Full text of "Furnace heating; a practical and comprehensive treatise on warming buildings with hot air"

See other formats

FURNACE
H E ATI N G

*\T

\ v

FURNACE HEATING

A PRACTICAL AND COMPREHENSIVE TREATISE ON
WARMING BUILDINGS WITH HOT AIR

WILLIAM G. SNOW

MEMBER

American Society of Mechanical Engineers
American Society of Heating a*>d Ventilating Engineers

FIFTH EDITION

REVISED AND ENLARGED

NEW YORK

DAVID WILLIAMS COMPANY

239 West Thirty-ninth St.
1915

COPYRIGHTED, 1909, 1915
BY DAVID WILLIAMS COiMPANY

VIM

PREFACE.

When " Furnace Heating " was first published the heat
unit basis of making heating computations was little used in
connection with furnace work.

Rule of thumb methods of figuring prevailed. The author
endeavored to reduce to a scientific and -practical basis the
computation of the grate surface necessary to meet given con-
ditions, the proportioning of pipes and registers, the design of
hot- water combination systems, the layout of fan-furnace com-
bination systems, etc.

It was aimed, by means of tables, to make the treatise con-
venient for ready reference.

Furnace Heating has been twice revised since its original
publication, and it is hoped that this revision will supply much
material for which there has been a demand.

The work has been considerably increased in size, many new
illustrations have been added, and the latter part of the book
is devoted to a collection of articles by others to whom it has
been intended to give due credit in each case.

WILLIAM G. SNOW.
BOSTON, 1915.

341781

TABLE OF CONTENTS

I. Furnaces ....... 7

Area of Air Passages — Joints — Materials Employed — Cast Iron
vs. Steel Plate — Types of Furnaces — Dome Furnace — Two Sec
tion Fire Pot Furnace — Steel Plate Furnaces — Grates — The Fire
Pot — Brick Lined vs. Cast Iron Fire Pots — Combustion Cham-
ber— Radiator — Evaporating Pan — Other Types of Furnaces —
Furnaces for Other Fuels — Gas Furnaces — Soft Coal and Gas
Furnaces — Heating Surface— Secondary Heating Surface — Ra-
diation and Convection — Heating Surfaces of Furnaces and
Boilers — Efficiency — Heating Capacity — Size of Furnaces for
Blocks — Manufacturers' Ratings.

II House Heating . . . . . -31

Comparative Merits of Furnaces and Other Systems — Location
of the Furnace — Foundation — Furnace Pit — Brick Setting —
Portable Setting — Portable vs. Brick Setting — Twin Furnaces
— Twin Furnaces vs. Separate Ones — Smoke Pipes — Chimney
jques — Area of Cold Air Box — Location of Cold Air Box — Ma-
terial of Cold Air Box — Cold Air Room — Cold Air Inlet — Air
Filters — Return Duct and Air Supply — Recirculated Air— Size
of Air Pipes — Velocity of Air in Pipes — Length of Hot Air Pipes
—Methods of Piping — Trunk Line Systems— Relation between
Grate Surface and Pipe Area — Risers or Vertical Flues — Sepa-
rate Risers — Location of Risers — Material of Pipes — Area and
Size ot Registers — Location of Registers — Floor and Wall Reg-
isters— Pattern and Finish of Registers — Registers — Manage-
ment ot a Furnace — Suggestions to Purchasers — Furnace Tests
-A Cold Day Test — Test in Another Dwelling — Heating from
Below Zero.

III. The Combination System .... 68

Hot Water and Hot Air — Direct Radiation — Hot Water vs. Hot
Air — Valves on Radiators — " Balance '' of the System — Heating
Surface in Furnace — Hot Water Combination Heaters — Direct
Radiating Surface — Indirect Radiating Surface — Heating Con-
servatories— Tapping of Radiators — Sizes of Pipes — Open Tank
vs. Pressure Systems — Expansion Tank and Connections — Sys-
tem of Piping — Steam Combination — Heat Given Off by Direct
Radiators.

IV. Air, Humidity and Ventilation ... 82

Composition and Impurities of the Atmosphere — Humidity —
Expansion of Air and Absolute Temperature — The Flow of Air
in Pipes — Velocity of Air in Flues — Importance of Ventilation
— Causes of Atmospheric Vitiation — Effects of Foul Air on
Health and Comfort — Necessity for Ventilation — Standards of
Ventilation — Compulsory Ventilation.

Table of Contents.

V. The Heating and Ventilation of School Build-
ings > 102

General Discussion — Relative Fuel Consumption — The Furnace
— School House Heaters — Air Passage in Furnace — Portable or
Brick Setting — Size of Furnace — Corridor Heater — Location of
Furnace — Cold Air Room — Fresh Air Supply — Return Air
Openings — Mixing Dampers — Location of Flues — Material of
Flues — Hood Above Flues — Area of Flues — Ventilating Flue
Dampers — Registers and Screens — Stack Heaters — Size of Stack
Heater — Arrangement of Stack Heater — Boiler with Coils in
Ventilating Flues.

VI. Heating of Public Buildings, Churches and

Stores • ';*• . . . . - . . 116

In General — Size of Furnace — Another Method to Determine
Size of Furnace — An Approximate Method to Determine Size
of Furnace— Area of Cold Air Box — Fresh Air Inlet— Location
of Furnace and Area of Flues — Location of Registers — Ventila-
tion— Size of Stack Heater — Janitorial Shortcomings — Hot
Water Combination — Smoke Pipes and Flues, — The Heating of
Stores — Cold Air Box and Registers.

VII. The Fan Furnace Combination System . . 132

Advantages — Application of the System — Location of the Fan
Location of Driving Apparatus — Size of Furnaces — Kind of
Furnaces — Area of Air Passages in Furnaces— Setting — Types
of Fans— Speed of Fans — Fan Capacities — The Motive Power
— Area of Ducts and Flues.

VIII. Temperature Control . ' » . . ., . 139

General Remarks— Types of Regulators — Damper Connections
— Operation of the Regulators — Control of Mixing Dampers.

IX. Estimates and Contracts .... 141

Forms and Blanks — Estimates — Specifications — Guarantee —
Payments.

X. Fuels, Miscellaneous Tables and Data . . 145

Fuels — Chimney Flues — Capacity of Coal Bins.

XI. Furnace Erection and Fittings . . .152

Furnace Fittings — Furnace Casing — Cold Air Supply — Cold
Air Box — CasingTops — Collars — Stock Fittings — General Hints
on Furnace Erection and Piping — Casing — Hood or Bonnet —
Casing Collars — To Attach Collars to Bonnets — Elbows — Reg-
ister Boxes — Side Wall Registers — Cold Air Connections —
Making Pipe — Elbows — Register Boxes — Shoes — Stack Offsets,
Elbows and Tees — Register Collars — Side Wai' Registers —
Fittings for Oval Pipes — Easy-flow Fitting for Boot — Another
Type of Transformation Elbow — Fittings Having Profiles in
Parallel Planes.

Table of Contents. 5

CHAPTER PAGE

XII. Miscellaneous Notes and Data on Furnace

Heating . . . . . . .213

Causes of Failure in Furnace Heating Systems — Directions for
Setting and Piping Furnaces — Location of Hot Air Registers-
Furnace Air Supply — Installing Furnace Plants in Old Houses
— Sizes of Small Pipes Based on Cubic Contents of Rooms —
Meaning of " Equivalent Glass Surface " — Proportions of Fur-
naces and Furnace Heating Systems — The Installation of Fur-
naces— Trunk Line System of Furnace Piping— The Control of
Air Leakage Around Windows — Leakage Around Different
Types of Windows — Testing a Furnace Plant in Warm Weather
— Test of a Fan-furnace Combination — Advantage of Air at
Relatively Low Temperature — Fan Furnace Heating — Use of
Small Electric Fans in Connection with Furnaces — Practical
Application of a Desk Fan — The Efficiency of a Desk Fan —
Miscellaneous: Fire Hazards of Heating Systems; Radiation
from Red Hot Iron; Suitable Size Coal to Use.

CHAPTER I.

FURNACES.

A furnace consists essentially of a stove within a casing. Air
is admitted to the space between the two, where it becomes heated,
rises, and flows through the pipes to the various rooms.

The earlier forms of furnaces were practically ordinary heat-
ing stoves incased in brick work. Such furnaces were very
deficient in heating surface, and consequently were wasteful in
the consumption of fuel. Various methods were adopted to in-
crease their heating surface and efficiency. Radiators were added
through which the gases would pass and lose a considerable por-
tion of their heat before reaching the smoke pipe. Projections
or extended surface in the form of pins or ribs were cast on the
fire pot, or the pot, in some cases, was made corrugated. In other
furnaces flues were added, through which the fresh air supply
would pass, surrounded by hot gases.

Small air flues, pins and ribs retard the flow of air over the
heating surface, hence are not so effective as, at first thought,
they appear.

AREA OF AIR PASSAGES.

Furnaces with sufficient heating surface properly arranged
and having the area of the air passages not greatly in excess of
the combined area of the warm air pipes will, with a steady fire,
deliver air at a fairly uniform temperature, even during strong
winds.

When the passages are too large the wind will force an excess-
ive amount of air through the furnace, much of which will fail
to come in contact with the heating surface, with the result that
the air issuing from the registers will vary greatly in velocity and
in temperature.

The examination of a number of well proportioned furnaces
showed the average area for the passage of air to be about 180
square inches per square foot of grate surface, equal to about ij4
square inches of free air-way to each square inch of grate surface.

8 Furnace Heating.

JOINTS.

A furnace should have as few joints as possible, consistent
with proper provision for expansion and contraction. These
forces are practically irresistible, and if proper allowance for their
action is not made, something must give way, causing, as a rule,
the leakage of gas. Where the sections join, a deep cup joint
packed with kaolin, asbestos cement or other suitable material
should be used, permitting a reasonable amount of " play " with-
out the escape of gas.

MATERIALS EMPLOYED.

The materials chiefly used in the construction of furnaces are
cast iron and wrought iron or steel plate. Much has been stated
(especially by the makers of steel plate furnaces) as to the ease
with which gases pass through cast iron at high temperatures.
The experiments most quoted, however, were made on thin plates
and under conditions unlike those existing in a furnace.
The best authorities on heating and ventilation agree that the
danger of contamination from this source is very slight, and is not
to be compared with that from ill fitting joints and other leaks
due to bad workmanship, or to causes having nothing whatever
to do with the kind of materials used.

CAST IRON VS. STEEL PLATE.

Cast iron furnaces may be built in almost any desired form
and arranged to present large radiating surfaces with few joints.

The variety in design with wrought iron or steel plate is much
more limited. The superior weight of cast iron furnaces over
those of other materials renders them less susceptible to sudden
variations in temperature with changes in the condition of the
fire. When once heated, the castings take longer to cool than thin
steel plate; consequently the temperature of the air passing
through the furnace is maintained more nearly constant. In
point of durability cast iron is thicker and less subject to corro-
sion than wrought iron or steel plate. It is, therefore, more
suitable for use in damp places.

Steel plate furnaces transmit heat readily, and with thor-
oughly riveted seams and well packed joints afford little oppor-
tunity for gas leakage.

Furnaces. g

TYPES OF FURNACES.

The better class of cast iron furnaces have a radiator,
generally placed at the top, through which the gases pass and be-
come cooled before reaching the smoke pipe. They have but
one damper, combined as a rule with a cold air check. Many
of the cheaper furnaces have no radiator whatever, in the true
sense of the term; the gases passing directly to the smoke pipe,

Fig. 1. — Cast Iron Furnace with Radiator at Top.1

carrying with them much heat that should be utilized. Such
direct draft furnaces are very wasteful, but find a market among
certain builders, whose chief requirement is that a furnace shall
have a large casing to deceive prospective purchasers as to its
actual capacity.

DOME FURNACE.

Fig. 2 shows a furnace of extremely simple construction. A
cast iron fire pot surmounted by a steel plate dome. Furnaces of

Furnace Heating.

this general type are often used in the cheaper classes of dwell-
ings. There is practically no flue travel for the gases.

While furnaces of this general design may be effective
heaters, they are not as economical in the use of fuel as are
those having a radiator of some sort through which the gases

Fig. 2. — Furnace with Cast-iron Fire Pot and Steel Plate Dome.

must travel in passing from the combustion chamber to the smoke
pipe.

TWO SECTION FIRE POT FURNACE.

Fig. 3 shows a cast iron furnace with a two-section corru-
gated fire pot and a corrugated combustion chamber on which
rests a cast iron radiator; a popular furnace in certain localities,
the principal advantage claimed being less likelihood of fire pot
cracking than if made in a single piece. All gases must pass

Furnaces.

through the radiator enroute to the chimney giving up a large
portion of their heat.

STEEL PLATE FURNACES.

In the ordinary steel plate furnaces (see Fig. 3), the gases
pass downward through a radiator located below the top of the

Fig. 3. — Cast-iron Furnace with Two Section Corrugated Fire Pot and Corrugated
Combustion Chamber.

furnace. In addition to the damper in the smoke pipe, a direct
draft damper is used, to give a direct connection with the funnel
when coal is put on, to prevent the escape of gas to the cellar.

GRATES.

No part of a furnace is more important to the user than the
grate. That much study has been put into their design is shown
by the many styles that have been put on the market.

Furnace Heating.

The plain grate, oscillating about a center pin, was for a long
time the one most commonly used. Such grates were usually
provided with a clinker door through which a poker could be
introduced to remove any refuse too large to pass between the
grate bars.

Grates of the draw center and dump center type followed.
In all these the removal of ashes takes place principally around

Fig. 4. — Cast-iron Furnace (Less Casing) with Steel Radiator.

the circumference, decreasing toward the center, where the mo-
,tion ceases. The action of such grates tends to leave a cone of
ashes in the center of the fire, causing it to burn more freely
around the edges. Vigorous shaking often results in depositing a
considerable quantity of unconsumed coal in the ash pit before the
ashes near the center of the grate can be dislodged. Different
forms of rocking grates have been used, which, though easy to
shake, have not proved effective in breaking up clinkers, and have
been liable to clog and restrict the passage of air through the fire.

Furnaces. 13

The most common type, the revolving triangular pattern, is now
used in many of the leading furnaces. It consists of a series of
triangular bars, having teeth. The bars are connected by gears
and are turned by means of a detachable lever. If properly used
when the fire is of proper thickness, this grate will cut off a slice
of ashes and clinkers over its entire area, with little, if any, loss
of unconsumed coal. Its action tends to break up the mass of
fuel, permitting the air to pass freely through the fire and causing
fresh coal to ignite quickly.

THE FIRE POT.

Fire pots are generally made of cast iron or of steel plate lined
with fire brick. The depth varies considerably, ranging from
about 12 to 1 8 inches. In cast iron furnaces of the better class
the fire pot is made very heavy to insure durability and to render,
it less likely to become red hot. Many furnaces have the fire pot
made in two sections, the makers claiming less liability of crack-
ing, and in case of repairs less expense, than with a pot made in
one piece. On the other hand, the latter presents fewer joints,
and in point of durability often lasts, with good management, more
than 20 years.

The heating surface of cast iron fire pots is often increased,
as previously stated, by corrugations, pins or ribs. Clinkers
never adhere to cast iron. To facilitate molding, a slight taper is
necessary in all cast iron fire pots. An excessive taper is unneces-
sary and misleading. In comparing the size of furnaces the aver-
age diameter of the fire pot should be used as a basis, in order to
allow for the difference in taper that may exist.

A fire brick lining is essential in a wrought iron or steel plate
furnace to protect the thin shell from the intense heat of the fire.
It is claimed for such fire pots that more perfect combustion is
obtained than in a cast iron pot due to the fact that the un-
burned carbon escaping from the fire is entirely consumed by
this intense heat before coming in contact with the comparatively
cold surface of the radiator. The fire requires less attention
and the air passing through the furnace is not likely to become
overheated. Brick lined pots are generally of the same diameter
throughout, no taper being necessary.

14 Furnace Heating.

BRICK LINED VS. CAST IRON FIRE POTS.

Since brick lined fire pots are much less effective than cast iron
in heat transmitting power, such furnaces depend to a great extent
for their efficiency on the heating surface in the dome and the
radiator. This is much greater as a rule than in cast iron fur-
naces.

When coal is put on, the direct draft damper is opened, which

Fig. 5. — Steel Plate, Brick Lined, Indirect Draft Furnace.

cuts out all the heating surface in the radiator ; the radiant heat
from the top of the fire is checked by the layer of fresh coal,
and as the heat from the fire pot must pass through about 2
inches of fire brick, it is obvious that until the gas has burned
off and the direct draft damper can be closed, comparatively
little heat is given off by the furnace, with the result that the
temperature at the registers will fall. Under similar conditions,
with a direct draft furnace having a cast iron fire pot, the heat of

Furnaces. 15

the fire will be readily transmitted through the sides of the pot
while the fresh coal is becoming ignited. No part of the heating
surface being cut off during this period, a more even temperature
at the registers will be maintained.

The overheating of the air may be avoided in any furnace by
selecting one so large that it will never be necessary to force it to
the extent that the surfaces become red hot. A fire hot enough
to heat a heavy cast iron fire pot to redness would be likely to have
the same effect on a portion of the thin dome oi a wrought iron
furnace.

COMBUSTION CHAMBER.

The body of the furnace above the fire pot, commonly called
the dome or feed section, provides a combustion chamber, which
should be of sufficient capacity to permit the gases to become
thoroughly mixed with the air passing up through the fire or
entering through openings provided for the purpose in or around
the feed door. In most furnaces this space is somewhat larger
than the capacity of the fire pot. In many of the cheaper ones,
however, it is very much restricted, resulting in incomplete com-
bustion of the gases and waste of heat.

RADIATOR.

The radiator, so-called, with which all furnaces of the better
class are provided, is separate from the dome or combustion
chamber and affords a sort of reservoir in which the gases are
retained in contact with the air passing through the furnace until
they have parted with a considerable portion of their heat. The
design of the radiator materially affects the efficiency of the furnace.

Radiators are built of cast iron, of steel plate or of a combina-
tion of the two. The former material is more durable, and can be
made with fewer joints, but owing to difficulties in casting radia-
tors of considerable hight, steel plate is often used for the
sides.

Fig. 6 shows a top view or plan of a cast iron radiator show-
ing the course of the gases which pass from the dome through
the short connection, then divide and pass in opposite direction
around the radiator to the smokepipe.

In some furnaces the connection between the dome and the

Furnace Heating.

radiator is nearly opposite the smoke outlet, the gases passing
around the entire ring instead of dividing as shown in Fig. 6,
one-half the gases going each way.

Steel radiators may be made any desired hight, and cast iron
for the top and bottom. The effectiveness of a radiator depends
on its form, its heating surface and the difference between the

Smoke Pipe

Fig. 6. — Plan of Cast-iron Radiator Showing Course of Gases.

temperature of the gases and the surrounding air. Its form
should be such that a thorough contact with the air passing
through the furnace will be secured. Owing to the accumulation
of soot, the bottom surface becomes practically worthless for
heating after the furnace has been in use a short time, hence sur-
faces to be continuously effective must be self cleaning.

As to the location of the radiator, if placed low down the
gases are surrounded by air at relatively low temperature, which
renders the radiator, foot for foot, more effective than if placed
near the top and surrounded by warmer air. If the radiator is
placed too low the cold air surrounding it near the base of a
furnace is likely to cause condensation of the gases and cor-
rosion. This also has the tendency of decreasing the efficiency
of the furnace.

Furnaces. ij

EVAPORATING PAN.

The evaporating pan, with which nearly all furnaces are pro-
vided, is sometimes placed where it will be of little service,
It is usually placed, however, above the level of the grate, where
there is sufficient heat to cause a rapid evaporation. Care should
be taken to keep the evaporating pan clean or the action of the
heat on the sediment in the bottom, in case the pan becomes
dry, is likely to cause a nauseating odor to pervade the house. To
insure a supply of water in the pan at all times a plumber's tank
and ball cock, properly connected, may be used with convenience.
The author considers it desirable to have the ball cock outside
the furnace where it is accessible rather than inside the casing.

OTHER TYPES OF FURNACES.

Several types of furnaces in common use differ materially
from those illustrated here, and it may be of interest to mention
that one has an unusually large amount of heating surface se-
cured by surrounding the fire pot by a number of vertical
castings triangular in section through which the air passes ;
another has a revolving fire pot made up of vertical bars, scraping
the ashes off by rotating the fire around a fixed grate; still
another form has a tubular radiator at the rear through which
the gases pass. This construction permits the furnace to be
very low and allows a good pitch to the pipe. Square fire pot
furnaces are also used to a considerable extent.

FURNACES FOR OTHER FUELS.

Thus far we have discussed only furnaces for burning hard
coal. In certain districts, however, this fuel is so expensive, as
compared with soft coal, natural gas or wood, that furnaces de-
signed to burn such fuels are in demand. Furnaces for burning
soft coal are designed to admit a quantity of heated air above the
fire to combine with the gases, to diminish the waste of heat and
the escape of free carbon, as soot, in the smoke. With all the
precautions that may be taken the deposit of soot is much greater
than with hard coal, necessitating more frequent cleaning of the
furnace and smoke pipe. On account of the large volume of

Furnace Heating.

smoke the pipe is made I or 2 inches greater in diameter than for
hard coal furnaces of the same size. A cold air check should not
be used, as it increases the deposit of soot by cooling the smoke.

In the natural gas districts furnaces are commonly arranged
to burn this most convenient of fuels. Such furnaces should
have a grate for burning coal, in case the supply of gas should,
from any cause, be cut off.

Wood furnaces, Fig. 7, are generally very simple in construc-
tion, little attention being paid, as a rule, to their efficiency, since

Fig. 7. —Portable Wood Furnace with Steel Radiator.

the cost of fuel where they are used is generally very low. The
smoke should be made to pass through a radiator as in ordinary
hard coal furnaces. The larger sizes are built to take ordinary
cord wood sticks 4 feet long. Smaller furnaces may be had for
burning sticks 2 to 3 feet in length. The smoke pipe must be
made larger than for hard coal furnaces of the same heating
capacity.

Coke may be burned in ordinary hard coal furnaces, but this
fuel is very bulky for a given weight as compared with coal, and
must be fed more frequently to keep the fire in good condition.

Furnaces. 19

GAS FURNACES.

Furnaces specially designed for burning artificial and natural
gas have a gas burning chamber from which the pot products
of combustion are made to pass through a series of sheet metal
flues so as to expose a large amount of surface for heating air

Fig. 8. — Gas Burning Ring in Place.

and to insure the extraction of their principal heat. For Fall
and Spring service their use with artificial gas for a few hours per
day has been found practicable in relation to fuel cost. For con-
tinuous service natural gas at its low cost may be used with
economy and convenience. Hard coal furnaces are in many in-
stances arranged to burn natural gas by the insertion of a gas
burning ring as a part of the fire pot.

2o Furnace Heating.

Fig. 8 illustrates the application of a "gas burning ring" to
a coal furnace.

The maker's description is as follows : The Ring occupies a
position in the center of the fire pot and with it the furnace will
burn either coal or gas or both without any changes whatever and

Fig. 9. — Soft-coal Furnace with Air-blast Attachment.

without disconnecting the gas pipes. It is so made that the gas
outlets cannot become clogged with ashes.

SOFT COAL AND GAS FURNACES.

In the case of soft coal furnaces ample space must be pro-
vided in the combustion chamber.

The castings must be exceptionally heavy to withstand the
effect of the intense heat.

Furnaces.

Fresh air is in some types admitted around the fire pot just
above the level of the fire, this air being first heated.

Fig. 9 shows a soft coal furnace with an air-blast attachment.
The cut clearly shows the course of the air and the method of

Fig. 10.— Underfeed Furnace.

heating it. The hotter the air admitted to the combustion cham-
ber the more effective the air-blast attachment.

Fig. 10 shows an underfeed furnace used principally for soft
coal. When used for hard coal the makers recommend that pea,
buckwheat or chestnut sizes be used.

The operation of the furnace is described as follows in the
literature published by the manufacturer:

By means of a plunger which slides in this coal-chute, and a

22 Furnace Heating.

light hickory lever, which operates the plunger, coal, which has
been placed in the hopper, is "pumped," or forced through the
chute, up onto the grate and underneath the body of burning
coal.

In forcing the fresh coal into the furnace the fire is pushed
upward and outward, the fresh coal being surrounded on the top
and sides by fire. In this way the fire is brought into direct con-
tact with the sides of the fire pot and dome — the most effective
radiating surfaces of the furnace.

The combustion is more rapid along the sides of the fire pot,
because of the air admitted through the grate, and as the com-
bustibles are entirely burned out of the coal, the refuse-ash is on
grate, which encircles the feed-chute, and is readily and easily
shaken down into the ash-pit.

HEATING SURFACE.

Taking up again the discussion of hard coal furnaces we come
to the question of heating surface. Many furnaces having ample
grate area for the work intended, fall short from lack of heating
surface. In cold weather such furnaces have to be forced, caus-
ing red hot surfaces, intensely heated air and lessened efficiency.
Surfaces unlike in character and location vary greatly in heating
power, therefore the kind, form and location of the heating sur-
face, as well as its area, must be considered in comparing fur-
naces. It is by no means certain that of several furnaces having
the same grate area the one having the greatest heating surface
will be the most economical heater. In some furnaces having an
unusually large amount of surface it will be found on inspection
that a considerable portion would soon become almost useless
from the accumulation of soot. In others a large portion of the
surface is lined with fire brick, or is so situated that the air cur-
rents are not likely to strike it.

Heating surfaces may be classified as follows :

1. Fire pot surface, lined or unlined.

2. Surfaces acted upon by the direct rays of heat from the
fire, such as the dome or combustion chamber.

3. Gas or smoke heated surfaces, such as flues or radiators.

4. Extended surfaces, such as pins or ribs.

Furnaces.

B«§*

s & «

co oo oo

t-^TtHT-i

iO <N CO T-I

r* in jo 10

c» e* <* oo

"'S^ESS

£, ^ @ « TH CO

^TO^S§
r*< <N l^ 00

gj eg S

S%~3

351

? =3 a

w -H OO<»<N j, . -e

E S O «' 'tf •* 22 TH fl'NS
pq c< ift<?i«T-i -^cdT)

3 g

* I

1 1
•fe-

3°S«3S

^ CJ (^ (N

^ 1

H « IH jo oo os

iO to CO W

t^i "2

g S ^03 «

ga-^s-ss|

S ;II s

lilii

41!!1
a f I is I
i*iH1

•s e- 1 * s »

•d g

> o <» 05 'S

O H ffi W M

24 Furnace Heating.

Their relative value is an interesting question, on which avail-
able data are lacking.

The total heating surface, as compared with that of the grate,
based on actual measurements of a number of furnaces of dif-
ferent makes sold in New England, is shown in Table I.

Various writers on heating recommend furnace proportions
ranging from about 50 to 70 square feet of heating surface per
square foot of grate. These proportions are much in excess of
those found in ordinary house heating furnaces, as shown in
Table I. Assuming a maximum rate of combustion of 5 pounds
of coal per square foot of grate surface per hour the above men-
tioned ratios give 10 to 14 square feet of heating surface per
pound of coal burned per hour.

Common furnace proportions would give about 10 square
feet of heating surface per pound of coal burned per hour at the
average rate throughout the heating season. By using larger
furnaces than customary to heat a given space the same ratio
may be obtained during cold winter weather, since by increasing
the size of the furnace the rate of combustion is diminished and
the heating surface per pound of coal burned increased.

In any line of furnaces of the same make and style it will be
found that the heating surface per square foot of grate is less in
the large sizes than in the smaller ones. For example, take two
furnaces, one with a 2oinch fire pot and the other with a 3o-inch
pot, both i foot deep.

The 2O-inch pot contains 2.4 square feet of heating surface
per square foot of grate surface.

The 3O-inch pot contains i .6 square feet of heating surface per
square foot of grate surface.

An advantage in the ratio of 3 to 2 in favor of the smaller
fire pot. About the same ratio will hold for the total heating
surface in the furnaces.

The great advantage in point of heating surface in small fur-
naces, as compared with larger ones, explains their greater pro-
portional heating capacity.

SECONDARY HEATING SURFACE.

In addition to the heating surface stated in Table I, the inner
casing of black iron forms a valuable secondary heating surface,

Furnaces. 25

absorbing the heat radiated from the body of the furnace and im-
parting it again by convection to the air passing over it. This
secondary heating surface is very important. Since the air pass-
ing through the furnace is heated only by convection — i. e., by
bringing it in contact with a heated surface, unless the radiant
heat from the furnace proper is absorbed by some secondary
surface, which in turn imparts it to the air, much of the heat
radiated from the body of the furnace will be wasted in over-
heating the cellar.

RADIATION AND CONVECTION.

With highly heated surfaces the loss of heat by radiation is
greatly in excess of that by convection.

Sir Wm. Thomson is credited with the statement that a stove
heated to 1200 odd degrees gives off 92 per cent, of its heat by
radiation and 8 per cent, by convection.

The formulas of Dulong show that with heated body at tem-
perature of 780 degrees and surrounding air and objects 60 de-
grees, loss of heat by radiation, as compared with that by convec-
tion, will be as 7.17 is to 2.23, and with temperature of 960 degrees
and surrounding air and objects 60 degrees, loss of heat by radia-
tion, as compared with that by convection, will be as 12.68 is to
2.348. The higher the temperature of the heated surface the
greater will be the loss of heat by radiation as compared with that
by convection.

HEATING SURFACES OF FURNACES AND BOILERS.

It may be of interest to compare the proportions given in Table
I with those in hot water heaters and steam boilers. In such ap-
paratus designed for house heating the amount of heating surface
per square foot of grate generally ranges from about 15 to i in
the smaller sizes to 25 to i in the larger ones.

EFFICIENCY.

One of the first items to be determined in estimating the heat-
ing capacity of a furnace is its efficiency, or the percentage of the
heat in the coal that may be utilized. The efficiency depends
chiefly on the area of the heating surface as compared with the
grate, on its character and arrangement and on the rate of com-
bustion. The proportions commonly found in furnaces of differ-

26 Furnace Heating.

ent types are shown in Table I. The rate of combustion required
to maintain a temperature of 70 degrees in the house varies, of
course, with the outside temperature. Taken for the entire sea-
son the rate is generally less than 2 pounds of coal per square
foot of grate per hour. In severe weather, however, a rate of 4 to
5 pounds per hour must be maintained. In tapered fire pots the
grate surface should be considered equivalent to the average area
of the pot.

It is apparent that the efficiency of a furnace decreases with an
increase in the rate of combustion to the point of forcing since the
more rapid the rate the less will be the amount of heating surface
per pound of coal burned, and the hotter will be the gases passing
to the chimney. On the other hand, a very slow fire is wasteful,
due to incomplete combustion resulting from insufficient air sup-
ply. In the absence of definite available data based on tests, it is
necessary in making calculations of the heating capacity to assume
an efficiency that may reasonably be expected in practice. One
pound of good anthracite coal allowing 10% ash will give off
about 13,000 heat units. Of this amount a furnace should
utilize from 50 to 70%, according to conditions.

A heat unit may be defined with sufficient accuracy for the
purposes of this work, as the amount of heat required to raise
the temperature of i pound of water i degree Fahrenheit.

The writer has assumed in the following calculations that
8000 heat units may be utilized per pound of coal burned at a
maximum rate of 5 pounds per square foot of grate per hour.
This allowance corresponds to an efficiency of about 60 per cent.

HEATING CAPACITY.

The heating capacity of a furnace is generally stated in terms
of the cubic space it is capable of warming. This measure is used
from custom, but since its relation to the exposure varies with the
size and shape of the building, it is more accurate to base the
capacity directly on the exposed wall surface.

The variation in the relation between the exposure and the
cubic space may be readily shown. For example, suppose we
have a house of plan shown in diagram A and another of the
same cubic contents shown in B :

Furnaces.

The relative exposure of A to B is as 160 to 200 = 4:5.
That is, while the cubic contents is the same in each the exposure
of B is 25 per cent, greater than that of A. The fact that the
exposure is used by many of the best engineers in calculating
the proportions of steam and hot water heating apparatus should
be a sufficient guarantee of its fitness. To determine the size of
the furnace required for a given exposure the latter should first
be reduced to equivalent glass surface (E. G. S.). To do this we
must know the heat transmitting power of walls of different
kinds and thickness, as compared with that of glass.

It is convenient and sufficiently accurate for ordinary calcula-
tions to consider i square foot of glass equivalent to 4 square feet
of well constructed wood and plaster, or brick walls. Hence to
reduce the area of the solid walls to E. G. S. divide by 4. Add

20'

40
EXPOSURE 160'

EXPOSURE 200'

Fig. 11. — Relative Exposures.

to this the glass surface in the windows and one-half the area
of outside doors. The sum is the total E. G. S. of the outside
exposure. Since i square foot of glass will transmit about 85
heat units per hour when the difference between the inside and
outside temperature is 70 degrees (A. R. Wolff), to ascertain the
total loss of heat by transmission multiply the E. G. S. by 85.
As to allowances for houses in exposed locations see note below
Table II. To this must be added the loss of heat by ventilation
or change of air.

If the air enters through the register at 140 degrees, which
may be considered a maximum temperature under zero con-
ditions, it is plain that one-half the heat supplied is carried away
by the air escaping at 70 degrees, the other half (neglecting
floors and ceilings) being lost through walls and windows. There-
fore, twice the amount of heat lost by transmission must be sup-
plied by the furnace.

In these computations it has been assumed that the factor 85

28 Furnace Heating.

is large enough to cover air leakage losses, since other authori-
ties use 70 B. t. u. per square foot of glass per hour with 70°
difference in temperature. Wolff originally used this allowance
but increased it to 85.

The leakage loss is really more affected by the character and
extent of the exposed surface of a room than by its cubic con-
tents, although the latter is commonly used as a basis for com-
puting the loss of heat by leakage, allowing an air change once
an hour for example.

Assuming that with a rate of combustion of 5 pounds of
coal per hour per square foot of grate surface 8000 heat units
are utilized per pound of coal burned in a well proportioned
house heating furnace, (grate surface being considered equiva-
lent to average fire pot area in the case of tapering pots), we
have 8000 X 5 = 40,000 heat units per hour per square foot
of grate surface transmitted to the air passing through the fur-
nace. Dividing the total loss of heat per hour (E. G. S. X 85
X 2) by 40,000 gives the required grate surface in square feet,
from which the diameter of the fire pot in inches may be readily
determined. Expressed as an equation this becomes

E. G. S. X 85 X 2

- = grate surface in square feet (a).

40,000

Now, reversing this process and assuming different grate
areas, we may compute a table showing the heating capacity of
furnaces expressed in the area of exposed wall to which they are
adapted. The glass surface, as compared with the total exposure,
may vary considerably in different houses, but from the inspec-
tion of a number of plans the writer has adopted, as a fair average
for those with windows of generous size, a glass surface equiva-
lent to one-sixth the total exposure of glass and walls combined.
Outside doors are reckoned as equivalent to one-half their area in
glass.

With a glass surface equal to one-sixth the total exposure,
and with solid walls equal to one-fourth their area in glass in their
power for transmitting heat, we have

E. G. S. of house = -j - _H (- x 7 ) [• exposure.
(6 \4 6' )

— °-375 exposure of glass and wall combined.

Furnaces. 29

Substituting in equation (a) this value of E. G. S. we have

0.375 exposure X 85 X 2

- = grate surface in square feet. Or
40,000

transposing: Total exposure = grate surface in square

feet X - ' ge v — G. S. X 627.4, from which equation

u-o/5 A 05 A 2

Table II is derived.

Table II.*— The Capacity of Furnaces Expressed in Terms of the Exposed Wall
Surface to Which They Are Adapted, to Maintain an Inside Temperature
of 70 Degrees with an Outside Temperature of 0 Degrees. Temperature of
Entering Airy 140 Degrees Rate of Combustion, 5 Pounds Coal per Square

Foot of Grate Surf ace per Hour.

Total exposure in square

Average diameter of fire Corresponding area feet to which furnace

pot in inches. in square feet. is adapted.

18 1.77 1,110

20 2 18 1,370

22 2 64 1,655

24 3.14 1,970

26 3 69 2,310

28 4.27 2,680

30 4.91 3,080

32 5.58 3,500

In exposed locations add from 10 to 15 per cent., according to the conditions, to the
actual exposure of the house and select a furnace with a rating corresponding most nearly
to the corrected exposure.

* In this table no allowance has been made for the higher efficiency of the smaller
sizes, due to their greater ratio of heating surface to grate surface. It has been assumed
t iat this advantage is to a great extent offset by the more rapid combustion common in
large furnaces and by the better care they generally receive.

Note in connection with Table II, in calculating the gross
exposure, to measure the entire distance around the house ; mul-
tiply this by the combined clear hights of the several floors to
be heated. The product will be the total exposure in square
feet. The kitchen walls are included, simply to serve as a rough
allowance for the loss of heat through floors and ceilings, which
if estimated separately would make the calculation less simple.
Where but a single room on a floor is to be heated, as for
example an attic chamber, add its exposed wall surface, making
proper allowances for any adjacent unheated space.

SIZE OF FURNACES FOR BLOCKS.

In estimating the size of furnaces for double houses, flats, or
houses in blocks, it should be borne in mind that in case an

3° Furnace Heating.

adjoining house is unoccupied the loss of heat will be con-
siderably increased. It is well, therefore, to provide for such a
contingency by adding to the actual exposure of the house one-
third the area of the party wall or one-third of the floor area, as
the case may be. In city houses, which may stand apart from
others for some time before the adjoining lots are built upon, the
loss of heat through the party walls must be taken into considera-
tion in estimating the size of the furnace. A solid brick wall of
this nature will transmit about two-thirds as much heat as an ordi-
nary wall having an average amount of glass. Hence add to the
area of front and rear walls two-thirds the area of the party walls.
Select a furnace having a rating in Table II most nearly corre-
sponding to the total exposure thus obtained.

MANUFACTURERS' RATINGS.

It may be of interest to note the rated capacity of furnaces as
stated in manufacturers' catalogues, the capacity being expressed
in terms of the cubic space in frame dwellings the furnaces are
rated to heat. Table III gives a fair average of the minimum
ratings of furnaces of the 'better class. Column d shows the
exposure corresponding to a given cubic space, assuming the
house to be square and the clear hights of the first and second
floors to be 9 feet and 8 feet 6 inches respectively. These expos-
ures are considerably in excess of those in Table II, indicating a
tendency on the part of manufacturers to overrate their furnaces :

Table III.

Diameter of Rated capacity in Exposed wall

fire pot in Area of fire pot in cubic feet for surface

inches. square fee frame dwelling. corresponding.

(a.) •;&.) (c) (d.}

18 1.8 8,000 1,500

20 22 10,000 1,670

22 2.6 14,000 1,980

24 31 19,000 2,300

20 37 26,000 "2,700

28 4.3 33,000 3,040

30 49 40,000 3,340

33 5.6 50,000 3,740

CHAPTER II.
HOUSE HEATING.

COMPARATIVE MERITS OF FURNACES AND OTHER SYSTEMS.

In first cost furnace heating is less expensive than steam
or hot water heating. The amount of fuel required is greater
than with either of the latter when direct radiation is used. Indi-
rect steam or hot water systems deliver air at a lower tempera-
ture, as a rule, than furnaces and consume more fuel.

As to the objections raised against furnaces, it may be said
that when installed in accordance with the building laws of most
cities the risk of fire is practically eliminated. The leakage of
gas and dust is more frequently due to faulty installation and
management than to any defect in the furnace. The gas and
dust are allowed to escape to the cellar, whence they are drawn
up into the rooms through the cracks in the cold air box, the
joints in the casing or the spaces around the pipes. For this
reason the cold air box should be carefully constructed.

With modern methods of proportioning the size of furnace,
pipes and cold air box, and with the more general use of fire places
and wide openings between rooms, good results are obtained.
The force causing air to flow through the pipes is, at best, slight.
They must, therefore, be carefully proportioned and the furnace
suitably located to secure a proper distribution under adverse
conditions. For warming country or seashore houses occupied
only part of the year, furnaces are particularly convenient. They
are always ready for use, and at the end of the season may be
left without precautions being taken against damage, as with hot
water or steam apparatus. Where wood is cheap excellent
results may be obtained with furnaces designed to burn that fuel.

LOCATION OF THE FURNACE.

A furnace should be so placed that the warm air pipes will be
of nearly the same length.

The air travels most readily through pipes leading toward the

32 Furnace Heating.

sheltered side of the house and to upper rooms. Hence pipes lead-
ing toward the north or west, or to rooms on the first floor, should
be given a preference in respect to length and size. The furnace
should be placed somewhat to the north or west of the center of
the house, or toward the points of the compass from which the
prevailing cold winds blow. See Figs. 29, 30, and 31.

FOUNDATION.

Having determined the location of the furnace, see that a
suitable foundation is provided of concrete or of brick. Excavate

Fig. 12. — Brick Furnace Foundation with Underground Cold Air Box.

and place the furnace in a pit, if necessary to obtain a proper
pitch to the pipes.

FURNACE PIT.

If a pit under the furnace is to be used, because of the better
distribution of the air around the furnace, care must be taken
to see that it is properly drained. All underground work should
be built of hard burnt brick laid in cement having two parts
of sand to one of cement. The thickness of the walls of the pit
may be 4 or 8 inches, according to its diameter and depth. A
large pier on which the furnace will rest should be built in the
center of the pit. The size of this pier will vary with the size

House Heating.

of the furnace. One 16 x 16 inches is common. The pier should
be set diagonally with reference to the opening from the cold air
box, to divide the current of entering air.

BRICK SETTING.

Having prepared the foundation and pier, set the bottom cast-
ing carefully in place, so that its center will coincide with that of
the foundation. Continue erecting the castings, packing the
joints with kaolin or other suitable material. This done, bolt the
front or shield firmly in place. Pack the joints around the door
frames with suitable cement or putty to prevent the leakage of
gas or dust. The inner and outer brick walls, each 4 inches in
thickness, with not less than 2 inches clear space between them,

Fig. 13. — Section on Line A A of Fig. 12.

may now be carried up (see Fig. 13)^ keeping the courses level.
Place irons over the openings for the cold air box, man-door and
front or shield.

The inside diameter of the circular wall is generally made 4
to 8 inches greater than the diameter of the radiator. The air
passage through the furnace should be equivalent to the combined
area of the warm air pipes. Light, hard bricks may be used for
the setting, to be well bedded in cement mortar consisting of not
less than one part of cement to three of lime mortar. The inner
circular wall should have a thin coating of cement applied.

When the walls have reached the proper hight, set a thimble
about 3 inches larger than the diameter of the smoke pipe, and
place the hot air pipes in position with their tops level. Give
them as sharp a pitch as possible. Build in carefully around them
and trim off their inner ends to conform to the circlar wall ; then

Furnace Heating.

lay on covering bars about 8 inches " on centers," with strips of
tin or galvanized iron between. Lay on these one course of
bricks dry, then another course in cement mortar, and plaster the
top. Another method of covering is to use two sets of bars with
one course of bricks on each, leaving a dead air space between
them.

Fig. 14. — Brick Set Furnace.

A better and more expensive method than either of those just
described is to use an inverted cone built of 22 or 24 inch galva-
nized iron (see Fig. 15). Rest this on the inner wall and lay one
or two courses of brick above on iron bars. The air chamber above
the top of the furnace castings should be at least 10 inches high,
to permit a free distribution of the air to the various pipes.

In some furnaces the cold air enters beneath cast iron trench
plates. In others they are omitted, the air entering through a

House Heating. 35

series of " pigeon holes " extending around the bottom of the
inner wall. An even distribution is thus secured.

PORTABLE SETTING.

The setting of a portable furnace is generally a very simple
matter. After the bottom casting has been properly set on a suit-
able foundation the other sections are placed in position, allowing
each to find its own bearing in the cup joints filled with kaolin or
other suitable material. This done the front or shield is bolted on
with joints cemented. The inner and outer casings are next
adjusted, then the collars are set in the top, and the furnace is
ready for the pipes. Building laws sometimes require a sheet
iron shield to be suspended from the ceiling above the furnace.

PORTABLE VS. BRICK SETTING.

Portable furnaces with galvanized iron casings have almost
entirely superseded those set in brick work. When properly

Fig. 15.— Section through Cone for Brick Set Furnace.

arranged with a double casing the loss of heat is no greater than
is necessary to keep the cellar of a country house at a proper
temperature. They occupy less space and are more accessible
in case of repairs than those set in brick. In city houses, with
the basement well protected, the loss of heat from* the furnace
easing is objectionable, and since the transmission of heat is less
with a brick setting the latter is sometimes used, although a metal
casing covered with non-conducting material would perhaps be
better. A brick setting has another advantage in its ability to
store heat, acting as a sort of temperature equalizer, absorbing
heat when the fire is intense and giving it out again when it
becomes low. Cracks in the circular wall are liable to occur,
however, which with certain forms of setting produce harnJul
effects.

TWIN FURNACES.

In large houses it is often a question whether to use one large
furnace or two smaller ones having a single top, known as twin

36 Furnace Heating.

furnaces (see Fig. 16). These will be somewhat more expensive,
but have certain advantages.

It has been pointed out in Chapter I that small furnaces have
more heating surface per square foot of grate than larger ones,
hence two small furnaces will present more heating surface than
one large one having their combined grate area. With furnaces of
the same make and type the greater the ratio of heating surface
to grate area the greater the efficiency.

Fig. 16. — Twin Furnaces or Battery System.

An advantage claimed for twin furnaces is the greater range
in heating capacity obtained by using one or both fires to suit
conditions. A single furnace of sufficient capacity to warm a
house in the coldest weather will require considerable skill in its
management to avoid overheating at other times. It is from lack
of such skill that houses are usually too warm during spring and
fall. In twin furnaces it is especially important that the air
chamber above them be roomy, to permit the easy flow of air to

House Heating. 37

the pipes. It can hardly be expected that the distribution of air
throughout the house will be as even in mild weather, when run-
ning but a single fire, as when both are in use, since the space for
the passage of air through the single furnace is not large enough
to admit a sufficient volume to fill all the pipes.

It is hardly necessary, however, that the distribution should be
perfect in mild weather, for with open doors the warm air ad-
mitted to the living rooms will soon become diffused throughout
the house. The pipes to the more important rooms should, if
possible, be connected with the hood above the same furnace, so
that the air will flow directly to them when running a single fire.
It is often advisable to combine two furnaces of different sizes
under the same top. The larger one can then be used until late
in the fall, when the smaller one can be added as an auxiliary.

The cold air box should separate in two branches before reach-
ing twin furnaces, each to have a slide or damper.

TWIN FURNACES VS. SEPARATE ONES.

It is often a question whether to use twin furnaces or two
furnaces placed separately. If it is found on laying out the sys-
tem with twin furnaces that some of the pipes must be made of
excessive length, it would be far better to discard this system and
use two separate furnaces placed some distance apart. These
will be somewhat less convenient to care for than twin furnaces,
but the shorter pipes will insure a more even distribution of warm
air. Unless the cellar is unusually high, permitting a sharp rise,
the length of the pipes should not greatly exceed 15 feet.

SMOKE PIPES.

Furnace smoke pipes range in size from about 6 inches in the
smaller sizes to 8 or 9 inches in the larger ones. They are gen-
erally made of galvanized iron of No. 24 gauge or heavier. The
pipe should be carried to the chimney as directly as possible,
avoiding bends, which increase the resistance and diminish the
draft. When" the draft is known to be good the smoke pipe may
purposely be made longer to allow the gases to part with more of
their heat before reaching the chimney. Where a smoke pipe
passes through a partition it should be protected by soapstone of

38 Furnace Heating.

the thickness of the partition and extending not less than 4 inches
from the pipe in all directions. A double perforated metal collar
may be used instead of the soapstone if desired, making it at least
8 inches greater in diameter than the pipe.

The top of the smoke pipe should not be placed within 8
inches of unprotected beams nor less than 6 inches under beams
protected by asbestos or plaster, with a metal shield beneath. The
connection between the smoke pipe and the chimney is frequently
very loose, allowing cold air to be drawn in, thus diminishing the
draft. A collar to make the connection tight should be riveted
to the pipe about 5 inches from the end, to prevent its being pushed
too far into the flue.

Where the pipe is of unusual length it is well to cover it to
prevent loss of heat and the condensation of gases.

CHIMNEY FLUES.

Chimney flues, if built of brick, should have walls 8 inches in
thickness, unless terra cotta linings are used, when only 4 inches
of brick work is required. Except in small cottage houses, where
an 8 x 8 flue may be used, the nominal size of the smoke flue
should be at least 8 x 12, to allow a margin for possible contrac-
tions at offsets, for undersized brick or for a thick coating of
plaster which is not necessary but which nevertheless is some-
times applied. A clean out door should be placed at the bottom.
A square flue cannot be reckoned at its full area, as the corners
are of little value. An 8x8 flue is practically no more effective
than one of circular form 8 inches in diameter. To avoid down
drafts the top of the chimney should be carried above the highest
point of the roof.

AREA OF COLD AIR BOX.

The cold air box should be large enough to supply a volume
of air sufficient to fill all the hot air pipes at the same time.

If the supply is inadequate the distribution is sure to be un-
equal, the cellar will become overheated from lack of air to carry
away the heat generated and the life of the furnace will be short-
ened. These points in many cases are not appreciated, if one may
judge by the absurdly small cold air boxes frequently used. If a

House Heating. 39

box is made so small or is throttled down so that the volume of
air entering the furnace is not large enough to fill all the pipes, it
will be found that those leading to the lee side of the house or to
rooms on upper floors will take the entire supply, and that addi-
tional air to supply the deficiency will be drawn down through
registers in rooms less favorably situated.

