LABORATORY INSTRUCTION MANUAL FOR ENG 3731: MATERIALS (AND METHODS) OF CONSTRUCTION

lifegunbarrelcityΠολεοδομικά Έργα

26 Νοε 2013 (πριν από 3 χρόνια και 6 μήνες)

60 εμφανίσεις












LABORATORY INSTRUCTION MANUAL FOR

ENG 3731: MATERIALS (AND METHODS
)

OF

CONSTRUCTION



PROCEDURES





Text compiled by Mr. Md. Rabiul Alam

(Laboratory Assistant)













1

















Laboratory # 1


“Tests on Aggregates”


(a)

Fineness Modulus o
f Fine Aggregate”


(b) Fineness Modulus of Coarse Aggregates”




















2

Laboratory I
: Tests on Coarse and Fine Aggregates


Purpose:

T
o obtain the fineness modulus
,

density
, and voids ratio

of fine and coarse
aggregate samples


Materials and Apparatus:



-

sa
mple of fine aggregate



-

sample of coarse aggregates



-

digital weighing scale



-

sieve sifter for fine aggregates

-

sieve sifter for coarse aggregates

-

various cleaning brushes (point and wire)


Procedure:


Part 1: Sieve Analysis of fine aggregate


S
tep 1:

Take 500g sample of fine aggregate

(as per CSA code provisions, the
aggregates
must be completely dry)
. This is determined by weighing the
material on a digital scale.

Also weigh each sieve of the mechanical
sift
er, and the pan, and record the weights.


Step 2:

Place the aggregate in the
top sieve of the well
-
cleaned
mechanical sifter
(sieves used are # 4, # 8, # 16, # 30, # 50 & # 100). This appa
ratus is used
for shaking the aggregate
s

(similar to
the
principle used in

a pain
t
-
mixing
machine) and sieving them
.

The mechanical

sifter

has a bottom pan (to
receive
the material passing # 100 sieve) and a lid t
o close

the sifter du
ring
the test. After p
l
a
cing the lid on the sifter, agitate the sifte
r for about 1
0
minutes.



Step 3:

Determine

the
weight of
aggregate
s

that are

retained i
n each of the

sieve
s
,

by weighing each of t
he sieves (along with the retained aggregates
)
,

and
subtracting
the weight of each sieve
.

Also

record all the weights of
aggregates retaine
d in
each of
the
sieves
. To ensure that all materials are
collected, clean each sieve carefully u
sing the proper type of brush. Use
the

paint

brush for the finer sieves,
the

copper brush for intermediate
sieves and
th
e
steel wire brush for the coarse sieves.

Als
o verify whether
the sum of wei
ghts of

a
ggregates
,

retained in all the sieves
,

and the
bottom

pan is equal to the initial weight of
the aggregates taken.


Step 4:

Tabulate the data and determine the percent retained in each sieve. From
these values calculate th
e (cumulative) percentage of material that would
have been retained in the sieve if the whole volume of material was to be
sifted in that sieve alone. Then add the percentage of material retained in
all the sieves and divide by 100 to get the fineness modu
lus. Also prepare

3

a column to determine the cumulative percentage passing through the
sieve to plot the fineness modulus curve

(as specified in CSA 23.1)
.


Step 5:

Plot a graph of percent passing by weight vs sieve size, with the limits
specified in CSA standard A23.1 for fine and c
oarse aggregate sizes
superimposed.


Note:


(1)
All sieves have to be cleaned prior to experiment
.


(2) See the sample calculations given at the

end of this secti
on.



PART 2: Sieve Analysis of Coarse Aggregate


Procedure:


Step 1
: Take 5000 grams of coarse aggregates by weighing the material in a
digital scale.

Weigh each of the
clean
sieve
, along with

the bottom pan,
and record their weights
.


Step 2
: Place

the aggregates in the mechanical sifter (sieve sizes used are 1 ½”,
1”, ¾”, ½”, 3/8”, & #4). This apparatus is used for shaking the material
(similar to the principle of
a
paint
-
mixing machine) and sieving it.


Step 3
: Determine

the aggregates that are retained

in each individua
l sieve, as
mentioned earlier in Part I,
and record the data. To ensure that all
materials
are collected, use the steel bru
sh to clean each sieve.


