American Society of Civil Engineers.

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

. (page 53 of 167)
Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 53 of 167)
Font size
QR-code for this ebook


648 TESTS OF CONCRETE IN" SEA WATER

"The specimens were made on the first floor of Building No. 107
at the Navy Yard, which was kept above freezing point but yet not
very warm. The feeding chute was removed, and materials were
shoveled directly from the floor to the mouth of the mixer, which was
about 3 ft. above. The materials were dumped into a water-tight
mortar box, from which they were shoveled into wheel-barrows and
wheeled to forms made of plank a short distance away. The specimens
were cast in a horizontal position. Embedded in each for its full
length and projecting from the end (forming a loop) was a f-in. square
twisted steel bar, bent in its middle (embedded near diagonal corners),
to give the specimens stiffness, so that they could be handled without
breaking. There was also embedded in the upper end of each specimen
for a depth of 8 ft. a steel pipe, 2^ in. outside diameter. This was
used merely to core a hole and, after the specimens were set, was
removed in the following manner: Before placing, it was coated on
the outside with paraffin. After the specimens were set, live steam was
turned into the pipe, which then was removed easily and quickly. In
the very top of each specimen was embedded a 3-in. pipe which was
threaded and capped. The object in coring this hole and capping it
was to see if sea water percolated into the center of specimens. (This
was used afterward for fastening hooks supporting the specimens when
the loops had rusted away.)

"The sand and stone were measured in accurately made frames.
In measuring they rested on a perfectly smooth hardwood floor. The
materials were struck smooth on top with a straight-edge. The frames
were made deep in order to avoid the large errors in volume which occur
when frames are very shallow and cover a large area. The sand and
stone were completely fed into the mixer first ; then cement was shoveled
in, which took about 1 min. After i min., water was added with pails,
being accurately weighed. This took about 1 min. to add. In every
ease, in as near 2 min. as possible after the first water was added,
the batch was dumped from the mixer. It took from 15 to 18 min.
to place a batch entirely in the forms, into which it was thoroughly
spaded and tamped. The workmanship was intended to reproduce as
nearly as possible actual commercial conditions.

"The first nine specimens, which were made from a standard
average composition Portland cement, were made by mixing together
equal parts of Vulcanite, Alpha, and Giant. These Avere thoroughly
mixed in the mortar box and then repacked by weight. The cement
free from iron was Blanc Stainless; the commercial Portland high in
alumina was Atlas; the commercial Portland cement low in alumina,
was a mixture of equal parts of Lehigh and Ilelderberg; the iron ore
cement practically free from ahmiina was Hermmoor Erz German
Portland cement; and the Portland cement, made from blast-furnace
slag and limestone, was Universal.



TESTS OF CONCRETE IN SEA WATEE



651



"The stone was broken trap rock from the quarries of E. Burnett,
Franklin Park ledge, Maiden, Mass. A cubic foot of stone weighed
90.24 lb. Determination of voids in stone was 50 per cent. The
stone passed a l:|-in. ring, with the dust screened out.

"The sand was clean and sharp, a little finer than usual, from
the bank of P. O'Riorden, at South Acton, Mass. Dry, it weighed
97.18 lb. per cu. ft. Determination of voids in the dry sand was
34.2 per cent. As used in the work, the sand contained 4.6% of
moisture by weight.

"With the above determined percentage of voids, the mixture,
nominally 1:2:4 but proportioned so as to have 10% excess of cement
over the voids in sand, and 10% of mortar over voids in the stone,
was actually proportioned 1 : 2^ : 4^.

"The specimens remained in the building for about 5 weeks. They
were then loaded on flat cars and moved to Pier 9 in the Yard, and
were hung from the cap log, as shown by Fig. 1. During the loading
of the specimens, they were calipered in each direction close to each
end and every 2 ft. along their length. Their length was also measured
and the length to which the core pipe was embedded. Before hanging
from the pier, they were weighed on platform scales.

"In the process of moving and setting. Specimen No. 22 was cracked
clear through, at about 6 ft. from the lower end, and Specimen No.
23 had a very small crack on the face, about the middle.

"From concrete left over from filling the forms, fifteen 8-in. cubes
were made of mix similar to Specimen No. 3, and fifteen others were
mixed similar to Specimen No. 9. These were put in a box for con-
venience in keeping them together; one-third were immersed per-
manently; one-third were supported at half -tide; and one-third were
kept in a dry place exposed to the weather."

The analyses of sand and stone made by Mr. Herbert L. Sherman
for the Aberthaw Construction Company, as given in Table 1, apply
to all the specimens.

TABLE 1. — Analyses of Sand and Stone Used in all Specimens.



Bank Sand.



Fineness, passing sieves, by weights.



Voids by volume 34.2 per cent.

Moisture '1-6 "

The sand Is very fine, but is clean and of good quality.