Common "thumb rules" are to make the cold air box two-thirds
or three-quarters the combined area of the hot air pipes. The area
of the box is governed by the capacity of the pipes and the expan-
sion of the air. In zero weather the maximum temperature of the
air leaving the furnace in a well proportioned system should not
exceed 140 degrees.

Each cubic foot of air admitted at o degree when heated to the
latter temperature is expanded to 1.325 cubic feet, or is increased
in volume about one-third; hence the cold air box need be only
three-quarters the area of the hot air pipes to fill them under the
conditions stated. A box the full area of the pipes would insure
an ample supply of air at all seasons, and its effective area could
be almost as easily regulated by the slide or damper as a smaller
one. Such a box would be of great assistance to avoid overheat-
ing in mild weather.

LOCATION OF COLD AIR BOX.

The cold air inlet should be placed where the prevailing cold
winds will blow into it, commonly on the north or west side of the
house. When the inlet is on the lee side, warm air from the fur-
nace is likely to be sucked out through the cold air box. Avoid
taking the air supply from narrow passageways between houses
from fear of the same action during strong winds. Whatever
may be the location of the entrance to the cold air box, reversals in
the direction of the air current therein may take place in the case
of very high winds blowing from a direction that brings the
entrance on the lee side of the house. The flow of air in the
proper direction may be re-established by closing the slide in the
cold air box and taking air temporarily from the cellar. A well
designed hot air system may oftentimes turn out to be a failure
due to improper location of the cold air inlet. For this reason

4o Furnace Heating.

the existing conditions should be carefully studied before the
location is fixed.

MATERIAL OF COLD AIR BOX.

The cold air box is generally built of matched boards. How-
ever well such a box may be put together, the wood soon shrinks
and joints open, allowing dust and cellar air to be drawn into the
furnace and discharged to the rooms. The wood work should be
kept at least I foot from the furnace and protected from radiant
heat. The connection between the wooden cold air box and the
furnace should be of galvanized iron or brick.

Galvanized iron is probably the best material to use. It may
be made practically air and dust tight, is fire proof, durable and in
harmony with the other parts of the apparatus. A cold air box of
this material costs more than a wooden one, but for first-class
work is worth the additional expense. In case the galvanized
iron cold air box is of considerable length, or passes through a
kitchen or laundry, it should be covered with non-conducting ma-
terial.

For an underground cold air box, hard burnt brick laid in ce-
ment should be the materials used. The bottom should be of brick
or concrete. The top may be covered with bluestone with close
joints or with bricks laid between covering bars and concreted
over the top flush with the cellar floor. (See Fig. 13.) Glazed
drain tile is often used for cold air ducts, especially in connection
with small furnaces. It is not advisable to use an underground
duct or pit where the ground is damp, for even if drained the walls
of the duct are apt to become mouldy and tke air to be unwhole-
some. If such conditions are encountered the duct should be
carefully protected with waterproofing.

COLD AIR ROOM.

A small room into which the air flows before entering the duct
leading to the furnace is sometimes provided. It acts as an equal-
izing chamber and overcomes to a great extent the effect of sudden
gusts of wind, making the flow of air through the pipes much
more uniform than with an ordinary cold air box. With this ar-
rangement less attention is required in regulating the slide. A
cold air room 6x6 feet in size is ample for a good sized house.

House Heating.

When the wind is likely to blow with unusual force baffle
plates may be used, as in Fig. 17.

Fig. 17. — Fresh Air Room with Dust Collector.
COLD AIR INLET.

The galvanized iron wire netting at the cold air inlet should be
at least 2/a-inch mesh ; a finer netting is unnecessary and cuts off
too much area. The frame to which it is attached should be no
smaller than the inside dimensions of the cold air box. The air is
frequently much restricted at this point. A door to admit air from
the cellar to the cold air box is generally provided. As a rule air
should be taken from this source only during high winds, ex-
tremely cold weather or when the house is temporarily unoccupied.

AIR FILTERS.

When the air supply is likely to be laden with dust, filtering
screens of cheese cloth or similar materials may be used. They
may be made in the shape of conical bags suspended in the cold
air room, or the cheese cloth may be attached to frames arranged
for convenient removal. The area of the screen should be at least
15 times the area of the cold air box. At best they are a bother
and must be frequently cleaned, but when properly arranged they
afford considerable relief from dust contained in the outside air.

42 Furnace Heating.

RETURN DUCT AND AIR SUPPLY.

In some cases it is advisable to return air to the furnace to be
reheated. Ducts for this purpose are common in places where the
winter temperature is frequently below zero. Return ducts, when
used, should be in addition to the regular cold air box. The great
amount of air supplied to a house by a well arranged furnace may
not be generally realized. Take, for example, an ordinary frame
house with seven or eight registers of average size. From data
at hand the air supply to such a house in winter weather is 800 or
900 cubic feet per minute, corresponding to a change of air about
once in 15 minutes. With the ordinary number of occupants this
volume gives so large a per capita air supply that during the day-
time a portion of the air may be returned to the furnace without
harm. In the evening, however, with gas burning, each jet vitiat-
ing the air as much as five or six persons, the air supply stated is

Fig. 18. — Return Air Duct with Damper and Back Draft Checks.

not any too much. It is apparent, therefore, that to obtain proper
ventilation at all times with a system arranged to take air from
either indoors or out, intelligent management is necessary. Be-
cause such management cannot be assured the return duct is not to

House Heating. 43

be generally recommended except where the climate is very
severe, when the best results are sought. Its use certainly re-
duces the amount of fuel required, but the tendency to econo-
mize too much by taking all the air from indoors is a serious
objection.

There are several methods of arranging a return duct. A sep-
arate connection may be made with the base of the furnace, or a
branch may be run from the main cold air box to the first floor,
with a mixing damper at the junction of the two arranged to shut
off a portion of the outside supply while admitting air from the
rooms above.

With the arrangement shown in Fig. 18 a portion of the air
supply for the furnace may be taken from out of doors, the
remainder being drawn through the register from the house.
The gossamer check dampers are essential to the proper working
of this device, for without them back drafts would be likely
to occur during winds.

RECIRCULATED AIR.

With the increase in the price of coal, greater attention is
paid than formerly to the recirculation of air. A house heating
furnace as commonly arranged supplies far more air than is
necessary for ventilation ; for example, suppose a furnace with
a 24-inch grate (average fire pot area) supplies I square inch
pipe area to each square inch grate surface, 24-inch diameter=
452 square inches=3.i4 square feet. With air velocities in pipes
ranging from say 250-350 feet per minute to first floor, 350-450
to second floor and 450-550 to third floor, or say 350 as average
velocity to first and second floors, no heat on third, gives 3.14
square feet X 35° = 1109 cubic feet of air per minute supplied
to first and second floor rooms combined. On the usual basis
of 30 cubic feet of air per minute supplied to each person, this
volume would provide for 37 persons, whereas there would be
hardly ten persons in a house heated by this size furnace.

During the night when sleeping room windows are supposed
to be open, as is the rule nowadays, there appears to be no valid
objection to returning the air from the house, shutting off the
outside air and saving fuel.

44 Furnace Heating.

During the day time in severe weather there is no objection
to supplying at least one-half the air supply to the furnace
through the recirculating duct from the house, which consider-
ably reduces the coal consumption.

It is difficult to arrange this so that air can be taken from both
outdoors and indoors at the same time and have no back drafts
of cold air through the recirculating register. Fig. 18 shows one
method by which this may be accomplished.

To avoid overheating the furnace in case nearly all registers
and the cold air box happen to be closed at the same time, it is
well to omit the pipe damper and the register blades or shutters
from one of the rooms, say the first floor hall,

SIZE OF AIR PIPES.

Much larger furnace pipes are now used than formerly. This
involves a greater original outlay and an increased running ex-
pense for fuel, but the householder is repaid by the more healthful
conditions secured through the supply of an ample volume of
warm air in place of a small volume of intensely heated air. The
pipes should be so proportioned that the several floors will be
heated equally.

Table IV, calculated as explained below, will be found useful
in determining their size. It must be borne in mind, however, that
in heating and ventilating work no rule or table can be successfully
used without a certain " coefficient of common sense " to allow for
varying conditions.

The main steps involved in the calculation of the table are :

1. The determination of the loss of heat through the walls,
windows and floor or ceiling of the room.

2. The volume of warm air required to offset this loss.

3. The velocity of air in the pipes.

The loss of heat is calculated by first reducing the total ex-
posure to equivalent glass surface. This is done by adding to the
actual glass surface one-quarter the area of exposed wood and
plaster or brick walls and one-twentieth the area of floor or ceiling
to cover the loss of heat to non-heated basement or attic. At
least ten per cent, is added where the exposure is severe, to
cover the increased loss of heat by transmission and by the leak-
age of air. The window area assumed in calculating the table is

House Heating.

•

1-1

„

Hi &

•S §

•<*

11 5

•§•§•8 5

l^ ?

•*

,§ 8 '£

coo o

£ t» > c

2 * J co'

85 1

•M

II II II 1

§| 1^

•

•3

: 1

§ . 1 i

•«

g $^ 1*

,_,

: «

C? "So

: : 5

,_,

: *

e«

«M

: •S

| « 8 ^

,_,

| £ •§ *"»

C-'

: 1 g

g I j| f

: ^ |

5 1 P J

,_,

,-1

£ SS, § S1

1 2 1 3 |

;

-H

p-aft 5 <»

1-1

a c a ^ *"

w S I* 1 «

®*s T£ f
3 g ° '"s

3J
O

33
0

0
<M

ssji g s

<D <D « O O .

ooo i t|

i ** 8 ' c

33
0

^c.

*^ O rrt

nan* S &§

lop

.E.g-.g. ^ -S |

l-sl 1 Js

•2 °° E-s

.S.S.9 • -gi g"^ g

8 | jf|

CD
H

os O ^H p 53*^^3

£ t§ J»^

';^ 3 If §

S 0- ^ l»

linn g> jg-g

f: « * o

: : tS ^£^

J|*|

—

•

1 1 111
': B SIS

1 § • ^

—
H

oo
oo

OS
OC

G-.

30
Oa
-0

o-.

CO
OS

: '

i IJllf

* <n Q) C 0

^ i IliU

TV, Showing the
ture at Register 1
issumed to be 10
.re Longer Than

r
X

oo
oo

00
•^ •v"*'

cc
oc

;;

»1 IP!!
i!} I|j|

000 1

•o

•aiooj jo mpj.M.

46 Furnace Heating.

one-fifth, or 20 per cent., of the entire exposure of the room.
From the inspection of a number of plans this ratio was found
to represent a liberal allowance for glass surface.

Having obtained the equivalent glass surface (E. G. S.), mul-
tiply by 85 (the loss of heat per square foot of glass per hour
with 70 degrees difference in temperature) . The product will be
the total loss of heat by transmission per hour.

Double windows when tightly put in transmit about three-
fifths as much heat as a single window.

The volume of warm air required to offset this loss depends
on its temperature, which generally ranges from 120 to 140 de-
grees in zero weather. Assuming the temperature of the entering
air to be 140 degrees and that of the room to be 70 degrees, the air
escaping at approximately the latter temperature will carry away
one-half the heat brought in. The other half, corresponding to
the drop in temperature from 140 to 70 degrees, is lost by trans-
mission. With outside temperature zero, each cubic foot of air at
140 degrees brings into the room 2.2 heat units. Since only one-
half of this, or i.i heat units, can be utilized to offset the loss by
transmission, to ascertain the volume of air per hour at 140 degrees
required to heat a given room, divide the loss of heat by transmis-
sion by i.i ; the quotient is the volume sought. This result divided
by 60 gives the number of cubic feet per minute. Having deter-
mined the volume of air required per minute, if we know the veloc-
ity with which it will travel through the pipes, their area in square
feet is readily determined by dividing the volume by the velocity in
feet per minute. This area is easily reduced to square inches, from
which the diameter of the pipe may be obtained. The table
avoids the bother of working out separately the size of each
pipe.

To illustrate the method just stated, take for example a room
two sides exposed, 14 x 16 feet x 9 feet high, on first floor, loss
of heat through floor neglected, cellar being warmed by waste
heat from the furnace and pipes. Glass = 20% total exposure
(14 + J6) 9 = 20% of 270 square feet = 54 square feet, leaving
270 — 54 = 216 wall, which divided by 4=54 square feet E.
G. S. Adding this to the actual glass, 54 square feet gives total
E. G. S. of 108 square feet. 108 square feet X 85 (the loss of

House Heating. 47

heat per square foot of glass per hour with 70° temperature dif-
ference) = 9180 B. t. u.

Add say 10% to allow for exposure. Total = 10098 B. t. u.
per hour by transmission.

As stated above with o° outside, 70° inside and 140° tempera-
ture of entering air, each cubic foot of warm entering air brings
in i.i B. t. u. available to offset loss by transmission, therefore
10098 -^- i.i = 9180 cubic feet air per hour must be supplied =
153 cubic feet per minute. With a velocity of 280 feet per

minute through the pipe, -^ X 144 = 70 square inches = 10

280

diameter pipe the same as in space opposite 14 in left hand
column and under 16 in upper line of Table IV.

(Weight of cubic feet of air at 140 degrees), 0.066 pound X
(increase in temperature from zero), 140 degrees -f- (specific heat
of air), 0.238 = approximately 2.2 heat units.

The specific heat of a body is the quantity of heat required
to raise the temperature of the body through i degree F., as
compared with that required to raise the temperature of an equal
weight of water i degree. The specific heat of air (at constant
pressure) is 0.2377; that is, approximately one-fourth as much
heat is required to raise i pound of air through i degree F. as
would be necessary to raise the temperature of i pound of water
the same amount.

VELOCITY OF AIR IN PIPES.

In calculating the table maximum velocities of 280 and 400
feet were used for pipes leading to the first and second floors, re-
spectively. These velocities are readily attainable in practice.
They are lower than those commonly assumed for straight vertical
flues, but this is accounted for by the greater resistance to the pas-
sage of air through the nearly horizontal basement pipes, and
through elbows, nettings and registers. The size of the smaller
pipes was based on lower velocities, according to their size, to
allow for their greater resistance and loss of temperature.

LENGTH OF HOT AIR PIPES.

Since long horizontal runs of pipe increase the resistance and
loss of heat it is unwise to extend them much over 15 feet in

48 Furnace Heating.

length. This rule applies especially to pipes leading to rooms on
the first floor or to those on the cold side of the house. Air tends
to move with the wind, not against it, hence pipes leading to ex-
posed rooms should be favored. Rooms having a fire place or
ventilating flue are more easily warmed than others.

Pipes of excessive length should be increased in size to allow
for the additional resistance. The loss of heat from them may
be diminished by a covering of asbestos or other nonconducting
material or by making them double, leaving an air space of ^ to
I inch between the two.

Long pipes or those leading to exposed rooms are sometimes
favored by attaching them to the furnace top near the center,
where the air is hottest, and by placing at their extremity inside
the casing an inverted funnel or hood.

Fig. 18a. — Cone Top Furnace and Pipe with Fig. 19. — Flat Top Furnace and Pipe with
Bevel Elbows. Bevel Elbows.

The pipes should pitch upward as sharply as possible, for the
greater the angle the less the resistance.

METHODS OF PIPING.

Several methods of piping are illustrated in the cuts. An
inspection of each will show that less pipe is required in Figs. i8a
and 19 than in the others. More careful measurements are nec-
essary, however, as each turn requires a special bevel elbow, so
called, made to suit the angle at which the pipe is placed. This
angle is fixed by the hight of the cellar and the distance from the
furnace to the register or riser. Lack of head room in low cellars
with long runs of pipe sometimes interferes with the adoption of
this method.

Figs. 20 and 21 show a cone or pitch top and a flat top fur-

House Heating.

nace piped with regular stock pattern square and 45-degree
elbows. This is a simple and fairly direct method of piping
and presents a neat appearance. Somewhat more pipe is re-
quired than in Figs. 18 and 19, but this is offset by the con-
venience of using stock elbows. Slip joints provide sufficient
"come and go" to make up for slight errors in measurements or
in the making of the pipes.

Fig. 22 shows a method used chiefly in low cellars to secure
the maximum amount of head room. It is the most roundabout
method and increases the resistance to the flow of air and is not
recommended.

Pipes should be kept at least 4 inches from the edge of the flat
top. This is especially important in furnaces having a large

Fig. 20.— Co
Fqrnace ai
with Stock!

ne Top
id Pipe
Elbows.

[

Fig. 21.— Flat Top Fig. 22.— Flat Top Furnace
Furnace and Pipe and Pipe with Square
withStockElbows. Four Piece Elbows.

space between the body and the casing, in order to cause the
air to hug the heating surface.

Generally pipes may be placed near the center of a flat top
than of a cone or pitch top. A damper should be put in each
pipe near the furnace.

TRUNK LINE SYSTEMS.

The method of piping illustrated in Fig. 23 has been used
successfully in certain sections and possesses these advantages
over separate pipes: (o) The friction is reduced; (b) the
loss of heat from the pipes is reduced; (c) less sheet metal is
required; (d) the appearance of the job is improved.

In designing this system the trunk lines have been subdivided
as follows:

Furnace Heating.

One 14-inch pipe supplies one 8-inch and one 11-inch.
One 16-inch pipe supplies one 13-inch and one 9-inch.
One 13-inch pipe supplies two 9-inch.
One 11-inch pipe supplies two 8-inch.

W//W//7///M W//////WA V////7/M//M •m

Fig. 23. Trunk Line System.

The relative areas are

14" = 154 sq.in., 8"+ 11" diameter
16" = 201 sq.in., 9"+ 13" diameter
13" = 133 sq.in., two 9" = 128 sq.in.
11"= 95 sq.in., two 8" = 100 sq.in.

145 sq.in.
197 sq.in.

The relative approximate frictional surface for a! given
length is

14" = 44
16" = 50
13"= 41
11"= 35

8"+ 11"= 6
9"+ 13"= 6<
two 9" = 56.
two 8" = 50.

The above tabulation shows very clearly the great difference
in frictional surface in trunk lines and in a pair of smaller
pipes of approximately the same aggregate area.

House Heating. 51

RELATION BETWEEN GRATE SURFACE AND PIPE AREA.

Furnace catalogs often give ratings expressed in aggregate
pipe area to which the furnace is adapted.

These ratings commonly allow from i to ij4 square inches
pipe area to each square inch grate surface (average fire pot
area).

On a heat unit basis, if I square foot G. S. is good for I
square foot pipe area, and I square foot grate gives off 40,000
B. t. u. per hour that are utilized in heating the air. then since
i B. t. u. will heat 50 cubic feet of air from o° through i°,
40,000 B. t. u. will heat 40,000 X 50 = 2,000,000 cubic feet per
hour through i°, or if air is raised 140° will heat 2,000,000
-4- 140 = 14,300 cubic feet per hour through 140°. To dis-
charge this volume, equal to — L — or 238 cubic feet per minute,

measured at o° F. through an area of i square foot would require
a velocity of 311 feet per minute.

The volume 14,300 cubic feet at o° is expanded to
,- 460+0

= j its volume at o° when heated to 140° = 18,652 cubic feet.

460'

This volume per hour would pass through a pipe i square foot
in area at a velocity of 311 feet per minute.

This corresponds well with the velocity of air in pipes to first
floor rooms, hence theoretically the rule to allow i square inch
pipe area to each square inch grate area (average fire pot area)
is shown to be approximately correct.

RISERS OR VERTICAL FLUES.

Some architects appreciate the advantages of round risers
instead of the usual shallow oblong, rectangular or oval form,
and provide partitions of sufficient depth to permit them to be
run. When such risers are located near the furnace they may
best be made the same size as the cellar pipes connected with
them. When they are some distance away the horizontal pipes
are generally made larger than the uprights. When vertical pipes
must be placed in single partitions, an important economy in
fuel and a much better efficiency attends making the studding

Furnace Heating.

5 or 6 inches deep and a much better job can be done than
where the ordinary 2x4 studs are used. The shallower the
pipes the greater the loss of heat and the greater the friction.
For these reasons risers should never be carried up in par-
titions having a nominal thickness less than 4 inches. Studs

MOULDING
St-ATE .STONE

>< / ROUND /\

\/ I WE 0

Fig. 24. — Oval Riser with Convex Register.

2x4 inches will shrink to a depth of 3^ inches, leaving only 3^2
inches for the flue, allowing a trifle for clearance.

It is often difficult to provide spaces for flues of proper area
to run up in ordinary partitions with studs' 16 inches on centers.
To run a riser large enough to heat a good sized room on the
second floor the studs would have to be set so far apart that the
plastering between them would not be firm unless very stiff metal
lathing were used. To give space for large risers the partition
may be thickened by nailing on furring strips, or in some cases a
breast can be built of sufficient size to contain a round pipe of the
full area required.

House Heating. 53

SEPARATE RISERS.

Each room should be heated by a separate riser. In some
cases, however, it is permissible to run a single riser, connected
with a tee or header at the top, to heat two rooms on an upper
floor, if a riser of sufficient size is provided. The distribution
with this arrangement is likely to be unequal during winds, the
air going more freely to the less exposed room.

A single flue is sometimes used to heat two rooms on different
floors, but such an arrangement should be avoided, if good ser-
vice is desired in both rooms at all times. When used a damper
is generally placed just above the lower register.

LOCATION OF RISERS.

A clear space of l/2 to i inch should be left between the risers
and the studs. The latter should be carefully tinned and the
space between them on both sides of the pipe covered with tin,
asbestos or metal lath. In some of the best work the risers are
made double throughout with an air space of l/2 inch or so between
the inner and outer shells. In other cases they are wrapped with
heavy asbestos paper. Protection of this character, however,
should never interfere >-ith the main object of heating, which is
of great importance to the health and comfort of the family for
a large portion of the year. If necessary use deeper studs or
resort to any method that will insure heating as well as protec-
tion. It sometimes happens that the cellar pipes are carelessly
pushed so far into the foot of the risers that the area of the flue
is seriously diminished. This should be guarded against by
beading the pipe a proper distance from the end.

If it can be avoided oval pipes should not be placed in parti-
tions opposite sliding doors from fear of warping the latter.

At the level of the first floor the space around each riser should
be stopped off with tin to prevent dust and cold air being drawn
up from the cellar.

Risers should be placed in inside partitions if possible. Where
they must be run in outside walls they should be made of larger
size as well as double, and if the wall is not back plastered the
outside boarding should be lined with asbestos sheathing. An
air space should be left between the outside wall and the riser.

Furnace Heating.

£•<

Is.

SiS

OS » S * OS

t» u > >
,500000

^00000

Coooo>i-H<M

a ooooo
>o c c c c a

1 22222

P ee as os as «s

S3 333

^ crcr o< cr n4

00 fli 6 ^

| II II I! II II

ft SSScu*

House Heating. 55

In city houses most of the risers may be placed in recesses
in the brick party walls. Wherever possible chimney breasts
should be utilized for running risers, as they can often be built
to accommodate round pipes, which are always to be preferred.

MATERIAL OF PIPES.

Bright charcoal tin is almost universally used for hot air pipes,
except those of unusually large size, which are made of galvanized
iron. In a dry atmosphere the tin retains its brilliancy and will
radiate less heat from its surface than any other suitable material.
In the best work, pipes 1 1 inches in diameter and smaller should
be made of IX tin. Those of 12 inches in diameter and larger
should be made of IXX tin or galvanized iron.

AREA AND SIZE OF REGISTERS.

The registers which control the supply of warm air to the
rooms, generally have a net area equal to one-half their gross
area. The net area should be 10 to 25 per cent, in excess of that
of the pipe connected with it. It is common practice to use reg-
isters having the short dimension equal to and the long dimension
about 50 per cent, greater than the diameter of the pipe. Thus,
for a 6-inch pipe use a 6 x 10 register ; for a 7-inch pipe, a 7 x 10
register; 8-inch, 8 x 12; 9-inch, 9 x 14; lo-inch, 10 x 14; 12-inch,
12 x 17; and so on.

Floor timbers are usually spaced 16 inches on centers, leaving
about 14 inches clear space between them. Registers as large as
10 x 14 inches may be set with either dimension perpendicular to
the timbers without the use of headers. The timbers may be
trimmed slightly if necessary to give clearance.

LOCATION OF REGISTERS.

The opinion is often encountered that registers should be
placed near exposed walls or the outer corners of rooms. On the
contrary they should always be placed as near the furnace as may
be practicable. This location will promote the natural circulation
of air in the room, will permit a sharper pitch to the cellar pipes,
will diminish the loss of heat, the resistance to the flow of air and

$6 Furnace Heating.

the cost. When registers are located near the outer walls the
resistance and loss of heat are so great that the flow of air is weak
and uncertain and the temperature at the register often barely
lukewarm. Registers should not be located below windows, as
under certain conditions a reversal in the direction of the current
of air in the pipe is likely to be induced by the current of cold air
descending along the glass.

The registers on the lower floor should be located with partic-
ular care, those in north or west rooms being favored with regard
to their distance from the furnace. Registers located on inside
walls about two-thirds the height of the room would be in ac-
cordance with the practice in schools and public buildings. This
location would be objectionable, however, in finely furnished
rooms. The dust entering with the air would be likely to dis-
color the walls. The discharge of air from a register so placed
would be more positive than from a floor register, due to the
greater hight of the former.

FLOOR AND WALL REGISTERS.

Registers in the lower story are generally placed in the floor
for convenience in piping. It is difficult to find space for wall
registers on that story without interfering with the proper location
of risers to the floor above. The registers in rooms on upper
floors can often be placed in the wall to advantage. This location
overcomes the necessity of cutting carpets and avoids the accu-
mulation of dust from sweeping.

Unless registers of the convex pattern, as in Fig. 24, are used
they must be boxed out to prevent the body of the register extend-
ing into the flue and cutting off a portion of its area. Wall reg-
isters are generally held in place by clips fastened to the register
box. By removing the register face these clips may be turned
over the edge of the body, which may then be drawn up to the
face by means of the screws at the corners.

The depth of ordinary floor register boxes should be about 6
inches. Slate or cast iron borders are used in connection with
floor registers. The wire netting in the register box should have
the edges turned down about y^ in. to raise it from the end of
the pipe, thus avoiding a restriction of the area at that point.

House Heating.

PATTERN AND FINISH OF REGISTERS.

Registers may be procured in many patterns and styles of
finishes, from the ordinary black japan to those of more elaborate
design with faces of solid brass or bronze. They may be coated
with white porcelain or be electro plated with nickel, silver or
other metal to harmonize with the surroundings.

REGISTERS.

Since the publication of the first edition of this treatise in
1899 a considerable advance has been made in the variety of reg-
isters on the market, also a large increase in the number of
sizes of regular stock patterns.

Table 30 gives the net area and depth of many of the sizes

<^m=^^>>

Fig. 25. — Single-valve Shallow
Register for Partitions.

Fig. 26. — Side Wall Register.

most commonly used; the approximate net area of others may
be readily computed by multiplying the gross area by two-thirds.

In certain open patterns, however, the net area is fully 80
per cent of the gross.

Cast iron predominates as the material most commonly used
for the construction of registers. Steel pressed into various
patterns is used to a considerable extent.

For shallow flues and for thin partitions, registers like the
one shown in Fig. 25 are used, these having no register box
projecting into the flue, cutting down the effective area.

$8 Furnace Heating.

In cases where two rooms, one above the other, are heated
from the same flue, the shutter back of the register face on the
lower floor serves as a deflector, insuring the proper discharge
of air. Otherwise, the upper floor is apt to "rob" the lower one.

Side wall registers in a variety of patterns have come into
use, one style being shown in Fig. 26.

This is a popular type of side wall or base board register. In
this make there is no grill work over the front, this being a solid
casting which, when pushed back, provides the desired opening
for hot air and serves as a deflector. The warm air is de-
flected away from the walls, keeping the dust from them.

Fig. 27.— Side Wall Register.

With side wall registers of these modern types a single flue
can be made to heat a first and second floor room owing to the
larger bottom flue opening that it makes possible so that by using
a pipe of sufficient size in basement, one furnace connection
serves. No cutting of carpets is necessary and more freedom is
given in the arrangement of furniture than when floor registers
are used.

On the second floor either convex or extra shallow side wall
registers may be used without obstructing the flue. It is well
to realize the advantage that the large bottom opening or flue
affords as is possible with this first floor register which takes
the supply from a flue about 7 inches deep or 3 inches deeper
than the studding, and fully twice the usual flue or riser
capacity. In Figs. 26 and 27 it will be noted that the face
of the register near the floor projects some distance in front

House Heating.

of the base-board, which is cut away to make room for the
register body.

In many cases wall registers are decidedly preferable to

Fig. 28. — Riser and Register.

those located in the floor ; the newer patterns satisfy this demand
and eliminate the objections that apply to convex registers.

Fig. 28 shows riser with register on first and second floors.

MANAGEMENT OF A FURNACE.

The following general principles apply to the management of
all hard coal furnaces and should be carefully observed if good
results are desired:

The fire should be thoroughly shaken once or twice daily in
cold weather.

Furnace Heating.

It is well to keep the fire pot heaping full at all times. In this
way a more even temperature may be maintained, less attention
required and no more coal burned than when the pot is only partly
filled. In mild weather the mistake is frequently made of carry-
ing a thin fire, which requires frequent attention and is likely to
die out. Instead, to diminish the temperature in the house keep
the fire pot brimful and allow ashes to accumulate on the grate
(not under it) by shaking less frequently or less vigorously. The
ashes will hold the heat and render it an easy matter to maintain
and control the fire. When feeding coal on a low fire open the
drafts and neither rake nor shake it till the fresh coal becomes
ignited. After the fire is well started the ashes may be shaken
down and fire banked.

Pig. 29. — Basement Plan (8' 0") of a Residence Heated by a Furnace System.

The air supply to the fire is of the utmost importance. An
insufficient amount results in incomplete combustion and a great
loss of heat. To secure proper combustion the fire should be con-
trolled principally by means of the ash pit, ash pit-slide or lower
draft-door.

The smoke pipe damper should be opened only enough to
carry off the gas or smoke and to give the necessary draft. The
openings in the feed-door act as a check on the fire and should be
.cept closed during cold weather, except just after firing, when with

House Heating.

a good draft they may be partly opened to aid the combus-
tion.

Keep the ash pit clear to avoid warping or melting the grate.
The cold air box should be kept wide open except during winds
or when the fire is low. At such times it may be partially (never
completely) closed. Too much stress cannot be laid on the im-
portance of an adequate air supply to the furnace. The symp-
toms of an insufficient supply are irregular and unequal distribu-
tion through the hot air pipes, a hot furnace casing and an over-
heated cellar.

It costs little if any more to maintain a comfortable temperature
in the house night and day than to allow the rooms to become so
cold during the night than the fire must be forced in the morn-

Fig. 30.— First-Floor Plan (10' 0") of a Residence Heated by a Furnace System.

ing, resulting in overheating the furnace, the formation of clink-
ers and the waste of coal.

In case the warm air fails at times to reach certain rooms the
air may be forced into them by temporarily closing the registers
in the other rooms. The current once established will generally
continue after the other registers have again been opened.

02 Furnace Heating.

It is best to burn as hard coal as the draft will warrant. Egg-
size is better than larger coal, since for a given weight small lumps
expose more surface and ignite more quickly than larger ones.
The large lumps do not lie so closely together and allow streams
of comparatively cool air to pass between them, hindering rather
than promoting combustion. The furnace and smoke pipe should
be thoroughly cleaned once a year. This should be done just
after the fire has been allowed to go out in the spring.

SUGGESTIONS TO PURCHASERS.

In purchasing a furnace it is often wise, when competition is
sharp, to select one a size larger than the dealer recommends. By
so doing a larger area of heating surface is secured, hence a
greater proportion of the heat generated will be utilized. The
coal capacity being greater less frequent attention will be neces-
sary, and as the fire will not require forcing, coal may be burned
without the formation of clinkers. Such a furnace will last much
longer and will give far better results and more general satisfac-
tion than a smaller one. The only advantage in buying the latter
is the small saving in the first cost, a saving which soon disap-
pears in repairs and waste of fuel.

FURNACE TESTS.

The following is taken from an article by the author of this
book which appeared in The Metal Worker, under the title,
"Some Data from Furnace Tests on the Rate of Combustion and
the Velocity of Air in the Pipes :"

Tests were made on the heating apparatus in a 29 x 35 foot
frame house with parlor, dining room and reception room on the
first floor, and four bedrooms and a bathroom on the second floor,
heated during one winter season by a brick lined wrought iron
furnace with a 22-inch firepot, and during the following season
by a cast iron furnace with a tapering firepot having an average
diameter of about 23 inches.

The brick lined furnace was tested during a 20 days' run in
midwinter. The average outside temperature during this period,
based on readings taken night and morning, was 26.3 degrees ;
total weight of coal burned, 2328 pounds ; rate of combustion per

House Heating. 63

square foot of grate per hour, 1.84 pounds. A cold day run was
made a little later in the season, the thermometer ranging from
7 degrees below zero to 8 degrees above. During the 24-hour test
coal was fed six times, the total weight amounting to 258 pounds,
making the average rate of combustion 4.07 pounds per square
foot of grate per hour.

The cast iron furnace was tested during a 32 days' trial, the
average outside temperature, based on three readings per day,
being 271/2 degrees. The total weight of coal burned was 4350
pounds ; the average per square foot of grate per hour being 1.97
pounds. During this test a record of room temperature was kept,
the average being fully 70 degrees.

A COLD DAY TEST.

During this test a particularly severe day occurred, the tem-
perature falling to 12 below zero. The coal burned during these
24 hours amounted to 300 pounds, giving an average rate of 4.35
pounds per square foot of grate per hour. Coal was fed seven
times. The firepot was red hot While the thermometer remained
below zero. The weight of ashes and unconsumed fuel passing
through the grate was 10 per cent, of the weight of Lehigh egg
coal supplied.

The house in which these furnaces were installed was of
ordinary frame construction, shingled on building paper and
plastered inside. The total cubic contents of rooms connected
with the furnace was 11,674 cubic feet. The total combined ex-
posed wall and glass surface was 1683 square feet.

It is to be noted that both furnaces used were inside the aver-
age rating given by reputable manufacturers to furnaces of their
size — namely, about 14,000 cubic feet. If based on the exposure
such furnaces are expected to carry approximately 1700 square
feet of combined wall and glass surface when the latter does not
exceed, say, one-sixth the total exposure. The exposure in this
case is practically the same as the above figure. The house had
storm windows. on the north and west sides, yet an average rate
of combustion of nearly 5 pounds per square foot of grate per
hour was found necessary to keep the rooms comfortable in severe
weather. This high rate requires pretty frequent attention and

Furnace Heating.

should be considered a maximum. The dimensions and other
data of the several rooms are as follows :

Rooms.

Dimensions.

Approximate
contents. Sides

Size of
register

First floor. Feet. Cu. ft. exposed. and pipe.

Dining room 13 x 18 x 8% 2,000 2 10x14 10

Parlor 14^x15x8% 1,850 2 10x34 10

Hall 14 x!8x8% 2,140 2 10x14 10

Second floor.

Bedroom 9 x 12 x 8 864 2 8 x 12 7

Bedroom 10 x 19 x 8 1,520 2 8 x 12 8

Bedroom 10 x 12 x 8 960 1 8 x 12 7

Bedroom 13 x 13 x 8 1,350 2 9 x 12 8

Bath 6 x7%x8 390 1 7x10 6

11,674

Anemometer tests were made with the following results :

Temperature Hori-

at Velocity zontal

Room. register. in pipe. Size pipe. run. Elbows.

First floor. Deg. F. Feet. Inches. Feet. 90° 45°

Dining room 116 418 10 8 1 1

Parlor .-. 114 429 10 2 2

Hall 146 465 10 4 1 1

Second floor.

Bedroom ; 100 252 7 16 2 2

Bedroom 104 320 8 12 2 2

Bedroom 104 510 7 2 1 1

Bedroom 127 570 8 2 1 1

Bath 103 286 6811

The above tests were made with cold air box wide open and
with little or no wind. The outside temperature was 5 degrees.
The register temperatures were lower than would have been neces-
sary to keep the rooms comfortable had it not been that they had
been warmed to a temperature considerably in excess of 70 de-
grees, and furnace drafts were checked to reduce the heat.

Other tests were made, closing all registers on the first floor,
giving velocities of over 500 feet in the rooms on the second floor
most remote from the furnace. Tests were made in 34-degree
weather, showing a velocity of only about 280 feet in rooms on the
first floor. Anemometer readings taken in the cold air box showed
a velocity of over 300 feet and a volume of 900 to 980 cubic feet
per minute, corresponding to an air change in the rooms heated
once in about 13 minutes.

Tests made in another house with outside temperature 24 de-
grees showed velocities in pipes leading to the first floor ranging

House Heating. 65

from 306 to 334 feet, the temperature at the registers ranging
from 104 to 109 degrees. Pipes leading to the second floor showed
velocities in excess of 450 feet per minute with slightly lower
register temperatures than on the first floor. The furnace in this
case had a 22-inch firepot. The total volume of air supplied to the
house per minute was 850 cubic feet.

TEST IN ANOTHER DWELLING.

Still another test, made in a different house, gave these results
for rooms located on the second and third floors, the test being
made in cold winter weather. It will be noted that the register
temperatures in this case are much higher than in the previous
tests :

Temperature of Velocity Horizontal

entering air. in pipe. Size pipe. run.

Room. Deg. F. Feet. Inches. Feet. Elbows.

Parlor 138 250 6x10 oval. 9 3

Library 120 210 6 x 7y2 oval. 4 2

Dining room 140 275 7 diameter. 15 2

Hall 151 450 6 x 8 oval. 7 2

Bath 108 280 6 diameter. 8 2

Bedroom 152 500 4% x 7% oval. 4 3

Rear bedroom 140 540 5 x 7 oval. 12 3

These tests give only a general idea of what velocities may be
expected under ordinary working conditions. From the above
and other data the author has adopted these velocities in making
furnace heating computations.

Approximate velocity in pipes leading to first floor, 280 feet
per minute ; to second floor, 400 feet per minute ; to third floor,
500 feet per minute.

During the test made in weather 12 degrees below zero the
temperature of the air delivered by the furnace was 113 to 115
degrees. When the outside temperature rose to 6 or 8 below zero
122 degrees were indicated by the thermometer placed at register
nearest the furnace. The maximum increase in temperature noted
was 130 degrees. The wind was blowing strongly into a wide
open cold air box. Had this been partially closed the maximum
temperature would doubtless have exceeded 140 degrees, which is
commonly used as a basis for computations in work of this kind.

Furnace Heating.

HEATING FROM BELOW ZERO.

Calculations of the loss of heat from buildings are generally
based on a difference of 70 degrees between the inside and outside

Fig. 31.— Second-Floor Plan (9' 0") of a Residence Heated by a Furnace System.

temperatures. In many parts of the country, however, the heat-
ing apparatus must be capable of warming the building to 70 de-
grees during weather of 10, 20 or even 30 degrees below zero,
corresponding to differences of 80, 90 and 100 degrees respect-
ively between the temperatures indoors and out. To compare the
loss of heat when the outside temperature is — 10 degrees with
that when the weather is zero we may assume, for convenience
in figuring, a building having an equivalent glass surface (E. G.
S.) of looo square feet. Now under zero conditions, with air
entering the rooms at 140 degrees, one-half the heat will be car-
ried away by the air escaping at approximately 70 degrees. The
other half will escape by transmission through the walls, win-
dows, floors and ceilings. The loss through 1000 square feet of
glass surface under the conditions named is 85,000 heat units per
hour (since i square foot of glass will transmit 85 heat units per
hour when the difference in temperature on the two sides is 70
degrees). The loss of heat by leakage or the escape of air is as
much more, making a total loss of 170,000 heat units per hour.
Now with an outside temperature of — 10 degrees, other condi-

House Heating. 67

tions remaining the same, the loss by transmission will be in-
creased in proportion to the difference in temperature, or will be

D X 85,000 — 97,000 4- heat units. The air entering at 140 and

70 + 10 80 ...

escaping at 70 degrees carries away-- - of the heat

140 -\- 10 150

brought in, the remaining — — escaping by transmission. Each
cubic foot of air admitted at 140 degrees brings in :

(Weight cubic foot air Rise in temper- Specific heat\

at 140°. ature. of air. 1—2.36 heat units.

0.066 X (140° + 10°) X 0.338 /

jjfi of which, or i.i heat units, will escape by transmission.
Hence, to provide for the loss of 97,000 heat units per hour in this

manner, as calculated above, '- — = 88,000 cubic feet (in round

numbers) of air at 140 degrees will be required; 88,000 X 2.36
= approximately 208,000 heat units per hour, as compared with
170,000 under zero conditions.

That is, 22 per cent, more heat will be required to maintain 70
degrees inside with — 10 degrees outside than to maintain the same
temperature with zero outside, the air admitted to the room to be
140 degrees in each case. The increased loss of heat calculated
in a similar manner for — 20 and — 30 degrees outside temperature
will be 46 per cent, and 73 per cent, respectively.

CHAPTER III.
THE COMBINATION SYSTEM.

HOT WATER AKTD HOT AIR.

In the combination system of heating, where both or either
air and water serve to convey heat from the furnace to the
various rooms, a slight saving in fuel is effected by causing the
gases to pass over water heating surface suspended above the fire.
Aside from this, whatever gain is made is at the expense of ven-
tilation, since in rooms heated by direct radiation the same air is
used over and over. The main reason for employing combination
hot water heating is to heat points too remote from the furnace to
be successfully heated by hot air. Plans of a residence heated
by a combination system are shown in Figs. 32, 33 and 34.

Living rooms should receive a continuous supply of warm
fresh air. This may be furnished most conveniently in the ordi-
nary manner through the furnace pipes, adding direct radiation
if necessary in exposed corners. To deliver fresh air at points too
remote from the furnace to be reached by an ordinary hot air
pipe an indirect hot water radiator may be used, suspended just
below the register and supplied with air from the furnace or
directly from out of doors. Valves should be omitted from such
radiators to avoid danger from freezing.

In finely furnished rooms indirect radiation may be used to
advantage in place of direct radiation when the appearance of the
latter is considered objectionable or when it is difficult to provide
space for them. When so used they may be arranged with a re-
turn duct and the air in the room rotated as in direct heating.
Under such conditions the heating surface is less effective than
when placed in the room, hence it must be liberally proportioned.
With this return air arrangement ventilation is eliminated.

DIRECT RADIATION.

The usual location for a direct radiator is near an outside wall
or below a window, although good results may be obtained in
rooms not too greatly exposed when the radiator is located near
one of the inner walls, especially in cases where efficient weather
stripping is used or tight double windows provided. Radiators

The Combination System. 69

should be set in as inconspicuous places as possible, provided
such location will be effective. Direct radiation may properly be
used in rooms where a constant supply of fresh aid is not re-
quired, as in bedrooms occupied only at night, when air may
be admitted through raised windows, or in halls not used as
living rooms. Unlike steam, the temperature of the water in
the radiators may be gradually reduced by throttling down the
supply with the valve. In rooms where heat may not be re-
quired for days at a time a small hole should be drilled through

pa a a a a a

Fig. 32.— Basement Plan (8' 10") of a Residence Heated by a Combination System.

the disk of the radiator valve to prevent the heat being entirely
shut off. Hot water radiators contain as a rule from i to i*/2
pints of water per square foot of surface.

HOT WATER VS. HOT AIR.

Owing to the capacity of water to store heat, rooms having
radiators are less subject to sudden changes in temperature than
those where hot air is used. Ordinarily this is an advantage, but
in living rooms which on certain occasions may contain an un-
usual number of occupants this feature is objectionable. It is sel-
dom noticed that a room has become overheated until the temper-
ature has risen considerably above the normal. Then the radiator

7° .x Furnace Heating.

valve is closed, but the water continues to give off its stored heat
for some time thereafter, which with the heat from the lights and
that from the bodies of the occupants makes it difficult to reduce
the temperature quickly. The act of closing a register shuts off
all the heat at once.

VALVES ON RADIATORS.

One or two radiators should be left without valves to prevent
all being shut off at once, which would cause the water in the

Fig. 33. —First-Floor Plan (9' 0") of a Residence Heated by a Combination System.

system to boil. Where the furnace is connected with but one
radiator the water must be allowed to circulate through it at all
times, whether heat is desired or not.

" BALANCE " OF THE SYSTEM.

One of the difficulties in a hot water combination system is
to secure a proper " balance " between the hot water and the
warm air, so that they will work harmoniously and one not heat
at the expense of the other. It is advisable to place in the hall or

The Combination System. 71

other convenient room both a register and a radiator, each of suffi-
cient size to heat the space, so that by using one or the other a
proper " balance " may be maintained.

HEATING SURFACE IN FURNACE.

The water heating surface in the furnace may be placed in
contact with the fire or suspended above it. In some heaters the
water is first brought in contact with the surface in the fire and
then ascends through a coil or cast iron section surrounded by the
hot gases. The tendency of the water heating surface to deaden
the fire with which it is in contact and to greatly diminish the air
heating capacity of the furnace limits its use. When the heating

Fig. 34.— Second-Floor FJlan (8' 6") of a Residence Heated by a Combination System.

surface is in contact with the fire, the water is maintained at a more
even temperature than when heated by a coil or section suspended
above it. With the latter the heating surface is acted upon chiefly
by the radiant heat from the top of the fire, which amounts to lit-
tle just after firing or until the fresh coal has become ignited. In
the meantime the temperature of the water falls. The heating
capacity of such surface may be varied to suit conditions. In
severe weather by carrying a high fire in contact with the coil or
section its capacity may be greatly increased.

When special castings cannot be procured for attaching a hot
water combination to a furnace, coils of wrought iron pipe are

Furnace Heating.

often used, placed either above or partly in the fire. They are
generally made of i% or i^ inch pipe, according to the radia-
tion supplied. The rating for various types of combination
heaters is shown in the following table.

HOT WATER COMBINATION HEATERS.

A few types of combination heaters are illustrated in the
following figures :

Fig. 36. — Base Section when Used without
Ring Section.

Fig. 35.— Dome Section.

Fig. 37. — Ring Section.

Fig. 35 shows a dome section suspended over the fire. When
additional heating capacity is desired one or more ring sections
as shown in Fig. 37 are placed above the base section when
the design of combustion chamber permits.

Figs. 38 and 39 shows another design of base section and
bell section for use above the fire.

Fig. 40 shows a combination heater section designed for use

The Combination System.

in a brick lined furnace, the hollow castings being inserted around
the fire pot in place of fire bricks. The discharge pipes are
joined above the fire as indicated.

Fig. 38. — Horseshoe Only.

Fig. 39.— Bell Only.

Fig. 40. — Three Long Regular Sections
Connected with Pipes and Discs.

Fig. 41. — Coil Combination Heater for
Furnace.

Table Via. — Showing Capacity of Hot Water Combination Heaters in Furnaces
Expressed in the Number of Square Feet of Direct Radiating Surface which May
Be Kept at 160 Degrees Temperature per Square Foot of Heating Surface in the
Combination Heater.

D-ription. s5SEff&.

A. Cast-iron sections suspended above the fire 15 to 20

B. Cast-iron sections in contact with the fire 40 to 60

C. A. and B. combined 25 to 35

D. Pipe coil suspended above the fire 20 to 25

E. Pipe coil buried in the fire 50 to 60

F. D. and E . combined 30 to 40

74 Furnace Heating.

DIRECT RADIATING SURFACE.

In estimating the total amount of radiation supplied by the
furnace the surface of the supply and return pipes should be added
to that in the radiators, unless the pipes are to be covered. In the
combination system with open tank sufficient radiating surface
should be provided to heat the rooms to 70 degrees in zero weather,
with a maximum water temperature not over 190 degrees. This
will leave a reasonable margin below the boiling point. If the
amount of surface is calculated on the thumb rule basis of cubic
space to be warmed the allowances in Table VII will be found safe
under ordinary conditions. Of course in determining the amount
of surface required for a given room due regard must be had for
its exposure, glass surface and the character of its walls.

Table VII.

For rooms with one exposed wall, allow 1 square foot of radiation for 30 to 40
cubic feet of space.

For rooms with two exposed walls, allow 1 square foot of radiation for 25 to 30
cubic fee* of space.

For rooms with three exposed walls, allow 1 square foot of radiation for 20 to
25 cubic feet of space.

For bathrooms and small exposed rooms, allow 1 square foot of radiation for
15 to 25 cubic feet of space.

Use maximum or minimum amount of surface given by above
rule according to the degree of exposure. For the pressure sys-
tem use about three-quarters as much surface as with an open
tank. For the purpose of permitting pressure and securing hotter
water without boiling use a mercury seal rather than a safety
valve.

If desired the radiating surface may be based directly on the
loss of heat through walls, windows, floors and ceilings. A con-
venient approximate method is to consider 4 square feet of ordi-,
nary wall equivalent in heat transmitting power to I square foot
of glass ; then reduce the exposure of the room to equivalent glass
surface by adding to the window area one-quarter the area of the
outside walls. Outside doors are to be estimated as equivalent to
one-half their area in glass. If the space below or above the

The Combination System. 75

room is cold, add to the equivalent glass surface one-twen-
tieth of the area of floor or ceiling. In the case of ordinary
cellars or attics in the body of the house it is hardly necessary
to add for heat losses through floors or ceilings. The total
equivalent glass surface thus obtained divided by 1.8 will give the
amount of radiation required with the open tank system. For
the pressure system divide by 2.4.

