American Society of Civil Engineers.

Transactions of the American Society of Civil Engineers (Volume 81) online

. (page 55 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 55 of 167)
Font size
QR-code for this ebook


No. 5, same proportions as No. 4, but mixed plastic.

No. 6, same proportions as No. 4, but mixed very wet.

No. 7, 1 part of cement, 3 parts of sand, and 6 parts of stone.
Mixed quite dry.

No. 8, same as No. 7, but mixed plastic.

No. 9, same as No. 7, but mixed wet.

No. 10 shall consist of a Portland cement which is free from iron.
One specimen shall be mixed, of the proportions 1:1:2, quite wet.

No. 11 same cement as No. 10, 1:3:6, wet.

No. 12 commercial Portland cement high in alumina, mixed
1:1:2, wet.

No. 13 same cement as No. 12, 1:3:6, wet.

No. 14 of a commercial Portland cement low in alumina, 1:1:2, wet.

No. 15 same cement as No. 14, 1:3:6, wet.

No. 16 of an iron ore cement practically free from alvimina, 1:1:2,
wet.

No. 17 same cement as No. 16, 1:3:6, wet.

No. 18 of Universal cement, 1:1:2, wet.

No. 19 same cement as No. 18, 1:3:6, wet.

No. 20 shall consist of the same materials and proportions as given
for No. 7, but shall be most thoroughly well mixed (much better than
commercial mixing), at the same time being quite wet.



674 TESTS OF CONCRETE IN SEA WATER

ISTo. 21 shall be the same as No. 7, but mixed with sea water, quite
wet.

No. 22 shall be mixed of /^ by weight of one part of standard
Portland cement as No. 7, yV P^^'t by weight of hydrated lime, 3 parts
of sand, 6 parts of stone, mixed wet.

No. 23 shall be of the materials given for No. 7 but in addition
shall contain Sylvester mortar hereinafter described, mixed wet.

No. 24 shall be of the materials in No. 7 and in addition shall
contain 5% by weight of the cement of finely pulverized clay, mixed wet.

In addition, there shall be fifteen cubes, 8 in. on each side, of 1
part of standard Portland cement, 1 part of sand, and 2 parts stone;
and fifteen others of the same cement, mixed 1:3:6. One-third of
these, after being thoroughly set, shall be permanently immersed in
water; another third shall be supported at about half tide; and the
rest shall be kept permanently dry but exposed to the weather.

There shall also be made briquettes, of the same cements and pro-
portions of mortar as above described for concrete, of such number
that they can be tested at intervals covering a period of years, some
of which shall be kept in the laboratory, some exposed to the same
conditions that the cubes will have; and enough briquettes shall be
made of standard sand to compare the relative strength between stand-
ard and the materials actually used.

Testing of Materials. — All cement shall be thoroughly tested for
all physical properties, and shall be subjected to chemical analysis. The
sand shall be subjected to a thorough mechanical and physical analysis
to ascertain the relative sizes of its grains and the quantity of foreign
matter which it contains, if any. The same tests shall be applied
to the stone. The various ingredients, such as sea water, hydrated lime,
Sylvester mortar, and clay, shall be tested, in order that the exact
nature of the material may be kno^\ai.

Sand. — The sand shall be of a good quality, commercial bank sand,
clean, coarse and sharp, as free from all impurities and foreign matter
as can be obtained commercially. The stone shall be of trap rock
which shall have passed a 2-in. ring and have been retained on a i-in.
ring. It shall be as free from dust and dirt as commercially practicable.

Mixtures. — All materials shall be proportioned by volume, as given
above for each specimen. A cubic foot of cement shall contain 100 lb.
The sand and stone shall be measured on the basis that a barrel of
cement contains 380 lb. net and measures 3.8 cu. ft. Samples of the
sand and stone shall be weighed so that the proportion by weight (as
well as by volume) will be known.

The quantity of water to be used shall be accurately measured for
each batch, and shall be proportioned by experiment, so that the dry
specimens shall just fail to show moisture when tamped in place; the



TESTS OF CONCKETE IN SEA WATER 675

plastic specimens shall just barely begin to quake when tamped; and
the wet specimens shall be proportioned so that the mortar will flow
easily off the blade of a shovel and shall be wet enough to flow readily
into place without being spaded or tamped.

