American Society of Civil Engineers.

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

. (page 42 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 42 of 167)
Font size
QR-code for this ebook


bearing in mind that this timnel was not in "ideal" or "made-to-order"
ground. The West Portal earth section was very difficult, and the rock
section was hard to break, as illustrated by the average explosives used
in the pioneer heading, that is, 15 lb. of 60% gelatine per cu. yd.

Mr. Dougherty's point, that all excavation should be taken out dur-
ing construction to allow for lining, is well taken. The extra excava-
tion can be done during construction for probably 10% of the cost of
enlarging imder traffic. No tunnel is entirely safe until it is lined.



AMERICAN SOCIETY OF CIVIL ENGINEERS

INSTITUTED 1852



TRANSACTIONS



This Society is not responsible for any statement made or opinion expressed
in its publications.



Paper No. 1391

UNUSUAL COFFER-DAM
FOR 1 000-FOOT PIER, NEW YORK CITY*

By Charles W. Staniford, M. Am. Soc. C. E.



With Discussion by Messrs. Frederic R. Harris, D. A. Watt, C. A.
Wentworth, Thomas H. Wiggin, T. Kennard Thomson, Charles
S. Boardman, WilliaxM M. Black, and F. E. Cudworth.



Synopsis.

This paper is descriptive of the design and construction of a steel
sheet-pile coffer-dam of unusual type. This dam, so far as known, is the
highest on record.

In outline the coffer-dam is a cellular steel sheet-pile structure,
designed for a pressure head, of soft mud and water, of about 65 ft.,
having the cellular spaces filled with earth and clay to form the water
seal, and held against the external pressure by an embankment of
rip-rap or broken rock. All the steel sheet-piling forming the cellular
spaces or "pockets" is in single pieces, many of which are more than
TO ft. long.

Descriptive and Historical.
On the refusal of the Secretary of War to grant any further exten-
sion of the pierhead line in the Chelsea District, thereby preventing
permanent additions to the present piers and thus making it impossible
to dock the new large transatlantic liners — more than 900 ft. long — at
that locality, the City of New York was obliged to secure another site



* Presented at the meeting of March 7th, 1917.



COFFER-DAM FOR 1 000-FOOT PIER 499

which would permit the construction of piers long enough for such
vessels; and on April 10th, 1913, the Commissioner of Docks, Mr.
E. A. C. Smith, presented a plan providing for a pier 150 ft. wide,
at the foot of West 46th Street, extending 1 000 ft. inshore from the
established pierhead line, and a half-pier at the foot of West 44th
Street, with a slip room of 360 ft. between them, and a temporary
slip, on the north side of the West 46th Street pier, 237 ft. 6 in. wide.
This plan was adopted by the Commissioners of the Sinking Fund on
April 30th, 1913, and steps were immediately taken to acquire the
necessary property.

It was apparent from the general configuration of Manhattan
Island in this vicinity that rock would be encountered over the inshore
portion of the site of the proposed pier and slips. This rock surface
was developed inshore from the existing bulkheads by wash-borings,
and outshore from the bulkheads in the then existing slips by shod
test piles. The development over the entire pier site out to the pier-
head line, by these means and the usual test piles, indicated that the
character of the river bottom varied from rock at an average depth
of about 20 ft. below mean low water at the inshore end to a very soft
mud at the outshore end, where it would be necessary to use fully
lagged piles, from 85 to 89 ft. long.

In order to make proper depth of water for the giant liners, drawing,
when loaded, about 38 ft., it was determined to provide for a depth
in the slips of 44 ft. below mean low water, as records show that
extreme low water has occurred 4 ft. below mean low water, and it
was decided that there should be a clearance of at least 2 ft. between
the ship's keel and the bottom of the slip, especially in that portion
where rock forms the floor of the slip.

The sub-surface conditions thus developed and the decision to
secure a depth of 44 ft. below mean low water in the slips plainly
indicated the necessity for adopting a type of pier which would reduce
to a minimum the quantity of submarine work to be done, and a
method of procedure which would insure safe, workmanlike, and
economical construction. Accordingly, a solid filled type was adopted
for the inshore portion, about 220 ft. long, consisting of concrete
retaining walls, to retain the filling and provide wharfage face, founded
on bed-rock, thus utilizing the rock bottom over this portion of the
pier area.