The 1.8 and 2.4 above are deduced as follows: Since
the heat given off per square foot of direct radiating surface per
hour when placed in rooms at 70° temperature is approximately
150 B. t. u. with hot water and 250 B. t. u. with steam, it follows
that with 85 as the heat loss per square foot of equivalent glass

surface, E. G. S.^— = 1.8 is the factor for hot water and -r—
85 #5

= 2.4 is the factor for steam.

Example computing radiation :

How much hot water radiation is required to heat a room
14 x 16 x 9, exposed 2 sides N. and W. and having 20% glass?

Exposure = 14+ 16X9 = 270 square feet.
Glass, 20 per cent = 54

Net, wall 216

E.G.S. of net wall = net wall -^ 4= 54

Total E.G.S. =actual glass +E.G.S. of net wall = 108

Add 20 per cent for exposure factor =approx. 22

130

Allowance of 10 per cent to provide for quick heating and to
cover leakage losses 26

Total E.G.S. -f allowances 156

This total 156-5-1.8 as above =87 square feet.

Ratio = cubic contents of 2016-7-87 = 1:23 cubic feet.
To compensate for the increased loss of heat due to winds add
at least:

Furnace Heating.

Fifteen to twenty per cent, for rooms having a northerly or
westerly exposure.

Ten to fifteen per cent, for rooms having an easterly ex-
posure.

To insure quick warming on cold mornings add at least ten
per cent, to the transmission losses. If a room has a large
cubical contents compared with the outside exposure, some
allowance should be made for bringing the air in the room in
addition to the wall losses. The above factors in conjunction
with the liberal heat loss allowance of 85 for glass will provide
for ordinary air leakage without computing this loss separately.

REG.

' Fig. 42.— Indirect Stack.
INDIRECT RADIATING SURFACE.

For indirect heating with pin radiators the sections should
have a depth of 10 to 12 inches io thoroughly warm the air. Fig.
42 shows the arrangement of an indirect stack.

To estimate the amount of indirect radiation when the air
supply is taken from the furnace add 25 per cent, to the amount of
direct radiating surface that would be required. When the air is
admitted to the stack directly from out of doors add at least 50
per cent, to the amount of direct radiating surface that would be
necessary. With indirect radiation for the first floor allow at
least i % square inches to each square foot of radiating surface
surface for warm air flue, that is an indirect stack of 100
square feet surface would by this rule require 125 square
inches cold air supply and 150 square inches warm air dis-
charge pipe.

The Combination System. 77

HEATING CONSERVATORIES.

For heating conservatories 154, i.H> or 2 inch pipes are gen-
erally used, run along the wall under the benches. Fig. 24 shows
a wall coil. One square foot of radiating surface is, with open tank
system, sufficient for 2 square feet of glass. In other words :

1 lineal foot of lJ4-inch pipe will carry $, square foot of glass.
1 lineal foot of IH-inch pipe will carry 1 square foot of glass.
1 lineal foot of 2-inch pipe will carry 1*4 square feet of glass.

TAPPING OF RADIATORS.

Hot water radiators are commonly tapped :

1 inch for radiators containing 40 square feet and under.
Ii4 inches for radiators containing 40 to 72 square feet.
IJi inches for radiators containing 72 square feet and over.

Unless otherwise ordered indirect radiators are usually tapped
2 inches, then bushed to the desired size.

SIZES OF PIPES.

The following sizes of flow pipes for the amount of radiating
surface stated will be found sufficient for ordinary runs :

Table VIII.— Capacity of Hot Water Pipes for Direct and Indirect Radiation.

1-inch pipe will supply 40 square feet of direct radiating surface.

l»4-inch pipe will supply 72 square feet of direct radiating surface.

1^-inch pipe will supply 125 square feet of direct radiating surface, or 80 square feet of
Indirect radiating surface.

2-inch pipe will supply 225 square feet of direct radiating surface, or 150 square feet
of indirect radiating surface.

2V6-inch pipe will supply 350 square feet of direct radiating surface, or 240 square feet
of indirect radiating surface.

3-inch pipe will 'supply 500 square feet of direct radiating surface, or 350 square feet
of indirect radiating surface.

OPEN TANK VS. PRESSURE SYSTEMS.

The open tank system is the safer. The pressure system
with closed tank and safety valve has been superseded by mer-
cury seal systems where there is no safety valve to possibly stick
on its seat and which give the advantages of the old fashioned
pressure system without its disadvantages. The advantages are
smaller radiators and pipes owing to the higher water tempera-
tures that may be carried.

The open tank system is most commonly used. Under certain

78 Furnace Heating.

conditions the water may boil and overflow, but if properly ar-
ranged this will do no harm and with ample radiating surface will
seldom occur. The surging in the pipes will call attention to the
fact that the apparatus is not working properly, and that either
more radiation must be turned on or the fire must be checked.

EXPANSION TANK AND CONNECTIONS.

The house tank is sometimes used as an expansion tank, but

L ' ' ' ^

N !

i — i)'

H r

H H 'V

® NO

Ipo^

Fig. 43.— Wall Coil.

this is unwise, as in case of boiling rusty water is forced into the
tank, rendering the house supply turbid and unfit for use.

A separate tank should be used, which may be provided with a
ball cock if desired to insure the proper water level being main-
tained. The expansion pipe must be so connected that the free ex-
pansion of the water cannot be interrupted.

Water expands about one twenty-fourth of its volume at 40 de-
grees when heated to 210 degrees. The expansion tank should
have a capacity equal to about one-twelfth that of the entire sys-
tem. The radiating surface divided by 50 gives the proper capac-
ity of the expansion tank in gallons. Care must be taken to locate
the expansion tank where there is no danger from freezing.

In the cheapest work no expansion tank whatever is provided,
the system being connected directly with the street service, full city
pressure of perhaps 80 pounds or more being maintained on the
system. In case of leaks from any cause the damage resulting
with such a pressure would be much greater than with either the
closed or open tank system.

The Combination System. 79

SYSTEM OF PIPING.

Two systems of piping are commonly employed. In one the
mains are run through the basement, taking off supply and return
connections to the various risers and connecting the expansion pipe
to the return near the heater. Fig. 44 shows a radiator on a two-
pipe system.

In the other, known as the " overhead feed," the flow pipe rises
directly to the expansion tank, the radiators being connected with

Fig. 44.— Single Valve Radiator Connection.— Two-Pipe System.

the drops or returns, as in Fig. 45, a single pipe serving for both
supply and return to radiators on several floors. No air valves are
required with this arrangement, since all air escapes from the ex-
pansion tank, located at the highest point. Nothing can interrupt
the circulation of water through the mains. Fewer pipes and con-
nections are necessary and the circulation is likely to be better than
with the two-pipe system. An ordinary expansion tank may be
used if desired, connected in the ordinary way and located above
the top of the lower feed distributing main, the latter being
vented.

STEAM COMBINATION.

Some furnaces may be fitted with a steam heating combination.
The advantages claimed for this system are quick heating ability
and the use of smaller radiators and pipes than in an open tank sys-
tem, with resulting economy in space and cost.

Among its disadvantages as compared with hot water may be
stated its sensitiveness to changes in the condition of the fire owing

Furnace Heating.

to the small amount of water in the system, steam going down
quickly with a deadening of the fire.

Unless vacuum valves are used there is no range of temperature
5n the radiators, as with hot water. With steam the boiling point

Fig. 45.— Kadiator Connections.— Overhead Feed System.

(212 degrees) must be reached before the radiators become hot.
The small water capacity involves frequent filling and damage is
likely to result from inattention. The apparatus with its additional

The Combination System.

valves and fittings is less simple than the hot water combination.
In estimating the steam radiating surface allow about six-tenths as
much surface as would be required using hot water radiators with
the open tank system.

TABLE^IX. — Welded Pipe, Steel or Iron.

1^4-inch and below, butt welded, proved to 300 pounds per square inch, hydraulic
pressure.

l^j-inch and above, lap welded, proved to 500 pounds per square inch, hydraulic
pressure.

TABLE OF STANDARD SIZES.

U«M

«!

!£J

l|l«

<O « O P

5 3

B^sl

&5*

S 8

•3

*e|

^-i 0

•a

la1-1^

4^0

f'3

t||

"SiS'S^

§2

•s-s

§ W2 O

§ -2 *W ' '

® Q

o »9

Ins.

Ft.

Ins.

Ft.

Lbs.

0.405

O.OH8

1.272

9.434

0.057

0.1288

2,500

0.24

0.54

O.U85

1 (.96

7.075

0.104

0.229

1,383.28

0.42

«/

0.675

0.091

2 12 L

5.658

0.191

0.3578

754.322

0.56

0.84

0.109

2.639

4.547

0.304

0.£54

473.84

0.84

1.05

0.113

3.299

3.638

0.533

0.866

270.016

1.12

1.315

0.134

4 131

2.904

0.861

1.358

167.246

1.67

11 ^i

1.66

0.140

5.215

2.301

1.496

2.164

96.257

2.24

nil

1.9

0.145

5.969

2.01

2.036

2.835

70.727

2.68

HJi

2.375

0.154

7.461

1.608

3.356

4.430

42.908

361

11V6

2.875

0.204

9 032

1.329

4.78

6.492

30.337

5.74

3.5

0.217

10.996

1.091

7.383

9.621

19.504

7.54

0.226

1.2.566

0.955

9.887

12.566

14.567

4.5

0.237

14.137

0.849

12.73

15.904

11.312

10.66

0.246

15.708

0.764

15.961

19.635

9.022

12.34

5.563

0.259

17.475

0.687

19.986

24 301

7.205

14.5

6 625

0.28

20.813

0.577

28.89

34.472

4.984

18.76

7.625

0.301

23 955

3.501

38.738

45.664

3.717

23.27

8.625

0.322

27.096

0.443

50.027

58.426

2.876

28.18

9.625

0.344

30.238

0.397

62.73

72.760

2.290

33.7

10.75

0.366

33.772

0.355

78.823

90.763

1.827

40.06

HEAT GIVEN OFF BY DIRECT RADIATORS.

Cast iron radiators with low pressure steam transmit approxi-
mately 250 heat units per square foot of surface per hour. Hot
water radiators on open tank system transmit about 150 heat
units per square foot of surface per hour.

A rate of heat emission of 1.6 B.T.U. per square foot of direct
radiating surface per hour per degree difference in temperature
between that of the heating medium inside the radiator and that
of the air in the room is a fair average value.

CHAPTER IV.
AIR, HUMIDITY, AND VENTILATION.

COMPOSITION AND IMPURITIES OF THE ATMOSPHERE.

Atmospheric air is a mixture composed of about 79 parts of
nitrogen and 21 parts of oxygen by volume, and in 10,000 vol-
umes there are from 3 to 5 volumes of carbonic-acid gas.

This gas in moderate quantities is not harmful, but it is nearly
always " found in bad company." It is mixed with the organic
matter exhaled from the lungs and thrown off by the skin. In
rooms having no special provision for ventilation the air must
be breathed again and again, constantly becoming more foul. The
porportion of carbonic acid in the air may be readily determined
by several methods. It therefore forms the most convenient
measure of the vitiation, since in occupied rooms the amount of
harmful organic^natter in the air is found to correspond with the
proportion of carbonic acid. This CO 2 index or standard has
long been commonly used, but the doctors of hygiene are now
inclining to a conviction that high room temperature and high
humidity are more to be guarded against than a slight excess|of
CC>2 in the atmosphere of occupied rooms. See pages 90 to 102.

When the number of parts of the latter exceeds 6 to 8 in 10,-
ooo of air, the room seems close to one entering from out of
doors and a slight odor is perceptible. By the process of dilu-
tion the air may be kept, within limits, at any desired degree of
wholesomeness. To maintain in a room continuously occupied
for a number of hours an atmosphere in which the carbonic acid
shall not exceed 6 parts in 10,000, an air supply of about 50 cubic
feet per minute per occupant must be admitted. To accomplish
this, much larger heating apparatus and flues than customary
would be required. The public has not yet been educated to a
full appreciation of what good ventilation really is. The commonly

Air, Humidity, and Ventilation. 83

accepted standard for schools is 30 cubic feet of fresh air sup-
plied per minute per occupant. This allowance will keep the
carbonic acid down to about 7.4 parts in 10,000 of air.

Churches generally have at least 50 per cent, more space per
occupant than schools, say, 300 cubic feet, and are occupied for
much shorter periods. Therefore a smaller air supply is con-
sidered sufficient for such buildings.

An allowance of 20 cubic feet per minute is common, and
some authorities recommend 1000 cubic feet per person per hour.
In halls, which generally have a greater number of seats to a
given space than the above classes of buildings, the air supply
should be based on a 20 cu. ft. per min. per capita basis, provided
this allowance will not change the air so frequently that uncom-
fortable drafts will be produced. In standard size school rooms
the air is changed, on the 30 cubic feet per capita basis, once in 7
minutes.

This is about as rapid a change as can be recommended with
inlets and outlets as commonly arranged.

In halls having perhaps only 100 cubic feet of space per occu-
pant, unless the openings were very carefully arranged, an air
supply of 20 cubic feet- each would be likely to give trouble from
drafts.

HUMIDITY.

The amount of moisture or water vapor contained in the at-
mosphere is expressed in terms of Actual Humidity, meaning the
number of grains of water vapor per cubic foot of space, or Rela-
tive Humidity, meaning the ratio expressed in hundredths, be-
tween the weight of moisture in the air and that contained in an
equal volume of saturated air at the same temperature. The
Dew Point is the point at which the saturation is complete, when
the vapor can no longer be held in suspension, but is deposited in
the form of dew.

The effect of humidity on bodily comfort is marked, a per-
son feeling far more comfortable on a hot, dry day, for example,
than on a muggy day with a much lower temperature. It is a
well-known fact that evaporation is accompanied by cooling,
which accounts for the greater comfort experienced when the
evaporation from the skin is rapid, as in a dry atmosphere.

84 Furnace Heating.

Table X.

Box says that when the air contains about —

85 per cent, water vapor we consider it .. .. damp.

65 per cent, water vapor we consider it moderately dry.

50 per cent, water vapor we consider it dry.

35 per cent, water vapor we consider it very dry.

^5 per cent, water vapor we consider it extremely dry

Billings states that no discomfort is experienced in an atmos-
phere with a relative humidity of 30 to 40, and that at the Boston
City Hospital no ill effects were observed with a relative humidity
of 15 to 21.

The air supplied by furnaces is moistened to a very limited
extent by means of the water evaporating pan. The capacity of
air to absorb moisture increases rapidly with rise in temperature.
For example, air at 72 degrees can absorb four times as much
moisture as air at 32 degrees. We commonly speak of air ab-
sorbing moisture; we really mean space.

Table XI —The Weight of Water Vapor per Cubic Foot of Saturated Soace nf
Different Temperatures.

Temperature.
0

Weight of vapor in
grains per cubic
foot.
0.54

Temperature.
50

Weight of vapor in
grains per cubic

4 09 — 4 approx

0.84

0 99 — 1 approx

7 99 — 8 approx

1.30

1095

1.97 — 2 approx.

14.81

£.88

100

19.79 20 approx.

1 pound avoirdupois = 7,000 grains.
Approximately 1,000 heat units are required to evaporate a pound of water.

Since the moisture that may exist in a given space increases
rapidly with a rise in temperature, as shown in Table XI, to main-
tain even a moderate relative humidity a great quantity of water
and a considerable amount of fuel will be required to evaporate it.

Take, for example, an eight or nine room house having an
air supply of about 800 cubic feet per minute. = 48,000 cubic feet
per hour. Outside temperature, 30 degrees.

Suppose the air entering the furnace has a relative humidity
of 65. Now I cubic foot of saturated air at 30 degrees tempera-
ture will contain approximately 2 grains of water vapor, hence
with relative humidity of 65 per cent., I cubic foot will contain

— X 2 = 1.30 grains. Each cubic foot of air entering at 30 de-
grees temperature will, on being heated to 70 degrees, ex-

Air, Humidity, and Ventilation. 85

pand to i. 08 cubic feet. A cubic foot of saturated air aV ^ie lat-
ter temperature will contain approximately 8 grains of moisture,
or with relative humidity 50, for example, will contain 4 grains.

Since the 48,000 cubic feet of air entering the furnace at 30
degrees becomes expanded to 48,000 X 1.08 = 51,840, at 70 de-
grees temperature, we have as the amount of water which must
be evaporated per hour to maintain a relative humidity of 50 in
the air at 70 degrees

51,840 cubic feet X 4 — 48,000 X i-3 = 154,960 grains = 22.14
pounds.

As about 1000 heat units are required to evaporate i pound of
water, 22,140 heat units will be required per hour, and assum-
ing that 8000 heat units are utilized per pound of coal burned, we

have — o'— -- — 2-77 pounds coal per hour = 66l/2 pounds coal
per day required merely to evaporate the water.

EXPANSION OF AIR AND ABSOLUTE TEMPERATURE.

Air expands and contracts with changes in temperature ac-
cording to a known law — viz., for each degree rise or fall in
temperature from 32 degrees F. air expands or contracts V««
of its volume at that temperature. If a cubic foot of air be
heated through 491 degrees from 32 degrees, or to 523 degrees,
it will double in volume. On the other hand, if a cubic foot of
air be cooled through 491 degrees from 32 degrees, or to 459 de-
grees below zero, it will theoretically contract --— of its original

bulk, or will entirely disappear. This point, 459 degrees below
zero, or more accurately 459.4 degrees, is known as absolute zero,
and is the point from which the expansion of air is reckoned in
determining its relative volume at different temperatures, the
volume being proportional to the absolute temperature. For con-
venience in making ordinary calculations 460 degrees F. below
zero may, with sufficient accuracy, be considered absolute zero.
Hence the absolute temperature of a body is equivalent to 460 de-
grees plus its Fahrenheit temperature. Suppose, for example,
we wish to determine how much space I cubic foot of air entering
a furnace at o degree F. will occupy when heated to 140 degrees

Furnace Heating.

F. Since the volume varies in proportion to the absolute tempera-
ture, we have :

Absolute temperature of air at 0° F = 0° 4- 460° = 460 | Volume at 0° is to volume at
Absolute temperature of air at 140° F = 140° + 460° = 600 f 140° as 460 is to 600.

Hence, volume at 140 degrees = — ^- X volume at o degree; vol-

400

ume at 140 degrees =1.3 cubic feet.

Table XII.— The Approximate Volume to Which 1 Cubic Foot of Air at 0° Will
Expand When Heated to the Temperatures Stated in the Table. Volume of
Air at 0° = 1 Cubic Foot.

Volume when heated to—

Degrees. Cubic feet.

10 =1.02

20 =1.04

30 =1.06

40 =1.09

50 =1.10

60 =1.13

70 =1.15

80 =1.17

90 =1.20

100.... .. =1.22

Volume when heated to—
Degrees.

110

120

130

140 ,

150 ,

200....

400..
500..

Cubic feet.

= 1.24

=1.26

=1.28

=1.30

=1.33

= 1.44

=1.65

=1.87

. . = 2.09

Table XIII.— The Weight of Dry Air per Cubic Foot at Different Temperatures.

Weight of a

Temperature. cubic foot

Degrees F. in pounds.

0 0.0864

12. .. 0.0842

2i 0.0824

32 0.0807

42. .. . 0.0791

52 0.0776

62... 0.0761

72 0.0747

82. .. 0.0733

92 0.0720

102. ... 0.0707

Temperature.

Weight of a
cubic foot
in pounds.

112.

122 0.0682

132 0.0671

142 0.0660

152 0.0649

162 0.0638

172 0.0628

182 0.0618

192 0.0609

202 0.0600

212... . 0.0591

THE FLOW OF AIR IN PIPES.

The resistance to the flow of air through pipes may be approxi-
mately stated as follows :

The resistance is proportional to the surface over which the air
passes and to the square of its velocity. In other words, the resist-
ance varies directly with the length of the pipe and the square of
the velocity and inversely as the diameter. With pipes of the same
length and air traveling at the same velocity the resistance will be
inversely proportional to the diameter.

Air, Humidity, and Ventilation. 87

VELOCITY OF AIR IN FLUES.

The velocity of air in a flue is governed by its hight and the
difference between the inside and outside temperature. Suppose
we have a flue I square foot in area and of hight h, represented in
Fig. 46.

The air in the flue is balanced by a column of colder outside air
of hight H, leaving an unbalanced force represented by the hight

o 1
Pig. 46. — Flue Diagram.

(h — H), tending to produce a velocity at the base of the flue
equivalent to that developed by a body falling freely through a
distance represented by the hight (h — H).

The velocity acquired by such a body, neglecting friction, is
expressed by the equation
v = V 2gh (a)

Here v = velocity in feet per second, g = the acceleration in
feet per second due to gravity, = 32.2 feet, h = the hight through
which the body falls — in this case represented by (h — H).

88 Furnace Heating.

Now let

tc0 = the weight per cubic foot of outside air.

wr = the weight per cubic foot of air in the flue.

to = the absolute temperature of the outside air = Fahrenheit temperature + 459. 4V

tf = the absolute temperature of the air in the flue = Fahrenheit temperature -j- 469. 4V

We have seen that the velocity at which the air enters the base
of the flue is expressed by

Now since the columns of air represented by h and H balance each
other we have weight of column h = weight of column H; or,

hwf^Hw0 (C) hence # = — F (d)

The density of the air. or its weight per cubic foot, varies inversely

as the absolute temperature ; hence we may substitute for — -, -^=°

T
equation (d) becoming H = h ~ (e)

J- F

Substituting this value of H in (b ) we have

(h-h ? =

Now the weight of air leaving the flue must be equal to the weight
of air entering — that is,

Velocity of air leaving flue x wp = velocity of air entering
flue x w0 ........................................................................ (g)

Velocity of air leaving flue =

Fk-nt£»ri-nor fliif* v in

.(h)

velocity of air entering flue x w0

Or, since the weight varies inversely as the absolute temperature,
Velocity of air leaving flue =

velocity of air entering flue x 7"P , ..

T ~ W

4 o

Equation (/) gives the velocity of the air entering the flue, hence
Velocity of air leaving or passing through the flue =

Air, Humidity, and Ventilation. 89

Allowing 50 per cent, for friction, and substituting the value of g
= 32.2, the velocity in feet per minute in the flue is

from which the followin table is calculated :

Table XIV. — The Approximate Velocity of Air in Flues of Various Rights

Outside temperature 32 degrees. Allowance for friction 50 per cent, in flue one square

foot in area.

of ,

Excess

of temperature of air in the flue over that out doors. —

flue. 10°

20°

30°

40°

50°

60° 70"

80U

90°

100°

120°

140°

Feet.

Velocity of air in feet per minute.

5 . 77

Ill

136

159

179

199 216

234

250

266

296

325

10 109

156

192

226

254

281 306

330

354

376

418

460

15 133

192

236

275

312

344 376

405

432

461

513

565

20 154

221

273

319

359

398 434

467

500

532

592

650

173

248

305

357

402

445 485

522

560

595

660

728

30 ..

.. 189

271

334

390

440

487 530

572

612

652

725

798

204

293

360

423

475

527 574

620

662

705

783

862

40 . .

...218

311

386

452

508

562 612

662

707

753

836

920

45 . .

.. 231

332

408

478

538

597 650

700

750

800

887

977

50..

...244

350

432

503

568

630 685

740

790

843

935

1030

60 .

267

383

473

552

622

690 750

810

865

923

1023

1125

.. 289

413

510

596

671

746 810

875

935

995

1105

1215

80..

..308

443

545

638

717

795 867

935

1000

1065

1182

1300

327

470

578

678

762

845 920

990

1060

1130

1252

1380

100 . .

.. 345

495

610

713

802

890 970

1045

1118

1190

1323

1455

The volume of air in cubic feet per minute discharged by a flue
equals the velocity in feet per minute multiplied by the area in
square feet. Knowing any two of these terms, the third may be

readily found.

volume volume

Velocity = - Area — — : — r-

area. velocity.

Example. — Find the area of a flue 20 feet high that will dis-
charge 3000 cubic feet per minute, when the excess of temperature
in the flue over that out doors is 40 degrees.

Opposite 20 in left hand column and under 40 on upper line is
the number 319, representing the velocity in feet per minute. The
volume 3000 ~ 319 = 9.4 square feet, the required area. In esti-
mating the effective hight of a warm air flue from a furnace, con-
sider the flue to begin 2 feet above the grate.

Table XV.— Wind Velocity
Weisbach defines winds as follows :

Scarcely appreciable wind 90 feet per minute equals 1 .02 miles per hour.

Very feeble wind 180 feet per minute equals 2.04 miles per hour.

Feeble wind 360 feet per minute equals 4.1 miles per hour.

Brisk wind 1080 feet per minute equals 12.3 miles per hour.

Very brisk wind 1800 feet per minute equals 20.4 miles per hour.

High wind 2700 feet per minute equals 30.7 miles per hour.

Very high wind ... ... 3600 feet per minute equals 40.1 miles per hour.

Violent wind 4200-5400 feet per minute equals 47.8-61.4 miles per hour.

Hurricane.. 6000 feet per minute equals 68.1 miles per hour-

The United States Weather Bureau defines a gale as a wind blowing 40 miles per hour.

90 Furnace Heating.

IMPORTANCE OF VENTILATION.

Under modern conditions, with buildings having a tight
construction and a relatively small accidental in-leakage of air,
the question of providing a sufficient supply of fresh air and the
removal of foul air, becomes an important one, especially in the
case of rooms which are crowded or occupied continuously for
many hours, with particular reference to moving picture theatres
and buildings of similar character.

CAUSES OF ATMOSPHERIC VITIATION.

The accumulation of carbon dioxide is the most commonly
mentioned cause of vitiation. When it is realized that this gas
is increased over a hundred fold in the air passing through the
lungs, it is not surprising that this component of the atmosphere
accumulates rapidly in occupied rooms.

An interchange of gases takes place in the lungs, called the
respiratory exchange, oxygen passing from the air to the body
and carbon dioxide from the lung cells to the air about to be
exhaled. The air discharged from the lungs is saturated with
water vapor.

In proportion to their weight children give off about twice
as much carbon dioxide as adults, hence the importance of
adequate ventilation in rooms occupied by little ones. Some
authorities state that the amount of air breathed may be averaged
as 15 cubic feet per hour, and that 0.6 cubic feet of carbon dioxide
is exhaled per hour by a person in repose ; that the vapor elimin-
ated by a person at rest is about il/2 oz. per hour, about one-
fourth of the vapor elimination coming from the lungs.

A great deal of effluvia or organic matter is carried through
the pores with the perspiration. It gives a foul odor to crowded
or poorly ventilated rooms. The more active the perspiration
the more free the effluvia elimination. An assembly of 1000
adults is said to give to the air nearly 100 Ibs. of perspiration
vapor per hour. Prof. Woodbridge states that "high humidity
increases the amount of decomposable matter present in an oc-
cupied enclosure, it hastens its decomposition, it accelerates its
diffusion and intensifies its putrefying: odor."

Air, Humidity, and Ventilation. 91

As to the accumulation of carbon dioxide in occupied rooms
to which atmosphere vitiation is commonly attributed, the late
A. R. Wolff, of New York, stated : "It is not the presence of the
carbon dioxide itself which causes injury, but the bad company
associated with its presence. The fact is that besides the car-
bon dioxide exhaled with the expired air there are also organic
matters and aqueous and other vapors, and at the same there are
given off from the pores of the skin organic secretions and mois-
ture, all of which taken together, and possibly acted upon and
made more detrimental in effect by the heat of the room, vitiate
the atmosphere and jointly are the sources of the trouble. . . .
They go hand in hand with the amount of carbon dioxide in
the room.

As to the degree of vitiation, the relative purity of the at-
mosphere is generally expressed in the number of parts by
volume of carbonic-acid gas contained in 10,000 parts or volumes
of air. The proportion of this gas contained in the atmosphere
may be easily determined by several methods, and it affords a
fairly good index of the relative number of micro-organisms
present and of the efficiency of ventilation."

In crowded rooms with the usual accompanying high tem-
peratures the water-vapor from the lungs and the perspiration
vapor soon saturate the air, and it is to this combination of
temperature and humidity that some writers attribute most of
the discomfort experienced. The author concurs in this view.

That moisture is present in crowded rooms in cold weather is
evidenced by the condensation on windows. When it is con-
sidered that this moisture is chiefly from exhalations from the
lungs and the elimination from the bodies of those present, it
would seem evident that such a component of the atmos-
phere must not only produce discomfort but be positively
harmful.

Macfie, in his work, "Air and Health," says : "Air containing
merely the carbon dioxide and moisture usually contained in
vitiated air will not produce the effect of vitiated air, therefore
must contain an additional constituent. This additional consti-
tuent, though undetected by chemists, is probably detected by
the nose for it is well known that air is oppresive and harmful,

92 Furnace Heating.

not so much in proportion to the amount of carbon dioxide and
moisture it contains as in proportion to its smelliness. The very
fact that the nose is so sensitive to such odors would seem to
suggest their harmfulness."

In addition to the carbonic-acid gas, the effluvia and the
humidity mentioned, which affect the comfort and well-being of
persons, are the dusts to which Dr. T. Mitchell Prudden's little
book, "Dust and Its Dangers," is devoted. Outer air contains,
of course, more or less dust which, when admitted to a building
tends to settle. Dr. Prudden observes "that even ordinarily
efficient systems of ventilation do not carry off any considerable
proportion of the dust particles from closed, still rooms, . . .
and that when, by a system of forced ventilation, we cause large
volumes of dust-laden air from out-of-doors to pass through
them, we are actually, so far as micro-organisms are concerned,
cleansing the air and sending it out much freer from germs than
when it entered, these having slowly settled as the air makes its
way from the entrance to the exit of the ventilating openings."
He says:

"When we consider the comportment of dust particles in
closed rooms, we see at once that the great renovating and
cleansing agencies which are so efficient out-of-doors are, except
on special occasions, absent, namely, the winds and strong air
currents and the more or less frequent and prolonged wettings.
. . . A rainfall to a certain extent tends to free the air of its
germs by washing them down. . . ."

Dr. Prudden points out that "we should always remember
that bacteria do not become detached from the surfaces or
materials on which they grow or are lodged while these are in
a moist condition." He remarks: "Ventilation is slowly becom-
ing recognized as important, but the removal of dust, which in
crowded places is very liable to be infectious, is not systematically
attended to." The most obvious means to prevent the accumu-
lation of dust within enclosures is to remove it from the entering
air.

We have briefly considered the "causes of atmospheric
vitiation." Now as to the effect on health, "the doctors
disagree."

Air, Humidity, and Ventilation. 93

EFFECTS OF FOUL AIR ON HEALTH AND COMFORT.

Billings observes in "Ventilation and Heating" that where any
room is occupied by human beings there is a definite, unpleasant
animal or musty odor, perceived by a person whose sense of
smell is of the usual acuteness and who enters from the fresh
outer air, the continued breathing of the air producing such
odor will be injurious to health."

The late Mrs. E. H. Richards of the Massachusetts Institute
of Technology states in her book, "Air, Water and Food," "That
a permanent or habitual lowering of oxygen in inspired air must
be harmful will be readily seen from a consideration of the
office of this gas in the body. (To Lavoisier and Laplace we owe
the knowledge that animal heat is derived from a process of
combustion. . . .)

"By the union of the oxygen with the substance found in the
tissues and brought to them by the circulating fluids of the body
from digested food, the heat necessary for the life and work of
the body is produced. This heat is needed to keep the tissues
at the temperature at which they can best accomplish their work,
to give mechanical power for the involuntary action of heart and
lungs for the process of assimilation and to furnish the energy
for all voluntary work and thought."

While the harmful effect of foul air may not be immediate
other than its effect on one's comfort or mental acuteness, it is
generally conceded that frequent and protracted exposure to
such air, as in the case of poorly ventilated school buildings,
results physiologically in a lowering of the vitality of the occu-
pants, rendering them more susceptible to disease and, considered
economically, results in a lessened efficiency on the part of both
pupils and teachers.

Playfair asserts that, in modern hygiene, "nothing is more
conclusively shown than the fact that vitiated atmospheres are
the most fruitful sources of disease."

Tuberculosis and pneumonia are most prevalent among per-
sons living or working in unventilated rooms. These diseases
are caused by specific bacteria, which for the most part gain
access to the air passages by adhering to particles which are
inhaled.

94 Furnace Heating.

Macfie, in his work, "Air and Health," says : "Any one who
compares his power of mental work in a pure and in a carbonic-
acid-laden atmosphere, even if the latter be dry and cool, will
find in the latter a considerable diminution.

He says: "Does such vitiated air as is ordinarily breathed
in human habitations cause ill health apart from the infectious
germs or infectious material it may contain? ... It is, of
course, almost universally believed nowadays that indoor air
rendered impure by respiration and combustion is harmful to
health. To bad air we attribute most of the anaemia, the pallor,
the neurasthenia, the general ill health of slum dwellers and
factory workers and most persons engaged in sedentary indoor
occupations."

As to the effect of dust, Dr. Prudden says :

"Very moderate amounts of dust particles in sensitive per-
sons cause such a degree of irritation of the respiratory organs
as either to deprive them of robust health or predispose them to
the acquirement of various diseases which with unirritated lungs
they would readily resist.

As to the bacteria . . . there are unfortunately a few
species which, when they once find lodgment in one place or
another in the organs of respiration, may grow and multiply,
and successfully resisting all the protective agencies of the body,
set up distinct and persistent and even fatal disease. Those
forms of bacteria which can, or in these regions commonly do
this, are insignificant in number in comparison with the harm-
less species with which dust is usually swarming. But few as
they are they have an extreme significance. If it were not for
these few species of disease-producing bacteria, most people
could perhaps afford to be as indifferent as they are to dust and
its dangers. . . ."

It has been pointed out among the causes of atmospheric
vitiation and discomfort that high temperature and humidity
have much to do with the oppressiveness of the atmosphere in
occupied spaces. In this connection Dr. Henry Mitchell Smith
of Brooklyn, N. Y., in a paper read before the Brooklyn Medical
Society, says:

"Records of the temperature in a large number of houses

Air, Humidity, and Ventilation. 95

showed . . . that it commonly ranged from 72 to 76°, and at
times, in very cold weather, 78° F. was recorded. Nevertheless
. . . rooms felt chilly when the recorded temperature in-
dicated that they were far too hot. It was often hard to believe
that the temperature was above 68° when it was actually 72°
and 74°.

It was at once apparent that some unrecognized factor was
responsible for this discrepancy between the temperature recorded
by the thermometer and one's sensations. Moreover, it was
found that the colder the weather the higher was the average
temperature maintained indoors. The reason for this is the in-
sufficient amount of moisture in our rooms in proportion to the
temperature (low relative humidity). The colder the weather
the lower will be the indoor relative humidity.

"The point to be emphasized is that every time we step out
of our houses during the winter season we pass from an at-
mosphere with a relative humidity of about 30 per cent, into one
with a relative humidity of an average of 70 per cent. Such a
sharp and violent contrast must be productive of harm, par-
ticularly to the delicate mucous membranes of the upper air
passages. Watery vapor what we term moisture, is as much
a part of the air as is oxygen ; absolutely dry air does not exist
in nature.

The skin and mucous membranes of the respiratory passages
are the principal sufferers, since these tissues are always kept
moist with their own secretions; and from them water is freely
abstracted to satisfy this large saturation deficit.

A moment's consideration shows that the prevailing prac-
tice of depending upon the thermometer as the sole guide in the
heating of buildings is not only inadequate and unscientific, but
it is often misleading. It is not sufficient to know only the tem-
perature if we desire either comfort or health, for the same tem-
peratures produce varying sensations of warmth or cold, de-
pending upon the relative humidity at the time existing.

It is unscientific and arbitrary to lay down a fixed tempera-
ture as a standard for living or sleeping rooms unless the rela-
tive humidity is indicated as well.

"Records from steam-heated apartments showed that the

96 Furnace Heating.

relative humidity was sometimes as low as 25 per cent., with a
temperature of 78° during a period of very cold weather. The
high temperature is necessitated by the chilling of the body by
the increased evaporation, evaporation being essentially a cooling
process."

"Thermostatic temperature control will not fill the require-
ment, for a constant temperature is constant in its effect only
if accompanied by a constant relative humidity. Moreover, prop-
erly moistened indoor atmosphere lacks all the oppressive dry
feeling so characteristic of the average artificially heated room.
The quieting effect of such an atmosphere is striking."

It was satisfactorily proved that one may live during the cold-
est weather with perfect comfort in a room at 65° F. where the
relative humidity is kept at about 60 per cent. During the experi-
ments upon the sensations produced by different percentages of
saturation, and in order to obtain the opinion of persons having
no knowledge of the existing conditions, one room was equipped
with a moistening apparatus and the temperature kept at 65° to
68°, with a relative humidity of about 60 per cent. ,An adjoin-
ing room, without a moistening apparatus and heated by an ordi-
nary steam radiator, had an average temperature of 72° to 74°,
with a relative humidity of 30 per cent. In every instance and
without at all knowing what the temperatures were in the two
rooms, the opinion was unhesitatingly expressed and the first
room was several degrees warmer than the second.

It is inconceivable that with otherwise perfect means of heat-
ing, provision for producing sufficient mosture to maintain a
higher relative humidity should have been so disregarded in all
but those elaborate systems applicable only to large halls and
public buildings.

As to desirable and practical relative humidities in rooms occu-
pied in winter by persons in health, taking into consideration the
cost of maintaining a high relative humidity in cold weather and
the trouble from condensation on windows. The author is in-
clined to favor a range from 40 to 50 per cent., according to the
weather, rather than the higher relative humidity mentioned in
Dr. Smith's paper, viz., 60 per cent. As to condensation on win-
dows, this will occur during cold weather when the indoor rela-

Air, Humidity, and Ventilation. 97

tive humidity is 40 per cent., and even somewhat less. When
double windows are used, as is common in northern latitudes,
there is little or no trouble from condensation.

The improved physical conditions of teachers and pupils in
moving from inadequately ventilated school buildings to those
equipped' with modern and efficient systems is a well-known and
admitted fact. Scientific tests have been conducted which have
proved these facts very conclusively.

In regard to dwellings, even though there be no method of
ventilation provided, the mere abundance of space per occupant
secures a certain air change, owing to the fact that no partitions,
floors or ceilings are perfectly tight, hence the greater the space
per occupant the greater the surface of surrounding walls, etc.,
and the greater the accidental air leakage, or spontaneous venti-
lation, as some put it.

As an example showing the results of improved ventilation, a
paper by Prof. C.-E. A. Winslow calls attention to the operating
room of the New England Telephone & Telegraph Co., at Cam-
bridge, Mass., a long room having a capacity of 30,000 cu. ft.,
extending from front to back of a business block. Fifty or sixty
women are employed in this room as operators. During the
warmer months no difficulty has ever been experienced in ven-
tilating the room by means of large windows at each end and by
the use of electric fans. In winter, however, it was impossible
to secure adequate natural ventilation without undue exposure
to drafts.

In the spring of 1907 a simple but efficient system of artificial
ventilation was installed. ... A marked improvement in the
comfort and general condition of the operators followed this
change and the betterment was sufficiently marked to show itself
in a notably greater regularity of work.

Statistics collected and tabulated show that prior to the in-
stallation of the ventilating system for the three winter months,
January, February and March, inclusive, 4.9 per cent, of the
force was absent in 1906 and 4.5 per cent, in 1907. With the
ventilating system in use the absence for the same months in 1908
fell to only 1.9 per cent., a striking reduction.

98 Furnace Heating.

NECESSITY FOR VENTILATION.

Having discussed the "causes of atmospheric vitiation" in
occupied spaces and the "effects of foul air on health and com-
fort" it would appear that the necessity of ventilation is. obvious.

Perhaps nothing has focussed the attention of the general
public on the necessity of fresh air so much as the crusade now
being waged against tuberculosis. Dr. Woods Hutchinson, in
his book "Preventable Diseases," brings out in a most vivid man-
ner the wonderful changes wrought in the prevention and treat-
ment of this disease. He says : "Fifty years ago belief was that
•consumption and all its attendant miseries were chiefly due to
exposure to cold. Now we know that, on the contrary, abundance
of pure, fresh air is the best cure for the disease, and foul air
and overcrowding is its chief cause. An almost equally complete
aboutface has been executed in regard to pneumonia.

"This much we are certain of already: that the majority of
so-called 'colds' have little or nothing to do with exposure to a
low temperature, that they are entirely misnamed, and that a
better term for them would be 'fouls'. . . . The best place
to catch them is not out of doors, or even in drafty hallways, but
in close, stuffy, infected hotel bedrooms, sleeping cars, churches
and theatres.

"The frequency of colds in winter is chiefly due to the fact
that, at this time of year, we crowd into houses and rooms, shut-
ting the doors and windows in order to keep warm, and thus
provide a ready-made hothouse for the cultivation and transmis-
sion from one to another of the influenza and other bacilli.

"At the same time, we take less exercise and sit far less in
the open air, thus lowering our general vigor and resisting power
and making us more susceptible to attack. Those who live out-
of-doors, winter and summer, and who ventilate their houses
properly even in cold weather, suffer comparatively little more
from colds in the winter time than they do in the summer."

Dr. Hutchinson advises "living and sleeping as much as pos-
sible in the open air. This helps in several different ways : first,
by increasing the vigor and resisting power of our bodies ; second,
by helping to burn up clean and rid our tissues of waste products

Air, Humidity, and Ventilation. 99

which are poisons if retained ; third, by greatly reducing the risks
of infection."

He advises us to learn to sit or sleep in a gentle current of
air all the time we are indoors.

Macfie, in his book, observes : "All the writers on ventilation
assume that ventilation which causes any perceptible motion of
cool air is not permissible. But why? Simply because the un-
natural habits of so-called civilized peoples render them unduly
sensitive to draughts ; and, through erroneous reasoning, cold
air and draughts are considered dangerous."

On the other hand Billings says : "We may write and talk as
much as we please about the horrors of bad air and the im-
portance of good ventilation, but we shall never induce people
to sit in cold draughts and shiver for the sake of pure air."

If the people at large could be educated to sit in perceptible
currents of warm air — not cold draughts by any means — the
work of heating and ventilating engineers would be much sim-
plified, for one of the limiting conditions in the ventilation of
rooms is the absence of perceptible draughts which is insisted
upon by the occupants and required by compulsory ventilation
laws. With the opinion at present held by people as to sitting
in draughts, it would be useless to expect that a ventilating
system involving perceptible draughts would continue to be oper-
ated in any building where persons have to sit for any length of
time. It is not pleasant to consider that in crowded unventilated
rooms the air must be rebreathed, nor is it pleasant to consider
the other causes of atmospheric vitiation within enclosures as
pointed out.

To keep the atmosphere of an occupied room wholesome, a
frequent change of air must be secured; if the space is so
large, the number of occupants so few and the air leakage through
walls or around windows and doors such that this accidental
ventilation is sufficient, well and good. This will, however, suf-
fice only in rare instances ; some dependable means must in most
cases be provided to furnish a minimum volume of fresh air per
minute for each occupant to meet present day standards.

In conclusion, adequate ventilation should be considered a
necessity in spite of the increased cost over heating only. It

ioo Furnace Heating.

contributes to health, efficiency and happiness by making us more
vigorous, keeping our bodies in a condition capable of warding
off disease.

In the foregoing the author has drawn freely from a paper
on "Ventilation in its Relation to Health," which he presented at
Cornell University in 1910.

STANDARDS OF VENTILATION.

Ventilation may be considered good, when measured on the
carbon dioxide basis, when the number of parts of CO2 in an
occupied room does not exceed from 6 to 7 parts in 10,000.
With 8 parts the air appears close to one entering from out-of-
doors. When the CO2 exceeds 10 parts in 10,000 the quality
of the air is noticeably bad, and produces a feeling of weariness
in a person breathing it for some time. While the CO2 basis
falls far short of meeting all requirements it is still commonly
used in lieu of a more comprehensive standard.

Air Supply Necessary. — The volume of fresh air that must be
supplied to keep the air in the room at a certain degree of purity
may be readily computed. For example: What volume of air
must be supplied to an occupied room to prevent the CO2 from
exceeding 7 parts in 10,000? Taking as a basis the commonly
accepted figure of 0.6 cubic foot as the amount of CC>2 given off
per person per hour, and 4 parts or cubic feet in 10,000 as the
proportion of CCh in the outside air; the fresh air admitted
absorbs 3 parts to reach the standard of 7 parts allowed, 3 cubic
feet of CO2 is taken up, which is equal to that given off by

3 -i- 0.6 = 5 persons. That is, 10,000 cubic feet of air containing

4 parts CO2 must be admitted per hour to 5 persons, or 2000
cubic feet per hour per person in order that the number of parts
of CO2 in 10,000 shall not exceed 7.

By similar computations, ' 6000 cubic feet per hour per per-
son will be found necessary to dilute the air to 5 parts of CO2 in
10,000 parts, 3000 cubic feet to dilute it to 6 parts, 1800 cubic
feet for 7.33 parts, 1500 cubic feet for 8 parts, and so on.

Where gas-lights are used, an additional supply of air must
be provided, since the vitiation of air caused by each jet is as
great as that caused by five or six persons.

Air, Humidity, and Ventilation. 101

Table XV a. — The Air Supply Commonly Accepted as Sufficient for Different Classes

of Buildings.

Class of Minimum, cu. ft. per hour Minimum, cu. ft. per min.

Building. per occupant. per occupant.

Hospitals 2400 40

Halls 1200 20

Churches 1200 20

Schools 1800 30

The volume of air which should be furnished for ventilation
should not be based solely on the number of occupants in the
rooms. The smaller the space per person the less must be the
supply per person to avoid draughts. Thirty cubic feet per min-
ute per occupant for example in a hall having only 100 cubic feet
of space per occupant would mean a change of air every 3^/3
minutes as against a 6 or 7 minute air change in a school room.

COMPULSORY VENTILATION.

Massachusetts was the pioneer in the matter of compulsory
ventilation. In this State the following requirements must be
included in the specifications accompanying plans for the ventila-
tion of school buildings submitted to the Department for ap-
proval.

1. The apparatus must, with proper management, heat all the
rooms, including the corridors, to 70° F. in any weather.

2. With the rooms at 70° F. and a difference of not less than
40° between the temperature of the outside air and that of the
air entering the room at the warm-air inlet, the apparatus must
supply at least 30 cubic feet of air per minute for each scholar
accommodated in the rooms.

3. Such supply of air must so circulate in the rooms that no
uncomfortable draught will be felt, and that the difference in tem-
perature between any two points on the breathing plane in the
occupied portion of a room will not exceed 3°.

4. Vitiated air in amount equal to the supply from the inlets
must be removed through the vent outlets.

5. The sanitary appliances must be so ventilated that no odors
therefrom will be perceived in any portion of the building.

CHAPTER V. .

THE HEATING AND VENTILATION OF SCHOOL
BUILDINGS.

GENERAL DISCUSSION.

For school buildings of suitable size the furnace system is
simple, convenient and generally effective. Its use is confined as
a rule to buildings having not more than eight rooms. For large
ones it must generally give way to some form of indirect steam
apparatus with one or two boilers, which occupy less space and
are more easily cared for than a number of furnaces scattered
about. Like all systems that depend on natural circulation un-
aided by fans the supply and removal of air is considerably af-
fected by changes in the outside temperature and by winds.

RELATIVE FUEL CONSUMPTION.

In small school buildings heated by furnaces the fuel con-
sumption per room is greater as a rule than in larger ones warmed
by other methods. This is not attributable, however, to a low
furnace efficiency so much as to other causes — viz. : The air sup-
ply is affected by winds to a greater extent than in large build-
ings in which the supply is governed by the speed of a fan. It
thus frequently happens that a greater quantity is driven through
the furnaces than is necessary for the proper ventilation of the
building. This involves a waste of heat. Small buildings have
a greater exposure in proportion to their cubic contents than
larger ones, hence their loss of heat by transmission is correspond-
ingly greater. The janitor service in such buildings is less effi-
cient and less skillful firing the rule.

THE FURNACE.

The furnaces used are generally built of cast iron, this ma-
terial being durable and easily made to present large and effective
heating surfaces. Several forms of furnaces have been designed
especially for this service.

102

The Heating and Ventilation of School Buildings. 103

SCHOOL HOUSE HEATERS.

Fig. 47 shows a cast iron furnace designed especially for
school house heating. Such furnaces are commonly used to heat
two standard class rooms each.

The makers give these dimensions in their description of this
furnace: Fire pot 34 inches inside diameter, 16 inches deep, over
8 cubic feet fuel capacity.

Fig. 47.— School House Heaters.

Note the corrugated fire pot and the combustion chamber
with outlets around the circumference leading to the radiator at
the top.

Such furnaces are rather high, but the basements of modern
school buildings are of sufficient height to receive them; further-
more the furnace is placed almost directly under the flues so
that the height does not affect the pitch of the pipes.

Fig. 48 shows a furnace with cast iron fire pot steel dome and
steel or wrought iron radiator.

The makers state that these are made with fire pot 28 inches
and 31 inches diameter.

Openings are provided in the front for access to the interior

IOA

Furnace Heating.

for cleaning purposes. It is of considerable importance to have
these cleanout doors easily accessible, as otherwise the cleaning
will be neglected.

Fig. 48. — School House Heater of Steel-plate Construction.
AIR PASSAGE IN FURNACE.

To adapt the larger sizes of house heating furnaces to schools
a much larger space must be provided between the body and the
casing to permit a sufficient volume of air to pass to the rooms.
The free area of the air passage should be sufficient to allow the
air to pass through with a velocity not greater than 400 feet per
minute.