No. 100


No. 50


No. 30


No. 20


No. 16


No. 10


No. 8


No. 6


14-inch.


4.8%


24.00/0


52.00/0


73.4%


86.406


91.0%


93.4%


96.1%


100.0%



652



TESTS OF CONCRETE IN SEA WATER

TABLE 1.— (Continued).



Broken Stone.



Fineness, passing sieves, by weights.



^-in.


3/io-in.


V4-in.


%-m.


^-iQ.


%in.


1-in.


l!^-in.


0.25JJ'


0.500/i,


1.750o


9.75%


25.25%


53.0%


80.75%


lOO.Oo/o



Voids by volume 50.0 per cent.

The stone is a broken trap rock of very good quality.



Table 2 shows the analyses of the various cements used in prepar-
ing the specimens.



TABLE 2.—
By Mr. Herbert



Brand


I


Vulcanite,

Alpha,

(iiant.

Mixed equally.

(Average alumina.)


Lehigh,

Helderberg,

Mixed

equally.
(Low alumina.)


Atlas.
(High
alumina.)


Blanc.
(White
Portland).


Percentage passing No
sieve


100


94.8

79.6

1 hr. 35 min.

4 hr. 45 min.

22
313

660
755


93.8

78.8

2hr. 10 min.

5 hr. min.

22
312
636
741

22

184

312

9

m

271
10

1.04
23.40

5.61

2.37
62.90

2.51

1.53
14 and 15


94. 2

76.2

1 hr.45min.

4 hr. 10 min.

22
342
677
744


94.4


Percentage passing No


200


75.0


Initial set


hr. 25 min.


Final set


3hr. 20 min.


Percentage of water


23


24 hr. neat, pounds tension.
7 days " " '•
28 days "
48 hr. moist air, pounds, ten-


277
549
726


Percentage of water


264
360
9
217
318
10

0.9fi
22.08
7.21
2.57
62.60
2.74

1..56
1 to 9, inclusive. (
20 to 24, inclusive. )


22
272
304
9
168
211
10

1.06
21.46
8.50
2.44
61.64
2.58

1.75
12 and 13


23


7 days, Ottawa sand, 1 :3

28 '■■ " " 1:3

Percentage of water


191

319

9




159


28 ' 1:3 ..


218


Percentage of water


10


Loss on ifctnition. Percentage.
Sihca (SiOoi,
Alumina (AloOs) "
lion oxide (FeoOs) "
Lime iCaO) "
Magnesia (MgO) "
Sulph. anhydride (SOs), Per-
centage


1.69
24.58

8.22

0.29
62.70

1.11

1.32


Specimen numbers


i


10 and 11



TESTS OF CONCRETE IN SEA WATER



653



The analysis of the water used in mixing the specimens is indicated
clearly enough by the analysis of tap water at the time, as stated in
the report of the Metropolitan Water and Sewerage Board. It is as
follows :

Analyses of Water (Report of Metropolitan Water and Sewerage
Board, 1909, p. 232.) Water taken from tap at State House,
Boston, Mass., February 1st, 1909.
Parts per 100 000.
Appearance: Turbidity very slight. Sediment, slight.

Color: Platinum standard. 20.

Odor : Cold : Decided geranium, Asterionella.

Hot: " " "

Residue on evaporation: Total 3.5. Loss on ignition, 1.2.

Cement Analyses.
L. Sherman.



Uuiver-


Hermmoor
Krz.
(Iron ore.)


Vulcan
Alpha.
Giant .
Pulv. e


ite. . . ) Mixed

> equally.

\ 95%


Vulcanite. . ) Mixed

Alpha V equally.

Giant ) m%

Hyd. lime, IOOq.


VulcaE
Alpha.
Giant .
Mixed


'te - I Mixed
("equally.




ay, 596.


sea water.


9.';.6


95.8

82.6
45 min.


Clay

Clay
Ihr.


99.0

58.C
40 min.


Hyd. lime 95.2

Hyd. lime 88.2
10 min.




78.8




35 miu.


1 hr.


15 min.


2hr.


26 hr.


4hr.


40 min.


4 hr.


4hr.


30 min.


22
315
608
717


22

0.0*
441
641

223

22

132

234

10

108

174

11

0.91
24.28
0.94
9.08
62.12
0.43




23
248
723
762


25

276
566
717




22
339
690
735


22
194
258




33
211
255
9
169
223
10

0.96
22.08
7.21 AI.,
2.57 Fea

3.74 -t;^


25
148
216
9
117
182
10

20.24 (18.73 1
1.33 and in

^,= [- 0.68
Os )

45.88

31.62


22

204


9
166
181


9
223


10

2.82
20.42

8.04

3.04
62.16

1.63


^'ater)
soluble


10
matter


1.73


1.75




1.56 fl CaOo 1.51






18 aDd 19


16 and 17




24 ffi 22




31



* No strength on 24-hr. neat test ; cement did not have hard set.



