or emerging technologies which can be utilized
to reduce the energy consumption and pollution control costs
of typical cupola operation. There were two primary motivating
factors behind this investigation. First, the rapid rise in
fuel costs has made cupola operation difficult for some foundries.
Second, the necessity of installing highly efficient pollution
control devices on the cupola has again placed a financial
burden upon the smaller foundries.
This report details those options available to the foundries
in terms of technological devices which will conserve energy and
capital. Included in this investigation were hot blast
recuperation, divided blast, oxygen enrichment and innovative
pollution control equipment.
This report was submitted by the Flynn and Emrich Company in partial
fulfillment of Contract Number 68-02-0286 under the sponsorship
of the Environmental Protection Agency.
111
-------
CONTENTS
Abstract iii
List of Figures v
List of Tables vii
Acknowledgements viii
Sections
I Conclusions 1
II Recommendations 2
III Introduction 3
IV Cupola Operation 4
V Recuperation 20
VI Oxygen Enrichment 48
VII Divided Blast 75
VIII Air Pollution Control 100
IX Discussion 108
X References 111
XI Appendices 113
IV
-------
FIGURES
No. Paqe
1 Cupola Characteristics 5
2 Price of Coke 1968-1975 19
3 Effect of Hot Blast on Coke Consumption 21
4 Effect of Blast Air Temperature on Melting Rate 23
5 Schematic of Heat Recovery System 26
6 Escher Recuperative Blast Heater 31
7 Plant Schematic, U.S. Pipe and Foundry,
Berlington, N.J. 33
8 Typical Escher Recuperative Arrangement 36
9 Inner Tube Fin Arrangement 37
10 Dimensions, McQuay Perfex Recuperators 38
11 Pressure, Temperature Relationship for
McQuay Perfex Recuperator 39
12 Recuperator Schematic, Berlin-Chapman Foundry 42
13 F.E.C.O.R. Recuperative System 46
14 Oxygen Cost Versus Consumption 50
15 Typical Oxygen Flow Control System 51
16 Methods of Oxygen Enrichment 60
17 Equivalent Enrichment Technique 62
18 Conceptual View of Oxygen Enrichment System 68
19 Typical Liquid Tank Installation 72
20 Effect of Varying Blast Distribution to Lower and
Upper Tuyeres on Furnace Performance 77
-------
FIGURES (continued)
No.
Paqe
21 Effect of Varying Blast Distribution to Lower
and Upper Tuyeres on Metal Composition 77
22 Effect of Tuyere Spacing on Furnace Performance
at Various Charge Coke Quantities 79
23 Effect of Tuyere Spacing on Metal Composition at
Various Coke Charge Quantities 79
24 Effect of Tuyere Spacing on Furnace Performance at
Various Blast Rates 80
25 Reduction of Charge Coke Consumption and Increase
in Melting Rate by Operating with Two Rows of
Tuyeres with Divided Blast Supply 81
26 Effect of Metal Temperature and Tuyere Spacing on
Carbon Content of Metal 83
27 Effect of Metal Temperature and Tuyere Spacing on
Silicon Content of Metal 83
28 Effect of Blast Rate on Metal Temperature at Various
Tuyere Spacings 85
29 Pattern of Lining Burn-out When Operating with One
and Two Rows of Tuyeres 85
30 Height of Burn-out for Cupolas of Various Sizes 86
31 Effect of Stage of Combustion on Element Loss in
Cupola 95
32 Relationship between Total Carbon Content of the
Iron and the Stage of Combustion within the
Cupola 97
33 Stage of Combustion in the Cupola vs. Melting 97
34 The Lone Star Steel Steam-Hydro Air Cleaning System 101
35 Energy Requirement vs. Emissions 103
36 Typical Steam Rate Required for Providing Draft 104
37 Schematic Diagram and Cross Sectional Diagram
of Laboratory-Scale Unit JL06
A-l Relationship between C02 and CO in Effluent Gases 114
VI
-------
TABLES
No. Page
1 Cupola Data 7
2 Typical Chemical Composition 8
3 Particle Size Disbtribution-Cupola Emissions 15
4 Cost Equations 17
5 Example Operating Data 22
6 Recuperative Hot Blast Systems 28
7 Escher Recuperator Price for Blast
Preheat of 950°F 29
8 Plant Data, U.S. Pipe and Foundry,
Berlington, N.J. 32
9 Plant Data, Chevolet Foundry, CMC
Saginaw, Michigan 35
10 Plant Data, Berlin Chapman Foundry,
Berlin, Wix. 41
11 Plant Data, U.S. Pipe and Foundry,
Union City, Calif. 43
12 Design Data, Central Foundry Division 47
13 Case History Summary - APCI 69
14 Divided Blast Operation Compared with Previous
Practice at Industrial Plants 88
15 Pressure Losses, Dust Concentrations, Gas and
Bed Material Flow Rates and Filtration
Efficiencies for a Number of Bed Feed
Materials 107
A-l Mean Specific Heats 115
A-2 Heats of Formation 117
A-3 Heats of Reaction 119
Vll
-------
ACKNOWLEDGEMENTS
This report was authored by York Research Corporation of Stamford,
Connecticut, on a subcontract from Flynn and Emrich Company.
Special appreciation is extended to Mr. James Turner III, President,
Flynn and Emrich Company for his assistance and cooperation.
Mr. Robert McCrillis of the Officer of Research and Development,
U.S. Environmental Protection Agency, Research Triangle Park,
North Carolina 27711, served as the Project Officer.
Vlll
-------
SECTION I
CONCLUSIONS
There exist various types of devices and technologies which
can successfully be used to reduce the cost of cupola operation.
Hot blast recuperative heat exchangers are now manufactured by
several companies. None of these systems appear to experience the
problems such as fouling which have detracted from such devices
in the past. Hot blast operation can reduce coke consumption
by up to 25 percent while reducing the exhaust gas volume to
the pollution control device. Divided blast operation can pro-
duce results similar to hot blast recuperators with a much lower
capital cost. Oxygen enrichment of the combustion air can also
affect coke consumption but appears more useful as a combustion
control device.
New types of air pollution control devices have also been studied.
The three devices included in this report appear to control pollu-
tion more economically than the conventional devices when applied
to the cupola.
There are many uses open to foundry engineers for waste heat
from the cupola, from in-plant heating to sand drying. Applica-
tions of waste heat must be evaluated on a plant by plant basis
since the plant layout and economic policy are usually the
deciding factors in such cases.
-------
SECTION II
RECOMMENDATIONS
Since this report deals with methods of reducing iron foundry
operational and air pollution control costs, it is recommended
that it be made available to all Environmental Protection
Agency personnel who deal with setting compliance schedules
for foundries.
It is recommended that a yearly update of this report be made
to take into account emerging technologies and that these
updates be made available to the foundry industry and compliance
officers.
It is also recommended that the EPA and ERDA aid in the design
of new foundries to demonstrate new pollution control and energy
conservation technologies on a trial basis.
-------
SECTION III
INTRODUCTION
The cupola is the device most often used for the melting of
gray iron. It is also generally considered to be the most
economical melting technique despite the fact it is an extremely
wasteful device from an energy standpoint. Two events in recent
years, however, have caused significant difficulties in the
use of the cupola. The first event was the Clean Air Act which
produced numerous air pollution regulations from federal, state
and local agencies. The second event was the substantial increase
in fuel prices.
These two occurrences have made the cupola a much more expensive
device to operate and maintain than would have been thought
possible ten years ago. All three problem areas; energy waste,
air pollution control and coke price increase, are part of
the difficulty involved in running a successful and economical
cupola operation. The adverse impact of each, however, can
be minimized by the application of existing technology. It
is the purpose of this manual to introduce these various technologies
to foundry management. While these technologies are for the
most part not new, they have not been utilized extensively
by the foundries due to the fact that they were not necessary
prior to the environmental movement and the escalation in energy
costs.
First the report details the operation of the cupola from an
energy viewpoint. Following this is a section devoted to air
pollution aspects of the cupola. A review is then made of
the fuel situation as it now exists in this country. After
defining these problem areas in depth attention is then focused
on solutions. Recuperative hot blast, divided blast, oxygen
enrichment and innovative pollution control devices are dealt
with both from a theoretical viewpoint and from the result
of an industry survey. The final section is devoted to outlining
options available to the cupola engineer on the basis of the
various technologies reviewed.
-------
SECTION IV
CUPOLA OPERATION
The cupola is simply a device for melting iron to a specific
chemical composition and at a specific temperature. The physical
characteristics of conventional type cupolas are shown in
Figure 1. In preparation for melting, the bottom doors of
the cupola are propped shut and a layer of sand up to 10
inches in depth placed above them up to the iron tapholes.
A layer of coke is placed on top of the sand and ignited usually
with natural gas. More coke added to a specified height above
the cupola's tuyeres,and then alternate charges of coke (mixed
with a fluxing agent) and iron are fed into the cupola through
the charging door.
Combustion air is forced into the cupola through the tuyeres.
The heat generated by combustion melts the iron charge and the
molten metal flows through the coke layers to the sand bottom
where it can be tapped. For an in-depth explanation of cupola
operating practices the reader is referred to the manual entitled
"The Cupola and Its Operation" (1)* prepared by the American
Foundrymen's Society.
The cupola is an energy intensive device with over 50% of the
heat generated being exhausted to the atmosphere. For any good
foundry operation an energy balance should be constructed. The
procedure for doing such, while somewhat involved, lays the basis
for the correct understanding of the cupola. An example of how
to perform a heat balance on the cupola is as follows.
ENERGY BALANCE
A. Heat Input
Table 1 and 2 list data from a typical cupola operation. It must
be noted that these data are from a relatively stable period of
cupola operation and that startup and shutdown data are not in-
cluded. Other data used in this analysis can be found in Appendix
A.
1. Total heat input of coke
Ibs of carbon burned = Ib of coke x % fixed carbon**
hr hr TOO
= 6583.2 x 0.92
= 6061.1 Ib/hr of carbon
*Numbers in brackets refer to references which are listed in
SECTION X, REFERENCES.
**Refer to Appendix B for English to Metric conversion factors.
-------
MELTING OF GRAY IRON
Stack
Skip- hoist rail
(I o-f Z )
Brick lining
Cast iron lining
Charging door
Charging
deck
Wind box
Skip-hoist rail
(lof 2)
Brick lining
Cast iron lining
Charging door
roclory lining
Water outlet
Blast duct
Tuyere
Iron trough
Tophole for iron
hole Is I8O°
opposite)
Sand bed
Door (I of 2)
Prop
Stack
Charging deck
Water flow between
inner and outer shell
Skip-hols_t rail
(I of 2 )
Brick lining
Cast iron lining
Charging door-
Water-cooled
tuyere
Slag dam
Slag and Iron
trough
Sand bed
Stack
Charging deck
Ooord of 2)
block
Prop
(Conventional Cupola )
(Water-cooled Cupola,water-wall)
Slag dam
Slag and Iron
trough
Sand bed
-Door(I of 2)
•Prop
(Water-cooled Cupola,flood cooled)
Figure I. Cupola characteristics.
-------
Since carbon yields 14,452 BTU per Ib combusted then the total
potential heat is:
6061.1 Ib x 14,452 BTU of carbon
hr Ib
or 87,595,017 BTU/hr
a. Carbon loss to dissociation
Not all of the carbon is combusted since some of it is lost in
the dissociation of water vapor in the combustion air to carbon
monoxide and hydrogen. The reaction is as follows:
H2°(q) + C ^ CO + H2 Eq. 1
Using a pyschometric chart and the data from Table 1 (i.e., dry
bulb temperature, wet bulb temperature) it can be determined
that there are approximately 6.18 grains of water per cubic foot
of blast air. If it is assumed that the blast air volume given
in Table 1 is approximately correct then the total rate of mois-
ture into the cupola is:
6'18 grains x 678,000 ft3 x 1 Ib
ft3 hr 7000 grains
or 598 Ibs of H20 per hour
From equation 1 it can be seen that 12 Ibs of carbon are required
to dissociate 18 Ibs of water vapor, then
598 Ibs of H20 x 12 Ibs of carbon
hr 18 Ibs of H20
or 398 Ibs/hr of carbon are used in this reaction.
b. Carbon loss to iron
Carbon content of iron charged %: 2.52
Carbon content of iron tapped %: 3.00
Carbon loss %: 0.48
Since the iron charged per hour is 56,000 Ibs/hr
..»!> *
56,000 Ibs x .0048
hr~
or 268 Ibs/hr carbon loss
-------
Table 1. CUPOLA DATA
Charging rate Ib/hr 56,000
Coke rate Ib/hr 6,588.2
Limestone rate Ib/hr 1,120
Blast volume ft3/hr 678,000
Blast temperature °F 70
Tapping temperature of metal
OF 2,762
C02 Content of stack gas % 12.00
CO Content of stack gas % 14.9
Dry bulb temperature of blast air
°F 70
Wet bulb temperature of blast air
°F 68
Barometric pressure in.HgA 29.92
Stack temperature °F 900
-------
Table 2. TYPICAL CHEMICAL COMPOSITIONS
Tapped Iron
-Carbon Content % 3.0
-Silicon Content % 0.60
-Manganese Content % 0.48
Charged Iron
-Carbon Content % 2.52
-Silicon Content % 0.70
-Manganese Content % 0.72
Slag
CaO Content % 22.00
FeO Content % 2.5
CaC03 Content of limestone % 98
Fixed Carbon Content of Coke % 92
-------
Net heat input from coke:
Carbon input 6,061.1 Ib/hr
Dissociation loss 398 Ib/hr
Iron loss 268 Ib/hr
Net 5,395.1 Ib/hr
Net heat input
5,295.1 Ibs carbon x 14,452 BTU
hr Ibs carbon
or 77,969,985 BTU/hr
2. Heat from oxidation reactions
a. Pounds per hour of slag
Flux added 1120 Ibs/hr
CaC03 content of flux 98%
CaC03 weight 1097.6 Ib/hr
CaCO3 moles 10.976
Since the same number of moles of CaO are formed as CaCO3 charged
then
CaCO in slag 10.976 moles
CaCO in slag (at 56 Ibs/mole) 614.7 Ibs
From Table 3 the CaO content of the slag is 22%.
Therefore, the total amount of slag produced per hour is
614.7 Ibs CaO x Ibs slag
hr .22 Ibs carbon
or 2794 pounds slag per hour.
b. Oxidation of iron
FeO content of slag 2.5%
Heat of formation 1,596 BTU/lb
Heat from oxidation: 2794 Ib slag x 0.025 Ib FeO x 1,596 BTU
hr Ib slag Ib FeO
= 111,480 BTU/hr
-------
Oxidation of silicon
% in charge
% in melt
% oxidized
silicon oxidized
0.70
0.60
0.10
56 Ibs/hr
heat from oxidation = 56 Ibs x 13,140 BTU
hr Ib
= 735,840 BTU
hr~~
d. Oxidation of manganese
3. Energy from combustion air
In this example the blast air was not preheated therefore, the
sensible heat input from the air is not significant.
4. Total heat input in BTU/hr
Potential heat from coke 87,595,017
Carbon loss from dissociation 5,751,896
Carbon loss from iron
pickup 3,873,136
Oxidation of iron 111,480
Oxidation of manganese 405,216
Oxidation of silicon 735,840
Total heat input BTU/hr 79,222,521
B. Heat Output
1. Iron melting
Mean specific heat of molten iron 0.2080 BTU
Ib°~F
Molten iron of flowrate 55,946 Ib
RT
Heat content of iron when tapped 55,946 Ib x 0.2080 BTU x
hr TE°F
2692°F
or 31,326,180 BTU/hr
2. Slag
a. Sensible heat of slag
Amount of slag 2,794 Ibs/hr
Mean specific heat 0.320 BTU
Ib°"F
10
-------
Heat content = 2794 Lb x 0.3202 BTU x 2692 F
hr Ibop
or 2,409,119 BTU/hr
b. Heat of reactions
From the heat of formations the reaction
A
CaO + SiC>2 •*• CaO • Si02 Eq. 2
requires 37,800 BTU. Since 100 pounds of CaCC>3 is required
to form one mole of CaO*SiC>2, the energy required per pound
of CaC03 is 378 BTU. Since there are 1,097.6 pounds of CaCO
(from 2a) J
1,097.6 Ib CaC03 x 378 BTU/lb
hr
or 414,892.8 BTU/hr
c. Total heat output from slag
Sensible heat of slag 2,409,119 BTU/hr
Heat of reaction 414,892.8 BTU/hr
Total heat content 2,824,001.8 BTU/hr
3. Decompositon of water
Besides tying up some of the carbon in the decomposition reaction
Eq. '1 some of the energy liberated by combustion is used in the
reaction. Since approximately 2,898 BTU/lb is required
2,898 BTU required x 598 Ibs of H20
Ib H20 hr
or 1,733,004 BTU/hr
4. Stack gas
a. Sensible heat
Knowing the flue gas composition the pounds per hour of each
constituent can be calculated. From Table 1
C02 12.00% - 0.10 moles
CO 14.9% - 0.182 moles
therefore, N2 73.2% - 0.718 moles
Since 5,395 Ibs of carbon are charged (having subtracted out
the appropriate losses) and since
11
-------
C02 + CO = 0.12 + 0.149 = 0.269 moles of carbon
or 3.228 pounds of carbon to produce the stated CO and C02
stack gas contents then the total moles of stack gas produced
are
moles of stack gas x 5,395 Ib of carbon
3.228 Ib of carbon hr
or 1671.3 moles of stack gas.
The stack gas therefore contains
1,671.3 x 0.12 = 200.556 moles of C02
or 8,824.5 Ibs of C02
1,671.3 x 0.149 = 249.02 moles of CO
or 6,972.7 Ibs of CO
1,671.3 x 0.732 = 1223.4 moles of N2
or 34,255 Ibs of N2
Adding in the H2 liberated by the dissociation reaction
H2 = 598 Ibs of H?0 V2 Ibs of H2 = 66.44 Ib
hr 18 Ibs of H20 hr
Using the mean specific heats taken from the tables in Appendix
B for the various gaseous constituents at 900°F
Heat content of C02=
0.2428 BTU x 830°F x 8,824.5 Ibs of CO?
Ib^F Er
or 1,778,348.5 BTU
hr
0.2564 BTU x 830°F x 6,972.7 Ibs of CO
Heat content of CO =
6,972.7 Ibs
hr
or 1,483,874.2 BTU
EF~
Heat content of N2 =
0.2544 BTU x 830°F x 34,255 Ibs of N?
Tb°~F hr
12
-------
or 7,233,011.8 BTU
hr
Heat content of H2 =
3.466 BTU x 830°F x 66.44 Ib of H?
lb°~F hr
or 191,133.2 BTU/hr
Total C02 1,778,348.5 BTU/hr
CO 1,483,874.2 BTU/hr
N? 7,233,011.8 BTU/hr
H2 191,133.2 BTU/hr
Total sensible heat
in stack gas 10,686,367.7 BTU/hr
b. Latent heat
Since the heat of combustion of CO is 4,346 BTU/lb then
6,972.7 Ib of CO x 4,346 BTU liberated
hr Ib of CO
or 30,303,354.2 BTU
hr
Total stack gas heat content = sensible + latent =
10,686,368 + 30,303,354 = 40,989,722 BTU/hr
5. Radiation from cupola
The radiation from the cupola cannot be adequately measured.
It is taken to be the difference between the heat input and output
from the cupola.
a. Heat input 79,222,515 BTU/hr
b. Heat output
1. Iron 31,326,180 BTU/hr
2. Slag 2,824,012 BTU/hr
3. Decomposition 1,733,044 BTU/hr
4. Stack gas sensible 10,686,368 BTU/hr
5. Stack gas latent 30,303,354 BTU/hr
Total heat output 76,872,954 BTU/hr
c. Radiation from cupola 2,349,557 BTU/hr
13
-------
As can be seen in this example 40,989,722 BTU per hour are lost
by exhausting the stack gas to the atmosphere. This amounts
to approximately 52% of the heat supplied to the system.
There are other energy inputs that should be included in an overall
energy balance of a cupola system. For example, there is the
energy expended in the charging system and other auxiliary equip-
ment. There are two additional energy inputs which are inherent
inputs to the system although they are not usually considered
in a normal energy balance. They are the ID fan and blast blower
energy requirements and the energy expended by the afterburners.
A cupola's fan horsepower requirements could go into the thousands
depending on the exhaust volume and pressure drop. As will be
explained in the section on air pollution the exhaust flow from
a cupola could run five fold over what the blast rate is due
to deliberate infiltration. If a venturi scrubber is used for
pollution control the system pressure drop would easily exceed
fifty inches of water. If we assume a 1500 hp fan this would
add another 3,819,600 BTU per hour for a total energy input of
83,042,115 BTU/hr.
Afterburners in a cupola system of the size used in the example
usually run 15 to 25 million BTU per hour. Usually after about
two hours, however, they can be shut down and a pilot burner
is sufficient. The( pilot usually is in the range of 3 million
BTU/hr. Applying these data to the cupola example for a 10 hour
operational period the afterburner would have an average energy
input to the cupola system of 6.4 million BTU per hour. This
would raise the total energy input of the example to 89,442,115
BTU/hr. The afterburner input adds directly to the sensible heat
of the stack gases increasing this percentage of input energy
to 53 percent. As will be seen in subsequent sections foundries
can no longer allow this energy to be wasted.
AIR POLLUTION CONTROL
Air pollution regulations have caused major difficulties for
the foundry industry due to the fact that the emissions from
a cupola have inherent properties which make them difficult to
control. Not only is the particulate concentration in a cupola's
exhaust high but also the particle size of the dust emitted
is usually below 10 microns in diameter (see Table 3). To control
this within acceptable limits, a pollution control device must
be highly efficient over the full range of particle sizes encountered
(usually an efficiency of 90% removal is the minimum required,
while a removal efficiency of 99% is a common requirement).