PORTABLE OR BRICK SETTING.

A galvanized iron casing is generally used in connection with
galvanized flues in buildings having wooden partitions. In brick
buildings the furnace setting and the flues are generally built of
that material.

The Heating and Ventilation of School Buildings. 105

SIZE OF FURNACE.

The size of the furnace is based on the loss of heat through
the walls plus that carried away by the air passing up the ven-
tilating flues or leaking out through other openings. Losses
through exposed floors and ceilings must also be included.

Assuming that a single furnace heats two rooms, which is
common practice, we should proceed to calculate the loss of heat
by transmission as follows: Suppose the school rooms to be of
average size, 28 x 32 x 12 feet, and to have 140 square feet of
glass, this amount being the average of a number of measure-
ments taken by the writer. Reduce the wood and plaster
or bricks walls to equivalent glass surface by dividing tbeir
area by 4. Reduce the floor or ceiling to equivalent glass
surface by dividing the area by 20 or 25, according to the con-
ditions. Add these equivalents to the area of glass in the
windows. The equivalent glass surface of the walls is equal to

(28 + 32) X 12 — 140 square feet r ^u

- = 145 square feet. The equiv-

4
alent glass surface of the floor or ceiling equals 28 X 32 — 44 g

square feet. Adding to these items the actual glass surface in
windows gives a total of 145 + 44.8 + 140 =- 330 square feet
approximately. Multiply this sum by 85 (the number of heat
units transmitted per hour per square foot of glass with tem-
perature of 70 degrees inside and o degree outside). The prod-
uct is the total loss of heat per hour by transmission. In this case
33° X 85 = 28,050 heat units, or for two rooms 56,100 heat
units.

To this must be added the heat carried up the ventilating flues
or lost by leakage. Assuming each of the two rooms to contain
50 occupants who are each supplied with 30 cubic feet of air per
minute, we have for the volume of air passing through the two
rooms per hour 2 (50 X 30 X 60) == 180,000.

Each cubic foot of air escaping at 70 degrees temperature
with the outside air at o degree carries away ij4 heat units, hence
the loss by ventilation is equal to 180,000 X 1/4 — 225,000 heat
units per hour. Adding to this the loss of heat by transmission

io6 Furnace Heating.

gives 281,100 as the total loss of heat per hour from the two
rooms. The furnace must be capable of imparting to the air
passing through it an equal amount.

With the more regular and skillful attendance it is safe to as-
sume a higher rate of combustion in school house heaters than in
those used in residences. Assume therefore a maximum rate of 6
pounds of coal burned per square foot of grate surface per hour.
The air passing over the heating surface is much greater in vol-
ume and lower in temperature than in house furnaces, therefore
we should expect even with the more rapid rate of combustion, to
obtain about the same efficiency as in the latter. With a large
volume of air passing through the furnace the average tempera-
ture of this air will be lower than in residence furnaces where
the quantity of air is smaller with the same size of furnace.
The transmission of heat and also the efficiency will therefore
be slightly higher.

Granting, then, that Sooo heat units per pound of coal burned
will be taken up by the air passing through the furnace we have
6 X 8000 = 48,000 heat units utilized per hour per square foot
of grate surface or average fire pot area. Hence to ascertain the
requisite grate area simply divide 281,100, the total loss of heat
per hour, by 48,000. The quotient is 5.86 square feet, equaling
about a 32-inch fire pot.

In determining the size of furnace required to heat rooms on
the more exposed sides of buildings, add as a factor of safety
10 to 20 per cent, to the estimated loss of heat as above computed.
It has been found in practice that furnaces with a 32-inch to 34-
inch fire pot and ample heating surface will heat two ordinary 50-
pupil rooms to 70 degrees in zero weather.

CORRIDOR HEATER.

Corridors may best be heated by a separate furnace. If it is
attempted to warm them from a furnace connected with the
schoolrooms the flow of air will be very uncertain and unsatis-
factory, since it tends to pass directly up the large vertical flues.
The size of the corridor furnace may be based on the exposure
according to Table II. A slight allowance should be added, how-
ever, to compensate for the cooling effect of outside doors at the

The Heating and Ventilation oj School Buildings. 107

beginning of sessions. Corridor registers should be set in the
floor to serve as foot warmers.

LOCATION OF FURNACE.

The furnaces, as in Fig. 49, should be located as nearly as pos-
sible under the flues with which they are connected to lessen the
resistance and loss of heat and to facilitate the arrangement of
mixing dampers. A pit at least 2 feet deep should be provided
under each furnace to permit an even distribution of the air over
the heating surface.

INLET ho gq Ft. INLETJlOSq. Ft

Fig. 49.— Basement Plan (9' 6") of School Building Heated by a Furnace System.
COLD AIR ROOM.

In school buildings a cold air room is far preferable to the
ordinary box. The flow of air is more regular and the resistance
to its passage to the furnace is reduced to a minimum. Less at-
tention need be paid to the location of the cold air inlet with
reference to the points of the compass than when an ordinary air
box is used. With large inlet and flues rooms can be successfully
heated when taking air from the lee side of the building. Port-
able furnaces are sometimes placed within cold air rooms. In
such cases they must be double cased throughout and the flues

io8

Furnace Heating.

leading from them be thoroughly protected with non-conducting
material to reduce the loss of heat.

FRESH AIR SUPPLY.

The net area of the cold air inlet should nearly equal the aggre-
gate area of the flues leading from the furnace. An inlet of gen-
erous size is especially important during mild weather, when the
air is heated and expanded but little and consequently has but
slight force as compared with zero weather conditions. A swing-
ing damper or slide should be used to regulate the flow of air
during winds and to shut it off at night. To work properly and
economically the furnace must have an adequate supply of air at

SCHOOL ROOM
ts'xSt'xll'

CO PUPILS, AIR SUPPLY 1500 CU. FT. PER MIN.

WIRE SCREEN
j^uS"*

SCHOOL ROOM
\ 28 x 32 'x 12'

50 PUPILS, AIR SUPPLY 1500 CU. FT. PER MIN.

Fig. 50.— First-Floor Plan (12' 0") of School Building Heated by a Furnace System,

all times. An oversupply during winds will be likely to occur
unless the inlet damper is intelligently managed.

RETURN AIR OPENINGS.

A duct or opening for returning air from the rooms to the
furnace, as shown in Fig. 50, should be provided for use while
the building is unoccupied, when the air supply for the furnaces
may be taken from indoors without harm and with economy in
fuel.

The Heating and Ventilation of School Buildings. 109

When cold air boxes are used they should be built of galva-
nized iron or brick. The building laws in many places prohibit
the use of wooden ones in public buildings.

Fig 51. -Sectional View of Furnaces on Line A A of Fig. 28.
MIXING DAMPERS.

At the base of each warm air flue is placed a mixing damper,
operated by a chain from the schoolroom above, as shown in Fig.
51. By means of this damper the teacher may regulate the tem-
perature of the room at will without seriously affecting the vol-
ume of air delivered, since the damper, in cutting off the supply
of warm air, simultaneously opens an equal area for the inflow

Furnace Heating.

of cold air, and vice versa. The damper should be arranged so
that the cold air will pass up at the rear of the flue and out at the
top of the warm air opening in the room. If allowed to pass up
the front of the flue the cold air is likely to descend on the heads
of the pupils. Cold air should enter the flue from below the mix-
ing damper. The weight of the damper will then keep it tightly
shut. If closed by pulling up on the chain, unless the latter be
drawn up perfectly taut, leakage of cold air will be likely to occur.

LOCATION OF FLUES.

•

The proper location of fresh air and ventilating openings to
secure the most thorough distribution throughout the room is a
matter that should be most carefully studied in laying out the sys-
tem. The locations which have been found to give good results
in practice, with rooms having exposures as indicated, are shown
in Figs. 52, 53 and 54.

INSIDE WALL

i i

INSIDE WALL

EXPOSED WALL
Fig. 52.

EXPOSED WALL
Fig. 53.

INSIDE WALL
Fig. 54.

Both warm air and ventilating flues are located along inside
walls. The entering air is discharged through an opening 3 or 4
feet below the ceiling, toward or along the cold outside walls. The
chilling effect of the latter causes the air to descend, to be drawn
across the seating space to the ventilating opening in or near the
floor.

When it is impossible to arrange flues in the desired positions
the air from the inlet may be directed to any part of the room by
deflectors or diffusers placed in front of the openings.

MATERIAL OF FLUES.

The flues are generally built of galvanized iron, No. 24 gauge
being commonly used, or of brick, with the inner surface smooth-

The Heating and Ventilation of School Buildings.

ly plastered. In some respects galvanized iron is superior, being
smoother and absorbing less heat while the building is being

IRON GRATING
FORRETURNIG
AIR AT NIGHT

SCHOOL ROOM
88x32x12

60 PUPILS
AIR-SUPPLY1500CU.FT.PER.MIN,

W|RE

30x34"

GALV. IRON DlVII

OPPOSITE
OPE.N
ON 20 FLOOR

8x1 C

SCHOOL ROOM

/VENT SHAFT 28x32x12

/OFFSETS IN ATTIC 50PUPIL8

TOCFNTHE LINE
FTOF BUILDING AIR SUPPLY

!5 WIRE SCREEN 1500 Cu.Ft.Per.Min.

* 30x34"
TOILET VENT12xio"
8x16"

Fig. 55.— Second-Floor Plan (12' 0") of a School Building Heated by a Furnace System.

warmed. On the other hand, brick ventilating flues absorb rain
that may be driven in, and can therefore be left open at the top
without any hood. Those of galvanized iron require a hood for
protection during storms to keep out the rain.

HOOD ABOVE FLUES.

The hood must extend far enough beyond the flue on all sides
to prevent rain beating in even when descending at an angle of 45
degrees. Louvers or slats are often used for further protection.

The area of the flue, divided by the combined length of two
sides, gives the proper clear hight between the top of the flue and
the under side of the hood.

AREA OF FLUES.

The warm air flues rise from the furnace to a height of 9 or 10
feet above the floor of the schoolrooms, discharging through a
wire screen or grill. The area of the flues is generally based on
a velocity of about 300 feet- per minute in those leading to the
first or second floors.

112

Furnace Heating.

In determining the size of flues from Table XIV, Chapter IV,
it is well to reckon on a difference in temperature between the air

Fig. 56.-Section through Ventilating Shaft on Line B B of Fig. 28, Showing Stack Heater

in the flue and that out of doors not greater than 40 degrees. A
flue based on the maximum difference in temperature existing in
zero weather will be altogether too small to provide the requisite
volume of air in mild weather. Theoretically, the greater the
hight of the flue the smaller the area required. Practically it is
often convenient to make the flues to both the first and second

The Heating and Ventilation of School Buildings. 113

floors the same size, generally about 24 x 30 inches for 5o-pupil
rooms requiring 1500 cubic feet of air per minute. One square
foot to every 10 pupils.

The tendency of the air to flow more readily to the upper
rooms and overheat them is counteracted by the mixing damper,
which cools the air in the flue and consequently diminishes the
velocity with which it ascends. Adjustable dampers may be
used in addition.

VENTILATING FLUE DAMPERS.

Dampers should be placed in ventilating flues to prevent the
escape of warm air at night and to regulate the discharge in severe
or windy weather, when over-ventilation is likely to occur. The
latter is accompanied by excessive inward leakage of cold air
around windows, causing chilly drafts.

REGISTERS AND SCREENS.

Wire screens or grills of open pattern are preferable to regis-
ters for school house work on account of the greater freedom they
afford to the passage of air. They are often made of ^-inch
wire, i ^4 -inch mesh, which gives a net opening equivalent to about
80 per cent, of the gross area. The frames are usually con-
structed of i" x i" x Yg" angle irons with holes drilled in one
leg of the angle to receive the wires. The other leg of the
angle is drilled for screws which attach the frame to the
wall.

To provide for the easy discharge of air the net area of wire
screens or register faces should be somewhat in excess of the area
of the flue.

In ordinary 5O-pupil rooms a wire screen of open mesh pattern
at least 30 x 30 inches, or a register not smaller than 30 x 36
inches, should be used for the warm air inlet. The ventilating
openings in or near the floor should, if possible, have an area
slightly in excess of that of the fresh air inlet.

The draft is so strong at the ventilating openings located in the
first floor of a building having two or more stories that a register
27 x 38 inches is generally large enough for a 5o-pupil room
located on that floor.

ii4 Furnace Heating.

STACK HEATERS.

It is customary to group the ventilating flues together in a
main stack or shaft, at the bottom of which is placed a stack heater
consisting of a small furnace or stove. The function of the latter
is to maintain, during mild weather, a sufficient excess of tem-
perature in the shaft to secure the requisite removal of air from
the rooms. Cast iron stack heaters are the most serviceable and
are most commonly used. The ordinary heating stove as applied
to this service is accessible only through a large door placed in the
side of the vent flue. This door often fits loosely, allowing an in-
ward leakage of cold air, thereby diminishing the effect of the flue.
Furthermore, the stove is so unhandy to care for that it is likely
to be neglected by the janitor.

A small furnace is much better adapted in every way to this
work.

SIZE OF STACK HEATER.

The size of the stack heater is governed by the hight and area
of the ventilating shaft and the volume of air to be discharged in
a given time.

The hight is generally but a few feet greater than the topmost
point of the roof, the area but little in excess of the combined area
of the 24 x 30 inch warm air flues, and the volume equivalent to
about 1800 cubic feet of air per hour per occupant. With such
conditions in the ordinary twro-story building a difference of near-
ly 20 degrees between the temperature of the air in the flue and
that out of doors will be required to produce the desired velocity
and air removal. That is, whenever the outside temperature rises
above 50 degrees, for example, a fire must be maintained in the
stack heater, its intensity to be increased as the outside tempera-
ture rises, in order to maintain an excess of temperature of 20
degrees in the shaft.

It is assumed that whenever the outside air closely approaches
the normal temperature of the room, windows will be thrown
open and an abundant circulation secured in that manner, thus
dispensing with the use of the stack heater. As a matter of
fact, the small stoves usually employed for this service are utterly
inadequate.

The Heating and Ventilation of School Buildings. 115

In ordinary two-story school buildings a stack heater having
y2 to 24 square foot of grate surface per standard 5o-pupil room
will maintain a nearly constant removal of air from the rooms
until a point is reached when all fires may be dispensed with and
windows opened without discomfort.

ARRANGEMENT OF STACK HEATER.

It is unquestionably best to bring the vitiated air into the ven-
tilating shaft — see Figs. 49 and 56 — below the stack heater. Ow-
ing to the lack of space and the increased cost of building drop
flues this arrangement is seldom carried out in ventilating rooms
above the first floor.

On the second floor and above the ventilating openings gener-
ally connect directly with the shaft. A curved damper hinged at
the bottom (see Fig. 56) and adjusted by a chain is used at such
openings. The ascending currents from below, passing rapidly
by the edge of this damper, tend to create a suction through the
ventilating openings. This, combined with the natural tendency
of the air to flow into and up the flue, is sufficient, as a rule, to
secure the desired removal of air from upper rooms. Also the
inflow of air from the warm air flues is usually more rapid than
in the first floor rooms and this tends to increase the outflow of
vitiated air from the vent flues.

BOILER WITH COILS IN VENTILATING FLUES.

In large school buildings heated by furnaces, to avoid the
bother of maintaining a fire in several stack heaters a small steam
boiler is sometimes used to supply coils placed in the ventilating
flues just above the openings from the rooms.

About 20 square feet of heating surface is generally allowed
for each ventilating flue from a 5o-pupil room, but with this small
amount the volume of air removed per minute will fall off rapidly
as the outside temperature approaches 70 degrees.

Steam is condensed so much more rapidly in coils thus placed
than in ordinary direct radiators that the actual heating surface
in the ventilating flues should be at least 2.5 to give the proper
boiler rating expressed in square feet of ordinary cast iron direct
radiation.

CHAPTER VI.

HEATING OF PUBLIC BUILDINGS, CHURCHES
AND STORES.

IN GENERAL.

Several features commend the furnace system of heating and
ventilation when properly applied in public buildings and
churches. The apparatus is the simplest of all and is comparative-
ly inexpensive. Heat may be generated quickly and when no
longer needed the fires may be allowed to go out without danger
of damage to any part of the system from freezing. When prop-
erly proportioned an air supply sufficient for ordinary require-
ments may be secured. Without further description a good idea
of such a system can be gained from the plans given of a town
hall, Fig. 57 showing the basement and Figs. 58 and 59 the first
and second floors, while Figs. 60 and 6 1 are details and sections.

In buildings similar to those illustrated in this chapter, in
which all the rooms are rarely used at the same time and are prac-
tically never fully occupied simultaneously, it is common prac-
tice to install an apparatus with switch dampers to direct the hot
air into either of the principal rooms or to divide it between them.

It is not necessary that an apparatus so arranged should be
large enough to heat the entire building to 70 degrees with a fre-
quent change of air. (Table XVII shows that the grate surface
necessary to heat 150,000 cubic feet of space with a 15-minute air
change will heat 250,000 cubic feet with a 3O-minute change.)

If the building is thoroughly warmed before occupancy, either
by rotation or by a slow movement of air, the chapel or Sunday
school in the case of a church may be shut off until near the close
of the service in the auditorium, when a portion of the warm air
may be diverted to it. When the service ends the switch damper is
thrown over and all the air is discharged to the Sunday school.
The mixing damper will prevent overheating.

SIZE OF FURNACE.

To determine the size of the furnace first reduce the entire ex-
posed wall to equivalent glass surface (E. G. S.) by adding to the

116

Heating of Public Buildings, Churches and Stores. 117

actual amount of glass one-fourth the area of solid walls. With
a non-heated attic reduce the ceiling to equivalent glass surface by
dividing its area by 20.

When there is no attic space and the room to be heated extends
to the roof, divide the roof area by 10, instead of 20, to obtain its
equivalent glass surface. Fig. 57 shows basement plan of a town
hall, while the first and second floor plans are shown in Figs. 58
and 59. Details and sections are shown in Figs. 60 and 61.

The basement is generally so warm that the loss of heat

COMBINED AflEA WARM AIR PIPE6944 SQ.IN.
EXPOSURE OF ROOMS WARMED, ABOUT 3100 SQ.FT.
EXPOSURE RATING OF FURNACE, 3500 SQ.FT. SEE TABLE II

Fig. 57.— Furnace System of Heating and Ventilating a Town Hall.— Basement Plan.

through the first floor may be neglected ; otherwise, divide its area
by 20 or 25, according to its construction, to reduce to equivalent
glass surface.

Having determined the equivalent glass surface multiply it by
85 (the loss in heat units per hour per square foot of glass with 70
degrees inside, o degrees outside). The prod act is the total num-
ber of heat units lost per hour by transmission. Add 5 to 10 per
cent, when the building is severely exposed.

To this must be added the loss of heat per hour by the escape of
air. Basing the air supply on the common allowance of 1000 cubic

Furnace Heating.

feet per hour per occupant, as stated in Chapter IV, we have:
Number of occupants multiplied by 1000 equals volume of air re-
quired per hour.

In case the seating capacity is unknown a change of air every
15 or 20 minutes may be assumed, or even a 3O-minute change
when the space per occupant is unusually large or the require-
ments not at all exacting. Since ij4 heat units are removed by
each cubic foot of air escaping at 70 degrees temperature in zero
weather, to ascertain the total loss of heat by ventilation multiply

r- . , H.irr.r>2 sqift*

pyw i»

J_ VJl/R^j-^

Fig. 58.— First-Floor Plan.

the volume of air removed per hour by i J4 \ add this to the loss by
transmission and the sum gives the total loss per hour, or T + V
= Q. When the heating is intermittent, unless provision is made
for returning the air to the furnace, add 10 to 15 per cent.

To determine the size of the furnace simply divide the total
loss of heat per hour from the building by the heat given to the air
passing through the furnace per square foot of grate. Assuming
a rate of combustion of 5 pounds of coal per square foot per hour
and 8000 heat units utilized per pound of coal burned, we have

Heating of Public Buildings, Churches and Stores. 119

5 X 8000 = 40,000 heat units per square foot of grate per hour.

Hence

= G S = average area of fire pot in square feet.

40,000

ANOTHER METHOD TO DETERMINE SIZE OF FURNACE.

When the walls are of greater thickness than 12 to 16 inches,
or where greater accuracy is desired than is obtained by using

Air Supply GOOO Cu.Ft.Per.Min. I >.

GRATING IN FLOOR

' 12

=B|

II a 'i 10"

14 COAT ROOM

10 x ic'x 9'

Fig. 59.— Second-Floor Plan.

the above approximate method, the values prescribed by A. R.
Wolff may be employed.

Table XVI— The L"ss of Heat By Transmission with a Difference of 70
Degrees Between the Indoor Temperature and that Outside.

The loss in heat units per square foot per hour by transmission for—

(A)

8-inch brick wall

12-inch brick wall

16-inch brick wall

20-inch brick wall .

24-inch brick wall

Single window

Ceiling (unheated attic)

Floor (unheated basement)

=<!!

22
18
16
14
85
5
4

For other differences than 70 degrees between the inside and
outside temperatures the loss of heat is increased or decreased pro-

120

Furnace Heating.

portionally. In using the above table simply multiply the wall area
of a given thickness by the corresponding figures in column B.
Add to this the loss of heat through the windows and that through

Fig. 60.— Sectional Elevation of North Furnace.

the floor or ceiling, then add about 10 per cent, to allow for winds.
The sum is the total heat transmitted per hour, to which must be
added the loss by ventilation, calculated as just explained.

Heating of Public Buildings, Churches and Stores. 121

Dividing the combined losses by transmission and ventilation
by 40,000 gives the grate surface in square feet, which is to be in-
creased, as previously stated, when the apparatus is fro be used
intermittently.

AN APPROXIMATE METHOD TO DETERMINE SIZE OF FURNACE.

It frequently happens that sufficient data are lacking to pursue
either of the methods of calculation just described. In such cases

W////////M

Fig. 61.— Section at A A, Showing Stack Heater and Ventilating Shaft

Table XVII will be found useful. This table is based on the loss
of heat by transmission plus that by leakage or escape of air from
buildings having an average glass surface. The combined loss of
heat divided by 40,000 gives the grate surface or average fire pot
area in square feet stated in the table.

122 Furnace Heating.

Table XVII.— Showing the Grate Surface in Square Feet Required to Heat
Buildings of Regular Form—i e., Without Extended Ells— When the Air
is Changed Once in 15, 20 or SO Minutes.

Square feet grate surface required when air

is changed every —

15 20 30

Cubic minutes. minutes. minutes,

contents. Square feet. Square feet. Square feet.

50,000 9.9 8.4 6.8

75,000 14 11.6 9.3

100,000 18 14.9 11.7

150,000 25.8 21.2 16.5

200,000 33.6 27.2 21

250,000 41.3 33.4 25.5

300,000 48.7 39.2 29.9

For severely exposed buildings add from 5 to 10 per cent, to the grate surface stated
In table to allow for winds. Add 10 to 15 per cent, for intermittent use.

When several furnaces are to be used, proportion them accord-
ing to the exposure appointed to each, the combined grate surface
of all to equal the amount stated in the table.

Table XXI, Chapter X, will be of assistance in determining
the diameter of fire pot in inches corresponding to a given grate
surface in square feet.

An inspection of Table XVII will show that the larger build-
ings require less proportionate grate surface than smaller ones,
since they have less exposure as compared with their cubic con-
tents. The loss of heat by transmission is correspondingly less.

AREA OF COLD AIR BOX.

In churches and public buildings the area of the cold air box
— see Fig. 62 — should be 90 to 100 per cent, of the combined ca-
pacity of the furnace pipes. This is especially important for heat-
ing and ventilating in mild weather, when a small amount of heat
but a large supply of air is desired. This can be secured only by
using large flues and cold air box.

FRESH AIR INLET.

The best location for the cold air inlet is on that side of the
building which faces the prevailing cold winds. It is often nec-
essary, however, to place it elsewhere to avoid making the box of
excessive length.

When the heating is intermittent, the use of a return duct —
see Fig. 63 — materially lessens the time and fuel consumed in
warming the building. This return duct may be run independ-

Heating of Public Buildings, Churches and Stores. 123

124

Furnace Heating.

ently to the furnace, or, as more commonly arranged, may be con-
nected with the cold air box, as shown in Fig. 63.

LOCATION OF FURNACE AND AREA OF FLUES.

The furnace should be located as nearly as possible under the
warm ai-r flues leading from it. For ordinary calculations it will
be found convenient to assume a velocity of 300 feet per minute
in flues leading to the first or second floors. Dividing the vol-
ume in cubic feet per minute by 300 gives the area of the flue in
square feet. For more exact calculations use Table XIV. In
determining the size of flues from this table it is well to select a

Fig. 63.— Detail of Return Air Connection at C, Fig. 41.

velocity corresponding to a difference in temperature not greater
than 40 to 50 degrees, in order that the flues shall be large enough
to provide a proper air supply at all times.

The remarks in Chapter IV with regard to the material of flues
and the arrangement of mixing dampers (see Fig. 64) apply here
equally well.

LOCATION OF REGISTERS.

It has long been the custom to locate the registers in the aisles,
placing the furnaces directly under them. There are several ob-
jections to this arrangement. The hot air ascends immediately

Heating of Public Buildings, Churches and Stores. 125

to the ceiling, causing an excessively high temperature at the top
of the room and a correspondingly great loss of heat through the
roof. The registers become the receptacles of dust and filth, over
which the fresh air must pass. It is better practice to discharge
the warm air through openings placed 7 or 8 feet above the floor,
as in schoolhouses.

The ventilating registers are placed, as in Fig. 65 in or near the
floors, in the best position to secure a thorough distribution of the

WARM AIR DUCT
TO AUplTORIUM
•M/V |y|."

Fig. 64. — Section at B B, Showing Mixing Damper and Switch Damper.

air throughout the seating space. Foot warmers should be lo-
cated in the entrance hall or near the doors, and heated by a
separate furnace.

VENTILATION.

Ceiling ventilators are generally provided, but should be no
larger than is necessary to remove the products of combustion
from the gas lights if these are used. If made too large much of
the warmest and purest air will escape through them.

126 Furnace Heating.

The ventilating system should be connected with a duct lead-
ing to a shaft, having a stack heater (Fig. 66) or a fan to accel-
erate the air current. In cold weather the natural draft will in
most cases be found sufficient. The construction and arrange-
ment of stack heaters has been fully discussed in the preceding
chapter.

SIZE OF STACK HEATER.

To determine the size of the stack heater is a simple matter.
Knowing the hight and area of the shaft and the volume of air in
cubic feet per minute to be moved, divide the volume by the area
expressed in square feet; the quotient is the velocity with which
the air must be moved. Next look in Table XIV in the line cor-
responding to the hight of the shaft and find the number most
nearly corresponding with the estimated velocity. At the head of
this column is given the excess of temperature that must be main-
tained in the shaft.

For example, suppose we have a shaft 60 feet high, of 8 feet
square area, and that 3000 cubic feet must be discharged per

^ooo cubic feet

minute; ^— — -. — = 375 feet velocitv. Following along
8 square feet

the line in Table XIV, opposite the hight of 60 feet in the column
at the left we come to the number 383, which most nearly corre-
sponds to the required velocity, 375. At the head of the column
in which the number 383 is found is the number 20, indicating the
excess of temperature that must be maintained in the flue.

Having determined the number of degrees through which the
air must be heated to secure a constant air removal regardless of
the outside temperature, the next step is to calculate the amount
of heat that must be supplied by the stack heater.

One heat unit will heat 55 cubic feet of air at 70 degrees
through i degree F., hence the amount of heat required to raise a
given volume through any number of degrees will be expressed

Volume of air in cubic feet per hour v , XT

by the equation : X Number

j j

of degrees temperature must be raised = Heat units required per
hour. This divided by 40,000 (the heat utilized per hour per
square foot of grate) gives the area of grate or average diameter
of fire pot.

Heating of Public Buildings, Churches and Stores. 127

128

Furnace Heating.

JANITORIAL SHORTCOMINGS.

The importance of the stack heater is very apt to be overlooked
by the janitor, who generally considers the heating as the all-im-
portant matter. Unless his work is under intelligent supervision,
which is seldom the case, the stack heater is quite likely to remain

Fig. 66. — Section at A A, Showing Arrangement of Stack Heater

idle and the flow of air through the ventilating registers to be very
sluggish.

Among other shortcomings of the janitor may be mentioned
taking the air supply from the cellar or from the return duct in-
stead of from out of doors, which should be the only source of
supply while the rooms are occupied; also, allowing insufficient
time to warm the building after a period of disuse, forcing the
fires until they are hottest about the time the occupants assem-
ble, resulting in overheating during the session.

Heating of Public Buildings, Churches and Stores. 1 29

HOT WATER COMBINATION.

It is often desired, when several furnaces are employed, to run
but one continuously, the others being used only when the audi-
torium or the entire building is occupied. When some of the
rooms are located at a distance from the furnace, as in Figs. 57,
58 and 59, the simplest way to heat them is by a means of a hot
water combination applied to the furnace, as described in Chapter
III.

SMOKE PIPES AND FLUES.

If the smoke pipes are very long the smoke is likely to become
so cooled that the draft will be seriously diminished, causing gas
to leak from the furnace into the basement. The liquid common-
ly called creosote, which condenses from the gases and oozes
from the pipes, is troublesome in certain places, besides rapidly
corroding the iron. These troubles may be avoided to a great ex-
tent by covering the pipe with non-conducting material.

If made tight and of ample size smoke pipes, in connection
with a good chimney 60 or 70 feet high, may be run 60 to 80 feet
horizontally without trouble. The smoke-flue may be run up in-
side the ventilating shaft to advantage, the waste heat stimulating
a more rapid ascent of the air.

THE HEATING OF STORES.

For heating small isolated stores or those at the end of blocks
the size of the furnace may be determined from the exposure, as
stated in Table II. For inside stores exposed only at the front
and rear the size of the furnace may be calculated in another way.

The space per occupant is generally so large, except in crowd-
ed districts, that the volume of fresh air to be admitted is seldom
considered in estimating the size of the furnace.

If its size is to be based solely on its ability to heat a given
space, regardless of air supply, we may proceed as follows:
Assume temperature of the entering air to :be 140°, that of the
room 70° and that of the outside air at zero. One-half of the
heat brought in is lost through the walls, floors and ceilings
by transmission before the air escapes at 70° temperature;
in other words, twice as much heat is supplied as that lost

130

Furnace Heating.

by transmission. One square foot of grate burning 5 pounds of
coal per hour will supply to the air passing through the furnace in
zero weather about 40,000 heat units, which is equivalent to that
transmitted by 470 square feet of glass. But since twice as much
heat must be supplied as that lost by transmission, 2 square feet of
grate surface will be required for each 470 square feet of glass, or
i square foot to 235 square feet of glass. Hence to find the square
feet of grate required, reduce the area of walls, floors and ceilings
to equivalent glass surface (E. G. S.). This divided by 235 =
G. S. required. The corresponding diameter of fire pot may be
found in Table XXI.

In narrow, deep stores in blocks the entire front and most of
the rear is generally glass. If not it should be so considered to
allow for the cooling effect of frequently opened doors. To pro-

Fig 67.— Register with Guard Having Marble Top.

vide for quickly warming narrow, deep stores — i.e., those in
which the depth exceeds, say, three times the width — add 25 per
cent, to the grate surface based on the exposure.

Where it is necessary to have basement doors open in winter
for the handling of goods, the loss of heat through the floor should
be added. Its equivalent glass surface equals one-twentieth its
area. With a tight basement the loss of heat through the floor
may be neglected. The equivalent glass surface of a ceiling with
non-heated attic above is equal to one-twentieth its area. When
the ceiling is directly under the roof with no attic space its equiva-
lent glass surface may be considered equal to one-tenth its area.

COLD AIR BOX AND REGISTERS.

The cold air box should be arranged with a branch, so that the
air may be used over and over to warm up quickly.

Heating of Public Buildings, Churches, andS'ores. 131

Having determined the size of the furnace the combined area
of the hot air pipes may be found by allowing about 1% square
inches of pipe area for each square inch of grate surface — i.e.,
average fire pot area. The net area of the registers should be 10
to 25 per cent, in excess of that of the pipes which supply them.

It is well to locate the registers in the walls or in front of
counters instead of in the floor. If floor registers must be used a
guard similar to Fig. 67 maytbe placed over them to prevent their
use as cuspidors and for scraping muddy shoes. Such an arrange-
ment is frequently found in railway stations.

CHAPTER VII.
THE FAN-FURNACE COMBINATION SYSTEfl.

ADVANTAGES.

The combination of a fan with furnaces has been successfully
applied in numerous instances, especially in the heating and ven-
tilation of churches and school buildings. The use of the fan ren-
ders this system capable of supplying a nearly constant volume of
air under all conditions of wind and weather. Reversals of the air
current in the flues, due to changes in the direction of the wind,
which sometimes occur in the simple furnace system, are pre-
vented.

APPLICATION OF THE SYSTEM.

The fan-furnace combination may be applied not only to
churches and schools, but to hospitals, public and other buildings
where a large and continuous supply of fresh air is required. In
comparison with other mechanical systems this one is less expen-
sive and simpler in its make up.

When arranged to rotate the air it is capable of warming
the rooms very quickly. The system is, therefore, well adapted to
buildings used intermittently. For buildings of good size, which
must be kept warm night and day, this method of heating must
give way to some form of steam apparatus. With the latter the
cost of power for driving the fan will be less than for gas, elec-
tricity or water, and the boiler fires may be handled more easily
than a number of furnaces.

LOCATION OF THE FAN.

The fan should be placed between the furnace and the fresh
air inlet to the building (see Figs. 68 and 69). The air will then
be forced, instead of drawn, through the furnace, as would be the
case with the fan placed beyond them.

The " blow through " arrangement has several advantages
over the other. At a given speed the fan will handle a greater

132

The Fan-Furnace Combination System.

133

weight of cold than of warm air ; hence, to deliver a stated volume,
the fan, when so arranged, may be run at a lower speed than when
handling air at a higher temperature, as in the " draw through "
arrangement. The lower the speed the less the noise and vibra-
tion. The air being under pressure, any leakage of gas or dust
from the furnaces is prevented. Branch pipes may be taken from
the main cold air duct before reaching the furnaces and be carried
to the mixing dampers placed at the base of the flues.

With the " draw through " arrangement this would be impos-
sible, as only warm air is handled by the fan.

LOCATION OF DRIVING APPARATUS.

The engine or motor and fan must be located where they will
be least likely to cause trouble from noise. The best location is

MOTOR

« S ><:- -

.INQi UNDERGROUND
PERI DUCT

1 1

: c§

S 14

«?

^3"

i i

Fig. 68.— Plan of Fan-Furnace Combination.

just outside the walls of the building in a room provided for the
purpose. If the apparatus must be located in the basement, the
fan and engine or motor must be placed away from piers, which
are likely to transmit vibration. To prevent sounds being carried
along galvanized iron ducts the following expedient is sometimes
resorted to. A section of the pipe about 4 inches long is cut away
and a sleeve of light canvas is slipped over the ends and fastened
by means of wires drawn up tightly, thus forming a flexible air
tight connection.

SIZE OF FURNACES.

The size of furnaces for schools, churches and public build-
ings is determined as explained in the chapters under those

134

Furnace Heating.

headings. Having calculated the grate areas, bear in mind that
small furnaces have more heating surface proportionally than
large ones. Hence they may be used to better advantage than a
single large furnace having their combined grate area.

KIND OF FURNACES.

The furnaces must be of the best materials and construction to
withstand the severe strain often accompanying intermittent use.
The ratio of heating to grate surface must be large. Extended
surface in the form of pins and ribs may be used to advantage to
break up the air current. To secure the best distribution of air
around the furnace a pit should be used.

AREA OF AIR PASSAGES IN FURNACES.

When the furnaces are intended to be run, at times, independ-
ent of the fan the space for the passage of air through them should

^^^^

Fig. 69. — Sectional Elevation of Fan-Furnace Combination-

be about equal to the combined capacity of the ducts supplied, or
sufficient to permit the required volume of air to pass at a velocity
of about 300 feet per minute. A higher velocity, say 600-800 feet
per minute, may be allowed in furnaces which are always used in
connection with the fan. If the space is too great the air will be
likely to be unequally heated, a portion passing through the fur-
naces without being brought into close contact with the heating
surface.

SETTING.

The furnaces may be set either in brick or galvanized iron, the
relative merits of which have been previously discussed. The

The Fan-Furnace Combination System.

135

joints must be tight, to prevent the leakage of air. It is well to
cover galvanized iron casings with plastic non-conducting mate-
rial.

The furnaces may be placed side by side, in battery, so-called,
or they may be set separately and connected with the fan by ducts.
The battery arrangement facilitates firing and attendance in gen-
eral, but furnaces so placed cannot be run as well independently as
those located near the rooms which they heat. With either ar-

Fig. 70. — Blower Type of Fan.

rangetnent, provision should be made for returning air from the
building when unoccupied.

TYPES OF FANS.

Two types of fans are used, the blower type, Fig. 70, like a
paddle wheel, where the air leaves the fan in a direction perpen-
dicular to the shaft, and the disk fan, Fig. 71, like a propeller,
where the air leaves the fan in a direction parallel to the shaft.

When the ducts are of considerable length the blower is pref-
erable to the disk fan, for with the former the air may be handled
against resistance without excessive expenditure of power. The
disk fan is adapted only to short lengths of pipe of large area.

If the resistance be increased by closing registers or dampers,
the volume of air delivered will be diminished, but the power con-
sumed by a disk fan will be greater. On the other hand, with

136 Furnace Heating.

fans of the blower type, if the resistance be similarly increased
the lessened delivery of air will be accompanied by a correspond-
ing reduction in power. Both types of fans may be pulley driven
or have direct connected motor.

SPEED OF FANS.

It has been found in practice that fans of the blower type hav-
ing curved floats operate quietly and give good results when run
at a speed corresponding to */2 ounce pressure— i e., a speed at

Fig. 71. — Disk Type of Fan with Pulley.

the circumference of the wheels of about 3600 feet per minute.
Higher speeds are accompanied with a greater expenditure of
power and are likely to produce a roaring noise or cause vibra-
tion. A much lower speed does not provide sufficient pressure to
give proper control of the distribution during strong winds.

FAN CAPACITIES.

The capacities and powers given in fan manufacturers' cata-
logues are often different from those obtained in practice, the
tables being based on other than practical working conditions.
They should therefore be used with caution.

The following tables, XVIII and XIX, are intended as a guide
in the selection of fans and motors, the former to be used where
the ducts are of considerable length, the latter where they are
short and of large area with easy turns :

The Fan-Furnace Combination System. 137

Table XVIII.— Air Delivery per Minute and the Appropriate Size of Motor
for Fans of the Blower Type.

Nominal size

Cubic feet

For belted

of fan.

Ordinary

of air

motor

Hight

Diameter

Width of

speed

delivered

use

of housing.
Inches.

fan.
wheel.

fan
housing.

^-ounce
pressure.

per
minute.

horse-
power.

580

1,600

465

2,600

390

4,500

333

6,000

293

8,000

260

11,000

100

233

12,500

Table XIX.— Air Delivery per Minute Against Slight Resistance and the
Appr priate Size of Motor for Fans of the Disk Type.

Cubic feet of air For belted

Disk fan delivered motor use
wheel. per horse-
Inches. Speed. minute. power.
12 1,000 600 H
18 800 1,500 H
24 500 2,300 . 1
30 410 3,500 1
36 380 5,700 l^i
42 330 7,800
48 280 9,900
54 250 12,500
60 • 230 16,000 3

THE MOTIVE POWER.

The driving apparatus generally consists of an electric motor,
although where electricity is not available a gas engine or water
motor may be used. The gas engine is the most expensive in
first cost, then the electric and water motors, in the order
named.

The cost per hourly horse-power where the amount is less than
5 horse-power per hour would be roughly say 5 cents for the gas
engine, 10 cents for the electric motor and 30 cents for the water
motor. The electric motor is the most convenient machine to
use. It may be easily controlled by a switch and starting box
or speed regulator. The latter should have an automatic device
to cut out the resistance coils whenever the current is interrupted
from any cause. The motor may be connected directly to the
fan shaft or it may be belted. Independent motors should be
slow speed and should rest on an adjustable base for convenience
in tightening the belt. In ordering always state the voltage and
kind of current.

The gas engine is the least quiet of the three machines, the
noise of the exhaust being difficult to overcome. This may be
done, however, by leading the exhaust pipe first into a cast iron

!^8 Furnace Heating.

pot or equalizing chamber, thence into a pit or dry well of large
capacity with a suitable outlet and vent. A water supply for cool-
ing the cylinder is necessary, and in some locations of the engine
this involves danger of damage from freezing in case of neglect.
The water motor is simple, quiet and convenient, but the cost
of running one at city water rates is generally prohibitive.

AREA OF DUCTS AND FLUES.

With the blower type of fan the size of the main ducts may be
based on a velocity of 1000 to 1200 feet per minute, the branches
on a velocity of 800 to 1000 feet per minute, and as low as 600 to
800 feet when the pipes are small. With the disk type of fan the
size of the ducts should be based on a velocity per minute not
greater than 1000 feet, preferably less, in order to keep the resist-
ance low.

Flue velocities of 500 to 700 feet per minute are permissible
with the fan combination, though it is better, when possible, to
keep the velocity as low as 400 feet. When 'the furnaces are
placed separately and are intended to be run independent of the
fan at times, the warm air flues should be based on a velocity of
about 300 feet per minute.

The size of registers may be based on about the same velocity
as last stated, adding 10 to 25 per cent, to offset the additional re-
sistance to the passage of air through them.

CHAPTER VIII.
TEMPERATURE CONTROL.

GENERAL REMARKS.

There is, perhaps, no device that contributes more to comfort
and convenience in the home during the winter months than an
automatic temperature regulator. These devices in various
forms give excellent results and are highly desirable. Not only
may an even temperature in the house be secured, but those
sudden and severe strains are avoided to which a furnace is often
subjected when regulated by hand. The fire is maintained so
evenly that the coal is burned to the best advantage, and few
if any, clinkers are formed.

TYPES OF REGULATORS.

These devices may be divided into two classes, one comprising
those in which the drafts are regulated directly by the temperature
of the air passing through the furnace, and the other those in
which they are governed indirectly by changes in the temperature
of the rooms.

In the former the difference in the rate of expansion of certain
metals is taken advantage of, to operate the dampers by means of
levers connected with them by wires or chains.

In the latter the thermostat placed on the wall of one of the
rooms is so constructed that a change in temperature causes a
metal strip or U shaped piece to open or close an electric or pneu-
matic circuit connected with a motor or diaphragm which operates
the dampers. Many thermostats are of the volatile liquid
changed type, the pressure generated within them due to increase
in room temperature serving to operate dampers directly or in-
directly through pneumatic control. For large installations the
pneumatic system is principally used.

The thermostat should be placed in room most nearly re-
presenting the average temperature of the house. It should be
located where it will not be subjected to cold drafts or to currents
of warm air from registers.

139

140 Furnace Heating.

DAMPER CONNECTIONS.

To give the best results the regulator should be connected with
both smoke pipe and ash pit dampers ; a sufficient air supply will
then be assured to promote proper combustion. The fire will re-
spond more quickly to the action of the regulator than when the
latter is connected only with the smoke pipe damper.

OPERATION OF THE REGULATORS.

Regulators acted upon by changes in the temperature of air
within the furnace serve to control the fire much as an ordinary
diaphragm regulator on a steam apparatus. When used with
house heating furnaces, regulators of this type must be set each
day with reference to the outside conditions. If these remain
nearly uniform an even temperature in the house will be main-
tained, but with sudden changes in the weather this type of regu-
lator, unless reset, is not capable of preventing variation in the
temperature of the rooms.

Regulators run by clockwork, which open dampers at any
desired time, are often used to automatically turn on the drafts in
the early morning. They generally consist of a simple alarm
dock with a ratchet or gear arranged to trip a lever, thus allow-
ing the weighted damper to open.

CONTROL OF MIXING DAMPERS.

In schools, churches and public buildings where mixing damp-
ers connected with warm and cold air ducts are used, they may
best be controlled by thermostats having a gradual movement.
These thermostats have no connection with the draft dampers of
the furnaces.

CHAPTER IX.
ESTIflATES AND CONTRACTS.

FORMS AND BLANKS.

In laying out furnace heating work it is desirable to have the
necessary items conveniently arranged on a printed form, either in
an indexed book or on loose cards or sheets, which may be filed
alphabetically. By the use of printed forms omissions will be
avoided and the data preserved in a form convenient for reference.

It is well to make a rough sketch of the 'house, giving outside
dimensions and showing the general arrangement of rooms and the
points of the compass. The items may well include : Date, name
and address of owner, location of house, name of architect, loca-
tion of house in regard to exposure to cold winds, list of rooms
with size and number of sides exposed, size of registers and pipes,
length of hot air pipes, length of smoke pipes, clear hights of base-
ment and floors above, square feet of exposed wall, size of furnace
adapted to the estimated exposure, combined area of hot air pipes,
area of air passages through furnace, area of cold air box. A
form of data card, 3^4 x 8 inches, used by a Boston company, is
shown herewith.

In computing the cost of furnace pipes it is convenient to allow
for elbows by adding a length of straight pipe equivalent in cost.
Two feet of straight pipe may be considered approximately equal
in cost to one elbow of the same diameter.

ESTIMATES.

Having determined the size of the furnace, pipes and registers
the cost of the job may be estimated. For an ordinary house heat-
ing apparatus the following are the principal items of expense :

Furnaces (number, kind, size, diameter fire pot, portable or
brick set), covering bars and man door for brick setting, smoke
pipe and check damper, fire tools, pipes and registers, stones,
boxes, nettings, plaster rings, floor flanges, dampers, furnace col-
Hi

142 Fdrnace Heating.

lars, covering tin or asbestos millboard, cold air box, galvanized
iron cold air neck, shield over furnace, labor in erecting, fares and
expenses, freight and carting, masons' or carpenters' work and
materials.

An estimate for heating schools, churches or public buildings
may include, in addition to the items stated above :

Galvanized iron heating and ventilating flues, mixing dampers,
chains and fixtures, regulating and shut off dampers, wrought iron
smoke stack in ventilating shaft, stack heater at base of venti-
lating shaft, steam boiler with coils in ventilating flues in place of
stack heater, pipes, valves, fittings and labor in connection with
same, hood for top of galvanized iron ventilating shaft.

A hot water combination heating estimate commonly includes
these items:

Water heating section or coil in furnace, radiators or coils,
pipes, valves, fittings, air valves, pipe covering, expansion tank
and fittings, labor of erecting, painting and bronzing, fares and
expenses, freight and carting.

SPECIFICATIONS.

The specifications should be clear and to the point, leaving no
opportunity for misunderstanding between the contractor and the
owner. The items just enumerated form the basis of the speci-
fications, which should describe each of them fully.

GUARANTEE.

Unless expressly stipulated to the contrary it is commonly
understood that the apparatus specified in a proposal for heating
is to be capable of warming rooms having registers or radiators
to an average temperature of 70 degrees in zero weather, when
operated continuously as directed by the contractor and that the
temperatures in the different parts of the room shall not vary
more than 5 degrees.

PAYMENTS.

On small jobs the entire payment is generally made on com-
pletion of the work ; on larger ones payments are made as the work
proceeds, on the certificate of the architect or engineer. In case a

Estimates and Contracts. 143

job is completed too late in the season to be tested in severe
weather and the owner is unwilling to have the final payment con-
strued to mean a final acceptance of the work, the contractor, if
responsible, can generally secure a prompt settlement by giving
the owner a written extension of the guarantee over another win-
ter, allowing ample time for a thorough trial of the apparatus.

Date

Location of
Architect

House

Room

Size

Sides
ex-
posed

Reg size

Pipe

size

Length of pipes

Run
off

Riser
el. col.
damp.

Total

On back of card show plan of house, with outside dimensions, arrangement of rooms and

points of compass.

Clear Ti'f s B'ment 1st 2d 3d 4th

Smoke Pipe ft.

Exposure, Severe Moderate Sheltered

Total Exposure sq. ft. Material

Furnace ; Use No Rated at sq. ft.

Combined Area, Hot Air Pipes sq. in.

Area, Air Passage in Furnace sq. in.

% Combined Area of Pipes sq. in.

Area, Cold Air Frame .sq. in.

The custom of allowing a portion of the contract price to re-
main unpaid until the apparatus has been tested in zero weather
is becoming less common among responsible contractors, not

144 Furnace Heating.

from a desire to dodge responsibility, but because in some sections
zero days are rare and it might be that an entire heating season
would pass without opportunity for a zero day test, thus tying up
the contractor's money. With a responsible contractor the owner
runs little risk to take the contractor's guarantee to make good
any defects that may occur in heating or in workmanship that
may appear during a second heating season, provided the weather
is too mild the first season to afford a proper test.

In heating contracts for schools, churches or public buildings
a bond for the successful completion of the work is often required.

CHAPTER X.
FUELS MISCELLANEOUS TABLES AND DATA.

FUELS.

Anthracite Coal. — Anthracite or hard coal consists almost en-
tirely of free carbon. It has a theoretical heating power of about
14,200 heat units per pound of combustible. It burns with a
bluish flame tinged with yellow, with no smoke.

Bituminous Coal. — Bituminous or soft coal contains about 50
to 80 per cent, of carbon ; as a rule, coal containing as much as
20 per cent, of volatile combustible is called bituminous. It has a
theoretical heating power of about 13,000 to 14,000 heat units per
pound of combustible. It burns with a yellow flame with smoke.

