John C. (John Cresson) Trautwine.

Concrete online

. (page 4 of 23)
Online LibraryJohn C. (John Cresson) TrautwineConcrete → online text (page 4 of 23)
Font size
QR-code for this ebook

lime. With this precaution, the cem, altho it may require more time to
set, than when fresher, does not otherwise very appreciably deteriorate in
many months.

61. Storage, under pressure, tends to the caking of cems, which, there-
fore, does not necessarily indicate deterioration.

62. Restoration by reburning. Cems which have deteriorated by
exposure, may be in great measure restored by reheating to redness.

63. If cem is stored in warm places, jt is apt to "flash" when
mixed with water, i. e., to set much more rapidly than it should.


See Digests of Specifications, A S C E, p 942 ; Engng Standards Comm
of Gt Brit, p 940; Report of Board of U S Engr Officers, p 937.

64. Thoro chemical tests of cem can of course be made only by


out sediment/ Powdered limes'tone or" cem rock, mixed with the cem,
causes violent effervescence, the acid giving off strong fumes until all the
lime carbonate is decomposed, when the yellow jelly forms. Quartz sand
remains undissolved. Reject cem containing these adulterants." Judson,
"City Roads and Pavements." The presence of slag is generally indicated
by the sulfur present, which causes a milky appearance, if the cem be agi-
tated in a solution of hydrochloric acid in water.

65. Fuller and Thompson found that cems, which failed to stand this
test failed also to set properly; while cems which passed it, also passed
more elaborate chemical tests. (Trans A S C E, Vol 59, '07, Dec, pp 73-4.)


Properties and Tests of Cement. Report of Board U. S. A.
Engineer Officers. Properties and tests of Portland, Natural and Puz-
zolan* cements. Digest of a Report of Majors W. L. Marshall and Smith S.
Leach and Capt. Spencer Cosby, Board of Engineer Officers, on testing Hydraulic
Cements. Professional Papers, No. 28, Corps of Engineers, U. S. A., 1901.

Unfortunately, tests for acceptance or rejection must be made on a product
which has not reached its final stage. A cement, when incorporated in masonry,
undergoes chemical changes for months, whereas it is seldom possible to
continue tests for more than a few weeks at the most.

A few tests, carefully made, are more valuable than many, made with less care.

Cement which has been in storage for a long time should be carefully
tested before use, in order to detect deterioration.

A cement should be rejected, without regard to the proportion of failures
among samples tested, if the samples show dangerous variation in quality or
lack of care in manufacture, and resulting lack of uniformity in the product.

The practice of offering a bonus for cement showing an abnormal strength
is objectionable, as it leads to the production of cements with defects not
easily detected.

For Portlandor Puzzolan cement, make tests for (1) fineness of grinding; (2)
specific gravity ; (3) soundness, or constancy of volume in setting; (4) time of
setting, and (5) tensile strength. For Natural cement* omit tests (2) and (3).

(1) Fineness. Cementitious quality resides principally, if not wholly, in
the very finely ground particles. Use a No. 100 sieve, woven from brass wire
No. 40 Stubs gage; sift until cement ceases to pass through. The percentage
that has passed through is determined by weighing the residue on the sieve.
The screen should be frequently examined to see that no wires have been

(2) Specific gravity-. The specific gravity test is of value in determining
whether a Portland cement is unadulterated. The higher the burning, short of
vitrification, the better the cement and the higher the specific gravity. If under>
burned, the specific gravity of Portland cement may fall below 3 ; if overburned,
it may reach 3.5. Natural cement has a specific gravity of about 2.5 to 2.8, and
Puzzolan about 2.7 to 2.8.