This efficiency requirement limits the choice of control equipment
to three devices; namely, electrostatic precipitators, high energy
scrubbers, and baghouses. Relatively simple devices such as
14
-------
Table 3. PARTICLE SIZE DISTRIBUTION-CUPOLA EMISSIONS (2)
Cumulative Percent by Weight
Diameter in Microns
Plant
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
-1 -2 -5
30% 50%
64 82
2
13 28
54
14
0.6 2
4
11
8
18
17
24
26
0 7 25
0 7 24
-10
65%
98
12
45
86
15
3
5.5
13
12
25
26
28
30
32
41
-20
82%
99
34
55
98
15
19
8
7
32
17
38
36
23
32
34
47
-50
90%
92
60
99
21
25
99
99
13.7
53
28
62
53
42
44
41
32
-100
99%
99
99
99
99
99
99
75
75
69
56
69
-200
99%
99
80
94
69
61
81
15
-------
wet caps and cyclones are eliminated. While all three of the
applicable pollution control devices mentioned are somewhat
hampered by these requirements, the effect is least upon the
baghouse. For further information on these devices the reader
is referred to Reference 2.
Finally the exhaust gases from a cupola have a carbon monoxide
content averaging around 13% . To comply with air pollution codes
and to avoid the possibility of explosions in the ductwork, the
exhaust gases are afterburned to convert the CO to C02- The
temperature of the exhaust gas is therefore usually in the range
of 2000QF by virtue of the CO combustion. This temperature is,
of course, far above what any pollution control device can handle
and consequently must be reduced. The two most common methods
of accomplishing this are dilution with ambient air and water
quenching. Both of these methods, however, present further pro-
blems. While diluting with ambient air is an effective method
of reducing the exhaust temperature it also significantly increases
the gas volume through the system. Since the cost of pollution
control equipment is directly related to the air volume to be
treated, this method results in a large cost increase. Water
quenching systems, while effective to a degree, are not useful
in reducing the exit temperature by more than about 500°F. Water
quench systems are also frequently incorrectly designed such that
not all of the water flashes into steam. This excess water
causes damage to the refractory and ductwork. If a baghouse
is used even with a well designed water quenching system, cool
spots in the baghouse could allow condensation of the water vapor
on the bags. This increases the weight on the bag, causes caking,
and bag failure. Due to the aforementioned reasons neither system
is used exclusively but rather a combination of the two techniques
is employed.
With the necessity of employing a highly efficient pollution
control device which incorporates a temperature reduction mode
the cost of such a system becomes an important consideration to
the foundry management. Table 4 shows the economics of air poll-
ution control for the three most prevalent devices. As noted,
the cost of any pollution control device is dependent on flow.
Using ambient air as a temperature controller can increase the
flow by a factor of four or more. The cost of controlling the
pollution from a large cupola runs into the millions of dollars.
Costs for smaller cupolas are often in the neighborhood of one
million dollars. Added to these is the cost of installation
which in many cases can be extremely high. This is due to the
fact that retrofitting a control device to an existing installa-
tion is difficult because of lack of available space or other
constraints. Often it becomes necessary to suspend equipment
from a superstructure since ground space is not available. Opera-
erational and maintenance costs for these devices is also quite high.
Venturi scrubbers, besides using large quantities of water which
16
-------
Table 4. COST EQUATIONS (2)
Equipment
Type
High energy wet
scrubber
Low energy wet
scrubber
Fabric filter
Investment Cost
Equation
1=49,519+2.84 gas
vol
1=43,519+8.97 gas
vol
1=38,744+2.05 gas
vol
1=55,000+8.95
vol
Limits of
Observation
6,000
-------
subsequently must be cleaned, have pressure drops frequently
in the range of 55 inches of water. Baghouses, while only contri-
buting about 6 inches to the pressure drop, typically have a
bag failure rate on the order of one to three bags per day.
All of these factors can be interpreted to mean that air pollution
control for a cupola will be expensive in terms of initial invest-
ment, operation, and maintenance costs.
While large nationwide foundries may be able to afford the expense
of pollution control, the smaller foundries usually have diffi-
culty. Unfortunately, there are too many instances where foundries
have spent money for a pollution control system only to discover
during performance tests that the equipment is not as efficient
as design specifications called for. Although this is sometimes
due to careless engineering on the part of the pollution control
companies more often it is the nature of the cupola that causes
the problems. Underdesigning of systems has been all too common
in the past; it has been necessary to scrap the entire system
and start anew in some cases.
In summary, numerous problems are associated with air pollution
control for the cupola. These problems necessitate expensive
solutions. While the need to protect the environmental quality
is accepted, the question of how to do so in the most effective
and economical way does not have a universal answer.
FUEL COSTS
The second major problem to confront foundries in recent years
is the sudden and drastic increase in the cost of coke. The
increase in coke prices over the last eight years is graphed
in Figure 2. The increases from 1968 to 1973 were caused by
two major factors; the Mine Health and Safety Act of 1969 and
air pollution regulations limiting the amount of sulfur in coal.
These two factors increased the price of the coal available for
production to coke thereby increasing the price of coke also.
The increase over this period averaged $5 per ton per year.
From 1974 to the present, however, coke prices have jumped enor-
mously, due primarily to the increase in oil prices. The increase
over this period averaged $3.45 per ton per month. This rapid
increase appears to be flattening out at the present with the
price of coke averaging $115 to $120 per ton. The complexities
of the world fuel situation make any fuel cost projection untrust-
worthy. It is extremely unlikely, however, that the prices will
go back down or even remain stable for very long.
18
-------
140
120
Figure 2. Price of coke 1968-1975.
-------
SECTION V
RECUPERATIVE HOT BLAST
INTRODUCTION
The idea of heating the combustion air of a cupola was known
over a hundred years ago. It was not until the 1950's, however,
that the idea began to be seriously investigated. It has been
shown that a hot blast has many advantages to its use including
a decrease in coke consumption and melting losses and an increase
in melting rate and iron temperature. It has also been reported
that hot blast enables the use of lower carbon contents in the
raw materials and allows for less sulfur pickup.
The decrease in coke requirements is the most important aspect
of hot blast. The energy in the air substitutes for energy
liberated by the coke, thereby reducing the amount of coke required.
The relationship between the degree of preheat and coke reduction
is shown in Figure 3. For a set tapping temperature a 100°F
increase in the blast air makes possible a 0.40% decrease in
the charge rate in the amount of coke used.
Example: Given the conditions in Table 5, what fuel cost savings
can be realized in a year's time with blast air at 1000°F?
2275 Ibs x .004 Ibs of coke reduction x 930°F
charge100°F
or 84.63 Ibs of coke saved per charge or 1.27 tons of coke
saved per hour. This is equivalent to 4,762.5 tons of coke
per year or a cost savings of $533,400 per year.
Instead of utilizing the hot blast to decrease coke consumption
it can be used to increase the metal tapping temperature. The
increase in the metal temperature is approximately 15°F for
every 100 F increase in blast air temperature.
Another advantage of the hot blast is to increase the productivity
of the cupola by increasing the melting rate. The melting rate
of a cupola is a linear function of the coke rate (i.e. percentage
of coke in charge). If hot blast is used and since the combustion
rate of coke is a constant then by decreasing the amount of
coke charged the melting rate will be increased in proportion
to the hot blast temperature. A decrease in the coke rate,
however, would result in a lower tapping temperature. This
can be compensated for, however, by a sufficient increase in
air temperature. As can be seen in Figure 4, for a given tapping
temperature the melt rate will increase. In such a case the
air volume must be the same despite a coke rate decrease since
20
-------
300 t
275 '
V)
CD
O
LL
O
O
cr
ID
O.
O
LU
CE
O
O
LU
IK
LU
*
O
O
250-.
225--
200
I ' 1 ' h-
468
AIR TEMPERATURE,F°x 100
10
12
Figure 3. Effect of hot blast on coke consumption .
21
(3)
-------
Table 5. EXAMPLE OPERATING DATA
Melting rate 30 tons per hour
Coke per ton of iron per
charge 275 Ibs
Ambient temperature 70°F
Hours per day of operation 15
Days per year operation 250
Cost of coke $112 per ton
22
-------
U)
14
13
12
I II
o
2 10
o
o>
c
200
54-in. cupola
Fixed tapping temperature.
400 6OO 800
Air temperature , F°
1000
1200
Figure 4. Effect of blast-air temperature on melting rafe .
( 3)
-------
the coke combustion rate must be held constant. If, however,
an increase in melting rate is not desired then benefit can
be attained by a decrease in the blast rate. This is obvious
since the amount of coke is decreased.
Silicon melting loss is also dependent on hot blast temperature.
An increase in the air temperature of 100°F appears to decrease
silicon loss by 1.5%. This is only a general empirical relationship
and can be affected by nature of the charge, slag composition,
etc. Sulfur pickup is also decreased by hot blast operation
since it is dependent on the amount of coke used and this, of
course, is decreased by the use of hot blast. There is also
a decrease in the melting loss of iron from hot blast operation.
This decrease is approximately 1% when converting from cold
to hot blast.
Another major advantage to hot blast is reduced emissions from
the cupola. This occurs due to either a reduced coke rate or
blast volume. While extensive research has not been performed
to correlate these factors with emission rates, limited work
has been performed (2) and the following regression equations
have been developed:
E = 57-6.6C + 0.1B for unlined cupolas Eq. 1
where E = particulate emissions in Ibs per ton of melt
C = metal to coke ratio
B = specific blast rate in SCFM per square foot furnace area
and E = .05 + .07B for acid lined cupolas
While the equations cannot be considered accurate enough for
purposes of prediction they do show the relative effect of the
variables.
Example: From the conditions listed in Table 5 and with B equal
to 300 SCFM/sq.ft. of furnace area, what % reduction in emissions
might be expected by converting to hot blast?
As before the amount of coke saved is 84.63 Ibs. This increases
the metal to coke ratio from 7.3 to 10.5. With a constant B
factor then using equation 1 the resulting decrease in emissions
is from 38.82 pounds per ton of melt to 17.7 Ibs, a decrease
in excess of 50%.
There are presently three methods available for preheating the
combustion air; namely, gas assisted firing, direct fired heat
exchanger and recuperative heat exchanger. The gas assisted
cupola is not generally recommended since natural gas while
cheap is scarce and also because gas firing in the cupola is
inefficient.
The direct fired heat exchanger has been the most popular method
used in this country. This was primarily due to two reasons.
24
-------
First, the heat exchanger necessary could be ordered off the
shelf since there were not any severe conditions (such as heavy
grain-loading) to contend with. Second, the price of oil made
the choice economically feasible. With, however, the rapid
increase in fuel oil prices, this can no longer be considered
a viable alternative. It appears, therefore, that the recuperative
heat exchanger is left as the most likely alternative.
RECUPERATORS
The idea of using the latent and sensible heat of the cupola
off-gas to preheat the combustion air has been achieved by recupera-
tive hot blast. A general operational schematic for accomplishing
this end is presented in Figure 5. Variations on this schematic
will be reviewed later in this section. The cupola off-gas
contains energy in two forms. The first is the sensible heat
of the air which is directly related to the temperature. The
second form is the latent energy which represents the CO content
of the gas stream. As can be seen from Figure 5 the cupola
off gas is mixed with ambient air infiltrated through the charge
door to provide oxygen for combustion and to lower the final
temperature by dilution. The gas-air mixture is immediately
afterburned to convert the CO to C02- Depending on the amount
of air added for dilution the temperature of the gas stream
exiting the afterburner is usually in the range of 1500 to 2200°
F. The gas stream temperature is then further reduced by water
quenching. In an effort to reduce the particulate grain loading
of the gas stream entering the heat exchanger a low pressure
drop cyclone is sometimes included in the system. The cyclone
will remove the large particles which may cause some damage
to the exchanger and also helps to reduce somewhat the possible
fouling of the heat exchanger surfaces. The gas stream then
is further cooled by the heat exchanger, followed by another
water quench to reduce the temperature to the design limits
of the air pollution control device. Since the most prevalent
control device for the cupola is a baghouse then the temperature
must be reduced to below 500°F. The temperature cannot drop
below 212°F or else condensation will occur.
There are four major modifications that can be performed on
the basic system. They are above or below charge door takeoff
and pollution control before or after afterburning.
The advantage to above charge door takeoff is that the infiltrated
air acts to reduce the temperature of the gas stream while supplying
air for the CO combustion. There are, however, two main drawbacks
to its use. First, if the ID fan is not sized properly dust
ladened air can escape through the charge door opening. The
second and primary disadvantage is that the increase in the
amount of air to be cleaned is usually fivefold that blown
through the tuyeres. Since pollution control cost is dependent
on the air volume to be treated this increase in flow has a
significant effect upon both capital and operating cost. One
25
-------
NJ
CUPOLA
AMBIENT AIR
FROM
CHARGE DOOR
NATURAL GAS
I
PARTICULATE
H20
AFTERBURNER —0
HEATED
Al R
WATER
GAS
AMBIENT
, AIR
t
PARTIAL | [ HEAT
CONTROL f "]
I
QUbNCH
H20
t
WATER
QUENCH
i
AIR P<
CONTROL
PARTICULATE
I
DISPOSAL
TO
STACK
Figure 5. Schematic of heat recovery system
-------
method of reducing the amount of infiltration is to minimize the
size of the charge door opening.
The principal advantage to below charge door takeoff is that
the amount of air infiltration into the system is much less than
that infiltrated with above charge door takeoff. This could,
of course, reduce the cost of pollution control and the horse-
power requirements of the ID fan. There are basically two differ-
ent ways of accomplishing below charge door takeoff. Normally,
the takeoff point is placed below about five feet of burden.
This would reduce the amount of infiltration by about half.
Another method is to place about 25 feet of burden over the
takoff point reducing infiltration to approximately zero. This
enables particulate removal to occur prior to afterburning.
Heat exchangers in such a system can be conventionally designed
without special provisions for fouling.
AVAILABLE SYSTEMS
Due to the relative lack of recuperative hot blast installations
in this country a survey of manufacturers and installations
was made. The purpose was to determine if available systems
were reliable. The following firms were identified as manufacturers
having installations in this country: Escher, Inc.; McQuay
Perfex;, Thermal Transfer; American Schack and Combustion Equipment
Associates. Other firms which have sold systems to foundries
which are not yet installed included F.E.G.O.K. and Q-Dot.
Finally, some engineering firms having experience with the cupola
have designed systems but have yet to market them extensively.
An example of this type of firm is Smith Environmental Corpora-
tion. An attempt was made to visit representative foundries
which had all of the various manufacturer's equipment installed
as well as receiving engineering descriptions of those systems
in the planning stages. Permission to visit the Allegheny Ludlum
Steel Corporation which had the only American Schack equipment
installed in this country was not granted for proprietary reasons
involving their duplexing process. Also, Q-Dot Incorporated
did not furnish any engineering information on their equipment.
All other requests for vital information and visits were granted.
Escher Incorporated
Escher's equipment accounts for 17 out of the 20 known recuperative
heat exchangers installed in this country. They supply both
the conventional "above charge door" takeoff system as well
as "below charge door" systems. They have both types of systems
in use in this country. A listing of a representative sample
of Escher's pricing can be found in Table £. These costs are
for recuperators accepting 100% of the flue gases and preheating
the combustion air to 950°F. These costs do not represent
installation costs which can vary appreciably depending on
the foundry arrangement.
The Escher heat exchanger is a counterflow type utilizing
27
-------
TABLE 6. RECUPERATIVE HOT
No. Installation Location Start-up
1 Glamorgan Pipe
Foundry
Chevrolet Grey
2 Iron Foundry
(1) Chevrolet
Grey
3 Iron Foundry
(2) Chevrolet
Nodular
4 Iron Foundry
Chevrolet
Grey
5 Iron Foundry
(3)
6 Auto Spec.
Mf. Co.
Chevrolet
Grey
7 Iron Foundry
(4) Chevrolet
Grey
8 Iron Foundry
(5)
9 U.S. Pipe &
Foundry
10 U.S. Pipe &
Foundry
11 U.S. Pipe &
Foundry
12 U.S. Pipe &
Foundry
13 East Jordan
Iron Works
14 Kelsey-Hayes
Gunite
15 U.S. Pipe &
Foundry
16 Chrysler Corp.
17 Eljer Pluming
Ware
18 Berlin-Chapman
Foundry
19 Allegheny-
Ludlum Steel
Corp.
20 U.S. Pipe &
Foundry
Lynchburg, 1969
Va.
1970
Saginaw,
Michigan
Michigan
Chattanooga,1971
Tennessee
Birmingham, 1972
Alabama
Bessemer, 1972
Alabama
Burlington, 1973
New Jersey
East Jordan,1973
Michigan
Rockford, 1974
Illinois
Anniston, 1975
Alabama
Fostoria, 1975
Ohio
Salem, .1976
Ohio
Berlin, 1971
Wisconsin
Pittsburgh, 1966
Penn.
Union City, 1960
Cal.
BLAST SYSTEMS
Manufact u r er Capacity T/H
Escher 45
Escher 35
Saginaw,
Michigan
Saginaw,
Michigan
Saginaw,
Michigan
Benton
Harbor ,
Michigan
Saginaw,
Michigan
Saginaw,
1970
1970
1970
1970
1971
1971
Escher
Escher
Escher
Escher
Escher
Escher
35
65
65
45
65
65
Escher 40
Escher 70
Escher 70
Escher 70
Escher 55
Escher 40
Escher 25
Escher 32
Escher 15
McQuay-Perfex 11
American Schack N.A,
Thermal Transfer 40
28
-------
Table 7. ESCHER RECUPERATOR PRICE LIST
FOR BLAST PREHEAT OF 950°F
Blast Volume
SCFM
2000
4000
6000
8000
13,500
19,000
Melt Rate
TPH
5
10
15
20
32
40
Cost
$ 35,000
49,400
68,500
86,800
132,900
171,500
29
-------
a radiation and convection section. The inner shell is surrounded
by an outer shell and a small portion of the blast air flows
between them. The major portion of the blast air flows through
the hollow fins located in the gas stream. The preheated blast
air is collected in a circular chamber prior to entry into
the blast piping leading to the cupola (see Figure 6).
To accomodate the expansion of the blast heater a flue gland
is provided between the heater and the flue gas inlet duct.
A flexible hot blast duct is also provided.
The hot blast temperature is maintained by controlling the inlet
flue gas temperature or by controlling the flue gas flow through
the heat exchanger.
Escher's equipment can be used in series or parallel applications
and can be followed by any type of pollution control device.
The materials of construction for the Escher heat exchanger are
a combination of stainless and mild steel to minimize cost.
One of the major points to be considered about any recuperative
device is how it is affected by the particulate grain loading.
The Escher units are self-cleaning devices by virtue of the
vertical arrangement, tube design and gas velocity through the
exchanger. Particulate also has an effect upon the allowable
incoming temperature since ash from the cupola can start to
form sintered deposits at 1800°F.
Plant Survey-
U.S. Pipe and Foundry, Burlington, N.J. Plant data can be found
in Table 8 and a schematic of the system in Figure 7. As can
be seen the parallel Escher units deliver approximately 25,000 SCFM
at 950°F to the cupolas. The hot blast allows for a 10 to 1
metal to coke ratio. The flue gas from the afterburners (modulating
control) goes to three spray nozzles which are temperature con-
trolled to reduce the temperature to below 1500°F. The heat
exchangers are one stage units with the exiting gas going directly
to the final quench and then to the baghouses.
When the plant visit was made the units had been in operation
for over a year with no sign of deterioration. A maintenance
check during the foundry's vacation period showed no sign of
deterioration in the structure or the material of the exchanger.
Some rubber seatings did have to be replaced but according to
the foundry personnel this was a quick inexpensive procedure.
Plant Survey-
There are five Escher units installed at the Chevolet Foundry,
CMC, Saginaw, Michigan. All units are of the "below charge
30
-------
OVER-ALL LENGTH _OFJHEATER _
CONVECTION SECTION (FINS)
COLD I BLAST
f INLET
u;
HOT I BLAST
OUTLET
HEATER INSULATION
FIN
INNER SHELL
OUTER SHELL
PART SECTION "A-A"
Figure 6 . Escher recuperative blast healer .
Courtesy of ESCHER, INC.
-------
Table 8. PLANT DATA, U.S. PIPE AND FOUNDRY COMPANY,
BURLINGTON, N.J.
Type of Product Ductile
Melt Rate TPH 58
Blast Rate SCFM 25,000'
Blast Temperature °F 950
Oxygen Enrichment % 2-4
Metal to Coke Ratio 10/1
Exhaust Flow Rate SCFM 67,800
Exhaust Temperature OF 1,800
Charge Door Opening Ft^ 25
Afterburner 106 BTU/hr 30
CO Content % 13
32
-------
FLUE GAS
FLUE
GAS
FINAL
QUENCH
VESSEL
FAN
FAN
BAG
HOUSE
BAG
HOUSE
FAN
BAG
HOUSE
BAG
HOUSE
FAN
Figure 7. Plant schematic , U.S. Pipe 8 Foundry Co., Burlington , New Jersey
33
-------
door" takeoff type. The first unit was installed in the late
sixties and performance after some initial difficulties has
proved satisfactory to the foundry management. There appears
to have been some trouble with the first installation primarily
due to excessive flue gas temperatures entering the heat exchanger.
After this was corrected the unit performed as designed and
there have been no problems with any of the other units. Routine
maintenance procedures have eliminated the need for any extensive
overhaul.
An innovative approach to energy conservation had also been taken
by GM engineers by using the energy from one of their cupolas
for plant heating. Approximately seventy million BTU per hour
are made use of in this manner contributing to substantial savings
in fuel cost. Plant data for the two different size cupolas
can be found in Table 9.
