Professional Documents
Culture Documents
BUREAU OF STANDARDS
George K. Burgess, Director
No. 272
FIRE RESISTANCE
OF CONCRETE
COLUMNS
BY
W.
S.
A.
H.
HULL, Physicist
INGBERG, Physicist
Bureau of Standards
FEBRUARY
PRICE
$1.25
25
24, 1925
CENTS
WASHINGTON
GOVERNMENT PRINTING OFFICE
1325
FIRE RESISTANCE OF
By W. A.
CONCRETE COLUMNS
ABSTRACT
made of 62 columns under working
CONTENTS
Page
I.
II.
III.
Introduction
Scope
637
638
of the investigation
638
2.
639
Furnace
Loading equipment
3. Temperature measurements
IV. Making of columns
1
Mixture used
2. Method of mixing
639
640
1.
640
2.
642
3
4.
5.
642
643
643
643
Consistency
Placing the concrete
Placing of thermocouples
643
644
644
V. Method of testing
1.
2.
645
645
646
&35
636
[Vol. 18
Page
647
1.
647
648
1.
2.
3.
648
VII. Results of
648
re-
inforced
648
(a)
hooping
(a)
(6)
(c)
gravel
concrete
column,
with
654
654
655
655
656
656
657
658
658
crete
658
659
659
661
661
661
661
661
662
668
669
669
67
plaster
on metal
on metal
lath forms
(6)
653
protection of
cinder concrete
3.
653
656
(d) Pittsburgh
649
Column protected by
673
lath forms
forcement
XII. Columns with 2>^ inches of protective concrete, with and without
secondary reinforcement
XIII. Discussion of test results
Test results obtained with columns of different types
1
.
674
676
677
678
683
685
688
688
688
(6)
(c)
690
(a)
690
..
re Resistance of Concrete
hUberg]
Columns
637
Columns protected by
Page
691
692
692
concrete
(c)
(d)
(e)
692
4.
2.
3
4.
XV.
693
694
on
elastic properties
695
XIV. Summary
1.
693
697
697
697
698
699
Fire-resistance periods
.'
699
700
1.
Method
2.
701
3.
4.
705
of derivation
fires
703
in buildings
XVI. Appendix
706
I.
INTRODUCTION
and the
of building materials
fire
program
The study
of the fire
would
establish,
if
made from
the
184.
638
to a
number
of
[Vol. 18
and
aid.
1.
In the
first
included.
It
types
(a)
Columns
ment.
(e)
Columns
From a
(a)
Columns
(b)
no
Columns
and
spiral reinforcement.
(c)
ment.
2
struction.
130.
Properties of
Some
of
hZberg]
639
ones,
In
the plain concrete columns the outer \]/2 inches were regarded as
protective concrete
and the
columns calcu-
lated accordingly.
2.
It
all
loss of
program
(a) Trap rock.
(b)
Blast-furnace slag.
(c)
(d)
gravel.
(e)
111.
showed
640
[Vol. 18
its
safety under
Consequently,
it
fire
conditions
is
Columns cast
(6)
in
all
(1) Those
Those with metal
Columns protected by
(c)
as protective
(2)
(2)
lath.
without binder.
III.
sectional
TEST EQUIPMENT
given in Figure
is
1.
1.
FURNACE
essentially,
of
three fixed walls and one moveable wall which inclosed a furnace
Hull
Ingberg]
641
mm
w72e//7forced
3ea77 of
Ooncrere Ba/Tarr?
>esrn7//7//?g
/y-ame
3n
Scale
w
Fig.
i.
Section through
18
test furnace
642
[va. 18
For
firing the
burners.
combustion to be
taken out at the top for the adjustment of the temperature
distribution.
2.
LOADING EQUIPMENT
disks.
3.
TEMPERATURE MEASUREMENTS
re Resistance of Concrete
fn gberg]
Columns
643
MAKING OF COLUMNS
IV.
1.
MIXTURE USED
be
cubic foot.
moisture
free, so
be nearly
artificially dried to
The weight
of aggregate per
obtained by weighing.
2.
METHOD OF MIXING
3.
CONSISTENCY
The consistency of the concrete made throughout the investigawas that obtained by adding from 8 to 10 per cent of water,
by weight, to the dry batch. This gives, with 1:2:4 concrete,
tion
which recommend
desired to
make
it
for investigative
work
in
which
it
it is a fairly definite consistency which can be reproduced with a fair degree of accuracy in the ordinary course of
hand mixing it is not far from the ideal consistency for reinforced
concrete, which building practice seems likely to approach as the
importance of the use of the right amount of water in concrete
becomes more generally appreciated and practice is adjusted
variables
accordingly.
644
[Voi.is
PLACING OF CONCRETE
from a
other words,
number
fairly large
it
number
of
it
same operation.
5.
PLACING OF THERMOCOUPLES
and about
inch apart were drilled in each of two rods for each column.
For each
of
these thermo-
practical,
U^J
their suffering
645
some displacement
To prevent such
dis-
was being placed. The frame was left midway between top and
bottom of the column until the concrete had partly covered it,
after which it was pulled up as fast as the concrete was placed.
Before the concrete reached the top the frame was taken apart
and removed. Before the ends of the hooks were covered the
The strings through
strings were cut and the crossbar removed.
the base plate were cut when the column was removed from the
base plate.
After a considerable amount of data had been secured on the
temperature progress at several points in columns of different types
and from different aggregates it was considered unnecessary to
continue placing thermocouples at depths beyond that of the
steel.
It will be noted that in tables giving temperature data
columns with numbers above 71 show temperatures for the steel
only.
V.
METHOD OF TESTING
1.
In
all of
before the
fire tests
646
[Vol. is
2.
Time.
Fig.
2.
DURATION OF TEST
had been
It
Ynber
647
This procedure was decided on in the belief that the critical time for a column which is exposed to fire is while the fire is in
progress, since at that time its strength presumably reaches its
failure.
it
resist
made
Two
in triplicate.
four-hour
in
fire test.
Those columns which did not fail at the end of the fire test
under the maximum load of the furnace loading equipment
(600,000 pounds) were tested in compression after cooling in the
1 0,000,000-pound machine in the engineering testing laboratory
of the Bureau of Standards.
VI.
1.
Among
is
itself.
cement of a
single
The concrete
batches.
648
[Vol. 18
2.
It
Results
Table
15,
steel of
No
3.
large
as well
tested
and
also of
fail
at the
maximum
fire test.
nection with
fire
resistance
and
is
four-hour
fire test
fire
VII.
1.
CALLY REINFORCED
paratively
little
and
others failing under their working load before the end of the test.
Ywbergl
649
a change in
The
most
At some part
what looks
practically
4
no protection
Woolson, Proc.
Am.
3954 25
after 2 hours.
130.
6, p. 433;
211, British
650
TABLE
1.
[Vol. 18
Time of
Age
at
time of
test
failure
in fire
test
maximum
Stress at
Maximum temperature
load
under
work-
fire
at
end
of
test
ing load
Aggregate
0)
u
1
eg
CO
CO
"0
2
3
5
73
74
75
1
.1
__|\
Elgin
(111.)
gravel
and sand
...
\
I
West Winfield
(Pa.)
limestone and
Pittsburgh sand
I
Blast-furnace
slag
and
Pittsburgh
New
sand
\i
co
77
78
6
6
4
7
6
4
4
4
4
24
.9
47
85
86
87
4
4
17
18
7
8
20
48
49
te
fca
i 11
O
OS
<
3i
CO
fa
0)
Lbs./
Lbs./
Lbs./
in.2
in.'
in.
15
c.
950
1,050
945
45
c.
410
460
355
240
290
210
350
150
275
250
100
105
240
180
120
100
6,340
50
20
980
780
4,880
660
605
1,993
2,120
4
10
"si
"3
4
3
23
1,145
3
3
42
43
3
4
i
>co
8 -?
"3
CO
>>
3
3
32
00
990
990
3
3
37
40
985
1,000
4,440
5,240
480
4,770
5,320
520
560
520.