The Sylvester mortar shall be mixed as follows, as given in Gillette's
"Hand Book of Cost Data", pages 389 and following: A light-colored
soft soap shall be dissolved in water, 1^ lb. to 15 gal. of water, and
3 lb. of powdered alum shall be mixed with each bag of cement. After
all the materials of the concrete are thoroughly mixed dry, water con-
taining the soap in solution shall be used to mix the materials wet to
the proper consistency.

Mixing. — All materials shall be thoroughly mixed in a batch con-
crete mixer of approved type. They shall be fed into the machine
as quickly as possible, and shall remain for 2 min. after the last
ingredient is added. For the specimen which is to be extra thoroughly
well mixed, the material shall remain in the mixer for 12 min.

If the weather is cold, all frost shall be removed from the sand
and stone by heating before the materials are placed in the mixer.

Placing. — The concrete, after mixing, shall be placed in the forms
as quickly and continuously as possible, and, after proper tamping,
shall not be again disturbed. In no case shall concrete which has been
allowed to stand for more than i hour be used, and in no case shall
i hour elapse between placing two batches of concrete in the same
mould.

Forms. — The forms shall consist of planed and matched spruce lum-
ber, so that they may be as tight as possible, and shall be braced so
thoroughly as not to spring under the pressure of green concrete.
The forms shall be left on for such a length of time that there shall
be no damage to the concrete when they are removed. The Inspector
shall be the judge as to when it is proper to remove the forms.

Reinforcement. — The columns shall be reinforced with an embedded
steel bar, about § in. square, which shall be bent into a U -shape and
run throughout the length of the big specimens near two diagonally
opposite corners, and the loop shall project a sufficient distance above
the top of column to permit of its being easily hooked in case it is
desired to remove it.

A hole, 3 in. in diameter, shall be cored in the upper 8 ft. of each
specimen, to permit of examination from time to time to see if water
has penetrated through the walls of the concrete to the hole.

Protection. — After the specimens are cast, they shall.be protected
from freezing, as far as possible, until they have set for a period of
at least 5 days, and, after the forms have been removed, they shall be
thoroughly and substantially braced, so that there shall be no danger
of their tipping over.



676 DISCUSSION ON TESTS OF CONCRETE IN SEA WATER

DISCUSSION



Mr. T. Kennard Thomson,* M. Am. Soc. C. E. — This is an interesting

paper on an interesting subject, and should bring out a full discussion
from experience in actual work.

The author calls attention to the fact that a mixture of one part of
cement to two parts of sand gave a better test than the mixtures which
were not so rich in cement. This is undoubtedly due to the fact that
a 1 : 2 mixture comes nearer to having the voids of the sand filled with
cement than a poorer mixture, and, of course, a concrete with all the
voids filled has a better chance of resisting the effects of frost and of
any injurious chemical which may be in the water or air ; and probably
explains why the greatest damage to concrete is generally between
high and low water.

Some years ago the speaker built a breakwater, off New Kochelle,
which is subjected to storms with the full sweep of Long Island Sound
behind them. A rock fill was used, up to the low-water mark, the depth
of water at low water being 25 ft. at the outer end.

From low water upward the stones were carefully placed to a height
of 14 ft. (except for the outer 50 ft.), on top of which a 12-in. concrete
coping was placed, as individual stones would have been displaced by
the storms, although the top of the breakwater was 9 ft. above ordinary
high tide.

At the outer end, for 50 ft., the built-up stonework was only 8 ft.
above low water, on top of which was placed a concrete block, 7 ft. high,
/ 10 ft. wide on the base, 5 ft. on the top, and 50 ft. long ; a small quantity
of reinforcement was used.

This breakwater has stood successfully the severe storms of the
Sound for some 11 years ; but this year it has been noticed that a small
quantity of the concrete has seriously deteriorated for a foot or so at
the bottom and for about a foot in depth.

It was a little surprising, however, that this destruction occurred
on the inside or sheltered portion of the breakwater, instead of outside,
where it was exposed to the storms. This may have been due to the
fact that on the exposed side of the breakwater any injurious chemicals
from sewage, etc., would be washed away instead of remaining in place,
as they would on the sheltered side.

It has been found impossible to get a concrete to withstand the action
of impure, hot water, in some power-plants. In one case the efforts
have been abandoned, and a wooden lining has been substituted for the
concrete.

It would seem that the three principal causes for disintegration of
ordinarily good concrete in sea water are: first, too coarse a mixture,

* New York City.