500



COFFER-DAM FOE 1 000-FOOT PIER



Holland American
North German Lloyd,
Hamburg Ameri




-Pierhead Line approved by Secretary of War March 2nd 1897

DETAILED LAYOUT
WEST 46th STREET IMPROVEMENT
FOR LONG STEAMSHIP PIERS
Pig. 1.



COFFER-DAM FOR 1 OOO-FOOT PIER 501

The larger quantity of rock excavation necessary with submarine
blasting over that within the colfer-dam, in order to make certain
that the excavation would be carried to the lines and grades required;
the advantage of being able to control the surfacing of the rock for
the pier and slip walls, and to prepare the rock surface for the
foundation of these walls; and the desirability of building these walls
"in air"; all indicated the wisdom of prosecuting the work within
a coffer-dam to be built so as to enclose the entire site of the work,
extending from the center line of West 44th Street northward to the
center line of West 47th Street, a distance of 780 ft.

Design.

Having decided on the coffer-dam method of construction, the type
to be used was made the subject of careful consideration. The area
to be enclosed prohibited the use of bracing, and the unbraced steel
sheet-pile type of coffer-dam was naturally suggested. Coffer-dams
of this type had been used previously by the United States Govern-
ment, at Black Rock and at Havana. There it was expected that
the filling within the pockets would make them stable against the
external pressure, unaided, but in both cases it was found necessary
to place an embankment of stone against the pockets, and in the
case at Havana, extensive bracing to the Maine was also placed, in
order to stop the continuous inward movement of the coffer-dam.

In a coffer-dam of this type the internal frictional resistance
of the filling takes the place of the cohesion in a solid, and analysis
would show that, with material having a natural slope of not more
than 3 to 1, failure would occur in vertical as well as horizontal shear,
though the resultant passes through the middle third of the base.

The material overlying the rock at West 46th Street is a clayey
mud, the natural slope of which is probably not as good as 3 to 1,
thus prohibiting its use to obtain stability.

It had been decided to dredge over the site of the coffer-dam to
30 ft. below mean low water, leaving the remaining material to furnish
support for the sheet-piles. The difficulty of removing the material
from within the pockets and substituting better, as well as the greater
cost of a coffer-dam of this type, due to the larger quantity of steel
sheet-piling required, led to the adoption of the type shown by Fig. 2.



502



COFFER-DAM FOR 1 000-FOOT PIER



It should be noted that, where no embankment is placed against
the pockets, the sheet-piles will be subjected to the pressure of the
entire depth of filling at all times, and as the water after pumping
may be expected to remain in these pockets for at least one-half their
depth, if not more, as was the case at Havana, these sheet-piles will
also be subjected to this hydrostatic pressure, necessitating a closer
spacing of the diaphragms than where embankments are used.




PLAN OF SHEET-PILE POCKETS
Fig. 2.

In the type of cotfer-dam adopted, the sheet-pile pockets are used
merely as a water seal, being supported against the external hydro-
static pressure by an embankment of rip-rap inside, and against the
pressure of the rip-rap, during construction and before pumping, by
a smaller embankment of earth outside.



COFFER-DAM FOR 1 000-FOOT PIER



503



This type of coffer-dam was adhered to, except along the crib bulk-
head at West 47th Street, where, in place of the sheet-pile pockets, a
single line of steel sheet-piling was driven, with an embankment of
rip-rap in front, the space between the sheet-piles and the crib being
filled with earth. Also, because of the limited space between the
inner end of the West 44th Street Pier and the — 44-ft. rock contour
line, two large cylinders filled with rip-rap were substituted for the
pockets and embankment at this point. The only leaks, as described
later, were encountered at these two places.




Fig. 3.

Pockets 6 and 22 were made 12 ft. wide and provided with sluice^
ways at a grade of 1 ft. below mean low water. These were put in
chiefly to maintain the water at the same level on both sides of the
coffer-dam during the construction of the closing pockets, thus avoiding
any imbalanced pressure due to changes of tide before the placing



504



COFFEE-DAM FOR 1 000-FOOT PIER



of the rip-rap embankment, and also to let the water into the coffer
after the completion of the excavation and construction.
f ■ The specifications called for the following:

"The steel sheet-piling shall be rolled, interlocking sheet-piling
similar and equal to that manufactured by the Carnegie Steel Com-
pany, type M-104, |-in. web, and weighing 38 lb. per lin. ft., or that
manufactured by the Lackawanna Steel Company, |-in., straight web

Key to extend for
fuU height of iralL



S.plj tar.paper secured to
end of wall bj a mopping
■y.f: yyycf; of.aspb<io cement.