5,620
17
4
6,890
4
4
4
85
85
4,870
50
21
2,260
465
110
100
54
55
4
4
16
2,420
3,000
610
560
190
239
100
110
2,700
480
Columns
77
and
78
In comparing the test results for columns 73, 74, and 75 with
1, 2, 3, and 5 (Table 1), account should be taken
of the fact that columns 1 to 5 were made near the beginning of
the investigation and columns 73 to 75 were made near the end of
it,
and therefore approximately two and one-half years later.
In the latter part of the investigation columns were made for the
purpose of determining the effect of a greater thickness of prothose for columns
tective concrete
mining the
Hull
Ingberg]
umns with
6.si
made at about
which
had
been made
the same time, as well as with such columns
those of columns with the usual type of protection,
ear Her.
Co/l//77/7S.
Looy
0/7? fc(6c//7c/
Ic/.
Tra/> /?ocA
P//fs/>urp/?
Jona"
6coo
of 4- hour f/n?
3000
/aoo
7)me cf /c///i/rv
Fig.
3.
Col Na 2,
.^fazm
Results of
-3/?r
tests
3A/-3-7m
Vfir
3/?r4wm.
fesr-
'
Wm\
f/;r
'
V/rr
-vertically
?/?r
and
laterally
reinforced
come
off fell at
the end of
652
[Vol. 18
oil
more than
its
was that
this shell
by
its
it is
by the concrete.
The Pittsburgh gravel
entirely
is
made up
of a mixture of different
There
is
fire test
there
this
may be
considered as
failure of the
The
spalling
?wberg]
re Resistance of Concrete
Columns
653
Columns
columns
(42
pebbles.
of
hours.
gravel
be noted that while this aggregate is distinctly different miner alogically from the Pittsburgh
gravel, yet the effects of the fire test on the columns from the two
aggregates are very similar.
or disrupt in the
fire test.
It is to
654
(d)
(III.)
[Vol. 18
Gravel.Both
the
sand and the gravel used in these columns came from pits near
Elgin, 111.
The chief difference between this sand and gravel and
those used in the gravel concrete columns discussed up to this
point was in that both sand and gravel were highly calcareous
instead of highly
siliceous.
Differences in
fire
resistance
are
In the
fire tests
much
made with
this
of 90 to 95 per cent.
There was no spalling
YnberH\
655
fourths
concrete
inch.
aggregates after
(/)
Furthermore,
was common
this
to all the
loosening
of
the
protective
fire test.
As in the case
explanation
Columns
No
6.S6
[Vol. 18
2.
COLUMNS
18
WITHOUT HOOPING
The
in Table 2
and Figure
TABLE
2.
Round
Columns
[Outside diameter, 18 inches; thickness of concrete outside the steel, 1H inches. Reinforcement: 2 per
inch diameter; ties, yi inch diameter, 12-inch centers. Effective area of
cent vertical, 8 round rods,
concrete, 168.7 square inches; area of steel, 3.53 square inches; effective area of column, 172 square
Age
at time
of test
Maximum temperature
Stress at maxiload
mum
at
end
of fire
test
Col-
umn
Aggregate
number
With-
Months Days
out
fire test
1
I
West Winfield
(Pa.)
(a)
38
22
limestone and \
11
7
7
7
2
24
41
20
Pittsburgh gravel in load-bearing portion; bituminous cinders in protective concrete; Pittsburgh sand
throughout
19
11
29
Pittsburgh sand
Column
I
Q
10
7
7
9
6
was loaded
to failure.
At
depth
At end
of four-
of vertical
hour
fire test
rods
C.
Lbs./in. 2 Lbs./in. 2
0)
1,365
1,320
1,010
630
700
steel
of
and
column
center
C.
C.
310
270
310
210
130
280
320
130
4,660
1,336
872
2,150
2,180
580
580
240
230
100
100
1,900
530
200
110
4,290
Maximum
In the
column
Midway beAt
tween center
first
test of
evidence
Technologic Papers
Fig.
4.
and 50)
of the
Bureau
of Standards,
Vol.
18
after failure
Fig.
6.
11
failure
after
Hull
Ingberg]
657
somewhat
Round Cotvmns
Pf/tsburan Grave/
P/ffsburgb 6and
5 fee/: % vertical
Long Id. f1ixed Grave/
lorry Id.
Sand
W. Mnfte/d.fPa) Limestone
P/ttsJburyb
Jand
6000
^SOOO
V 9000
^JOOO
'Loads si sra/ned
Profecfed Jbu
ffe'concrete
(b)
(triz/e
nof or /hee.
Protected by
/ concrete
Protected Jbo
/^'concrete
In
had been
was
observed to start.
Cracks developed rather slowly. After two
hours of firing portions of the outer concrete were observed to be
up
form of slabs as the test progressed, but none of these slabs fell off.
Near the end of the test
the bulging had progressed to such extent that in one place the
portions of the shell appeared to be pushed out about 8 inches
from their original position.
bulging.
Portions broke
in the
658
[Voi.rs
Limestone.
columns in
The few cracks which appeared were so
the tests 21 and 22.
fine that they were indicated mainly by the darker color of the
surface of the concrete fringing them.
The protective concrete
remained intact during the fire test. The columns were loaded
After cooling it
to failure while hot at the end of the fire test.
was found that with the exception of that portion which came in
the region of the actual fracture of the column the protective
concrete was still solid on the column.
(d) Pittsburgh Gravel Concrete Column, With Protection of Cinder Concrete. Column 41 was made by wrapping
tar paper around the outside of the reinforcement, thereby dividing
the space within the form, so that one kind of concrete could be
placed inside and another kind outside the reinforcement. Pittsburgh gravel concrete was used in the load-bearing portion and
cinder concrete from bituminous cinders in the outer portion.
In the test of this column there were yellowish flames playing
about the column during a large part of the test, indicating that
gases were being distilled out of the cinders, tar paper, or both.
Cracking of the protective concrete was first observed after 50
minutes of firing. The cracking was somewhat extensive, but
not similar to that in the columns in which spalling took
place.
Cracks opened wider in the course of the test, one crack
appearing to be about 3 inches wide at the widest part. None
(c)
There was
little
(Pa.)
of the material
3.
COLUMNS
18
fell
off.
Fig.
8.
failure
after
Hull
Ingberg]
TABLE
3.
Plain
659
Concrete Columns
[Outside diameter, 18 inches; thickness of concrete considered as protective material, iy2 inches; reinforcement, none; effective area, 180 square inches; working load, 81,000 pounds=450 lbs./in.2]
Age
Maximum temperature
Stress at maxiload
at time
of test
mum
at
end
of fire test
Col-
Aggregate
umn
num-
With-
ber
out
Months Days
of four-
At 2%
inches
At 5%
inches
hour
from
from
fire
surface
sur-
C.
1,070
950
C.
At end
fire
test
12
14
15
Column
14 failed
of 3
10
23
10
test
Lbs./
Lbs./
in.2
in. 2
835
0)
"3,"656"
face
At
center
of col-
umn
450
240
C.
205
120
^tfcrn?
Sree/.
fi/<73f
Fig. j
4.
COLUMNS
ft/rn&ee ^/a<?
Co/c//77/7S
Z//er//ca/.
Trap
/2gc#
Csustea Z/Me&e/X'
'
|[
vertical reinforcement
and no hooping
25
8.
and 26 (Table
In columns
66o
[Vol. 18
TABLE
4.
Square
Columns
[16-inch square columns. Thickness of concrete outside the steel, 1% inch. Reinforcement: 2 per cent
vertical, 4 rods, 1 inch diameter; ties, Vi inch diameter, 12-inch centers.
Effective area of concrete, 156
square inches; area of steel, 3.14 square inches; effective area of column, 159 square inches; working load
92,000 pounds =578 lbs./in.*]
Maximum
Age
at time
of test
temperature
at
end
of fire
test
Col-
umn
Aggregate
num-
With-
ber
Months Days
out
fire
test
Lbs./
in.2
(
West Winfield
(Pa.) limestone
and Pittsburgh
sand
Pittsburgh sand_
26
28
7
8
24
4
44
45
4
4
8
24
29
30
32
8
8
7
14
27
25
51
52
53
4
4
4
26
56
57
4
4
25
At
depth
At
center
fire
of
vertical
column
test
rods
hour
"5,"690"
of
Lbs./
in.a
C.
1,005
830
C.