DISCUSSION ON" TESTS OF CONCEETE IN SEA WATER 677

in which the voids of the sand or stone are not filled — thus allowing Mr.
chemicals to accumulate; second, alternate freezing and thawing, which Thomson,
do far more damage when the voids are not filled ; and third, the action
of injurious chemicals assisted by heat.

The remarks in the discussion about sand recall an attempt of a
railroad contractor, many years ago, to make the railroad haul free of
charge for 110 miles a fine beach sand, because the local sand appeared
to contain too much loam. The speaker made three sets of tests : first,
with regular testing quartz; second, with the loamy sand; and third,
with the fine beach sand. The tensile strength developed by the
briquettes of the quartz and loamy sand were almost the same, and
those made with the clean but fine sand gave only half the strength of
the others. As strength in the concrete was what was wanted, the work
was finished with the loamy sand.

J. J. Yates,* Assoc. M. Am. See. C. E. — The question of the action Mr.
of sea water on concrete is of great importance, as it affects the per- ^^'^^^*
manency of such a large number of important structures which have
been, or may be, built in sea water. This paper adds much valuable
information on a subject that has received far too little study.

The speaker, however, thinks that one important phase which has
not been considered is injury to concrete structures by floating ice
or floating debris, resulting in the breaking up of the surfaces and
the exposure of the interior concrete to the action of sea water.

From observations of existing structures, extending over a period
of about 15 years, the speaker is led to the opinion that, in setting,
a surface is formed on concrete which will resist successfully the action
of sea water on the cement; and that, as long as this surface is main-
tained in good condition, no destructive action due to sea water may
be expected. In other words, he is of the opinion that concrete
surfaces are first injured or broken up by drifting material, and
that, after they are thus broken up, the concrete is not able to
resist the action of the sea water. Observations indicate that this
action, after it once starts, continues, and is accelerated by the wearing
away of the disintegrated surfaces by drifting ice and other material,
or by the alternate freezing and thawing of the concrete surfaces
between tides. By one of these means, the disintegrated concrete is
removed, continually exposing a fresh surface to the action of the
sea water.

As examples of the action of sea water on concrete, the following
are given:

The concrete masonry of the old center piers of the Hackensack
and Passaic River Bridges of The Central Railroad Company of J^ew
Jersey were constructed about 1870 of Rosendale cement. The

* New York City.



678 DISCUSSION ON TESTS OF COXCRETE IN SEA WATER

Mr. concrete of these piers was deposited in a circular, wrought-iron,
floating caisson, which was sunk in position to a depth of about 12 ft.
on a previously prepared pile foundation. About 1888 or 1889 the
iron superstructure was renewed, making use of the old piers. When
inspected about 1902 it was found that the iron caisson had com-
pletely disappeared, to a depth of several feet below low water, and
that the concrete had been eaten away between high and low water
to a depth, in some cases, of about 2 ft. These piers were repaired by
patching with concrete behind a steel plate band extending around
each pier.

The piers were not exposed to drifting material, as they were pro-
tected or practically enclosed by pile and timber fenders of a type
generally used at draw-bridges. Considerable trouble was experienced,
however, from ice forming around the piers. In fact, during extreme
high tide conditions, the operation of the lower of the draw-bridges
was seriously interfered with by ice clogging the bearing wheels on
the center pier.

In 1912 both the substructure and the superstructure were removed,
and temporary structures were built at one side for the purpose of
building new bridges.

During 1903 to 1905 The Central Railroad Company of New Jersey
replaced its draw-bridge over Newark Bay, about | mile from its mouth
or junction with Staten Island Sound, with two bascule spans. The
seven concrete piers required were deposited in air within wooden
floating caissons which were sunk to a depth of about 30 ft. on prepared
pile foundations. Sea water was not allowed to come in contact with
the concrete until it had had time to become thoroughly set.

The concrete was composed of 1 part of cement, 3 parts of sand,
and 5 parts of 1^-in. broken trap rock. The following is a represen-
tative test analysis of the cement used:

Sulphuric acid 1.38% Magnesia 1.54%

Gravity 3.06 Water and steam tests. O. K.

Initial set 1 hr. Hard set 4 hr. 30 min.

Residue, No. 200 sieve. 27.5% Residue, No. 100 sieve. .8.0%

One-day test 385 lb.