The temperature may vary between 60 and 80 F. Any approved form of
volumenometer or specific gravity bottle may be used, graduated to cubic centi-
meters with decimal subdivisions. Fill the instrument lo zero of scale with
benzine. Take 100 grains of sifted cement that has been previously dried by
exposure on a metal plate for 20 minutes to a dry heat of 212 F., and allow it to
pass slowly into the benzine, taking care that the powder does not stick to the
sides of the graduated tube above the fluid, and that the funnel, through which
it is introduced, does not touch the fluid. The approximate specific gravity will
be represented by 100 divided by the displacement in cubic centimeters. The
operation requires care.

(3) Soundness, and (4) setting qualities. The temperature should
not vary more than 10 from 62 F. For Portland cement use 20, for Natural 30,
and for Puzzolan 18 per cent, of water by weight. Mix thoroughly for 5 minutes.
On glass plates make two cakes about 3 inches in diameter, % inch thick at the
middle and drawn to thin edges, and cover them with a damp cloth. At the end
of the minimum time specified for initial set, apply needle JL- ^ ncn diameter,
weighted to l / pound. If an indentation is made, the cement passes the require-
ment for initial setting. Otherwise the setting is too rapid. At the end of the
maximum time specified for final set, apply the needle -J^- inch diameter, loaded
to one pound. If no indentation is made, the cement passes the requirement for
final set. Otherwise the setting is too slow.

Generally speaking, both periods of set are lengthened by increase of moisture,
and shortened by increase of temperature.

*By Portland cement, in this report, is meant the product obtained by
calcining intimate mixtures, either natural or artificial, of argillaceous and
calcareous substances, up to incipient fusion. By Xatural cement is meant
one made by calcining natural rock at a heat below incipient fusion, and grind-
ing the product to powder. By Puzzolan is meant the product obtained by
grinding slag and slaked lime, without subsequent calcination.



Recommendations of Board of U. S. A. Engineer

Officers. Continued.

In gaging Portland cement in damp weather, thesamples should be thoroughly
dried before adding water. This precaution is not deemed necessary with
Natural cement. Sufficient uniformity of temperature will result if the testing
room be comfortably warmed in winter, and if the specimens be kept out of the
sun in a cool room in summer, and under a damp cloth until set. Temperatures
may vary between 60 and 80 F., without atf'ecting results more than the
probable error in the observation.

Boiling test. Place the two cakes under a damp cloth for 24 hours. Place
one of them, still attached to its plate, in water 28 days ; immerse the other in
water at about 70 F., and let it be in a rack above the bottom of the receptacle;
heat the water gradually to the boiling point, maintain the heat for 6 hours ana
then let cool. The boiled cake should not warp or become detached from the
plate, or show expansion cracks. If the cold-water cake shows evidences only
of swelling, the cement may be used in ordinary work in air or fresh water for
lean mixtures, but if distortion or expansion cracks appear in it, the cement
should be rejected.

Accelerated tests are not generally recommended, but where a test must
be made in a short time, the boiling test is considered about the best. It not
only gives short-time indications, but at once directs attention to the presence
of ingredients which might lead to disintegration. On the other hand, it may
lead to the rejection of a cement which would behave satisfactorily in actual
work and which would stand the test after air-slaking. Sulphate of lime, while
enabling cements to pass the boiling tests, introduces an element of danger.

(5) Tensile tests are preferred to flexural or compressive tests. Sand
tests are the more important and should always be made; and neat tests should
be made if time permits.

A cement which tests moderately high at 7 days, and shows a substantial
increase in strength in 28 days, is more likely to reach the maximum strength
slowly and retain it indefinitely with a low modulus of elasticity, than a cement
which tests abnormally high at 7 days with little or no increase at 28 days.

Use briquettes of the form recommended by the American Society of Civil
Engineers,* measuring 1 inch square in cross-section at place of rupture, and
held by close-fitting metal clips, without rubber or other yielding contacts. The
tests should be made immediately after taking the briquettes from the water.

Neat tensile tests. Use unsifted cements. For Portland cement, use
20; for Natural, 30; and for Puzzolan, 18 per cent, water by weight. Place the
cement on a smooth non-absorbent slab ; in the middle make a crater sufficient to
hold the water; add nearly all the water at once, the remainder as needed ; mix
thoroughly by turning with the trowel, and vigorously rub or work the cement
for 5 minutes.