775
995
680
868
1,000
1,000
250
280
2,360
2,420
660
630
100
100
2,278
690
100
1,905
770
1,855
690
""5,"630"
160
165
"4~7(J6~
12
5
12
At end
of four-
"4,~480"
The appearance
was necessary to work the
pump
fact that it
Fngberg]
of
830
of failure at the
lbs. /in.
66
end of the
column 44
fire test,
In the
test
and continued
After 2 hours and 15 minutes of firing all the outer concrete was gone from the corners of the column with the exception
steadily.
(c)
(Pa.)
Limestone.
(e)
end
New
fire
column
test of this
failed
shock.
5.
MENT
The main results of the tests made are given in Table 5. There
was considerable difference in the behavior of columns 33 and 35
in the fire test.
In column 33 disturbance in the outer concrete
662
[Vol. 18
was indicated
Age
at time
of test
A
Y
Stress at maxiload
Maximum temperature
At end
At
way beAt
depth of tween center
mum
at
end
of fire test
Col-
umn
Aggregate
Mid-
number
Without
Months Days
fire
test
of four-
hour
vertical
fire test
rods
steel
of
and
column
center
Lbs./in.2 Lbs./in.2
Column
35 failed
33
35
36
at
end
7
10
7
of
4
5
6
C.
1,145
~~4,~840~
690
C.
560
C.
480
(0
two hours.
still
in place,
although some cracks opened wide, so that the spiral steel could be
seen through one of them in the latter part of the test. After
one and one-half hours of firing the shell of outer concrete was
observed to be bulging considerable in various places. After the
test
had been
fell off,
ways
Hull
"I
IngbergA
663
such protective material will stay in place during a fire, particuif applied to gravel concrete, which itself has a tendency to
The test data of Table 6 compared with the
spall and come off.
larly
data of Table
TABLE
show the
6.
results obtained
Columns
[Outside diameter of concrete, 18 inches; column diameter increased to 20 inches by plastering; thickness
Reinforcement: 2 per cent vertical, 8 round rods, yi inch diameter;
of protective material, iy
2 inches.
inch diameter, 2-inch pitch, 2 spacers. Effective area of concrete, 168.7 square inches;
1 per cent spiral,
area of vertical steel, 3.53 square inches; effective area of column, 172 square inches; working load, 822
!bs./in.2]
Stress at maxiload
Age
at time
of test
mum
Col-
At end
number
at
end
Tested
umn
Aggregate
Maximum
Months Days
coid
At
depth
of four-
after
hour
four-
fire
hour
of
vertical
test
fire
rods
test
Lbs./in.2 Lbs./in.2
14
16
West Winfield
(Pa.)
limestone
76
and
Pittsburgh sand
12
5
6
16
25
10
23
C.
temperature
of fire test
Midway beAt
tween center
steel
of
and
column
center
C.
C.
5,970
5,615
375
410
460
155
160
90
110
6,050
410
135
90
"37270~
1
Columns 4, 6, and 19 were given the additional protection of 1 inch of plaster composed of Portland
cement and sand with a small percentage of lime hydrate. The plaster covering was reinforced with
expanded metal.
2 Column
76 was plastered with a mixture of Portland cement and crushed bituminous cinders, increas-
light
iK to 2K inches.
thickness of approximately
if/
It
position
plaster
First coat:
664
[Vol. 18
would be practical and probably less expencoat of plaster and after this coat had
sufficiently
apply
hardened
the expanded metal or other reinforcing material directly over the first coat and then complete
the plastering. In case the main object was to safeguard against
has indicated that
sive to apply the
it
first
and one
test.
by
width
fairly
regular,
None
of the
had
fallen off
the
fire test
connection that
tions
made
when column
18,
concrete of column 4 may have cracked during the fire test, and
that the deformation which occurred when the load was increased
Columns
m^
J.
had no
(Table
the
At any
6)
fire
The
jn broken.
if
combination of
The record
4.
the
final
of the
rate,
insula.
it
665
column 6
it.
is
have
there
fire test
fire
The
visible.
column.
When
this
after cooling, it
was found
friable
throughout
its thickness,
so,
though
plaster
was
than
soft
The
plaster
on the half
of the
of the
furnace was soaked with water at intervals during the day after
After standing five days
between the condition of this
that which had not been watered
was no apparent
there
difference
On
3954 25
it
666
Standards
[Vol. 18
all of it
final
was made.
and that
of 17
and
its
2j4 inches
18, similar
columns
The ultimate
test
much
difference
exceed 400
C.
(752 F.)
is
in
fire
are
damage by fire.
Column 76 was one that had been made
safe
against
serious
structural
is
performance of a kind of plaster made from Portland cement and crushed bituminous cinders, the latter taking
ll
Y^b er
Where
667
plaster
fire
crushed to a suitable fineness to take the place of sand, had expansion characteristics entirely different from and apparently
much more favorable than those plasters made from cement and
Columns 76 and 69a were accordingly plastered with such
sand.
a mixture. The cinders were the refuse from bituminous coal
cinders,
as
Per cent
Volatile combustible
10.
Fixed carbon
28. 33
Ash
61. 13
Total
Two
100.
No
54
00
metal
this
plaster in place.
In the
fire test
inches wide and 4 feet high fell off after 10 minutes of firing.
At the end of 20 minutes a long and rather wide crack was showing in the plaster on one side of the column, and the entire surface
668
IX.
[Vol.
The
column to
TABLE
7.
it
in place
is
that
it
18 inches.
Columns
[Outside diameter, 18 inches. Protective material applied by plastering. Reinforcement: 2 per cent verinch diameter; 1 per cent spiral,^ inch diameter, 2-inch pitch, 2 spacers. Effectical^ round rods,
tive area of concrete, 168.7 square inches; area of vertical steel, 3.53 square inches; effective area of column,
172 square inches; working load, 822 lbs., in. -J
Age
at
time
of
test
Stress at maxi-
mum load
Col-
Tested
umn
Aggregate
number
Months Days
At end
cold
of four-
after
hour
fire test
four-
hour
fire test
Lbs./in.2 Lbs./in.2
r
2
3
63
65
66
67
8
6
7
71
72
88
S89
9 90
69a
9
4
4
4
4
4
4
4
5
17
26
4
22
1
1
14
1,770
"i,~290~
"~4,~935~
~~5,"345"
"Z, 455"
1,280
3,120
1,885
2,000
3,200
Maximum
at
end
At
depth
of vertical
rods
C.
590
440
765
185
570
780
510
685
675
520
temperature
test
of fire
Midway beAt
tween center
steel
of
and
column
center
C.
185
220
335
95
250
245
C.
90
100
150
95
110
175
1 Column
63 was made 16 inches in diameter and plastered with a mixture of Portland cement, lime
hydrate, and sand, increasing its diameter to 18 inches.
2 Column
65 was made 16 inches in diameter and plastered with a special plaster containing Portland
cement, asbestos, and sand, increasing its diameter to 18 inches. This column did not fail at the end of
the four-hour fire test under the maximum furnace load, 3,480 lbs./in. 2
3 Column 66 was cast in
a form made by covering the spiral reinforcement with metal lath and plastering on the metal lath with cement plaster.
4 Column
67 was made by pouring concrete in a form made of metal lath wrapped around the reinforcing
Poultry netting used for
steel, column then pla'stered with a mixture of gypsum, lime, and kieselguhr.
binder in plaster. Total thickness of protective material, 1% inches; thickness of plaster, approximately
1 inch. This column did not fail at the end of the four-hour fire test under the maximum load of the furnace
loading equipment, 3,480 lbs./in. 2
6 Column
71 was made 16 inches in diameter and plastered with a mixture of cement, lime hydrate, and
sand, increasing its diameter to 18 inches.
6 Column
72 was cast in a form made by covering the spiral reinforcement with metal lath. Protective material added by plastering with a mixture of Portland cement, lime hydrate, and sand.
7 Column
88 was a duplicate of 65.
8 Columns
Pro89 and 90 were cast in a form made by covering the spiral reinforcement with metal lath.
tective material added by plastering with a mixture of lime hydrate, gypsum, and sand. Total thickness
of protective material, i>2 inches. Thickness of plaster approximately 1 inch.
No metallic binder in
.
plaster.
8
Column
fire test in
69a
was
originally cast in a
cinders.
re Resistance of Concrete
?ngberg]
1.