HORIZONTAL SECTION AT VERTICAL
JOINTS OF PIER AND SLIP WALLS



Dotted lines India&te X
top ot Waet Wall of [T
Pier.



4;^' Not Use Ihan 6*




WhewTer present rock
fiorface has & dlope of gre*t«r
than 1 to 10 transrerselj of
the trail, it shall be benched
ae Indicated.



^i- l.o' Not less than 6*
C3^it^ ElcaT'n-44




Atoraee Wall Sect.
i:iev. of Rook from -44 tOr48
at back line of wall.



TYPICAL CROSS-SECTIONS OF
PIER AND SLIP WALLS



FlQ. 4.



section, and weighing 37.2 lb. per lin. ft., or that manufactured by
the Jones and Laughlin Steel Company, 12-in., special I-beam section,
and 5-in. locking bar, weighing, together, 36.1 lb. per lin. ft., and
shall develop a tensional strength in the interlock between two adjacent
piles, when tested on pieces about 3 in. long, of not less than 9 500 lb.
per lin. in."

The contractor selected the type made by the Lackawanna Steel
Company, shown on Fig. 2, which proved very satisfactory. The



PLATE VIII.

TRANS. AM. SOC. CIV. ENGRS.

VOL. LXXXI, No. 1391.

STANIFORD ON
UNUSUAL COFFER-DAM
FOR 1 000-FOOT PIER




PROFILE OF COFFER-DAM ON C.L. OF POCKETS



COFFEE-DAM FOR 1 000-FOOT PIER 5G5

three-way connection pile used at the intersection of the transverse
and outer walls of sheet-piling weighed about 85 lb. per ft.

Stresses in Sheet-Pile Walls of Pockets.
Due to the pressure of the retained filling, the sheet-piles are all
in tension transversely. Their resistance to this tension is limited
by the strength of the interlock, which tests show will exceed 9 000 lb.
per lin. in. When the filling is done in a proper manner, the pressures
from each side of the transverse walls will balance each other, and
the tension will be entirely due to the pressure against the longitudinal
sheet-pile walls.

p d

where t = the tension, in pounds per linear inch, in the transverse walls ;
p^the pressure, in pounds per square foot, against the longi-
tudinal walls at the level considered; and
d = the distance between the transverse walls, in feet.

The longitudinal sheet-pile walls are arcs of a circle, and, assuming
them to be subjected to a radial pressure from the filling, the tension
per linear inch is :

_p E
^1 — 12"'

where B is the radius, and tj^ = t where R =^ d.

The original plan required the pockets to be filled with the material
remaining on their inshore sides after dredging. It was considered
possible, therefore, that they might be entirely filled before depositing
the rip-rap embankment. In this case the sheet-piles would be sub-
jected to the pressure of about 58 ft. of filling. Assuming the weight
of the submerged river mud at 80 lb. per cu. ft. and its natural slope
at 3i to 1,

p = 0.57 X 80 X 58 = 2 600 lb. per sq. ft.,

and as r = d, the maximum stresses in the longitudinal and trans-
verse walls are equal and given by

2 600 X 24
t = -f^ = 5 200 lb. per lin. in.

The possibility of avoiding unduly straining the sheet-piles by
depositing some of the embankment before completely filling the



506



COFFEE-DAM FOR 1 000-FOOT PIER



pockets was considered, and was actually found advisable, as explained
later.

Tests on Model Embankment.

To obtain information regarding the probable behavior of a rip-rap
embankment under various conditions of pressure, some tests were
made with a model, one-twentieth the size of the embankment, placed
againrt the inshore side of the sheet-pile pockets, or 42 in. in height

Grade on front line top of wall= + 9,65 '
, / .. .. rear .. = +9.71' |4', 3'6'

219.0^ >^V>i




PLAN



I RJDgljolt gJDEle bitt-east-Steel



=|b=^M=Hfe=jp^






^'^ ^ 'I ,




m



This Burfaoe to be produced by "jLdl^''_i^
ChttDneling or bj Drillingand Broadling.

: 1 r.rrA.' nt Ruck in Slips - 14'^



FRONT ELEVATION SOUTH WALL OF PIER

Column foundation




-*j4'rt Single bitt



Single bitt
Grade on frontline top of wall = V 9.65'
.. .. rear " .. •< .• =^.9.7x'

PLAN
Fig. 5.



at the rear. These tests are incomplete, but, though it was found
necessary to discontinue them, they are given here for whatever value
may be derived from them.