654



TESTS OF CONCRETE IN SEA WATER



Ammonia: - Free, 0.0016

• -1 m n ^^.^o ( Dissolved 0.0108
Albuminoid, Total 0.0128 ' ^ , ^ ..

] Suspended 0.0020

Chlorine: 0.38

Nitrogen: as Nitrates, 0.0060; as Nitrites, 0.0000.

Oxygen consumed: 0.17

Hardness : 1.7

The chemical analysis, by A. L. Babbitt, Navy Yard Chemist, of
the salt water of Boston Harbor in which the specimens are immersed,
at low tide, is as follows (Samples of sea water taken February 1st.
1916, from slip between Piers 8 and 9, for use in report of concrete
test specimens) :

Specific gravity 1.0220

Total solids 32 100 parts per million.

Silica 1.6

Iron and aluminum oxides .... 3.6

Calcium 365.9

Magnesium 1 138.7

Sodium 9 163.0

Potassium 508.6

Sulphate radical (SO,) 2 232.2

Chlorine 16 350.0

Hypothetical Combinations.

Parts per Grains per

million. U. S. gallon.

Silica 1.6 0.10

Iron and aluminum oxides 3.6 0.20

Calcium carbonate 780.4 45.5

Calcium sulphate 181.2 10.6

Magnesium sulphate 2 637.7 153.8

Magnesium chloride 2 370.1 138.2

Sodium chloride 23 290.5 1 357.8

Potassium chloride 966.3 56.3

And at high tide as follows:

Specific gravity 1.0217

Total solids 31 820.

Silica 1.2



parts per million.



TESTS OF CONCRETE IN SEA WATER



655



Iron and aluminum oxides ....

Calcium ,

Magnesium 1 117.1

Sodium 9 041.0

Potassiiim 503.4

Sulphate radical (SOJ 2 286.6

Chlorine 15 980.0



3.6 parts per million.
354.5



Hypothetical Combinations.

Parts per Grains per

million. U.S. gallon.

Silica 1.2 0.07

Iron and aluminum oxides 3.6 0.20

Calcium carbonate 844.2 49.2

Calcium sulphate 39.6 2.3

Magnesium sulphate 2 831.5 165.1

Magnesium chloride 2 132.1 124.3

Sodium chloride 22 980.4 1339.8

Potassium chloride 956.4 55.9

Table 3 shows the scheme of the tests. For the purpose of this
report, they have been arranged in eleven series, according to the
cement used. Those in series II to XI, inclusive, in every case
have parallel tests in Series I, which is the principal one.
Table 3-A gives, for each specimen :

(a) The important dates;

(h) The materials of which made;

(c) The proportions.

The percentage of water used, as given in the tables, is not an abso-
lute guide to the quantity in the concrete, because, during a portion of
the tests, a snow storm caused the sand and stone to become wetter
initially.

The progressive condition of the specimens is shown in Table 4
where the information is collated for convenient reference.

The observations as to condition of the specimens were made at
various times by Mr. E. S. Phelps, DeWitt C. Webb, M. Am. Soc. C. E.,
Civil Engineer, U. S. Navy, and by the writer.



656



TESTS OF CONCRETE IN SEA WATER



TABLE 3.



Propor-
tions.


Series I.

Mixed Cement.

Avkrage Alumina.

AL2O3 7.210.0

MgO 2.740'o


Series II.

Blanc

Cement.

AL0O3 8.3305

MgO 1.11%


Series III.

Atlas

Cement.

AL2O3 8.505

MgO 2.58%


Series IV.
Helderberg
AND Lehigh

Cement.
AL2O3 5.61%
MgO 2.510/0


Series V.

Hermmoor
Erz Iron Ore

Cement
AL2O3 0.94%
MgO 0.43%




No. of
Speci-
men.


Water, in
pounds per
cubic foot.


No.


Water.


No.


Water.


No.


Water.


No.


Water.


1:1 : 3
1:1 :2
1:1 : 2


1
2
3


Dry 9.3

Wet 10.4
Very wetl2.5


10


Wet 10.2


13


Wet 11.0


14


Wet 11.7


16


Wet 11.1


1 : 24 : 4)^
1:24:4^
1 : 21^ : 44


4
5

6


Very dry 5.9
Wet 7.1
Very wet 12.3


















1 :3 :6
1:3 :6
1:3 : 6


8
9


Very dry 5.8
Wet 7.7
Very wet 11.3


"]


Very )
wei V
12.1 )


H


Very )

wet y

11.3 \


H


Very )
wet >
11.7 5


"\


Very )
wet V
10.2 )



Note.— The weight of water, in pounds per cubic foot, takes

A very careful examination, by three experienced observers, was
made in April, 1914, at low tide, of all the specimens, and they were
graded and classified on the spot in accordance with their condition
at the time. The estimates of grade and class were on the basis of
durability or serviceability of the concrete, as if it were in actual
service. The examination was made after the specimens had been in
the water 5 years, since which time there have been no changes of note.
The various classes, especially the upper ones, grade into one another
with definite, but without particularly strong, dividing lines. The
specimens in each class are arranged from left to right in the order of
durability.