Columns
669
the column.
coat
is
Column 71 was plastered when the column was 14 days old and
was tested 107 days after it was plastered. The behavior in the
fire test was so nearly like that of column 63 that no separate
account of the test
2.
is
necessary.
670
Vol. is
of 88.
wide, extending
20
T/MB
FlG.
9.
considered important.
the four-hour
fire test
in both cases
^Urg]
671
COLUMNS CAST
IN
afterward.
In placing the metal lath to serve as a form over the steel reinforcement care was taken to place the lath right side out, or right
side up, so that the concrete could not escape as readily as if
the lath were placed the other way. In making column 67 the
consistency was kept sufficiently fluid so that the concrete would
spread out somewhat on the mixing floor instead of standing up
in a mound.
It was not as wet, however, as could be made in
hand mixing and not as wet as is commonly used in construction
work. The concrete placed in the lower half of the column was
well poled during the placing and filled out against the spiral and
around the spiral steel fairly well. A little mortar oozed out
through the metal lath and trickled down, but the loss of material
was not important. The concrete was not so well poled during
the pouring of the upper half, and in consequence did not fill out
so well around the spiral steel and against the lath.
In some
places large voids remained after the concrete was placed.
In
spite of these voids the
fire test.
top.
make by hand
practical to
in one batch
672
[Vol. x8
fill
Over a large
out completely
The coarse
steel.
aggregate banked up against the spiral and held the fine material
back.
Not enough concrete passed through the spaces in the
spiral to
fill
the small
amount
of space
Approximately a gallon of
lath.
bottom of the column, on the floor, during the operation. The
concrete was placed quickly in this column so as to determine
whether any considerable pressure would develop against the
metal lath form in the lower portion, due to the weight of the
The pouring was completed within approxiconcrete above.
mately 30 minutes. Practically all the pressure was exerted
against the spiral instead of against the metal lath.
In the
construction work.
would make
in the strength of a
column
is
somewhat
An
by column 67 when tested cold after the four-hour fire test shows
that this was a good column, imperfectly filled out as it was.
This was the only column made in the metal lath forms which
Hull
Ingberg
673
should prove to be useful a better covering than the ordinarymetal lath would be devised for the purpose.
case of
plaster
had
fallen
674
[Vol. 18
would obviously be
desirable in case columns or other members were to be plastered
for fire protection to use a plaster that would stay in place under
fire and also one having better insulating properties.
(b) Column Protected by Special Insulating Plaster on
M^tal Lath Forms. Column 67, like 72, was cast in a metal
lath form and plastered afterwards.
Some preliminary work had
been done to determine the properties of various combinations of
cement, kieselguhr, lime, and gypsum with the purpose of finding, if possible, a combination which would make a plaster of good
insulating efficiency and of such physical properties that it would
stay in place under fire conditions without any special binder or
reinforcement to hold it in place. One of the most promising
mixtures was selected for the protection of column 67, but on
account of a tendency to shrink when heated it was thought best
to make some extra provision against portions falling off to be
the ordinary gravel concrete protection.
It
by
consisted of 56
and hydrated
plaster
lime.
drawn
fairly tight
third coat of plaster concealed the poultry netting, the finish coat
At the end
of the first
be a
little
At the end
finish coat,
appeared
and
all
the
appeared to be in
Hull
I
Ingbergj
675
/fir
&?
4?
2,Jir
20
77M
Fig. 10.
No. 67
plaster
Just how this column would have fared without the poultry
netting to hold the outer plaster in place is, of course, problematical.
It may be that if a reasonable proportion of asbestos
were included in the mixture for the outer coat it would stay in
place without any metal mesh like the poultry netting, but this
was not tried. The outstanding feature of this test was the
final
676
[Voi.18
column was only i 2 inches, and that the protection was much
more effective than that of columns 4, 6, and 19, which totaled
2y2 inches in thickness. Even columns 68 and 69, with a total
thickness of 2 if inches protective material, 2-3^ inches of which
was gypsum, were not as well protected, as shown by the final
temperatures (Table 8), as was this column. One inch of this
special plaster gave better protection, judged by the temperatures
in the column at the end of the four-hour fire test, than 2^ inches
of gypsum (plaster of Paris) which is known to have exceptional
Y
It does
seem worth
while, however, to
1^
or 2 inches of protective
concrete and that the saving in size and weight would be considerable.
4.
number
it
in place.
As might be expected,
it
was
Fig. ii.
Views
hZberg]
re Resistance of Concrete
Columns
6jj
fire
lbs. /in.
test,
subjected to a four-hour
fire
test
fire test,
and
was
on column 76
(p.
667).
The
8 inches
and the
total
inch.
was used.
The
sort of material
cinders are probably superior for the given purpose to those used
678
[Voi.18
in these tests.
It
is
much more
favorable expansion
% i/erf/co/, /% sp/ro/
Crave/
dond
/''concrete
ProtecfeJ J>y:
dfcpfipsi//77 form
Fig.
Results of tests
12.
X.
On
also with
'/^concrete
Z&pyjpsi/m faro?
W?r
/"c'/rffpfifr
v^V
pfosfer
*g^
/'jfec/'a/ pfas/tr
COLUMNS CAST
IN
gypsum forms
GYPSUM FORMS
gypsum
by
21 inches long.
Hull
Ingberg]
TABLE
8.
Columns
Cast in
679
Gypsum Forms
[Outside diameter of concrete, 15y2 inches; thickness of form, 2^ inches; total thickness of protective
inch diameter; 1 per cent
material, 2f inches. Reinforcement: 2 per cent vertical, 8 round rods,
inch diameter, 2-inch pitch, 2 spacers. Effective area of concrete, 168.7 square inches; area of
spiral,
2
vertical steel, 3.53 square inches; effective area cf column, 172 square inches; working load, 822 ibs./in. ]
Age
at time
of test
Stress at maximum
load
Col-
number
at
end
Tested
umn
Aggregate
M aximum temperature
Months Days
At end
cold
At
With-
of four-
after
depth
out fire
hour
four-
fire
hour
of
vertical
test
fire
rods
test
test
160
61
68
69
70
4
4
4
{
2
1.205
2,800
"1,126"
2
3
C.
810
185
305
190
of fire test
Midway beAt
tween center
steel
of
and
column
center
C.
260
150
95
90
C.
90
110
85
90
6,480
1 No anchorage
was provided for the gypsum form on column 60. Form was anchored to concrete of
61 by means of wires. In the forms for columns 68, 69, and 70 light metal reinforcing material,
poultry netting, or expanded metal was used.
2 Column
69 did not fail under the maximum load of the furnace equipment, 3,480 lbs./in. 2 and was subsequently tested as column 69a with plaster protection.
column
The
tabular form.
vations
made
There was
in the
two
sufficient similarity
fire
680
[Vol.
visible,
mud
partially exposed.
fell off.
When
test,
is
such that
it
its
thickness.
made
in Table 8 were
in
gypsum forms
15 V&
The form
sections for
test,
by
like
first
as in the tests of 58
gypsum covering
fell off
after
with
its
When
filled
filled
with concrete.
forming a system of reinforced concrete pegs progypsum form. The gypsum covering cracked in
this fire test very much as in the tests of the columns in gypsum
forms already described.
In this case, however, the gypsum
covering supported by the concrete pegs stayed in place until the
fine concrete,
fnjberg]
had been
test
utes,
68
at which time
it
fell
off rapidly,
make the load test after the fire test was completed. Since the gypsum covering fell off in the test of column 61, 15 minutes before
the end of the test, the spiral steel on this column was undoubtedly
at a
fire test.
One
of the
3954 25
682
Where
[Vd.i8
At the end
that part of
For some reason the gypsum seemed firmer and in much better
condition after this test than after that of 68.
stripped of the
may
?t!2berg]
683
drawn, but the strength of this column suggests the thought that
the concrete may have developed greater strength because of the
supply of water stored in the gypsum form.
The method of casting concrete columns in precast forms is
not new. Columns have been cast in large construction work in
forms cast from cinder concrete. Although the forms were not
costly, the use of this method was found to add to the cost of the
If it should
construction, and it has not come into common use.
become common practice in course of time to employ special
safeguards against fire damage in structures built from the kinds
of concrete which now appear to require them, it may be that the
Cinder
use of precast forms would be practical in some cases.
concrete and other kinds of concrete which do not spall easily,
while not having heat retarding properties equal to those of the
gypsum, would have advantages in other respects and should be
given thorough consideration in case the use of precast forms is
contemplated.