The testing apparatus provided a bottom having the average slope
of the rock floor at West 46th Street, 6 to 1, and also glass sides, 3 ft.
apart, in order to observe the effect of pressures on the mass. The



COFFER-DAM FOR 1 000-FOOT PIER 507

board, against which the embankmeut was placed and by which the
pressures were transmitted to the embankment, was fitted so as to
slide inward on the bottom and between the glass sides. The embank-
ment consisted of broken stone weighing as placed about 93 lb. per
cu. ft., and varying from ^ to | in., which is about one-twentieth
the size of rip-rap. The pressures were obtained by using two calibrated
steel car springs, which were compressed by a nut working on a
threaded bolt, passing partly through each spring, and fastened to
the backing blocks attached to the base of the apparatus. The pressures
applied were determined from the decrease in length of the springs,
a gauge being used for each spring, giving the pressure directly in
pounds.

In order to determine the behavior of the embankment under pres-
sure, vertical lines were marked on the ground-glass sides, and along
them columns of small wooden blocks were laid up in the embank-
ment. These columns of blocks moved with the embankment, and
the extent of their displacement from their original positions was
measured as the pressure was increased. Figs. 6, 7, 8, and 9 are
reproduced from photographs showing various views of this model.

Six tests were made with this model imder different conditions, with
the results shown by Figs. 11, 12, and 13.

In Test 1, the board transmitting the pressure extended the full
height of the embankment, or 3 ft. 6 in. The embankment had a berm
of 10 in., and the pressure was applied at one-third the height from
the bottom.

In all the tests the embankment was 3 ft. 6 in. high, but, in all but
Test 1, the board was only 33 in. high, being placed against the upper
33 in. of the embankment, the lower edge of the board being approxi-
mately in the horizontal plane through the toe. The pressure in
Tests 2, 3, and 4 was applied at one-third the height of the board
from its lower edge. Tests 5 and 6 will be referred to later.

The object of these tests was to determine the extent of the distor-
tion which would take place in such an embankment under pressure,
as well as its ultimate resistance.

Figs. 11 and 12 show the distortion throughout the mass in Tests
1 and 2, obtained by measurements of the displacement of the blocks
of wood referred to. Also, the curves of deflection of the embankment
with the increase of pressure are shown by Fig. 13.



508 COFFER-DAM FOR 1 OOO-FOOT PIER

In Test 1, at a pressure of about 2 375 lb., a tendency to break,
about through the horizontal plane through the toe, was observed,
and is indicated by the breaks in the lines in Fig. 11 at this pressure,
as well as in the break in the curve of deflection shown in Fig. 13.

The pressure above the horizontal plane through the toe corre-
sponding with 2 375 lb. on the whole board is 1 460 lb. The weight
of the embankment above this plane is about 1 900 lb., and the corre-
sponding value of [X is 0.77, as compared with 0.83 indicated by the
natural slope of 1.2 to 1.

A second break occurs in the deflection curve in Fig. 13 at a
pressure of 2 750 lb. Assuming this as an indication of sliding up
the inclined bottom, taking the weight of the embankment at 2 380 lb.,
the tangent of the angle made by the bottom with the horizontal at
J, and substituting those values in the formula,

_ P — tan. a W
^ ~ TF-h tan. a P'

we have a value for ^, the coefficient of friction of the stone on the
wood bottom, ^ 0.83. This high value is presumed to be due to
the probability of the stones becoming embedded in the wood to some
extent.

Considering the break in the deflection curve for Test 2, Fig. 13,
at 1 280 lb. as an indication of sliding along the horizontal plane
through the lower edge of the pressure board, the weight of the
embankment above this plane being about 1 750 lb., the resulting value
of /A for broken stone on broken stone is 0.73. The critical pressure
of 1 000 lb. in Test 3 gives a value for jx of 0.80. These values com-
pare with 0.83 indicated by the natural slope of 1.2 to 1.

The movement of the embankment in these tests, accompanying
each increase of pressure, would apparently cease in a very short
time. The readings of both pressures and movement were then taken,
and the next increase of pressure effected. It became evident, how-
ever, that, particularly for the higher pressures, further movement will
occur when more time is allowed. This is illustrated in Test 2, Fig.
12, under pressure 8, in which case the pressure against the embank-
ment dropped from 1 805 to 1 565 lb. over night, accompanied by a
corresponding movement of the embankment. To determine the effect
of time, Test 4 was made on an embankment similar to that used in




Fig. 6. — Side View of Model, Showing Displacement of Columns of Wooden
Blocks at End of Test 1, When the Pressure Against the Embank-
ment HAD Reached 2 810 Lb. The Black Vertical
Lines Show the Original Position of the
Columns of Wooden Blocks.