One- week test 716 lb.

One- week test, sand 315 lb.

The concrete was deposited under the supervision of experienced
men, who stated that all the piers were treated in the same manner.

Shortly after the timber caisson was removed. Pier 5 was injured
between high and low water by a floating barge. This injury was
repaired by patching with concrete. Examinations of this pier in 1909
showed that the natch had disappeared entirely, and that the concrete



DISCUSSION ON TESTS OF CONCRETE IN SEA WATER 679

was being eaten away gradually between high and low water at the Mr.
point of injury. All the remaining piers except two showed some
deterioration of the concrete between high and low water. In the
case of Pier 5 the concrete was eaten away to a depth of 20 in. Two
piers showed practically no deterioration. An examination by a diver
in 1909 showed that there was no deterioration of any of the piers
below a point about 1 ft. below mean low water. From observations,
the piers most affected were those which had the greatest exposure
to floating ice or drift.

Attempts were made to get samples of the disintegrated concrete
for the purpose of making a chemical analysis to determine, if prac-
ticable, the action of the sea water. This was found impracticable,
but when the samples were first removed and still wet, a large number
of crystals of a white salt were visible, and these seemed to disappear
entirely as the concrete dried out.

These piers have been repaired several times — once with a cement
gun, which latter repair lasted about 4 years. The repairs have some-
what retarded the disintegration of the concrete, but the destructive
action has been found to be practically continuous. It is very notice-
able that the action is not general over the whole surface, but is
rather confined to spots, such as the injured section of Pier 5.

Another example may be cited, not particularly for the disintegrat-
ing action of the sea water, but rather on account of the fact that a
sample of the white salt, mentioned in reference to the Newark Bay
Bridge, was obtained and analyzed; this is the bridge of The Central
Railroad Company of New Jersey over the Shrewsbury River about a
mile from its outlet in Sandy Hook Bay, and built in 1913. Ten
piers were constructed in sea water, and of these, two showed signs
of failure about a year after they were built. These piers were rebuilt,
and an investigation into the cause of the failure developed the fact
that, in both of them, the concrete of the foundations, which had been
deposited in water with a tremie, had been deposited at low tide, when
the chute to the tremie was very flat. The concrete mixer had been
on a floating barge, and the low tide had caused the flat condition of
the chute which resulted in the concrete being mixed very wet and
the tremie not being kept full of concrete during the process of
depositing.

The concrete foundations of these piers failed completely, and
when removed showed stratification, consisting of alternate layers of
a putty-like substance and of sand and gravel with a little cement, or
rather, a very weak mixture of concrete. Samples of this mixture were
taken, and the small crystals of white salt were plainly visible when
the concrete was first removed and before it had time to dry out.
Enough of this white powder was collected, and an analysis was made



680 DISCUSSION ON TESTS OF CONCRETE IN SEA WATER

Mr. by Mr. C. M. Chapman, Engineer of Tests, of Westinghouse, Church,
Yates. j^QYT and Company, a copy of which is as follows :

"We have made an analysis of the sample of white powder which
was scraped from the sample of disintegrated concrete from the Shrews-
bury River bridge pier, with the following results :

"Free Silica Si O^ 57.64%

^^"'^^"^^ {It'o'f '-44

Calcium Ca^.' 6.83

Magnesium Mg 8.66

Chlorides CI 1.82

"The large amount of silicia is undoubtedly due to the grains of
sand which were scraped off the concrete with the sample. Eliminating
the silica from consideration, we find that the sample is made up of
approximately equal parts of magnesium, calcium, and iron alumina
oxides. The magnesium is somewhat in excess of the calcium, which
indicates a transfer of the calcium and magnesium bases between the
cement in the concrete and the salt in the sea water. The magnesium
of the sea water replaces the calcium of the cement. This action
can only take place where the sea water has free access to the cement,
or, in other words, at points where the concrete is quite porous. In
a dense concrete such action could not take place, or would be so
slow that it would be almost negligible."

At the south or up-stream end of one of these piers, when examined
about 6 months ago, there was disintegration over a small area to a
depth of about 2 to 3 in. between high and low water. This end of
the pier is very much exposed to damage from floating ice.