Place the briquette mold on a glass or slate slab. Fill the mold with consecu-
tive layers of cement, each to be % i ncn thick when rammed. Give each layer
30 taps with a soft brass or copper rammer weighing 1 pound, having a face %
inch diameter or 0.7 inch square, and falling about % inch.

After filling the mold and ramming the last layer, strike smooth with a trowel,
tap mold lightly on side, to free cement from plate, remove the plate, and leave
for 24 hours, covered with a damp cloth. Then remove the briquette from the
mold and immerse it in fresh water, which should be renewed either continu-
ously or twice in each week during the specified time.

Tensile tests with sand. For Portland and Puzzolan cements, use 1
part cement to 3 parts sand ; for Natural or Rosendale, 1 to 1. Use crushed
quartz sand, passing a No. 20 standard sieve, and being retained on a No. 30
standard sieve.

After weighing carefully, mix dry the cement and sand until the mixture is
uniform, add the water as in neat mixtures, and mix for 5 minutes. The con-
stituents should be well rubbed together.

For maximum strength in tested briquettes, Portland cements require
water = 11 to 12% per cent, by weight of constituent cement; Natural, 15 to
17 ; and Puzzolan, 9 to 10.

A machine which applies the stress automatically and at a uniform rate

' * See page 944.

CEMENT. . 939

Recommendations of Board of U. &. A. Engineer
Officers. Continued.

of increase is preferable to one controlled entirely by hand. The stress
should be increased at the rate of about 400 tbs. per minute. A rate materially
greater or less than this will give different results.

The highest tensile strength from each set of briquettes made at any one time
is to be considered the governing test.

Field tests are recommended, whether or not the more elaborate tests
above described have been made. In connection with tests of weight and fine-
ness, and observations of texture and hardness in the work, field tests often
suffice for well-known brands, showing whether the cement is genuine and
whether it is reasonably sound and active. Pats and balls of neat cement from
the storehouse, and of mortar from the mixing platform or machine, should be
frequently made. Estimate roughly the setting and hardening qualities by
pressure of the thumb-nail ; hardness of set and strength by breaking with the
hand and by dropping upon a hard surface. The boiling test may also be used.
Should the simple tests give unsatisfactory or suspicious results, then a full series
of tests should be carefully made.

A cement may be rejected if it fails to meet any of the following requirements,


Portland. Natural. Puzzolan.
Slow. Quick.
Fineness. Percentage to pass through a No.

Specific gravity. Betweei

, 3.10
. 3 25




Time of setting. Initial,

not less than
nor more than





Final not less than

45 m

nor more than

10 h.

2.5 h.


10 h.

Tensile strength, neat,

f 7 days f

. 450




Tbs. per sq. in.


. 540




Tensile strength. With

sand, as in (5).

Ibs. per sq. in.

f 7 days f

. 140
. 220




*92 per cent, is quite commonly attained by high-grade American Portlands,
but rarely by imported brands. For the latter, use 87.
f Reject any cement not showing an increase at 28 days over 7 days.



American Society for Testing Materials.

Digest of Specification adopted by the Society, Nov 14, 1904.
See Amendments of 1908.*

Adopted by Assn of Am Portland Cement Mfrs, June 10,
1904,* and by Am Ry Eiig & Maiiit of Way Assn, Mar 21, 1905.*

1. Packages. Brand and mfr's name plainly marked thereon. Bag
to contain 94 Ibs net. Bbl Portland = 4 bags; nat, 3 bags.

2. Tests in accordance with recommendations of Comm of A S C E, p
942. "Cem, failing to meet the 7-day requirements, may be held awaiting
the results of the 28-day tests before rejection."