Step 4
: Tabulate the data and determine the percent retained, and the percentage
that would have be
en retained in each sieve, if that sieve alone was used
to sieve the whole volume. The fineness modulus is obtained by adding
the percentage of material retained in all the sieves and dividing it by
100.


Step 5
: Plot a graph of percent passing by weight v
s. sieve sizes.


Note:

All sieves have to be cleaned prior to experiment.


Part 3:
Bulk
Density

and Voids Ratio

of F
ine A
ggregates


Step 1
:
Take a bucket (of volume, say, 2.5 litres)

and weigh it
.
Place
the
aggregates in the
bucket (whose volume is already determined by
dividing the weight of the volume of water that is contained in the
completely fill
ed bucket by the weight density of water
. In addition
,

the
bucket must be properly dri
ed befo
re placing fine aggregate in it
) to fill

1/3
rd

its
capacity
each time. Rod each portion

25 times.
The steps are

4

repeated till the bucket in is
filled with fine aggregates.
Remove excess
aggregate using the tamping rod as a straight edge.


Step 2
: Determine the weight of
fine
aggregates by weighing the bucket full of
aggreg
ates, an
d subtracting the
weight

of
empty bucket.


Step 3
: Determine the
bulk
weight density of
fine
aggregates by dividing the
weight of the aggregates by the volume of the bucket.

Step 4
:

Also determine the volume of voids in the fi
ne aggregate, using the
following equation.



%

voids in fine aggregates

=


{
[
(S
p.

Gr.

of
fine aggregates
)
*

w

-


bulk

density
]/[

w
*(Sp. Gr.
o
f
fine aggr.)]
}
*100


Part 4:
Bulk

Density and

Voids
Ratio
in C
oarse Aggregates



Repeat
the procedures given in Part 3

for f
ine aggregates
, and obtain the bulk
density
and voids ratio for coar
se aggregates.



Part

5: Sample Ca
lculations for Fineness Modulus,
Bulk Density and Voids Ratio


1.

S
i
eve analysis of a 1000
-
g sampl
e of fine ag
gregates resulted in the following
data. Find

the fineness modulus.


Sieve size

# 4

# 8

# 16

# 30

# 50

# 100

Weight
retained (g)

26

130

240

252

210

138



Calculations:


Sieve #

Weight retained

(g)

Percent r
etaine
d

Percent coarser

Percent finer

4

26

2.6

2.6


8

130

13.0

15.6


16

240

24.0

39.6


30

252

25.2

64.8


50

210

21.0

85.8


100

138

13.8

99.6





Cumulative =
308




Fineness modulus = 308/100 = 3.08



5

2.

Fin
d the volume of voids in a 3 cubic

y
ar
ds

of coarse
aggregates, of bulk density
equal
to 102 lbs/(cubic feet). The specific gravity

of particles is 2.65.


% of Voids =
[
(SG*
W
-

B)
*100]/(SG*W)


Spec
ific gravity,
SG

= 2.65

Density of water,
W

= 62.4 lbs/(cu. ft.)

Bulk density,
B

= 102 lbs/(cu. ft.)


% of void
s = [
(
2.65*62.4
-

102
)*100
]
/(2.65*62.4)



= 38.3%


Volume of voids =
(
38.3/100
)*3 =
1.143 cu. yards





= 1.143*27 = 31 cu.ft.






6


Materials and Apparatus
(Laboratory # 1)
:



Figure # 1











Coarse aggregate


Fi
ne aggregate


Figure # 2











Digital weighing scale



Figure # 3

















Sieve for fine aggregates



Sieve for coarse aggregates









-----
kg


7

























Labroratory # 2


“Tests on Cement Mortar”


(a)

Normal Consistency of Cement Mortar


(b)

Setting Time of Cement Mortar


(c)

Compressive Strength of Cement Mortar Cubes

(Casting it earlier to test it later)






















8


Laboratory II
: Normal Consistency & Se
tting Time and the Determination of
Compressive Strength of Cement Mortar.


Purpose:

To prepare samples of cement mortar specimens for: (1) Determination of
normal consistency and setting time tests; and (2) compressive strength of
mortar cement test.