The arrangment of the system can be found in Figure 8. Due to
the size of the cupolas, however, it was necessary to install
a by-pass cooler. It is the heat from the by-pass and evaporative
heat exchanger that is being used to supply the plant heating.
An interesting point to note concerning the operation of these
cupolas is the high coke rate. According to the foundry engineers
this is due to the fact that there is no pig iron in the charge
material. As a result the carbon pickup must come from the coke
and thus the high coke rate. It is also interesting to note
that the blast temperature is given in a 400°F range. Again,
metallurgical considerations are the prime reasons for a given
blast temperature and the cupola operators have greater control
of the final product by manipulating the blast temperature. All of
the Escher systems are capable of operating over a blast temperature
range of approximately 400°F. The degree to which this range is
used depends upon the particular foundry operating practice.
McQuay Perfex Inc.
Perfex's recuperative heat exchanger was designed to provide
hot blast to a cupola with a minimum of maintenance. The heat
exchanger consists of 14 tubes, 92 inches long and 12 inches
in diameter. The tubes are equipped with both internal and
external fins to aid in heat transfer. The internal fins through
which the dust laden off-gas from the cupola is forced are de-
signed to present an inclined spiral shape to the gas flow.
These fins are cleaned daily by an air lance and can be used
during operation. A representation of the inner fin arrange-
ment can be seen in Figure 9.
McQuay Perfex presently offers two sizes of heat exchangers.
Data on the size of the units can be found in Figure 10 and
data on pressure drops, flow and blast temperatures is presented
in Figure 11. The cost of the two units is $30,000 for the
smaller unit and $41,000 for the larger.
34
-------
Table 9. PLANT DATA, CHEVOLET FOUNDRY, CMC,
SAGINAW, MICHIGAN
Type of Product
Melt Rate TPH
Blast Rate SCFM
Blast Temperature °F
Oxygen Enrichment %
Coke Rate %
Exhaust Flow Rate SCFM
Exhaust Temperature °F
Afterburner 106 BTU/hr
CO Content %
I.D. Fan HP
Grey
35
18,000
600-1000
Max. 4
18
60,000
2000
17
18
350
Type of Product
Melt Rate
Blast Rate
Blast Temperature °F
Oxygen Enrichment %
Coke Rate %
Exhaust Flowrate SCFM
Exhaust Temperature °F
Afterburner 106 BTU/hr
CO Content
I.D. Fan HP
Grey
55
25,000
600-1000
Max. 4
18
70,000
2000
17
18-20
1250
-------
COMBUSTION
CHAM8CR
BLAST
HCATER
I
LU
CUPOLA
- ACCESS
FLUE CAS DUCT
MOT AIR
OISCHARCE
FLUE SAS COOLER
COOLING
AIM FAN
^//////>>>i£J$^'/^
Figure 8 - Typical Escher Recuperative Arrangement
Courtesy of Escher, Inc.
-------
Figure 9. Inner tube fin arrangement.
37
Courtesy McQuoy-Perfix
-------
M5^
C2
P2
A
2'-0"
2 '-6"
B
3'-0"
4'-0"
c
6"
6"
D
I'-O"
r-o"
E
19'- 0"
23'- 0"
F
5'-0"
4'-9"
G
3'-0"
3'- 6"
H
I3'-0"
I4'-0"
J
3'-0"
4'-0"
K
73"
85"
oo
oo
CLEANOUT
OPENING
HOT BLAST
INLET/OUTLET
HOT BLAST
INLET/OUTLET
t
J
i
-K-+-H-
CLEANOUT
OPENING
CUPOLA GAS
" INLET/OUTLET
INLET/OUTLET
B DIA.TYP.
Figure IO. Dimensions, McQuay-Perfex recuperator
Courtesy M c Quay-Perf
-------
o
.
P
00
HI
LU CC
N V)
N CO
O LU
14-
12-
10-
8-
6-
4-
2-
PRESSURE LOSS Vs. FLOW
HOT
BLAST
a/
2,000
4,000
6,000 8,000
FLOW. SCFM
ASSUMPTIONS:
I. Hot blast inlet temp. = 0°F
2. Cupola gas inlet 1emp.= I200°F
3. Unit 85% clean
4. Flow as shown schematically
10,000
HQT BLAST
CUPOLA GAS
800
i-
LU
CO
So
°0
CO U.
< Z
CD
s«i
LU LU
K- cr
1 i-
x a;
§£
0- 2
a. LU
< h-
750 -
700 -
O 650 -
600 -
550 -
500'
HOT BLAST OUTLET TEMPERATURE Vs. FLOW
CONFIG.I CONFIG.
C-2 D-2
3.0
I
2.8
2.6 2.4
CUPOLA GAS FLOW /HOT BLAST FLOW
I
2.2
-7
g a.
z o
i- °
LU tt
'4 N (/>
N (/)
O LU
z a:
V)
-3
-2
_ i a. r>
r ' < o
12,000
850
-800 fc
^-750 ,_
o
< u
_i •
CD U.
-700
-650
-600
LU
cr
S^
a.
550
2.0
Figure II. Pressure,temperature relationship for McQuay-
Perfex recuperator .
39
Courtesy of McOuay- Perfex
-------
Plant Survey-
Berlin-Chapman Foundry, Berlin, Wisconsin cupola operating data
can be found in Table 10. The recuperative heat exchanger has
been in operation for four years with no sign of deterioration.
The total cost of cleaning the heat exchanger for a year's period
is estimated at $3,000. A schematic of the system layout is
depicted in Figure 12. The pollution control device is a scrubber
type system with the trade name "Air Sweep" and it is marketed
by McQuay Perfex Inc.
The heat exchanger is composed primarily of carbon steel materials
with the exception of the fins which are stainless steel. The
hot blast operation has increased the melting rate of the cupola
by 25 percent. Due to the increase in melting rate no change
in the coke percentage was achieved.
Thermal Transfer Corporation
Thermal Transfer Corporation has one installation of a heat
recovery system on a cupola. The installed unit has been in
operation for over fifteen years and the design is a conventional
shell and tube type with no inherent design features to overcome
fouling.
Plant Survey-
U.S. Pipe and Foundry, Union City, California cupola operating
data is presented in Table 11. The main disadvantage of this
system is that it is not self-cleaning. A shot-cleaning system
is employed and there were numerous start-up problems associated
with it. Subsequent problems with the system would indicate
that such a design should not be used in the future. U.S. Pipe
and Foundry and Thermal Transfer should be commended, however,
for being farsighted enough to install a heat recovery unit
and make it work when they did.
American Schack Company Inc.
At present American Schack has one recuperator installation
in this country at Allegheny Ludlum Steel Corporation. Due
to the proprietary nature of the process permission to visit
the facility was not granted.
The installed recuperator is a below charge door type which
has the take-off point below approximately 15 feet of charge
material. The off-gas is then water quenched and cleaned by
use of a scrubber type system marketed under the name of the
"Theisen Disintegrator" by American Schack.
40
-------
Table 10. PLANT DATA, BERLIN-CHAPMAN FOUNDRY,
BERLIN, WISCONSIN
Type of Product
Type of Cupola
Melt Rate TPH
Blast Rate SCFM
Blast Temperature °F
Coke Rate %
Exhaust Flow Rate SCFM
Exhaust Temperature °F
Pressure Drop Across Recuperator
(Hot Side) Inches of Water
Pressure Drop Across Recuperator
(Cold Side) Inches of Water
Grey, Ductile
#5, Whiting
11
4,000
700
13.3
13,462
1200
3.4
4.0
41
-------
Clean exhaust to atmosphere
•—i
^r
NJ
I. AIR SWEEP
Z. PRE-QUENCH
3. PRE-CLEANER CYCLONE
4. CYCLONE DUST COLLECTOR
5. PERFEX RECUPERATO
6. VENTURI WASHER
7. SEPARATOR
8. PRIMARY SUCTION FAN (ZOO H.P.)
9. STACK
IO. RECYCLE PUMP
I I . HYDRO CYCLONE
12. CLASSIFIER
13. SLAG AND CUPOLA DIRT COLLECTOR
14. BLAST INTAKE STACK
15. BLAST AIR BLOWER (50 H.P.)
16. HOT BLAST TEMPERATURE CONTROL SYSTEM
17. HOT BLAST DUCT
Slag* H20 —^--y_-_
\
Figure 12. Recuperator schematic, Berlin-Chapman Foundry
Courtesy of McQuay-Perfex, Inc.
-------
Table 11. U.S. PIPE AND FOUNDRY, UNION CITY,
CALIFORNIA
Melt Rate TPH
Blast Rate SCFM
Blast Temperature °F
Coke Rate :
Exhaust Flow Rate SCFM
Exhaust Temperature °F
Afterburner 106 BTU/hr
CO Content %
30-40
11,000-13,000
600-900
8.2
50,000
1400
3
Not available
43
-------
The cleaned gas has a heating value of approximately 50 BTU/SCF.
All of the gas is combusted by an ignition burner of 500,000
BTU/hr capacity. This burner is in constant operation to insure
complete oxidation of the CO. Exhaust gas is continuously analyzed
for CO, H2 and 02 both for combustion and safety considerations.
In the case of high 02 or H2 concentrations a valve is automatically
activated exhausting the cleaned gas to the atmosphere.
The recuperator at Allegheny Ludlum has been in use since 1966
and has not needed any major repair work outside of tube replacement,
The major benefit to pollution control prior to combustion is
that the heat exchanger does not need any special design consi-
derations to overcome fouling.
F.E.C.O.R. Industries
This system is similar in concept to American Schack in that
the exhaust gas is cleaned prior to combustion. A schematic
of the F.E.C.O.R. system is presented in Figure 13. As can be
seen the gas take-off point is below the charge door. There
is approximately twenty feet of charge material above the take-
off point. The exhaust gas from the cupola is water quenched
to about 160°F prior to entering the venturi scrubber. The
gases are monitored for 02, H2 and CO and a sophisticated control
system is used to switch the gas stream into a by-pass mode
if so indicated. The by-pass system includes an afterburner
to combust the CO. During by-pass mode operation or during
start up and shut down, hot blast operation can be accomplished
by firing the heat exchanger with oil or natural gas.
The F.E.C.O.R. system is being installed at the Central Foundry
Division of General Motors Corporation in Defiance, Ohio. Design
data for the proposed facility can be found in Table 12. The
F.E.C.O.R. system is not suited for retrofitting to existing
installations due to the necessity of keeping a twenty foot
charge burden above the take-off point. The cupola must be
designed to hold the charge and constructed to handle the extra
weight. Also, the top of the cupola must be able to be sealed
during melt down after the charge is dropped to prevent oxygen
infiltration.
i
Combustion Equipment Associates, Inc.
The original intent of the EPA contract under which this report was
written was to conduct an extensive testing and evaluation program
on the integrated energy recovery-air pollution control system
purchased by the Flynn and Emrich Company from Combustion Equipment
44
-------
Associates (C.E.A.)- Reference 4 contains a detailed description
of the system including projections of the operation cost savings.
The system, however, was never made operational due to severe
design deficiencies centering on undersizing of the ducts and
fan. Reference 5 discusses these shortcomings in detail and
and describes the instrumentation selected for the test program.
The computer model which was to be verified with the test data
and then used to extend the results to other operations is also
described.
The novel feature of the F&E installation was the dry media heat
exchanger which employed ceramic balls as the heat transfer
media. Projections indicated that such a device would allow for
operation at a higher temperature and would result in a more
compact unit than other types of heat exchangers. Since the
unit was never made operational, no definite conclusions can be
drawn. However, the basic concept is still valid and is worthy
of further investigation. CEA is no longer marketing the
system. Mr. Paul Vandenhoeck, holder of the patent on the heat
exchanger, is now with Richmond Tech-Air Corporation, Mountainside,
New Jersey.
ECONOMIC EVALUATION OF RECUPERATIVE HOT BLAST
There are two primary economic advantages to recuperative hot
blast operation. First, with hot blast operation a coke reduction
of 20% can be experienced for an average cupola. Second, with
recuperation, the cost of hot blast operation after initial
capital cost is limited to the fan amperage required to overcome
the increased pressure drop and minor maintenance. To determine
if the capital expenditure for the recuperator is justified a
number of points must be considered. First, it must be determined
that hot blast operation will not adversely interfere with the
product composition and that the space requirements of the recupera-
tor can be met. If the charge material composition remains the
same then a decrease in coke can be expected. A life cycle cost
analysis of the recuperator should then be performed and the results
compared with the savings expected from the coke reduction. A
decision should then be made consistant with the company's fiscal
policy.
The plants visited during this investigation did not include any
which were retrofitted with recuperators but rather the recuperators
were constructed to be fitted to the cupolas from initial start
up of the plant. Data on fuel savings were gathered from other
investigators as was previously mentioned but are not inconsistent
with data given in this report. Those plants which have high
coke percentages usually use what is considered poor charge materials,
The excess coke was utilized to increase the carbon content of
the product.
A point to keep in mind is that when going out for bid the con-
tractor should guarantee not only the equipment but the coke
percentage and product composition. This is a practice common of
the more reliable manufacturers.
45
-------
By-pass Blower
Quencher—\J
JL
Cupola Off-gas
Compressor
Afterburner
Dwell Chamber
ABDC
Blower
Purge Stack
JLj
J — l^
/ \
«
=5
3
•
i
y»
Primary Ai
-i 1.1 i >~
'.
>
1 * " U
i .
Secondary
i
Heater
rJ h-b- r^ 1 .
j u_l! u_^ U — S
^)
"*~^-Combustor
Figure 13. FECOR Recuperative system.
Courtesy of FECOR IND.
-------
Table 12. DESIGN DATA, CENTRAL FOUNDRY DIVISION
Type of Cupola
Type of Product
Melt Rate TPH
Metal Temperature °F
Blast Rate SCFM
Blast Temperature °F
Temperature to Exchanger °F
Oxygen Enrichment %
CO Content %
Minimum Combustion Content of
Exhaust Gas BTU/C.F.
No. 4 Whiting
Grey Iron
63
2700
30,000
1200
4600
2-4
18
60
47
-------
SECTION VI
OXYGEN ENRICHMENT OF THE CUPOLA WIND
INTRODUCTION
It has been well known for many years that a controlled addition
of pure oxygen to enrich the combustion air in gray iron cupolas
can have marked beneficial effects upon the performance of this
type of furnace.
For the most part oxygen enrichment has been used on an intermittent
basis, generally as a corrective measure. It has been found
that the time required to restore normal operating temperature in the
cupola after unscheduled interruptions can be reduced by 50% or
more with oxygen enrichment.(6) The reason for its effective-
ness is that an immediate response is obtainable by using 2 to
4% oxygen enrichment levels without adjustment of the blast velocity.
(7) This is due to the fact that a faster combustion reaction
is brought about by the concentration of oxygen in the blast,
with a corresponding decrease in the nitrogen volume. The nitrogen,
acting as an inert ingredient in the blast air, tends to cool
the flame and carries combustion heat up the stack. Some of this
heat is recovered by the incoming charge, but adding more air
causes excessive gas velocity through the coke bed, increasing
heat loss and giving rise to channeling along the cupola wall
with attendant radiation loss and refractory erosion.
Significant raw material savings have been achieved by reducing
the wind rate and adding 2 to 3% oxygen continuously.(6) This
procedure produces higher flame temperature and a shorter oxida-
tion zone in the oxygen enriched coke bed, providing a melting
atmosphere richer in CO at any given metal-coke ratio without
sensible heat losses from nitrogen. The important result of
this is to melt iron with 15 to 20% less coke than normal without
the metal oxidation which would have occurred with conventional
combustion air rates.(7) The lower velocity of gaseous materials
for each ton of iron melted allows more even burning of the
coke and reduces the danger of channeling.
It should also be noted that the stack emission load is reduced
in proportion to the drop in blast volume. The extent of this
reduction is dependent upon the amount of air influx at the
charging door. An air pollution control system that is marginal
in its ability to treat the volume of stack gas produced by
conventional practice will improve in performance with the
reduced volume provided with oxygen enrichment.
Additional production benefits have been derived by the use
of oxygen enrichment during cupola startups. Case studies have shown
48
-------
that the time from initiation of the furnace blast to the first
tap have been reduced by as much as 58%, with increases in metal
first-tap temperatures of approximately 50°F. (6) In this type
of application the cost of oxygen consumed is more than offset
by savings realized through the reduction or elimination of
iron normally pigged, and a reduction in misruns and faulty
castings produced due to "cold" metal defects after initial
start-ups.
The use of oxygen enrichment on anything other than an intermittent
basis has until fairly recently been limited to the larger foundries
due to the relatively high cost of liquid oxygen as compared
with the cupola raw material savings that could be realized by
the application of enrichment techniques. The unit cost of oxygen
is primarily dependent on consumption levels as illustrated in
Figure 14.
Spiralling raw materials costs in the past five or six years
have, however, forced a change in this philosophy. Improvements
in air separation techniques and plant design have maintained
the cost of liquid oxygen nearly constant over the past two decades,
while foundry raw materials and labor costs have roughly doubled.
In light of these recent developments it can be seen that adopting
continuous oxygen enrichment techniques, which require a relatively
low initial capital outlay, can enable even the small, low output
foundry to obtain significant net cost savings.
DESCRIPTION OF CONTROL SYSTEM
Oxygen can be added to the cupola air most conveniently by inserting
a small bent pipe into the blast main, downstream of the metering
orifice used to measure air flow. This pipe is of a specific
size, depending upon the oxygen flow desired, with a closed end
and a series of drilled holes beyond the bend. These holes are
symmetrically positioned so that the jets of oxygen will enter
the blast main at right angles to the direction of the air flow.
The drilled section is inserted into the blast main parallel
to the air flow and then welded in place. Turbulence in the
wind main and at the tuyeres is generally sufficient to provide
mixing of the oxygen with the incoming blast air.
Due to the highly oxidizing nature of pure oxygen the control
system incorporates an interlocking device which is activated
by the flow of air to the cupola. In the event of sudden elec-
trical or mechanical shutdowns or wind spills a solenoid valve
is activiated to stop the flow of oxygen. Turning on the wind
automatically reestablishes the oxygen supply at a pre-set air
flow rate.
It should be noted that an oxygen enrichment control system such
as the one described above (and illustrated in Figure 15) can,
in some cases, be purchased from the manufacturer as a complete
49
-------
1.00
0.2
0.3 O.4 0.5 0.6 O.70.809I.O 1.5 2.0 3.0
MONTHLY OXYGEN CONSUMPTION (MILLION F3)
4.O 5.O 6.0 7.O 8.0 9.O 10.0
15.0
Figure 14. Oxygen cost vs. consumption .
Courtesy of Air Products and Chemicals
-------
LIQUID
OXYGEN
STORAGE
TANK
BLOWER
MOTOR
ui
OXYGEN
DIFFUSER
I
PRESSURE
REGULATOR
VAPORIZERS
FLOW
ELEMENT
SHUT OFF
VALVE
SHUT OFF
VALVE
—Dft—
PRESSURE
REGULATOR
DROP FOR
AUXILIARY
OXYGEN
PRESSURE
GAUGE
, FLOW
_! CONTROL
VALVE
FLOW
INDICATOR
CHECK
VALVE
SOLENOID
VALVE
Figure 15. Typical oxygen flow control system.
Courtesy of Air Products and Chemicals
-------
package. It can be installed without disturbing normal cupola
operation and without modifications to the cupola itself. The
initial capital investment is nominal when compared with the
savings that can be realized by its effective implementation.
In addition, there is no extensive training period or additional
crew required to operate the flow control system. The procedures
involved can be easily and quickly learned and the manufacturers
generally offer instructional programs with the sale of their system.
INDUSTRY SURVEY
There are numerous industrial firms involved in the manufacture
and supply of liquid gases. Of these companies, however, only
a small number are actively engaged in the design and development
of the oxygen enrichment systems used in grey iron foundries.
In the early stages of this project, communication was established
with a number of those companies which are presently marketing
systems of this type.
One of the first companies contacted was Airco Industrial Gases,
located in Murray Hill, New Jersey. Airco has been promoting oxygen
enrichment technology for several years now. They do not manu-
facture package systems, but they will design the system layout
and supply all the necessary hardware to meet plant specifications.
They will also provide technical back-up during the installation
and start up period and will instruct the operating personnel in
correct procedures for operation, maintenance and repair of the
associated equipment.
Our initial contact was with a Mr. J.W. Estes who is the manager
of combustion metallurgy and development at Airco and has contri-
buted much to the development of the oxygen enrichment technology
being utilized in the Airco systems today. Mr. Estes and Mr. J.
Blessing co-authored an informative paper on "The Potential
for Oxygen Enrichment of the Cupola in the Small Foundry."(8)
In this paper are included some pertinent facts of interest to
foundrymen who are considering economic and operational alterna-
tives to compensate for the recent exponential rise in coke prices.
The most obvious benefit derived from the implementation of oxygen
injection is to effectively increase the average charge ratio
(metal/coke). Airco has conducted a number of in-depth studies
to compare the before and after operating parameters of cupolas
incorporating oxygen enrichment systems developed by Airco. In
these studies coke reductions of 40 pounds per ton to as much as
70 pounds per ton of metal charged have been obtained.(8)
Oxygen enrichment also reduces the effective stack height of the
cupola. Many cupolas have stacks too short for efficient
preheating. In striving for maximum production operators often
increase pressures at the windbox, forcing a greater volume of
combustion gases up the stack. This practice results in excessive
heat losses and large top fires. Oxygen injection corrects these
conditions by increasing the flame temperature which shortens
52
-------
the coke combustion zone. This permits a reduction in the
total wind volume which effectively maintains the sensible heat
in areas where it can perform useful work.
Mr. S. Fredricks, Sales Manager of the Combustion Metallurgy
and Development Division at Airco, points out "that other benefits
that make oxygen injection profitable are higher steel-to-scrap
charge ratios vs. pig iron and castings, and the use of large
section sizes. Of course, this is only true if the cupola is
operated to make use of the increased carbon pickup and spout
temperatures possible with oxygen."