XI.
and the age at the time of test were to be avoided, a comnew series of 18-inch columns with spiral as well as vertical
reinforcement, with Pittsburgh gravel aggregate, was made up,
part of the series (columns 73, 74, and 75, Table 1) having plain
protective concrete and 77 and 78 having secondary reinforcement
or binder in the form of expanded metal in the outer concrete.
The consistency of all these columns was made somewhat wetter
than that of the early ones in order that the concrete might flow
properly through the expanded metal and completely fill the form.
The test data for this series of columns are given in Table 1.
The behavior of the ones without the secondary reinforcement
crete,
plete
684
(73
and
[Vol. 18
(a),
with
made
in this
In the
way.
fire
test of
column 77 evidence
of shell failure
was
much
better test records than similar columns without the secondary reinforcement, but distinctly poorer records than all
columns of the same type that were made from nonspalling
aggregates, although none of the latter had any metallic binder in
Hull
Ingberg]
685
XII.
CRETE, WITH
better
TABLE
9.
Columns
With
2%
[Outside diameter, 20 inches; thickness of protective concrete, 2V2 inches. Reinforcement: 2 per cent
inch diameter, 2-inch pitch, 2 spacers. Effective
vertical, 8 round rods,
inch diameter; 1 per cent spiral,
area of concrete, 168.7 square inches; area of vertical steel, 3.53 square inches; effective area of column, 172
square inches; working load, 822 lbs. /in.-]
Age
at time
of test
Stress at maximum
load
Maxi-
mum
temCol-
At
umn
Aggregates
number
With-
Months Days
out
fire
test
Tested
end
cold
of
after
four-
four-
hour
hour
fire
fire
test
test
perature
at end
of fire
test at
depth
of vertical
179
180
181
2
82
83
4
4
4
4
4
84
1,495
1,640
23
11
C.
1,000
960
"5,"56"
5,115
4,950
rods
5,155
Columns Nos. 79, 80, and 81 had no metallic binder or reinforcement in the protective concrete.
Columns 82, 83, and 84 had a light wide mesh expanded metal in the protective concrete.
410
470
686
As
[Voi.18
apparent from the test results given in the table, the proby 2^ inches of concrete without secondary
reinforcement, while somewhat more effective than that of i}4
inches of the same kind of concrete without secondary reinforcement, was much less effective than that of 1 ]/2 inches of concrete
with secondary reinforcement. The behavior of column 79 and
that of 80 in the fire test were so nearly alike that the account of
is
tection afforded
the observations
Shell failure
made
was
The
34 minutes a slab about 2)4 feet long had fallen off. So far as
could be judged, the average thickness of this slab was rather
small, apparently about 1 inch. The protective concrete continued
The outer portion came first in rather
to break up and fall off.
thin slabs, but the action continued, the freshly exposed concrete
breaking up and falling off in its turn. After 2 hours and 13
minutes steel was exposed to an important extent. (In the test
of column 80 the steel was exposed over a considerable area after
1 hour and 49 minutes.)
After three hours it was observed that the portion of the protective concrete which was still in place was continuing to break up.
Fig. 13.
Views of specially protected columns (68, 80, and 82) after failure
when load was increased following the four-hour fire test
Hull
1
IngbergJ
687
/to
zo
T/ME
Furnace and column temperatures, 20-inch Pittsburgh gravel concrete column
No. 83, with expanded metal in the 2%-inch thickness of protective concrete
Fig. 14.
may
secondary reinforcement.
688
XIII.
1.
[Vol.
The
resistance of columns
same amount
of vertical reinforcement.
2
to 3,480 lbs./in. for these columns.
When
30 days after the fire test, these limestone gravel columns and
crushed limestone columns showed an average ultimate strength
approximately 80 per cent as high as that of similar columns that
had not been fire tested.
The apparent cause of the great difference between the fire
resistance of the columns made from spalling aggregates and those
made from nonspalling aggregates was the loss of the protective
The proconcrete by the columns with the spalling aggregates.
tective concrete commenced to crack early in the fire test and then
to break up into slabs and fall off, so that after approximately one
hour of firing the steel and load-bearing concrete would begin to
be exposed, and the exposed area would increase in extent so that
Hull
"J
Ingberg]
689
during the latter half of the test the column would have little or
no protection. The temperature data of Tables 1 and 10 show
the effect of this loss of protective covering on the temperature
progress in the steel and the load-bearing concrete.
TABLE
10.
Temperature
Temperatures
Age
time
of test
a
3
a
a
CO
03
3
o
CO
*o
at
depth of vertical
rod (centers of
rods 2& inches
from surface of
columns)
at
U A
H
s
O
CO
CO
u
3
o
3"
"
CM
CO
A
&
Temperatures
center of column
(5V& inches from
surface of column)
CO
S3
o
J3
"
Oi
CO
CO
3*
A
m
c. c. c. c. c. c. c.
Gravel concrete columns not
greatly affected by spalling.
8
9
77
78
6
6
2
3
Gravel
concrete
columns
4
4
4
4
90
35
30
100
85
130
105
215
180
10
10
50
50
285
370
390
375
410
465
490
490
485
530
570
570
260
210
370
370
430
( )
(48
\50
21
110
95
54
\55
4
4
7
16
135
125
330
310
470
440
590
530
/ 85
no
86
170
255
270
380
385
460
465
9
4
17
4
430
630
450
635
17
7
c. c. c. c.
110
100
90
80
80
7
7
<>C.
30
20
60
30
30
n<
30
17
18
21
22
CO
305 1390
320
445
207
340
300 ( 2 )
160 3 250
165 ^250
110
200
180
190
3
5
CV]
145
160
100
145
90
100
900
46
47
73
74
,3.
65
75
40
70
45
50
1,035
950
*960
980
'780
3
o
-1
1937
840
972
812
985
855
825
900
715
CO
3
o
50
70
682
805
620
735
410
470
700
425
4
7
9
CO
rt<
260
310
130
140
120
160
110
125
at
a.
175
200
346
275
H
3
o
100
120
5C0
515
495
425
CO
3
o
100
100
350
385
300
245
19
11
Temperatures
center of column
610
625
650
555
165
185
110
100
at
point
t
80
100
90
110
130
180
130 1230
140
290
100
210
150 ( 2 )
90 3 100
100 uoo
110
110
no
85
100
85
100
50
85
50
10
30
40
40
75
100
100
130
185
215
20
20
25
580
750
570
730
30
50
50
85
105
100
105
220
145
170
325
230
10
20
30
95
100
85
85
50
85
95
50
80
110
100
110
80
90
90
150
was plaster
Plaster of Portland cement and sand (considerable portions of
plaster fell off during
fire test in all cases)..
Plaster of
sand
Plaster of gypsum, lime
hydrate, and kieselguhr with wire binder.
3
4
5
6
63
66
72
125
120
230
255
430
275
480
80
cement and
crushed cinders
69a
115
280
425
520
80
100
145
230
25
95
95
115
230
290
335
405
420
505
55
100
130
185
35
85
100
100
110
150
22
65
85
130
180
35
55
75
95
20
40
75
95
26
88
67
Final temperatures for column i read at 3 hours and 40 minutes because of failure of column.
Final temperatures for column 43 read at 3 hours and 00 minutes because of failure of column.
Final temperatures for column 46 read at 3 hours and 30 minutes because of f ailure of column.
Final temperatures for column 47 read at 3 hours and 40 minutes because of failure of column.
Final temperatures for column 74 read at 3 hours and 20 minutes because of failure of column.
See note to Table 7.
It
the failure of the outer concrete of these columns was due to the
690
[Voi.18
type of aggregate and not with those made from another type
appears to be due to difference in expansion of different minerals
at high temperatures.
The fact that the loss of protective concrete was more rapid and more extensive in the case of hooped
columns than with either round or square columns without hooping may be due, in part, to the greater load carried by the hooped
columns in the fire test, but it seems probable that it was due in
larger measure to the opportunity for definite division or parting
interposed between the protective concrete and the concrete of
the column proper presented by the hooping. A summary of
temperatures attained in 1 8-inch round columns with and without
hooping is given in Table 10.