Coke. — Ordinary gas house coke, commonly used for domes-
tic purposes, is a by-product from the distillation of gas from
bituminous coal. It consists almost entirely of carbon, ignites
quickly, and gives an intense clear fire. The weight of coke by the
bushel may be estimated by allowing 50 bushels per ton. It is
commonly sold by the chaldron, equal to 36 bushels.

Wood. — The American Society of Mechanical Engineers con-
siders 2^ pounds of dry wood equivalent in heating power to I
pound of coal. On this basis:

1 cord of hickory or maple is equivalent to 1800 pounds of coal.
1 cord of beech or oak is equivalent to 1300 pounds of coal.
1 cord of pine is equivalent to 800 pounds of coal.

To put it roughly, a cord of hard wood is equivalent to a ton of
coal.

Gas. — Natural gas varies greatly in heating power accord-
ing to its composition. For equal volumes, ordinary coal
gas has about two-thirds the heating power of natural gas of
average composition, water gas (uncarbureted) about 30 per
cent., and producer gas about 13 per cent. In round number, 25,-
ooo cubic feet of natural gas is equivalent in heating power to a

145

146

Furnace Heating.

ton of coal. Professor Jacobus states that " The number of cubic
feet of water gas required to produce the same heating effect as
that produced by burning i ton (2000 pounds) of Lackawanna
coal is 91,780 cubic feet of uncarbureted gas, or 40,590 cubic
feet of carbureted gas."

Petroleum. — Crude petroleum has a specific gravity of 0.83 to
0.93 (i. e., for equal volumes it is 83/ioo to 93Aoo as heavy as water).
The heating power of i pound of the crude oil is a trifle less than
21,000 heat units. Refined petroleum oils have specific gravities
ranging from 0.628 to 0.792, with heating powers from 28,087 to
26,975 heat units respectively.

CHIMNEY FLUES.

Table XX. — The Appropriate Orate Surface or Fire Pot Area for Chimneys of
Various Sizes and Bights, Based on a Rate of Combustion of 6 Pounds of
Hard Coal per Square Foot of Grate Surface per Hour.

Diameter of

chimney.

Inches.

-Hight of chimney in feet.

Approximate grate surface. Square feet.

Square or rec-
tangular

Flue.

8x8

8x 12
12 x 12
12 x 16
16 x 16
16 x 20
20 x 20
20 x 24
24 x 24

Table XXL— Area of Fire Pot in Square Feet.

Diameter.
Inches.
18 . ..

Area.
Square feet.
. — 1 76

Diameter.
Inches.

Area.
Square feet.
— 427

19 ...

— 1.97

— 459

— 218

— 4 90

— 2.40

— 5 25

— 2.64

— 5.68

— 2.88

— 5.83

. . — 3.13

34..

— 6.80

— 3.40

35..

— 6.67

— 3.68

— 7.06

27...

.. -3.98

CAPACITY OF COAL BINS.

For convenience in estimating the capacity of coal bins, an al-
lowance of 40 cubic feet per ton of 2000 pounds of anthracite
egg coal is approximately correct. This rule is on the safe side
and on this basis a ten ton bin would require 400 cubic feet or
say 8 feet x 9 feet, piling the coal about 6 feet high.

Fuels. — Miscellaneous Tables and Data.

147

According to Pouillet the following temperatures have been
observed for iron at different stages of incandescence:

Table XXII.— Colors of Iron at Different Temperatures.

Faint red 525

Dark red 700

Faint cherry 800

Cherry 900

Bright cherry 1 ,000

Dark orange 1,100

Bright orange 1,200

White heat 1,300

Bright white 1,400

Dazzling white 1,500

Table XXIII.— Power of Various Substances to Transmit Heat.

Peclet gives the relative heat transmitting power of various
substances as follows :

Copper 64.00

Iron 29.00

Zinc 28.00

Lead 14.00

Coke 4.96

Marble 3.13

Limestone 1.82

Glass 0.82

Burned clay 0.60

Gypsum 0.48

Pine wood, parallel to fiber 0.17

Pine wood, across fiber 0.09

Oak, across fiber 0.25

Cork 0.14

India rubber.. . 0.17

Brick dust 0.15

Woodashes 0.08

Linen 0.05

Cotton 0.04

Paper (gray), unsized 0.03

Table XXIV.— Colors of Fires.

The same authority gives the temperature of the fire corre-
sponding to its color as follows :

Temperature.
Color. Fahrenheit.

Red, just visible 977

Red, dull.... 1,290

Red, cherry dull 1,470

Red, cherry full 1,650

Red, cherry clear

Temperature.
Color. Fahrenheit.

Orange, deep 2,010

Orange, clear 2,190

White heat 2,370

White, bright 2,550

White, dazzling 2,730

Table XXV.— Radiating Power of Various Substances.

According to Dulong and Petit the relative radiating power
with same difference in temperature for different substances is as
follows :

Polished silver 16

Polished brass 32

Red copper 20

Zinc 30

Tin 27

Polished sheet iron 56

Leaded sheet iron 81

Black sheet iron... .. 345

Rusted sheet iron 419

New cast iron 895

Rusted cast iron 419

Glass 373

Soot 500

Building stones 449

Wood.. ...449

148 Furnace Heating.

Table XXVI.— Conducting Power of Various Substances.

Representing the conducting power of gold by 1000 the con-
ducting power of other substances is represented, according to
Depretz by the following figures :

Platinum 981

Silver 973

Copper 898

Iron 374

Zinc 363

Tin 303

Lead 180

Marble 23

Porcelain 12

Brick earth 11

Table XXVIL— The Weight of Galvanized Iron Pipe, the Areas and Circumfer-
ences of Circles.

In
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
22
24
26
28
30
32
34
36
38
40
42
44
46
48
50
52
54
56
58
60

ameter Approx. Circum- , Weight of p
ipe. area. ference. No. 28 No. 26 No. 24
ches. Sq. inches. Inches, gauge, gauge, gauge.
0.7854 3.14 ...

ipe per running foot.
No. 22 No. 20 No. 18
gauge, gauge, gauge.

No.16
gauge.

3.1416

6.28 . .
9.42 0.
12.56 1.
15.70 1.5
18.84 l.i
22.00 1.'
25.13 1 .{
28.27 2
31,41 2.
34.55
37.70
40.84
44.00
47.12
50.28
53.41
5554
59.69
62.83
69.11
75.39
81.68
87.96
94.24
100.53
106.81
113.00

....

7.07

1 ....

12.57

19.64

I 1.4
I 1.7

r 2.0

) 2.2
1 2.4
J 2.7
2.9
8.2
3.4
3.7

1.8
2.1
2.5
2.8
3.1
8.4
3.7
4.1
4.4
4.7
6.0
5.4
5.7
6.0
6.3
6.8
7.3
8.0
8.7
9.4
10.0

... 28.27
38.49

The heavy faced figures Indi-
cate the weight of pipes
commonly built of the gauge
stated at the head of the
column in which they occur.

. 50.27

. 63.62

78.54

. 95.03

113.10

132.73

153.94

"6.i
6.5
6.9
7.3
7.7
8.2
8.9
9.7
10.6
11.4
12.2
13. 0
13.9
14.6
15.5
16.2

....

'.'.'.'.

!!!!

176.72

201.06
.... 226.98

25447

283.53
814.16

....

...

880.13
452.39

11.5
12.4
13.4
14.4
15 3
16.3
17.2
18.2
19.1
20.1
21.01
22.0 f
22.9
23.9

area of
iference

18.7
20.0
21.2
22.4
23.7
24.9
26.1

27.4

29.8
31 O
32.2
33.6
34.9
36.1
37.4

30.7
32.2
J33.7
J 35.2
36.7
38.2
39.7
41.4
43.0
44.5
46.0

530.93

615.75

.. 706.86

804.25

907.92

101788

1134 12

11938

1256.64

125 66 ....

1385.45
152053

131 94

138 23 ....

...1661.91
180956

144 51 ....

150.79

1963 50

157 08

2123.72
2290.23
.. 2463.01

163.36
169.64 The diameter squared X 0.7854 =
175.93 a circle
182.21 The diameter X 3.1416 = circuit
188.49 of a circle.

2642.09
2827.74

Weight of Galvanized Iron Sheets in pounds per square foot, United States Government

Gauge 28

Weight in pounds 0.78

0.91

1.16

1.41

1.66

tie

16
2 66

In the larger sizes of pipes the elbows are commonly made
with the inner radius of the bend equal to the diameter of the pipe.

Fuels. — Miscellaneous Tables and Data. 149

This gives an easy turn. Such elbows may be figured at double
their actual weight to allow for the extra cost of making. This
weight may be estimated by multiplying the diameter of the el-
bow in inches by T\, which gives double the length of the elbow
expressed in feet. This length multiplied by the weight of the
pipe per running foot, as shown by above table, gives double the
weight of the elbow.

To estimate the weight of square or rectangular pipe, it is ap-
proximately correct to find in the table the circumference corre-
sponding most nearly to the sum of the four sides of the pipe, ex-
pressed in inches. Opposite the circumference thus found is
given the weight of the pipe per running foot.

Table XXVIII.— Weight of Black Sheet Iron, United States Government
Standard Gauge.

Thickness Weight in pounds

in per
Gauge. inches. square foot.

10 ^ 5.625

32 & 4.375

14 ^ 3.125

16 ^ 2.5

18 & 2.0

20 & 1.5

22 ^ 125

24 -& 1.0

26 -rfo 0.75

28 Jj 0.625

~ Intermediate gauges 11, 13, etc., have weights midway between the even gauges
stated in table.

Table XXIX.— Weight of Plate Iron in Pounds per Square Foot.

Thickness Weight in pounds

of plate per

in inches. square foot.

A 7.65

i^ 10.20

tfc 12.75

y% 15.30

& 17.85

% 20.40

& 22.95

y8 25.50

% 30.60

% 35.70

1 40.80

Cast iron weighs about 0.26 pound (roughly J^-pound) per cubic inch.

Furnace Heating.

Table XXX — Showing the Size, Net Area and Depth of Registers.

Size of
opening.
4^x 6>$

4x 8..

4x10

Net area of
opening.
Square
inches.
19
21
27

Depth
open.
Inches.

4x13 .................... 85

4x15 .................... 40

4x18 .................... 48

4x21 .................... 58

4x24 ................ 64

5x 8 .................... 27

5x 9 .................... 30

5x10 .................... 33

5x11 .................... 37

5x12 .................... 40

5x13 .................... 43

5x14 .................... 47

5x15 .................... 50

5x16 .................... 53

5x17 .................... 57

5x18 .................... 60

6x 6 .................... 24

6x 8 .................... 32

6x 9 .................... 36

6x10 .................... 40

6x12 .................... 48

6x14 .................. 56

6x16 .................... 64

6x18 .................... 72

6x20 .................... 80

6x22 .................... 88

6x24 .................... 96

6x 28 .................... 112

6x32 .................... 128

7x 7 ................... 33

7x10 ............. . ...... 47

7x 12 .................. 56

7x14 .................... 65

8x 8 .................... 43

8x10 .................... 53

8x12 .................... 64

8x13 .................... 69

8x14 .................... 75

8x15 .................... 80

8x16 .................... 85

8x18 .................... 96

8x21 .................... 112

8x24 .................... 128

8x27 .................... 144

8x30 .................... 160

9x 9 .................... 54

9x12 .................... 72

9 x 13 .................... 78

9x14 .................... 84

9x15 .................... 90

9x16 .................... 96

9x 17 .................... 102

9x 18 .................... 108

9x 19 .................... 114

9x20 .................... 120

9x22 .................... 132

9x24 .................... 144

9x 25... .......... 150

9x 26 .................... 156

9 x 28 .................. 168

9x 30 .................... 180

10x10 .................... 67

10x12 .................... 80

10x14.,. . 98

Net area of

Size of Square

opening. Inches.

10x16 107

10x18 120

10x20 133

10 x 24 160

11x17 125

12 x 12 96

12 x 14 112

12 x 15 120

12 x 16 128

12 x 17 136

12 x 18 144

12 x 19 152

12x20 160

12x24 192

12x30 240

12x36 288

14x14 131

14x16 149

14x18 168

14x20 187

14x22 205

15x25 250

16 x 16 171

16 x 18 192

16x20 213

16 x 22 235

16x24 256

16x28 298

16x32 342

18x18 216

18x21 252

18x24 288

18 x 27 324

18x30 360

18x36 432

20x20 267

20x22 294

20x24 320

20x26 347

20x28 374

20x30... 400

20x32 428

20x36 480

21 x 21 294

21x25 350

21x29 406

21 x 33 .... 462

21x37 518

24x24 384

24x27 432

24 x 30 480

24x32 512

24x36 576

24x45 720

27x27 486

27x38 684

30x30 600

30x36 720

30x42 840

30x48. 960

36x36 864

36x40... 960

36x42 1,008

36 x 48 1,152

38x38 963

38 x 40 1,013

38 x 42 1,064

Depth
open.
Inches.
3%

Registers are made in many other sizes than those stated above.

Fuels. — Miscellaneous Tables and Data.

Table XXXI —The Lowest Temperature Recorded at Various Places in the
United States During a Period of ten Years from 1886-1895 inclusive.
Compiled by the Author from Reports of the Weather Bureau.

ALA.— Montgomery

AT?Ty jPrescott

A±u^. -| Tucson

ARK.— Little Rock

jLos Angeles ....

\ Sacramento

COLO. — Denver.
CONN.— New Haven

FLA.— Jacksonville

GA.— Atlanta

1886. 1887. 1888. 1889. 1890.
. 15 13 18 21 21
._ 4 8-12—8 3

1892. 1893. 1894. 1895.
20 17 13 8

.10 0 7 17

. 36 33 31 32

.34 28 19 31

_H _18 —20 — 7

— 2 — 5 — 4 — 3
22

16
34
29
— 8
4

31 22 28 30 27
8 9 13 14 17

22 18

11 1—2

31 32 34

28 26 28

_17 _ 2 — 8 —15

0 —3 — 5 — 7

29 24 14 14

13 8 4 0

IDAHO1 Idaho Falls.

* *

—32

ILL.— Chicago
IND.— Indianapolis.
IND T —Fort Sill

—14
1

—15

—18
— 6

— 7

—11
— 1

— 5
4
6

— 8
— 3

— 5

—15

— 9

—15
—14

lOWA-Des Moines
KAN.— Dodge City
KY.— Louisville
LA — New Orleans ....

.—20
.—18
.— 5
28

-24

—17
— 5
21

—27
-18
8
29

—13

- 8
6
32

—18
— 6
13
30

-10
— 0
7
30

—26
—11
4
23

-16
— 7
—10
29

07
—15
— 5
21

—18
—14
-10
16

MASS — Boston

— 5

— 1

+ 2

— 4

MD.— Baltimore
ME — Portland

— 5

7
— 15

9
- 12

12
4

16
— 4

12
5

1
9

7
15

1
11

MTHTT J Detroit
MICH. /Marquette.

MINN.— St Paul ... .
MISS.-Vicksburg
MO.— St. Louis
MONT.— Helena
N. C.— Charlotte
NEB.— Omaha
NFV (Carson City.. ..
j Winnemucca . . .
N. D.— Bismarck . . .
N. H.— Manchester
w T j Atlantic City . . .
N- J']New Brunswick..
N. MEX.— Santa Fe . .
XT AT- ) Albany

.—12
—15
.—36
. 17
.—10

11
.—18

'9
—36

'—'3
—10

— 3

-21
—36
10
-10

8
—22

—'3
-44
— 4
— 2

— *8
—15

— 7
—27
—41
18
—12
-41
16
—25
—10
—28
—37
—11
2

— '2
10

— 8
-21
—25
24
0
—15
13
-10

—14
-34
— 9
2

— 'i

— 5

-1- 8
—12
—22
24
4
—29
19
-14
—22
—23
—35
— 6
10

— 2
4

2
—12
—25
22
4
—24
19
— 9
0
- 8
—33
— 7
14

— 'e

-3
—10
—25
16
— 2
—22
18
—26
2
3
—34
— 3
9

1
5

- 10
—19
-26
20
— 2
—42
5
—16
8
—19

- 9
— 4
-10
5
g

—11
—17
—25
15
—11
—26
2
—22
— 7
—11
—33

5
— 1
0
11

— 8
—16
—26
4
—12
—17
1
-20
— 4
—14
-39

— io

—11
1°

N- Y'lNew York
OHIO — Columbus.

0
—11

6
— 5

2
2

9
5

8
5

1
4

— 3

OKLA. Oklahoma City.

—11

roj^ j Baker City....

— 14

01UM Portland .
p. (Philadelphia
1 Pittsburgh .
R. I.— Narragansett Pier.
S. C.— Charleston
e r» J Pierre . . .

17
0
— 9

'. 22

9
8
4

— 2
2

23
2
— 1

10
9
5

23
12
9

—11

20
10
2
— 1
25
—30

8
0
— 3
— 4
20
26

18
4
— 4
— 7
14
28

25
— 3
— 6

| Yankton

—24

—29

—28

—18

3°

TENN. -Nashville.
TEX.— San Antonio
UTAH-Salt Lake City..
VA. — Lynchburg. ...
VT.-Northfield . ..

— 2
. 26
. 5
4

- 2

17
9
6
-21

11
—17
11
—24

12
28
5
7
—32

16
21
— 6
19
—22

17
25
0
16
—17

10
19
— 1
10

26
4
— 6

— 2

16
— 1
7
31

— 6
11
0
— 3
17

m A OTT J Olympia.
WASH- 1 Spokane
W. VA.— Parkersburg.
WTS La Crosse

. 23
. 14

—11

— 2

12
12

4
23

7
-23
4
23

21
-10
8
24

24
— 5
0
20

28
—19
-11
' °6

21
— 2
— 4
19

27
8
- 8

WYO.— Cheyenne. . .

—16

— 7

—29

—17

—20

Lowest
in ten
years.

—12
11

— 2
31
19

—20

14
0

—28
—32
—18
—15

— 7
—27
—18
—10

— 7
1

—15
—12
—27
—41
4
—12

—22
—28
-44
—11

— 4
—10
—11
—15

— 3
—12
—11
—17

— 2

— 3

— 9

— 7
12

—30
-32

— 8
11

—17

— 6
—32

— 2
—80
—11
-42

CHAPTER XL
FURNACE ERECTION AND FITTINGS.

This chapter is made up of a series of articles reprinted from
"Metal Worker" and other sources. Some of the statements
contained therein are not in full agreement with rules laid down
by the author in the main portion of this treatise. Where sizes
obtained by rules given by various writers in this chapter are not
so large as when based on rules stated in Chapters I. to X. in-
clusive, the reader is cautioned to use the larger sizes.

All figure numbers in this and chapters that follow refer to
illustrations in the text in which they appear.

The reader's attention is called to the fact that in the follow-
ing pages the pronoun "I" or the words "the author" refer to
the writer of the article and not to the author of the preceding
pages, Win. G. Snow.

FURNACE FITTINGS.

In the preparation and construction of fittings three general
rules should be strictly adhered to. i. Adaptability. 2. Construc-
tion conforming to the laws of air currents. 3. Due regard to the
economy of stock and labor. What we understand by adaptability
is that the fittings should be so constructed that they will be adapt-
ed to the work and not that the work must be adapted to the fit-
tings. Again, every fitting should be so made as to adapt itself to
as many different situations as possible in order to avoid the neces-
sity of keeping a large variety on hand.

In the earlier days of furnace work, when materials of every
kind were expensive and labor cheap, it was the main thought in
the making up of stock to save material. But at the present time,
when materials are cheap and labor comparatively high, it is often
found to be economy to sacrifice material to save labor.

FURNACE CASING.

The first fitting to be used after the furnace castings have been
set is the casing. We wish to'say , however, before taking up theques-

152

Furnace Erection and Fittings. 153

tion of casing, that in these articles we shall have reference only to
the style of furnace known as portable furnaces, or furnaces with
sheet iron jackets or casings. The question of merit as between
the portable and brick set does not enter here. We only know that
a large percentage of the furnaces used are portable, and we are
compelled to take things as we find them, not as we may think
they should be. The making of the casing is governed somewhat
by the construction of the furnace. All casings should be double,
an inside and an outside one, with a space between of at least an
inch, with a provision for a free circulation of air from the base to
the top. But to have a proper double casing requires casing rings
made for that purpose, and all manufacturers do not make them
that way. Where they are not so made a substitute must be pro-
vided that will come as near to it as possible. Sheets of black iron
may be suspended from the rings on the inside, or sheets of tin
grooved together and hung in the same way will answer. For the
double casing black iron is used for the inside and galvanized for
the outside. The inside casing should be made to fit in its place
on the ring the same as the outside, so that there will be no chance
for it to get out of place while in use. We frequently hear com-
plaints by those using furnaces that at times the air coming from
certain registers is cold even when there is a good fire. The cause
of this trouble is usually found in the fact that the top radiator of
the furnace is so much larger than the body of the furnace and
gives so great a space between the body of the furnace and the cas-
ing that much of the air passes up along the casing and does not
come in contact with the furnace castings or body and is therefore
not heated. Some manufacturers remedy this by making the
lower section of the casing smaller, using a flaring ring and larger
upper section. Where this is not done the difficulty may be
overcome by using trench plates or deflecting plates, which are
usually made in sections of sheet iron, cut on the circle of the cas-
ing and wide enough to take up a part of the space. These plates
are hung just above the ash pit by straps hooked on the casing
ring, or they are sometimes hung at the under side of the top
radiator. In either case their object is to compel the air to come
in contact with the furnace castings before passing into the pipes.
Care must be taken, however, that these plates do not take up

154

Furnace Heating.

too much space and thus interfere with a free and sufficient flow
of air through the furnace chamber.

COLD AIR SUPPLY.

In regard to the admission of the air supply to the furnace I
think it will be generally admitted that the best way is through a
pit under the furnace ; but circumstances will not always allow of
this, and the supply must be taken in through the casing above the
base, when a chute or galvanized iron box connecting the casing
with the cold air box should be used. This is a rectangular box of
a size that will give sufficient capacity for the air to be supplied.
This chute should never be higher than the ash pit of the furnace
and of a length that will keep the wooden box connecting to it at
least i foot away from the furnace at the nearest point. To make
this chute so as to give a good connection to the casing proceed as
follows in the case of a chute 10 x 30 inches to fit a 4O-inch casing :
Provide a sheet of iron of a width that will make the box the re-
quired length. From one end mark off 10^/2, then 30, then 10
and last 31 inches, as shown in Fig. i. Draw lines at these marks,

A*--.

Fig. 1. — Marking Sheet for Air Chute.

as shown. With trams set to strike a 4O-inch circle strike circle,
touching at points as shown. Cut notches i inch deep as shown
and turn lock V> inch wide up square on the ioJ/2 -inch end ; also
y2 inch lock way over on opposite end and in opposite direction.
Now brake up the square at dotted lines and double seam at cor-
ner. Next run one of the circled edges through the crimper y^
inch deep ; then with a .mallet on a stake turn this edge up square,
which will bring a flange to rivet or bolt to the casing, in which
punch holes about 4 inches apart, and punch similar holes on both
ends.

There are two ways of putting on this chute. One is to bolt
on and the other to rivet on. If riveted it must be put on before
the casing is placed around the furnace, the work being done as
follows : Set lower section of casing on the base to hold it in

Furnace Erection and Fittings.

155

shape ; then hold the box against the casing where it is to be put
on and mark all around, also mark holes for rivets. Always have
the chute close to base. Now take the casing off of base and cut
hole inside the marks far enough so that the hole will be i inch
smaller all around than the inside of chute except at the bottom,
where it is cut close to base. Now rivet on, the helper holding a
head on the inside of casing and the man riveting on the outside.
After the chute is riveted on turn the i inch that the hole was cut
smaller than the chute over on the inside of the chute, thus mak-
ing a strong and tight job. It will be observed that, there being
no flange turned on the bottom of the chute, when it was riveted
on part projected through the casing. When the casing is put in
its place this part of the chute projecting through the casing is
turned down over the inside of the base rim, making a tight con-

Fig. 2. — Side Connected Cold Air Chute.

nection on the bottom as well as the top. Care must be taken
when the inside casing is put on that an opening tlie size of the
chute is cut out opposite the opening. If the chute is bolted on it
is done in the same manner, only it can be left until the casing is
on and the balance of the job completed. The position of the
chutes should always be at the back of the furnace if possible, un-
less two cold air boxes are used, one entering on each side.

But if only one box is used and it is necessary to put it on
the side the chute should connect to the casing in the form shown
in Fig. 2, in order to give as nearly equal distribution of air as
possible. To mark out chute of this style proceed as follows for

156

Furnace Heating.

a chute 10 x 30 inches to fit 4O-inch casing, opening at side on line
with back : Strike a 4O-inch circle .on bench or floor and from any
point on the circle draw tangent line a b indefinitely; 30 inches
from and parallel to a b draw line c d at least 12 inches long. At
right angles to c d draw line c b. For stretch out let a b, Fig. 3,
represent one end of sheet of iron from which chute is to be made,
itf/2 inches from and parallel with a b draw line c d, equal to c d}
Fig. 2. At right angles with c d draw line to e on a b; 30 inches
from and parallel with c d draw line / g, equal to line a b in Fig.
2 ; 10 inches from / g draw line h i of same length. Let / k repre-

V i k

Fig. 3.— Pattern for Side Chute.

sent end of sheet 31 inches from h i. From / on line / k mark
point /. Now with trams set for 4O-inch circle strike circles d g
and f /. Then will e d g i I j a be the pattern required. Brake at
lines c, f and h and double seam at corner, and you have a chute
the required shape. Turn flange and put on as described for the
ordinary chute, only that the flange must be turned on the side of
the chute that will bring it on the required side of the furnace.

COLD AIR BOX.

If the cold air box in the cellar is made of wood, nothing
but thoroughly seasoned matched lumber should be used. For
if green or poor lumber is used it will soon shrink apart, and
thus not only will it be useless as an air conductor, but it will allow
the dust from the cellar to enter and be distributed all over the
house, thus giving rise to the oft-repeated remark that furnaces
are always dirty. Care also should be taken in building the box to
see that its capacity is not reduced at any point of its entire length
by cleats or braces inside or by making angles that reduce its effi-
ciency, for the capacity of a box or pipe is no larger than its meas-
urement at its smallest point.

A much more tight and satisfactory duct attends the use of

Furnace Erection and Fittings.

157

galvanized sheets and then is entirely under the control of the
furnaceman as to angles, shape, size, dampers, and other details.

CASING TOPS.

" Casing tops " are called by some " bonnets," by others
" hoods/' etc. For uniformity we will designate them " bonnets."
There are two regular styles of bonnets in general use, known as
the flat and pitched bonnet. The flat bonnet is one that is made
low or flat with the intention of taking the pipes from the top of
it, while the pitched bonnet is made higher so as to take the pipes
from the sides. In both the sides or body is made more or less
flaring, the difference being that in the flat bonnet the sides are
made low in order to get as much room as possible on the top for
the pipes, while the pitched bonnet is made high enough to take
the pipes from the sides.

The hight of the flat bonnet is determined by the hight re-

Fig. 4.— Pattern for Part of Bonnet.

quired for air space between the castings and the bonnet, it being
understood that the top of the casing is on a line with the top of
the furnace castings. The air space between the top of the cast-
ings and the top of the bonnet varies somewhat according to the
size of the furnace. In all sizes up to and including a size requir-
ing a 4O-inch casing a 6-inch space will be sufficient ; up to and in-

158 Furnace Heating.

eluding a 46-inch casing a 7-inch space; above that size from 9
to 10 inches.

The hight of the pitched bonnet will be determined largely by
the size of pipes to be taken from it, as it is necessary to have the
bonnet high enough to take out the largest size pipe to be used,
and as the space required between the casting and a pitched bon-
net is greater than for a flat bonnet the above rule governing the
hight of the bonnet provides for this, for ordinarily the larger the
furnace the larger will be the pipes used. It should be noticed
here that the net slant hight of the pitched bonnet should be about
2 inches higher than the largest pipe to be used in order to give
room for dovetailing in the collar.

In the construction of these bonnets the first thing to be done
is to strike out a pattern for a section of the side or bodv. To do
this proceed as follows : Required, a flat bonnet for a 4<>inch cas-
ing, the flare or pitch of the side to be an angle of 23 degrees, or 5
inches to the foot (which makes a good proportioned bonnet),
and 6 inches high. This will call for a body 40 inches in diam-
eter on the bottom and 34 inches in diameter at the top. As the
body of this bonnet cannot be cut in one piece it must be made in
sections, and these sections, to have as little waste as possible,
should be cut across the sheet and at the same time be as large as
possible. As galvanized iron sheets 30 inches wide are about as
wide as are generally used we will suppose that size for cutting
the pattern. Referring to Fig. 4, let a b c d represent sheet of
iron 30 inches wide ; 6 inches from and parallel with the end a b
draw line e f, which will represent the hight of body. From a
on line a b mark point g one-half the diameter of bottom of body,
in this case 20 inches; 17 inches from e on line e f mark point h,
which will represent one-half the diameter of top of body. Draw
line from g through h until it intersects edge of sheet a d at i.
Now with i as center describe arcs / k and e h. Then will / k h e
represent a section of body required. But as this section is not the
full width of the sheet, by extending arcs / k and e h to edge of
sheet at m and n and drawing line from n to i we will have a sec-
tion cutting the full size of the sheet. This description will ap-
ply to the cutting of a pattern for a pitch bonnet as well, or for
any similar pattern that is the frustum of a cone.

Furnace Erection and Fittings.

159

We now have the pattern for the bonnet, but before we pro-
ceed to construct one we must determine what kind of a flat bon-
net we will have, as there are several ways of making the bonnets,
One way of making them is to double seam a straight rim 2l/2 of
3 inches wide on the bottom to fit the casing ring, and double seam
a sand ring on the top, as shown in Fig. 5. Another way is to
crimp the bottom to fit the casing ring, and double seam on the

Fig. 5.— Bonnet with Double
Seamed Bottom.

Fig. 6,— Bonnet with Crimped
Bottom.

sand ring, as shown in Fig. 6 ; others peen on the sand ring. The
bonnet, as shown in Fig. 5, makes a very strong and pretty bon-
net, but it requires so much labor in its construction that in these
times of sharp competition beauty must be sacrificed and economy
practiced wherever possible. The bonnet shown in Fig. 6 an-
swers every purpose, and can be made much more cheaply. To
make this bonnet rivet together enough sections to reach around
the ring to be used, with some to spare, draw around the ring
tight and mark, add I inch to this to allow for top, cut off and rivet
together. Crimp the bottom edge i inch deep, sufficient to make
it fit the ring snug. As every furnaceman knows, casing rings
vary in size for the same size furnace. Hence the crimped edge
is very convenient, as a little tapping with a hammer will vary it
to suit the variations in the rings.

After the bottom edge is crimped and fitted to the ring turn an
edge on the top with the small turner. If it is intended to double
seam the sand ring on turn a small edge, then get out the cover
so it will fit nicely with a small edge and peen it on. Now for the
sand ring : Get out strip 2^4 inches wide, and long enough to reach
around the top of the bonnet. Draw it tight around the edge of
the cover, just peened on, and mark where it laps. Laying the
strip out on the bench, measure back from mark just made 2.^/2
inches, making it that much smaller, and cut off. Notch for wire
and a i-inch lap. Put in the rod or wire desired, and form up
and rivet together with two rivets. Now run the edge not wired

160 Furnace Heating.

through the turner, and mark lightly % inch deep, then crimp
quite heavy to this bead. Next with mallet and some solid stake
lay off as a flange and square the part crimped. On this flange
turn an edge with turner that will fit tight over edge of cover and
peen down snug. Now turn the bonnet upside down on the
bench. Then with some suitable stake or iron, held firmly in the
corner on the inside, bring the bonnet over the edge of bench, and
with mallet double seam cover and sand ring together to the body
of the bonnet.

The object of turning so wide a flange on the sand ring is that
when finished it will have a square shoulder or offset, as shown at
a a, Fig. 6, making a much neater and stronger seam than it would if
the double seam took up the whole of the flange, leaving no shoulder.

If it is not desired to double seam the sand ring on the dif-
ference in the process is simply to turn a wider edge on the body
(say 3-16 inch wide), and a similar wide edge on the cover and
sand ring and peen down well and leave it without double seam-
ing. It is not necessary to lay off quite so wide a flange on the
sand ring where if is not double seamed on. A workman who has
never made one of these bonnets may have some difficulty at first,
but with a little practice it will be found that they can be made
easily and rapidly.

If it is desired to double seam a straight rim on the bottom to
fit the casing ring, as shown in Fig. 5, instead of making the body
i inch larger in circumference than the casing ring, as described
above for crimping, make the body il/2 inches smaller in circum-
ference and lap i inch and lay off flange on the body and double
seam to rim in a manner similar to that described above in double
seaming the sand ring to the body. What has been said in re-
gard to the construction of the flat bonnet will apply as well to the
construction of the pitched bonnet, as shown in Figs. 7 and 8, the
only difference being that the sides are made higher.

The bonnets shown in Figs. 9 and 10 are also very similar, the
only difference being that the top or cover of the bonnet is con-
caved and has no sand ring. The covers to these bonnets are
usually pitched from 5 to 6 inches to the center. This provides
room for sand, and at the same time its form is such as to have a
tendency to more equally distribute the air to the pipes.

Furnace Erection and Fittings.

161

The style of bonnet shown in Fig. 10 for a pitched bonnet is
the most practicable and at the same time the most economical
style of all. It is simple in construction and quickly made, an-
swers every purpose and looks well when done. In making this

Fig. 7. — Pitched Bonnet with
Double Seamed Bottom.

Fig. 8. — Pitched Bonnet with
Crimped Bottom.

style of bonnet proceed the same as described for other styles of
bonnet up to the point where the edge is to be turned on the body
for the cover. Turn this edge now of a good width (say ^ inch),
square and smooth, and proceed as follows to make the cover : Re-

Fig. 9.— Bonnet with Concave
Top.

Fig. 10.— Bonnet with Concave
Top and Crimped Bottom.

quired, a cover 30 inches in diameter with 6-inch pitch to center.
Referring to Fig. IT, draw line a b, equal to one-half the diam-
eter of cover when finished, including edge of same width as on
body. Let a c represent the pitch, then will c b be the slant hight.
With c as center and c b as radius strike circle, as shown. Now
with dividers set to a & step from b along circle six times to d.
Draw line d c, cut out circle. Allow for lap and cut out piece
d c b, join d b, and rivet together. Turn the edge and snap on
bodv and peen down tight and smooth.

There are occasions when special bonnets are required, as, for
instance, a bonnet from which one large pipe equal to the full
capacity of the furnace is to be taken from the top. If the work-
man has become familiar with the principles involved in the above

162

Furnace Heating.

described bonnets he will have no trouble in applying them in the
construction of any special bonnet he may require. For making
a bonnet of this kind follow the instructions given above for strik-
ing out the pattern for the body of a bonnet, making the top I
inch in diameter larger than the diameter of the pipe required to
be taken from it. Then in place of putting on a cover peen or
double seam on (double seam is best in this case) a collar to fit the

Fig. 11.— Pattern for Concave Top.

pipe required, in the same manner as described above for putting
on a sand ring.

COLLARS.

The next thing in order to be considered will be the collars for
connecting the pipes to the bonnet, and we will take up the styles
to be used on the flat bonnet first. There are several ways of
making these, all of which are so simple that our only excuse for
describing them is that this article may reach some one that is in
need of help in the rudiments of furnace work. One way of mak-
ing these is to form up and rivet together a strip of galvanized
iron about 3 inches wide to fit the required size pipe. Run through
the large turner about y2 inch from edge, throwing a heavy bead
or swedge out on the ring. Notch about ^4 mcn wide up to bead.

Furnace Erection and Fittings. 163

Cut hole in bonnet that will just fit collar at bead. Insert the
notched edge and drive them over tight on the inside of the bon-
net. Another way is to turn a flange of, say, V^ inch wide on one
end of the collar, then rivet a strip of i% inches wide on the in-
side of this end of the collar, allowing it to extend about Y^. inch
beyond the flange. Notch to flange and put in bonnet as de-
scribed above. Collars for the flat bonnet can be completed in the
shop and put in the bonnet on the jobs as required.

Collars for the pitched bonnets have to be made somewhat
different. They must be made to fit the side of the bonnet, and,
as it is not known what angle the pipe will assume for which it is
intended, it must be fitted on the job, hence they must be made
longer to allow for trimming. For all sizes up to and including
10 inches they are made 9 or 10 inches long, larger than that 12
inches long. They are usually made flaring so they will nest to-
gether, the small end to fit the pipe. They should be riveted to-
gether, but so riveted as to be smooth on the outside to allow the
pipe to slip over without catching on the rivets. This is done
by forming the burr from punching on the inside, and then ham-
mering the rivets down flat without using a set. These collars
are supposed to be dovetailed in on the job in the following man-
ner : First, the collar must be trimmed to fit the bonnet at an angle
to correspond with angle of the pipe for which it is intended. This
is best done by drawing a line or wire from a point opposite the
center of the register box to the point on the bonnet where the
collar is to be put in.

After the collar is trimmed to fit, mark around it on the bon-
net with pencil and cut out the hole. Now with dividers mark
around l/2 inch from the trimmed end and notch J^ inch apart to
this mark. Then with pliers turn every other one of these notches
out square. Then insert the remaining ones in the hole in the
body and hammer them over tight on the inside of bonnet. Time
was when they were taken on the job and fitted and marked, and
then taken back to the shop and flanges turned on them and strips
riveted in and taken back and put in the bonnet. That made a
good job " all right.'' But then that was " befo' de wah," when
furnace work was not done for nothing and a year's supply of
coal thrown in.

164 Furnace Heating.

The constant cry heard by the workman nowadays is, " Get
there," and he must provide ways and means to do it. It is con-
sidered quite a " trick of the trade " to trim a collar for a pitched
bonnet and do it nicely and quickly. Again, it requires some skill
and good judgment to locate the collars properly. For there are
several things to be taken into consideration : First, they must be
so located that all the pipes can be taken out, which is sometimes
quite difficult to do. Then they should as far as possible be so
located as to have the pipes run direct to the box or stack without
an angle. Again, the pipes having the most work to do should
have the preference of location on the furnace. It is not so diffi-
cult to locate the collars on a flat bonnet, as the pipes can be swung
around one way or the other as desired. They should, however,
be put in a circle around and as close to the center as possible. If
it is desired to give some long pipe a decided preference the collar
may be put in the center. Avoid having a collar directly over
the feed chute. The hottest part of the furnace is usually at
the back side; put the longest pipes there if possible.

In connection with putting in collars in pitched bonnets it
is recommended that the collars should all be trimmed and fitted
and marked before any of the holes are cut, so that if necessary
to make a change it can be done, and after all are fitted the
bonnet may be taken off and the collars put on the cellar bottom.
But before taking the bonnet off be sure to mark it in some way
so it will go back in exactly the same place. The next fitting in
order is the round pipe and elbows.

It is impossible to show every type of fitting in the limited
space available in this chapter. Those who would like further
information on the layout and erection of furnace work are re-
ferred to "Piping and Heavy Iron Work" and "Furnace and
Tinshop Work,' ' two of the volumes in the series entitled
" Practical Sheet Metal Work and Demonstated Patterns.' ' The
"New Worker Pattern Book" also gives the principles by which
any pattern problem may be developed.

STOCK FITTINGS.

Two groups of stock fittings for furnace work are shown in
group Figs. i2a and 126.

Furnace Erection and Fittings.

165

The dampers shown herewith are particularly recommended
as often the ones made in the shop are not equipped with a
suitable device for adjusting them in any desired position.

Casing Collar,

Style A,
For Top of Bonnet

Casing Collar, Style C.-For
Side of Straight Bonnet

Hot Air Pipe Damper

Style B,

For Side of Bonnet,
Having 15 or 60 degree Pitch.

Adjustable Elbow

Group Fig. 12o. — Stock Fittings.

The adjustable elbow is also a very convenient fitting, and
register boxes for setting in the floor may as well be purchased
ready made.

These stock fittings will be found very convenient and will

i66

Furnace Heating.

save much labor which under certain conditions can be better
employed otherwise, then in case of a fall rush of furnace business
these fittings enable the furnace man to make prompt deliveries.

No.5
G.oup Fig. 126. — Stock Fittings.

The clips shown in Figs. 4 and 8 are perhaps the most con-
venient method of holding in place registers which must be set
vertically.

Furnace Erection and Fittings. 167

GENERAL HINTS ON FURNACE ERECTION AND PIPING.

The following methods are used quite generally in the Central
and Western States :

CASING.

The first fitting to be used after the furnace has been set, is
the casing. This applies to the furnace that is generally used,
and called a Portable Furnace, by which is understood a furnace
encased in sheet iron jackets or casings. The question of merit
between the portable furnace or one encased in masonry, will
not enter into this subject. The style of the casing is governed
by the construction of the furnace. Most furnaces, have the
base ring and one casing ring through the center of the furnace,
and another below the bonnet.

The width of the sheets that the different sections of the
casing is to be made of, is governed entirely by the size and
style of the furnace, and the position of the casing rings.
Extra care should be given to the base ring of the furnace to
see that this is absolutely plumb and level, as this is the real
foundation of the casing itself. Much depends on the casing
being made absolutely tight, as fine ashes, dust and dirt will pass
through very small openings in the casing, and from there be dis-
charged through the pipes and into the rooms above.

Complaints of dirty furnaces, can generally be attributed to
the careless casing of same.

The first, or lower section of the casing should be cut the
proper width between the casing ring and the second ring. If a
solid front is on furnace, it should be attached to the one side,
properly bolted up, and brought around the casing ring very
tightly, and bolted to the other side. The same process should
be followed with the second section. The second section of the
casing is generally made double, with some furnaces a double
ring is sent, so that an inside casing can be put in place. Where
a double ring is not sent, corrugated metal should be bolted or
riveted to the inside of sheet. The object of this inner casing is
to prevent the heat from penetrating the casing, and being lost in
the cellar, the space between this double casing or the outside cas-
ing and corrugated sheet, allows a circulation of air between them.

1 68 Furnace Heating.

HOOD OR BONNET.

The upper section of the casing is the hood or bonnet. This,
in every instance, comes above the front and therefore is complete
in itself. Much depends on the construction of the furnace
bonnet. Experience has shown that the bonnet should be of
sufficient height to act as a reservoir for the accumulation of
warm air to be distributed through the pipes that are attached to
same. Bonnets are made in many different styles, some with
straight sides and flat top, others with straight sides and concave
tops. Some are made with flaring sides, and these have both flat and
concave tops. Bonnets with flaring sides are used more generally
on account of the angle of same, providing a proper elevation for"
the pipes. With the flat top bonnet, the collars can be taken
out of the top. This is very satisfactory in high cellars, but can
not very well be used in shallow cellars. The top of the bonnet
should be at least 8" over the castings. Where collars are taken
from the side, the bonnet must be high enough to admit of the
larger pipe that may be used, and care should be taken that the
collars are all as near to the top of the bonnet as possible, and
that all of them are on a line at the top.

To prevent the heat from going through the top of bonnet,
it is generally covered with sand or an asbestos fibre.

CASING COLLARS.

The next thing to be considered is the casing collar. This
is a short tube or pipe to be attached to the bonnet of the casing.
They are usually made flaring, so that the pipe will readily fit
the same. They should not be any larger than is necessary to
make a good joint; anywhere from 2" to 6" is the proper length.

After cutting the sheet of the proper length for the size of
collar to be made, it should be formed and riveted, and an edge
turned on the large end, with turning or burring machine. Next
a strip of metal should be inserted in the large end, projecting
about 1A" below the edge. This strip to be riveted. Then the
projecting part of this strip should be notched in, so that it can
readily be turned over on the inside of the bonnet.

For flat tops, this casing collar is made straight.

For side^collars, this should be cut to the radius of the casing
ring.

Furnace Erection and Fittings. 169

Formerly, it was a practice to fit these casing collars after
the furnace had been set, cutting them, so that they would fit in
a straight line from the furnace to the register. To do this it
was necessary to go back to the turn on the edge, and rivet in the
extra strip. This required considerable waste of time, and as the
price of labor is the one important factor in the furnace construc-
tion, it is now policy to cut the casing collar on the same angle
and if necessary use an extra elbow or angle.

TO ATTACH COLLARS TO BONNETS.

Hold the completed collar to the bonnet, and scribe on the
inside of same, cut out this opening, and insert the projected edges,
bend them over on the inside, drawing them up tightly. This
will leave the formed edge on the outside, and another flange on
the inside will make the casing collar absolutely tight.

Continue with all of the collars in the same order, spreading
them around the casing as much as possible, and as near as can
be on a straight line to the register.

ELBOWS.

All right angle elbows should be 4 p.c. To prevent the neces-
sity of carrying different angles, these should be made adjustable,
and can then be turned to any point that may be required. The
size of sheets that these elbows are to be cut from, is the same as
round pipe. Very seldom an elbow is used next to the casing
collar, but in nearly every instance, it requires one at the register
or connection on pipe leading to the rooms to be heated. Pipe
and elbows should be put together properly, to prevent any dust
or dirt getting into the joints, and attached to the collar at one
end and to the register box or boot at the other.

These cellar pipes usually are covered with asbestos paper,
which insulates them and at the same time, covers any opening
that may otherwise be left at the joints.

When floor registers are used, a register box should be made
to fit the register or border. These boxes can either be made
with a flat bottom and collar of proper size, attached, or can be

170

Furnace Heating.

made funnel shaped, from a square top to the round collar at
bottom.

SIDE WALL REGISTERS.

Side wall registers are more generally used for first floor work
than they were formerly, and many different styles and sizes of
same are being made. The use of the side wall register is, without
doubt, a success. They are made to fit the ordinary studdings,
and to project from the wall enough to make the bottom of the
opening of proper size and capacity to admit for the pipe necessary
to heat the room. These boxes or heads, must be made to fit

Damper—-'

Fresh Air Inlet •
*>'mesh wire

Overhead Co/d Air Duct
with full supply of outside
or inside air at all times

Fig. 13o. — General Arrangement of Ducts and Registers.

the register. This can be of single material, either tin or galva-
nized iron or of double construction. The latter is now more
generally used, and proper fittings are made by different manu-
facturers for this purpose, which can be bought at less cost than
they can be made in the ordinary tin shop.

The connection to these boxes or heads, is by means of a boot,
which can be made either straight, with a round collar at the
bottom, as shown in Fig. 20, or, if an offset is used, as shown in
Fig. 21.

COLD AIR CONNECTIONS.

In Fig. 130 is shown the method, of a leading furnace manu-
facturing company, of making the connections for the cold air
when the overhead scheme is adopted, which would be so in the

Furnace Erection and Fittings. 171

majority of cases. If the re-circulating system is not used the
circulating register is omitted and a damper attached so that
cellar air can be taken into the cold air box.

MAKING PIPE.

We will now consider briefly the making of warm air pipes.
Where any large amount of round furnace pipe is to be made I
think it economy to use the standard sizes of pipe stock found in
the general market. The sizes are as follows: For 8-inch, 20x26^;
for q-inch, 2ox 29^/2; for jo-inch, 20x323/2; for 1 2-inch, 20x38^.
These sizes make a joint of pipe of the respective sizes 20 inches
long without waste and may be had in 1C or IX gauge. They
are also very convenient sizes for general use in the shop. How-
ever, if a small dealer does not wish to carry such a variety of
stock then the most economical method of making all sizes of pipe
from one size of stock is to groove together a number of sheets,
say 25, and roll up in a roll. Set the gauge on the squaring
shears for the required size and have the roll on the floor in front
of the shears. Let the man take the end of the roll and put
through the shears and the helper hold it against the gauge, re-
peating the operation until the roll is cut up. As one end of the
joint must be cut a trifle smaller than the other for the small end
it will be necessary from time to time to cut off a small piece to
square the sheet, but this waste will be but a mere trifle and will
include all the waste there is. Having cut the sheet turn the locks,
form up groove and solder together in lengths of four joints each.

Almost every man has a way of his own for soldering pipe.
Some solder it on the bench and some use a trough, either an-
swering the purpose for small quantities. But if a large amount
of pipe is made it will pay to construct a device by which it can be
neatly and quickly done, and such a device can be made by any
f urnacemen as follows : Construct two frustums of cones of gal-
vanized iron with base 13 inches in diameter and top 6 inches in
diameter and 7 or 8 inches high, as shown at A and A, Fig, 136.
Fasten a head on each end of these cones. Before putting on
punch a hole in the center of each that will admit a piece of i-inch
gas pipe. After the heads are in place fasten in a piece of i-inch
gas pipe long enough to be soldered to both heads of the cone and

172

Furnace Heating.

extend from the bottom 3 or 4 inches. Now from i-inch lumber
construct two brackets, as shown at B and B. Let the uprights
be, say, 10 inches high and the bottom of one of them 4 inches
long and the other equal to the upright and 5 or 6 inches wide. In
the center of the uprights and 7 inches from the bottom bore a
hole just large enough to admit the pipe in the cones. In the one

Fig. 136.— Device for Soldering Pipe.

with the short arm bore hole for 3/2 -inch bolt, as shown, with
which to fasten to the bench. In the other cut a slot in the center
of the bottom j£ inch wide and, say, 4 inches long. Provide a
bolt with a lever nut with which to fasten this one to the bench.
Now fasten the one with the short arm firmly to the bench and
the other at a distance from it that will allow a length of pipe be-
tween after the cones are inserted in the ends, holding them just
snug enough to turn and not slip, and fasten with the lever nut.
It may be found necessary to put a boss or washer around the pipe
where it extends through the bottom of the cone to avoid too much
friction against the bracket in turning.