The following is a copy of the report on classifying and grading
the specimens :

"April 1st, 1914.

"Specimens graded largely as to condition at point of maximum
deterioration.

"Specimens in order of durability.

"Elevation of tide, 1 ft. above mean low water at time of ex-
amination.



TESTS OF CONCKETE IN SEA WATER



657



Scheme of Tests.



Series VI.

Universal

Cement
Al^Os 8.O4O0
MgO 1.6yo/o


Series VII,

LIKE Series I.

Concrete

EXT HA
WELL MIXED.

AL0O3 7.2106
MgO 2.740-0


Series VIII.

LIKE Series I,

using .=ea

WATER.


Series IX,

LIKE

Series I,

WITH LIME.


Series X.

LIKE

Series I,

WITH

Sylvester
Solution.


Series XI,

LIKE

Series I,

with
596 Clay.




No.


Water.


No.


Water.


No.


Water.


No.


Water.


No.


Water.


No.


Water.




.8-;


Wet)
11.5 (■






















Series XII.




























H


Very 1
wet V
13.03 )


H


Very )

wet y

9.7 \


■i


Very )

wet >
9.7 )


H


Very 1

wet y

10.7 )


H


Very)
wet V
9.6 )


H


Veryl
wet V
9.5 )


Series XIII.



no account of moisture in the ingredients before mixing.

Class 1 : Perfect condition ; No. 14.

Class 2: Nearly perfect condition; Nos. 3, 16.

Class 3: Very fine condition, but with slight deterioration; Nos.

18, 12, 10, 2.
Ciass 4: Yery good condition, but with some deterioration; Nos.

6, 5.
Class 5: Good condition, but with deterioration apparent; Nos.

15, 20, 24, 17, 1.
Class 6 : Fair condition ; deterioration apparent ; concrete soft in

spots near low- water mark; Nos. 9, 11, 13.
Class 7: Poor condition; decided deterioration; concrete shows

deterioration at low water; Nos. 4, 8.
Class 8: Very bad condition; concrete disintegrated and badly

eroded above low water; Nos. 22, 19, 23.
Class 9 : Exceedingly bad condition ; concrete totally disintegrated

and eroded above low water line. No. 7."

The following specimens showed deterioration between the 1914
and 1916 inspections, arranged in the order of greatest deterioration:
Nos. 22, 19, 23, 7, 11, 4. The remaining specimens show no noticeable



658



TESTS OF CONCRETE IN SEA WATER

TABLE 3-A.— Data Kelating







Dates: (All in 1909.)












01














t


3
3
3


7


Quite dry.


Jan. 11


Jan. 13


Feb.


27 -j


Equal parts of
Vulcanite, Alpha,
and Giant.


\


1:3:6


4


8


Plastic.


Jan. 11


Jan. 13


Feb.


H


Equal parts of
Vulcanite, Alpha,
and Giant.


\


1:3:6


4


9


Wet.


Jan. 11


Jan. 13


Feb.


H


Equal parts of
Vulcanite, Alpha,
and Giant.


I


1:3:6


4


10


Wet.


Jan. 12


Jan. 14


Feb.


27 -{


Blanc
(white Portland).


1


1:1:2


3


11


Wet.


Jan. 12


Jan. 14


Feb.


27 -j


Blanc
(white Portland).


\


1:3:6


m


12


Wet.


Jan. 12


Jan. 14


Feb.


27 -j


Atlas, high in alu-
mina.


1


1:1:2


3


13


Wet.


Jan. 12


Jan. 14


Feb.


27 ]


A.tlas, high in alu-
mina.


1

1"


1:3:6


4


14


Wet.


Jan. 13


Jan. 15


Feb.


23 J


Equal parts of
Helderberg and
Lehigh.


i


1:1:2


3


15


Wet.


Jan. 13


Jafl. 15


Feb.


H


Equal parts of
Helderberg and
Lehigh.


[


1:3:6


4


16


Wet.


Jan. 15


Jan. 18


Feb.


23 1


Iron ore (Herm-
moor Erz) (Ger-
man).




1:1:2


3



TESTS OF CONCRETE IN SEA WATER

TO Mixture of Concrete.



659



Weights (All in Podnds) :


i

D




,




1


u -


CO




d

^ O

o


o u;


n


d

a

o




Online LibraryAmerican Society of Civil EngineersTransactions of the American Society of Civil Engineers (Volume 81) → online text (page 53 of 167)