Please note: this text may be incomplete. For more information about this OCR, view
About OCR text.
Vol. LXXXVI
MAY 27, igii
No. 2254
A NEW SOURCE OF RENT IN TALL BUILDINGS
The Type of Foundation Invented for the Bankers' Trust Building
Means a Saving of ;_$60,000 and a Handsome Yearly Income
By ALLEN E. BEALS.
WHAT wiil he the evolution of the
Cleopatra needle-like structures now
being erected m congested New York? Tlie
day is tast approaching when it will L>e no
longer economical to build commercial
structures on very small plots. Whole
blocks must be utiiiKed and the city will
carry on its activities halt a liundred feet
below the sidewalk. A construction prac¬
tice, far in advance of present day imag¬
inings, will cut deep into building costs
and at the same time double and treble
the dividends on structural investments.
Stiil more startling is tlie fact that a be¬
ginning toward this ideal already has been
made. True, it is a comparatively small
one. but nevertheless it points the ten¬
dency unmistakably. Engineering and
archilectural peers have solved the princi¬
pal problem, the question of foundations,
aud have applied it in recent construction
works as a nucleus to what in the near
future probably will he a universal system
in sections where land values are excess¬
ively high,
A difference of from $00,000 to $100,000
in foundation costs is an item sufflciently
important to appeal strongly to any tiuild-
er, hut it sinks into insignificance when
the mere matter of type of foundations
permits the owners of the finished struc¬
ture to rent four or more floors which
heretofore have been given over to stor¬
age, service and other purposes not di¬
rectly revenue producing. If the huilding
could he made tenantable from bed rock
to penthouse, and, If every building in a
whole hlocli could have this advantage, the
increased revenue producing capacity of
the square would force the item of taxes
and maintenance into a position of minor
instead of major importance. Here, then,
is the chief factor in speeding the day of
the "Syndicate Block."
Most of our skyscrapers and large huild¬
ings rest on caissons sunk to bedrock or,
in some cases, to hardpan. These caissons
vary in size and length according to the
weight they must support and the distance
they have to be driven into the earth to
flnd a flrm and permanent position. These
are driven, in most cases, in criss-cross
fashion at different points in the lot or
plot. Upon these caissons rest the col¬
umns supporting the superstructure. It is
evident that a number of columns and
braces, massed in the foundation floors of
a building, rob fhe subterranean floors of
a vast quantity of room and, in conse-
quence, they frequently are unused.
fhe problem has been to flnd a system
of foundation construction so as to elim¬
inate these columns. In most cases they
have made it necessary to give over to
service and storage purposes floors which
could be made profltable.
This was one reason why the Gillender
Building at Wall and Nassau streets be¬
came economically unprofltahle so early in
its existence. In its basement were three
caissons sunk to hardpan, two of which
measured 12x27 feet and the third 15x:27
feet. The floors helow the street level were
so interwoven with columns and strusses
that they were of little real value. Dur¬
ing its fifteen years of life the corner upon
which it stood became almost the most
valuable in the city, which means that it
had a revenue producing capacity perhaps
greater than any single piot of its size in
the world. It was, therefore, imperative
that .every inch of space be utilised.
The architectural problems in this build¬
ing were enormous, but here was one
which seemed, for a time, almost unsur-
mountahle. Messrs. Trowbridge & Living¬
ston, the architects; T. Kennard Thomp¬
son, the consulting engineer, and officials
of Marc Eidlitz & Son, the general con¬
tractors, and of the Foundation Company
had many conferences on the subject, and
it was finally decided to depart from gen¬
eral practice and cofferdam the whole plot
by sinking a series of eighteen interlock¬
ing caissons around the outside wails of
the building lo bedrock. This was a de¬
velopment of the principle applied in cof¬
ferdam foundation work under the Mutual
Life and Commercial buildings and the
Stock Exchange. In the case of the Bank¬
ers' Trust Building, however, no interior
caissons were used.
This cofferdam was seven feet thick
from sidewalk to bedrock, about 67 feet
below the street. It was made entirely
of concrete of 1:2:4 . mixture and rein¬
forced with !54-irich round bars between
BANKERS' TRUST BUILDING.
The height to which the building is in¬
closed in the picture corresponds to the
full height of its predecessor.
LuWEi: FLOORS OF BANKERS' TRUST
BUILDING.
Showing steel frame rising from grillage
at top of cofferdam walls.
the caissons. The whole was waterproofed
with hydrolithic cement so that seepage,
carried off by small drains running be¬
tween the footings of the walls, amounts
to only about a quart a day. although the
pressure against the walls is about one
and one-half tons to the square foot, at a
depth of fifty feet where the quicksand
rests upon hardpan, which in that part of
the city, runs from ten to fifteen feet deep
on hard gneiss bedrock.
There were on the site three old caissons
put in by the O'Rourke Construction Co.
in ISOG, which were of no use in the new
building arrangement. They had to come
out. The cofferdam type of construction
solved this difficulty, as it did the others.
Once the dam was finished, open excava¬
tion began and the old caissons were taken,
out by blasting, exactly as the other mate¬
rial, such as quicksand and hardpan, were
removed.
As the excavation work progressed, how-
ex er, the tremendous pressure on the walls
of the cofferdam had to be taken care of.
Other buildings in the vicinity had their
foundations on hardpan and precautions
had to be taken that this was not permit¬
ted to shift. It was necessary to put in
temporary timber strusses from wall to
wall, fore and aft. but the problem was to
install them in such a way that the steel
erectors could lower and set fioor beams
and girders, which, as the accompanying
sketch shows, form the permanent bracing-
for the walls. Here was a nice piece of
engineering calculation to leave room
enotigh for the erectors to work, and at
the same time prevent^ a general cave-in,
due to the outside pressure of 3.360 pounds
to the foot.
So great was this pressure that no tim¬
ber was used in the caissons. This omis¬
sion was in itself a novelty, and it illus¬
trates the progress being made in caisson
work. The accompanying sketch shows thp
construction detail of the caissons there