3. Qualities. Natural Portland

Sp gr, cem thoroly dried at 100 C.* min 2.8 * min 3.1

Loss of wt, on ignition . ... *

Fineness. Percentage, by wt:

Residue on No. 100 sieve max 10 max 8

on No. 200 sieve max 30 max 25

Time of setting, mins, initial min 10 min 30


min 30 f min 60

\ max 180 \ max 600

Tensile strgth,

Min requirements,* Ibs per sq inch; briquettes 1 inch square section.
Briquettes must show no retrogression in strgth during specified


1 day in moist air in all cases.
Neat Natural Portland

24 hours 50 to 100 150 to 200

7 days 100 to 200 450 to 550

28 days 200 to 300 550 to 650

1 part cem, 3 parts standard sand.

7 days 25 to 75 150 to 200

28 days 75 to 150 200 to 300

Soundness (constancy of volume)

(For normal and accelerated tests, see

digest of A S C E Specfns, p 945) to stand to stand

normal test. normal and


Anhydrous sulfuric acid max 1.75%

Magnesia max 4.00%

Engineering Standards Committee of Great Britain,

Adopted Nov. 23, 1904.

1. Consignments of from 100 to 250 tons to have expert testing and
chemical analysis. For consignments of less than 100 tons, makers shall, if
required, give certificate, for each delivery, that cem meets this spec'n.

2. Samples. Test samples to be taken as soon as bulked at factory
or on the work, at consumer's option. Samples to be taken from each
"parcel," each sample consisting of cem from at least 12 diff positions in
same "heap," mixed together and spread out, 3 ins deep, for 24 hours, at a
temp between 58 and 64 F.

* Amendments adopted by Am Soc for Testing Materials, Sep 1908:

Strength. The means of the values given shall be taken as the
required minima where these are not specified.

Natural Cement. Omit specification for specific gravity.

Portland Cement. Specific gravity. For "thoroly dried at 100 C,"
read "ignited at a low red heat."

Loss of weight, on ignition, > 4 %.



Requirements. Engineering Standards Committee of

Great Britain. Continued.
3. Fineness.

per lin inch per sq inch
76 5,776

180 32,400

Wire woven, not twilled.


diani, ins



Residue not

to exceed


22.5 %

4. Tensile strength.

Test room temperature, 58 to 64 F.

Water, fresh, renewed every 7 days. Temp 58 to 64 F.

Paste, smooth, easily worked, that will leave the trowel cleanly in a com-
pact mass.

Briquette, filled, not rammed, into mold resting upon an iron plate, and
left until cem has set. Briquette kept in damp atmosphere 24 hours; then
in water until broken. Clips. See Fig. 1.





Fig 1. Briquet and Clips.

Load, start at zero. Add 100 Ibs each 12 seconds.

Neat test. 6 briquettes at 7 days, and 6 at 28 days. Av of the six ac-
cepted as the tensile strgth of the cement. 7 days, < 400 Ibs per sq. inch;
28 days, < 500.

When 7 day test is betw

400 and 450 Ibs per sq. in ..................... 25 per cent.

450 and 500 ........ .................... 20

500 and 550 ..... ' " .................... 15

550 and over ...... " .................... 10 "

Test with sand. By wt, 1 cem, 3 standard sand from Leighton Buzzard,
thoroly washed and dried. Sand must pass No. 20 sieve of 0.0164 inch
wire, and remain on No. 30 sieve of 0.0108 inch wire. Mixture thoroly
wetted, but without superfluous water. 7 days, 120 Ibs per sq inch; 28
days, 225. Increase, from 7 to 28 days, not less than 20 %.


Requirements. Engineering Standards Committee el
Great Britain. Continued.

Time, mins
. < in.,. maximum minimum

Quick 30 10

Medium 120 30

Slow 300 120

"Set" has occurred when needle, loaded with 2J^ Ibs, with flat end Vic
inch square, fails to make an impression.