Steel scrap melts at higher temperatures than iron scrap or pig
iron. Much more carbon must be dissolved from the coke to finish
the iron with a proper analysis when higher steel charges are
used. With oxygen enrichment, steel scrap can be increased to
as high as 50% of the charge without an abnormal increase in
coke. Oxygen injection also allows the use of heavier than normal
metal sections. Conventional operation with large scrap, such
as ingot molds, ingot butts or large motor blocks, results in
chilled metal when the pieces reach the cupola well only partially
melted. According to Airco, oxygen enrichment results in complete
melting of these sections without chilling the tapped iron.
Other difficult-to-melt materials such as briquetted borings
can be melted with lower losses.
Another interesting point brought up by Mr. Fredricks is the
fact that, in his opinion, "....the biggest incentive to use new
methods for reducing foundry costs and to increase production
have been pollution control devices."
Common problems encountered in normal cupola operation are bridging
of charges, channeling immediately above the tuyeres, and non-
uniform combustion. The latter is probably caused by the varying
back pressures created by the tail-end systems during the course
of a melt. Nothing can substitute for a well calibrated air volume
control system, however, enrichment of the cupola air with only
2 to 4 % oxygen can help to eliminate bridging and channeling.
By raising the percent of oxygen per cubic foot of air the blast
rate may be lowered by as much as 16% without changing the total
oxygen content. (8) This wind reduction promotes much more stable
combustion and, as pointed out earlier, the lower air volume can
be invaluable when the existing emission control system is incapable
of handling the full wind rates of the cupola.
Mr. Estes suggests (8) that the most effective way to learn the
response of the cupola to oxygen additions is by gradually increas-
ing the level of enrichment. Response of the cupola to oxygen
is not linear but occurs in steps. The first increment to 1%%
will seem to have little effect, but at 2% the effectiveness will
increase rapidly and become most apparent at 2% to 3%. The cupola
53
-------
should be operated for a full day at 1%, 2, 2% and 3% enrichment
to learn the most effective level for a given cupola operation.
Coke cuts will be necessary for maintaining a uniform temperature
and chemistry at the 2, 2% and 3% levels. Operation beyond the
3% level is not recommended unless the demand for additional melt
rate can justify the additional oxygen. In most cases, the most
effective coke savings can be obtained at the 2% to 3% enrichment
level. A number of interesting observations have been made con-
cerning the effect of oxygen on coke rate, melt rate, temperature,
chemistry and lining erosion. (8.)
Coke Rate
A metal/coke ratio is selected to provide the proper spout tempera-
ture and the specified carbon pickup for a given charge make-up .
Low metal/coke ratios are generally required for high tap tempera-
tures and high steel content charges. At high metal/coke ratios
the resultant spout temperatures will be low and carbon pickup
considerably less, thus requiring charges containing pig and scrap
iron in order to ensure sufficient carbon pickup. Oxygen injec-
tion, however, will allow lower coke rates with reductions of
20% being possible using 2^ to 3% enrichment. (8) The reduction
in coke rate also has a distinct effect on melt rate since the
same coke can melt more charge in a given time.
Melt Rate
Melt rate is determined by metal/coke ratio and rate of combustion
is determined by the oxygen driving force. The addition of oxygen
to a level of 23.9% 02 (normal 02 level in air is 20.9%) in the
wind increases the total oxygen content in an amount equivalent
to a 19% increase in wind rate. This oxygen enriched blast tends
to raise the flame temperature of the combustion zone which in
turn accelerates the melting process. A 20% reduction in coke
with the same wind rate also increases the melt rate proportionately
since 20% more metal can be melted with the same coke.(8)
Temperature
Temperature is also related to the metal/coke ratio and the combustion
conditions in the cupola. The use of continuous oxygen will
result in elevated temperatures unless adjustment is provided
by reduction in the coke and/or blast rate. In foundries where
production demands require extended heats and where the tap metal
must be consistently maintained within certain design specifica-
tions several test runs are generally conducted to arrive at
the most effective air/coke rates and tap temperature for that
particular operation.
Chemistry
Silicon loss is more pronounced with low metal temperature and
high coke beds and thus enrichment can be useful in reducing
54
-------
silicon loss in acid lined furnaces. The use of continuous en-
richment in amounts of 2 to 4% can reduce silicon losses suffi-
ciently to allow a reduction of 50% ferrosilicon, equivalent to
5 Ib per ton which could add up to a sizeable savings in raw
materials costs over the course of a year's operation.
Scrap Loss Reduction
With the increasingly high costs of capital investments such
as pollution control systems, manpower and raw materials, reduced
scrap loss is an important consideration to reduce operating
costs. With oxygen enrichment even the first metal out of the
cupola is often hot enough to form thin-section castings, and
need not be pigged for reuse. Airco has found that in foundries
producing specialty castings, the saving of one casting per month
can be sufficient to pay the entire month's oxygen fee.
Lining Erosion
Since erosion of the cupola lining is a function of temperature
and time, it can be stated that a substantial increase in spout
temperature can be expected to result in increased lining wear.
If the temperatures are kept at the normal level by reducing
the blast rate and/or the coke charge the lining wear will be
no greater with oxygen, but the profile of wear may change some-
what. Many acid lined cupolas are operated successfully with
continuous enrichment for as long as 17 hours on a routine basis;
this includes flush tuyere cupolas with limited wind penetrations.
Overtime Costs
Occasionally it has been found that smaller foundries require
certain operating personnel to perform two functions. For example,
certain individuals may make molds in the morning and pour in
the afternoon. In this case, any delays will compel the foundry
to hold the crew over for a few hours. The higher metal tempera-
tures and increased melt rate provided by oxygen enriched com-
bustion air might make up for the loss of time.
Included in the Airco study were a series of interesting case
histories which illustrate the operational benefits of oxygen
enrichment which are being utilized by eight small foundries equipped
with Airco injection systems. These results typify the success-
ful application of the oxygen enrichment process in this segment
of the grey iron industry.
Typical Case Histories - Airco (8)
Foundry A - Melt rate originally 9 T/hr
42" lined cupola - cold blast
Metal/coke ratio 6.6:1
55
-------
Foundry A realized a melt rate increase of approximately 20% with
an enrichment level of 2% while an increase of 30% was achieved
at the 3% enrichment level.
In many smaller foundries, it is not unusual for molders to assume
that their work is complete when the pouring operation begins.
In some cases, the pouring is done by some of the molders. The
substantial increase in melt rate achieved by this foundry with
the resulting reduction in O.T. wages paid for the annual oxygen
expenditure in one month. In addition to the labor savings,
use of FeSi was reduced by 30% while coke reduction amounted
to 17.5%. The most significant advantage realized by this foundry
was the large reduction in overtime wages.
Foundry B - Melt rate originally 11.5 T/hr
54" Cupola cold blast
Metal/coke ratio 6.6:1
An increase in the melt rate is not necessarily a desired feature
in every application of oxygen enrichment. This particular foundry
had other objectives which were of greater importance.
Coke was in short supply to such a degree that a 20% reduction
in shipments was experienced. At a 2%% enrichment level, this
foundry was able to operate a full five days on four days supply
of coke.
Since no increase in production was required, this same enrichment
level (2%%) allowed a reduction in wind of 10% which in turn
lightened the load on the dust collection system.
The most profitable return on the investment in oxygen to this
foundry was achieved by the total elimination of stack mounted
afterburners. These afterburners were fueled with natural gas
which was rising rapidly in cost with the additional danger of
curtailed use. They are no longer required. A reduction in
the use of ferrosilicon was also achieved.
Foundry C - Melt rate originally 8 T/hr
48" lined cupola - cold blast
Metal/coke ratio 6.6:1
This is another example of a foundry required to operate five
days on a four-day supply of coke, which was achieved with a
enrichment level.
Additional cost reductions were realized by substituting cast
scrap (purchased and returned) for the more expensive pig iron.
No change was observed in metal chemistry.
This particular foundry had a normal scrap loss of 15% of finished
castings. Oxygen enrichment has helped to reduce this figure
to 6 to 8% by offering increased metal temperature.
56
-------
Foundry D - Melt rate originally 6.5 T/hr
48" lined cupola - cold blast
Metal/coke ratio 6.6:1
Here is an instance where coke reduction, increased melt rate
and the reduction of FeSi use were of minor importance. This
foundry specialized in the production of castings weighing appro-
ximately 7 tons each. To pour castings of this size required
the use of two large ladles permitting simultaneous pouring.
The increased metal temperature obtained by an enrichment level
of 2% (approximately 110°F) has reduced scrap loss by 65%.
Foundry E - Melt rate 6 T/hr
36" lined cupola - cold blast
Metal/coke ratio 6.6:1
The single purpose of oxygen enrichment in this foundry was to
maximize coke reduction. A 2^% enrichment level permitted a
coke savings of 22%.
Foundry F - Melt rate 11 T/hr
54" lined cupola - warm blast
Metal/coke ratio 8:1
The foundry had used oxygen on an intermittent basis for a number
of years, primarily for faster recovery following lunch hour,
coffee breaks, etc.
While maintaining a metal temperature of 2750°F and a melt rate
of 11 T/hr, the following changes were permitted with an enrich-
ment level of 2% to
1. Wind volume was reduced by 16% resulting in reduced bag house
loading.
2. Coke consumption was reduced by 18%.
3. A minor substitution of a portion of the pig iron with
costly scrap was also achieved.
Foundry G - Melt rate 5 T/hr
36" lined cupola - cold blast
Metal/coke ratio 7.2:1
The use of oxygen enriched wind probably had the greatest impact
on any foundry in the region at this particular location. Airco
was able to provide a combined coke reduction of 40% by means
of reducing bed height and adjusting the iron-coke ratio. A 50%
reduction in scrap loss was achieved by providing continuously
hotter metal on a more consistent basis even after temporary
wind stoppages.
57
-------
Proper pollution control had previously required the use of oil
fired afterburners. Due to increased combustion efficiency these
burners were eliminated resulting in substantial reduction in
operating costs in view of the rapidly escalating cost of fuel
oil.
Another company which appears to be the leader in the development
of oxygen injection technology is Air Products and Chemicals,
Inc. in Allentown, Pennsylvania. They are primarily involved
in the manufacture and supply of liquified gases for commercial
and industrial use, however, their Applied Research & Development
Group has been working closely with foundrymen to create cost
effective alternatives to compensate for the spiralling rise in
raw materials costs and the cost of pollution control equipment.
A field trip was arranged to APCI's Corporate Headquarters in
Allentown, Penn. Mr. D.R. Ruprecht directed the day's activities
which included the inspection of the various in-house design,
testing and fabrication facilities utilized in the development
of various oxygen enrichment systems, along with a brief inspection
of the countless other cryogenic devices being developed for a
broad range of industrial applications. Following the tour a
conference was held in which a number of pertinent questions were
resolved and certain misconceptions regarding oxygen enrichment
systems were corrected. The afternoon was spent touring a nearby
foundry which is presently utilizing an oxygen injection system
which was designed, fabricated and installed by APCI.
Among the major topics of discussion was the question of safety.
The reluctance of some foundrymen to adopt oxygen enrichment is
understandable when one considers the highly oxidizing nature
of pure oxygen. APCI is well aware of the potential hazards in-
volved with the use of oxygen and has gone to great lengths to
ensure the integrity of their systems. As an example of the type
of precautionary measures considered by APCI Mr. Ruprecht pointed
out that in large systems where oxygen is submitted to high velo-
cities special care is taken in selecting the materials to be
used in the piping system to avoid the possibility of sparks produced
by ferrous particles entrained in the supply stream. In addition,
regardless of system size or complexity, all piping and related
accessories are chemically cleaned and dried prior to assembly.
All valves and fittings are selected for their ability to prevent
leakage and minimize wear. Interlocks for protecting against elec-
trical failures and temporary wind spills are standard equipment
in all APCI oxygen systems and in the interest of keeping component
costs down Mr. Ruprecht has stated that on the larger injection
systems the solenoids used are installed on the instrument air
lines which provides a safe, positive emergency shut-off while
eliminating the need for costly in-line shut-off devices.
58
-------
Another interesting point that was clarified during our discussion
was the fact that the large liquid oxygen storage vessels and
the accompanying vaporizing systems do not have to be purchased
by the user. This is a point that has often been misunderstood
by many foundrymen, because in the course of the literature search
no mention was made concerning this. The only cost for which
the perspective user is responsible is for the concrete pad on
which the storage tank and vaporizing equipment are mounted and
a monthly rental fee which covers installation and maintenance
both of which are performed by the supplier.
Mr. Ruprecht also pointed out that the average pressure of the
storage system is approximately 120 psi which is much lower than
the pressures found in oxygen bottles commonly used in the foundry's
repair shop which are often well over 2000 psi. Another bonus
which can be derived from a conversion to oxygen enrichment is
the opportunity to use the oxygen supply for various auxiliary
plant applications such as lances, oxyactelene burning and countless
other uses and at a cost considerably lower than if purchased
in individual cylinders. In addition APCI will conduct a study
of the average oxygen consumption rates and will establish a regular
supply and maintenance schedule to ensure continuous operation
of the enrichment system.
Mr. Ruprecht co-authored an informative paper with Mr. A.L. Love
which was published in the AFS Transactions. (6) in this paper
he points out that oxygen enrichment can be applied in three basic
ways:
1. Continuously to optimize raw material savings
2. Continuously to increase production
3. Intermittently for quicker startups and greater cupola control.
The foundryman maximizes the benefits with each practice by con-
trolling three variables: wind rate, oxygen content of the wind
and the period of enrichment (continuous or intermittent use).
These benefits are achieved to greater or lesser degree depending
on the method of application and the operating parameters of the
cupola.
PRODUCTION INCREASE APPLICATIONS
In general, the minimum production rate required to economically
justify a production increase using oxygen enrichment is approxi-
mately 15 tons per day, 3 days per week. Oxygen enrichment can
provide production increases of 25% and more.v^) This is achieved
through enrichment by simply increasing the flow of oxygen to
the cupola. More oxygen results in higher combustion rates thus
higher melt rate. For example, a cupola operating with a wind
rate of 1000 SCFM (standard cubic feet per minute) contains 210
SCFM of oxygen (Figure 16). A 2% enrichment practice would add
59
-------
Supplemental Enrichment
21% 0.
79% N
210 CFM
790CFM
1000 CFM Air
26CFM
Oxygen
236 CFM
790 CFM
1026 CFM
Enriched Air
23% 0.
77% N.
21% 0.
79% N.
1000 CFM Air
Equivalent Enrichment
210 CFM
790 CFM
23 CFM
V V Oxygen
187 CFM
703 CFM
^s
210 CFM
703 CFM
890CFM Air
23% 0,
77% N.
913 CFM
Enriched Air
Figure 16. Oxygen enrichment techniques.
Courtesy of Air Products ond Chemicals
-------
26 SCFM of oxygen to increase the oxygen content of the air to
23%. A non-enriched wind rate of 1124 SCFM would be required
to achieve the same total oxygen flow provided in the 1026 SCFM
of enriched air. The difference, of course, is heat absorbing
nitrogen. A 2% continuous enrichment practice can provide a
10% to 15% increase in production while a 3% practice offers 20%
to 25% increased production. As oxygen enrichment levels are
increased beyond 5%, however, production gains become less signi-
ficant.
Oxygen enrichment may be used intermittently to increase cupola
melt rate quickly after shutdowns and during periods of heavy
casting line demand. During periods of extremely high casting
line demand, the oxygen enrichment level may be raised from 4% to
5% providing an increase in melt rate of as high as 30%. This
type of intermittent enrichment practice is used primarily to
avoid casting line downtime due to lack of hot metal.
RAW MATERIAL SAVINGS APPLICATIONS
Again, a cupola must produce more than fifteen tons of iron per
day and operate at least three days per week to economically
justify oxygen enrichment with raw materials savings. When addi-
tional melt rate is not required, the oxygen enrichment practice
can provide substantial savings in raw material consumption.
An equivalent wind enrichment practice would be employed to maximize
raw material utilization without increasing the melt rate.
The most common reduced wind enrichment level is 2%. Our sample
1000 SCFM wind rate is again used to illustrate the practice (Figure
17). The wind would be lowered to 890 SCFM and 23 SCFM of enrich-
ment oxygen added. This practice would maintain the same melt
rate achieved with 1000 SCFM air, but with 9% less total volume
entering the cupola. Eliminated is 87 SCFM of nitrogen, which,
as discussed earlier, is a thermal burden on the cupola. By im-
proving the thermal efficiency of the cupola, oxygen enrichment
offers several areas for raw material savings: (9)
1. Coke Savings - The increased thermal efficiency of the cupola
allows a substantial reduction in the amount of coke needed for
smooth operation. A 2% enrichment practice conservatively offers
a 10% reduction in coke consumption. This savings alone normally
offsets all or a large part of the oxygen cost.
2. Calcium Carbide (CaC2) Reduction - Calcium carbide is used
in the cupola to increase metal temperature and increase melt
rate. However, it is very expensive and becoming increasingly
harder to obtain. With an oxygen enrichment practice, calcium
carbide consumption can be significantly reduced. Savings of
50% or more may be realized due mainly to the increased temperatures
that are achievable with oxygen enrichment.
61
-------
to
Reduce wind
by 11%
23 CFM
Oxygen
2 I % 0,
79% N.
2IOCFM
87 CFM
2IOCFM
790 CFM
703 CFM
703CFM
23% 0,
77% N,
1000 CFM Air
890 CFM Air
913 CFM
Enriched Air
2iO CFM OXYGEN
913 CFM Enriched Air
23% or 2% Enrichment
Figure 17. Equivalent enrichment.
Courtesy of Air Products and Chemicals
-------
3. Pig Iron Substitution - Pig iron is a carbon and silicon rich,
but expensive charge material. With oxygen enrichment many foundries
have been able to substitute lower-grade, less expensive raw materials
for a portion of the pig iron charge. This is made possible through
the higher temperatures and improved heat transfer and thermal
efficiency achieved in the combustion zone with oxygen enrich-
ment. The substitution of cast scrap for pig offers two signi-
ficant areas of potential savings. Experience has shown, however,
that the substitution of large quantities of steel scrap for
pig iron may require the addition of small amounts of silicon
bearing charge additions such as FeSi to provide proper metal
chemistry. The net result, however, is usually a substantial
reduction in total charge.
ADDITIONAL BENEFITS OF OXYGEN ENRICHMENT (9)
Improved Cupola Control
Variations in normal cupola temperature and chemistry can occur
frequently due to differences in scrap makeup and size, quality
of coke and the moisture content of the air. Adjustments in
the oxygen enrichment level can compensate for these variances.
Oxygen enrichment can also be a valuable tool for quickly bringing
the cupola to normal temperature after start up or interruptions
due to mechanical outages. The time to restore a cold cupola
to normal operation has been reduced by as much as 50%, resulting
in much less pigging and waste of cold, first tap metal.
Reduction in Casting Rejects
Due to the higher temperatures realized with the oxygen enrich-
ment practice and the ability to increase iron temperature as
desired, substantial reductions in rejects due to cold metal may
be possible. Oxygen enrichment has proven to be particularly
effective during start up by bringing the metal to the desired
temperature in a shorter period of time. Experience has shown
that as much as 50% of rejects due to cold metal can be eliminated.
Increased Metal Temperature
Increasing the percentage of oxygen entering the cupola results
in higher temperatures in the combustion zone. The net effect
on metal temperature is an increase of 10°F to 50°F or more de-
pending on the practice. This additional temperature can provide
more desirable casting characteristics and result in fewer rejects
on the casting line.
Greater Slag Fluidity
The increased melting zone temperature produced by oxygen enrich-
ment results in a more fluid slag. Consumption of slag fluidizers
63
-------
such as fluorspar can be reduced by 50% or more without sacrificing
proper slag properties. Oxygen enrichment reduces slag bridging
around the tuyeres and results in a smoother running furnace.
In a subsequent study conducted by APCI seven representative case
histories were recorded which demonstrate each of the three methods
of application and cover the normal range of operating conditions
where raw materials savings, production increases or greater cupola
control can be profitably achieved with oxygen enrichment. By
comparing his operation with those described in the case histories,
the foundryman can estimate the magnitude of savings available
to him with the oxygen practice. In some of the cases mentioned,
foundry management chose to keep actual dollar savings with oxygen
enrichment confidential. Sufficient data has been given, however,
to effectively evaluate the oxygen technique. Based on this data,
minimum production levels for the profitable application of oxygen
enrichment were estimated.
CASE HISTORIES (9)
The first three foundries looked to oxygen enrichment primarily
to increase cupola melt rate. In addition, foundries two and
three also achieved significant raw material savings. (Foundries
four and five employed oxygen primarily to reduce raw material
costs.) The last case history is an example of a foundry using
oxygen as a control tool to increase metal temperature and melt
rate during start up and when needed during the operating day.
Oxygen Enrichment for Increased Melt Rate
Foundry #1 - Cupola Size (ID): 86", Water Wall
Wind Rate: 13,000 CFM, 1000°F
Production: 450 T/D, Malleable Iron
This malleable iron foundry was able to increase cupola melt
rate by 20% with 3% continuous supplemental enrichment. This
represented an additional 90 tons of metal each day. The daily
oxygen cost for this practice is $1,030 or $11.43 per extra ton
melted.
In this particular case, an increase in metal temperature was
also achieved which resulted in a small decrease in fuel con-
sumption for the holding furnace.
Foundry #2 - Cupola Size (ID): 60", Acid Lined
Wind Rate: 5500 CFM, 40QOF
Production: 270 T/D, Gray Iron
The second example illustrates an operation which needed addi-
tional melt rate during most of the day to meet fluctuating
casting line demand. Here a continuous 2% supplemental enrich-
ment practice was employed when additional melt rate was needed.