Fig 7. — Front View of Model, Showing Slope of Broken Stone Embankment.




-Side and Rear of Model, Before Application of Pressure in Test
Showing Columns of Wooden Blocks Laid Up Along
Black Lines on Glass Sides.




Fig. 9.



-Side View of Model at End of Test 2, When Pressure had Reached
1 810 Lb. Note Displacement of Wooden Blocks.



COFFER-DAM FOR 1 000-FOOT PIER



513



^Feet




514



COFFER-DAM FOR 1 000-FOOT PIER



Test 2, with a berm of 10 in. The springs were compressed mitil they
indicated a pressure of 1 060 lb. The accompanying movement of the
embankment at the top was 0.07 in. At the end of 1 week, the springs
indicated a pressure of 950 lb., and this drop in pressure was accom-
panied by an additional movement of 0.045 in. at the top of the
embankment. At the end of 1 month, the pressure had dropped to
900 lb., with a further displacement of the embankment at the top
of 0.08 in. Observation of the embankment at the end of the second
month indicated no further change either in pressure or displacement.



TEST 1



1


1 225 lb.


z


1555 ..


3


1 970 ■'


i


2 375 "


6


2 550 ■'


6


2 750 "


7


2 810"


Scale of Inches




Fig. 11,

This test does not show that the embankment would not have come
to rest in time under a pressure greater than 900 lb. constantly
applied, which could not be determined with the apparatus used,
as any movement of the embankment necessarily results in a par-
tial release of the springs, and consequently a decrease in pressure.

Tests 5 and 6 were undertaken to determine the effect of applying
the pressure at a point above the center of gravity of the embank-
ment as well as the effect of friction along the face of the embank-
ment in contact with the board.



COFFER-DAM FOR 1 000-FOOT PIER



515



The embankment used in these tests was the same as that in Test
2, except that the width of the berm was about 6 in. In the case of
Tests 5 and 6, the pressure was applied at two-thirds of the height
above the bottom of the board. In Test 5 the board was free to move,
as in the previous tests, and in Test 6 the board was anchored at the
bottom with hinges. The results are given by the curves of deflection
of the top and bottom of the board, as shown in Fig. 13.

The curve for Test 5 indicates an ultimate resistance of about
600 lb. to a pressure applied at two-thirds of the height, which about



TEST 2.

Scale of Inches



*8 =
+ 9 =
10 =

n=

12=



865
1U80
1280
1380

uao

1570

1805-1505

1985

1585

1750

1810



* Springs indicated a drop of
pressure to 1565 pounds over night,

t Pressure was lowered after
No.9 to about lOUO pounds with-
out effect on embankment and
then raised to No.lO,




Fig. 12.

agrees with the theoretical resistance, on the assumption of a coeffi-
cient of friction of 0.83 developed along horizontal planes in the mass.
This is only one-half the resistance of the similar embankment used in
Test 2 under pressure applied at one-third of the height.

The tendency of the rear of the embankment to rise under pres-
sure, "carrying the board with it, was very evident in Tests 1, 2, and 3.
In Tests 5 and 6 the movement was entirely horizontal. To deter-
mine the effect of friction between the board and the face of the
embankment, the board, in Test 6, was anchored as mentioned pre-



516



COFFEE-DAM FOE 1 000-FOOT PIEE



Deflection of Embankment,
M in Inches.

n OiO



Deflection of Top of Embankment, in Inches.
























1%%.






\6°\V






"P^




T






1


k°^ I'sk^






^U5&„


1 1


i^SJ^4 1 1


\\4..rW''^Lj


1 %


"■o^ \ W I


1


^o-^-l IX


1


°l^l 1 °


1


1 r \



•nwW



Deflection of Embankment
in Inches.
















































































O

o

i

•6
o

p
w

tr














































































-




-


-












































































































-


<


I




































H






































\


'i,






















1

E




















































Vo„




































\


V


















\


















o\


^r.


















Vrt"


















^Vi


















^


So


















\o


















\




















S



COFFER-DAM FOE 1 000-FOOT PIEE 517

viously. The result is sho\\ii by the lower curve. Fig. 13. Considering
the resistance developed along horizontal planes in the embankment, and



Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 42 of 167)