I have given the outlines of a crude device of this kind, know-
ing that many improvements will suggest themselves to any one
who attempts to make one and also knowing that whoever suc-
ceeds in perfecting one will be well repaid for his trouble when he
comes to use it. It will be observed that the sizes of these cones
are such that all sizes of pipe from 8 to 12 inch may be soldered on
the device. The slot in the one bracket is to allow the bracket to
move back and forth to allow the length of pipe to be put on and
taken off.

It is hoped that the suggestions contained in these pages will
stimulate the reader to devise ways and means to facilitate the
installation of furnace heating plants.

Furnace Erection and Fittings.

173

ELBOWS.

As furnace pipe after it is made cannot be put up without el-
bows, we will consider them next. All right angle elbows for fur-
nace work should be four-piece elbows, to provide for an easy flow
of air. They should be made strong and neat and at the same time
by methods permitting rapid work. The first requisite in making
elbows is a perfect pattern, and to procure such a pattern for a
four-piece elbow proceed as follows : Upon any horizontal line set

Fig. 13c. — Elevation of Four-Piece Elbow.

off the diameter of the elbow required, as a b, Fig. 13^. Upon
line a b extended establish point c at a distance from b equal
to one-half the diameter of the elbow. With c as center and c a
as radius describe arc a d. At right angles with line a b draw
line cd. Now divide arc a d into six equal spaces, as i, 2, 3, 4,
5, d. Draw lines from i, 3, 5 to c. With c as center describe
arc b e. Then a deb will represent the elevation of the elbow
required. As a four-piece elbow is composed of two half sections
and two whole sections, a i 8 b will be the first half section ;
1378 the first whole section; 3567 the second whole section,
and 5 d e 6 second half section.

174

Furnace Heat-ing.

As we shall have no use for any of the lines in developing the
pattern except those composing the first half section, we shall give
no further attention to the others. To develop the pattern proceed
as follows : At right angles to a b drop lines a f and b g indefinite-
ly. At a convenient distance from a b describe half circle of diam-
eter of elbow, and draw line h i at right angles with a f, cutting
ths center of circle. Divide this half circle into any number of
equal spaces, as I, 2, 3, 4, etc. Then draw lines parallel with b g
and a f from points 2, 3, 4, 5, 6, cutting line i c, as shown. Now
draw any horizontal line, as A B, Fig. 14, equal in length to the

'1334567654321
Fig. 14.— Pattern for Elbow.

circumference of the elbow. At right angles to this line erect cen-
ter line C. Divide the spaces on each side of the center line C into
as many spaces as there are spaces in the half circle, numbering
them from i to 7, and draw lines indefinitely from these points at
right angles with A B. Now with dividers transfer the distances
from the points on line a b Fig. 13$, where lines i, 2, 3, 4, etc.,
cross to points where same lines cut i c to corresponding lines and
numbers in Fig. 14, as shown. Then a line drawn with free hand
through these points will be the pattern for first piece or half sec-
tion of elbow. It is well in developing that part of the pattern as
represented in Fig. 14 to use a sheet of iron or tin. Then when
this first section is obtained it may be cut out and all the others
marked from it by simply turning it over and allowing for the
distance required in the throat (which should not exceed il/4
inches) and mark around. For tin elbows it is necessary to use
only one section of the pattern.

Furnace Erection and Sittings. 175

If there is but a small number to be made use the first section,
as follows : Get out the sheet of tin of the required length and
width and cut with small and large end. Let the pattern represent
the first section of the large end. Lay the pattern on the sheet, al-
lowing for the length of the large end. Then turn the pattern
over, allowing for the width of the throat, and mark around the
upper side, repeating the operation until the last piece, leaving the
length required for small end. If a large number of elbows is to
be made it will pay to construct a device for marking them more
rapidly, which may be made at very little expense, as follows :
From a sheet of galvanized iron cut a third section of the elbow,
allowing the ends to extend l/2. inch longer than the regular pat-
tern. For convenience in handling this pattern should be stiff-
ened in the following manner : Cut a piece of galvanized iron of
the same shape as the pattern, but about }4 mch smaller all around.
This is bumped up with the raising hammer until it assumes an
arched form. It is then fitted to the flat surface of the pattern and
soldered all around the edges. Before the raised part is soldered
to the pattern an opening is cut in it oval in shape, so that the fin-
gers of one hand may be inserted for holding the pattern firmly to
the sheet to be marked.

In connection with this pattern a board with pins is made for
marking the different sizes, as shown in Fig. 15, and may be con-
structed as follows : Provide a board 36 inches long, 18 inches
wide and I inch thick. Across the board at the proper distances
from the ends let in flush with the top three strips of iron I inch
wide and ^ inch thick, as shown by A A, B B, C. C. The distance
between the two strips A A, measuring from center to center,
should be about ^2 inch more than the circumference of an 8-inch
elbow, locks included ; strips B B, the distance equal to the circum-
ference of a Q-inch elbow, and strips C C, the same for lo-inch
elbow. Before strips are fastened in their place drill holes in
them into which No. 8 wire nails will fit snug. The first row of
holes, D D D, should be drilled so that the distance from the edge

of the board to the under side of the hole will equal the length
required at the throat of the first section and large end of the
elbow. The second row of holes, E E E, should be at a distance
from holes D D D equal to the length required for the large end

Furnace Heating.

of the second section, including the length required at the throat
F, and including also the thickness of pins. Drill holes in like

Fig. 15.— Board for Marking Patterns.

manner in strips B B for 9-inch elbows and stops C C for lo-inch
elbows.

In marking out elbows cut the sheet the required length and
width. Supposing it to be for an S-mch elbow it will lie between
the pins D D and E E in strips A A. Bring the sheet down to the
edge of the board. Lay the pattern on the sheet and against the
lower side of pins D D and mark around upper edge of pattern.
Then move pattern to the upper side and against pins E E and
mark around both edges of the pattern, and the elbow is com-
pletely marked out.

It will be observed that I have not taken into consideration the
12-inch size of elbow, for the reason that a board for a 1 2-inch
size would be so large that it would be cumbersome, and
as there are comparatively few 1 2-inch elbows used they can
better be marked out the other way. After they are marked out
they should be notched and the locks turned before they are cut
out. When grooved together the seam of the third section should
be soldered, as it is so short it is liable to slip out in turning the
edges. It is well also to tack the small end to prevent it slipping
apart when being crowded into a piece of pipe.

If the above instructions in regard to obtaining an elbow pat-
tern are closely followed it will be found to be a comparatively
short method and will produce a correct pattern for a right angle
elbow.

Furnace Erection and Sittings.

177

Tin elbows at the furnace, however, should not be at right
angles, for every pipe should have some elevation. It is best,

Fig. 16a.— Bevel Elbow.

therefore, to make the elbows at an angle of about 80 degrees.
Hence it will require a pattern made accordingly.

To cut a pattern for a four-piece elbow other than a right angle
will require a little different process, as follows : In Fig. i6a let afe
be the required angle. Bisect the angle by the line / c, which may
be done in the following manner : On lines a f and / e establish
points b and d at any equal distance from /. From these points
as centers strike arcs g and h. Then draw line / c through the in-
tersection of these arcs. Now draw line a c, whose length will
equal one and one-half the diameter of the elbow, and at right
angles to a f ; also draw line e c at right angles to e f. Now with
c as center strike arcs a e and i j, and then proceed to develop the
pattern in the same manner as described for right angle elbow.
This will give us a pattern for the regular four-piece elbow for

178 Furnace Heating.

general use. But in practice it is often found necessary to have
elbows with more bevel than this. For this purpose it will be
found that taking the first and fourth sections and putting the two
together will make a very convenient bevel, and using the first,
third and fourth pieces will make a three-piece elbow of another
very convenient bevel, and the two will meet almost any demand
that will be made for bevels.

In regard to the stock used for elbows I would say that I think
it poor practice to use the full width of the sheet (20 inches) for
any elbow up to and including 12 inches, for the following rea-
sons : If a flat top is used the elbow should be set as close to the
top as possible, hence the large end must be short. The elbow that
connects to the register box should come up close to the timber
and in order to do so the small end must be short. Of course they
can be cut off to fit, but that takes time and stock is wasted.
Hence the large end of the elbow should not be over 2 inches long
at the throat and the small end not over 4 inches long at the throat.
When elbows are made this way 14 inches will make an 8-inch el-
bow, 15 inches a Q-inch elbow, 16 inches a lo-inch elbow and 17^2
inches will make a 1 2-inch elbow. Then we have saved a 6-inch
strip from the 8-inch elbow, a 5-inch strip from the 9-inch elbow,
a 4-inch strip from the lo-inch and a 2^/2 -inch strip from the 12-
inch. Further along we will explain how these strips can be used
to advantage.

In making galvanized iron elbows I think it well to use sep-
arate patterns for each section. These should be riveted together
and the rivet holes punched in the patterns so they can be marked
at the same time the elbow is marked out. The taper for the small
end may be all in the last section and may be made as follows:
After the patterns are cut out and before the holes are punched in
the last section draw in the holes toward the outer end of the sec-
tion enough to have it fit into the large end nicely ; then form it up
and fasten it lightly and trim it until it is true across the bevel end ;
then take it apart and use it for the pattern. The object of having
the taper all in one section is that in making it there is no danger
of getting the sections mixed, and also in cutting out the pieces for
the elbow it can be done in the square sheet without regard to large
and small end.

Furnace Erection and Fittings.

179

In closing these remarks on elbows I would like to make
this observation: That I think it will pay any dealer who
intends to do any furnace work at all to take the time when
not busy and cut out a set of patterns for tin elbows from 8
to 14 inches inclusive (above that size they should be made of
galvanized iron) and patterns for smoke pipe elbows from 6 to
8 inches inclusive, and have them hung up in their place in
the shop ready for use when wanted. Considerable time will
be saved if this is done.

The next fitting in order is the register box for the floor register
in Fig. i6b. This box should be about 4^" deep when finished,
with a collar 3^" or 4" long, for all sizes up to and including 10 x

Plaster of Paris _ ,. Slate or- Iron

Floor / if required f Register t Border

Fig. 166. — Floor Register, Showing Construction and Setting.

14 inches. For 12 x 15 box they should be 6 inches deep, with
collar 4 inches long. The body of all register boxes should be as
shallow as possible and allow a free circulation of air from the
box through the register, in order that the elbow connecting the
box may come up as close to the timber as possible. A register
border should be used with all floor registers, and the box should
fit the border snugly. In order to have the box fit the border it
is necessary to make the box a little flaring. They can be made
quicker that way and make a much better box. Of course they
cannot flare much or there will not be room in the bottom for the
collar. To make such a box proceed as follows : First get out pat-
terns for one side and one end of the box, of the required depth*

180 Furnace Heating.

allowing for a ^-inch flange on the top and an edge on the bot-
tom to double seam. Allow for double lock on each end of the
long piece and single lock on each end of the short piece for double
seaming the pieces together on the corners. Cut both the long
and short piece flaring ^ inch on each side, so that when the box
is finished it will be *% inch each way smaller at the bottom than
the top. Now take the 5-inch strips cut from the Q-inch elbows,
as mentioned above, and groove them together in a strip. From
this strip cut the bodies for the boxes, reversing the pattern each
time to avoid waste. After the pieces are cut out and notched
they can be taken to the folder and the locks on the ends, the edge
for the bottom and the flange can all be turned with the folder be-
fore leaving the machine. Then double seam them together at
the corners and it is ready for the bottom. But before putting

Fig. 17.— Register Box with Collar Passed Through.

on the bottom the collar must be double seamed in. I say " double
seamed in " because a collar should never be put in any other way.
As there are several ways of double seaming in the collar I have
thought it best to describe briefly three different methods.

The first is as follows : Prepare the bottom and turn the locks
ready for double seaming on the body, then cut the hole the re-
quired size that will allow the collar to pass through snugly after
an edge has been turned up square in the same direction as the
locks for double seaming. Before passing the collar through the
hole thus prepared turn an edge on one end way over, as far as
possible, with the burring machine. Now pass the collar through
the hole until the edge turned on the collar will hook over the edge
turned on the bottom, as shown in Fig. 17, then close the locks
together with pliers, and holding on beakhorn at a} double the

Furnace Erection and Fittings. 181

seam over on the bottom. Then double seam the bottom on and
the box is completed.

The second method is to prepare the bottom as before, cutting
the hole, allowing for an edge to be turned way over and in the
opposite direction from the one shown in Fig. 17. Now turn an
edge over square on the inside of one end of the collar so that it
will hook over the edge on the bottom snugly, as shown in Fig.
1 8, and holding on the stake peen down the edges on the inside of
the collar. Then, holding on the stake at a, drive the seam over
on the side of the collar with the side of the mallet. This is a
very good method, as the seam on the inside of the collar prevents
the elbow or pipe from being forced up into the box above the
bottom.

For the third method prepare the bottom as before, cutting
the hole and turning the edge the same as in Fig. 18. Now, on

Fig. 18.— Collar with Seam on Inside.

one end of the collar, with the small turner, lay off a double edge
or flange, then with the burring machine turn an edge back so it
will hook over the edge on the bottom, as shown in Fig. 19. Now,
holding on the stake at a, with a mallet drive the seam down flat
and smooth and you have made a connection that is solid, smooth
and quickly done. I think this much the best method of the three,
and when the workman gets accustomed to doing it that way he
will be much pleased with it.

Occasions arise at times when it is necessary to use a larger
than 12-inch pipe and 12 x 15 register, such as 14 or 16 inch pipe.
As a rule it is difficult to use the larger sizes on account of the
hight of cellar. , I have found that two 9-inch pipes run to a 14 x
1 8 inch register answer the purpose and work nicely. If a larger
amount of air is required use two lo-inch pipes and 14 x 22 inch

182 Furnace Heating.

register. Another difficulty that is met at times is that owing to
conditions that cannot be overcome the furnace, of necessity, must
be so located that most if not all of the pipes run off from one side
of the furnace, and where there are a number of pipes it may be
difficult to get them out of the top. In such cases I have some-
times found it necessary to connect two registers to one pipe, and
have had success in doing so by the following method : I will sup-
pose a stack leading to room on second floor, at a long distance
from the furnace. About on a line with this pipe is, say, a 9 x 12
floor register, at a shorter distance from the furnace. I would
run a lo-inch pipe from the furnace to a point just beyond the 9 x
12 box. From the top of the pipe take a 9-inch tee direct into the
9 x 12 box, having the end of the lo-inch pipe close to the tee.
Then put a reducer in the lo-inch pipe at this point to 8 inches,

Fig. 19.— Third Method

and connect with 8-inch pipe to the stack beyond. The propor-
tions of pipes and registers can be changed to suit circumstances,
but I would not suggest putting two floor registers on the same
pipe. These and many other ways out of difficulties will suggest
themselves to the workman if he will use his head as well as his
hands and keep his eyes open.

We now come to the partition pipes or stacks for conducting
the air through the partitions to the rooms above. There are sev-
eral different styles of pipe in use and almost every shop has its
peculiar methods of manufacture. Experience has convinced me,
however, that the style of pipe best adapted to the requirements
of the work — the most readily made and the most economical in
construction — is the rectangular pipe, commonly known as square
pipe. The sizes must vary more or less according to the condi-

Furnace Erection and Fittings. 183

tions met. The partitions usually set in private houses are 4
inches, but- as the studding is cut more or less " scant," it is sel-
dom that a pipe larger than 3^ inches the one way can be used.
Hence these conditions virtually establish the size one way. As
the studding is usually set 16 inches from center to center we have
about 14 inches of space that regulates the size of pipe the other
way. There are other things, however, to be taken into consid-
eration, and a very important one is the size of stock from which
the pipe is to be made. The sizes of tin plate that is adapted to
this work are comparatively limited. Hence the size of pipe must
conform somewhat to stock at hand. As the 20 x 28 inch size of
tin plate is a regular stock size and is usually on hand with all
dealers, the sizes of pipes that I would suggest are nearly all those
that can be made from this size stock without waste. There is no
necessity of more than two sizes of pipes for 4-inch partitions and
one size for 6-inch partitions. For the 4-inch partitions 3^ x 10
and 31/2 x iijA. For 6-inch partitions $l/2 x 13, and these sizes
can all be made from 20 x 28 inch stock.

SHOES.

The first fitting to be taken into consideration in connection
with stacks is the one connecting the round cellar pipe from the
furnace to the stack and known as bottom collars, shoes, heads or
boots. For convenience we will designate these fittings as shoes,
of which there are two regular styles — namely, the straight shoe,

Fig. 20.— Straight Shoe. Fig. 21.— Offset Shoe.

as shown in Fig. 20, and the offset shoe, as shown in Fig. 21. The
straight shoe is a fitting rectangular at the top and round at the
bottom and straight at the back, as shown in elevation, Fig. 22.

184

Furnace Heating.

As shown in the elevation this shoe is composed of two pieces —
namely, the transition piece or body and the collar, the' body and
extension being made of one piece. To make this body in one
piece it is necessary to have a pattern cut expressly for it, and
being of an irregular shape I will give a brief description of how
to lay it out, as follows : Desired, a straight shoe for 3 x 10 stack,
with 8-inch collar. Referring to Fig. 23, let A B C D represent
the icctangle 3 x 10. Draw line E F through center indefinitely,
as shown. With center on this line strike an 8-inch circle tangent
to line B C at E. Divide one-half of this circle from E to F into

Fig. 22.— Elevation of Straight Shoe.

any number of equal parts, as 1,2, 3, etc., having the spaces not
larger than i inch. From the points i to 6 draw lines to corner at
C and from points 6 to 1 1 draw lines to corner at D. To develop
the pattern draw any horizontal line, as G H, Fig. 24. Erect lines
G I and J K at right angles to G H, in length equal to G F, Fig. 22.
Now with dividers transfer the several distances from C to points
i to 6, Fig. 23, to line G H, Fig. 24, as shown at G, i, 2, 3, etc.,
being sure to mark corresponding numbers to avoid confusion.
(As it will be noticed that the distance C 5, Fig. 23, has the same
position on line from G, Fig. 24, as C 3, and C 6 as C 2, care and
accuracy will be required at this point to avoid mistake.) From
these points draw lines to I. In like manner transfer distances
from D, 6, 7, 8, etc., Fig. 23, to line G H from J, as shown, and
draw lines from these points to K.

Furnace Erection and Fittings.

185

For the pattern proceed as follows, assuming that the seam is to
be in the center at back of shoe, as shown at E, Fig. 23 : Draw any
vertical line, as L M, Fig. 25, equal in length to B C, Fig. 22. At
right angles to L M, Fig. 25, draw line L N equal in length to E C,
Fig. 23. Draw line from N to M, which should be equal in length
to K. n, Fig. 24. Now with N as center and K 10 to 6 inclusive,
Fig. 24, as radius strike arcs 10, 9, 8, 7, 6, as shown in Fig. 25.
Then with dividers set same as equal spaces in half circle, Fig. 23,
step from point n, Fig. 25, to arc 10, and from 10 to 9, and so on
to arc 6. Draw line from point 6 to N. Now with N as center and
C D, Fig. 23, as radius strike arc O, Fig. 25. Draw line from this
point to point 6, which should be equal to I 6, Fig. 24. Then with

Fig. 23.— Plan of Straight Shoe.

O as center and I 5 to i inclusive, Fig. 24, as radius strike arcs 5,
4, 3, 2, i, Fig. 25. Then with dividers set as before step from point
6 to arc 5, and from 5 to 4, and so on to i. Draw line from i to O.
Now with O as center and E C, Fig. 23, as radius strike arc P, Fig.
25. Then with i as center and G F, Fig. 22, as radius strike arc
that will intersect with arc P, Fig. 25. Draw line from i to P and
from P to O. Then with free hand draw line through points I, 2,
3, etc., to ii. Then will i M L N O P be one-half the pattern for
the body of shoe, less the extension piece.

For the extension piece proceed as follows : Extend lines I P

1 86

Furnace Heating.

and M L, Fig. 25, 2 inches to Q and T. With dividers set at 2
inches and with O as center strike arc R, and with N as center
strike arc S. Draw line from Q to intersect arc R and line from
arc R to arc S and from S to T. Draw line from point where lines
cross at R to O and from S to N. Then will i Q R S T M be one-
half the pattern for the body and extension piece for the required
shoe. The other half of the pattern can be duplicated from this or
the body can be made in two pieces, which is desirable for large
size shoes. Allow for all locks on this pattern.

In making up this shoe proceed as follows : When the body has
been marked and cut out cut the extension piece, Fig. 25, from
point R to O, and from point S to N, and the same on other half
of body, and turn the locks. Then on the beakhorn stake break
from O to i and from O to 6 and from N to 6 and from N to M,

Fig. 24.— Drawings for Straight Shoe Pattern.

breaking square at points O and N, but very slightly at points i, 6
and M, and the same with the other half ; then on a round stake
form the lower end round. After it is formed up and grooved
together make the 8-inch collar about 2 inches wide and peen on
the bottom, and the shoe is complete. It will be noticed that the
corners of the extension piece will be open. But as the stack will
set over this piece the open corners will be no objection. If it is
desired to have the corners solid it may be done by making the ex-
tension piece separate and double seaming it on the body after it
is formed up. But this way of making will require much more
time in making without adding much to the value of the fitting,
and I think will not be found advisable. Made in one piece it can
be done rapidly and with little stock.

Furnace Erection and Fittings. 187

There are several styles of cleats used in making the connec-
tions with the stacks, any of which may be used according to cus-
tom or desire. In Fig. 26 is shown a very convenient cleat that is

Fig. 25.— Pattern for Straight Shoe.

used as shown in Fig. 22. In Fig. 20 is shown another style that
is much used and is riveted on as shown.

We come next to the offset shoe, as shown in Fig. 21. This
shoe will be found to be a little more difficult of construction, as
well as requiring a little more skill in developing the patterns.
But the value and usefulness of the fitting will repay the labor in
obtaining it. In order to give a better explanation for construct-
ing a shoe of this style, we will suppose a shoe to be required to fit

Fig. 26.— Cleat for Connecting with Stack.

a 3 x 10 inch stack, with 4-inch offset and 8-inch collar. Let A B
C D E F G H I, Fig. 27, represent the elevation of the desired
fitting, as shown in perspective in Fig. 21. It will be observed that
this fitting is composed of three sections — namely, the collar, the
body or transition piece, and the offset piece. A G H I repre-

i88

Furnace Heating.

sents the collar, A B F G the body and B C D E F the offset piece.
These sections are made separate and joined together after they
are made. The only difficult part of the fitting is the body or
transition from round to rectangular. It will be found in con-
structing fittings of this design that it is just as necessary to know
how to draw an elevation of the fitting required as to be able to
develop the pattern for it, hence we will take up the elevation first.
Draw any horizontal line, as A G, equal in length to the diameter
of the collar required, and at right angles with A G draw line G F
2 inches long. Parallel with and 4 inches from G F erect line E J

Fig. 27.— Elevation of Offset Shoe.

indefinitely. Parallel with and 3 inches from E J draw line D C.
It will be observed by a glance at the elevation that when this fit-
ting is placed in the position for which it is designed the point C
will be at the floor line. Hence the distance between the point C
and the bottom of the body at G should be as short as possible to
prevent the shoe from extending below the floor timbers. It will
be seen, therefore, that the line G F must be as short as possible,
and as 2 inches is about as short as can be worked that length be-
comes arbitrary for all sizes. To establish point C on line D C
set the dividers equal to D E and place them on line D C at such a
point that they will strike arc K F, and this will be point C, as

Furnace Erection and Fittings. 189

shown. Now with dividers set same as before, with # as center
strike arc L. Now draw line indefinitely from C touching arc L.
Draw line from A at an angle of 45 degrees with A G, intersecting
line C at B. Draw line from B to F. Then will A B F G be the
body or transition piece. It will be noticed that the line B F is
longer than D E, hence the rectangular end of the body will be
larger than the stack. This must of necessity be so in order not
to contract the fitting between F and C. The angle of the line A B
is not arbitrary, but can be changed to meet requirements, as, for
instance, if this same fitting require a 9-inch collar the angle of the

Fig. 28.— Elevation of Body.

line A B would have to be changed in order to avoid having B F
too long. But a little practice will make all these points clear.

As this section of the fitting is the only part requiring the de-
velopment of the pattern we will proceed to give an explanation
of the manner of doing it. To avoid confusion we have drawn a
separate elevation of the body, as shown by A B C D of Fig. 28.
With center of line A D as center strike arc A D ; divide this arc
into any number of equal spaces, as i, 2, 3, 4, etc. From these
points erect lines at right angles with and touching line A D, as
12, 13, 14, etc. From points 12 to 16 inclusive draw lines to B.
From points 16 to I inclusive draw lines to C. Then will these
lines represent the bases of sections to be used in the development

i go

Furnace Heating.

of the pattern. To produce the sections proceed as follows : Dra\\
any horizontal line, as A B, Fig. 29. At right angles with this line
at points A and B erect lines A C and B D, equal in length to one-
half the length of the long side of the rectangle end of this section.
It being in this case 10 inches one-half would be 5 inches. Now
with dividers transfer the distances from B n to 16 inclusive, Fig.
28, to A ii to 16 inclusive, Fig. 29, as shown. In like manner
transfer the distances C 16 to I, Fig. 28, to B 16 to I, Fig. 29, as
shown. From these points on line A B, Fig. 29, erect lines 12, 10,
II, 9, etc., equal in length to corresponding lines in Fig. 28; also

10 15 14 13i2 U

Fig. 29.— Drawings for Pattern

lines 16 6, 17 5, etc., to correspond with lines in Fig. 28. It will
be noticed that points I and n have no hight. Now draw lines
from points 6, 7, 8, etc., to C and from points 6, 5, 4, etc., to D.
To develop the pattern draw any perpendicular line, as A B, Fig.
30, equal in length to A B, Fig. 28. At right angles to A B, Fig.
30, draw line A C, equal in length to A C, Fig. 29. Draw line
from C to B, which should equal C n, Fig. 29. Now with C,
Fig. 30, as center and C 10, Fig. 29, as radius strike short arc 10,
Fig. 30; in like manner strike arcs 9, 8, 7, 6, Fig. 30, as shown.
Now with dividers set at the distance used in stepping the circle,
Fig. 28, step from point 11, Fig. 30, to 10, from 10 to 9 and so on
to 6. Draw line from 6 to C. Now with C as center and B C, Fig.
28, as radius strike arc D. Then with 6 as center and 6 D, Fig. 29,
as radius strike arc that will intersect with arc D, Fig. 30. Dra\*
line from point of intersection to C and 6, as shown. Now with

Furnace Erection and Fittings.

191

D as center and D 5, Fig. 29, as radius strike arc 5, Fig. 30. In
like manner strike arcs 4, 3, 2, i, Fig. 30. Then with dividers set
the same as before step from 6 to arc 5 and from 5 to 4 and so on
to i. Draw line from I to D. Then with i as center and C D, Fig.
28, as radius strike arc E, Fig. 30. Then with D as center and A
C, Fig. 29, as radius strike arc that will intersect with arc E. Draw
line from D to point of intersection at E and from E to i. Now
with free hand draw line through points i, 2, 3, etc., to n. Then
will i nACDEibe one-half the naked pattern for section i of
Fig. 27. The other half may be obtained by duplication. Notice
that if the pattern is made in one piece the seam should be at the

Fig. 30.— Pattern for Body of Offset Shoe.

back at C D, Fig. 28, or if desired the body may be made in two
pieces.

After the pattern is obtained and the body is cut from it it is
necessary to form it properly in order to have it assume the desired
shape. To form it break over beakhorn on lines i D, D 6, 6 C, C
1 1 and the other half the same. Break sharp at points D and C
and lightly at points i, 6 and n, forming the rectangle at D and
C and allowing the other end to be formed round. A little practice
will overcome any difficulty at this point. The offset section is
made in four pieces similar to the regular stack elbow.

There may be found a little difficulty in getting the angle B F,
hence we will describe the manner of getting the end piece BCD
E F. Draw any right angle, as a b c, Fig. 31. Establish point d

IQ2

Furnace Heating.

on line a & at a distance from b equal to the difference between B
and G D and C and A D, Fig. 28, as shown by dotted line B X,
Fig. 28. With this point as center and with dividers set equal to
B C, Fig. 28, or B F, Fig. 27, strike arc cutting line b c at e. Draw
line from d to e. At right angles with b c and at a distance from
e equal to the required offset (in this case 4 inches) draw line / g

Fig. 31. -Method of Getting Angle.

indefinitely. Parallel with / g and at a distance equal to D E, Fig.
27, draw line h i. Now with e as center and F C, Fig. 27, as radius
strike arc cutting line h i at i. Draw line from i to d. Let * /
equal C D, Fig. 27. Draw line from / to k. Then with
i as center and / k as radius strike arc / m. Draw line from e tan-
gent to arc / mf as shown. Then will d i j k I e be the naked pat-
tern for the end pieces for offset section. Allow for all locks.

The front and back of this section is double seamed in and then
the section is double seamed to the body at B F, Fig. 27. Before
the body is double seamed to the offset section the collar is peened
onto the body, thus completing what we have termed an offset
shoe, shown in perspective, Fig. 21.

It may be found necessary to have two special stakes to make
this shoe to advantage, as shown in Figs, 32 and 33. Both of
these stakes are very simple and can be made of wood and cast in

Furnace Erection and Fittings.

193

any foundry. Fig. 32 is a round head stake, the standard about
i]/2 inches square. The head is half circle about 5x1^ inches
and is used in double seaming the back on the offset section, and

LJ/ JL

Fig. 82.— Bound Head Stake.

will be found useful for many other fittings. Fig. 33 is a special
stake for double seaming the offset section to the body, with
dimensions about as shown.

It may seem from the length of the description of this fitting
that it will be an expensive fitting to make. But it will not be
found so by any one who will take time to work it out and get ac-
customed to making it, as a dozen of them can be made in seven

Fig. 33.— Special Stake.

hours. The shoe when done is the best fitting of the kind in the
market, and this style, together with the straight shoe, will be
found to meet all requirements for stack connections.

194

Furnace Heating.

STACK OFFSETS, ELBOWS AND TEES.

The next fitting in order will be the stack offset, as shown in
perspective, Fig. 34. This fitting, which is used, as its name in-
dicates, to make an offset in the stack, which is frequently desir-
able, can be cut from the sheet without waste. But before any-
thing can be done toward obtaining an idea of the sizes and shape
to be cut out it is necessary to know what we want, which can
only be found by drawing an elevation of the desired article.
And as it is sometimes as difficult to draw an elevation as it is to
obtain the pattern, we will give an idea of drawing the elevation
first. Required, an offset 3 x 10 inches to offset 4 inches at an
angle of 45 degrees. Draw any right angle, as A B C, Fig. 35.

Fig. 34.— Stack Offset-

B D C

Fig. 35.— Elevation of Stack Offset.

Draw line D E, in length equal to the length required for one end
of the offset and at a distance from B A equal to the narrow side
of the stack, 3 inches. Draw line C J indefinitely at a distance
from D E equal to amount of offset required, 4 inches. As the
point F on the line C J establishes the angle of the offset, and as
that angle is required to be 45 degrees, it will be seen that it must
be at a hight from dotted line K L equal to the amount of offset,
4 inches. Draw line from F to E. Draw line G H parallel with
F E at distance equal to B D. Draw line G I parallel with F J
and at a distance equal to B D and of a length equal to the required

Furnace Erection and Fittings.

195

length of the upper end of the fitting. Draw lines I J, G F and
H E. Then will B H G I J F E D be an elevation or outline of
the required offset. It will be readily seen how the angle and
amount of offset for this fitting may be changed to suit require-
ments.

To lay out the pattern for this fitting proceed as follows : As
a sheet of 20 x 28 tin is ample for this fitting complete, let A B
C D, Fig. 36, represent said sheet. It is desired that the seam
shall be in one of the wide sides of the fitting. Parallel with and

Fig. 36.-Pattern for Stack Offset.

at a distance from A B to half the width of the wide side of the
stack for which it is to be used draw line E F ; at a distance from
E F equal to the narrow side of the stack draw line G H ; at dis-
tance from G H equal to the wide side of the stack draw line I J ;
at a distance from I J equal to narrow side of the stack draw line
K L ; at a distance from K L equal to one-half the wide side of the
stack draw line M N. Draw lines O P and T U at a distance
from B C equal to D E, Fig. 35. Draw line R S at a distance
from B C equal to B H, Fig. 35. Draw lines P R and S T as
shown. Then will O P R S T U M B be the first section of the
required fitting. At a distance from irregular line O to U equal

196

Furnace Heating.

to line E F, Fig. 35, draw parallel line V W. The piece between
these irregular lines will form middle section H G F E of Fig.
35, and the remainder of the sheet will form the third section.
Allow for locks and cut on line M N. Allow J^ inch for locks on
middle section. Before cutting out sections notch at O V and
U W and turn the locks. Now cut out the sections and form
square at lines E F, G H, I J and K L. Double seam the sections
together, as shown at H E and G F, Fig. 35, and you have the
required fitting. This fitting, it will be seen, is designed to offset
a stack the narrow way of the pipe. It not infrequently happens
that it is desired to offset the stack the wide or flat way, and this

Fig. 37.— Stack Elbow.

Fig 38.— Pattern for Elbow.

offset can be laid out and made by the same process as the othei
by simply producing elevation, as Fig. 35, with the distance B D
equal to the wide side of the stack, and the distance B E, Fig. 36,
equal to one-half the narrow side of the stack and E G equal to the
wide side, and so on to the end.

The next fitting is the elbow, as shown in perspective, Fig. 37.
This fitting is very useful and simple in construction. It is made
in four pieces, double seamed at the corners. The only parts re-
quiring a pattern are the two end pieces. To mark these out pro-
ceed as follows : Draw any right angle, as a b c, Fig. 38. Then
draw right angle d e f, equal in length to the required length of
the elbow at the throat (usually 4 inches). Parallel with and at
a distance from line d e equal to the width of the narrow side of
the pipe draw line a i. Parallel with and at a distance from e f
equal to the narrow side of the pipe draw line c i. At points I

Furnace Erection and Fittings. 197

inch each way from e draw line g h. With center of this line as
center strike arc / k, touching lines a i and i c, as shown. Then
will dhgfckja represent the pattern for end piece. The object
of cutting off the corner g h is to provide a more easy flow around
the corner. Allow for locks and turn in opposite directions. Dou-

Fig. 39.— Stack Tee.

711

Fig. 40.— Pattern for Tee.

ble seam in a piece fitting the circle from a to c of a width equal
to the width of the wide side of the pipe when finished, then seam
in a similar piece formed to the shape of the throat, and the elbow
is completed. In a similar manner an elbow for the wide way of
the pipe may be constructed.

The next fitting will be the tee or branch, as shown in per-
spective at Fig. 39. It is best to have a pattern for the body of
this tee, which may be drawn as follows : Draw right angle equal
in length to the required length of the branches of the tee (usually
about 4 inches), as A B C, Fig. 40. Opposite and at a distance
from B C equal to the narrow side of the pipe draw right angle
D E F. Parallel with and at a distance from lines A B and E F
equal to the narrow side of the pipe draw dotted line M N. Par-

198 Furnace Heating.

allel with line M N and at a distance from the same equal to the
wide side of the pipe draw dotted line O P. Parallel with and at
a distance from line O P equal to the narrow side of the pipe draw
right angles G H I, J K L, with distance between the points H
and K equal to the narrow side of the pipe. Draw lines A L and
G F and C D and J L Then will ABCDEFGHIJKLbe
the pattern for the three sides or body of the required tee. Cut
out and brake square at lines M N and O P. Double seam in the
two pieces to fit the angles and of a length equal to the wide side
of the pipe, and the fitting is completed.

It is best in cutting out to cut at bevel lines at corners B, E, K
and H. These lines are found by marking back from the corner
I inch each way and drawing line from these points. The object
is to provide an easy flow for the air around these corners. A tee

Fig. 41. -Square Register Collar. Fig. 42.— Groove in Strip.

for the flat way of the pipe can be constructed in the same way,
changing the distance between C D and J I to the wide side of the
pipe and the distance between dotted lines M N and O P to the
narrow side of the pipe.

Fig. 41 in perspective represents a square register collar to
dovetail into a stack. To make this collar of a plain strip of tin
and then notch it, and in putting in the stack turn one notch in
and one out, is a very unworkmanlike manner of doing it. To
rivet a flange on all around takes too much time and is not very
neat. The best way of making these collars is as follows : Get
out strips of the required width and in length equal to one side
and one end. With the folder (which should be a 3O-inch folder)
turn an edge lengthwise of the strip ^4 incn wide and press down

Furnace Erection and Fittings.

199

flat. At this point it is necessary to describe a tool that is re-
quired to make this collar. Take a piece of cast or wrought iron
3 or 4 inches wide, and if cast iron 3 or 4 inches thick (less will do
if wrought iron), and at least 30 inches long, to a machine shop
and have a groove cut through the center of one side the entire
length, y$ inch wide and ^ inch deep. When this has been pro-
vided, set the double edge of the strip in this groove and bend back
each way and flatten down to the stake with mallet, making a
strip as shown in Fig. 42, the flange and strip being in one piece.
Brake square to the required size and double seam two pieces to-
gether at the corner. Then notch to the flange and it is ready to
dovetail in the stack, making a strong and neat job and one that is
quickly done.

Fig. 43. —Circle Top Collar.

Fig. 44. -Double Stack Head.

For making circle top collars, as shown in perspective, Fig. 43,
another device is required and may be made as follows : Take the
lower front roll of the stove pipe formers to the machine shop and
have a groove cut in it similar to the one mentioned above at about
6 inches from the end nearest the handle. Then get out two strips
for the circle top collar of a length that will bring the seam at the
top of the circle and in the center of the square end. Mark how
far on each piece it will be necessary to form it to have the two
pieces make the required circle ; then form in the rollers, allow-
ing the flange to run in the groove. If a round collar is required
it can be made and formed in the rollers in the same way, either of
tin or galvanized iron. This is an excellent way of making fur-
nace collars for flat tops.

200 Furnace Heating.

In the perspective, Fig. 44, is shown what is known as a dou-
ble stack head or side wall box. This is simply a piece of the
stack with two collars of the required size dovetailed in with par-
tition between and top end closed. The proper way to make these
is to get out a strip 20 inches wide and long enough to make the
body for the required size. Before forming up, cut out the holes
and dovetail in the collars, then form up and put in partition.
The collars should be set about 3 inches from the top, and the par-
tition should come up to the top of the collar, leaving a space be-
tween the partition and the top of the head. Then when one reg-
ister is closed and the other open the air can pass over the top of
the partition and out the other register. To close the end of this
stack head it is not necessary to solder the end piece on or double
seam it on, as is generally done, but proceed as follows : Cut out
a piece of tin i inch larger each way than the size of the head.
For instance, if the head is 3 x 10 inches, cut the piece 4x11.
Cut the corners so they will measure about ij^ inches across the
cut. Now with folder turn an ordinary lock on all four sides and
all one way. Then turn it over with locks down and turn up an
edge about 2/g inch wide on the four sides, and we have a square
countersunk end piece. Next cut each corner of the end of the
head that is to be closed straight down about % inch. Drop the
end piece into the end of the head, allowing the small locks on the
end piece to hook over the end of the head. Then with the sharp
end of the hammer drive the corners over onto the end piece and
close down tight with pliers. Finally, with pliers or mallet flatten
the edges that hook over the end of the head down tight all around,
and you have a neat, light and solid job without solder or double
seam, and one that can be done very quickly. This manner of
putting in an end piece will apply to any square pipe or box that
requires one or both ends closed.

SIDE WALL REGISTERS.

While discussing registers and register boxes, it is well to
consider baseboard registers. These are set partly in the wall
and partly on the floor, and possess the advantages of the floor
register and common wall register without having their disadvan-
tages. The deflector plates throw the air away from the walls,

Furnace Erection and Fittings.

201

thereby avoiding discoloring them. The deflector is of consider-
able importance in securing the discharge of the required amount
of air. See Figs. 45 and 46.

In Fig. 45 a cellar pipe is shown connected by the usual
elbow to a transition piece fitted to a collar in the register box;

-Wall
Register

2'*4'Stud~

Metal Lath
'and Plaster

Elbow

Fig. 45.

Fig. 46.

the box has also a collar inserted in its top for the wall pipe leading
to an upper story. Fig. 46, however, illustrates the manner in
which two registers of this kind, set in one register box, are utilized
to heat adjoining rooms on the same floor.

Circumstances and general conditions govern the methods of
making the boxes and fittings for these registers. Manufacturers
as a rule, give detailed instructions in their catalogs.

202

Furnace Heating.

FITTINGS FOR OVAL PIPES.

As oval, or strictly speaking, flat pipe with semi-circular
ends, is a shape popular with many furnace men, it would seem
advisable to discuss the making of fittings for this shape.

Naturally the shape of the riser does not govern the shape
of the cellar pipe which should be round in any case, whether for
individual piping from furnace to risers or for a trunk line system.
Of course, trunk lines of square piping as installed by many ad-

Boaf-..

Fig. 47.

vocates of this system, is a different proposition and calls for special
treatment of all fittings. About the first fitting, therefore, to be
affected by the shape of the riser is the starter or boot, and in Fig.
47 is shown a boot transforming from a round cellar pipe, to the
shape of the riser and having an offset to pass over a girder or wall.
At this point of the discussion it is seemingly advisable to
state that it is not the intention of the publisher to burden a book
of this scope with lengthy expositions of pattern drafting when
the same is more adequately presented in special books on the
science of the development of the patterns for sheet metal work.
Therefore, the readers are referred to the problem on page 124
of volume 9 of "Practical Sheet Metal Work and Demonstrated

Furnace Erection and Fittings.

203

Patterns," and problem 209 on page 393 of "The New Metal
Worker Pattern "Book," for a complete demonstration of the
method of obtaining the pattern for the fitting shown in Fig. 47.

Fig. 50.

-Beam

-Tin or
Asbestos
Lining

Fig. 49.

Many tinsmiths prefer to make a fitting like that shown in
Fig. 48, probably because of the ease in laying out the patterns

204

Furnace Heating.

and making it, although, as can be readily seen, it is not so scien-
tific a fitting as that shown in Fig. 47. The fitting shown in Fig.
48 is simply a joint of round pipe with one of its ends stopped
with a head, which can be double- seamed like the bottom of a
can and a joint of flat pipe inserted in this. Various applications
of this fitting are shown in Figs. 49 and 50.

In laying out the fitting in Fig. 51, draw the end view of the

I 1 1

1 11

*~ Half Profile
of F/at Pipe

! '

0 O 123456789 !(.

] 1

! i]

Lijjjjjj!

Half Net Pattern of Flat Pipe

Profile of •-*
Round Pipe

1 1

-Y

Hat 'f Pattern of Round Pipe
with cutout for Flat Pipe

Fig. 51.

tee, to the top of the flat pipe part, attach half of its profile as
shown. Divide the semi-circular part of the profile into equal
parts as from i to 9. Drop parallel lines down to the circular
profile. Continue the line 10-0 to the right and stop off on it
the space on the flat profile as o to 10 and drop parallel lines as
shown. Intersect these with lines drawn from the intersection

Furnace Erection and Fittings.

205

on the circular profile as shown and the usual method of tracing a
line through the points of intersection will give the net half-
pattern of the flat pipe part of the tee. To develop the pattern
of the round pipe, draw the line XY and place on it the girth of
the round pipe as A to 5, and repeat, as shown. At A and A draw
lines of a length to suit the length of pipe required, these lines to
be at right angles to the stretchout line X-Y, draw lines to connect
these and then the rectangle LMNP is half the pattern of the
joint or length of round pipe part of the tee. For the outline of
the part to cut out proceed in this fashion, at right angles to XY
and through the points i-o to i-o on this line, draw lines which
are to be of a length, each side of XY, as similar lines are in the
half profile of the flat pipe. A line drawn through these points
is the cut out as shown by R.

EASY-FLOW FITTING FOR BOOT.

Many readers would prefer a true transformation fitting in
lieu of the one shown in Fig. 48, and so Fig. 52 has been prepared

Plan

Fig. 52.

to convey an idea for a fitting that meets the requirement. The
fitting is really a three-piece elbow with the first piece being the
regulation first piece of a three-piece square elbow for a round
pipe. Similarly, the third piece is the first piece (or the third
piece) of a three-piece square elbow for a flat or oval pipe, having
the miter along the wide side of the piece. The second piece is
the transition from the shape and position of the first piece to
that of the third.

When designing this elbow it is well to bear in mind that it

206

Furnace Heating.

is usually placed in a clamped position and hence should be as
compact as possible while preserving the full capacity throughout
the fitting. The throat therefore should be as small as practicable,
and when drawing the elevation, prior to developing the patterns,
the customary quarter circle is described as at B in Fig. 53. This
quarter circle is divided into four equal spaces, as per the rule for
obtaining the rise of the miter line of elbows. From the center
draw lines through the first, second, fourth and fifth division

Half Pattern of HI
Also Stub Pattern for
three piece square
elbow, and two piece

Half Pattern of I ,
A/so Stud 'Pattern i!
for three iece

two

Fig. 53.

points (skipping the third) on the quarter circle, as for instance
B 7-0.

On line 7-0 draw the half profile of the flat pipe as shown,
continuing the line from 7 and o until they meet the miter line.
Do likewise on line B 8-14 with the half profile of the round pipe.
Connect the points of intersection on the miter lines which gives
the elevation of the elbow with attached half profiles.

Diligent search through many books on pattern cutting
reveals but one elucidation of the development of the patterns
for this object. It is on page 137 of Vol. 9 of ''Practical Sheet
Metal Work and Demonstrated Patterns," and arbitrarily shows

Furnace Erection and Fittings. 207

miter lines, whereas binding the rise of the miter lines as here
explained is more practical and economical. This exposition will
be the basis for the development of similar problems. So to
develop the patterns, divide the half circular profile into say six
equal spaces and the semi-circular ends of the flat profile into
three equal spaces, to correspond with the round profile by
having a total of six spaces in both semi-circles. Number
these spaces. From these division points in the profile draw
the lines to the miter lines as shown and connect by solid and
dotted lines.

For the pattern of the flat profile piece number three, one
proceeds like this: To the right continue line 0-7 B and place
thereon the spaces of the flat profile and drop the usual parallel
lines which in turn are intersected by parallel lines, projected
from the miter line, all as shown; which, after tracing a line
through the intersection points, gives one-half the net pattern of
piece number three. Do likewise and as shown for the pattern of
piece number one. As was mentioned, these patterns will do for
three piece square elbows, that of the flat profile, of course, is
for an elbow when the turn is along the wide side. Also, two
pieces of number one joined together will make a 45 deg. offset
for round pipe and similarly, two pieces of number three will
make an offset of 45 deg. for a flat pipe along its wide side.

Before the pattern for piece number two can be developed
it is necessary to determine the true lengths of the solid and dotted
lines of the elevation. Therefore, as in Fig. 54, draw a horizontal
line and place thereon the distances of the solid lines in the eleva-
tion, as for instance, 3 to n is 3 to n on the elevation of piece
number two in Fig. 53. Erect verticals from these points as shown.
On the first vertical the spaces of the flat profile are set, as for
example, 3 A of Fig. 53 is 3 A of Fig. 54, and so on. On the other
verticals the spaces of the round profile are set as 12 C in Fig. 53
is 12 C in Fig. 54. The same procedure is followed for the dotted
lines in Fig. 55, exercising due care to have the dotted lines join
the correct points, as shown.

The pattern may be started to suit one's fancy, still it is a
good idea to always first make the triangle representing the flat
part of the transition which gives a substantial basis for triangu-

208

Furnace Heating.

lating the more complex portions of the pattern. So then, as in
Fig. 56, draw a line of a length coincident with the length of 3/D
to 4/D of Fig. 53, From 4 in Fig. 56 describe a short arc of a
radius equal to the length of the line in Fig. 54 marked A B . And
from 3 in Fig. 56 describe an arc intersecting the one previously
drawn from 4, the radius of this arc to be equal to the length of
the line in Fig. 54 designated A/D. Connecting this point of
intersection (marked 1 1) with lines to 3 and 4 realizes the triangle
aforementioned. Now, from 4 in Fig. 56 swing an arc the radius
of which is equal to the length of the dotted line in Fig. 55 labelled
A/B. On this arc step the distance i ix/iox of the miter cut of the
pattern of piece number one in Fig. 53 and mark it 10 in Fig. 56.
From point 10 in Fig. 56 as a center describe a short arc of the
radius equal to the line on verticals 5 and 10 in Fig. 54, From

II 1213

.__--1

-ft!
I ii
Ml

Fig. 54.

Fig. 55.

Fig. 56.

4 to this arc in Fig. 56 step the distance 4x to $x on the miter cut
of pattern piece number three in Fig. 53. Continue like this
until the pattern is completed on both sides of the triangle 3,11,4
in Fig. 56 for one-half the net pattern of piece number two; re-
membering to take the spaces on the miter cuts in Fig. 53 for like
space in Fig. 56 and also that the lines 7, 8, o, 14 of Fig. 56 are
shown in their true lengths in the elevation of Fig. 53 .