64
-------
During periods of reduced casting line demand the wind rate was
decreased and the oxygen adjusted to maximize coke savings.
Daily production was increased an average of 8% with increases
in melt rate as high as 15% during peak casting line demand.
Coke consumption was reduced by an average of 4%. Also, normal
additions of proprietary flux were reduced by 75% as a result
of higher melting zone temperatures with oxygen. The coke and
flux savings offset most of the oxygen cost.
Foundry #3 - Cupola Size (ID): 50", Acid Lined
Wind Rate: 4500 CFM, Cold Blast
Production: 40 T/D, Gray Iron
This gray iron foundry needed a 10% increase in melt rate and
a 20°F to 30°F increase in metal temperature. They were able
to achieve these goals with 2% enrichment. In addition, increased
carbon and silicon levels in the hot metal permitted the operator
to substitute steel scrap for a portion of the pig iron charge
and reduce coke consumption. The pig iron charge was reduced
from 30% to 25% of the metallic charge. A small increase in
the silicon briquette charge was required to make up for some
of the silicon lost in the steel substitution. Coke consumption
was reduced by 5% to 10% depending on coke quality. Evaluation
of the process economics showed that the savings in pig iron
and coke offset the cost the oxygen and resulted in a net savings
of approximately $90 per day.
These first three foundries are fairly representative of foundries
using oxygen enrichment to increase cupola melt rate. In addition
to a production increase of 20% or more, gray and ductile cupola
operators can usually expect some raw material savings, parti-
cularly in coke. Coke savings are not normally achieved when
increasing melt rate of a malleable cupola due to the inherently
lower carbon levels in a malleable operation.
Oxygen Enrichment for Raw Material Savings
Many foundries which do not need additional production but can
still benefit with oxygen enrichment. These foundries would
utilize the equivalent or reduced wind enrichment practice dis-
cussed earlier. An example of one such operation is the fourth
case history.
Foundry #4 - Cupola Size (ID): 54", Acid Lined
Wind Rate: 5500 CFM, 900°F
Production: 200 T/D, Gray Iron
Due to coke quality and supply problems, this foundry had to
supplement its coke charge with calcium carbide to achieve the
desired metal temperature and melt rate. By decreasing the wind
65
-------
from 5500 CFM to 4900 CFM and enriching with 2% oxygen, the cupola
operator was able to reduce coke consumption from 220 pounds
per ton of iron to 190 pounds. In addition, the calcium carbide
charge (10 pounds per ton of iron) was eliminated while still
maintaining desired melt rate, temperature and chemistry levels.
Raw material savings totalled $660 per day with a daily oxygen
cost of $400 resulting in a net savings of $260 per day.
Foundry #5 - Cupola Size (ID): 60", Acid Lined
Wind Rate: 5000 CFM, Cold Blast
Production: 95 T/D, Gray Iron
This gray iron foundry is using oxygen to reduce coke consumption
and improve metal temperature. However in this operation, rather
than reduce the normal coke splits, it was decided to reduce
the number of coke boosters needed throughout the day by smoothing
operating fluctuations with oxygen enrichment. Oxygen is used
at an average of 2% enrichment with adjustments above and below
this level being made as operating conditions dictate. This
practice has resulted in the elimination of an average of 1000
pounds of daily coke booster. In addition, the improved carbon
pick up resulting from the higher combustion zone temperatures
permits substitution of purchased scrap for a portion of the
pig iron charge. As can be seen below, coke and pig iron savings
offset the cost of oxygen. The higher metal temperatures have
resulted in a reduction in cold metal rejects.
OXYGEN
CONVENTIONAL ENRICHMENT
Coke (Ibs/ton Fe) 287 287
Coke Boosters (Ibs/
day) 1800 800
Pig Iron (Ibs/ton
iron) 80 50
Tap Temperature 2780°F 2850°F
$/DAY
Coke and Pig Iron Savings 160
Oxygen Costs 145
NET SAVINGS 15
While most foundries using oxygen enrichment achieve sufficient
savings to offset the cost of an oxygen practice, oxygen is not
economical for all. Generally, a foundry must be producing more
than 30 tons per day to justify oxygen enrichment on raw material
savings alone. The following outlines an evaluation made for
a small gray iron foundry, which shows that oxygen enrichment
would not be economical for this operation.
66
-------
Foundry #6 - Cupola Size (ID): 42", Acid Lined
Wind Rate: 2700 CFMf Cold Blast
Production: 32 T/D, Gray Iron
This foundry did not need any additional melt capacity or metal
temperature. The cupola charge consisted of 70% cast scrap and
30% returns with a metal to coke ratio of 10 to 1. In this operation
the 10% to 15% coke savings possible with 2% enrichment would
offset only 30% to 50% of the oxygen costs as shown below.
$/DAY
15% Coke Savings 53
Oxygen Cost 100
NET COST 47
Oxygen Enrichment for Cupola Control
While some operations, particularly smaller foundries, cannot
justify the use of continuous enrichment for production or raw
material savings, they can usually realize economic benefit using
oxygen intermittently for faster start ups, improved operation
control and higher temperatures.
Foundry #7 - Cupola Size (ID): 46", Acid Lined
Wind Rate: 3800 CFM, Cold Blast
Production: 150 T/D, Gray Iron
The final example outlines the results one such foundry achieved
using oxygen enrichment intermittently. Fluctuations in casting
line demand at this particular operation made it necessary to
go off wind frequently during the day. This resulted in low
casting temperatures particularly after these off wind periods.
With oxygen enrichment, when the cupola is brought back on wind,
2% to 3% enrichment is used to quickly bring the metal temperatures
up to the desired level resulting in a substantial reduction in
cold metal rejects. The enrichment practice has also reduced
the amount of pigging required during start up by over 50%.
Finally, the higher metal temperatures have allowed this foundry
to eliminate their 1000 pounds per day CaC2 additions representing
a savings of $50 per day. On the average, oxygen is used four
hours out of the sixteen hour operating day. Thus, including
CaC2 savings, this foundry greatly reduced pigging and cold metal
problems at a net oxygen cost of only $35 per day. (See Table
13 for a summary of the case histories.)
COST ANALYSIS OF OXYGEN ENRICHMENT SYSTEMS
Figure 18 represents a conceptual view of an oxygen enrichment
system. This system can be broken down into three areas:
67
-------
00
AIR
OXYGEN
SUPPLY
CONTROL
PANEL
ORI
Zl
ERS
r
i
PRESSURE
REGULATOR
Figure 18. Oxygen system.
Courtesy of Air Products and Chemicals
-------
Table 13. CASE HISTORY SUMMARY - APCI
Cupola Inside Diameter Wind Rate
Production
Savings Area
en
Oxygen Enrichment
for Increased
Melt Rate
Foundry #1
Foundry #2
Foundry #3
B. Oxygen Enrichment
for Raw Material
Savings
Foundry #4
Foundry #5
Foundry #6
86 inch water well
60 inch acid lined
50 inch acid lined
54 inch acid lined
60 inch acid lined
42 inch acid lined
13,000 CFM
hot blast
5,500 CFM
hot blast
4,500 CFM
cold blast
5,500 CFM
hot blast
5,000 CFM
cold blast
2,700 CFM
cold blast
450 T/D 20% increase in
malleable iron melt rate
270 T/D
gray iron
40 T/D
gray iron
200 T/D
gray iron
95 T/D
gray iron
32 T/D
gray iron
8% production increase
15% increase in melt
rate
75% reduction in
flux additions
10% increase in melt
rate
5% pig iron reduc-
tion
10% coke reduction
14% coke reduction
Elimination of calcium
carbide additions.
55% reduction of coke
boosters
15% coke reduction
Oxygen enrichment
uneconomical.
-------
Table 13. CASE HISTORY SUMMARY - APCI (Continued)
Cupola Inside Diameter Wind Rate Production
C. Oxygen Enrichment
For Cupola Control
Foundry #7
46 inch acid lined
3,800 CFM
cold blast
150 T/D
gray iron
Savings Area
Reduction of cold
metal rejects.
50% reduced pigging
Elimination of
calcium carbide
additions.
-------
1. Liquid oxygen storage and vaporization system
2. Oxygen supply line
3. Oxygen enrichment flow control system
Liquid Oxygen Storage and Vaporization System
Since relatively large volumes of oxygen are needed for oxygen
enrichment, a bulk oxygen storage system is normally required.
The system consists of a tank, vaporizers and normally some type
of pressure regulating equipment (see Figure 19). The oxygen
is stored as a liquid (at -300°F) in a double-shell cryogenic
tank. In liquid form, a large amount of oxygen can be stored
in a relatively small volume (one gallon liquid oxygen equals
115 standard cubic feet of gaseous oxygen).(9) When oxygen is
needed for cupola enrichment or oxygen lances, etc., the cold
liquid is automatically transferred from the oxygen tank to
a vaporizer. The ambient air vaporizer is the most common type.
In this unit the liquid oxygen absorbs heat from the surrounding
air and changes to a gas. As it changes, it expands providing
the pressure needed to deliver oxygen to the cupola area. Pressure
is normally regulated before it leaves the tank area.
The only capital investment the foundry makes in this area of
the system is for a concrete foundation. The oxygen tank and
vaporizer are normally leased from the oxygen supplier. This
frees the foundryman from the responsibility of servicing and
maintaining the fairly sophisticated cryogenic equipment. In
addition, with ambient air vaporizers no power is required except
to pump the liquid oxygen from the tank truck to the storage
tank during deliveries. When extremely high oxygen flow rates
are needed electric or steam assisted vaporizers may be required.
Oxygen Supply Line
From the tank area the gaseous oxygen is piped to the cupola,
maintenance shop, or any other oxygen use point. The oxygen
supplier will usually provide assistance in the design and
sizing of the supply line.
Oxygen Enrichment Flow Control System
The equipment used to control the flow of oxygen into the cupola
(and thus the oxygen enrichment level) can range from simple
to quite sophisticated. A basic flow control system (see Figure
15) includes:
1. Pressure regulator
2. Flow measuring device
3. Flow control valve
4. Electrical shutoff valve
5. Check valve
6. Diffuser
71
-------
LIQUID STORAGE
TANK
AMBIENT
AIR VAPORIZER
HOUSELINE
PRESSURE
STATION
Figure 19. Typical liquid tank installation.
72
Courtesy of Air Products 8 Chemicals
-------
The system is normally interlocked with the cupola air blast
to insure that oxygen cannot flow without sufficient air flow.
Foundries operating more than a few hours a day would find a
recording type system valuable for establishing and maintaining
an economical enrichment practice. If frequent changes in the
cupola wind are made during the operating day, an automatic control
system may be desirable. An approximate cost range for each
type of system is provided below. (9)
PRICE RANGE
1. Manual System with Flow Indicator $1000-$3000
2. Manual System with Recorder and
Remotely Operated Valves $3000-$6000
3. Automatic Ratio Type System $5000-$8000
The wide cost range is due primarily to the size of the flow
components required for the maximum oxygen flow rate.
In summary, the foundry's capital investment in the oxygen en-
richment process includes:
1. Oxygen Tank Foundation
2 . Oxygen Supply Line
3. Flow Control Equipment
For most operations, the total investment is $5000 to $20,000. (9)
In order to determine the cost effectiveness of oxygen enrich-
ment, the foundryman must first decide which method of appli-
cation will best serve his needs. If his cupola produces over
15 tons of hot metal per day, raw materials savings and reduced
wind rate benefits are possible with oxygen enrichment. The
production increase application is called for if he has unused
casting capacity and the extra product can be sold. Regardless
of his production level, he can achieve greater cupola control
with the oxygen practice.
Next the foundryman must carefully evaluate all the benefits
of oxygen enrichment. It is best to calculate the maximum and
minimum savings that can be achieved with each benefit. Even
intangibles, like the cost of downtime on the casting line, due
to the lack of hot metal, should be considered. The case histories
previously cited, however, provide a representative picture of
savings that might be achieved for similar applications and opera-
ting conditions.
73
-------
Finally, oxygen consumption should be calculated. A rough estimate
may be obtained by multiplying wind rate by 2% by the on wind
time. For example, the monthly oxygen consumption for a cupola
operating at an 8000 CFM (225 m3/min) wind rate and 2% continuous
oxygen enrichment with 10 hours per day on wind time would be
approximately (8000 CFM (225 m3/min) x 2% x 60 min/hr x 10 hr/day
x 20 days/month) = 1,920,000 ft3(55,000 m3).
Oxygen cost may be estimated using Figure 14. The unit cost
of oxygen is primarily dependent on consumption level, but the
oxygen price also varies considerably with the distance from
source to consumer, the type of vaporization required and other
factors. Additionally, U.S., Canadian and European oxygen prices
differ. At the 1.9 million ft3 (55,000 m3) per month consump-
tion level of the example, oxygen cost is roughly $0.22 per hundred
cubic feet as derived from Figure 14. Monthly oxygen cost would,
therefore, be approximately $4400.
Comparing total savings to total oxygen cost determines the economic
feasibility of oxygen enrichment. Economic evaluations of oxygen
enrichment are, of course, greatly dependent on the price of
the raw materials saved and/or the price attached to increases
in production. Since these may change significantly over a one
year period, annual re-evaluation of oxygen enrichment would
be prudent. Clearly, the tendency in recent years has been towards
the use of oxygen enrichment.
74
-------
SECTION VII
DIVIDED BLAST TECHNIQUES
INTRODUCTION
The concept of introducing blast air into the furnace through
single, double, or multiple rows of tuyeres has been with us for
several years. Claims have been made at various times that by the
use of two or more rows of tuyeres, higher metal tapping temperatures
can be obtained for a given coke charge or, conversely that charge
coke consumption can be reduced for a given metal tapping temperature.
An interesting prototype was the balance blast cupola, introduced by
BCIRA over forty years ago.(10) The system was constructed so that
the relative amounts of air admitted to each of the several levels
could be adjusted to give a maximum of carbon monoxide at the bottom
and a minimum at the top of the combustion zone, thus utilizing all
of the potential heat of the coke.
In principle, the concept of balanced blast appears valid, however,
the control and monitoring systems available at that time were incapable
of effectively proportioning and balancing the amount of blast among
the various rows of tuyeres. It was therefore left to the individual
judgement of the cupola operators to establish and maintain optimum
conditions throughout the burn.
Though the system realized a limited degree of success it was never
adopted on a large scale and most foundries chose to remain with the
more conventional single row of tuyeres.
In light of the recent exponential rise in raw material costs, along
with the propects for increasingly stringent pollution control stan-
dards, foundrymen have begun to reexamine the possibility of improv-
ing cupola operating efficiency and economy by the use of two rows
of tuyeres.
A comprehensive series of tests were conducted by the British Cast
Iron Research Association (BCIRA) at their cupola plant in Alvechurch,
Birmingham, England, in an effort to define the conditions under which
these improvements could be obtained. The authors feel that these
are probably the most detailed and informative tests available to
date and that they are well worth noting in this publication.(10)
TEST PROCEDURES
The cupola used for these tests had an internal diameter of 30 in.
(76.2 cm). Two windbelts were provided: one for the supply of air
to four lower tuyeres, (each of 4 in. (10 cm) diam.) and the other
supplying air to four upper tuyeres of the same diameter.
The distance between lower and upper tuyeres could be varied by the
75
-------
provision of several alternative rows spaced at one foot intervals,
the rows not in use being sealed off. The flow of blast air to each
lower and upper row of tuyeres was measured, recorded and automati-
cally controlled.
The cupola was continuously tapped and slagged. The temperature of
the metal was measured continuously by a Platinum/Platinum-13% Rhodium
thermocouple immersed in the clean metal compartment of the front
siphon box and recorded on a strip chart recorder.
The tests were conducted as follows:
Varying the Distribution of the Blast Air Between Lower
and Upper Tuyeres
These tests were carried out under three different combined condi-
tions of total blast rate, coke charge and spacing of the lower and
upper rows of tuyeres. At each set of conditions, the blast dis-
tribution to upper and lower tuyeres was varied. Figure 20 shows
that for each set of conditions investigated, increasing the pro-
portion of total blast supply to the upper rows of tuyeres and corres-
pondingly reducing that to the lower row:
1. Increased tapping temperature of the metal until the blast was
divided approximately equally between lower and upper rows. On fur-
ther increasing the proportion of air supplied to the upper tuyeres,
temperature decreased.
2. Reduced melting rate slightly.
3. Increased combustion efficiency and stack gas temperature.
Figure 21 shows that increasing the proportion of air supplied to
the upper tuyeres:
1. Increased carbon content of the metal, a maximum value being
obtained when the blast was equally divided between the two rows
of tuyeres.
2. Had little significant effect on manganese and sulfur contents
of the metal.
3. Tended to reduce silicon content of the metal slightly.
Varying the Distance Between the Lower and Upper Tuyeres
The results already obtained indicated that for a given spacing of
the lower and upper tuyeres, maximum temperature was obtained when
the air supply was divided equally between the two rows. A further
series of tests was carried out to examine the effects of varying
distance between the two rows of tuyeres to determine optimum spacing
and to find if this was influenced by the amount of charge coke used
and the blast rate.
76
-------
u.
o
Q.
z
s
2700
26OO
2700
27OO
3.5
3.0
3.5
3.O
[
o
o
a
"E
; lor
30_
9OOp
700
9OOp
700L
I 2OO,
IOOO
-•2
-fi.3
-•1
0 2O 4O 60 80 IOO
PERCENTAGE OF BLAST SUPPLIED TO UPPER TUYERES
KEY^ I COKE 9% BLAST RATE l,6OOft%«in TUYERES 30Sn APART
2 u 12% ,. i, I.3OO u ii (Bin u
3 (l 15% i. Ii 1,600 u u 54 in »
Figure 20. Effect of varying blast distribution to lower
and upper tuyeres on furnace performance. (10)
o
a>
KEY:
3.5
3.0
3.5
3.0
3.8
• 3.3
2. I
°-5c _.
0.3 t r
0.5 r »_
0.3 t
0.5 P ^.
0.3 t
o.nr
o-oet
. o.i i
aoat *~
O.I 11
o.
-• I
0 20 40 . 6O BO IOO
PERCENTAGE OF BLAST SUPPLIED TO UPPER .TUYERES
I COKE 9% BLAST RATE I^OOftVmin TUYERES 30 In APART
2 n 12% >. ii 1,300 • u 18 in
3 -i 15% u „ |,6OO u u 54 in
Figure 21 .Effect of varying blast distribution to lower and
upper tuyeres. (10)
-------
Vary Charge Coke Quantity and Distance Between Upper
and Lower Tuyeres at Constant Blast Rate
Tests were conducted at charge coke quantities of 9, 12 and 15% by
weight of the metal charge; at each charge coke quantity, the dis-
tance between the two rows of tuyeres was varied. In addition, one
test was made at a charge coke quantity of 7% with a tuyere spacing
of 30 in. (76.2 cm). The blast rate in all tests was 1600 ft^/min
(45 m^/min) and this was divided equally between the two rows of
tuyeres (except, of course, when only one row was used where upon
the total blast quantity was delivered to this row). Figure 22 shows
that at each charge coke quantity, increasing the distance between
two rows of tuyeres:
1. Increased metal temperature until a maximum value was attained
when the tuyeres were spaced between 30 in. (76.2 cm) and 42 in.
(106.7 cm) apart (i.e., approximately 36 in. (91.4 cm); when tuyere
spacing was increased further, metal temperature decreased from the
maximum value.
2. Slightly reduced melting rate.
3. Increased combustion efficiency and stack gas temperature.
Figure 23 shows that by varying the distance between the two rows
of tuyeres:
1. Carbon and manganese contents of the metal attained maximum values
and sulfur content a minimum value when the tuyeres were spaced approx-
imately 36 in. (91.4 cm) apart.
2. Silicon content of the metal decreased progressively as the dis-
tance between the two rows of tuyeres was increased.
Vary Blast Rate with Constant Coke Charge
Results of these tests showed that the optimum spacing of two rows
of tuyeres was independent of the amount of charge coke used. To
determine if it was influenced by blast rate, tests were carried
out at blast rates of 1300, 1600 and 1900 ftVmin. (37r 48 and 54
m^/min.) at various tuyere spacings, the charge coke quantity being
kept constant at 12% by weight of the metal. Figure 24 shows that
optimum distance between the two rows of tuyeres (i.e., the distance
at which maximum metal temperature was obtained) was not influenced
by blast rate.
BENEFITS DERIVED FROM DIVIDED BLAST
Figure 25 shows the relationship between charge coke quantity and
metal temperature at a blast rate of 1600 ft-^/min. (45 m^/min.) for
operation with one and two rows of tuyeres. In the latter case the
cupola was arranged with optimum spacing of the two rows of tuyeres
and with the blast divided equally between them.
This illustration shows that at a given charge coke quantity, the
metal temperature increased by 80° to 90°F (45° to 50°C) by operating
73
-------
2900
0 T6 20 30 40 50 60
Distance between lower 8 upper tuyeres, In.
Figure 22. Effect of tuyere spacing on furnace performance
at various charge coke quantities (blast rate ISOOftVmia
= 45m3/min.) (10)
10 20 30 40 50
Distance between lower Supper tuyeres, In.
60
Figure 23. Effect of tuyere spacing on metal composition at
various coke charge quantities (blast rate I600ft3/min.
= 45m3/min.) (10)
-------
u.
o
HI
a:
ill
o.
s
UJ
<
UJ
ID
is
H h-
UJ
o: o-
i 1
2800
2700
2600
5.0
4.0
3.0
60
50
40
Crt O
i 8 * 30
O o
o Q
1200
u.
o
I 100
5 in 1000
o "^
* < 900
o tt
-------
CO
2900
2800
UJ
tr
cc
UJ
a.
S.