6. Soundness. LeChatelier test. Expansion not to exceed 12 mm
after 24 hours aeration; 6 mm after 7 days.

7. Specific gravity. Not less than 3.15, when sampled and hermeti-
cally sealed at makers'. Not less than 3.10, when sampled after delivery to

8. Analysis.

Water, > 2 %, whether added or naturally absorbed from the air.

Calcium sulfate, > 2 % of wt of cem, calculated as anhydrous calcium

Lime, > enough to saturate the silica and alumina.

Insoluble residue, > 1.5 %. Magnesia, > 3 %. Sulfuric an-
hydride, > 2.5 %

American Society of Civil Engineers.

Digest of report of Committee on Uniform Tests of Cement,* Jan '03, as
amended Jan '04 and Jan '08.

1. Selection of samples left to discretion of engineer. Number of
samples and quantity to be taken from each package depend upon impor-
tance of work, upon number of tests to be made and upon facilities for
making them. Where conditions permit, sample one bbl in ten. Individual
samples may be mixed, and av tested; but, where time permits, test sepa-

2. Barreled cement to be sampled through a hole made in the center
of a stave, midway between the heads, or in the head. Bagged cement to
be sampled from surface to center.

3. Samples to be coarsely screened thru a No. 20 sieve.

4. Chemical analysis may show adulteration in the case of cems
rich in inert material, but is not conclusive evidence of quality. Committee
recommends method proposed by Committee on Uniformity &c., New York
Section of the Society for Chemical Industry, see E N, '03, Jul 16, p 60;
ER, '03, Julll,p49.

5. Specific gravity test. Le Chatelier's method recommended. Figl.
Flask, D, 120 cubic centimeters (cc); neck about 9 mm diam and 20 cm

long, with bulb, C; vol, betw marks, F and E, 20 cc. Neck graduated, to 0.1
cc, above F. Neck of funnel, B, enters neck of flask, and extends to top of
bulb, C. Use benzine (62 Baume naphtha) or kerosene free from water.
During the operation, in order to avoid variations in the temperature of
this liquid, the flask is kept immersed in water, in a jar. Two methods, viz:

(a) Flask filled to lower mark, E. Weigh out 64 grams (2.25 oz) of the
cem powder, cooled to temp of liquid. Thru the funnel, B, introduce the
cem powder gradually until surf of liquid reaches the upper mark, F. Then
64 grams, minus wt of powder remaining unused, = wt, w, which has dis-
placed 20 cc and

Specific gravity = w / 20.

(b) Fill, with liquid, to lower mark, E, as before. Add the entire 64
grams cem powder, liquid rising to some division of the graduated neck.

*Geo. S Webster, Richard L. Humphrey, Geo. F. Swain. Alfred Noble,
Louis C. Sabin, Spencer B. Newberry, Clifford Richardson, F. H. Lewis,
W. B. W. Howe. A S C E, Proceedings, Jan '03, Feb '04, Feb '08.



Tests. Am Soc Civ Eng-rs. Continued.

The reading of this division, plus 20 cc, is the vol, v, displaced by 64 grams
of the powder; and

Specific gravity = 64 /v.

6. Fineness. Sieves should be circular about 20 cm (7.87 ins) diam,
6 cm (2.36 ins) high, with pan 5 cm (1.97 ins) deep, and a cover.

Sieves should be of wire cloth,

No. 100, 96 to 100 meshes per lineal inch; wire 0.0045 inch diam.

No. 200, 188 to 200 " " " " " 0.0024 "

Use 50 grams (1.76 oz) or 100 grams, cem; dried at 100 C (212 F).
Hand sieving preferred. Use No. 200 sieve until one minute continuous
sieving, at about 200 strokes per minute, passes not more than 0.1 %. Weigh
residue, and treat it similarly on No. 100 sieve. A small quantity of large
steel shot, placed in the sieve, expedites the work. The results should be
reported to the nearest 0.1 %.

Sp grav Flask.

Fig 2.

Vicat Needle Apparatus.

7. Xormal consistency. The percentage of water, used in making
the pastes, for tests of strgth, soundness and setting, vitally affects the
results. Normal consistency is determined as follows :

The quantity of cem, to be subsequently used for each batch in making
the briquettes, but not less than 500 grams, is kneaded into a paste as under
"Mixing," 1 12, quickly formed into a ball, with the hands, and tossed
six times from hand to hand, held 6 ins apart. The ball is then pressed
thru the larger opening of the Vicat needle apparatus into the rubber ring,
7 cm (2.76 ins) diam, 4 cm (1.57 ins) deep, smoothed off below, and placed
on the glass plate. Its upper surf is then smoothed off with a trowel.
The point of the Vicat needle is then brought into contact with the upper
surf of the sample, and the cyl is allowed to descend. The paste is of the
normal consistency when the needle penetrates to a depth of 1 cm (0.39 in).
With this rather wet paste, the committee believes that variations, in the
amount of compression to which the briquette is subjected in molding, are
likely to be less than with a drier paste.

8. Setting. Vicat needle, 1 mm (0.039 in.) diam, loaded to 300 grams
(10.58 oz). Setting has begun when needle ceases to pass a point 5 mm
(0.20 in.) above the upper surface of the glass plate; and has terminated
when the needle does not visibly penetrate the mass. Test pieces to be
kept damp, during test, by being stored in a moist box or closet, or placed
on a rack over water in a pan and covered by a damp cloth, the cloth resting
upon a wire screen, so as not to touch the test pieces. Keep needle clean;

/^" "->v





Tests. Am Soc Civ Engrs. Continued.

as cem, adhering, seriously vitiates results. Time of setting is materially
affected by temp of mixing water, by temp and humidity of air, by the per-
centage of water used, and by the amount of molding the paste receives.

9. Standard sand. Crushed quartz objectionable, "especially on ac-
count of its high percentage of voids, the difficulty of compacting in the
molds, and its lack of uniformity." Comm recommends natural sand from
Ottawa, 111. Sand to pass a No. 20 sieve, with wire diam = half the diam of
spaces betw wires; < 99 % to be retained on a similar No. 30 sieve after 1
minute of continuous sifting of a 500 gram sample. The Sandusky Portland
Cement Co., Sandusky, O.. has agreed to furnish such a sand at actual cost
of preparation.

10. Standard briquette. See Fig. 3.


Fig 3. Briquet.

Fig 5. Clip.

Fig 4. Gang Mold.

11. Molds, "of brass, bronze or some equally non-corrodible material;"
sides strong enough to resist spreading. Gang mold, Fig 4, recommended,
because the greater quantity of mortar, required for it, conduces to uniform-
ity of results. Molds to be "wiped with an oily cloth before using."

12. Mixing. Proportions stated by wt; quantity of water stated as
percentage of dry material.

Metric system recommended.

Temp of room and mixing water as near 21 C (70 F) as practicable.

Sand and cem thoroly rm::e;i dry. Mixing done on some non-absorbing
surf, preferably plate glass. If an absorbing surf is used, it should first be
thoroly dampened.

Quantity of material, mixed at one time, depends on number of test
pieces to be made; about 1000 grams (35.28 oz.) convenient to mix, espe-
cially by hand methods.

Hand mixing and hand molding recommended. Material weighed, and
placed on mixing table, and a crater formed in the center, into which the
proper percentage of clean water is poured; material on outer edge turned
into crater by aid of a trowel. As soon as the water is absorbed, the opera-
tion is completed by vigorously kneading with the hands for an additional
1 H minutes. A sand-glass affords a convenient guide for the time of knead-
ing. The hands should be protected by gloves, preferably of rubber.

Molds filled immediately after the mixing is completed, material pressed
in firmly with the fingers and smoothed off with a trowel, without mechani-
cal ramming; material heaped up on the upper surface of the mold. In

1 2 4 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Online LibraryJohn C. (John Cresson) TrautwineConcrete → online text (page 4 of 23)