Ithaca, N. Y., which it is worth while to illustrate at this point in order
Fig. 119.— View of a Dam at Ithaca, N. Y.
Digitized by
Google
STATE GEOLOGIST. 181
that confidence may be strengthened in the value of concrete for struc-
tures placed under trying conditions.
The Ithaca Water Works Company wished to increase and de-
velop their source of supply for the city of Ithaca but felt that they
could not put in sufficient capital to build a large dam with a gravity
section. Several plans and estimates were made but all were too
high in price.
l^r/jca/c5ect/on on CenUrL/ne.
Fig. 119a. — Cross Section of the Ithaca Dam.
Mr. Gardner S. Williams, Associate Professor of Civil Engineer-
ing at Cornell University, was engaged as consulting engineer to
prepare plans and he designed the dam illustrated in figures 119
and 119a, accomplishing a reasonably large storage for a minimum
construction. The dam was originally designed to be 90 feet high
buti was reduced to its present height in building. It is built in a
narrow slate or shale gorge about 90 feet wide at the point where the
Digitized by
Google
182 ANNUAL RErORT
dam is located. The dam is built of concrete reinforced at the face
and back with 3 by 3-16 inch longitudinal steel bands spaced 4 feet
apart and interlaced with wire mesh.
Five-eighth inch tie rods connect the two systems of reinforce-
ment. Both back and face of dam are faced with vitrified brick or
block. The dam is about 7^ or 8 feet thick and is built in a partial
dome shape, convex side upstream. The radius of the horizontal cur-
vature is 100 feet.
During construction it successfully withstood a severe overtop-
ing flood. Messrs. Ross F. Tucker and Thomas M. Vinton built the dam.
Fig. 120.— Sidewalk Composed of Reinforced Concrete Slabs,
According to the St. Louis Expanded Metal Company's System.
Digitized by
Google
STATE GEOLOGIST. 183
METALLOID SIDEVALKS.
The St. Louis Expanded Metal Company have a new design for
sidewalks which would seem to have some commendable features.
By the use of sheets of expanded metal, concrete sidewalk blocks
five or six feet square and about 2 inches thick are made at some
central factory, and the blocks are then shipped to the site ready
to set in place like stone slabs.
The foundation consists of two trenches filled with cinders upon
which the concrete stringers at the edges of the slabs rest. The
center portion of the block does not rest upon the ground, hence is
not affected by the heaving due to frost.
It is claimed that this form is no more expensive than the usual
concrete walk and it can be laid at any season of the year, without
interrupting traffic. Figure 120 illustrates the method.
FIRE PROOFING, FIRE TESTS AND FIRES.
The fire proof qualities of concrete and concrete-steel are becom-
ing better known each year. Theoretically, a good fire proofing
material must have two qualities, it must be able to resist sudden
changes of temperature within its own structure without disintegra-
tion and it must be a non-conductor of heat. The porous nature of
concrete, and especially of cinder concrete, prevents heat frgm rapidly
penetrating the mass. A comparatively small section of cinder
concrete will prevent the temperature of enclosed steel from becoming
high enough to destroy its strength or to badly expand or warp it. Tests
seem to prove that concrete has both of the desirable qualities.
Mr. E. Lee Heidenreich, the agent for the Monier system in the
United States, says he has heated Monier plates 2 inches thick, one
foot wide and three feet long to a temperature of 1,200 degrees, Fah-
renheit, and cooled them off by plunging them into cold water with-
out showing any deteriorating effect. This would indicate not only
that the concrete could successfully resist great sudden changes in
temperature, but also that the coefficient of contraction and expansion
of concrete and steel are nearly similar. For if this were not true,
the variation of expansion and contraction of the metal in the plate
would have ruptured the concrete. The test mentioned in the
description of the Roebling system of reinforced concrete also bears
witness to the power of concrete to resist the destructive effects of
fire.
New York: City Fire Tests. — In 1901,* the Department of Build-
ings of New York City conducted a series of fire tests quite severe in
character, which nearly every firm in the tests, using Portland cement
in any form, successfully passed. The department specified the
^Engineering News December 26, 1901.
Digitized by
Google
Fig. 121. — Ruins of the McMahon Cracker & Biscuit Company's Building, Burned
October 8th, 1901, at Chicago, III. Only the Armored Concrete Worl< Remains Intact.
Digitized by
Google
STATE GEOLOGIST.
185
method of procedure, the size and shape of the house to be con-
structed and the arrangement to be made for firing.
The partitions and test walls were erected by the firm interested
in the particular fire proofing submitted for test.
The houses were all 9^ by I4>4 by 12 feet high. The test walls were
from 2 to 4 inches thick. The houses were built with a flue in each
corner to provide for thorough circulation and quick heating. Grates
were built 3 feet above the foundations and heavy protected doors were
provided so that access was given for firing and observations. Ker-
osene, pine and hard wood were used for fuel. The temperatures
Were to be kept from 1,700 degrees to 2,100 degrees, Fahrenheit, for
one hour and then water from the city mains was to be thrown on to
the heated walls for 25^ minutes through fire nozzles with the regular city
pressure.
The test that is of particular interest here was the one made
upon the Sprickerhoff partition. This partition was built of concrete
blocks 3 inches thick plastered one-half inch thick on both sides with
King's Windsor "browning mortar." The blocks were composed of i
part Portland cement, i part sand, and 5 parts steam ashes — ^the blocks
being laid in a cement mortar of i cement to 2 sand. The tempera-
ture reached 1,868 degrees, Fahrenheit. Water was applied which
stripped the plaster, but left the concrete portion unharmed and as
straight and plumb as before the test.
There is nothing so convincing, however, as the actual test upon
buildings in service.
McMahon Cracker and Biscuit Company's Building* — The build-
ing of the McMahon Cracker Company, located in Chicago, burned on
October 8, 1901. The entire structure was totally wrecked, except the
portion holding five large bake ovens, each weighing 200 tons. These were
situated from the third to the fifth floors and were supported by steel
columns protected by a concrete shell composed of i part Louisville
natural cement and 4 parts soft coal cinders, enclosed in a wire form
plastered with cement mortar on the outside. The steel work in the
other portion of the building was twisted and ruined and the walls,
left unsupported, fell. But the heaviest portion of the building sup-
ported by concrete protected steel remained standing. Figure 121
shows the results of the fire.
The Borax Company's Building* — ^The Pacific Coast Borax Com-
pany's building situated at Bayonne, N. J., was destroyed by fire,
April II, 1902. A large portion of the main building — the footings,
walls, posts, girders, floors and a few partitions — were of steel con-
crete. The floors were concrete slabs 4 to 5 inches thick resting on
and being monolithic with concrete beams, which were 4j^ inches
Digitized by
Google
Fig.. 122. — ^The Effect of Fire on Steel Construction. The
Pacific Coast Borax Company's Worlcs, Bayonne, N. J.
Fig. 123. — Effect of Fire on Reinforced Concrete Construction, Pacific Coast
Borax Company's Worlcs. Ttie Fire was Exceptionaiiy l-iot at Tiiis Point.
Digitized by
Google
STATE GEOLOGIST. 187
wide, 28 inches deep and 24 feet long, spaced 3 feet center to center.
The columns were of solid Ransome steel concrete, being 21, 19 and
17 inches square for the first, second and third stories, respectively.
The walls were 16 inches thick, having 9 inch hollow spaces in the
center. The concrete was made of Atlas Portland cement and
crusher run of basaltic rock passing a i inch ring screen. No sand
was. used, the fine crushed basalt taking its place. The proportions
varied in different portions of the work from I part cement and 5
crushed stone to i cement and 6^ stone.
One wing where the crystallizing tanks were placed was of steel
frames with wooden walls, partitions and fixtures. Much inflammable
material was stored all through the building at the time. The roof
was of wooden beams and light supports, with board and tar cov-
ering. There were many wooden partitions in the main building,
much wooden framing for the shafts, machinery supports, stairways
and bins.
The heat of the fire was sufficient to fuse copper in several places
in the building. All steel posts and girders were warped out of all
semblance to their proper shape. Several large tanks set upon the
roof, one a concrete tank 6 by 6 feet and 50 feet long, weighing 33 tons,
fell through to the fourth floor without doing any injury. The clear
fall from roof to floor was 14 feet. One steel tank weighing 18 tons
fell corner-wise, apparently, and that cracked two or three concrete
floor girders and broke a small hole through the floor. With the
exception of the cracked floor beams, no damage was done to the
concrete portion of the building except to crack off the plaster here
and there and to smoke up the walls and ceilings so that a coat of
paint or plaster was required to put them in presentable shape.
The floors held three and four inches of water after the fire
without showing any leakage.
Figure 122 shows the results of the fire in the one story portion
of the building where there was not a great amount of combustible
material. Figure 123 shows the condition of the concrete at the
point where the fire was the most intense and where the hole was
punched in the floor doing the greatest damage.
The test was so conclusively in favor of concrete for a fire proof
material that the company in rebuilding made all of their partitions,
the roof, bins and the machinery and shafting supports out of concrete.
The writer visited the works on August 28, 1902, just before
the rebuilding was completed; the plant had been in operation, how-
ever, for some months, as it took but little work to get the. main
building in condition to operate the plant. The work in progress was
the laying of the roof.
In making the roof, planed boards were used for the forms.
They were carefully put up and thoroughly supported. A thin coat
Digitized by
Google
188 ANNUAL REPORT
of white plaster one-half inch or less in thickness, apparently carrying
some plaster of Paris, was spread upon the forms and then the iron
bars and concrete were put in place. The white mortar acted as a
plaster or surface coat for the underside of the roof.
Within the building heavy machinery, crushers, filter presses,
tanks, etc., were set upon the floor or hung from the ceiling with no
extra strengthening attempted. Borax, weighing 340 pounds per barrel,
was stacked three barrels high over large areas. In fact, much of the
floor, so the general manager said, had been tested in actual service by
loads of 1,000 to 1,650 pounds per square foot.
Baltimore Fire* — The Engineering News* employed several experts
to investigate and report upon the fire proof construction within the area
swept by the great Baltimore fire of February 7, 1904. Extracts from
these various papers read : "Comparing the efficiency of concrete and the
hollow blocks as fire-proofing materials, there is no doubt but that the
concrete made from steam boiler cinders and Portland cement, made
the best showing."
"Generally speaking concrete and terra cotta protected successfully
the steel columns."
"As fire and water resisting materials, terra cotta and concrete
have given a very good account of themselves."
"The buildings were all gutted, but the concrete floors were all
apparently in first-class condition. The iron work was not exposed
and the concrete did not appear to be disintegrated."
"The concrete constructiont (speaking of the Fidelity and Guar-
anty Company's building) endured the fire practically uninjured, a
notable demonstration being the fact that the cantilever extensions
of the floors in front and rear remained intact, and the attic floor
carried a tier of columns reaching to the roof which it had never been
designed to support."
"In a few cases where the heat was the greatest, fine surface cracks
were seen in the beams and ceilings and small portions of the concrete had
flaked off, but nowhere sufficient to indicate serious injury to the structure."
One of the floors in this building was afterward tested by loading it
with a uniformly distributed load of 225 lbs. per square foot and it only
showed 1-16 inch deflection under the load. The concrete used in the
United States Fidelity and Guaranty Company's building was composed
of I cement, 3 sand and 5 crushed granite. Two other buildings were of
so called fire proof construction.
*EnQineering News March 8, 1904.
•jr Engineering jBccord March 18, 1904
Digitized by
Google
STATE GEOLOGIST. 189
CHAPTER V^
SPECIFICATIONS FOR CONCRETE MATERIALS.
In the following pages the effort has been made to show what safe-
guards may be thrown around the use of cement for the numerous class of
structures which have been described in the foregoing chapters. . It will
be, of course, undertsood, that the framing of specifications for a piece of
work of great importance where either large sums of money or human life
is involved in any failure of the structure as in dams, bridges, floors, etc.,
is a work which properly belongs to the trained engineer and to him alone.
Each piece of work of this sort must be made a special study and take
into consideration all the local and unusual features of the case as well as
considerations based on the normal qualities of the materials.
But there are a very large class of uses of less importance, where
no such results depend, and there is no reason why the architect, con-
tractor, builder, owner and in fact any person of intelligence, but devoid
of engineering training, may not with safety and satisfaction to himself,
prescribe the conditions under which the cement structure shall be built.
Accordingly a number of specification forms which have been pre-
pared by skilled engineers and cement users, are here reproduced in order
to serve as models upon which others may draw their contracts.
CEMENT.
The cement may be of any brand of American or foreign Portland
cement which will meet the requirements of these specifications.
Condition of Delivery. — It must be delivered in original packages
labeled with the brand and the name of the manufacturer. These pack-
ages may be either barrels or bags, but must be well protected in either
case from air and moisture. Any broken packages may be rejected or
used at the option of the engineer in charge of the work.
Time of Delivery* — The contractor shall furnish the cement upon
the work at least ten days before it is to be used, in order that time may be
given to make the necessary tests.
Housing* — It shall be stored in dry, well ventilated buildings for
work of any magnitude ; and for work of less importance it shall be safely
stored and protected from moisture in any form.
Digitized by
Google
190 ANNUAL REPORT
TESTS OF PORTLAND CEMENT^
The cement shall stand the following tests; any that fails to meet
these tests will be rejected and the contractor shall immediately remove
the same from the work.
Soundness. — Two pats of neat cement, with thin edges, will be made
up<3n glass plates, and allowed to attain permanent set in moist air.
Twenty-four hours after making, one pat will be placed in water having
a temperature between 50 and 70 degrees F., and must withstand indefinite
exposure without checking, softening or distortion. The other pat will
be placed in some form of a steamer over cold water, which shall be
brought to the boiling point and maintained at this temperature for three
hours and then allowed to cool slowly. The pat shall not show any signs
of distortion, cracking or softening under this test.
Fineness* — The cement shall be so finely ground that after being
thoroughly dried by heating it 94 per cent, shall pass through a No. 100
standard sieve, woven from No. 40 Stub's wire guage.
Activity* — Initial set shall not occur in less than 40 minutes, and
final set in less than one hour and 30 minutes nor more than six hours.
The time of setting shall be determined by Gilmore's wires, or Vicat's
needle.
Tensile Strength* — The standard section of briquette shall be used.
The neat cement shall be mixed into a rather dry, stiff paste by the addition
of from 17 to 20 per cent, of its weight of water. In a mortar of sand
and cement, water to the amount of 10 to 13 per cent, of their combined
weights shall be used, the amount depending upon the character of the
cement and sand. The mortar shall be firmly pressed into the molds
with the thumbs, filling the molds in three layers of about equal thickness
and smoothing off both sides with a trowel. Briquettes shall remain in
moist air for twenty-four hours and the remainder of the time, until
tested, they shall remain in water at a temperature of about 60 to 65
degrees F.
Seven day tests of neat cement shall show not less than 450 pounds
per square inch, and not less than 550 pounds for 28 day tests. Briquettes
made of i part cement to 3 parts standard sand shall stand a test of 150
pounds per square inch at seven days.
SPEaFICATIONS FOR SAND^
All sand used for mortar shall pass a No. 10 sieve and 80 per cent, of
ic shall be retained upon a No. 74 sieve. It shall be a silicious sand, as
sharp as can be obtained within reasonable limits of cost. It shall be free
from all vegetable and organic matter, and shall not contain more than
10 per cent, by weight, of clayey or loamy material.
Digitized by
Google
STATE GEOLOGIST. 191
SPEaFICATIONS FOR STONE^
The aggregate shall consist of crushed trap rock, granite, hard lime-
stone, or other material equally hard and durable which shall meet the
approval of the engineer. The broken stone shall be free from vegetable
or organic matter in any shape and free from mud and dust or from lumps
of clay or clay covered fragments. When sand is to be used in the con-
crete, the stone shall be screened to pass through a ( — ) inch ring and
retained by a screen of J4 inch rings. The stone should be thoroughly
wet before mixing with the mortar. When it is desired to use screenings
with the crushed stone, the proper proportion of sand to be used shall be
determined by analysis.
SPEaFICATIONS FOR CONCRETE^
Proportions* — The concrete shall consist of one part cement, 2 to
4 parts of sand, and 4 to 10 parts of crushed stone measured by loose vol-
ume. (The proportions must vary to suit the character of the work and
the requirements which the concrete must meet.)
Mixings — The sand and cement shall be thoroughly mixed dry, then
sufficient water added to make a plastic or wet mortar and the whole thor-
oughly mixed again. The stone, having been previously wet down, shall
then be added and the whole mass thoroughly mixed until every particle
of stone is coated with mortar. The concrete thus mixed should be imme-
diately placed in position and rammed until excess mortar shows over the
entire surface. (Where surfaces are to be exposed, either a facing of I
cement to 2 sand shall be used and placed in position as the mass of con-
crete advances, or else the stone should be forced back by thrusting down
flat bladed shovels along the face of the form, thus allowing the richer
mortar to run in next the form which makes a smooth impervious surface.)
No concrete which has attained partial set shall be used. The engineer
will be guided by the tests in establishing a limit of time beyond which
any concrete once mixed shall not be used in the work.
Digitized by
Google
192 ANNUAL REPORT
SPECIFICATIONS FOR PORTLAND AND NATURAL CEMENT.
Adopted May, 1903, by the American Railway Engineering and
Maintenance of Way Association.
PORTLAND CEMENT.
J* Definition* — Portland cement is a product of the mixture of
clay and lime-carbonate in definite proportions, calcinated at a high
temperature and reduced to a fine powder.
2. Packages. — Cement shall be packed in well-made wooden bar-
rels lined with paper, or in strong cotton or paper sacks. Each
package shall be plainly marked with the brand and name of the man-
ufacturer, and the net weights shall be exact and uniform.
3. Weight.— One barrel shall contain not less than 376 pounds
of cement, and four sacks shall be equivalent in weight to one barrel.
4. Condition. — All cement shall be delivered in sound packages,
undamaged by moisture or other causes.
5* Storage* — Cement must be stored until used in a perfectly dry
place in such manner as will insure it from all damage.
6. Rejection. — ^All cement failing to meet the requirements of
the specifications may be rejected, and all rejected cement, whether
damaged or rejected for other causes, shall be removed at once from
the company's property.
7. Tests. — All cement shall be subject to the foUcrwing tests:
(1). The selection of the sample for testing, the number of pack-
ages sampled, and the quantity taken from each package, must be
left to the discretion of the engineer, but each sample should be a
fair average of the contents of the package from which it is taken.
At least one barrel in every ten should be sampled.
(2) Cement in barrels should be sampled through a hole made in
the center of one of the staves, midway between the heads, or in the
head, by means of an auger or sampling iron similar to that used by
sugar inspectors. If in bags, it should be taken from surface to
center.
(3) All samples should be passed through a sieve having twenty
meshes per linear inch in order to break up lumps and remove foreign
material. For determining Ihe characteristics of a carload of cement
the individual samples may be mixed and the average tested; where
time will permit, however, each sample will be tested separately.
8. Fineness. — ^Not less than 94 per cent, of the cement tested
shall pass through a No. 100 standard sieve. The standard sieve shall
be circular, about 20 cm. (7.87 ins.) in diameter, 6 cm. (2.36 ins.) high,
and provided with a pan 5 cm. (1.97 ins.) deep and a cover. The wire
cloth in the sieve to be woven (not twilled) from brass wire having
a diameter of 0.0045 ins. This cloth to be mounted in the frame
without distortion; the mesh should be regular in spacing and for a
No. 100 sieve shall contain not less than 96 nor more than 100 meshes
per linear inch. The cement to be thus tested shall be thoroughly
dried at a temperature of 100** C. (212 degrees Fahr.) before sieving.
Digitized by
Google
STATE GEOLOQIST. 193
9. ScU— (1) Initial set shall not occur In less than thirty (30)
minutes.
(2) Final set shall not occur in less than one hour nor more than
ten hours.
(8) The time of setting shall be determined by means of the
Vicat needle apparatus as recommended by the Committee of the
American Society of Giyil Ehigineers upon uniform tests of cement in
conjunction with the Committee of the International Association for
Testing Materials.
(4) Using a paste composed of neat cement and water, of nor-
mal consistency, the initial set is said to have commenced when the
needle ceases to pass a point 5 mm. (0.20-in.) above the upper surface
of the glass plate in the Vicat apparatus, and is said to have ter-
minated the moment the needle does not sink visibly into the mass.
(5) The paste is of normal consistency when the cylinder of
the Vicat apparatus penetrates to a point in the mass 10 mm. (0.39-in.)
below the top of the ring.
(6) The amount of water required to make a paste of normal
consistency varies with different cements, but will be found to be
approximately 20 per cent, of the weight of the cement. It should
have a temperature of 70 degrees Fahrenheit.
)0* Socmdness*— (1) A pat of neat cement 2^^ to 3 inches in diam-
eter, %-inch thick at center, tapering to a thin edge, and allowed
to take its final set in moist air, must withstand indefinite exposure