UJ
I-
UJ
2700
2600
2500
Two rows of tuyeres with
optimum spacing
7.2%/ll% 20%
Coke
I
I
10 II 12
CHARGE COKE,%
13
14
15
16
Figure 25. Reduction of charge coke consumption and increase in melting rate by
operating cupola with two rows of tuyeres with divided blast supply ( blast rate
leOOft'/min. -45ms/min. )(10)
-------
with two rows of tuyeres. It also indicates the extent to which
charge coke consumption could be reduced for a given metal tapping
temperature by using two rows of tuyeres and a divided blast supply.
For example, a charge coke consumption of 15% was required to pro-
duce a tapping temperature of 2732°F (1500°C) when operating with
a single row of tuyeres whereas, with two rows, this temperature
was obtained for a charge coke consumption of 10.8%. At a blast
rate of 1600 ftVmin (45 m^/min.(the melting rate obtained with
15% charge coke was 3.41 tons per hour with one row of tuyeres).
When using two rows of tuyeres and 10.8% charge coke, the melting
rate was 4.07 tons per hour. Thus, operation with two rows of
tuyeres allowed charge coke consumption to be reduced by 28% and
the melting rate to be increased by 19% compared with operating
with one row of tuyeres while permitting a tapping temperature of
2732°F (1500°C) to be obtained. Figure 25 also shows the extent
to which charge coke consumption could be reduced and the melting
rate increased for other required levels of metal temperature.
Within the range studied, the reduction of coke consumption amounted
to between 20 and 30% of that required when operating with one row
of tuyeres. The corresponding increase in the melting rate was
between 11 and 23%.
Effect on Carbon and Silicon Content of Metal
Figure 26 shows the relation between metal temperature and carbon
content of the metal at various tuyere spacings, the results relat-
ing to the tests carried out at a constant blast rate of 1600 ft^
/min. (45 m3/min.) and various charge coke quantities. At each
tuyere spacing, carbon content increased with metal temperature
but (as shown by the bottom diagram, in which lines on the upper
diagram have been combined) for a given metal temperature, carbon
content of the metal increased slightly as the distance between
the two rows of tuyeres was increased.
Figure 27 shows the relation between metal temperature and silicon
content of the metal. As in the case of carbon content, the silicon
content increased with metal temperature at a given metal tempera-
ture as the distance between the two rows of tuyeres increased.
Thus, by operating with two rows of tuyeres with optimum spacing,
carbon content was approximately 0.06% higher and silicon content
approximately 0.17% lower than obtained when operating with a single
row of tuyeres, metal temperature being identical in both cases.
The carbon equivalent value (C% + Si% + P%) was not therefore
3
affected by varying tuyere spacing at constant metal temperature.
Optimum Blast Rate
Figure 28 illustrates the effect on metal temperature of varying
the blast rate at a constant charge coke quantity of 12% when operat-
82
-------
0\°
(0
u
2.4
2.2
2.0
2.4
2.2
2.0
2.4
2.2
2.0
1.8
2.4
2.2
2.0
1.8
2.2
2.0
2.4
2.2
2.0
I.8
TUYERE SPACING Oln.
TUYERE SPACING 18 In.
TUYERE SPACING 3d in.
TUYERE SPACING 42ln
TUYERE SPACING 52 lit
TUYERE SPACING
.0llM8lnjoin.
42in.
94 in.
2500 2600 2700 2800 29OO
METAL TEMPERATURE, ° F
KEY: A 7% Coke • 9% Coke
O 12% Coke • 15% Coke
Figure 26. Effect of metol temperature and tuyere
spacing on carbon content of metal. (10)
3.6
3.4
3.2
3.0
3.6
3.4
3.2
3.6
.. 3.2
§>••
03.4
rH 3.2
•H
W 3.8
3.6
3.4
3.8
3.6
3.4
3.2
TUYERE SPACING Oln.
TUYERE SPACING 18 In.
TUYERE SPACING 30 In.
TUYERE SPACING 42in.
TUYERE SPACING 54 In.
TUYERE SPACING (In.)
5430
'42
18
2500 260O 2700 2800 2900
METAL TEMPERATURE, °F
KEY: A 7% Cok« • 9% Coke
O 12% Coke • 15% Coke
Figure 27. Effect of metal temperature and tuyere
spacing on silicon content of metal. (10)
8.1
-------
ing with a single row of tuyeres and with two rows spaced 18, 30,
42 and 54 in. apart (47.5, 76.2, 106.7 and 137.2 cm). The curves
have been drawn so they are consistent with the relationships between
metal temperature and tuyere spacing of various blast rates shown
in Figure 24. They show that operation with two rows of tuyeres
did not affect optimum blast rate which, at 12% charge coke, was
1750 ft3/3jin (49.5 m3/min.) corresponding to a specific blast rate
of 360 ft3/min (107 m3/m2/min.).
Burn-out Pattern and Spacing of Tuyeres
The burn-out pattern in the cupola studies is illustrated in Figure
29. The burn-out extended further upwards when using two rows
of tuyeres but it was not as deep, so that operating with two
rows of tuyeres had little effect on the quantity of material re-
quired for repair of the lining.
The distance from the center of the tuyeres to the top of the burn-
out when operating with a single row of tuyeres was approximately
36 in. (91.4 cm) and varied little with the charge coke quantity
or blast rate. It was also observed, that, when operating with
two rows of tuyeres, the burn-out extended approximately 36 in.
(91.4 cm) above the center of the upper row of tuyeres. This was
a very clear indication that the use of an upper row of tuyeres ex-
tended the height of the operating coke bed. The improved performance
obtained with divided blast is believed to be due to this extension
of the coke bed, which improves heat transfer to the molten metal
passing through it, rather than to any substantial combustion of
carbon monoxide by the air entering the upper row of turyeres.
The distance between the tuyeres and the top of the burn-out 36 in.
(91.4 cm) corresponded very closely to the optimum spacing of the
tuyeres. It was therefore thought that this might provide a criterion
for spacing the tuyeres on cupolas of varying sizes and practices.
To obtain information on the burn-out pattern in cupolas of different
sizes, several foundries were requested to provide details of cupola
lining wear. The cupolas varied from 28 in. (70.12 cm) to 87 in.
(221.0 cm) internal diameter and covered a wide variety of practice,
the charges varying from high pig iron and cast iron scrap mixes
to high steel scrap mixes. The height of the burn-out for these
cupolas is shown in Figure 30, from which may be seen that this is
not a function of the cupola diameter. Additionally, in the great
majority of cases, the height of the burn-out was within the range
33 to 48 in.(83.3 to 121.9 cm). Inspection of Figure 22 shows that
within this range of tuyere spacings, variation of the spacing had
very little effect on furnace performance. It was therefore tenta-
tively suggested that in designing or converting cupolas for opera-
tion with two rows of tuyeres, the rows should be spaced between
about 30 in. (76.2 cm) and 42 in. (106.7 cm) apart. From experience
of cupolas up to 73 in. (185.4 cm) internal diameter converted to
operate with two rows of tuyeres, this recommendation appears to
have been correct.
84
-------
2700
2650
2600
2800
2750
270O
2650
2800
2750
2700
2650
2800
2750
2700
2800
2750
2700
2650
TUYERE SPACING
"• 0 in.
ISIn.
• 30ln.
42 In.
54 In.
1300 1600 1900
BLAST RATE.ftVmln.
Figure 28. Effect of blast rate on metal temperature at
various tuyere spacings. (coke charge 12%) (10)
36ln
36in
ONEROW^-'"">X"V—''^'TvroRoWS
Figure 29. Pattern of lining burn-out when operating
with one and two rows of tuyeres. (10)
85
-------
60
oo
£ 40
H
O
cr
oo 30
LL.
O
I
O
LU
20 30 40 50 60
CUPOLA DIAMETER, in.
70
80
90
Figure 30. Height of burn-out for cupolas of various sizes.
(10)
-------
Coke Bed
The use of two rows of tuyeres produces a deeper hot coke bed and
so the quantity of bed coke needed is greater than when operating
with one row of tuyeres. Due consideration should be given to the
extra bed coke required when attempting to arrive at an estimate
of the overall saving in coke consumption.
INDUSTRIAL APPLICATION IN THE U.K.
Since the initial tests were performed a number of cupolas have been
installed or converted to operate with divided blast and more instal-
lations and conversions are planned. Details of divided blast cupolas
operating performance reported by eleven foundries and the corres-
ponding information relating to performance obtained previously when
operating the cupolas with one row of tuyeres, are given in Table
14.
The cupolas studied vary between 29 and 73 in.(73.7 and 185.4 cm)
internal diameter and between about 2 and 25 tons per hour in melting
capacity. The distance between the two rows of tuyeres is, in all
cases, between 30 and 42 in.(76.2 and 106.7 cm) irrespective of cupola
diameter. The proportion of the total quantity of blast air supplied
to the lower tuyeres in different installations varies slightly between
50 and 60% and, correspondingly, that supplied to the upper tuyeres
varies between 50 and 40%.
Table 14 shows that the benefits of divided blast operations reported
by the foundries have, in some cases, been smaller and in others
greater than those evaluated under the controlled experimental con-
ditions of operation and measurement at the BCIRA cupola plant.
Thus, when advantage has been taken of divided blast operation to
reduce charge coke consumption, this amounted to between 2.0 and
5.5% by weight of the metal, in some cases with a reported increase
in tapping temperature. On the other hand, where divided blast
has been introduced without reducing the charge coke consumption,
increased tapping temperatures of 70°to 140°F (40°to 80°C) have been
reported. In some instances, other altered features of cupola design
or operating practice accompanying the changeover to divided blast
operation may have contributed to improvement in performance. Some
foundries reported an increased melting loss of silicon with divided
blast operation, but at others, this has not been detected.
Some of the plants have been able to take advantage of the increased
rate of output made possible by the conversion of existing cupolas
to divided blast operation. At others, where coke consumption has
been reduced, the cupolas have been operated at a reduced blast rate
possibly, in the view of some operations, to the detriment of obtain-
ing the maximum possible saving of coke consumption.
The advantages of using two,rows of tuyeres with a divided blast
demonstrated at the BCIRA cupola plant have therefore been generally
confirmed by the results reported from foundries which are encour-
aging further adoption of the process as a means of obtaining oper-
ating economies.
87
-------
Table 14 DIVIDED BLAST OPERATION COMPARED WITH PREVIOUS PRACTICE AT INDUSTRIAL PLANTS
One Row of Tuyeres Divided Blast _
(10)
Cupola " Cupola
Internal Charge Metal Production Internal Charge Metal
Plant Diameter Coke Temperature Rate Diameter Coke Temperature
No. jn. % °F tons/hour in. % °F
36
Production
Rate
tons/hour
Notes
17.2
2820
4.5
36
12.7
2820
Producing whiteheart malleable iron.
Charges consist of pig iron (6-7%),
silicon pig (1.7-2.3%), returns 60%,
steel scrap 32%. Level of charge sili-
con increased by 0.2% for divided blast
operation.
60 14.0 2500
12.3 60 10.5 2500
16.8 Producing iron for ingot molds. Charges
consist of 60% heavy ingot mould scrap,
40% pig iron. No change in charge make-
up for divided blast operation.
3 44 19.0 2730
7.8 44 13.5 2770
Producing iron for duplexing for SG
iron castings. Charges consist of 70%
steel, 30% returns with ferro-silicon
briquettes. Level of charge silicon
increased by 0.12% for divided blast
pger^at ion.
28 17.5 2370
2.7 33 13.5 2730
Producing iron for general engineering,
machine tool and textile machinery cast
ings. Charges generally contain 30-40%
steel scrap.
54 12.5 2700 12.9 54 10.5 2700
12.9 Producing iron for automobile castings.
Charges consist of 35% pig iron, 40%
return scrap, 25% steel. No change in
charge make-up for divided blast opera-
tion. Silicon content of iron pro-
duced 0.1% lower than when operating
with 1 row of tuyeres but carbon equi-
valent value the same.
6 42 14.0 2520
7.8 42 9.0 2640
Producing iron for ingot moulds.
Charges consist of 67% pig iron, 32%
ingot mould scrap. No change in charge
make-up for divided blast operation.
Divided blast cupola operated with closed
instead of open slag hole as in previous
practice
7 72 12.5 2460 20.0 72 12.5 2590
Producing iron for ingot moulds. Cupola
converted to continuous tapping and
slagging for operation with divided
blast. ______
30 13.4 2560
3.6 30 11.0 2700
9 73 17.2 2790 16.8 73 13.8 2820
Producing grade 14 iron for general
castings. Divided blast applied to
new cupolas with increased shaft
height of 9 ft compared with previous
cupolas. Alterations to charge make-
up not connected with divided blast
operation prevent comment on effect of
divided blast on silicon loss or car-
bon pickup.
10 32 12.0 2620
3.4 29 12.0 2700
Producing grade 14 iron for bearing
housings and pump bodies. Because of
improved casting yield since intro-
duction of divided blast(but not ne-
cessarily because of it), charge con-
tains additional 5% steel & 1 addition-
al silicon briquette per 5 cwt charge
to replace 5% return^ scrag.
11 54 18.0 2700 11.£
54 , 15.5 2700
Producing castings for automobile in-
dustry. Typical charge-32% pig iron,
43% returns, 25% steel. Levels of
charge silicon & manganese increased
by 0.07%.
-------
NORTH AMERICAN EXPERIENCE WITH DIVIDED BLAST
One of the objectives of this study was to establish a liaison with
a number of foundries and individuals actively involved in experi-
mentation and development of divided blast techniques throughout
the industry. Among those contacted was Mr. Barry J. Davies, P.E.,
Manager of Foundry Development for Canron Ltd. of Hamilton, Ontario.
The Canron foundry division has been utilizing divided blast on two
of their cupolas for approximately two years and in that time Mr.
Davies states that they have realized a 27% reduction in their net
coke consumption with no appreciable drop in spout temperatures.
Fuel conservation was the primary incentive for instituting the di-
vided blast techniques, however, a number of unexpected benefits
accrued as the system was refined. The first, and most obvious,
improvement in cupola operation that was noticed was in the visual
appearance of the stack emissions. Though no comprehensive studies
were done to explain this reaction it would appear that a more effi-
cient combustion of the stack gases, in the upper stages of the cu-
pola, occurred with divided blast. This theory gains merit when
it is noted that the stack gas temperature at the exhaust fan was
reduced from approximately 550°F to 350°F.(11) This is a very signi-
ficant development from an energy conservation standpoint for a
greater portion of the sensible heat of the stack gas is obviously
being utilized in the furnace where it is most useful. This temper-
ature drop is one aspect not highlighted at all in the BCIRA study,
and heat losses at the charging doors and from the stack have been
an inherent problem with cupolas since their inception. Increased
melting rates in the range of 10 to 25% have also been achieved with
no appreciable drop in spout temperatures.(11)
Mr. Davies sighted another operational problem which was inadvertently
solved with the institution of divided blast. In their operation
they were having problems with hot spots near the end of each heat.
With divided blast, however, there was a difference in the pattern
of refractory burnback. As can be seen in Figure 29 this burnback
extended over a higher vertical range, but did not extend as close
to the shell as did the conventional cupola burnback pattern, thus
eliminating the hot spot problem. This could be a very important
factor for foundries where extended heats are necessary to meet pro-
duction demands.
When the divided blast system was first adopted at Canron the opera-
tors experienced a very significant drop in their silicon contents
as well as a sizeable reduction in the spout carbon content despite
the fact that their temperatures were always maintained at a very
high level. After much experimentation Mr. Davies says that they
have been able to minimize these losses by maintaining a precisely
balanced blast equally divided between the upper and lower tuyeres.
He states that, despite arguments calling for 40/60% ratios between
the upper and lower tuyeres, it has been their particular experience
that even a small amount of extra air through the bottom tuyeres
tends to promote increased silicon and carbon losses. Davies further
suggests that in order to ensure the equal distribution of the blast
89
-------
air the divided blast installation should include a new wind belt
with an independent row of tuyeres and a separate air weight con-
troller for each wind belt. This would facilitate the maintenance
of a constant flow rate to each row of tuyeres by continually cor-
recting for changes in back pressure due to variations in the den-
sities of the individual coke charges along with changes in emission
control system conditions during the course of the melt. The Canron
facility has employed this type of air control system on their di-
vided blast cupolas and at the present time their silicon yields
are approximately 85% and their spout carbons are reported to be
very close to the theoretical value.
Another problem which they have encountered during the past two years
has been slagging of the bottom row of tuyeres. Mr. Davies has been
told that a number of cupolas in the United Kingdom have experienced
a similar problem and that by going to smaller sized tuyeres, there-
by increasing the velocity of the blast, they were able to alleviate
this slagging problem. At Canron they are experimenting with various
size inserts in the lower tuyeres and although the problem has not
been completely eliminated they are hopeful that continued testing
will produce favorable results.
During the last phase of this project arrangements were made for
a visit to Canron Ltd.'s Stewart Street plant (the site of the con-
verted cupolas). Mr. Davies and Mr. Bud Meyers conducted us on a
tour of the facilities, during which time we familiarized ourselves
with the physical layout of the units and discussed the new operating
parameters being encountered with the divided blast system.
The two 48" acid lined, cold blast cupolas are located side-by-side,
which allows for consolidation of the blast air control system.
This arrangement also facilitates the transition from one unit to
the other as the cupolas are used on alternate days. Based on the
experimental work done by the BCIRA, a 36" spacing was chosen between
the upper and lower tuyeres and because of space limitations the
new wind boxes are concentrically mounted around the existing wind
boxes. There are six tuyeres in each row with the second row on
staggered centers above the tuyeres in the first row.
The main blower pipe splits into two separate mains each leading
to an individual windbox. The volume of air going through each is
controlled with two separate air controllers maintaining a flow rate
of approximately 4200 cfm equally divided between the two wind boxes.
This instrumentation services both cupolas thus reducing the initial
installed cost of the divided blast system, while ensuring a well
balanced air supply to both units.
The divided blast system has not created any problems with metal
chemistry and carbon pickup and silicon recovery levels have remained
consistent with those experienced during the conventional operation.
However, Canron is presently producing a very soft iron for pouring
large bottom plates (stools) for steel mill ingot molds, and finds
it necessary to raise the tapping carbon from 3.20% up to 3.80-4.05%.
This is accomplished by blowing graphite with nitrogen through a
lance into the iron accumulated in their 10 ton fore hearth.
90
-------
The addition of a second set of tuyeres requires a corresponding
increase in coke bed height. At Canron, this necessitated raising
the coke bed by 3 feet, using 3650 Ibs. of bed coke instead of the
2600 Ibs. used in the original single row operation.
Despite the increase in the size of the initial coke bed Canron has
achieved significant savings in raw material costs since they began
utilizing the divided blast techniques. The present system has per-
mitted them to drop from 240 Ibs. of coke per split to 170 Ibs. while
maintaining their metal charge at 1800 Ibs. This has resulted in
a drop from 13.5% by weight of metal, or a metal to coke ratio of
7.4 to 1 down to 9.4% by weight of metal, a metal to coke ratio of
10.6 to 1. In spite of the reduced consumption of coke there was
no loss in spout temperature in the 2750 to 280C°F range. No extra
splits are required during the 8 to 10 ton per hour melt campaigns
which often extend for 14 hours per day. So the net effect has been
to reduce the coke charge by 27% even after accounting for the larger
initial coke bed.
At the time of our visit Mr. Davies was experimenting with a coke
charge of 150 Ib. which represents a metal to coke ratio of 12 to
1. The effect of this ratio on metal chemistry has not yet been
documented, however, should these results prove favorable, Mr. Davies
envisions a total system coke savings on the order of $250,000 per
year.
A look at the overall economic picture shows that the initial capital
investment for the conversion of both cupolas to divided blast cost
is approximately $18,000. The installation was easily accomplished
during a regularly scheduled shut down. Mr. Davies suggests that
$9,000 per cupola may be a bit higher than usual due to space limita-
tions unique to the Canron Foundry. Since the inception of the di-
vided blast techniques, Canron has achieved an average annual savings
of approximately $170,000 which clearly seems to justify their ini-
tial investment.
Additional coke savings are being achieved with the divided blast
cupolas by eliminating coke from the last few charges and securing
the upper row of tuyeres. By running the cupola with only the single
row of tuyeres the cupola operator is able to burn the coke bed al-
most completely down before dropping bottom.
Mr. Davies firmly believes that, in view of the escalating cost of
coke, (which in Canada averages $100-$110 per ton) divided blast
is certainly a viable approach to fuel conservation. He points out,
however, one important factor which must not be overlooked. When
you install a new upper row of tuyeres 3 feet above the existing
tuyeres, extend the height of your coke bed accordingly; that is,
your new coke bed must be 3 feet above the one you are presently
using. This increase in coke bed must be amortized over the number
of charges that you put into the cupola daily in order to determine
your true net coke savings. Therefore, the cupolas having an extended
91
-------
melting campaign, which is the case at the Canron Foundry, will find
the divided blast more economical than those cupolas which use a
short melting campaign. It is conceivable that if your melt is short
enough, the extra bed coke required would outweigh the savings in
coke splits which you could operate successfully with and you could
possibly end up with a net increase in consumption of coke.
During the course of the industry survey, it was suggested by a number
of foundry managers that York Research Corporation contact Mr. Edgar
J. Hopkins of Alabama By-Products Corp., a major coke supplier for
much of the southeast. Mr. Hopkins is an acknowledged authority
on divided blast systems and has acted as a consultant to several
foundries, supplied by Alabama By-Products Corp., during the design,
installation and start up phases of their conversion to divided blast.
Mr. Hopkins points out that unlike most cupola modifications the
divided blast components can be fabricated and installed by in-house
personnel. He further notes that the down-time required, particularly
for the smaller cupola applications, can be reduced to a few hours
by having all the tuyere and wind box additions pre-fabricated and
cut-outs for the existing wind box and cupola shell clearly marked
prior to the scheduled shut down.