In conclusion it is to be said that it can be employed instead
of the fitting depicted by Fig. 48 in the situations presented by
Figs. 49 and 50 as well as in Fig. 48. In both fittings when used
as an offset boot as in Fig. 48 a 45 deg. elbow or offset is required
for the flat pipe as can be seen. This elbow has the turn on its
narrow side, and in consequence the pattern of Fig. 53 for an off-
set will not do. By simply turning a quarter around the profile
of the flat pipe in Fig. 53 so that the long axis 0-7 instead of being
horizontal is vertical as in Fig. 5 7 ; the same procedure would then

Furnace Erection and Fittings.

209

be followed and a three piece square elbow obtained, also a 45
deg. offset, when the turn of the elbow is along the narrow side
of the flat pipe. It is an excellent idea to always have the rise of
the miter lines for the first and last pieces of a fitting of this kind,
so that it coincides with the rise of miter line of some number of
pieced elbow; for example, the miter lines in Fig. 47 were for a
four piece elbow, and having the patterns already developed for
those elbows that much labor is saved, for then those patterns
would do for these parts of the fitting.

ANOTHER TYPE OF TRANSFORMATION ELBOW.

In Fig. 5 7 is shown the different views of an elbow transform-
ing the same as that of. Fig. 52 except that the flat pipe is in a

Side Elevation End Elevation

n If Profile
'Flat Pipe

Plan

Fjg. 57.

Fig. 58.

different position, that is to say the turn or miter line is on the
short side of the flat pipe. The patterns for this fitting are
obtained in essentially the same manner as was done for that of
Fig. 52. However, Fig. 58 was prepared to show the way the
elevation is drawn so that the reader would not be confused in
the placing of the half profiles in their correct position. Pieces
one and three are parts of elbows and offsets. The patterns for
all fittings of a like nature, for example the boot Fig. 47, are de-
veloped by exactly the same procedure as outlined for Fig. 53, the
elbow shown in Fig. 57 like that in Fig. 52 can be employed as a

210

Furnace Heating.

starter or boot, providing though that no girder or wall requires
an offset, which by the way applies to Fig. 52 also. Fig. 57 can
also be used under the floor and between beams to connect different
risers as in Fig. 50 and in many other positions which no doubt will
come to the mind of the reader.

Another important fitting which is in the same category as
these is a reducing elbow for round pipes of different diameters
as is illustrated in Fig. 59. Although this fitting has no flat

Plan

Fig. 59.

triangular sections as in an oval pipe, it nevertheless has its
patterns developed by precisely the same process as the others.

FITTINGS HAVING PROFILES IN PARALLEL PLANES.

One of the most common fittings is that termed a straight
starter or boot as shown in Fig. 60. This fitting, as with the others
discussed, transform from a round shape to an oval with however
this difference, it has no turn; that is to say both pipes would be
in line or speaking geometrically, the profiles are in parallel planes
and a few fittings in this class will now be discussed.

The development of the patterns in Fig. 60 are clearly ex»
plained in Problem 188 of "The New Metal Worker Pattern
Book."

In Fig. 6 1 is shown a boot that has the profile of the flat pipe
placed centrally to the profile of the round pipe, or in other words

Furnace Erection and Fittings.

211

the long axis of the flat profile is in the same vertical plane as the
axis or diameter line of the round profile. A demonstration of
the pattern cutting for this problem can "be found in "The New
Metal Worker Pattern Book" and also is very ably discussed on
page 107 of Vol. Q of "Practical Sheet Metal Work and Demon-
strated Patterns."

Still another fitting of similar nature to these is a reducer for
round pipe. That shown in Fig. 62 is when the profiles in plan
are not concentric so that the fitting has a straight back similar

to Fig. £0. The pattern problem is for a scalene cone and is
demonstrated by many problems in "The New Metal Worker
Pattern Book." If the profiles be concentric in plan, presenting
then a fitting like that of Fig. 61 the pattern problem is then
simply a cone development. Should, however, the profiles be
eccentric in plan and so that one is outside, or partly so, of the
other the problem then becomes identical to Fig. 58 and would
be similarly developed. On page 92 of Vol. 10 of "Demonstrated
Patterns and Practical Sheet Metal Work" is presented a solution
of this problem, but it is not recommended because the intersecting
lines of the collars are parallel. This restricts the area of the
transition piece of the fitting. These miter lines should be as in

212

Furnace Heating.

Fig. 58, hence, as was said, the problem is similar and should
have its patterns developed in the same manner.

Two other fittings that come under the same classification
as the immediately foregoing are those shown by Figs. 63 and 64.
Fig. 63 makes a quarter turn in a line of flat piping for cross parti-
tions and directly in the corner, while Fig. 64 also makes a quarter
turn of a line of flat piping it does so centrally as can be seen.
This fitting is decidedly more scientific and practical than the
square box with attached collar which is so often used, and an
interesting exposition of these problems is presented on page 108

Plan

Fig. 63.

of Vol. 9 of "Practical Sheet Metal Work and Demonstrated
Patterns," it to be remembered that both fittings would be de-
veloped by the same process.

In concluding this discussion of fittings and the like it is to
be said that numerous other fittings would be presented and
discussed would space allow. Those chosen are representative
ones, and in the books referred to herein the reader may find a
large number of other interesting problems pertaining to furnace
work. The reader is also reminded that the publishers are always
anxious to assist and that they maintain a large consulting staff
of experts and will gladly help and advise readers who have prob-
lems to solve and cannot find the solutions in books already
published.

CHAPTER XII.

MISCELLANEOUS NOTES AND DATA FROM VARIOUS
SOURCES ON FURNACE HEATING.

The principle of heating a room with warm air was introduced
by Benjamin Franklin in 1742. His stove of that date contained
a chamber surrounded by iron plates and fed by a cold air box,
the openings for the escape of the warm air being in the sides
or jambs at the top of the chamber. The warm air furnace of
the present day is identical in principle, but more elaborated.

Fig. 65. — Embryo Idea of a Fan Furnace Apparatus, 1870.

In the B. F. Sturtevant catalogue of 1870 what appears to
have been the embryo idea of a fan furnace apparatus is shown.

213

214 Furnace Heating.

Fig. 65, reproduced from a cut therein shown, serves very clearly
to give an idea of the arrangement whereby the heated air from
the hot air furnace was to be drawn through a connecting pipe
to the fan and thence discharged to any desired point. There
appears, however, to have been no general application of this
style of apparatus. — The Metal Worker.

In an article entitled " Early Hot Air Furnaces " in The
Metal Worker the writer stated that "it is probable that the
modern hot air furnace is the development of a large cast iron
stove placed in a brick chamber, having one or more registers
directly above it. Just who was the first man to improvise this
heating apparatus, or when it was done, is difficult to learn, al-
though a great many people would be willing to thank him for
the excellent heating system which has been developed from his
experiment. The date, while it cannot be fixed with certainty,
was in all probability prior to 1836. There is an impression among-
many of the older hot air furnacemen that experiments in this
line were numerous in the vicinity of Hartford, Conn., and along
about 1840 a number of different hot air furnaces are known to
have come into existence. The construction of the early furnaces
shows that the principle of heating with hot air had received con-
siderable study, and that some of the experimenters had a keen
appreciation of the principles involved and also the necessity of
making economy and efficiency go hand in hand."

In this article a cut of the Culver furnace, made in 1845, *s
shown with a firebrick firepot. The products of combustion were
carried from the top of the radiator to a series of pipes at the
back, so arranged that an indirect draft could be effected by forc-
ing them to pass down one pipe and up another, until the final
outlet was reached ; or, by opening a damper a direct draft could
be secured. The furnace was used with brick setting. Another
pattern of the furnace was put on the market in 1846 with a cast
iron firepot; This furnace had a cast iron radiator at the back of
the furnace, through which the products of combustion were
forced to pass, the durability of cast iron as compared with
wrought iron for withstanding the moist air of the summer season
having been noticed.

Another illustration shows a furnace in which the products of

Miscellaneous Notes and Data. 215

combustion pass up through tubes. A cut of a furnace popular
in 1860 is shown, having vertical wings or flanges cast on the
firepot to give extended surface. The products of combustion
pass to a large radiator, then called the Globe crosshead radiator,
above the combustion chamber and to a supplementary radiator
with diving flue.

CAUSES OF FAILURE IN FURNACE HEATING
SYSTEMS.

C. E. Oldacre, in an article in The Metal Worker, writes as
follows :

After investigation of hundreds of heating plants, running
well up into the thousands, I assign the following as the principal
causes of many failures in hot air heating that have occurred in
the past — but not occurring so frequently as we more closely study
and clearly understand our various undertakings :

Furnace too small.

Furnace improperly located.

Draft not sufficient.

Cellar pipes not properly arranged.

Cellar pipes not properly proportioned.

Cellar pipes too small.

Some cellar pipes too large.

Insufficient pitch to cellar pipes.

Friction from the use of two-piece elbows.

Too much friction in various fittings used.

Failure to use fittings that provide easy turns.

Too much friction at bottom of stack or flue.

Lack of protection to cellar pipes when subject to currents of
cold air.

Lack of fresh air duct.

Insufficient size of fresh air duct.

Lack of means for adjusting fresh air duct.

Fresh air duct taken from wrong side of house.

Fresh air duct taken from a point affected by adverse air
currents.

216 Furnace Heating.

Fresh air ducts closed entirely by slides.

Fresh air ducts wrongly connected to furnace.

Improper arrangement of return duct, where used in connec-
tion with fresh air duct.

Screen of too small mesh used over fresh air inlet.

Heat flues too small.

Heat flues improperly located.

Heat flues not proportioned to their work.

Heat flues not protected in outside walls.

Heat flues not protected at other cold points.

Heat flues diminished in size at various joints of stack.

Heat flues diminished in size in changing shape of same.

Heat flues diminished in size by register body projecting too
far into flue.

Heat flues diminished by too small dampers.

Too many heat outlets on one stack.

Registers too small.

Lack of ventilation.

Too much cold air entering through loose fitting doors and
windows.

DIRECTIONS FOR SETTING AND PIPING FURNACES.

The following directions are reprinted by permission from the
catalogue of the L. J. Mueller Furnace Company :

Determining the proper size and location of furnace and regis-
ters, also size of air conducting pipes, is a matter of judgment in
each special instance, the successful operation of the plant depend-
ing on these important requisites. The construction and exposure
of the building, prevailing winds and climatic conditions, also
favorable or unfavorable location of the furnace and registers
must all be considered. In all instances a furnace a size larger
than absolutely necessary will be more economical, more durable
and in every way more satisfactory than one just large enough,
to do the work required.

Miscellaneous Notes and Data. 2 1 7

The furnace should be placed as nearly central to the rooms
to be heated as possible, favoring that direction from which the
prevailing winds blow. In setting the castings see that they are
perfectly level on the foundation, and that the faces of the mouth-
pieces of ashpit and feed section, dome, or that of radiator, as the
case may be, are plumb, so that the door shield will properly fit
against them. Mix dry cement with water to the thickness of
mortar. Thoroughly cement all joints with this, excepting the
flanges on the door shields ; on these use asbestos cement. Spread
this carefully around the shield flanges and also in the cup joint
where the shields join ; then place the shields in position and draw
them up tightly and evenly with bolts. See that the smokepipe
fits tightly over the smokepipe collar, and do not allow it to project
into the chimney flue. Before connecting the smokepipe with the
chimney see that there is a good draft and that the flue is clear
of obstructions, such as brick, mortar or soot. Carefully line with
tin all woodwork in close proximity to the smokepipe, leaving
space for circulation of air around it.

Registers without valves must always be used where
but one is installed for each furnace, or where several registers
are placed in the same room, taking the entire capacity of the
heater.

Collars attached to the side of the hood in case of port-
able furnaces, or connected to the inner brick wall in case of
brick set furnaces, must be placed close to the top, and have
their upper sides on a level with each other, irrespective of
their size.

The warm air pipes in the basement should be straight and
have all the elevation possible (not less than I inch to I foot). If
necessary to make turns, avoid all sharp angles. The only power
that moves air through the pipes is that caused by the tendency
of heated air to rise ; avoid horizontal and crooked pipes. Protect
all warm air pipes from cold air currents, because these will chill
the pipes and stop the circulation of air within. Pipes exposed to
cold currents, or where they pass through cold rooms in the base-
ment, should be made double or wrapped with air cell asbestos
paper. Provide all warm air pipes with dampers close to the
furnace.

2 1 8 Furnace Heating.

The partition pipes or stacks must be made double, or, if single,
covered with asbestos paper to protect the woodwork, and also
prevent the loss of heat. We recommend double pipes. The
stacks should be connected to the basement pipes by means of
shoes or boots. In case a chimney flue is used for a warm air
duct, a single tin pipe should be placed inside of it. Warm air
pipes should not be placed in outside walls. Stacks leading to
the second floor can be about 25 per cent, smaller in capacity than
the warm air pipes connecting them to the furnace, on account
of the increased velocity of air in vertical pipes.

The cold air supply, if taken from the outside, should enter
preferably that side toward which the prevailing winds blow dur-
ing the winter, which is usually from the west and north. The ca-
pacity of the cold air duct should be equal to three-fourths of that
of all warm air pipes. The cold air duct must be provided with
a suitable slide or damper to regulate the supply, and the outside
opening should be protected with a wire guard of not smaller than
*/2 -inch mesh. If it can be conveniently arranged, we recommend
the building of a cold air room. This can be built of brick or
wood, but care must be taken to have it tight.

If the air supply is taken from the inside it should be of the
full capacity of all warm air pipes.

We recommend the use of both outside and inside air, enabling
the user in severe weather and during the night to use inside air.
Where the same air duct is used for outside and inside air, it must
be provided with a damper or slide, so that the air can be taken
from either source.

For stores, churches, halls and other buildings where the space
to be heated is all one large room, the best and cheapest manner
of installation is to place directly over the furnace one large
register face with border, having a capacity equal to that of the
furnace, connecting the casing hood to the register border with a
discharge pipe of the same size as the register face, thus saving
the cost of a register box. When so installed the whole capacity
of the furnace is discharged through the register face, and there
being no heat lost through radiation from warm air pipes, and
but little friction to overcome, this gives the furnace a greater
capacity than it would have for dwellings. When set in this

Miscellaneous Notes and Data. 219

manner a cold air register face or faces equal in capacity to the
warm air register face should be used for conducting the cold
air from the room to be heated back to the furnace.

Remember that the successful working of the furnace depends
largely on the chimney. The furnace smoke flue should be a
separate one, with no other openings or connections, as straight
as possible, of the same area from top to bottom, extending several
feet above the highest point of the roof, and provided with an
ashpit door below the smokepipe opening. We recommend that
the furnace smoke flue be not less than 8x12 inches inside meas-
urement. However, an 8 x 8 inch flue with a good draft may
answer for heaters with an 8-inch or smaller smokepipe. No
smoke flue should be less than 8 inches in depth. Long, narrow
flues, such as4xi2or4xi6 inches, are no good.

LOCATION OF HOT AIR REGISTERS.

A writer in The Metal Worker, has this to say regarding the
location of hot air registers :

When registers are located near inside walls less pipe is neces-
sary and a sharper pitch may be obtained than when they are
placed near outer walls. On the other hand, the loss of heat
through the ceiling will be greater. This is of little consequence
except on top floors. When possible registers should be located
about midway of partition to permit the warm air to reach all
points along the exposed walls with nearly equal ease. The rapid
circulation caused by the downward currents along the cool out-
side walls, coupled with the upward current of inflowing air from
the register near inner wall, gives an even temperature through-
out the room. The current of hot air is flattened on striking the
ceiling and passes without perceptible draft over the heads of
the occupants to the outer walls. In effect this is similar to that
produced by the overhead system of heating mills, where coils
of steam pipes are hung from the ceiling a few feet from the ex-,
posed walls. Even when there are no machines or belts to stir
up the air this system works well.

Following the same theory of circulation, it is the established

220 Furnace Heating.

custom in school houses to place the warm 'air inlets on inside
walls. Those advocating the placing of registers near outer walls
may refer to the practice of so locating them in indirect steam
and hot water work. This is done, however, chiefly from con-
siderations of economy in piping, since when the stacks are placed
near the exposed walls both cold and warm air pipes may be
made very short. With a furnace system having registers simi-
larly placed the hot air pipes would stretch from one side of the
house to the other, their excessive length reducing the pitch and
increasing the friction and loss of heat.

When registers are placed below windows the upward current
of hot air meets a downward current from the glass, which tends
to retard the flow through the pipes. Back drafts through such
pipes are more likely to occur (in case the cold air box is insuf-
ficiently open) than through short pipes having a sharper pitch.

As bearing on this subject the effect of the location of direct
radiators may be cited. They are commonly placed under win-
dows: i. To counteract down drafts. 2. Because they give off
the most heat in that position. 3. Because such location seldom
interferes with the arrangement of furniture. The objections to
a furnace register location near outer walls have no force when
applied to radiators. With evenness of temperature and comfort
in rooms of moderate size and glass surface the location of the
radiator has little to do. Wherever placed the warm air will
seek the cold walls and a continuous circulation will be established.

FURNACE AIR SUPPLY.

The Metal Worker published an editorial of interest on the
above subject, which is reprinted here :

Laws to compel the change of air, in school buildings in par-
ticular, by taking fresh air from out of doors, warming it and
then sending it into a building, with provision to exhaust the air
previously contained therein, have had a noticeable influence in
the East on the method of supplying furnaces with air. It is quite
a common custom, and one that is growing, to provide the fur-
nace with a duct connected at the bottom and leading to a point
outside of the building, so that the exterior air can readily flow

Miscellaneous Notes and Data. 221

in, to pass over the heated surface of the furnace and be distrib-
uted through the building by means of the hot air pipes. The
size and location of this air supply duct have remained an un-
solved problem to many in the furnace trade, although experi-
enced men favor running it from the most exposed side of the
building and providing it with a capacity equal to from two-
thirds to three-quarters of the area of the combined hot air out-
lets from the top of the furnace. Evidently this custom is by no
means universally observed, at least in some sections, in the West.
The conditions there are somewhat different from those obtain-
ing in the Eastern part of the country. The Western winters
are apt to be more severe throughout their entire length, and the
period at which the mercury ranges below zero is much longer
extended. Consequently the heating of buildings with furnaces
is a somewhat more difficult problem in that section. Quite a
strong favor is shown to the use of return air ducts in the West.
In many instances provision is made to take some air from out
©f doors, but the damper in this 'section of the air supply duct is
frequently closed as soon as severe weather is experienced and
the supply of air for the furnace is taken from the inside of the
building. Whatever may be urged against this practice from an
advanced sanitary standpoint, the arguments are strongly in its
favor from an economic point of view. There is no question but
that the building in which the heating is so arranged can not only
be more readily warmed and the temperature more evenly main-
tained, but also with a much smaller consumption of fuel than
if the entire air supply was taken from out of doors with the
mercury from zero to 20 or 40 degrees below. The fact that this
method of using furnaces has been customary for many years
also strengthens the position of those who advocate it. When
the question of the purity of the atmosphere in -such buildings is
raised it is pointed out that the buildings are occupied by com-
paratvely few people to vitiate the atmosphere, and that a suf-
ficient change to maintain a satisfactory purity is effected through
the natural leakage around the crevices of the windows and the
building generally, in addition to the large amount of air that
will naturally be admitted through the opening and closing of
doors.

222

Furnace Heating.

INSTALLING FURNACE PLANTS IN OLD HOUSES.

The following extracts are reprinted from an article by M. L.
Kaiser in The Metal Worker:

Floor Registers. — The better way to provide the requisite
area of warm air flues for the first floor is to place the register in
the floor. The householder and his wife will sometimes refuse to
consider the placing of floor registers, however, on account of the
cutting of carpets. The fact remains that by using a floor register
the entire area of the leader pipe, whether it be 8 inches or 14

BASE MOULD

Fig. 66. — Installing Furnace Plants in Old Houses. — Method of Arranging fof
Side Wall Register.

inches in diameter, may be made available, while the maximum
available area with a partition flue placed in a house already
built is reached at 4^4 x 12 inches for the first floor and 3^4 x
12 inches for the second floor. The objection that the floor
register collects dust is also true of the wall register, as any
one who has removed a wall register which has been in use can
testify. The argument in its favor is that the floor register may
be easily removed for cleaning, while the old people of the house
will be quick to appreciate the advantage of being able to warm
the feet over the floor register.

Arranging Side Wall Registers. — To obtain the maximum
area from 4^ x 12 inches, or 57 square inches, for the
first floor partition flues, it is necessary to cut away the lath
and plaster back of the basebord, and let the asbestos cov-
ered tin pipe rest against the baseboard and flush with the

Miscellaneous Notes and Data.

223

finished wall surface. To make this effective there should
be some means of so placing the register that it will not
extend into the tin flue, as the flue would thereby be re-
duced in area just as surely as though the entire flue were the
size of the space remaining. One way to accomplish this is to
miter a I x 3 inch strip around the register opening and nailed
to the studding, with the bottom ends resting on the baseboard.
The base mold may either be finished against the strip or mitered

r¥

Fig. 67. — Side View of Side Wall Register Connection.

around it, as shown in Figs. 66 and 67. The face of the strip is
flush with the base, and the register flanges rest against the strip
at the top and sides and against the baseboard at the bottom. The
edges of he strip under the register flanges should be covered with
tin and asbestos. A convex register used in connection with this
plan entirely obviates the obstruction of the flue by the register
body.

SIZES OF SMALL PIPES BASED ON CUBIC CONTENTS OF ROOMS.

From a perusal of various rules given in manufacturers' cata-
logues, the subjoined has been prepared as representing a fair
average. These rules afford a rough check as to pipe sizes de-
termined on the basis of equivalent glass surface :

For dwellings allow i square inch of pipe area in first floor

224 Furnace Pleating.

living rooms to each 20 to 25 cubic feet of space. In second floor
sleeping rooms allow I square inch of pipe area to each 30 to 35
cubic feet. In bathrooms allow I square inch of pipe area to each
15 to 20 cubic feet of space.

For churches and halls an allowance of i square inch of pipe
area to each 40 to 50 cubic feet of space will give a rough approxi-
mation as to pipe sizes.

Examples: Living room, 16X16X10 = 2560 cubic feet.
Divide by 25 = 102 square inches. Use 1 2-inch pipe.

Sleeping room, 14 X 15 X 9 = 1890 cubic feet. Divide by 35
= 54 square inches. Use 8-inch pipe.

Bathroom, 6X9X9 = 486 cubic feet. Divide by 20 = 25.
Use 6-in, if short run.

MEANING OF "EQUIVALENT GLASS SURFACE."

In response to a question as to the meaning of the words
" wall surface," " glass surface," and " equivalent glass surface,"
and their use, this answer was given in The Metal Worker:

By measuring the length of the outside walls of the room to
be heated and multiplying by the hight of the ceiling, the wall
surface in the room is determined, after the glass surface exposed
in the windows has been subtracted from it. The glass surface
is obtained by taking the number of windows in the room and
adding together the total amount of square feet of surface pre-
sented in each. The equivalent glass surface is determined by
assuming that the cooling effect of 4 square feet of exposed wall
surface is equal to that of I square foot of glass surface. So,
after subtracting the glass surface presented in the windows from
the wall surface and dividing this amount by 4 and adding the
result to the glass surface in the windows, the equivalent glass
surface exposed by the room is found. Dividing this by the cross
sectional area of the pipe gives the ratio of pipe area to equiva-
lent glass surface (E.G.S.).

A number of furnace heating systems have been described in
The Metal Worker, and the proportion of hot air pipe area to the
space heated has been given in each case. By a comparison one
can satisfy himself as to what is the proper proportion between

Miscellaneous Notes and Data. 225

the area in the hot air pipe and the space to be heated. A definite
rule cannot be readily given, but with good heating work it will
not vary greatly for first floor rooms from I square inch of area
in the hot air pipe to from 25 to 30 cubic feet of space. We have
given the wall surface and equivalent glass surface in these cases
so that the proportion between them and the area in the hot air
pipe can be studied, as these are the factors in the work to be done
which are most important to be considered. In order to keep a
building at a comfortable temperature during the cool season it
is necessary to continually supply the heat which is lost through
the walls and through the glass, consequently it will be better
practice to consider the wall surface and the equivalent glass sur-
face than to consider the cubic space alone.

Those who are looking for a rule for determining the size of
hot air pipes required for rooms will find it much safer to reduce
the wall surface of a room to equivalent glass surface than to
follow some of the rules using cubic feet of space as a basis that
have formerly been used.

PROPORTIONS OF FURNACES AND FURNACE HE4TINQ

SYSTEMS.

The following extracts are taken from an article by J. J. Black-
more in the Engineering Magazine:

Some manufacturers advocate large firepots, others deep fire-
pots. Some use a comparatively small amount of heating surface
over the firepot, and claim that highly heated surfaces do not have
a detrimental effect on the air, while others claim that large sur-
faces over the firepot give the best results. A careful comparison
will show that the best and most expensive furnaces of all reput-
able makers have a heating surface definitely proportioned to the
size of grate, and that the proportion of heating surface is larger
than it is in cheaper grades of heaters. This indicates that large
surface areas for the air to impinge upon have been found advan-
tageous. All manufacturers are not agreed on this point, however.
It is a somewhat difficult task for the lay mind to determine which
of the various kinds is the best.

Without trying to settle the question, I will describe the condi-

226 Furnace Heating.

tions under which a furnace has to perform its work and how the
heat it gives off may be utilized. The first task of the furnace is
to burn the fuel properly — i. e.} it must have a chamber where the
various elements in the fuel and air may be united to produce
combustion. This function of the furnace has a much greater
importance than is usually ascribed to it, and, as a result, losses
from imperfect combustion are frequent. In the burning of fuel
rather more than two-thirds, under certain conditions, may be
burned to carbonic oxide, an intermediate product of combustion ;
and, unless this gas can be further converted into carbonic acid,
most of the heat which the fuel might have developed goes up the
chimney with the smoke. Carbonic oxide is a combination of
i part oxygen with I part carbon, usually written CO. The addi-
tion of i part of oxygen will complete the combustion and develop
all the heat which the fuel can yield.

If the draft of a furnace is poor, or if the firepot, or combustion
chamber, is too small, enough oxygen will not be brought into
contact with the fuel, or gases, to enable them to give off the heat
that is in them; hence it is important that a good chimney flue
should be provided, and that the furnace room should be properly
supplied with air. The furnace should have a space above the
fuel at least three times as large as the firepot, to allow the gases
room for combustion. The size of chimney required depends, of
course, on the size of the house, but a furnace should not be con-
nected to a flue less than 8 x 12 inches, and houses containing
more than 20,000 feet of space should have larger flues.

In a pound of the average grade of anthracite coal there are
about 14,000 units of heat ( i unit is the amount of heat necessary
to raise i pound of water from 60 to 61 degrees F.) In burning
to carbonic oxide (CO) from 4000 to 4500 units only are given
off ; the rest may all be lost through the fault of a poor draft or a
badly constructed furnace.

If a furnace is constructed with a large firepot and only a small
amount of heating surface above it a large portion of the heat will
be wasted (no matter how perfect the combustion may be), for
the reason that the air coming into contact with the outer surfaces
cannot carry off the heat as rapidly as it is generated, and the
surplus escapes up the chimney.

Miscellaneous Notes and Data. 227

We will now consider how the heat is taken up by the air as
it comes into contact with the heated surfaces of the furnace.
One thousand cubic feet of air at the temperature of zero weigh
'86.4 pounds, and, as the specific heat of air is 0.238 and the tem-
perature of the air delivered through the registers should be 140
degrees, there would be absorbed by 1000 cubic feet 2878.4 units
of heat, as follows : a X b X c X d = x, in which a represents
1000 cubic feet of air at zero, b the weight of a cubic foot at zero,
0.08641 ; c the specific heat of air, 0.238 ; d the number of degrees
to which the air is heated, 140 ; and x the heat units absorbed by
1000 feet of air. To change three times an hour the air contents
of a house having a capacity of 20,000 cubic feet absorbs in zero
weather 172,704 units of heat, equal to 12.33 pounds of coal per
hour, presuming no waste of heat. But even in well constructed
furnaces there is a loss of 25 per cent. ; hence it would be necessary
to burn 16.44 pounds of coal per hour to do this amount of work
in zero weather. As a fire burns actively for 16 hours and at
one-half its capacity for 8 hours in the 24, we have 20 hours at
the rate of 16.44 pounds per hour, or a consumption of 328.8
pounds per day, or, again, very nearly i ton of coal in six days.

Taking the average winter temperature in the northern portion
of the United States as 40 degrees it would be necessary to heat
the air 60 degrees, requiring 6l/2 pounds of coal per hour, or, for
200 days; 13 tons of coal.

THE INSTALLATION OF FURNACES.

The following is from a paper read by R. S. Thompson, Spring-
field, Ohio, and Jas. H. Brown, Rochester, N. Y., to the conven-
tion of National Association of Master Sheet Metal Workers,
Cleveland.

The author, W. G. Snow, recommends that without recircu-
lated air i sq. inch of grate (average fire pot area) should be pro-
vided for \Yt square feet of E. G. S.

The furnace should have i square inch of grate surface to
2 2-10 square feet of equivalent glass surface. Locate the furnace
as nearly as possible to the center of the area to be heated. This
will usually result in the pipes radiating more uniformly in all

228 Furnace Heating.

directions from the furnace and secure better results than if the
greater number are taken from one side. If found necessary to
vary this on account of chimney or other obstruction, place it to
the side of the center toward the prevailing winds.

If the furnace casing is made with a truncated cone hood, there
should be an inverted cone of tin inside the top to divide the cur-
rent of hot air and assist in distributing it to all the pipes. If a
flat top casing is used group the pipes as near the center as pos-
sible, where they will get the hottest air.

An inner lining of tin riveted to the casing will lessen the loss
of heat in the cellar, but by all means suspend a black sheet lining
about an inch from the inside of the casing. This will act as a
powerful supplementary radiator. The relative radiating power of
tin is given as 27, while that of black sheet iron is 345. As air is
heated only by contact with a hot surface, it will be seen that
these black sheets very materially increase the heating capacity
of the furnace.

The use of asbestos lining is open to objection, and it is a
question if it serves any good purpose.

The capacity of each hot air pipe should be proportionate to
the size of the room to be warmed, allowance being made for ex-
posure and glass surface. If more than one register is used on
a pipe the size should be increased proportionately. Tables are
published giving definite information on this point. A good gen-
eral rule is to allow I square inch of cross sectional area of hot
air pipe to 2TV square feet of equivalent glass surface. A
more conservative rule is given on page 225. Very good results
are obtained by the use of deflecting registers where from two to
four registers are served by one pipe.

Cellar pipes should in all cases be run straight where conditions
permit. Use elbows made with as large a sweep as possible. It is
stated that a 1 2-inch elbow with a 6-inch throat has a resistance
equal to 121 feet of straight pipe, while an elbow of the same size
with a 6o-inch radius has a resistance equal to 8 feet of straight
pipe.

The fresh air duct should have a capacity of at least two-thirds
the aggregate area of all the hot air pipes. It is good practice to
supplement this by the use of a cold air exhaust pipe from the

Miscellaneous Notes and Data. 229

hall on the first floor. If this is done the combined area of the
two should be equal to the combined area of all the hot air pipes.
The inlet should be on that side of the house which will result in
the air traveling with the prevailing winds, not against them.

A damper or slide should be provided, but it should not be
made so that the passage of air can be entirely shut off.

If the duct is run overhead care should be taken that the ver-
tical shaft does not drop too near the furnace. There is danger
that the air may become rarefied by heat radiated from the furnace
and cause a back draft or outflow instead of an inflow.

The draft of the furnace should be controlled by a lift check
damper, connected with the smokepipe. An excellent method of
attaching it is to extend the smoke tee down vertically for about
2 feet below the smoke collar and attach a QO-degree elbow on
the lower end. In this elbow place a lift check damper. In this
arrangement there is no danger of escaping gas. 'The check
damper and the direct draft in the ashpit door should be connected
by chains with a plate on the first floor, from which point they
may be operated.

Mark each hot air pipe near the furnace, designating the room
which it serves, so that the dampers may be operated in the cellar
without confusion.

Stipulate in the contract that the owner is to furnish a chimney
of good and sufficient draft.

To ascertain the wall surface in a house wholly exposed, with
no re-entering angles, add extreme length to extreme breadth,
multiply by combined hight of ceilings and multiply product by 2.

To ascertain the number of cubic feet of air per minute at a
temperature of 140 degrees required to maintain a temperature
of 70 degrees, with the outside temperature at zero, divide
the number of square feet exposed wall surface by 2. (This
approximate rule is fairly close when glass surface is equal to
about J the total exposure of glass and wall combined. — W. G.
Snow.)

To ascertain in square feet the area of air supply divide the
exposed wall surface by 600.

To ascertain the grate surface required where all outside air
is used divide the exposed wall surface by 900. When all inside

230 Furnace Heating.

air is used divide by 1500. The product is square feet of grate
surface.

To ascertain area of leader pipe for a first-floor room where
pipe is not over 15 feet long and has no bad bends, divide exposed
wall surface of such room (in square feet) by 3. The product
gives area in square inches. If pipe is over 15 feet long add 20
per cent.

To ascertain the area in square inches of the leader pipe for a
second-floor room divide the number of square feet exposed wall
surface of such room by 6, if pipe has no bad bends and is not
over 15 feet long. If over 15 feet long add 25 per cent. If over
25 feet long add 50 per cent

The area of a perpendicular stack should be two-thirds that
of the leader pipe feeding it.

A 45-degree horizontal bend in a leader pipe should be com-
pensated for by an increase of 20 per cent, in area. A 9O-degree
bend should be compensated for by an increase of 30 to 40 per
cent, in area.

In The Metal Worker, an editorial study of the rules given
above by R. S. Thompson and Jas. H. Brown was given as follows:

The rules all refer to the exposed wall surface of a building
taken as a whole. In other words, it is not necessary to measure
the windows and take the glass surface into account, and the wall
surface also into account, and finally get the area of the so-called
equivalent glass surface. Incidentally, it may be remarked, such
rules must obviously be of a more or less approximate character,
but if they can be shown to have a rational evolution, they are far
better than no rules at all.

If we let W S stand for the wall surface of the building, these
rules stated in their simplest terms are as follows :

1. (Extreme length of house -f extreme breadth) X combined hight of ceilings

X 2 - W. S.

2. W. 8. -^ 2 ~ cubic feet air per minute at 140 degrees.

3. W. S. -j- 4 = cubic feet air per minute at 210 degrees.

4. W. S. -j- 600 == square feet of air supply duct.

5. W. S. -5- 900 = square feet grate when outside air is used.*

6. W. S. -=- 1500 = square feet grate when inside air is used.

7. W. 8. of any first-floor room -v- 3 = square inches leader pipe.f

8. W. S. of any second-floor room -r- 6 = square inches leader pipe.f

9. Two-thirds of leader pipe = area of perpendicular stack.

* See discussion on page 233. t See discussion on page 234.

Miscellaneous Notes and Data. 231

The correctness of rule No. I will be apparent to any one if he
will sketch the plan of any house, providing there are no re-entrant
angles, like courts, to the building, the authors stating that the
rule was for exposed walls without re-entrant angles. Adding the
extreme length and the extreme breadth of a house gives one-half
of the distance around it, and multiplying this by the ceiling
nights gives the area of this exposed wall, while multiplying this
product by 2 gives the total wall area.

The correctness or approximation of rule No. 2 can be indi-
cated as follows: In the average type of house one-sixth of the
total wall surface is of glass. That leaves five-sixths for the area
of the exposed wall proper. As 4 square feet of wall surface in
the average building is equivalent to I square foot of glass, then
one-quarter of 5-6, or 5-24, of the entire wall surface can be re-
garded as having the same heat transmitting properties regarding
it as glass, as the whole of that same 5-6 exposed wall regarded
as it actually is. So the actual 1-6 of glass and the 5-24 that are
equivalent to glass make it that 1-6 -f- 5-24 = y% of the wall
surface, regarded as glass has the same heat transmitting prop-
erty as all of the glass and all of the exposed wall combined. It
is commonly accepted that I square foot of glass with 70 degrees
indoors and zero outdoors will lose 85 heat units per square foot
per hour. This is equivalent to 1.4 heat units per minute per
square foot. As every 'square foot of wall surface is regarded
as y% foot of glass, every square foot of the wall surface will thus
lose in a minute ^ of 1.4, or 0.525 heat unit per square foot per
minute. (The author, W. G. S., recommends J in place of f as
stated). If the air is assumed as being admitted into the room
at 140 degrees and cooled to 70 degrees, each cubic foot of
air will give up i.i heat units in being cooled the 70 degrees.
As this heat is transmitted through the walls and by the fore-
going calculation is shown to amount of 0.525 heat unit per
minute for every square foot of the wall surface, as many cubic
feet of air will be needed to provide the heat passed through
i square foot of wall surface as i.i is contained in 0.525, which
is about 0.48, or \. That is, there will be required about
\ cubic foot of air for every square foot of wall surface,
which is what the rule stated, dividing the wall surface by 2.

232 Furnace Heating,

(The author considers 210° too high a temperature at which
to supply air.)

Rule 3 may be shown approximately correct also in the follow-
ing way: If the.air is admitted into the room at 210 degrees and
leaves it at 70 degrees, it has a- range of 140 degrees for giving
up the heat necessary to offset the loss through the exposed wall,
or twice as much as needed when the air is admitted at 140 de-
grees. As the air has twice the range in temperature, it needs to
be but half as much in quantity, so that instead of dividing by 2
the wall surface can be divided by 4.

Rule 4 has the following basis : By rule 2 it is shown that the
quantity of air needed in a minute is obtained by dividing the wall
surface by 2. If the velocity of air in the supply passages to the
furnace is 300 feet per minute, the necesseary area would be de-
termined by dividing the air volume by 300. Dividing one-half
of the wall surface by 300 is the same as dividing all the wall
surface by 600.

The derivation of rule 5 does not bring the close results ob-
tained in the case of the rules preceding. As the following will
show, grates of the small size given by the rule would require a
construction of furnace and method of operation that would allow
for getting, say, 10,000 heat units from each pound of fuel burned
and of burning the fuel at a rate between 5^ and 6 pounds of coal
per square foot of the grate per hour. It is acknowledged that
some authorities give the figure 10,000 as the amount of heat that
can be absorbed from I pound of coal in a house heating appa-
ratus, but it is safer to figure on 8000 or 9000. With regard to
the development of the rule, it will be recalled in a preceding para-
graph that it was shown that the heat lost through the exposed
wall was 0.525 heat unit per square foot of the wall per minute.
It is assumed that the air is admitted into the furnace at zero de-
grees and heated to 140. The heat loss just stated is that com-
pensated by the cooling of the air from 140 degrees to 70 degrees.
The rest of the heat, represented in the fact that the air warmed
from zero escapes at 70, is equal to the amount of heat offsetting
the heat losses through the exposed wall, so that the total heat
required of the furnace is twice that represented by the losses
through the exposed wall. The total heat required is thus

Miscellaneous Notes and Data. 233

2 X 0.525, or 1.5 heat units per minute, or 63 heat units per
hour per square foot of wall surface. If we allow 5.5 pounds
of coal burned per square foot of grate per hour, the total
heat delivered into the furnace at 10,000 heat units per pound
is 55,000 heat units. As I square foot of exposed wall needs only
63 heat units per hour from the furnace, the total amount of heat
obtained from I square foot of grate is sufficient to heat 55,000 -f-
63 = 873 square feet of wall surface. If, therefore, I square foot
of grate will supply enough heat for 873 square feet of exposed
wall, i square foot of wall will need 1-873 square feet of grate,
or the total number of square feet in the grate will be found by
dividing the wall surface by 873. This, it will be seen, is nearly
equal to 900, the figure given by the authors. If each furnace
were credited with the capacity to absorb 8000 heat units from
each pound of coal, and the coal were burned at the rate of 5
pounds per square foot per hour, the grate surface would be found
by dividing the wall surface by 630, a figure which is considerably
different from 900.

If all the air is to be taken from the inside, as specified by rule
6, it would be expected that theoretically only one-half the grate
surface would be required, for the reason that theoretically the
air is heated over and over again and there is no loss by ventila-
tion, which was shown to be one-half of the total heat require-
ments. Instead of that the authors divide by 1500, which indi-
cates that the heat needed when all inside air is used is 60 per
cent, of that needed when all outside air is used.

Rule 7 indicates that a velocity of air in the leader pipe is taken
at about 200 feet per minute. By rule 2 it was shown that the air
required in the system or for a room, for that matter, is half a
cubic foot for every square foot of wall surface. At a velocity of
216 feet per minute, the required area of the leader pipe in square
feet would be obtained by dividing y2 by 216, which is 1-432
square foot for every square foot of wall surface. As there are
144 square inches in a square foot, the area of the leader pipe is
thus equivalent to 144-432, or 1-3 square inch for every square
foot of wall surface, as given by the rule. (By wall surface is
meant total exposure of wall and glass combined. This rule
checks closely with Table IV, page 45, for rooms on first floor.

234 Furnace Heating.

The author, W. G. S.) It will be remembered that this rule is for
cases where the cellar pipe is not over 1 5 feet long and has no bad
bends. If the pipe is over 15 feet long it is advised to add 20 per
cent, to the area.

Rule 8 shows that the velocity for second floor rooms is taken
at twice that for first floor rooms, in as much as the divisor is twice
that used for the first floor rooms. If the cellar pipe is over 15
feet long it is advised to add 25 per cent, to the area and if over
25 feet long to add 50 per cent.

Rule 9 is that the area of the perpendicular stack should be
two-thirds that of the leader pipe fitting it. This indicates that
the pipes are proportioned for velocities half again as great as
that in the cellar pipes — that is, 300 for first floor rooms and 600
for second floor rooms. The first figure is that commonly given
for first floor rooms, but the 600 ft. per minute is higher than
ordinarily is allowed. The modification of this rule as set down
by the authors is that a 45-degree horizontal bend in a leader pipe-
should be compensated for by increase of 20 per cent, in area and
a 9O-degree bend by an increase of 30 to 40 per cent, in area.

TRUNK LINE SYSTEM OF FURNACE PIPING.

By F. D. GODDARD.

The trunk line system of running furnace pipes is no new idea ;
it has been used for years, but has usually been done in a careless
manner and without much consideration as to maintaining pro-
portions and areas. There are, however, many things that can be
said in its favor and little to be said against it. In planning the
layout of the cellar pipes for a furnace plant where two or more
risers or uptakes are to be taken from one main cellar pipe, care
should be taken to maintain the area of the main pipe; it should
be equal to the combined area of the branches supplied. This is
an important consideration and should never be overlooked.

Another feature to which careful consideration should be given
is the importance of having the top of the line straight, avoiding
anything that will add friction or prevent a free flow of the air.
It is not good practice to combine a first floor pipe with a second
or third floor connection ; two first floor pipes may be taken from

Miscellaneous Notes and Data.

235

one main with good results. The best method is to take the
branches from the end of the main, leaving length enough in the
connections to place a damper. When it is desired to take off a
branch at some point between the end and the furnace, the con-
nection should be made with a Y-branch and the area of the
branch passing beyond should be maintained equal to what it
supplies. Under this system, when properly executed, the whole

2-8'

Area 138"

Fig. 69. — Trunk Line System.

job is simple and compact and does away with many features that
are an objection in the ordinary single-pipe installation.

Fig. 69 shows the cellar piping for an average ten-register
house, with three pipes to the first floor and seven pipes to the
rooms above. The same top is also shown having the collars for
each 'separately. It will be seen that the furnace man will have
some little maneuvering to do if he gets the collars all in and the
pipes satisfactorily run without interfering with the head room
about the furnace and getting the collars properly spaced. When

236 Furnace Heating.

collars are placed too near together or too near the outer circum-
ference of the furnace top, there is not so good a distribution of
the warm air. It is the practice of some furnace men to cut the
furnace collars in near the edge of the top. This is all wrong;
the collars should invariably be at least 4 in. from the edge.

In the illustration there are three trunk lines shown, one line
supplying three risers and two lines supplying two risers each;
this arrangement allows the use of three cellar pipes in place of
seven, clears the space above the furnace and allows room enough
to properly cut in the collars without decreasing the area capacity.
If each one of these risers had a separate pipe to the furnace there
would be six 8-in. pipes and one 7-in. pipe. The combined area
of these pipes would be 338 sq.in.; as run in the illustration there
are two n-in. pipes and one i4-in. pipe, and the combined area of
these three pipes is 344 sq.in. or 6 in. more than the seven smaller
pipes.

The outside circumference of these pipes shows a large gain in
favor of the trunk line system in less exposed surface. The com-
bined circumference of the seven separate pipes is 172 in., while
the circumference of the three main pipes amounts to 113; thus
it can be seen that the exposed surface of the three main pipes is
about 66 per cent, of that of the seven separate pipes. This is
an important gain, a gain that counts in the efficiency of the
apparatus. The larger cellar pipe contains a greater volume of
air, has less exposed surface and consequently does not cool so
quickly.

In cases where two or more branches are taken from one main
there is frequently one of the branches that does not convey the
air as freely as the others. In such a case the dampers to the
more freely working branches can be partly closed, thus forcing
the air into the weaker pipe ; this would be impossible if the pipes
were run separately. The air in a small pipe, if carried a con-
siderable distance, is liable to be cooled before reaching the point
of delivery, its volume being so small.

The furnace man who installs a few furnace plants under this
system in accordance with correct methods will be suprised at the
results, both in satisfaction to his customer and to himself. Some-
one will advance the argument that the average shop cannot get

Miscellaneous Notes and Data. 237

out the fittings for a job of this description. To such the writer
will say that there is hardly a furnace manufacturer but would
be pleased to send patterns of such fittings, and there are the pipe
and fittings concerns who would be only too glad to supply the
manufactured articles. All that is required out of the regular is
the Y-branches, two-way, three-way and four-way branches,
patterns of which once obtained can be varied to suit conditions.
As to the cost- of the trunk line system compared with the re-
gular single pipe job, the writer is not prepared to say. Investi-
gation so far as made would indicate that the trunk line costs less
after the furnace man has the patterns and understands the
method.

THE CONTROL OF AIR LEAKAGE AROUND WINDOWS.

By HAROLD M'GEORGE.

Air leakage around windows is a matter that, up to a compara-
tively recent date, has received but little attention. The heating
engineer in calculating the amount of radiation required for a
house or building has accepted an arbitrary factor for glass loss.

Early in the year 1907, H. W. Whitten, heating engineer, ran
into a circumstance that to him was of such importance that he
decided thereafter to devote his entire time and attention to win-
dow leakage and its prevention. The Mt. Royal Apartment
House, Baltimore, Md., had been built and equipped with an
efficient heating apparatus. During the first season this apparatus
was very satisfactory, supplying the necessary amount of heat
with reasonable economy of fuel.

This building is located near the Union Station, where dust and
smoke are prevalent. To overcome this nuisance the building
after standing one year was equipped with metal weather strips.
It was then observed that the temperature of the rooms was too
high and could only be reduced to normal by reducing the steam
pressure to the lowesr point consistent with circulation. Clearly
the radiation, which was barely sufficient originally, was too large
for the new condition. Acting on this assumption, the radiating
surfaces were reduced nearly 25 percent, and the difficulty rem-
edied. The result of the joint action of the weather strip .and

238

Furnace Heating.

reduction in heating surfaces was to lower the coal consumption
35 per cent. Mr. Whitten came to the conclusion that if the
stated reduction in radiation was due to the metal weather strip
then there must be some way of securing a definite formula or
calculation whereby the saving in radiation could have been
figured in the original calculation. To arrive at such a basis a
number of interesting tests were made.

The first experiment made by Mr. Whitten, in conjunction
with Ralph Collamore of Detroit, was with a double tapered sheet
iron cone; to one end was connected a motor driven pressure
blower, to the other end an anemometer; in the middle a frame
into which sash of varying clearances could be placed. Three
styles of sash were tested, one having -£$ in. clearance, which is
usually termed by builders a tight window; one of ^ in. clearance,
a loose window, such as is usually found in average house con-
struction, and one fitted with metal weather strip in accordance
with the foregoing specification. The results obtained are shown
in the accompanying table:

LEAKAGE AROUND DIFFERENT TYPES OF WINDOWS.

Air pressure
in inches.

Corresponding
wind velocity in
miles per hour.

2 x 4 ft. ordinary window. Cu. ft. of
air per minute passing through sash.

Window equipped
with metal weather
strip. — Cu. ft. of
air per min. passing
through sash.

Sash of A-in.
clearance.

Sash of j"j-in.
clearance.

0.03

7-75

7-4

0.05

10.78

15-0

....

O.O8

13.66

1 .4

O.I

15.11

24.7

* ••*

2.8

0.2

21 .61

37-4

5-3

0-3

26.48

47-4

8-4

0.4

30.37

57-4

9-6

o-5

34-18

65-1

10.7

0.6

37-44

103

73-i

ii. 7

0.7

40-45

112

80.7

12.4

0.8

43-24

121

88.1

14.0

0.9

45.86

130

95-0

15-6

I.O

48.34

137

101.4

16.1

1-3

55-19

1 60

124-5

20.7

The author (W. G. S.) would state in this connection that
when the late A. R. Wolff introduced heat loss values into this

Miscellaneous Notes and Data. 239

country based on German standards his first charts showed a
heat loss through glass of 70 B.T.U. per square foot per hour of
70 degrees difference in temperature.

Later he found it advisable to increase this amount to

85.

Now Peclets values and those of other experimenters are not
far from 70 B.T.U. per square foot of glass per hour for 70 degrees
F. difference in temperature.

Wolff's increase was evidently due, in the opinion of the author,
to the necessity of allowing for the air leakage around the
windows.

This the author has found to be a matter of extreme im-
portance.

In computing heat losses or computing the size of a furnace or
heating apparatus the kind of windows should always be taken
into consideration ; are they plain double hung, are they casement
windows or have they transoms ? Have they plain wooden sash
or is the sash of steel with steel frames? Are they fitted with
metal weather strips or are double windows used in winter?

The author has found steel sash to be very leaky and the leaks
very difficult to overcome.

Inside double windows of the casement variety are to be
recommended for use in connection with steel sash.

A liberal allowance over the usual heat loss must be allowed
for casement windows or for transoms.

When one considers the matter it is evident that the glass area
alone is a very -poor index of the heat loss through windows; for
example two rooms of the same size may have exactly the same
glass area, yet one may have twice as many windows as the other,
in which case it stands to reason that with windows equally tight
the air leakage will be much greater in the second case as in the
first due to the greater aggregate length of cracks around the
window sash.

While there appears to be no rules which make due allowance
for the lineal feet of cracks around windows the question of
the number of windows into which the glass area is divided
must receive consideration in determining the heat losses for a
room.

240 Furnace Heating.

TESTING A FURNACE PLANT IN WARM WEATHER.

The following answer appeared in The Metal Worker in reply
to the question whether a test of a warm air plant can be made
at a time when the weather is 22 degrees above zero so as to tell
if the furnace would heat the rooms to 70 degrees in zero
weather :

The subject of equivalent heating powers under different con-
ditions has been discussed in connection with direct steam and hot
water heating, but for indirect or furnace heating the problem is
more troublesome, owing to the greater number of variable factors.
Heating a house to a certain temperature during above zero con-
ditions proves little as to the heating capacity of the furnace in
zero weather unless the volume of warm air delivered is known.
If this be measured with an anemometer, then the cubic feet per
hour at register times the weight of i cubic foot of air at that
temperature times the excess of temperature over outside air
times the specific heat of air (0.238) will equal the heat units per
hour delivered to room. Combining the results of similar tests
and computations in all the rooms gives the total effective output
of heat by the furnace. If this sum total equals or exceeds the
estimated heat loss in zero weather the furnace should easily do
the work under the latter conditions. The heating surface will
then be more effective, since colder air is brought in contact with
it, and the chimney draft will be stronger. To compute the heat
loss in zero weather see page 45 of this book. Note the percentage
corrections for exposure to cold winds.

The statement is sometimes heard that it makes no difference
whether a room is heated by a large volume of warm air or a
smaller volume of hotter air. Those familiar with hot air heat-
ing know, however, that the total quantity of heat that must be
supplied to keep the house at the desired temperature is affected
to a marked degree by the temperature at which the air is de-
livered by the furnace. When air enters at 140 degrees, for ex-
ample, with outside temperature o degree, I cubic foot at the
higher temperature brings in 2.2 heat units, of which i.i are avail-
able to offset the loss by transmission, the remainder escaping
with the air at 70 degrees. With air at, say, 120 degrees, these

Miscellaneous Notes and Data. 241

figures are 1.94 and 0.8 1, respectively. To compensate for a given
transmission loss through walls and windows more air at 120 than

at 140 degrees, in the ratio of — 5— = 1-36, would be necessarv.

O.ol
1.36 X 1-94 2 64

The total heat would be as— — — — 1-— = 1.2; that is, 20

i X 2.2 2.2

per cent, more heat would be required at the lower temperature.
This shows the importance of noting the inlet temperature when
testing a furnace. A register temperature below 120 degrees in
zero weather, with the fire in good condition, indicates an over-
supply of air and consequent waste of heat, or insufficient heating
capacity.

TEST OF A FAN-FURNACE COMBINATION.

Tests of a Kelsey generator, as the hot air furnace manufac-
tured by the Kelsey Heating Company is known, were made un-
der the direction of Prof. Wm. Kent and were reported in The
Metal Worker in part as follows:

Air was supplied through a 48-inch Sturtevant disk type fan
driven by a 5 horse-power electric motor. The arrangement is
shown in Fig. 70. The horizontal cold air intake was designed
to conduct the forced supply of air to the generator [furnace]
casing, but the special arrangement was provided of taking from
this intake a supply of air to the ashpit of the generator. It ap-
pears that this scheme was chiefly to obtain additional means for
regulating the rate of combustion, or, in other words, for
maintaining the condition of the fire uniform throughout the
trials.

The means for measurement of the air handled were as fol-
lows : At a hight of 6 feet above the floor line in the vertical
downtake the cross section of the shaft was accurately measured
and laid off in 24 sections or rectangles, so that an anemometer
could be placed opposite each rectangle, and the average of the 24
readings of the instrument thereby obtained to secure a figure for
the average velocity in the downtake. It has been proved that the
air in such a passage does not travel in currents of equal velocity
over the whole cross section, and this is a common method for
eliminating the errors which would arise by taking a few readings

242

Furnace Heating.

Miscellaneous Notes and Data. 243

of the instrument at places selected more or less at random. The
cross section of this downtake, according to accurate measure-
ments, was 11.88 square feet.

To determine the variations in the pressure produced by the
fan and to see that these were not of an excessive character a
U-shape water gauge was connected to the horizontal intake, and
its fluctuations were noted every few minutes. The vertical and
horizontal air passages were carefully made of tongued and
grooved material, and in order to prevent any further leakage of
air all joints were carefully covered by strips of building paper
on both the inside and outside. The instruments were standard-
ized in order to make the measurements as accurate as possible.

The heater was a No. 30 Kelsey generator with 211 square feet
of heating surface and a grate area of 4.91 square feet. The
velocity of the air was read in each of the 24 rectangular spaces
in the main air supply shaft once an hour, and the whole was
carefully averaged to determine the true velocity, as already men-
tioned. The thermometer was placed centrally in the air intake
to record the temperature of the incoming air. Above the dome
of the generator and on a level with the top of the bonnet a ther-
mometer was hung in the center, which recorded the temperature
of the heated air emerging from the generator. The other ap-
paratus used consisted of a thermometer in the smokepipe, a
hygrodeik on the roof and another near the generator to determine
the percentage of moisture in the air both before and after passing
through the generator. Analyses were taken of chimney gases in
the usual manner.

The test was taken by what is called the standard method —
that is, having a given thickness of fire in the firepot at the be-
ginning of the test and the same amount of coal in a similar con-
dition at the end of the test, all coal supplied in the meantime
being carefully weighed. The temperatures were recorded every
15 minutes. Observations were made for eight hours on two sepa-
rate days, and the average results are shown in the accompanying
table.

This table besides giving the results of the tests on the two
days mentioned includes also a supplementary test conducted on
March 23, in which the velocity of the air was measured at the

244 Furnace Heating.

outlet of the generator. These results are tabulated under the
column marked test C.

In connection with the table the following explanations may be
in order : The third line, giving the weight of the vapor in each
cubic foot of the air is, of course, determined by multiplying the
weight of moisture which air at the temperature given can hold in
suspension by the percentage of humidity. The weight of dry air
per cubic foot for the given temperature can be obtained directly
from tables for the weight of air as varied by the temperature.
The fifth line gives the average of the readings of the anemometer
in velocity expressed in feet per hour as a matter of convenience
in making the further calculations. The loss in leakage, it is un-
derstood, includes not only the air delivered to the ashpit, but pro-
vides for leakage that inevitably took place through the boards
of the intake passages.

The numberof British thermal units absorbed by the dry air
per hour is, of course, obtained by multiplying the weight of dry
air delivered to the heater per hour as tabulated by the specific
heat of the air, and this product by the range of temperature,
which, for example, in test A was 96 degrees. The specific heat
of air, which is the number of British thermal units which are
required to raise I pound of air through i degree F., is sometimes
taken at 0.2375, so that for the 19,821 pounds dry air delivered
per hour in the case of test A the number of heat units absorbed
in increasing the temperature 96 degrees is equal to 19,821 X
0.2375 X 96 = 451,872.

In calculating the heat units absorbed by the vapor per hour it
will be interesting to note that the specific heat in this case was
taken at about 0.3, although the specific heat for water in the liquid
state is I. For example, the number of heat units absorbed in the
case of test A are 70 X 0.3 X 96 = 2016. The total number of
heat units absorbed by the mixture is, of course, the sum of those
absorbed by the dry air and those absorbed by the vapor.

In determining the number of heat units given up in combus-
tion, shown in next to the last line of the table, the number of
pounds of coal burned per hour was multiplied in each case by
14,700, the coal used being credited with a heat emitting value of
14,700 B. T. U. per pound.

Miscellaneous Notes and Data.

24c

Test A. Test B. Test C

Average temperature of the cold air, degrees F. 39 58^4 52%

Per cent, saturation or humidity of the cold air. 71 56% ....

Number pounds of vapor in each cubic foot of

the air 0.000281 0.000434 ....

Number pounds of dry air in each cubic foot of

the air 0.079004 0.075754 0.062

Average velocity of air through measuring orifice,

feet per hour 26,400 26,220 28,860

Average volume of air through measuring orifice,

cubic feet per hour 313,620 311,494 270,660

Average volume of air lost through leakage, cubic

feet per hour 62,724 62,299

Average volume of air delivered to heater, cubic

feet per hour 250,896 249,195 ....

Number pounds of dry air delivered to heater per

hour 19,821 18,878 16,781

Number of pounds of vapor delivered to heater

per hour 70 108 ....

Average temperature of the warm air, degrees F. 135 152*4 178

Average difference in temperature between warm

and cold air 96 94 125*4

B. T. U. absorbed by the dry air per hour 451,872 421,496

B. T. U. absorbed by the vapor per hour 2,016 3,102 ....

B. T. U. absorbed by the mixture per hour 453,888 424,598 502,678

Average number pounds of coal burned per hour 36 33% 38S/T

B. T. U. given up in combustion per hour 529,200 492,450 564,900

Per cent, efficiency of the generator =

B. T. U. absorbed bv mixture

- = 85.7 86.2 88.9

B. T. U. given up by coal.

The last line or efficiency of the generator is the percentage
of the heat supplied by the coal that is absorbed by the air deliv-
ered from the generator, and is consequently the quotient of the
heat absorbed by the mixture of dry air and vapor divided by the
heat given up by the coal.

The Metal Worker stated editorially in regard to the above
tests : " It is common knowledge that where, say, three furnaces
have been installed for a given building on the basis of the re-
quirements with gravity operation two of the furnaces have suf-
ficed for the severest demands when the air supply has been
forced. It will be noted that as much as 450,000 B. T. U. were
absorbed per hour from the heating surfaces in the generator in
question, which for the 211 square feet of heating surface in the
generator is 2135 B. T. U. per square foot per hour — a figure
which is remarkably high for the heat delivered by steam pipe
coils in forced blast work. The air delivery through the generator
was at the rate of over 1300 cubic feet per square foot of the

246 Furnace Heating.

heating surface per hour. How far these figures can yet be ap-
plied in calculations of heating systems without reference to the
nature or details of the furnace needs further tests of the same
praiseworthy character as those discussed."

ADVANTAGE OF AIR AT RELATIVELY LOW
TEMPERATURE.

. There are advantages in supplying air at, say, 120 degrees in
zero weather. There is less tendency for the air to remain at the
ceiling than when admitted at a higher temperature, thus promot-
ing a better circulation in the room and a nearer approach to a
uniform temperature throughout. On the other hand, the lower
the temperature of the air supply the greater must be the quantity
to supply the number of heat units necessary to make good the
loss through exposed walls and glass, consequently the more fre-
quent the air change and the greater the fuel consumption.

A source of annoyance in furnace heating systems is the con-
trol of the air supply, an insufficient supply causing injury to the
furnace from overheating and an unwholesome air supply from
the same cause.

FAN FURNACE HEATING.

George W. Kramer, an architect with a wide experience in
fan furnace heating, advocated in The Metal Worker extended
surface in the form of vertical ribs or flanges on the furnace, and
stated that by providing these the efficiency of the furnace is
greatly increased in this class of heating.

E. T. Child states in The Metal Worker as to fan motors in
connection with the fan furnace system of heating:

" The motive power of a fan may be furnished by a water or
electric motor or by a gas engine. For many reasons the electric
motor, if current is available, will be the most satisfactory. The
running cost is rarely high, the noise is very slight and if direc-
tions are followed it may be operated by attendants not at all
familiar with the detail of the construction. The motor should be
provided with a starting box, or a speed regulator. These are
ordinarily provided with what is called an automatic release, which
shuts down the motor in case the current is accidentally cut off.
This prevents the motor from being damaged by receiving the full

Miscellaneous Notes and Data. 247

current at once. In starting a motor the starter should be moved
slowly, one notch at a time, to let the motor gain speed, otherwise
there is danger of burning the motor out. In many localities, how-
ever, electricity is not available, while gas is very cheap, and gas
engines have been used to a very considerable extent for fan pro-
pulsion. When a gas engine is used care must be taken to prevent
the sound of the exhaust from being transmitted through the
building. This may be accomplished by running the exhaust pipe
to a cast iron tank or equalizing chamber, from which it may be
run into a pit or dry well of large capacity with proper outlet. In
gas countries the gas engine furnishes the cheapest power, and
for this reason it is to be recommended. Water motors are some-
times used, but the cost for water makes them practically out of
the question in large towns and cities. If high pressures and
low rates may be obtained the water motor will be found a very
satisfactory source of power."

USE OF SMALL ELECTRIC FANS IN CONNECTION WITH FURNACES.

What may be termed fan furnace heating on a small scale is
the application of a small office type fan 12 to 16 inches in diam-
eter to accelerate the flow of air in the cold air box. One large
concern engaged in furnace heating has suggested that mention
be made in revised " Furnace Heating " of this use of such fans,
stating that the work it has done in this line has been very satis-
factory, especially in cases where return air is used.

PRACTICAL APPLICATION OF A DESK FAN.

From The Metal Worker is taken the following description of
the use of a small fan in furnace heating by F. N. Jewett of the
Wagner Electric Mfg. Company, Chicago :

Having difficulties in heating some of the rooms in his house,
he devised the following scheme, which operates satisfactorily:
Herewith, Fig. 71, is given a diagrammatic cross section through
the basement and lower part of the first floor of the house. The
usual fresh air intake or cold air box, B, is equipped with a swing-
ing door weighted so as to be self-closing. Connected to this
door is a rope, which passing over pulleys and through the floor,
permits opening and closing the fresh air intake from the first
floor. At C an opening was made in the cold air box and a round

248

Furnace Heating.

galvanized iron duct 16 inches in diameter was run out and up to
a large register, A, placed in the first floor. In this duct a 1 6-inch
desk fan was installed.

With this arrangement it will be seen that should any one
room or any paYt of the house fail to heat properly, the door B
may be closed, the register A opened, and the fan operated. Then
the cold air on the floor of the house is drawn down through' the
register A, blown rapidly through the cold air box and mechan-
ically forced up through all the warm air ducts and registers, and
the heating not being dependent upon gravity, the air must go
where desired. The fan is also found of great value early in

Fig. 71. — Mr. Jewetfs Sketch Showing Fan Used with Furnace.

the morning in rapidly warming up the house, which usually has
cooled off during the night.

The experience of a neighbor of Mr. Jewett, J. R. Cravath,
Western editor of the Electrical World, whose heating system
includes the use of a furnace, is as follows :

" We find it heats the house much quicker in the morning. It
forces the air into rooms that it would be difficult to heat without
the fan. The cost of operating the fan is about the same as 16
candle-power incandescent lamp. At Chicago prices it is about
y2 cent per hour. An ordinary i6-inch electric fan is used. The
fan need not be run more than an hour or so in the morning. Then

Miscellaneous Notes ard Data.

249

all of the house will be heated in good shape all over. On ex-
tremely cold or windy days it might be necessary to run the fan
all day. The cost of installing the fan is from $20 to $25."

THE EFFICIENCY OF A DESK FAN.

From an article in The Metal Worker the following practical
suggestions are taken:

The capacity of the fan may be greatly reduced if some pro-
vision is not made to prevent eddies or back currents. This

SUCTION

Fig. 72. — Natural Movement of Air from Desk Fan.

FLA T SHEE T IKON SHIELD

SUCTION

PRESSURE

Fig. 73. — Back Flow Prevented by Shield.

SHEET IKON CONE. SHIELD

SUCTION

PRESSURE

Fig. 74.— Eddies Avoided by Cone Shield.

should be done by installing in the duct a sheet metal cross-
partition with circular opening the size of the fan. The illustra-
tions, Figs. 72, 73 and 74, will give some idea of how the fan will
act, and will be readily understood by the practical men who are
engaged in furnace heating.

250 Furnace Heating.

In the first of these is given the plan showing the side walls
of the air duct of sheet iron or wood, and showing the location of
the fan and the movement of the air from the suction side to the
pressure side. In this illustration it will be noted that the blades
of the fan throw the air off in every direction, and that when
there is no wind guard or shield there is a tendency for some of
the air to be drawn back of the fan to the side where the suction
makes it easier for the air to flow than on the pressure side, where
some resistance is being encountered.

In order to avoid this a wind guard or shield made of sheet
iron or thin board should be placed across the duct with a hole in
the center, practically the same size as the fan, and the fan should
set close as possible to it. While a perfectly straight shield or
guard will serve a good purpose, nevertheless there will still be
eddies on the pressure side of the fan at its outer edge. This, in
my opinion, however, is not sufficient to warrant any elaborate
effort to avoid it, particularly in a square duct. In a round duct
it is a less difficult matter to make a cone shield or guard extend-
ing just back of the center of the fan blade, as shown in the last
diagram. This allows the air to spread, but it directs it forward,
overcoming the eddy and possibly contributing sufficient increase
to the efficiency of the fan to warrant the employment of the cone
guard where the best results are sought or there is no effort to
confine the furnaceman in his cost.

It is possible that inasmuch as the fan will not always be in
the air supply duct, and the fire may be started up depending on
the gravity air supply, a better method of providing a fan and
wind guard would be to arrange for it to be on a special slide,
which can be slipped into or out of the air duct as may be desired.
This will avoid a reduction in the size of the air duct at the point
where the fan would be likely to interfere with the proper amount
of supply when gravity would be the only force driving the cold
air to the furnace, and under such conditions the wind guard
might severely interfere with the supply, so that some pipes on
the furnace would not work and deliver the hot air expected from
them.

Where an inside air supply is used there is advantage in setting
the fan under the return air supply register face. Then the face

Miscellaneous Notes and Data. 251

can be lifted and the fan taken out in mild weather, when the
furnace will do its work without help or the fan is needed else-
where in the house for cooling in the summer.

MISCELLANEOUS.

FIRE HAZARDS OF HEATING SYSTEMS.

The great diversity of devices and methods in use for house
\varming, due to the varying conditions of fuel supply and climatic
requirements, says Insurance Engineering, presents to the under-
writer a range of fire hazards which compels the careful study of
all the conditions to enable him to suggest to the user the proper
safeguards to prevent disaster.

The surveyor or inspector finds his attention called to problems
ranging from the old-fashioned open fireplace, with wood for
fuel, such as grace and make cheerful our country homes, to the
wood or coal burning stove, and up to the more modern and com-
plicated steam, hot water and hot air furnaces, now so common
in the equipment of city and town houses, with a sprinkling of
natural gas grates or stoves, and an occasional encounter with the
kerosene oil device, each demanding special expert knowledge as
to construction and use, and the exercise of good judgment in the
suggestions necessary to make safe such defects as may be discov-
ered upon investigation. * * * Considering the record of
fires from heating apparatus, the only conclusion to be reached
is that carelessness is at the base of each of these accidents, either
as a defect in the original installation of the apparatus, or as a
result of recklessness and the neglect in its after use and care.
With a desire to consider briefly some of the salient points of
hazard incident to the methods of house warming, we take up the
different devices in the sequence of the statistical record as above
noted.

Stoves and Stovepipes. — Where such devices are to be used
for heating or cooking, they should be free from cracks or other
imperfections which would admit of the escape of coals or sparks ;
should be set upon solid platforms of brick, or in metal pans with
raised sides and legs 3 inches in hight, in either case being large
enough to extend well in front of the ashpit and thus protect the

252 Furnace Heating.

floor. Stoves should not be set within 18 inches nor their pipes
within 10 inches of any woodwork, lath or plaster partition or
other combustible material, except when conditions will not permit
otherwise, and then all combustible material should be protected
with bright tin sheeting, with a space of not less than y2 inch
between it and the combustible.

Stovepipes should be well and frequently supported by wire ;
each joint should overlap the other toward the stove and be care-
fully riveted to prevent the escape of sparks. Fires should not be
dumped into the ashpit except upon a bed of dead ashes of not
less than 2 inches thickness. All pipes should enter the chimney
or flue horizontally and in plain sight. Pipes entering flues ver-
tically, or passing through blind or unused attics, where they
may not be frequently inspected, are prolific causes of fires, par-
ticularly in country houses.

Hot Air and Indirect Steam Heating. — The furnace should be
set upon a very solid foundation in order to prevent the sagging
or cracking of its walls. The top or dome of the enveloping walls
should not be less than 18 inches from the unprotected woodwork
or lath and plaster ceilings, and its smokepipe or flue should be
a like distance from combustibles. The ashpit should be sunken,
or the floor in front of it be of brick, stone or concrete, not less
than 36 inches wide. The inclosing walls of the furnace should
not be less than 12 inches from all combustible material,
and the inlet or cold air duct should be entirely of iron or
other metal.

Hot air flues or conduits should be made of heavy, bright tin
plates with well soldered lock seams, and be kept at least 10 inches
from all woodwork or other combustibles. Where it is necessary
to carry them through or into wooden or lath and plaster partitions
the flues should be double — /. e., one inside of the other, with an
air space of not less than l/2 inch between the two, and be properly
braced to insure rigid separation throughout. Where register
boxes are set in floors or partitions, the woodwork should be
framed around them to leave an air space of not less than 2^2
inches, and be protected by flashings of bright tin extending from
the outer edge of the register opening to and through the floor
beams or partition. Each register should be set in a frame of

Miscellaneous Notes and Data. 253

slate or soapstone not less than 2^2 inches wide and i inch thick,
firmly and well set in cement or plaster of paris.

At least one of the registers of the system should be so ar-
ranged as to insure its being constantly open, either by the re-
moval of the vanes of the valve or by wiring the same open, so
that closing would be impossible. Heater firepots should be care-
fully examined before use each season, to discover and remedy
defects due to the burning out of their walls or the destruction of
the luting at the joints of the same, whereby coals might fall into
the surrounding air space and ignite dust or other combustibles
which are liable to accumulate therein.

Natural Gas. — Heating by this means is restricted to limited
areas of the country, and when the supply is sufficient to insure
a full supply at constant pressure, the hazards of its use are quite
mild, provided the piping has been properly installed and has
passed a rigid test for leaks, and the flues for carrying off the
products of combustion have been constructed for the purpose.
When, however, the supply is weak and restricted and the conse-
quent pressure variable, the hazard of its use is vastly augmented,
for with low pressure and small supply the user is inclined to
open the valve in the supply pipe to its fullest extent in order to
secure a good blaze ; and when, later, the pressure is increased
from any cause, the small blaze is turned into one of great in-
tensity and power, and is liable to ignite combustibles at a distance.

The only practical means of reducing this hazard Ihs in the
use of an automatic high and low pressure regulating valve in
the main supply pipe whereby the flow of gas will be automatically
cut off when the pressure either rises above or falls materially
below normal, at which it is set to act. Such controlling devices
are open to purchase in the districts where natural gas is a factor,
and the use of the same should be made obligatory.

RADIATION FROM RED HOT IRON.

In answer to the query how much heat is radiated from an
iron casting weighing i pound when raised to a red heat, the
casting being incased in a sheet metal jacket with a flue to and
from so that it is supplied with fresh air, The Metal Worker gave
this statement :

254 Furnace Heating.

The specific heat of cast iron is 0.129, water being equal to I,
or i heat unit per I pound of water raised I degree F. One pound
of cast iron heated to 1000 degrees F. above the atmospheric tem-
perature will therefore contain 1000 X 0.129 = 129 heat units;
which, if transferred to air circulating through a jacket as de-
scribed, without loss by radiation from the outside of the jacket,
should heat 542 pounds of air i degree F. The specific heat of
air being 0.2377 of a unit per pound, and as about 13 cubic feet
weigh i pound, the total amount of air heated i degree F. would
be about 7000 cubic feet. This is the gross theoretical result with-
out loss of heat otherwise. The practical result will be a stream
of intensely hot air issuing from the jacket at first and gradually
cooling to normal temperature, the total radiation of which will
be, as above stated, 129 heat units.

SUITABLE SIZE COAL TO USE.

These hints as to sizes of anthracite coal are given in the cat-
alogue of the Thatcher Furnace Company :

For small furnaces, stove coal; medium furnaces, stove and
small egg coal.

For large furnaces, stove and large egg coal.

Do not use what is known as furnace coal. It is too large and
not suitable for warm air furnaces.

Use a hard grade of coal in furnaces adapted to this kind
of fuel.

INDEX

PAGE

Absolute Zero 85

Acceleration of Gravity 87

Advantages of Furnace Heating. . 31
Air, Advantages of Low Tempera-
ture 246

Air, Bad, and its Effects 93

Air Change. . 28, 48

Air, Composition of 82

Air, CO2 Standard 91

Air, Dry, Damp, etc 84

Air, Expansion of 85, 86

Air, Flow of, in Pipes 86

Air, Foul, Effect of 93

Air, Leakage around Windows. 28, 237

'Air Passages in Furnaces 7, 104

Air Pipes, Size of 44

Air, Quantities for Buildings 101

Air, Recirculation of 43

Air, Relative Volume 86

Air Supply. See Cold Air Supply.

Air Supply for Combustion 60

Air Supply per Person. ... 46, 83, 108

Air Supply Required 101

Air, Table of Velocities in Flues . . 89
Air Temperature at Registers,

27, 46, 64, 65, 246

Air Velocity in Flues 46, 51, 87

Air, Weight of 46

Air, Weight of Cubic Foot Dry . . 86

Air, Weight of Saturated 84

Angles, Method of Getting 192

Area of Cold Air Boxes 38

Area of Fire Pots 13

Asbestos 228

Ashes, Removal of 61

Atmospheric Vitiation 90

Balance of Combination Systems. 70

Bevel Elbows 177

Billings, on Ventilation 93, 99

Blackmore, J. J., on Furnace Sys-
tems 225

Boards for Pattern Making 176

Bonnet 168

Bonnets for Furnaces 158

Boots 205

Brick Settings 33, 34

British Thermal Unit

Brown, J. H., Rules by. ...

PAGE

. 26
. 230

Calculation of Heat Losses 46

Capacity of Furnaces 29

Carbonic Acid 82, 90

Casing 167

Casing Collars 168

Casings for Furnaces 152, 167

Casing Tops 157

Cast Iron vs. Steel Plate 8

Check Dampers, Gossamer 42

Chimney Flues 38

Chimney Flues, Table of 146

Church Heating 116, 127

Cleats for Connecting with

Stacks 187

Coal. . . 145

Coal Bins, Capacity of 146

Coal Consumption for Schools. . . 102
Coal Consumption per Square

Foot of Grate 24, 26, 28, 63

Coal, Suitable Sizes to Use 254

Coils, Pipe 78

Coke 145

Cold Air Box Area 38, 122

Cold Air Box Construction. . . 33, 154

Cold Air Box, Halls. . : 122

Cold Air Box, Location '. . . 39

Cold Air Box, Material 40

Cold Air Box, Window 156

Cold Air Boxes for Stores 130

Cold Air Connections 1 70

Cold Air with Return Air Connec-
tions 42

Cold Air Room 40

Cold Air Room, Schools 107

Cold Air Supply 154, 220

Cold Air Supply, Schools 108

Collars 162, 164

Combination Systems. ... 68, 79, 129
Combination Systems, Heaters for 72

Combustion Chamber 15

ombustion, Rate of . . . 24, 26, 28, 63

ondensation on Windows 91

onducting Power of Substances. 148
Conservatories, Heating 77

255

256

Index.

PAGE

Contracts and Estimates 141

Convection of Heat 25

Corridor Heater for Schools 106

Cost of Fan Operation 137

Cost of Furnaces, Comparative.. . 31

Covering for Furnace Pipes 48

Cubic Contents, Rule for Furnace

Pipes 224

Dampers, Gossamer 42

Dampers, Vent Flue 113

Depth of Fire Pots 13

Desk Fans, Cone Shields for 249

Desk Fans for Furnace Heating,

247, 249
Directions for Setting and Piping

Furnaces 216

Direct Radiating Surface Combi-
nation Systems 74

Direct Radiation, Location of. ... 68

Dome Furnaces 9

Double Windows 46

Ducts and Flues, Fan System .... 138
Dust and its Dangers 92

Early Forms of Furnaces 7

Early Heating, Fan Furnace 213

Effect of Foul Air on Health and

Comfort 93

Effect of Wind on Furnaces 7

Efficiency of Furnace 25

Elbows 169, 173, 179

Elbows, Bevel 177

Equivalent Glass Surface. 28, 46, 224

Erection of Furnaces 167, 227

Estimates and Contracts 141

Evaporating Pan 17

Expansion of Air 85

Expansion of Water 78

Expansion Tanks 78

Exposure in Relation to Cubic

Contents 27

Failures in Furnace Heating 215

Fan Capacities 136, 137

Fan Furnace, Combination Sys-
tems 132, 138

Fan Furnace Heating with Elec-
tric Fans 246

Fan Furnace System, Early

Method 213

Fan Motors, Location of 133

Fan Operation, Cost of 137

Fan System Flues 138

Fans, Types of 135, 137, 247

F'rebrick Lining for Fire Pots .... 13

PAGE

Fire Hazards in Heating Systems. . 25 1

Firepot Areas 146

Firepots 10, 13

Firepot, Two-section 10

Fires, Colors of 147

Fittings, for Warm Air Pipes . 165, 166

Fittings, Oval 202, 203

Fittings with Parallel Planes 210

Flow of Air in Pipes 86

Flue Arrangement for Halls 120

Flues, Area of for Schools 1 1 1

Flues for Furnace Systems 138

Flues for Indirect Radiation 76

Flues, Location of, for Schools. . . no
Flues, Material of, for Schools. . . no

Flues, Velocity of Air in 87, 89

Foundations 32

Foul Air, Effect on Health and

Comfort 93

Fuels 38, 145

Furnace Bonnets 158

Furnace Capacities, Cubic Con-
tents Basis 30

Furnace Capacities, on Exposure

Basis 29

Furnace Capacities, Schools 105

Furnace Capacities, Table of 122

Furnace Casings 152, "167

Furnaces, Comparative Cost of

Installation 31

Furnace Erection. . 152, 167, 216, 227

Furnace Failures 215

Furnace Fittings 152, 164

Furnace Hoods 168

Furnace Installation 216, 227

Furnace Installation in Old

Houses 222

Furnace Joints 8

Furnace Location 31, 107, 124

Furnace Management 59

Furnace Materials 8

Furnace Pan 17

Furnace Pipes, Size of 44

Furnace Pipe, Table of Sizes 45

Furnace Piping, Methods of . . . 48, 49

Furnace Pit 32

Furnace Proportions 24

Furnace System Proportions 225

Furnace Tests 62, 65, 240

Furnaces, Air Passages 7

Furnaces, Cast Iron vs. Steel Plate, 8
Furnaces, Directions for Setting.. 216

Furnaces, Early Patterns 214

Furnaces, for Double Houses. ... 29
Furnaces, for Different Fuels. . 17, 20
Furnaces, for Halls, etc 121, 122

Index.

257

PAGE

Furnaces, for Schools. . . 103, 104, 105

Furnaces, for Wood 18

Furnaces, General 7

Furnaces, Heating Capacity 26

Furnaces, Heating Surface of. . 22, 23
Furnaces, Location of . . . 31, 107, 124

Furnaces, Management of 59

Furnaces, Types of 9, 17, 20

Furnaces, Under-feed 20

Galvanized Iron for Cold Air

Boxes 40

Galvanized Iron, Weights of, etc. . 148

Gas 145

Gas Burners for Coal Furnaces. . . 19

Gas Furnaces 19

Glass Surface Equivalent 224

Goddard, F. D., on Trunk Line

Systems 234

Gossamer Check Dampers 42

Grates 1 1

Grate Surface 24

Grate Surface and Pipe Area 51

Grate Surface, Coal Burned per

Square Foot of 24, 26, 28, 63

Gravity, Acceleration of 87

Grills 113

Guarantees 142

Halls, Heating of 117

Heat, Convection of 25

Heat Losses 27, 46

Heat Losses from Pipes 48

Heat Losses through Walls. . . 66, 119
Heat Transmitting Power of Sub-
stances 147

Heat Transmission from Direct

Radiations 81

Heat Units per Cubic Foot of Air. 46
Heaters for Combination Systems . 72
Heating Capacity of Furnaces,

29, 30, 105, 122

Heating Computations 46

Heating Conservatories 77

Heating of Stores 129

Heating Surface for Hot Water

Combination Systems 71

Heating Surface, Furnaces and

Boilers 25

Heating Surface, Gas Furnaces. . . 19
Heating Surface of Furnaces,

22, 23, 245

Heating Value of Coal 28

Hoods 168

Hoods for Vent Flues in

Horizontal Pipes, Length of 47

PAGE

Hot Air Pipes, Size of 44, 45

Hot Water Combination Heater

Sections 72, 73

Hot Water Combination Systems. 129
Hot Water Heater Sections, Ca-
pacities 73

Hot Water Heating, Cubic Con-
tents Rule 74

Hot Water Heating, Down Feed . 80
Hot Water Heating, Pipe Sizes. . . 77
Hot Water Open Tank vs. Pressure

Systems 77

Hot Water Piping Systems 79

Hot Water vs. Hot Air 69

House Heating 31

Humidities, Desirable 96

Humidity 82, 83

Humidity, Evaporation Necessary

to Secure 84

Importance of Ventilation 90

Indirect Radiation 76

Indirect Steam Systems 252

Iron, Color of, at Different Tem-
peratures 147

Iron Plates, Weight of 149

Iron Sheets, Weight of 149

Janitoral Shortcomings 128

Joints in Furnaces 8

Kelsey Warm- Air Generator 242

Leakage of Air 237

Location of Furnaces 31

Location of Radiators in Furnaces. 16

Low Cellars, Piping for 49

MacFie, Air and Health 94, 99

Management of Furnaces 59

Manufacturers' Ratings of Fur-
naces 30

Mixing Damper, Control 140

Mixing Dampers 109

Motors 137

Necessity of Ventilation 98

Nonconducting Covering 48

Open Trunk vs. Pressure Systems . 77

Oval Pipes 54

Oval Pipes, Fittings for 201

Overhead Systems, Hot Water. . . 79

Patterns for Elbows 173, 190

Patterns for Stack Offsets 195

Index.

PAGE

Payments 142

Petroleum 146

Pipe Area, Relation to Grate Area. 51
Pipe Area, Store Heating. ....... 131

Pipe Capacities for Hot Air 228

Pipe Coils 78

Pipe, Galvanized, Table of Dimen-

mensions 148

Pipe Sizes, Hot Air 77

Pipe Soldering 172

Pipes, Flow of Air through 86

Pipes, Furnace 48

Pipes, Hot Air, Construction of . . 171

Pipes, Making of 171

Pipes, Material of 55

Pipes, Oval 54

Pipes, Sizes of 81, 224

Pipes, Welded Steel or Iron, Di-
mension of 8 1

Piping for Low Cellars 49

Piping, Furnace 48

Piping Furnaces, Directions for. . 216

Piping Systems, Hot Water 79

Portable Settings 35

Power, Cost of 137

Pressure Systems, Hot Water. ... 77
Proportions of Furnaces and Sys-
tems 225

Prudden, Dr., on Dust 92, 94

Public Building Heating 116

Radiating Power of Substances,

H7, 253

Radiation and Convection 25

Radiation from Iron 253

Radiation, Ratio to Space 74

Radiator Connections 80

Radiator, Furnace 15, 16

Radiators, Heat Given Off by. ... 81
Radiators in Furnaces, Location

of 16

Radiator Tappings 77

Radiator Valves, Combination

System 70

Ratio, Heating Surface to Grate

Surface 23

Recirculated Air 43

Recirculation of Air 42

Registers 55, 56, 57, 58, 170, 200

Registers and Screens for Schools . 113
Registers, Dimensions of and Area 150

Registers, for Stores 130

Registers, Location of . . . 55, 124, 219
Registers, Side- wall 170, 200, 222

PAGE

Resistance to Air Flow 86

Return Duct 42

Risers ' 201

Risers, Furnace Pipe 51

Risers, Location of 53

Sheet Iron, Weight of 149

School Corridor Heaters 106

School Furnaces 103, 104

School Furnaces, Location of .... 107

School Furnaces, Sizes of 105

School Heating 102

Schools, Fresh Air Supply 108

Secondary Heating Surface 24

Setting of Furnaces, Directions.. . 216
Sheet Metal, Erection of. ... 154, 156

Shoes 183, 192

Side- Wall Registers. ... 170, 200, 222

Sizes of Air Pipes 44

Smith, Dr. Henry Mitchell, Tem-
perature and HumMity. ..... 94

Smoke Pipes 37, 129

Soft Coal Furnaces 20

Soldering Pipes 172

Specific Heat 47

Specifications 142

Stack Elbows 196

Stack Heaters.. ... 112, 114, 121, 126

Stack Offsets 194

Stack Tees 197

Stakes 193

Standard Sizes of Pipe 81

Standards of Ventilation 100

Steam Combination Heaters 79

Steel Plate Furnaces 1 1

Steel Radiators for Furnaces 16

Stock Fittings 164, 165, 166

Stores, Heating of 116, 129

Substances, Heat Transmitting

Power of 147

Table of Air Velocities in Flues . . 89

Tappings for Radiators 77

Tees 197

Temperature, Absolute 85

Temperature Control 139

Temperature, Determination by

Colors 147

Temperature, Effect of 95

Temperature in Various Cities,

Table of 151

Temperature of Air Supply,

27, 64, 65, 246

Temperature Regulators 139

Temperatures, Color of Iron Pro-
duced by 147

Index.

259

PAGE

Test of Fan-Furnace Systems 241

Testing Furnaces in Warm

Weather 240

Testing Plant, Furnace 242

Tests, Furnace 62

Thompson, R. S., Rules by 230

Town Hall, Heating 117

Transformation, Elbows 209

Transformers 205, 209

Transmission Losses through Walls 119
Trunk Line System of Furnace

Heating 49, 234

Twin Furnaces 35, 36

Two-Section Fire Pot 10

Underfeed Furnace 21

Velocity of Air in Pipes and Flues,

47, 64, 87

Velocity of Wind 89

Vent Flue Coils 115

Vent Flue Dampers '. 113

Ventilation . . 82

PAGE

Ventilation, Compulsory 101

Ventilation, Importance of 90

Ventilation, Necessity for 98

Ventilation of Public Buildings. . 125

Ventilation of Schools 83

Ventilation Standards 100, 101

Wall Factors for Heat Losses .... 116

Water, Expansion of 78

Water Vapor, Weight of 84

Weather Bureau Records 151

Weather Stripping 237

Weight of Air 86

Weight of Black Iron Sheets 149

Wind, Effect of, on Furnaces. ... 7

Wind Leakage, Allowances for. . . 120

Wind Velocities 89

Windows, Condensation on 91

Winslow, Prof. C.-E. A., Ventila-
tion Results of 97

Wolff Data 27

Wood 145

Wco 1 Furnaces 18

Best Books on Heating

Snow's Principles of Heating

One of the best reference and text books ever published. It contains well-
written chapters on designing and installing all types of heating apparatus
and equipment, with 59 special tables calculated by the author during his
extensive practice, and many valuable extracts from the reports of investi-
gations and experiences of other eminent engineers.

CONTENTS. — Heating Power of Fuels, Boilers and Combination Heaters;
Gas, Oil and Electricity vs. Coal, and the Capacity and Fuel Consumption
of House-heating Boilers; Heat Given Off by Direct Radiators and Coils;
The Loss of Heat by Transmission; Computing Radiation; Heating Equiv-
alents; Specific Heat; Humidity; The Heating and Co9ling of Air, etc.;
Heating Water; Capacities of Pipes for Hot- Water Heating; The Flow of
Steam in Pipes and the Capacities of Pipes for Steam-Heating Systems and
for Steam Boilers; Modified Systems of Steam Heating; Hot- Water Heating
by Forced Circulation; Central Steam-Heating Plants and Mill Heating;
The Steam* Loop; Non-Conducting Coverings; Miscellaneous Tables and
Furnace Tests.

224 Pages. 6 x 9 in. 62 Figs. 59 Tables. Cloth, $2.00.

Snow's Pipe Fitting Charts

A well-classified series of charts in bound form, showing modern methods
of planning, installing and inspecting the various types of heating systems,
with concise information regarding each plate. Valuable data on ventilating
systems and the weight of ducts are also given.

CONTENTS. — Piping for Hot- Water Heating; Piping for Steam Heating;
Boiler, Engine and Pump-Room Connections, Castings, etc.; Drawings of
Piping and Apparatus; Galvanized Iron Work; Miscellaneous Articles
Relating to Piping.

284 Pages. 6 x 9 in. 220 Figs. Cloth, $2.00.

Lincoln's Steam and Hot-Water Heating

A text-book for the student and mechanic covering the principles of
design and installation of all systems of heating, with a series of test questions.

CONTENTS. — Physics of Heating; Systems of Heating; Boilers; Radiation;
Pipe and Fittings; Low-Pressure Steam Heating; Exhaust-Steam Heating;
Hot-Water Heating; Special Systems of Heating; Hot-Blast Heating.

168 Pages. 6x9 in. 103 Figs. Cloth, $1.50.

Fuller's Designing Heating and Ventilating Systems

This book treats on the practical application of the best engineering rules
and formulas in every-day use. It tells how to lay out a Steam, Hot-Water,
Furnace and Ventilating Equipment for any kind of a building. The entire
contents are presented in a simple and easily understood manner.

The subject matter has been adapted from lecture courses given by the
author before various institutions interested in heating and ventilating,
including Y. M. C. A. and other classes. It gets right down to practical
heating in the very first chapter, illustrates and describes the detailed use of
exactly the same methods that the most proficient engineers use. It tells
how to determine the sizes and proportions of equipment for every-day
work.

CONTENTS. — The Heat Unit; Heating Value of Coal; Relation of Heat
and Work; Loss of Heat from Buildings; Properties of Saturated Steam;
Transmission of Heat through Radiators and Coils; Flow of Steam in
Pipes; Pressure Drop in Steam Mains; Gravity Systems; Vacuum Systems
and Vacuum Valves; Piping for Vacuum Systems; Vacuum Pumps; Indirect
Systems; Boilers; Hot-Water Systems; Furnace Heating; Hot-Blast Heating;
Ducts for Factory Ventilation; School Ventilation; Theatre Ventilation;
Hotel Ventilation; Heating Stacks; Vento Heaters; Arrangement of
Heaters; Individual Duct Systems; Fans; Estimating Static Pressure;
Fan Motors; Temperature Regulation; Recirculation of Air; Air Washers;
Estimating Coal Consumption for Heating.

220 Pages. 6 x 9 in. 78 Figs. 37 Tables. Cloth, $2.0O.

King's Progressive Furnace Heating

A complete modern guide to warm-air furnace heating installation, with
working plans and methods of estimating. Chapter 18, pages 178 to 236,
presents a complete treatise on the construction of sheet -metal furnace fit-
tings, with detailed pattern-layouts and methods for the operator, written
by the well-known sheet-metal expert, William Neubecker.

CONTENTS. — The Chimney Flue; The Furnace — Character, Size, Location
and Setting; Pipes, Fittings and Registers; Installation of the Furnace;
Trunk Line and Fan-Blast Hqt-Air Heating; Estimating Furnace Work;
Intelligent Application of Heating Rules; Practical Methods of Construc-
tion; What Constitutes Good Furnace Work; Ventilation; Ventilation
by the Use of Propeller Fan; Humidity and the Value of Air Moistening;
Recirculation of Air in Furnace Heating; Auxiliary Heating from Furnaces;
Temperature Regulation and Fuel Saving Devices; Fuel: Its Chemical
Components and Combustion; Cement Construction for Furnacemen;
Construction and Patterns of Furnace Fittings; Rules, Tables and Informa-
tion; Recipes and Miscellaneous Data.

280 Pages. 6 x 9 In. 189 Figs. Cloth, $2.50.

Lyman's Steam and Hot-Water Heating Estimate and
Contract Record Book

A new and thoroughly practical form for estimating small or large jobs,
listing all the items of cost so that all the estimator has to do is fill in the
quantities and prices opposite each item.

It is of exceptional value as a check against loss through the omission of
some of the elements of cost, for it lists everything apt to be required and
thus serves as a reminder. It was carefully edited and finally approved by
eight heating contractors before it was printed.

The money columns are double, so that the estimate can be carried in the
first and the actual cost in the second, if desired; or estimates may be itemized
by groups in the first column and totals carried into the second.

There are 100 forms, each of which is numbered, and an index is provided
with columns for recording the name, number, date and file.

208 Pages. 9 x 13 in. 1OO Estimate Forms. Cloth, $2.00.
Carpenter's Heating and Ventilating Buildings

A clear and concise discussion of the principles of heating and their appli-
cation in the design and practical construction of the various systems and
apparatus employed in heating and ventilating buildings, with a valuable
collection of useful tables and other data.

583 Pages. 6 x 9 in. 290 Figs. Cloth, $3.50.

Button's Hot-Water Supply and Kitchen Boiler Connections

This new book provides specialized and reliable information on the
installation and repair of hot-water service, with examples covering all
conditions likely to arise in practice. Special attention is given to installing
special apparatus and making proper connections for them. The illustra-
tions are clear and well detailed, so they show all the construction at a glance.

CONTENTS. — Principles of Heating, Combustion, Transmission of Heat, etc.
Corrosion of Water Fronts, Boilers and Pipes, Deposit of Sediment, etc.
Water Fronts, Coils and Heaters. Examples of Range Conditions for Vari-
ous Conditions. Variations in Connections to Suit Special Conditions.
Multiple Connections with Tank and Pressure Supply. Supply Connections
and Distribution. Hot Water Circulation in Large Buildings. Double Boilers.
Connections and Distributing Pipes. Heating Water by Gas. Heating Water
by Steam Coils and by Injecting Steam, and by Coils in Heating Apparatus.
Utilizing Excess Heat in Warming Rooms and Domestic Appliances. Air
Locking, Expansion of Water, Relief Pipes and Valves. Common Complaints
and Their Remedy; Repair Kinks. Typical Examination Questions on the
Theory and Practice on Hot-Water Supply Installation.

211 Pages. 6x9 inches. 151 Figs. Cloth, $1.50.

PehPs Everready Pipe and Elbow Chart

A time- and labor-saving device for quickly and accurately laying out all
kinds of pipe and ventilation work, with 54-page book of tables for figuring:
the weight of elbows and ducts; wrought-iron and steel sheets; galvanized
sheets; steel angles; round, square and flat bar steel; revolutions of fans;
air yelocity; flue and register dimensions; pressure and horse-power; sizes
of suction traps; friction in elbows; capacity of round and rectangular ducts,
etc.

54 Pages. 4H x 6% in. Cloth, $1.00.

Tables for Ventilating Ducts

A set of six tables for estimating the superficial area and weights of gal-
vanized sheet-iron rectangular ducts such as are'usually employed in ventilat-
ing-work. The tables embrace all sizes from 1 inch to 60 inches square.

Heavy Bristol Board. 7 x 9 in. 50 cents.

Baldwin On Heating

Describes the various types of apparatus for heating and ventilating large
buildings and private houses by steam, water and air, with useful tables and
data.

404 Pages. 5x8 in. 141 Figs. Cloth, $2.50.

Thompson's House Heating by Steam and Water

Improved methods of installing heating apparatus in the home. Short
method of -computing radiation and heat losses, with graphic charts showing
boiler power and coal consumption.

265 Pages. 7 x 10 in. 268 Figs. Cloth, $3.00.

House Chimneys

Useful to every one. It describes the best way to overcome every imag-
inable chimney trouble so the layman can locate and fix it.

77 Pages. 5 x 8 in. 74 Figs. Cloth, 50 cents.

Starbuck's Questions and Answers on Steam and Hot-
Water Heating

It takes up systematically and concisely the subjects of boilers, flues,
fuel and combustion, radiation, radiator connections, character and use of
valves, pipe and fittings, the various systems of steam heating and hot-water
heating, ventilation, vacuum and vapor heating, accelerated hot-water
heating.

135 Pages. 4^x7 in. 57 Figs. Cloth, $1.00.

PLUMBING AND SHEET METAL WORK

We specialize on books relating to Seating, Plumbing, Sheet-Metal
Work, Carpentry and Building, Estimating, House Planning, Hard-
ware. Special Catalogs describing all the books on each of these subjects
will be sent free promptly on receipt of request.

DAVID WILLIAMS COMPANY

Publishers of METAL WORKER

231-241 West 39th Street New York

RETURN CIRCULATION DEPARTMENT

TO "ifr 202 Main Library 642-3403

LOAN PERIOD 1

LIBRARY USE

This book is due before closing time on the last date stamped below

DUE AS STAMPED BELOW

REC.CIR.OCT 1

278

FORM NO DD 6A 12m 6'76 UNIVERSITY OF CALIFORNIA, BERKELEY
BERKELEY, CA 94720

®i

^3BBII^HBHI^^^H^HHBH^^B? . ifHIBHI^^^^HH

341781

UNIVERSITY OF CALIFORNIA LIBRARY

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Furnace heating; a practical and comprehensive treatise on warming buildings with hot air"

See other formats