(b) Round Columns Without Hooping.
In columns of this
type from limestone aggregate the strength at the end of the fourhour fire test was approximately 50 per cent of that of a similar
column not fire tested, whereas the columns made with gravels
of spalling types
explained, in
of the protective concrete in the course of the fire test (Table 2).
(c)
spalling
nonspalling aggregates
rock.
Square
limestone,
of
In the
fire tests
of square
Hull
I
Ingberg]
691
TABLE
11.
Temperature
Age
at
time
of
test
Col-
umn
of
of
num-
of temperatures
two
vertical
rods
Temperature at center
of column
column)
ber
4
2
3
3
4
1
1
2
Months Days hour
hours hours hours hour hours hours hours
Gravel
concrete not
affected by spalling..
Gravel
concrete
affected
Blast-furnace-slag concrete
Trap-rock concrete. _
Tests
C.
C.
C.
C.
C.
C.
24
250
485
640
750
80
100
100
150
24
780
570
985 1,000
780
900
90
95
no
260
235
105
44
45
165
250
280
29
175
30
14
27
210
420
425
550
550
630
630
70
80
100
90
100
100
51
53
4
4
5
12
150
140
385
370
560
570
680
690
50
90
100
56
160
390
560
670
by /
spalling
Limestone concrete
C.
25
C.
greatly
25
cold weather.
2.
CAUSE OF SPALLING
Sci.
p. 1; 1914.
692
3.
many
[Vol. 18
aggregates,
localities it
and
it
now universally
provided.
and one from limestone, were plastered with 1 inch of cementlime-sand plaster. Light expanded metal was provided in the
plaster to insure its staying in place in the fire test.
As shown in
Table 6, this protection was so effective that there is no evidence
in the test results that the ultimate strength of these columns was
not as high after the fire test as before. It will be noted that the
maximum
steel
C.
Columns Protected by Plaster in the Place of Protective Concrete. The substitution of plaster for protective
(6)
is
given in
7.
fire test,
Two columns
plaster containing
plaster
Hull
1
Ingberg]
One column
693
of
(67)
lime,
TABLE
12.
Temperature
Cast in
Age
at time
of test
Column
number
Average of temperatures
of two vertical rods
(centers of rods 3M
inches from surface
of form)
to
to
>>
{d)
4
4
......
of
temperatures
Temperature at center
of column
u
s
S3
u
3
u
3
.a
Xi
.3
.3
.3
,3
*H
CM
CM
en
C.
68
69...
Average
45
45
c.
75
65
c.
c.
C.
100
85
255
175
25
25
C.
50
55
c.
70
65
c.
90
90
C.
25
20
CM
c.
40
40
.3
co
.3
c.
60
60
c.
85
90
rela-
On
694
[Vol. 18
TABLE
13.
Temperature
Average
Average
Age
at
time of
test
u
S
tures
cal rods
of rods
of
inches
Columns
greatly affected
Gravel
spalling:
C
S
A
o
CO
3
o
to
CN1
CO
vertical
of
umn
at center of
column
(6Vs inches
CO
3
o
Temperatures
rod
of col-
from surface
column)
CO
3
o
two
3
o
CO
3
o
cvi
CO
of
CO
CO
3
o
79
80
82
83
4
4
4
6
19
12
5
10
90
80
280
460
795
700
960
805
90
80
145
150
270
295
400
445
85
120
110
185
190
165
265
290
290
355
370
390
to
CO
3
o
Ao
A
CO
c. c. c. c. c. c. c. c. c.
by
concrete
CO
3
o
cf
and center
13
J3
"3
temper-
midway
between center
points
(centers
3&
from surface
column)
of
atures
c.
c.
c.
55
80
40
90
90
105
85
110
Columns
Columns originally
18 inches in
diameter with the added protection of 1 inch of Portland
16
23
20
55
25
85
105
80
105
115
90
140
155
130
50
10
90
metal
(e)
tective Concrete.
Secondary
with
(82, 83,
78)
gravel
and
columns with
The \]/2 inches of
in
84).
subjected to the
fire test.
re Resistance of Concrete
?i!gberg]
4.
Columns
695
The length, of time the columns withstood the fire tests under
working load and the additional load sustained at the end of the
fire test or subsequent to cooling give information on the loadcarrying capacity of the columns during the fire and subsequent
to cooling.
As a further aid in determining the feasibility of
reuse of columns in buildings after exposure to fire, information
on the effect of the fire on the stress-deformation ratio or modulus
of elasticity is desirable.
It is
and at
its
conclusion the
full
and load test in the hot condition is between two and three times
larger than with comparable columns on first loading without fire
This
test, the modulus of elasticity being correspondingly lower.
can be considered as due to the effect of the fire on the concrete,
particularly on that in the outer portion of the column and in
involved.
For the columns that failed in the fire test or in the loading at
the end of the fire test no deformation or strength data for the
cooled condition after fire test were obtainable. The effect of the
fire was without doubt more marked than for the columns that
withstood the fire and excess load in the heated condition, but it
appears doubtful that the resulting deformation characteristics of
the concrete would preclude reuse
strength were present.
would
if
The cold-drawn
696
[va.
18
4500
r-^5
mo
3300
3000
%
s?
2500
*
*
2000
H
/ k
A^'>""
k
^-^'H
^>
<4
/<
.^"'"
*A
1500
<,
f<
,-A
.>*
1000
yy^
500
JOOZ
Fig. 15.
.0004-
.0006
.0003
.0010
.OOIZ
.00/4
.00/0
.OOI5
.0020
.0012 .002*
0016
U/v/r
Fig. 16.
JX&S
MO
.0012
#0/4
JP0/6- .0013
.0020
.0022
oo2g.
4, 5,
and 6
DeFcaMAT/OM
Deformation curves for limestone concrete columns Nos. 18, ig, and 20
YngierA
697
XIV.
1.
SUMMARY
The
and in
made from
fires
certain types
in buildings
is
due to
damage by
fire.
395425
698
[Vol. 18
ment and
It has
exposure to
fire is
re Resistance of Concrete
Ynjberg]
aggregate
4.
699
which appears
say, below 500
C. (932 F.)
where necessary or desired. The problem of providing
reasonably low
feasible
Columns
is
is,
of course, simpler
when a
nonspalling
used.
and to adjacent
buildings.
ject to
concern
is
members
of a given type
almost
failure
all conditions.
The ability to come through fires without
under working load does not imply freedom from damage
The
PERIODS
which results
can be reduced and made available for use. It is
derived from the results of tests, with suitable margins taken for
variations in material, workmanship, design, and size of columns
fire-resistance period is a convenient unit to
of fire tests
700
[Vol. z8
may
be
applied but not obtaining in the relatively few test specimens and
With
safe
The
METHOD OF DERIVATION
the standard
fire
is
taken as two-thirds
the group.
if
subjection
the
to
(3,480 lbs./in.
2
)
maximum
while hot.
is
fire test is
load
The
of
six-hour classification
is
given
if
tions.
?ni erg]
re Resistance of Concrete
Columns
701
The periods derived from the results of this investigation accordmethod outlined above are given in Table 14. The
resulting periods cover a range from one and one-half to six hours
or over, depending mainly on the column size and type, concrete
aggregates used, and thickness and type of protection. It is
ing to the
seen that for columns not smaller than 16 inches in outside diam15 inches square a variation in safe-resistance period
eter or
six
and calcareous.
inches of concrete or
sum forms
by
inches thick.
casting the
column
in reinforced gyp-
702
TABLE
14.
Fire-Resistance
iva. 18
[Concrete mixture to be not leaner than 1:2:4 Portland cement, sand, and coarse aggregate, by volume.
Details of design and loading to conform with generally accepted standards]
Protection
Mini-
mum
Type
of
column
Shape
area of
solid
Fireresist-
Coarse aggregate
mate-
Thick- ance
period
ness
Description
rial
Plain concrete
Round..
Sq. in.
200
200
forced.
Do_--i.
Do...
Round .
Round or
Do
...do.....
Inches Hours
Sixty per cent or more
quartz, chert, or granite.
do
...
Plain concrete,
same
as
column.
do
VA
2V2
2V2
200
200
do
Trap rock or blast-furnace slag.
do
do
IH
VA
200
Limestone or calcareous
do
1H
square.
VA
110
do
do
m
m
VA
do...
IVi
VA
as
2V2
as colreinforced with
iy2
2Y2
2%
iy2
2y
do
ite.
Do
Do
...do
...do
200
200
Do
...do.
200
Do
...do
200
do
Concrete
umn
expanded metal
wire mesh.
ite.
...do
250
do
Plain concrete
Do
...do
250
do
column.
Concrete
do
expanded metal or
wire mesh.
One and one-half inches
umns
Do
...do
250
same
of p ain concrete
2A
4
or
Do
same
VA
same
as colum i
covered
with 1-inch Portland
Do
_..do
250
Limestone
or calcareous
do
Do
...do
250
One-half inch
of consame as column
covered with a 2 inch
thick unreinforced gypsum mold within
crete
which column
Do
...do
250
do
Same
that
is cast.
as above, except
gypsum mold
is
reinforced
mesh.
Do...
...do
200
do
One-half inch
of consame as column
covered with 1-inch
unreinforced Portland
cement or gypsum
plaster not leaner than
1 : lYi by volume with
crete
f^l ^]
TABLE
14.
Fire-Resistance Periods
of
703
Concrete Columns
Continued
Protection
Mini-
mum
Type
of
Shape
column
area of
solid
Fireresist-
Coarse aggregate
mate-
ance
Thickperiod
ness
Description
rial
Round or
Sq. in.
200
square.
Inches Hours
Sixty per cent or more
quartz, chert, or granite.
One-half inch
of conas column
covered with 1-inch
unreinforced Portland
same
crete
cement-cinder plaster,
1 3 by volume surface
:
column hacked or
column cast in metal
of
lath form.
Do
...do
200
do
One-half inch
of consame as column
covered with 1 inch of
unreinforced Portland
cement asbestos and
crete
sand
plaster of propor-
1 y% 2 by volume
surface of column
hacked or column cast
tion 1
Do
...do
200
do
or equivalent.
tions.
tections
iyi inches.
replacting
It
1
is
is
obtained by
Portland cement or gypsum plaster. Somewhat greater resistance is obtained with unreinforced cement-cinder plaster and
cement-asbestos-sand plaster due to somewhat better insulating
forced.
3.
fire
resistance of
704
subjected to
fire
and water
tests.
[Vol. 18
or
square vertically reinforced and round spirally hooped and vertically reinforced, the concrete aggregates employed being trap
rock and dolomitic limestone, except in the columns for the fire
and water tests, portions of which were made with siliceous gravel
All columns
concrete.
same as
Two
had a
2 -inch
concrete being
little affected,
cracked and
the case of the hooped column.
from a
tion
away the
if/& -inch
concrete
applica-
fire
exposure
and some
columns being quite generally exposed and the bars and spiral
hooping in the portion made with siliceous aggregates in the case
of the round columns.
After cooling the columns sustained at
failure loads averaging four times the design working load.
The
condition of the columns after the fire and water treatment was
not such that they would be subject to early failure in case of a
recurring fire, with the limitation on permissible percentages of
reinforcement prescribed by current standards of practice. Recurring fires are likely to be considerably less severe than an
unchecked fire, considering the effects of the preceding fire and the
water application. The conclusion was, therefore, reached with
reference to the columns in the investigation that the fire and
water tests developed nothing that would modify conclusions
from the results of the fire-endurance tests. This is significant,
since in the series reported in this paper no fire and water tests
of the adjacent portions, the vertical bars of the square
6
i
B.
184,
pp. 48 and
184, p. 113.
53.
184,
pp. 158-159.
?ngberg]
705
were made, and the results from the other investigation indicate
that for reinforced concrete columns of ordinary design conclusions can be safely based on results from fire tests alone.
Comparing the resistance periods deduced from the results of
the other investigation, vertically and combined laterally and
allowed
No
fire
resistance of concrete
columns, other than the one above referred to, have been made
under conditions comparable with the present investigation, the
To apply
some knowledge
fires
is
necessary of
and merchandizing.
The greatest amount
fires
fires
184, p. 213.
706
[Vol. 18
exposure in the tests have occurred. For residenand office occupancy some experimental investigation recently conducted at the Bureau of Standards, as also
observation of the effects of fires in buildings, indicate that fires
five
hour
tial,
institutional,
fire
the
first
XVI.
TABLE
15.
Tension
APPENDIX
Tests of Spiral Reinforcement
Modulus
of elas-
Specimen No.
Diameter
Ultimate
Elonga-
Reduction
strength
tion
in area
ticity in
million
pounds per
square
inch
Inch
a
b
c
0.316
.316
.312
Lbs./in.
Per cent
Per cent
104,500
106, 000
103, 000
3.4
3.4
44.8
54.9
44.7
30.4
31.4
29.1
104, 500
3.4
48.1
30.3
Hull
Ingbergj
TABLE
16.
1:2:4
:
Mortar cylinders
Concrete cylinders
Compressive
Column num-
Num-
ber
Age
Average
compres-
ber of
Aggregates
cylin-
sive
ders
strength
Months
Days
3
3
7
5
5
15
6
7
12
6
9
9
2
10
8
7
9
19
11
do..
do
do
do
do
-do.
6
7
29
10
6
7
7
6
7
9
7
7
8
13
23
10
7
7
9
10
7
8_.
9
10
12.
6
7
7
do
14
15.
25
.do
do..
26
27
do
do
28
33
34
35
do
do
do
do
36
....do.
5
5
5
5
.do
..do
41...
58
59
601
61
64
65
66
67..
4
2
4
6
20
do...
do
do
17
do...
do
do
do
3
3
9
5
4
4
4
do
do....
70...
do
do
do
3
3
72..
4
9
8
4
3
68
69
71
24
7
3
3
3
do
63._.
3
3
11
23
4
4
4
4
4
26
4
22
strength (average
of 5)
Lbs./in.2
i__.
2__
3...
707
days
Lbs./in. 2
1,593
1,948
1,665
28 days
Lbs./inJ
2,240
2,203
3,202
2,554
2,684
3,365
3,618
3,010
3,400
3,500
4,160
2,980
1,343
1,586
1,316
2,610
3,314
2,966
2,972
2,960
2,950
2,133
1,915
2,344
1,752
1,900
2,940
2,760
2,967
3,186
3,265
3,135
2,709
3,965
3,037
2,910
1,811
1,808
2,025
2,098
1,564
2,810
3,410
3,683
3,040
2,397
2,722
1,850
1,864
2,362
2,558
2,980
2,500
2,208
1,514
1,670
4,403
2,920
3,609
3,875
3,488
3,200
2,316
1,816
2,330
1,930
1,755
2,062
2,202
2,772
2,364
2,678
2,368
2,380
2,066
2,560
2,263
2,186
2,020
2,291
2,444
2,350
2,460
4,500
2,740
4,0^0
2,540
2,075
2,620
2,130
3,270
2,024
2,920
4,450
3,340
4,250
2,975
2,550
3,123
2,993
4.82C
3,988
4,700
4,370
74
do
do
77...
do
3
3
3
....do
do
3
3
3
4
4
2,450
2,324
2,440
2,114
12
2,834
2,180
3,234
8
10
7
7
7
17
23
4
7
2
4,240
4,060
2,862
4,137
3,849
4,660
2,071
2,050
2,061
1,838
2,083
1,950
2,870
3,299
3,760
2,586
3,159
3,050
73..
78.
80
82._
83._
88._
do
do
....doTotal
18....
19._
20._
21
22.
1
190
Aver age .
17._
4
4
4
4
by
5
5
6
5
7
7
16 inches.
708
TABLE
16.
Compressive
Col
Num-
ber
Mortar cylinders
Age
cylin-
sive
ders
strength
Months
West Winfield (Pa.) limestone
and Pittsburgh sand
23
do....
..do
30
32
.do...
Total
...do
.do ....
..do....
*
_
6
3
3
1
Total
of 5)
days
28 days
6
7
8
8
7
29
3
14
27
25
3,464
3,278
3,097
3,527
3,400
Lbs./in. J
2,106
1,862
1,946
2,206
1,882
28
3,664
2,005
3,109
4
4
4
4
11
7
8
24
3,162
3,160
2,529
3,560
3,156
3,770
2,750
4,493
4,798
4,605
Lbs./in.*
3,510
2,996
2,890
3,129
2,952
13
12
3,046
3,225
4,632
7
22
11
2
2,400
2,569
2,560
3,204
1,760
2.628
1,904
2,053
2,590
4,275
3,417
2,625
3,367
3,643
2,438
2,294
3,465
2,262
2,123
2,096
2,482
2,646
2,049
3,213
3,600
2,892
3,902
3,518
3,110
Long Island
37..
38.-.
gravel
and sand.
do
39
do....
7
5
8
12
46-.
do
do
47
Total..
23
Blast-furnace slag
48
and
Pitts-
49
50-.
51-.
52-.
do..
do
do
do
do
53
3
3
_.
3
3
Total
55-.
56
57
Jersey
trap
rock
21
5
12
2,858
2,370
2,328
2,965
2,609
2,644
13
2,629
2,276
3,372
4
4
4
4
7
16
5
12
2,540
3,108
2,200
2,238
2,030
1,830
3,078
2,960
2,445
3,053
3,030
26
and
3
3
3
3
.do
do
do
Total
12
4
4
4
4
18
New
S4
85
86.
strength (average
69
42
Average
Days
Lbs./in.*
6
7
7
7
7
do
._
Average
compres-
ber of
Aggregates
43
44
45.
Con.
Compressive
umn num-
24
29
[Voi.is
Total
10
2,522
2,102
2,085
2,434
3,365
4.
2,260
3,365
3,620
3,180
3,906
4,010
5,010
2,554
3,100
3,172
2,940
4,570
3,745
4,210
4,000
2,306
3,463
2,238
3,133
5__
75__.
76-.
81__
!
84
87
89
90.
Grand average.. .
Washington,
INDEX TO VOLUME
A
Accumulator.
18
Page
Damaged hosiery
Page
boxes, causes
157
619
Deformation
Airplane struts
243
Airships
609
permanent
Derry, A. T., John R. Freeman, jr., and, Effect of hot-rolling conditions on the physical
properties of a carbon steel
Tuckerman and, An
the mooring spindle of the Shen-
Aitchison, C. S., L. B.
analysis of
andoah
Aluminum
rectifiers
Analysis,
gases
by
305
609
Duck
Dyes used
Apparatus
E
243
465
229
Elasticity
609
107
Electrical,
381
Beams
243
305
parison of American
and
foreign clays as
paper fillers
Blanket tests, apparatus
337
Blankets, properties
529
Box dimensions
157
229
107
of iron
ductivity
by
465
Emissivity of paints
171
End restraint
381
of
columns
column strength
Epstein, Samuel, Henry S. Rawdon and, The
nick -bend test for wrought iron
Engesser's theory of
method
381
115
107
243
F
Fabric heat-transmission apparatus
Fahlman, Everett G., Robert J. Anderson and,
Development of a method for measurement
of internal stress in brass tubing
Federal Specifications Board specification
53
595
229
Fibers, paper
443
101
Fillers
337
327
635
189
635
thermal-con35
steel
35
Electrolytic rectifiers
No.
dioxide, analysis
529
277
Efficiency in rectifiers
Bicking, George
547
Eccentricity of tubes
curved
609
443
101
for stains
35
Camera
Carbon
hooks
465
thermal-conductivity
method
of
547
W. Schoffstall and.
Cardboard boxes
157
Cartridge-bag cloth
567
157
Cement brick
107
609
waste
157
silk
567
Clay brick
107
Clays
337
Coblentz,
W. W., and C.
IV. Hughes,
Emissive
Columns
381, 243
635
Girder
Grab method
(1
by
by 3
inch)
443
635
beams
327
Curtis,
H.
Sillman,
in flight
L.,
wrought iron
W. H.
camera
115
for
243
Heat
insulation (textiles)
529
value, discussion
studying projectiles
595
data
189
and
443
Wadleigh, and A. H.
35
Tom
Greene,
381
Crystallinity of
547
305
305
443
595
595
709
..
7io
Page
Page
apparatus
Heat-transmission
for
textile
fabrics, specifications
595
operation
595
blankets
529
method
H. D., and J. P.
performance
Honey,
Hooks
609
wrought iron
Nick -bend
test for
Numbered
cotton duck
115
443
specification
443
35
Schrodt,
Theory and
465
of rectifiers
solubility of sugars in
277
Output of rectifiers
Oxygen, effect on red-shortness
305
Hosiery
157
box designing
measurement
157
157
boxes
packing
157
157
203
Hot-rolling
547
465
of iron
327
P
Packing, improved, of men's hosiery
Paints, emissive tests
Palmer, P. E., and E. R. Weaver, Thermalconductivity method for the analysis of
gases
W.A., andS. H.
of concrete
101
making
337
of rectifiers
465
Permeability, textiles
529
air
columns
635
method
35
I
Ingberg, S. H.,
157
fibers
171
W. A. Hull and,
529
water vapor
Photographing projectiles in
529
flight
189
Physical properties
Plaster as fire protection for concrete col-
umns
Fire resistance
277
Iron, red-shortness of
327
Projectiles,
columns
635
camera
for
studying
levulose
381
colorless
waterproofing materials
L
Ladies' hosiery boxes
157
M
A
study
and
waste used
with an appendix
.
465
327
277
157
of fiber
115
327
fire resistance of
635
547
of
5 29
595
107
277
157
567
of fibers
465
Red-shortness of iron
Reinforced concrete columns,
McGowan
composition
Microscopy
Rectifiers
Schenke, E.
567
Reference standards
paper
Methods
465
Sale, P. D.,
567
and estimation
Rectification
of
171
203
Schoffstall,
fication
171
of silk
position standards.
171
measuring
Rolling
W.
R., Charles
of
paints
Kessler,
A. A. Mercier,
methods
277
McGowan, F.
189
Jackson, Richard F., and Gillis Silsbee, Saturation relations in mixtures of sucrose, dex-
and
203
189
photographing, in flight
J
trose,
547
635
of concrete
171
337
Performances
157
35
Paper
boxes
W. W. W.
Hughes, C.
Hull,
107
381
203
101
443
of
157
Index
to Vol.
18
711
Page
Schrodt, J. P.,
performance of
Secant formula
Selectivity,
rectifiers
465
measurement
of,
radio receiving
203
and,
camera
for
381
Tantalum
465
measurement
of,
sets
and
tubes
609
567
battery separators
Solubilities of sucrose
619
and dextrose
595
Textile specification
443
Textile test
277
Specification
443
Spindle, mooring
609
Standard boxes
Stang, Ambrose H., Herbert L. Whittemore
and, Tests of some girder hooks
157
305
Steel
54 7
tubes
Stone preservatives
Storage battery
243
methods
443
529
529
rectifiers
465
Tolerance, textiles
443
Transverse strength
Transverse tests of bricks
243
Tubes
243
229
189
465
rectifiers
W
Wadleigh, W.H..A.H. Sellman, H. L. Curtis
camera for studying projectiles in
and,
619
305
flight
381
Strength, brick
columns
tubes
Stress
157
381
Waste silk
567
243
305
381
229
189
107
Waste reduction
columns
Strip method,
465
Vibrating
619
on columns
609
Velocity of projectiles
Strain-gauge measurement
107
V
Valves, electrolytic
separators
tests
Thermionic
in sugar
mixtures
595
heat insulation
re-
443
243
Textiles, blankets
277
203
107
textiles
'
levulose
381
Tests, brick
107
columns
619
Shaw, Merle B., and George W. Bicking, Comparison of American and foreign clays as
Silsbee, Gillis,
for brick
paper fillers
Shenandoah
Silk waste
609
apparatus
radio receiving
rectifiers
Testing
189
Sensitivity,
studying projectiles in
flight
619
243
sets
Sellman,
Page
Sulphuric acid, effect on wood
Surface treatments for masonry
gases
Weaving of
381
stress
35
567
107
in metal tubing
229
Structures, deformed
609
Struts
243
277
method
on red-shortness
nick-bend test
" nick-bend clause in specifications
structure and mechanical properties
305
609
619
115
'
35
of iron
'
327
115
115