Mr. Hopkins reports that savings of 25-35% in coke charged, 25-50%
savings on refractory due to the change in the burnout pattern, and
a 20% increase in melt rate have been achieved by three foundries
which he recently assisted in converting to divided blast.(13)
These foundries, Macin Parts Corp. (Anderson, South Carolina), Bahan
Textile Machinery Company (Greenville, South Carolina), a subsidiary
of MPC, and Cox Foundry and Machine Company (Atlanta, Georgia), make
gray and ductile iron jobbing castings in a wide range of sizes.
The lengths of their heats vary from two to six hours.
At the time of the conversion, the foundries employed either No.
3% or No. 4 cupolas, acid lined to a 36" inside diameter. Cold blast
was used in all three cases. Two of the cupolas utilized six lower
tuyeres. When converted, three of the lower tuyeres were blocked
off and raised to a height of 36" above the remaining tuyeres. The
third cupola had four tuyeres, and four additional tuyeres were in-
stalled 36" above them. The total conversion time for all three
cupolas ranged from seven to thirteen manhours. Existing air weight
controls and blower systems were not altered.
Each of the cupolas was converted to divided blast using the same
formula. The total area for both the upper and lower tuyeres was
set at 6.5% of the inside diameter. Of this 6.5%, 40% was allotted
to the upper tuyeres and 60% to the lower tuyeres. This uneven dis-
tribution was chosen because it is believed that there is less re-
sistance to air flow at the level of the upper tuyeres (due most
probably to the difference in density of the charge at that level),
making the smaller tuyeres more efficient.(13)
92
-------
Since the installation of divided blast, the foundries have prepared
the cupolas as before with a good bed burn in. However, the total
bed height has been raised from 20" to 24" above the old established
bed height. This gives a bed height of 48-52".
The charges used by the three foundries in making gray an^ ductile
iron are similar. For gray iron, the charge is 30% steel, 30% pig
iron and 40% returns. This provides an in-going analysis of appro-
ximately 2.90 to 3.00 carbon and silicon charge of 1.80 to 2.00.
Before installing divided blast, the foundries were using a seven
or eight to one coke ratio, and now, with divided blast, they are
all using a ten to one coke' ratio. The result has been improved
carbon pickup with the final carbons ranging from 3.50 to 3.70 which
is a 60 to 70 point pickup on a 10 to 1 coke ratio. The silicon
loss is approximately 20% - about twice the loss encountered when
utilizing the conventional single row of tuyeres. With the divided
blast, temperatures have ranged from 2780° to 2840°F.(13)
The charge for ductile iron consists of 25% steel, 25% low carbon
sorel-metal and 50% ductile iron return. Some silicon is charged
into the stack as well. This gives a charge going into the cupola
of 2.50 carbon and 1.4 silicon. Formerly, a six to one coke ratio
had been used to maintain temperature and carbon pickup. Now with
divided blast, an eight to one coke ratio or 90 pounds of coke to
750 pounds of metal is used. Temperatures have ranged from 2820°to
2890°F, and final carbons from 3.70 to 3.90. Again a 20% silicon
loss has resulted with the final silicons ranging from 1.10 to 1.20. (13)
In addition to coke savings and increased melt rates, the foundries
have realized a number of other advantages with divided blast. One
benefit which has been noted is that even though the burnout pattern
is over a wider area, the total amount of refractory used, during
the course of a melt, is less and the foundries are often getting
two or three heats before having to reline the cupola.
Another unexpected bonus achieved with the use of divided blast was
the realization that even after as much as a 30 minute shutdown,
the cupolas recover more quickly than before. With the divided blast,
the cupolas are normally back up to 2820°F within five minutes after
start up. This reaction appears very similar to that achieved by
adding 3% oxygen enrichment to the blast.
An additional plus noted after the installation of the divided blast
and when clean scrap is used, is that there is almost no visible
plume. Any plume which is visible generally dissipates within ten
to twenty feet from the stack. This is attributable, it is believed,
to more complete combustion in the upper stages of the cupola as
a result of the redistribution of the blast air and the reduction
in the coke ash released.
93
-------
The primary disadvantage reported after conversion to divided blast
has been the increased silicon loss which, as noted earlier, is approx-
imately 20% compared to 10% with the single row of tuyeres. It
should be noted that, in our earlier discussions with Mr. B.J. Davies
of Canron Ltd., he stated that he had experienced similar difficulties
following Canron's conversion to divided blast. After much experi-
mentation he found that by installing independent air weight con-
trols for each wind box he was able to reduce the silicon losses
to a minimum by insuring a more uniform distribution of the blast
air between the upper and lower tuyeres.
While all three foundries had a very good cupola operation before
the conversion, the consensus is that the operations have improved
as a result of the divided blast. Mr. Hopkins warns, however, that
this system is not a cure-all and does not compensate for poor cupola
practices.
CONTROL OF METAL COMPOSITION
A major problem generally associated with divided blast systems is
a sizeable increase in silicon losses as well as a drop in spout
carbon content when compared with the element recovery levels asso-
ciated with a conventional single tuyere cupola. Much discussion
and years of research have been devoted to the solving of these pro-
blems, but to date, no clear cut and definitive answers have been
formulated. There was, however, an authoritative study presented
by the American Foundrymen's Society, in their third edition of
"The Cupola and Its Operation."(1) In the discussion, a detailed
comparison is made of various stages of combustion (percentages of
CO2 in the effluent gases) and their effect on the composition of
the final tap metal.
Silicon and Manganese
The determination of the carbon dioxide content of the stack gases,
together with the composition of the metal being tapped at the cupola
spout, affords a convenient means of relating the losses of silicon
and manganese during the process of melting to the various stages
of combustion. A correlation between gas composition and element
loss can be obtained from a knowledge of the carbon dioxide concen-
tration of the gases at the time a particular iron sample is taken
for analysis.
Based on such data Figure 31 graphically shows the relation between
the stage of combustion as determined by the carbon dioxide content
of the effluent gases, and the percentage loss of silicon and manganese.
The latter figure was obtained by deducting the net silicon and man-
ganese present in the final tap metal from the gross silicon and
managanese in the initial cupola charge. These curves show that as
the percentage of carbon dioxide (a moderately oxidizing gas) increases,
the loss of both silicon and manganese also increases. Conversely,
of course, the reverse situation is also true.
94
-------
14 13 12 II
STAGE OF COMBUSTION, PERCENT C02
10
95
Figure 31. Effect of stoge of combustion on element loss in cupola.
(I )
-------
From such data it is quite apparent that efficiency of combustion
is, in part, restricted by the losses of silicon and manganese when
considered from a purely economic viewpoint. It is also obvious that
these losses, unless definitely controlled, will have a direct effect
on the physical properties of the metal.
Carbon
The actual process by which carbon pick-up in the cupola is accom-
plished is still a somewhat debateable question. One view is that
the process is one of carbonization in which carbon monoxide reacts
with the metal to form the carbide. The AFS manual refers to a
series of tests performed by Rambush and Taylor which showed that
the carbon content of steel bars increased only slightly while they
were still solid but increased rapidly as they began to melt. Another
study done by Lownie et al was noted which reported similar results
and demonstrated that carbon pickup in the cupola well, where the
metal is in intimate contact with hot carbon, can account for but
a small percentage of the total.
It would appear that the carbon pickup is principally through car-
burization and is dependent, therefore, on the carbon monoxide con-
tent of the gases in contact with the melting and molten metal,
hence indirectly a function of the stage of combustion. To add
substance to this concept, reference was also made to a series of
analyses of metal and effluent gas obtained simultaneously as re-
ported by Massari.
These results, shown in Figure 32, indicate that carbon pickup in-
creases with a decrease in carbon dioxide content (increase in car-
bon monoxide content) of the gas. Although difficult to accomplish,
this data indicates that, within certain limits, the total carbon
content of the iron can be controlled if the conditions in the
furnace bed can be satisfactorily controlled. Variations in carbon
pickup with fuels of different reactivity are probably due to actual
differences in the CO/C02 ratio in the upper portion of the bed.
ECONOMIC CONSIDERATIONS
When attempting to establish an ideal operating efficiency in a
system, such as the divided blast cupola, one must look at the overall
effect of manipulating the level of combustion as a method of
achieving product control. Just as certain metallurgical benefits
result from careful control of the combustion processes of the
cupola, so too it can be shown that definite economic advantages
can also be realized.
Although not synonymous, melting efficiency (iron to fuel ratio)
and combustion efficiency (carbon dioxide content of the stack gas)
are closely related. A higher percentage of carbon dioxide in the
effluent gases is indicative of the more complete combustion of
the coke, and liberation of a greater proportion of the potential
heat in the fuel. Consequently, the cost of coke per ton of iron
96
-------
j* ^.UU
§3.70
r 3.60
f 3.50
CD
3 3.40
< 3.30
*" 3.20
**"
**
x*
X*
^x
X*
X
r1
15 14 13 12 II
STAGE OF COMBUSTION, % CO,.
HOT BLAST CUPOLA
10
Figure 32. Relationship between total carbon content of the
iron and the stage of combustion within the cupola. (' *
2.00
1.80
0.80
0.60
0.40
0.201
15 14 13 12 II 10
STAGE OF COMBUSTION, %C02(Vol.)
Figure 33. Stage of combustion in the cupola vs. melting.''
97
-------
melted will decrease with an increase of carbon dioxide in the effluent
gases, as illustrated in Figure 32. Hence, solely from the stand-
point of coke economy, it is evident that a definite savings will
result from more complete combustion.
However, it is also shown in Figure 33 that, as the efficiency of
combustion increases, the losses of silicon and manganese increase.
This means that, as the percentage of carbon dioxide in the stack
gases increases, the cost due to silicon and manganese melting losses
will increase. Consequently, aside from the metallurgical aspects
previously discussed, maximum economy is not attained with maximum
efficiency of combustion, since the saving in coke will not counter-
balance the cost of the elements lost.
If the summation of the cost of coke and element loss per ton of
iron melted is plotted against the stage of combustion (percentage
of carbon dioxide in the effluent gases) as in Figure 33, it will
be seen that the composite curves pass through an ideal minimum
point. These data, for a hot blast cupola, show that maximum economy
is obtained when the carbon dioxide concentration of the effluent
gases is held between 13 and 14 percent in the case of low silicon
iron, and in the vicinity of 12 percent when the metal melted con-
tains 1.40 percent silicon.
When these stages of combustion are achieved for the two irons con-
sidered, the most economical quantity of coke is consumed to melt
a given tonnage of iron with the minimum cost of alloys. For an
equivalent cold blast cupola melting operation, similar economies
will result when the carbon dioxide content of the stack gases is
from 1.5 to 2.0 percent lower.
In addition to these economies which can be accomplished by combus-
tion control, operation losses are minimized. It is obvious that
a control system that ensures greater uniformity of melting will
reduce shop losses and rejections.
The capital invested in combustion control equipment can be amor-
tized in a reasonably short time as a result of the various economies
which can be affected.
Most authorities agree that it is just as essential to utilize a
reliable means of controlling the quantity of air entering the cupola
as it is necessary to measure the quantity of coke and iron, and
to regulate the composition of the charge. However, the manner
in which the air is controlled merits close consideration.
A constant weight of air entering the cupola does not necessarily
result in a constant or uniform stage of combustion in the melting
zone, since the latter may vary depending upon the type, size, and
chemical resistivity of the fuel, in addition to other factors.
A constant volume of air likewise does not at all times contain
the same weight of oxygen with which to burn the coke, since the
9R
-------
concentration of oxygen (wt/cuft) will vary with barometric pressure
and atmospheric temperature. Similarly, the weight of oxygen de-
livered will vary with the humidity of the air since any water vapor
present replaces an equal volume of air. As a consequence the stage
of combustion will, of necessity, vary even if the blast control
equipment is designed to maintain a constant volume of air.
In view of the previous discussion relating to the metallurgical
aspects and economics of melting, it appears that for optimum opera-
ting conditions a more satisfactory control could be obtained by
keeping track of the carbon dioxide concentration in the effluent
gases and adjusting the blast volume, when necessary, to maintain
uniformly the most satisfactory rate of combustion. Though the
AFS discussion was geared to cupola . operation in general it takes
on added significance when related to the element loss problems
associated with divided blast systems. There are a number of gas
analyzers described in the AFS manual and in view of the vastly
improved control technology available today this approach to com-
bustion control may well be the answer to the element recovery pro-
blems which have vexed the proponents of divided blast for so long.
99
-------
SECTION VIII
AIR POLLUTION CONTROL
INTRODUCTION
As stated in the section on cupola operation air pollution control
for the cupola is an expensive proposition. It has been shown
that hot blast, divided blast and oxygen enrichment can decrease
the particulate grain loading from the cupola and can also decrease,
by varying amounts, the gas volumes to be treated. Conventional
pollution control costs can thereby be reduced significantly.
It was decided, however, to briefly investigate innovative pollu-
tion control equipment which could be applied to the cupola in
expectation of further reducing control costs.
STEAM-HYDRO AIR CLEANING SYSTEM, LONE STAR STEEL
This system is a highly efficient (99%+) pollution control device
which can be powered by the waste heat of the cupola. A schematic
of the system can be found in Figure 34. The CO laden air is com-
busted in a waste heat boiler and the steam produced is utilized
both to provide the required draft and scrub the particulate.
Preliminary particulate removal can be accomplished in the atomizer
section. This section has water sprays which can be used to re-
move the heavier particulate and cool the gas stream. Some agglo-
meration takes place in this section preparing the smaller diameter
particles for the second phase removal. The steam is injected
just prior to the mixing tube where collision between the water
droplets and the particulate take place. Encapsulation, nucleation
and droplet growth occur enabling sub-micron particulate to be
brought to a manageable size for removal by low pressure drop cyclone
A shock wave pattern is created in the mixing tube which induces
massive turbulent action for maximum scrubbing efficiency.
The device was tested for efficiency by Southern Research Institute
under a contract with EPA.(15) To quote from their results:
"The collection efficiency of the Steam-Hydro air cleaning system
is quite high. As measured using conventional (Method 5) techniques
on a source producing particulate having a mass mean diameter
of about 1 micron, the efficiency was measured at 99.90 and 99.84%
for two days of testing. Measured fractional efficiencies were
about 90% at 0.01 micron, about 70% at 0.05 micron, 85% at 0.1
micron, 99.9% at 0.5 micron, 99.99% at 1 micron and 99.6% at
5 microns. The minimum in the fractional efficiency is probably
real, but the actual value is somewhat uncertain because of diffi-
culties in making diffusional measurements in the time variable
open hearth process."
Costs
The average capital cost of the Lone Star System is between $3
and $5 per SCFM treated depending on the materials of construc-
100
-------
PARTICLE
ACCELERATOR
INJECTION WATER
STEAM NOZZLE
INLET
ATOMIZER WATER
INLET
SAMPLING
LOCATIONS.
FLUE GAS FROM
WASTE HEAT BOILER
CYCLONES
ATOMIZER SLURRY
CYCLONE
SLURRY
Figure 34. The Lone Star Steel steam-hydro air cleaning system.
101
Courtesy of Lone Star Sieel
-------
tion. Another method of partially estimating capital costs is
to determine the size of the waste heat boiler required. Knowing
the grain loading required by pollution regulations a graph
such as Figure 35 can then be used to determine steam require-
ments. Since the system pressure drop can also be calculated
the steam required to provide the proper draft can be obtained
from Figure 36. The higher of the two steam rates is then used
for pricing of the waste heat boiler. In the system tested the
following condition prevailed:
Steam usage: 7300 Ib/hr at 250 psig
Water usage: 61 gal/min (of which 55 gal/min is recoverable)
Air Flow: 13000 SCFM with a system back pressure of 6 inches
H20
Outlet particulate loading: 0.0007 gr/DSCF
When using the waste heat to power the boiler the only operating
cost incurred is the water and pumping cost which would be equi-
valent to any normal scrubber operation. The primary advantage
to this sytem lies in the fact that an I.D. fan is not required
for the draft. For a venturi type scrubber pressure drops of
60 inches of water are common and this necessitates the use of
a high h.p. I.D. fan to obtain similar removal efficiencies.
UNITED MCGILL, ELECTROSTATIC PRECIPITATORS
Electrostatic precipitators (ESP) have been rarely applied to
the cupola in this country for a number of good reasons. Dust
resistivities are high and an appreciable amount of electric
power is consumed. The necessity of low velocities for charging
purposes increases the size of the unit and the cost. Wire
breakages causing shutdowns are common and shutdowns of indivi-
dual sections of the ESP result in non-compliance with air pollu-
tion regulations. The United McGill ESP due to an innovative
design overcomes many of these inherent difficulties. The elec-
trostatic field is generated by sharp needles mounted on both
edges of the electrode plates. This supposedly allows the ESP
to operate at a lower voltage and amperage than conventional
units which use suspended wire or rigid frames as discharge elec-
trodes. The discharge electrodes are electrically positive which
means that the positive ions used as the charge carriers build
up an effective space charge. This would inhibit sparking and
arcing to a large degree. As a result the distance between the
discharge plate and the collector plates can be decreased enhancing
the collection efficiency. If increased collection efficiency
is not required, then the size of the ESP can be reduced and
the velocity through the unit increased. Since the field outside
of the immediate vicinity of the discharge needles is electro-
statically stable agglomeration takes place at an increased rate.
Another advantage to United McGill's ESP is that there is twice
the amount of collection plate area since the conventional wire-
plate arrangment is replaced by two plates of opposite polarity.
Continual rapping is not required to prevent back corona since
the electrostatic plate voltages are not high enough to cause
102
-------
U.
U
^C
.E
o
t_
0>
•«
tn
g
CO
CO
•s
2
LU
01
.001
0001
.c
^_
X.
^
x
N
k
!
W
^
W^
^^^
^
s
V
h^
V
v
s
s
k
v
s
s
s
^
^
%
^
^
x
^S
^
s
15 .06 .07 .08 .09 .10 .12 .14 .16 .18 .20 .24
STEAM RATE, Ib/lb gas
Figure 35. Energy requirement versus emissions
103
Courtesy of Lone Star Steel Company
-------
.16
.14
CO
o
en
.- -12
LU
ce
UJ
V)
.10
.06
.06
-2 -3 -4 -5 -6
DRAFT AT UNIT INLET, INCHES WATER
-7
-8
Figure 36. Typical steam rate required for providing draft, Lone Star Steel Company
-------
dialectric breakdown of the dust layer. The needle points, due
to their positioning, are self-cleaning and do not need a cleaning
system. Since rapping frequency is reduced, reentrainment of
the collected dust is consequently reduced. The ESP is usually
operated at an inlet temperature of about 500°F and is insulated
to protect the materials against acid corrosion caused by cold
spots in the system.
United McGill has had success in marketing their ESP in Japan
for cupola applications. Due to the innovative design it is
believed that the device may well be able to find a market in
similar applications in this country.
PACKED BED FILTER SYSTEM
The Canadian Centre for Mineral and Energy Technology has been
investigating the use of packed beds for air pollution control.
The advantages to such a system are: (16')
1. "Both particulate and gaseous emission abatement can be per-
formed by a single unit."
2. "Hot stack gases may be treated, not only avoiding the pro-
blem of gas cooling but also permitting heat recovery from the
cleaned gas as well."
3. "If the charge materials and particulate effluent are of
a suitable nature, the process can be used to simultaneously
preheat charge material and collect and recharge particulate
effluent, providing significant savings in both energy and materials,
4. "Considerable flexibility in filtration, pressure drop and
gas reactivity characteristics can be achieved by modifications
in the sizing and composition of the granular packed-bed materials."
A pilot study was done to determine the feasibility of such a
filter arrangement. Figure 37 shows a schematic diagram of the
process itself and a cross-sectional diagram of the laboratory-
scale unit. The vertical shaft labeled "C" includes inlet and
outlet manifolds and inlet and exhaust louvres which control
the flow of the bed material in the shaft. A baffle is located
within the shaft above the filtration section forcing the bed
material to flow against the wall containing the inlet manifold
and louvres. The vertical column attached to the top end of
the shaft contains the gas entry port "B" and the bed material
entry port "A" at the upper end. The hot gases and fines from
the bed material are then transported to the bed through duct-
work "D". "E" represents a screw-feeder which controls the feed
and removal rate of the bed material and collected dust.
Table 15 shows the results of the study and as can be noted the
efficiency for the device is quite high. A commercially sized
unit (7500 SCFM) is near completion at a foundry in Winnipeg.
The system is being marketed under the trade name "Hi-TEC System"
by Prasco Ltd. of Canada.
105
-------
x .
COARSE BED MATERIAL
HOT GAS PLUS
PARTICIPATES
SIZED BED MATERIAL
•LOUVRES
Figure 37. Schematic diagram and cross-sectional diagram
of laboratory-scale unit.
106
-------
Table 15. PRESSURE LOSSES, DUST CONCENTRATIONS, GAS AND BED MATERIAL
FLOW RATES, AND FILTRATION EFFICIENCIES FOR A NUMBER OF BED
FEED MATERIALS (16)
Packed Bed
Material
Type
M
M
M
M
M
M
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
Q
V
V
V
BB
BB
BB
CC
CC
CC
DD
DD
DD
Rate,
Ib/min
0.97
0.62
0.44
0.96
0.98
0.98
0.62
0.36
0.96
1.21
0
0
0
0
0.50
1.22
1.74
2.20
1.44
1.44
1.68
0.19
0.38
0.76
0.28
0.53
0.89
0.04
0.23
0.44
Filter
Pressure
Drop
in
w.g .