Another example of the deterioration of concrete in sea water is
one of the piers of the bridge of The Central Railroad Company of
New Jersey over the Elizabeth River at Elizabethport, N. J. Due
to delays which were beyond control, it was necessary to deposit the
concrete for these piers in freezing weather. Parts of the surfaces of
the piers were injured slightly by freezing, which resulted later in a
peeling of the surfaces. These surfaces were patched with a cement
gun, and have shown no further peeling, with the exception that the
patching on the down-stream end of one of the piers has fallen off.
Where the injured surface of this pier is between high and low water
the concrete has disintegrated to a depth of from 4 to 6 in.

The water of this stream is salt, but is apt to carry a large mixture
of sewage and fresh water. The pier surfaces are covered with slime
between high and low water, but, with the exception noted, show no
deterioration such as might result from the action of salt water.

In regard to the construction of the New Hackensack and Passaic
River Bridges by The Central Railroad Company of New Jersey in
1912-14, it may be of interest to note that, due to experience with



DISCUSSION ON TESTS OF CONCRETE IN SEA WATER 681

the disintegration of concrete in sea water, it was decided to build Mr.
those portions of the piers between high and low water with a granite ^**®^-
facing, backed up by a concrete interior. This granite facing was
carried from a point 2 ft. below mean low water to a point 2 ft. above
high water.

Practically the whole of the 12 000 cu. yd. of concrete in the foun-
dation of these piers was deposited in water with drop-bottom buckets.
The channel piers extended to a depth of 30 ft. below low water, and
the concrete was deposited continuously within timber coffer-dams up
to a point 2 ft. below low water. About a day after the completion
of the concrete, the water was pumped down, the laitance was removed
from the upper surface, and the construction of the remaining portions
of the piers was carried on in the air.

The concrete was composed of 1 part of Portland cement, 2 parts
of washed Cow Bay sand, and 4 parts of washed and graded gravel.

In experimenting with different concrete aggregates, it was found
that when Cow Bay sand and gravel were used, 2 to 3 in. of laitance
would result in depositing a depth of 15 ft. of concrete under water.
When Cow Bay sand and l^-in. broken trap rock (dust screened out)
were used, the result was about double that quantity of laitance. For
this reason practically all the work was done with the gravel.

Examinations of the under-water surfaces by a diver, shortly after
the completion of the structure, developed no defects, except that on one
of the piers there was found a wedge-shaped horizontal band of laitance,
from 4 to 6 in. wide, a section of which the diver was able to remove
to a depth of about 4 in. Investigations showed that this band of
laitance was due to several days' stoppage of the work of depositing
the concrete at this level, and that, although attempts had been made
to remove the laitance with a clam-shell bucket and the help of a
diver before the depositing was again started, the efforts were not
entirely successful.

Annual general inspections of all the structures on the Central
Railroad are made by the Bridge Engineer, and no disintegration
similar to that found in salt water has ever been observed in concrete
piers in fresh-water streams, although in many cases the concrete is
exposed to damage from floating ice. Concrete surfaces have been
fotmd injured (probably by floating ice), but no disintegration has
taken place. This rather disproves the theory that the disintegration
of concrete in sea water, between tides, is due solely to the alternate
freezing and thawing of the concrete.

The result of the speaker's experiences with the structures cited
leads him to believe that concrete can be used safely in sea water, pro-
vided there is no danger of the surfaces being injured by floating
material. Where such action is at all likely to take place, it has been



682 DISCUSSION ON TESTS OF CONCRETE IN SEA WATER

Mr. decided to use a stone facing backed up with concrete. The speaker
^^^^' has no hesitation in using properly deposited concrete to a point 2 ft.

below low tide, or to a height that is below any likelihood of injury

by floating material.

Mr. J. R. McClintock,* M. Am. Soc. C. E. — The speaker had an experi-

Ciintock. ence, at Clarksburg, W. Ya., with disintegration of some concrete
beams across the top of an open coagulating basin. These beams are
normally above the water level, but at times, owing to careless operation
of the pumps, they are alternately submerged and exposed, causing
in cold weather very severe freezing conditions.

The basin was constructed 7 years ago, and after about 5 years
probably two-thirds of these beams were almost completely disinte-
grated, the concrete looking practically like a loose shale. The other
beams, although under the same conditions, were practically intact.

As far as known, there was no difference in construction methods.
They were all built by the same gang, with the same foreman, and
presumably of the same materials. It is possible, however, that in
some of these beams use was made of a local stone which is subject to



Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 55 of 167)