Mat
erials
:

-

Sand




-

Portland Cement




-

Water


Apparatus:

-


Vicat Apparatus



-


Digital weighing scale

-


Moulding tray for cubes

-


Beaker (for taking water)

-


Universal Testing Machine

-

Tamping rod

-

Mixing bowl

-

Trowel

-

A stop
-
watch


Procedure:


PART I: Normal
Consistency and Setting Time



Step 1:

Obtain 500g sample of Portland cement by weighing on the digital scale.


Step 2:

Obtain 145ml of water in a beaker, which gives a water/cement ratio of
0.290:1 (this ratio may vary between 0.25 to 0.31).


Step 3:

Comb
ine both the Portland cement and water in the mixing bowl. Wait
approximately 30 seconds for water absorption.



Step 4:

Start mixing using the mixing machine at a slow speed for 30 seconds.


Step 5:

Turn off the machine for 15 seconds and scrape down the
mortar from the
sides of the bowl using a scraping stick.


Step 6:

Turn on the mixing machine again, for approximately 60 seconds at
medium speed.


Step 7:

Within a minute, remove the mortar from bowl, press the mortar into the
conical ring and put it on t
he Vicat Apparatus, using a rectangular glass
(on bottom), and smooth the top with a trowel.



9

Step 8:

In the Vicat Apparatus release the large
-
sized plunger (10 mm) to
penetrate the sample. Find the depth of penetration. If the penetration of
the plunger i
s 10
±

1.0 mm, then the consistency is called the normal
consistency. Otherwise, decrease or increase the percentage of water and
mix the cement and water once again to carry out the above steps till the
normal consistency is obtained.


Step 9:


Once the normal consistency is determined, th
en use the same paste t
o
determine the initia
l
and final setting times.

In this case, the
1
-
mm
-
diameter Vicat needle is allowed to penetr
ate the setting c
ement paste.

Start the timer when this cement

paste was

ma
de.


Step 10:

When the 1
-
mm
-
diameter Vicat needle

penetrates to a d
istance of 25
mm, then the
time taken by the
cement

paste (from its mixing to the
present time)

is said t
o

be

its initial setting time.


Step 11:

When

the 1
-
mm
-
diameter Vicat needle

just pen
etra
tes the surface of the
setting c
ement pas
te

and stops, then th
e time taken from the time of mixing

of

the paste to this time is called t
he final setting time.



PART II: Compressi
ve Strength of Cement Mortar


Step 1:

Obtain 1375 g sample of sand, 500 g sample of Portland Cement by
weighing material on a digital scale (in this the ratio
of cement to sand is
1:2.75
.

T
his

ratio can
also
be 1:1.5
, 1:2
,
or
1:2.5, which can be tried by
other grou
ps

at the instructor’s discretion
.


Step 2:

Obtain 240 ml sample of water in a graduated cylindrical beaker, which is
equivalent to the specified water/cement ratio of 0.48:1 (the ratio can
vary from 0.45 to 0.60 and this ratio can be changed at the instr
uctor’s
discretion).


Step 3:

Combine the two dry samples (Portland cement and sand) in the mixing
bowl and mix for approximately 5 minutes.


Step 4:

Slowly add water to the mixing bowl and mix the contents until the
desired consistency is achieved.


Step
5:

Remove mixed contents from bowl and place it in even portions in five
cubes in the moulding tray, which has already been wiped on the inside
with the given oil.


Step 6:

Tamp all five cubes to ensure uniform filling of the moulds. Smooth the
top with tr
owel.



10

Step 7:
Cure the specimens for fourteen days in water so that it may attain the
desired design compressive strength.


Step 8:

The cubes are removed from water, area of loading face of cubes
determined, and then placed in between plates of Universal
Testing
Machine.


Step 9:

Apply load to the specimen faces that are in contact with the machine.



Step 10:

Apply load until failure, and note the load.



Step 11:

Compute and record compressive strength of cement mortar.



11


Laboratory II: Normal Consiste
ncy and Setting Time and the
Determination of Compressive Strength of Cement Mortar.


Apparatus:
















Digital Weighing Scale









Vicat Apparatus












Mold tray for cubes












Beaker










Cement










Mortar cubes













Universal Testing Machine

-------
kg

240

ml


12




















Laboratory # 3


“Tests on Concrete”


(a)

Compr
essive Strength of Concrete Cylinders

(Casting now and Testing later)


(b)

Slump of Concrete





















13



Laboratory III
:
Slump Test and Compressive Strength of Concrete Cylinders


Purpose:

To make and test concrete specimens for: 1) Slump test on co
ncrete; and 2)
Compressive strength tests on cylinders at 7 and 28 days.


Materials and Apparatus:

-

cylindrical moulds for concrete cylinders (3)

-

conical hollow cylinder for slump test

-

sampling and mixing pans

-

scales

-

concrete

-

scoops

-

tamping rod

-

Concrete Mix
er


Procedure:


Part
I: Making and Curing of Concrete Compression Cylinder Test Specimens.


Step 1:

Get the weight of material needed for making one concrete cylinder

(
150
mm diameter, and 300 mm high
)
, knowing that the average weight
density of concrete is 2300 kgf/m
3
(varies between 2,20
0


for small
sized aggregates
-

to 2,500 kgf/m
3



for very large
-
sized aggregates).


Step 2:

From the ratio of 4:2:1 (by weight) for coarse aggregate: sand: cement
calculate the material required knowing that we need enough concrete for
3 cylinders plus a
n extra 20% for wastage (the ratio of 4:2:1 can be varied
between 3:1 ½ :1 to 6:3:1, at the instructor’s discretion).

Verify your
calculations with th
e instructor

before proceeding with the experiment.


Step 3:

Determine
the
amount of water
,

given that the ratio for cement to water
was specified as 1:0.55 (this ratio can vary between 0.45 to 0
.60, and can
be varied at the instructor’s discretion).



Step 4:

Weigh amount of coarse aggregate,

sand, cement and water needed.


Step 5:

All dry materials are added to the mixer
, one after the other,

and
mixed

d
ry
. First the coarse aggregate is added to the
mixer and then the sand;


both
a
re mixed dry f
or one minute, before cement is added to the
mixer
.


Then
t
he cement is added and mi
xed th
oroughly.


Step 6:

Slowly add water to mixer and mix until the desire
d consistency is
obtained.



14

Step 7:

Coat the
inside of the
cylinders with
the specified mineral
oil. Place

the
concrete mixture in

the cylinder to

one
-
third of its volume,

and tamp

the

layer 25 times.

Then fill to

next

one
-
third and once again tamp the layer 25
t
imes. Finally fill
the remaining portion of t
he cy
lin
d
er

with concrete and
tamp it once again 25 times
.


Step 8:

Level
the
top of mould
with
a
trowel
,

and tap the sides to eliminat
e air
pockets


Step 9:

Set moulds aside
, and cure them in water, before testing them

for 7
and 28

d
ays strengths.


Step 10:
Before testing the cylinder
s

in the universal
testing machine for

their


c
ompressive strengths
, the cylinders must be capped with the
proper

c
appi
ng
m
aterial (to load the concr
ete cylinders
with a
uniform
ly

d
istributed

load
,
o
ver its surface
)
.




Part II: Slump Test


Step 1:

Make concrete for the test, as given in Part

I,
sufficient for fill
ing the
slump cone

(approximately 40

to 5
0 lbs concrete
).


Step 2
:

Carry out step 1 to step 6 as stated above, and fill the hollow conical metal
cone with concrete mixture in 1/3 portions making sure to rod 25 times
for each p
ortion and to tap sides.


Step 3
:

Lift metal cone slowly straight up, and measure the vertical distance by
which the to
p

surface of concrete settles
(or slumps)
down, to the nearest
10 mm with respect to the
height of the
slump cone
.



15

Laboratory III
: S
lump Test and Compressive Strength of Concrete Cylinders


Apparatus:















Hollow cylindrical mould


Hollow conical mould for slump test






















Scoop








Concrete Mixer

















Sampling and Mixing Pans














16


















Laboratory # 4


“Tests on Masonry”


(a) Modulus of Rupture of Brick

(only demonstration)


(b) Compressive Strength of Brick

(only demonstration)



(c) Compressive Strength of Concrete Masonry Block

(only demonstration)


















17


Laboratory IV:

Modulus of Rupture and Compressive Strength of a Brick and
Compressive Strength of a Concrete Masonry Block


Purpose:

To determine: i) the modulus of rupture: (ii) compressive strength of a brick;
and ii) compressive strength

of a concrete masonry block


Materials:

-

red clay bricks (2)

-

masonry block (1)


Apparatus:

Universal Testing Machine


Procedure:


Part I: Modulus of rupture of brick



Step 1
: Measure and record the dimensions and weight of the red clay brick.


Step 2:

Place the brick in the Universal Testing Machine in the three
-
point

Bending mode..


Step 3:

Place a steel plate on the top of brick and apply load to the upper surface
until rupture (bending fracture).



Step 4:

Observe and record maximum load. Also expla
in the reason for the



characteristic failure, observed during testing.


Part II: Compressive Strength of Masonry Block and Clay Brick



Step 1:

Measure and record the dimensions of masonry block (for both brick and



Masonry block).


Step 2:

Place masonr
y block in the University Testing Machine between sheets of
wafer board.



Step 3:

Apply load until failure occurs.



Step 4:

Observe and record maximum load.



Step 5:

Repeat s
teps 1
-
4 for the strength test
on

the
clay brick
.






18


Apparatus:























Brick placed in Universal Testing Machine, in the three
-
point bending mode


























Masonry block placed in Universal Testing Machine




Note: The arrows repr
esent the force applied by testing machine







19



















Laboratory # 5


“Tests on Wood and Wood Fasteners”


(a) Flexural Stress in Wood at Proportional Loads


(b) Modulus of Elasticity and Modulus of Rupture


(c) Compression Strength of Wood Para
llel to Grain



(d) Pullout Strengths of Steel Fasteners for Wood



















20





Laboratory V: Flexural and Compressive Strength Tests on Wood and Wood
Fasteners.


Purpose:

To determine (i) Flexural strength of wood at proportional limit; (ii) Modulus

of rupture (at failure of beam in flexure test); (iii) Modulus of elasticity of
wood (within the proportional limit range); (vi) Compressive strength of
wood at failure; and (v) Pullout strengths of metal fasteners for wood (various
types of nails).


Mate
rials and Apparatus:

(a)

Test samples of wood and metal fasteners

(b)

Universal testing machine

(c)

Deflecto
-
meter

(d)

Dial gauge

(e)

Measuring tape


Procedure:


I.

For parts (I), (ii) and (iii):

F
lexural

Strength Tests on wood



Step 1:

Measure dimensions of wood samples.


Step 2:

Position the wood sample on t
he bending supports of the Universal testing
machine, over a span length of 28 inches, and place the deflectometer,
below the center of the beam specimen.


Step 3:

Apply the load at the center of the specimen at equal intervals and at a
uniform rate (75 lb
s

per minute
).


Step 4:

The load and deflection values are recorded at the above intervals (remove
the deflectometer after the elastic limit of wood has been exceeded).


Step 5:

Continue to apply the load at equal intervals till the specimen fails in
flexure and reco
rd the maximum load and the type of failure in wood.


Step 6:

Plot the load vs. deflection curve and determine the slope of the diagram
(P/

), where P is any load within elastic limit and


is the corresponding
deflection (this gives the stiffness of the b
eam).


Step 7:

Calculate the flexure strength at the end of the elastic limit of P vs.


plot.


Step 8:

Calculate the modulus of elasticity of the wooden beam [y = PL
3
/(48 EI)].


21


Step 9:

Calculate the modulus of rupture of wood=
)
2
/(
)
(
2
max
bh
L
P
.

Step

10:

Tabulate the results and present in a report form.


II.

For part (iv)

-

Compressive Strength of Wood


Step 1:

Measure the dimensions of the wood sample.


Step 2:

Position the wood specimen under the cross
-
head of testing machine (at
the center).


Step 3:

Apply the load till the specime
n fails in compression.


Step 4:

Determine the compressive strength of wood as [Applied Load/(cross
-
sectional area)].


Step 5:

Present the results.


III.

For part (v)
: Pullout Strengths of Wood Fasteners


Step 1:

Insert the steel fasteners (nails) into wood.


Step 2:

Fix the wood onto the test mac
hine.


Step 3:

Fasten the nail to the top cross
-
head of the testing machine and apply
tensile load (to pull out the nail fastener).


Step 4:

Note the maximum load required for pulling out the nail specimen.


Step 5:

Present the results.


















22


La
boratory V
:

Flexural and Compressive Strength Tests on Wood and Wood
Fasteners
.