2.5-7.5
5.0
7.0
5.0
5.0
3.5-6.5
5.0
5.0
5.0
5.0
1.4
1.6
1.9
2.0
2.2
2.2
2.2
2.2
5.0*
5.0*
5.0*
3.5
3.5
3.5
3.5
3.5
5.0
3.5
3.0
3.0
Dust
Type**
MCD
ii
ii
ii
it
ii
MCD
ii
n
ii
GB
n
n
n
n
"
n
11
MCD
n
n
MCD
n
n
MCD
n
n
MCD
n
n
Characte
Inlet
Cone . ,
gr/SCF
135
134
134
45.5
87.8
81.6
135
132
82.4
85.1
56.6
65.8
137
150
148
118
153
130
30.5
31.4
35.7
141
144
153
144
142
82.6
38.0
81.2
118
ristics
Outlet
Cone. ,
gr/SCF
0.050
0.036
0.054
0.175
0.196
0.156
0.011
0.023
0.056
0.196
0
0.14
1.28
3.02
4.22
1.19
0.880
0.420
1.91
1.59
0.637
0.061
0.082
0.109
0.010
0.016
0.091
0.023
0.051
0.068
Filter
Efficiency,
%
99.96
99.97
99.96
99.62
99.78
99.81
>99.99
99.98
99.93
99.77
100
99.79
99.07
97.98
97.15
99.00
99.42
99.67
93.74
94.94
98.22
99.96
99.94
99.93
>99.9
99.99
99.89
99.94
99.94
99.94
Gas
Flow
Rate,
SCFM
10
10
10
15
15
15
10
10
15
15
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
15
10
10
10
*Pressure varied widely, figure reported is average pressure drop,
**MCD - Mixed Cupola Dust
GB - Glass Beads
107
-------
SECTION IX
DISCUSSION
This section of the report is intended to give an overview of the
technologies investigated and to discuss briefly other techniques
to save energy such as waste heat use and general energy conserva-
tion.
COMPARISON OF RECUPERATION, DIVIDED BLAST AND OXYGEN ENRICHMENT
To directly reduce coke consumption while maintaining cupola chemistry,
a foundry engineer has the three aforementioned technologies to
choose from in theory. In fact, however, the choice is usually
quite limited by present operation. Recuperative heat exchangers
are ideal for future plants and those already with hot blast opera-
tion. Divided blast is a cheap, easy method to greatly reduce
coke consumption in an existing cold blast operation. Oxygen en-
richment has proven itself to be an almost indispensible tool for
the foundry engineer but is, in some instances, too expensive to
justify on a small scale basis and is frequently misused by operating
personnel.
For new installations, recuperators offer numerous different ways
to recover the majority of the energy wasted by the cupola. Not
only can this energy be used for hot blast operation but a new
plant can be designed to use the energy for in-plant heating
(as was done at Chevolet Foundry) or for various drying processes.
The latter use is possible when the gas is cleaned first and then
transported to the combustion site under low oxygen conditions.
Oxygen can only be kept at this low level in those systems whose
gas take-off point is below 15-20 feet of charge. All of the
manufacturers mentioned in the recuperator section market these
type systems. In retrofitting a recuperator three problem areas
must be considered carefully by the engineer. First, there is
a high capital cost involved. Second, ground space should be
available for installation. Costs can be sharply increased if
superstructuring is necessary. Finally, since about one-half
of all cupolas are now externally fired and since these heat ex-
changers are relatively new, a careful financial analysis must
be performed to determine if it is worthwhile to convert to recupera-
tors. In some cases the heat exchangers are designed for natural
gas firing only and conversion cost to oil firing (if conversion
is possible) should be considered in the costing since cutbacks
of natural gas in many areas of the country is now certain.
Divided blast is the best system for those foundries having cold
blast cupolas and a need to reduce coke consumption. The tech-
niques recommended in this report can give 20 to 25 percent re-
duction in coke consumption with a very small capital outlay
for an average cupola.
108
-------
Finally, oxygen enrichment is also capable of reducing coke con-
sumption but in the majority of the foundries which were contacted
it was used principally as a combustion tool. The primary dis-
advantage when compared to the other two techniques is that the
operating cost is higher due to the outlay for oxygen. If,
however, coke continues to rise in price oxygen enrichment will
become increasingly competitive. It is important to note that
oxygen enrichment is frequently used as a cure-all for a bad
melt by inexperienced operators. Such use can be discouraged
by a thorough education of the operators. This caution also
holds true for all other techniques. Often the benefits of
a new melting technique can be lost if the operator is not
properly educated in its use and expected results.
If these three techniques are to be subject to life-cycle cost
analysis in conjunction with pollution control equipment, it
must be remembered that a decrease in pollution control costs
should result from the use of either recuperative hot blast,
divided blast or oxygen enrichment. A decrease in air flow
results in a decrease in the size of the pollution equipment
needed. A decrease in the coke percentages results in a lower
grain loading which allows for a slightly less efficient control
device to be used.
Waste Heat Use
The energy recuperated for hot blast from a cupola's exhaust
gas accounts for less than 50 percent of the energy content of
the gas. Among the technically feasible possibilities for use
of this energy are the following: in-plant heating, sand and
core drying, coke drying, scrap pre-heating, electrical generation,
steam and steam generation for compressors and fans. All of
these uses are indeed possible. The main drawbacks for actual
use are plant layout, low return on investment, and high initial
costs. The larger foundries with a number of cupolas and BOF's,
while having the initial capital for such schemes, have plant
layouts either non-conducive or too limited for actual inception.
The smaller foundries on tfre other hand do not have the capital
and could not afford the low return on investment. Probably
the only way most of these waste heat concepts will be put into
use is in a new plant or by a coincidence in plant layout as
in the Chevolet Foundry.
It is strongly recommended, 'however, that the Environmental Pro-
tection Agency be involved in assisting in the design of any
new facility and that all possible efforts be expanded in making
maximim possible use of the .cupola's waste heat.
Energy Conservation
While attention has been focused on the amount of energy wasted
by the cupola, it should be kept in mind that a foundry consumes
109
-------
and wastes energy in many other ways also. In order to alleviate
this the first step to take is to conduct a complete energy audit
of the entire facility. Gas, oil, coke, electricity should all
be monitored for a set period of time. Often devices such as
afterburners are oversized and the gas consumption can be reduced
without increasing the CO emissions. Cutbacks on lighting load
and rescheduling of shifts to maximize off-peak hours is another
way to reduce energy consumption. An energy audit, however,
is necessary to systematize the conservation effort. Once the
audit is completed those systems which consume the most energy
should be carefully checked. If a sizable proportion of the
total energy is going to in-plant heating (usually around 7 to
10%) then the plant boiler should be inspected and tuned. Also,
conservation methods such as reducing excessive ventilation or
leakage should be instituted by the housekeeping personnel.
Insulation on boilers and steam lines should be examined and
replaced if necessary. Leaks in steam and compressed air lines
should be corrected. Old electrical equipment and lighting should
be replaced.
While no data ia available on how much energy a foundry can save
by a conservation program, most engineers agree that for any
industry considerable dollar savings can be realized. In many
cases foundries do not have personnel trained in energy conser-
vation but contract private consulting firms to conduct a program
for them.
110
-------
SECTION X
REFERENCES
1. The Cupola and Its Operation. American Foundrymen's
Society, Third Edition, 1965.
2. Systems Analysis of Emissions and Emission Control in the
Iron Foundry Industry. A.T . Kearney and Company, Inc.,
Chicago, Illinois. Contract Number CPA 22-69-106. The
Environmental Protection Agency, February, 1971.
3. Hot Blast Affects Coke Efficiency. Foundry Facts, Washington,
D.C. No. 36, March-April, 1975.
4. Coursey, J., and J.F. Turner, III. Iron Foundry Cupola
Recuperative Emission Control Demonstration-Design Manual.
Flynn and Emrich Company, Baltimore, Maryland. EPS 650/2-74-
004, Environmental Protection Agency, Raleigh, North Carolina,
January, 1974, 51 p.
5. Turner, James F. III. Iron Foundry Cupola Recuperative Emission
Control Demonstration. Flynn and Emrich Company, Baltimore,
Maryland.EPA 6UO/2-76-004, Environmental Protection Agency,
Raleigh, North Carolina, January, 1976, 52 p.
6. Love, A.L. and D.R. Ruprecht, Determining the Economic Feasi-
bility of Oxygen Enrichment of the Cupola Wind. AFS
Transactions, pp. 437-441, 73-111.
7. Fredricks, S.L., J.W. Estes, An Economic Evaluation of the
Successful Use of Oxygen Enrichment on Cupolas in Terf gray
Iron Foundries. AFS Transactions (72-40), pp. 189-192.
8. Blessing, J. and J.W. Estes, The Potential for Oxygen Enrich-
ment of the Cupola in the Small Foundry. A paper prepared by
Airco Industrial Gases, 575 Mountain AVenue, Murray Hill, N.J.
07974.
9. Oxygen Enrichment of the Cupola. A technical report prepared
for York Research Corporation by Air Products and Chemicals,
Inc. following a field trip to corporate headquarters in
Allentown, Penn. pp. 1-14, September 19, 1975.
10. Leyshon, H.J. The Divided Blast Cupola. AFS Transactions,
Paper No. 73-56, pp. 202-207, presented at the 77th AFS
Casting Congress, Montreal, May, 1973.
11. Davies, Barry J. A technical paper presented at the
American Foundrymen's Society Ontario Chapter, October 5,
1973.
Ill
-------
12. Davies, B.J. and A.O. Bain. New Cupola Technology. Modern
Castings, pp. 46-47, September, 1975.
13. Hopkins, Edgar J. Case Study: "Three Foundries Convert to
Divided Blast." Foundry Facts No. 32, pp. 1-3, March-April,
1974, American Coke & Coal Chemicals Institute.
14. Fletcher, J.E. The Balanced Blast Cupola. BCIRA Research
Report, No. 91, pp. 1-11, October, 1930.
15. Smith, W.B. and T.D. McCain, Lone Star Steel Steam-Hydro
Air Cleaning System Evaluation. Southern Research Institute,
Birmingham, Alabama. 36 p. Contract Number 68-02-1308,
The Environmental Protection Agency, April, 1974.
16. Buhr, R.K. and R.D. Warda. A New Packed-Bed Gas Cleaner
for Pyrometallurgical Applications. Department of Energy,
Mines and Resources, Canada Centre for Mineral and Energy
Technology. Ottawa, Canada. Internal Report MRP/PMRL-75-2(R)
January 14, 1975, 16 p.
112
-------
APPENDIX A
113
-------
O
o
LL)
O
LU
Q.
34
32
30
28
26
24
22
20
18
16
14
12
10
8
6
4
2
0
V
\
\
\
%co
\
\
= 1.65;
\
\
>
>(2I.O
\
\
-%CO
1
C02
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.O
9.0
10.0
1 1.0
12.0
13.0
14.0
IS.O
16.0
17.0
18.0
19.0
20.0
21.0
\
\
*)
\
\
>
PERCENT
CO
34.7
33.0
31.4
29.7
28.1
26.4
24.7
23.1
21.5
19.8
18.2
16.5
14.9
13.2
1 1.6
9.9
8.3
6.6
4.9
a. 3
1.7
0.0
\
\
i
\
\
N2
65.3
66.0
66.6
67.3
67.9
68.6
69.3
69.9 ~
70.5
71.2
71.8
72.5
73.2
73.8
74.5
75.1 ~
75.7
76.4
77. 1
77.7 ~
78.3
79.0
\
\|
\
4 6 8 10 12
PERCENT CARBON DIOXIDE
14
16 18 20
A-I. Relotionship between CO, and CO in effluent gases.
114
-------
A-l. MEAN SPECIFIC HEATS
t(°F)
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
2700
2800
2900
3000
3100
3200
02
0.2188
0.2203
0.2221
0.2240
0.2259
0.2279
0.2299
0.2318
0.2337
0.2355
0.2373
0.2390
0.2406
0.2420
0.2434
0.2448
0.2461
0.2473
0.2484
0.2495
0.2506
0.2517
0.2527
0.2536
0.2545
0.2554
0.2562
0.2570
0.2578
0.2585
0.2593
0.2600
Cri=H/t-60
H2
3.420
3.434
3.442
3.448
3.452
3.455
3.458
3.462
3.466
3.470
3.475
3.480
3.487
3.494
3.501
3.510
3.519
3.. 528
3.538
3.549
3.460
3.572
3.584
3.596
3.608
3.620
3.632
3.644
3.656
3.668
3.680
3.692
Btu
H20
0.4448
0.4472
0.4499
0.4529
0.4562
0.4597
0.4634
0.4674
0.4715
0.4757
0.4800
0.4844
0.4888
0.4932
0.4976
0.5021
0.5066
0.5111
0.5156
0.5201
0.5245
0.5289
0.5334
0.5375
0.5415
0.5456
0.5496
0.5536
0.5575
0.5614
0.5652
0.5688
per Ib per
N2
0.2482
0.2485
0.2488
0.2493
0.2500
0.2509
0.2520
0.2531
0.2544
0.2558
0.2572
0.2586
0.2600
0.2614
0.2628
0.2642
0.2656
0.2669
0.2682
0.2695
0.2707
0.2719
0.2732
0.2742
0.2753
0.2764
0.2774
0.2784
0.2793
0.2802
0.2811
0.2819
deg F
CO
0.2485
0.2488
0.2493
0.2501
0.2511
0.2522
0.2535
0.2549
0.2564
0.2580
0.2596
0.2611
0.2627
0.2642
0.2657
0.2672
0.2686
0.2700
0.2713
0.2726
0.2739
0.2751
0.2763
0.2774
0.2784
0.2794
0.2804
0.2814
0.2823
0.2831
0.2840
0.2848
C00
0.2022
0.2086
0.2145
0.2201
0.2253
0.2301
0.2346
0.2388
0.2428
0.2465
0.2500
0.2533
0.2564
0.2593
0.2620
0.2646
0.2671
0.2694
0.2716
0.2737
0.2757
0.2776
0.2795
0.2813
0.2830
0.2845
0.2860
0.2875
0.2889
0.2902
0.2915
0.2927
115
-------
A-l. MEAN SPECIFIC HEATS (Continued)
Slag
(solid)
t(°F)
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
2000
2100
2200
2300
2400
2500
2600
2700
2800
2900
3000
3100
3200
*alpha
**gamma
Slag
(liquid)
Pig Iron
(solid)
2%
Carbon
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.1809
.1893
.1967
.2032
.2089
.2139
.2183
.2223
.2258
.2291
.2321
.2349
.2374
.2397
.2418
.2437
.2455
.2473
.2490
.2506
.2522
.2538
.2555
.2571
0
0
0
0
0
0
0
0
0
.3256
.3239
.3224
.3210
.3198
.3187
.3176
.3166
.3156
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.1200
.1223
.1245
.1275
.1292
.1309
.1327
.1346
.1367
.1392
.1422
.1458
.1500*
.1671**
.1658
.1647
.1638
.1631
.1626
.1622
.1620
.1619
.1618
.1618
.1619
.1620
.1622
.1624
.1627
4%
Carbon
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.1308
.1326
.1353
.1403
.1404
.1409
.1417
.1428
.1441
.1457
.1475
.1497
.1523*
.1622**
.1616
.1611
.1608
.1606
.1604
.1604
.1604
.1605
.1607
.1608
.1611
.1613
.1616
.1619
.1623
Pig Iron
(liquid)
2%
Carbon
0
0
0
0
0
0
0
0
0
0
0
0
0
.2208
.2187
.2168
.2151
.2136
.2122
.2110
.2099
.2089
.2079
.2070
.2062
.2054
4%
Carbon
0.2221
0.2197
0.2176
0.2157
0.2140
0.2124
0.2109
0.2095
0.2083
0.2072
0.2062
0.2052
0.2043
116
-------
A-2. HEATS OF FORMATION
Compound
A1203
Al203-Si02
CaO
CaC2
CaO-Si02
2CaO-Si02
3CaO-Si02
CaO-Al203
2CaO-Al203
3CaO-Al203
4CaO-Al203
12CaO-7Al203
3CaO-Al203.2si02
CaO-Al203-6Si02
4CaO-Al203'Fe203
CaS
C02
CO
Ccoke
H20(g)
H20(l)
FeO
Fe304
Fe203
FeO'Si02
2FeO-S102
Fe3C
FeSi
FeS
MgO
Molecular
Weight
101.96
162.05
56.08
100.09
184.42
64.10
116.17
172.25
228.33
158.04
214.12
270.20
326.28
1386.68
390.38
518.58
485.98
72.15
44.01
28.01
12.01
18.016
18.016
71.85
231.55
159.70
131.94
203.79
179.56
83.91
87.92
40.32
84.33
k cal
per gram
molea
399.09
644.6
151.9
288.45
556
15
378.0
538.5
688.4
551C
704
861
1026
4617
1303
1828
1211
115.3
94.052
26.416
-2.38d
57.798
68.317
63.7
267.0
196.5
276
343.7
-5.0
19.2
22.72
143.84
266
Btu
per Ib
mole*3
718,360
1,160,300
273,400
519,210
1,001,000
27,000
680,400
969,300
1,239,100
992,OOOC
1,267,000
1,550,000
1,847,000
8,311,000
2,345,000
3,290,000
2,180,000
207,500
169,294
47,549
-48,280^
104,036
122,971
114,700
480,600
353,700
497,000
618,700
-9,000
34,600
40,900
258,910
479,000
Btu
per
lbb
7045.5
7160
4876
5187.4
5430
420
5857
5627
5427
6280C
5920
5740
5660
5993
6008
6345
4485
2877
3846.7
1697.6
-356d
5774.7
6825.6
1596
2076
2215
3760
3036
-50
412
465.2
6421
5680
117
-------
A-2. HEATS OF FORMATION (Continued)
Compound
MgO'Si02
2MgO-Si02
MnO
Mn304
Mn203
MnO 2
MnS
MnO-Si02
Si02
Na2C03
S02
Ti02
a. At 25°C.
b. At 60°F.
c. Estimated.
d. Based on P
Bur. Stds.
Molecular
Weight
100.41
140.73
70.93
228.79
157.86
86.93
87.00
131.02
60.09
106.00
64.07
79.90
.H. Dewey and D
, 21,457 (1938)
k cal
per gram
mole
357.9
488.2
92.0
331.4
232.1
124.5
48.2
302.5
205.1
270.3
70.96
218.0
.R. Harper:
.
Btu
per Ib
mole
644,200
878,800
165,600
596,500
417,800
224,100
86,800
544,500
369,200
486,500
127,730
392,400
J. Res.
Btu
per
Ib
6416
6244
2335
2607
2646
2578
997
4156
6144
4590
1993.6
4911
, Nat'l
118
-------
A-3. HEATS OF REACTION
Reaction
ccoke + °2 = C02
Ccoke + %®2 = ^
CO + ^02 = C02
^C02 + hC k = CO
kcala
96.43d
28.79d
67.636
-19.42d
Btub
173,570d
51,820d
121,745
-34,960d
Btu per lbc
14,452 /lbCd
4,315 /lbCd
4,346.5/lbCa
-2,910 /lbCd
H2°(g) + ccoke =
CO + H2
CaC03 = CaO + C02
CaC03-MgC03 =
CaO + MgO + 2C02
•29.00d -52,210d -2,898 /lbH2Od
-42.5 -76,500 -764 /lbCaC03
-72
MgC03 = MgO + C02 -28
130,000
-50,000
-700 /IbCaMg
(C03)2
-600 /lbMgC03
a. At 25°C.
b. At 6QOF.
c. Estimated.
d. Based on P.H. Dewey and D.R. Harper: J. Res., Nat'l
Bur. Stds., 21,457 (1938).
119
-------
APPENDIX B
CONVERSION FACTORS
Environmental Protection Agency policy is to express all
measurements in agency documents in metric units. When
implementing this practice will result in undue cost or
lack of clarity, conversion factors are provided for
the non-metric units used in a report. Generally, this
report uses British units of measure. For conversion
to the metric system, use the following conversions:
To convert from
BTU/lb-F
BTU/min
BTU/lb
cfm
Of
ft
gal
gpm
hp
in. we
Ib
lb/ft3
oz/in2
psig
To
J/kg-K
W
J/kg
m-Vsec
°C
m
m3
m^/sec
W
N/m2
kg
kg/rn-^
N/m2
N/m2
Multiply By
4184.
17.573
2326
.0004719
5/9(0F_32)
.3048
.003 x 785
.00006309
745.7
248.84
0.454
16.018
430.922
6,894.757
120
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-76-071
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Energy Conservation Techniques for the Iron
Foundry Cupola
5. REPORT DATE
March 1976
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Dennis J. Martin
York Research Corporation
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Flynn and Emrich Company
3001 Grantley Avenue
Baltimore, Maryland 21215
10. PROGRAM ELEMENT NO.
1AB013; ROAP 21ARO-002
11. CONTRACT/GRANT NO.
68-02-0286
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 6/72-12/75
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTESpr0ject officer for tnis repOrt is Robert C. McCrillis, Mail
Drop 62, Ext 2557.
is. ABSTRACT
Tne j^po^ gjves results of an investigation of various existing or emerging
technologies which can be used to reduce the energy consumption and pollution control
costs of typical cupola operations. The investigation was prompted by: the increasing
difficulty of cupola operation for some foundries caused by the rapid rise in fuel costs;
and the financial burden on smaller foundries resulting from having to install highly
efficient pollution control devices on cupolas. The report details options available
to foundries in terms of technological devices which will conserve energy and capital.
Included in this investigation were hot blast recuperation, divided blast, oxygen
enrichment, and innovative pollution control equipment.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI I'icld/Group
Air Pollution
Energy
Conservation
Iron and Steel
Industry
Foundries
Cupolas
Cost Effectiveness
Fuel Consumption
Air Pollution Control
Stationary Sources
Energy Conservation
Hot Blast Recuperation
Divided Blast
Oxygen Enrichment
13B
10B
14A
11F
13M
21D
18. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (Tliis
Unclassified
21. NO.
PAGE
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |